The Oxford Handbook of Genocide Studies (PDFDrive) PDF

the oxford handbook of
GENOCIDE
STUDIES
This page intentionally left blank
the oxford handbook of
...................................................................................................................................................
GENOCIDE
STUDIES
...................................................................................................................................................
Edited by
D O NA LD BLOX H AM
AND
A. DIRK MOSES
1
3
Great Clarendon Street, Oxford OX2 6DP
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
# The Several Contributors 2010
The moral rights of the author have been asserted
Database right Oxford University Press (maker)
First published 2010
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose the same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloging in Publication Data
Library of Congress Control Number: 2010920519
Typeset by SPI Publisher Services, Pondicherry, India
Printed in Great Britain
on acid-free paper by
CPI Antony Rowe, Chippenham, Wiltshire
ISBN 978–0–19–923211–6
1 3 5 7 9 10 8 6 4 2
CONTENTS
..........................................
List of Contributors ix
Editor’s Introduction: Changing Themes in the Study of Genocide 1

DONALD BLOXHAM AND A. DIRK MOSES
PART I: CONCEPTS
1. Raphael Lemkin, Culture, and the Concept of Genocide 19

A. DIRK MOSES
2. ‘Ethnic Cleansing’ versus Genocide? 42

BENJAMIN LIEBERMAN
3. Gender and Genocide 61

ELISA VON JOEDEN-FORGEY
4. The State and Genocide 81

ANTON WEISS-WENDT
5. Genocide and Memory 102

DAN STONE
PART II: INTERDISCIPLINARY PERSPECTIVES
6. The Law and Genocide 123

WILLIAM A. SCHABAS
7. Sociology and Genocide 142

MARTIN SHAW
8. Political Science and Genocide 163

SCOTT STRAUS
vi contents
9. Anthropology and Genocide 182

KEVIN LEWIS O’NEILL
10. Social Psychology and Genocide 198

PAUL A. ROTH
11. Philosophy and Genocide 217

MARTIN SHUSTER
PART III: PREMODERN AND EARLY

MODERN GENOCIDE
12. Genocide in the Ancient World 239

HANS VAN WEES
13. Early Medieval Europe: the Case of Britain and Ireland 259
JAMES E. FRASER
14. Central and Late Medieval Europe 280

LEN SCALES
15. Colonial Latin America 304

NICOLAS A. ROBINS
16. Rethinking Genocide in North America 322

GREGORY D. SMITHERS
PART IV: GENOCIDE IN THE LATE MODERN WORLD
17. Genocide and Mass Violence in the ‘Heart of Darkness’:

Africa in the Colonial Period 345
DOMINIK J. SCHALLER
18. Genocide at the Twilight of the Ottoman Empire 365

HILMAR KAISER
19. Mass Deportations, Ethnic Cleansing, and Genocidal

Politics in the Later Russian Empire and the USSR 386
NICOLAS WERTH
contents vii
20. The Nazi Empire 407

CHRISTOPHER R. BROWNING
21. Twentieth-Century China: Ethnic Assimilation and Intergroup Violence 426

URADYN E. BULAG
22. Political Genocides in Postcolonial Asia 445

ROBERT CRIBB
23. State-Sponsored Violence and Secessionist Rebellions in Asia 466

GEOFFREY ROBINSON
24. National Security Doctrine in Latin America: the Genocide Question 489
DANIEL FEIERSTEIN
25. Genocide and Population Displacement in Post-Communist 509

Eastern Europe
CATHIE CARMICHAEL
26. Genocidal Warfare in North-East Africa 529

ALEX DE WAAL
27. War and Genocide in Africa’s Great Lakes since Independence 550
OMAR MCDOOM
PART V: THE CONTEMPORARY WORLD:

RULES AND RESPONSES
28. The United Nations, The Cold War, and its Legacy 579
GERD HANKEL
29. Military Intervention 597

ALEX J. BELLAMY
30. Punishment as Prevention? The Politics of Prosecuting Génocidaires 617

DONALD BLOXHAM AND DEVIN O. PENDAS
31. From Past to Future: Prospects for Genocide

and its Avoidance in the Twenty-First Century 638
MARK LEVENE
Index 661
LIST OF CONTRIBUTORS
......................................................................................
Alex J. Bellamy is Professor of International Relations and Executive Director of

the Asia-Pacific Centre for the Responsibility to Protect at the University of
Queensland, Australia. His most recent book is Responsibility to Protect: The Global
Effort to End Mass Atrocities (2009).
Donald Bloxham is Professor of Modern History at the University of Edinburgh.
He is author of The Great Game of Genocide: Imperialism, Nationalism, and the
Destruction of the Ottoman Armenians (2005), winner of the Raphael Lemkin
Award for 2007; The Final Solution: A Genocide (2009); and Genocide on Trial:
War Crimes Trials and the Formation of Holocaust History and Memory (2001). He is
editor, with Mark Levene, of the Oxford University Press monograph series Zones
of Violence.
Christopher R. Browning is the Frank Porter Graham Professor of History at the
University of North Carolina at Chapel Hill. His publications include Ordinary
Men: Reserve Police Battalion 101 and the Final Solution in Poland (1992) and The
Origins of the Final Solution: The Evolution of Nazi Jewish Policy, September 1939–
March 1942 (2004).
Uradyn E. Bulag is reader in social anthropology at the University of Cambridge.
Author of Nationalism and Hybridity in Mongolia (Clarendon Press, 1998), and The
Mongols at China’s Edge: History and the Politics of National Unity (Rowman and
Littlefield, 2002), his interests broadly span East Asia and Inner Asia, especially
China and Mongolia.
Cathie Carmichael is a Senior Lecturer in European History at the University of
East Anglia. She is author of Ethnic Cleansing in the Balkans (2002) and Genocide
before the Holocaust (2009), co-editor of Language and Nationalism in Europe
(2000) with the late Stephen Barbour, and co-author of Slovenia and the Slovenes
(2000) with James Gow.
Robert Cribb is Professor of Indonesian History at the Australian National Uni-
versity. His publications include The Indonesian Killings of 1965–1966: Studies from
Java and Bali (1990), Gangsters and Revolutionaries: The Jakarta People’s Militia and
the Indonesian Revolution 1945–1949 (1991), and (with Li Narangoa) Imperial Japan
and National Identities in Asia, 1895–1945 (2003).
x list of contributors
Daniel Feierstein directs the Genocide Chair at the University of Buenos Aires. He
is Researcher at the Consejo Nacional de Investigaciones Cientificas y Tecnicas and
Director of the Center of Genocide Studies at the Universidad Nacional de Tres de
Febrero. He is author of Genocidio como práctica social (2007). His publications also
include State Violence and Genocide in Latin America: The Cold War Years (co-
authored with Marcia Esparza and Henry Huttenbach, 2009).
James E. Fraser is Senior Lecturer in Early Scottish History and Culture at the
University of Edinburgh. His publications include The Battle of Dunnichen 685
(2002) and From Caledonia to Pictland: Scotland to 795 (2009).
Gerd Hankel is Senior Research Fellow and Lecturer in International Public and
International Criminal Law at the Hamburg Institute for Social Research. His
publications include contributions to Antonio Cassese (ed.), The Oxford Compan-
ion to International Criminal Justice (2009), and ‘Rwanda. A Small Nation in Africa’,
in Madelon de Keizer and Ismee Tames (eds), Small Nations: Crisis and Confronta-
tion in the Twentieth Century (2008).
Elisa von Joeden-Forgey teaches History at the University of Pennsylvania. She is
the author of articles and book chapters on race and colonialism in German history
and is currently writing a book on gender and genocide, entitled Killing God: The
Family Drama of Genocide.
Hilmar Kaiser received his PhD from the European University Institute, Florence,
Italy. He specializes in Ottoman social and economic history with a special
emphasis on the Armenian Genocide. Currently he works and lives in Ankara,
Turkey.
Mark Levene is Reader in Comparative History at Southampton University. He is
involved in a four-volume project on Genocide in the Age of the Nation-State, vols. 1
and 2 of which were published in 2005. Much of his current work is about the
relationship between rapid anthropogenic climate change and violence. See Crisis
Forum, http://www.crisis-forum.org.uk, and the Rescue!History network, http://
rescue-history-from-climate-change.org/indexClassic.php, of which he is founder.
Benjamin Lieberman is Professor of History at Fitchburg State College. He is the
author of Terrible Fate: Ethnic Cleansing and the Making of Modern Europe (2006),
and ‘Ethnic Cleansing in the Greek-Turkish Conflicts from the Balkan Wars
through the Treaty of Lausanne: Identifying and Defining Ethnic Cleansing’, in
Steven Bela Vardy and T. Hunt Tooley (eds), Ethnic Cleansing in Twentieth-Century
Europe (2003).
Omar McDoom is a lecturer in comparative politics at the London School of
Economics and Political Science. He has previously held research fellowships at
Harvard’s Kennedy School and Oxford University. He is currently writing a book on
list of contributors xi
the Rwandan genocide entitled Why They Killed: Security, Authority, and Opportu-
nity in Rwanda’s Genocide.
A. Dirk Moses is Associate Professor in History at the University of Sydney. He is
the author of German Intellectuals and the Nazi Past (2007), and editor of Genocide
and Settler Society: Frontier Violence and Stolen Indigenous Children in Australian
History (2004), Colonialism and Genocide (2007, with Dan Stone), and Empire,
Colony, Genocide: Conquest, Occupation and Subaltern Resistance in World History
(2008). He an associate editor of the Journal of Genocide Research.
Devin O. Pendas is Associate Professor of History at Boston College. He is the
author of The Frankfurt Auschwitz Trial, 1963–1965: Genocide, History, and the Limits
of the Law (2006), as well as numerous articles and chapters concerning the history of
Holocaust trials, transitional justice, and the history of international law.
Kevin Lewis O’Neill is an assistant professor in the Department and Centre for the
Study of Religion at the University of Toronto. He is the author of City of God:
Christian Citizenship in Postwar Guatemala City (2009).
Nicholas A. Robins is a Lecturer in the Department of History at North Carolina
State University. His publications include Priest-Indian Conflict in Upper Peru: The
Generation of Rebellion (2007), Native Insurgencies and the Genocidal Impulse in the
Americas (2005), Genocide and Millennialism in Upper Peru: The Great Rebellion of
1780–1782 (2002), and The Culture of Conflict in Modern Cuba (2003).
Geoffrey Robinson is Professor of History at UCLA, where he teaches and writes
about Southeast Asia, political violence, and human rights. His works include: The
Dark Side of Paradise: Political Violence in Bali (Cornell University Press, 1995); and
“If You Leave Us Here, We Will Die”: How Genocide Was Stopped in East Timor
(Princeton University Press, 2010).
Paul A. Roth is Professor of Philosophy at the University of California-Santa Cruz.
His publications include Meaning and Method in the Social Sciences (1987) as well as
numerous articles on topics ranging from naturalism in epistemology to explana-
tion in history. He is a member of the editorial board of the Journal of the
Philosophy of History.
Len Scales teaches medieval European history at Durham University. He has a
particular interest in the pre-modern history of ethnicity and nationhood. His
publications include (ed., with Oliver Zimmer) Power and the Nation in European
History (2005).
William A. Schabas is Professor of Human Rights Law at the National University
of Ireland, Galway, where he directs the Irish Centre for Human Rights. He is the
author of Genocide in International Law, The Crime of Crimes (2nd edn, 2009)
and Introduction to the International Criminal Court (3rd edn, 2007).
xii list of contributors
Dominik J. Schaller is a lecturer and researcher at the Ruprecht-Karls-Universität

Heidelberg. He is co-editor of The Armenian Genocide and the Shoah (ed., 2002),
Late Ottoman Genocides: The Dissolution of the Ottoman Empire and Young Turkish
Population and Extermination Policies (2009), and The Origins of Genocide: Raphael
Lemkin as a Historian of Mass Violence (2009).
Martin Shaw is Research Professor of International Relations at the University of
Sussex. As a sociologist who writes on global politics, war, and genocide, his recent
books include War and Genocide (2003) and What is Genocide? (2007). He writes a
regular column on genocide and war for http://www.opendemocracy.net
Martin Shuster is a PhD candidate at the Humanities Center of the Johns Hopkins
University. He is interested in Kant, post-Kantian philosophy, and Jewish thought
and philosophy.
Greg D. Smithers is a lecturer in the Department of History at the University of
Aberdeen. He is the author of Science, Sexuality, and Race in the United States and
Australia, 1780s–1890s (2008) and, with Clarence E. Walker, The Preacher and the
Politician: Jeremiah Wright, Barack Obama, and Race in American History (2009).
Dan Stone is Professor of Modern History at Royal Holloway, University of London.
His books include Breeding Superman: Nietzsche, Race and Eugenics in Edwardian
and Interwar Britain (2002), Constructing the Holocaust: A Study in Historiography
(2003), Responses to Nazism in Britain, 1933–1939 (2003) and, as editor, The Histori-
ography of the Holocaust (2004) and The Historiography of Genocide (2008).
Scott Straus is Associate Professor of Political Science and International Studies at
the University of Wisconsin at Madison. He is author of The Order of Genocide:
Race, Power, and War in Rwanda (2006), winner of the 2006 Award for Excellence
in Political Science and Government from the Professional and Scholarly Publish-
ing Division of the Association of American Publishers, and of Rwanda, Intimate
Enemy (2006).
Alex de Waal is programme director at the Social Science Research Council (New
York), Senior Fellow at the Harvard Humanitarian Initiative, and co-director of
Justice Africa. He has written and edited several books related to war, famine, and
genocide, including Famine Crimes: Politics and the Disaster Relief Industry in
Africa (1997), Islamism and Its Enemies in the Horn of Africa (2004), and War in
Darfur and the Search for Peace (2007).
Hans van Wees is Professor of Ancient History at University College London. He is
the author of Status Warriors: War, Violence and Society in Homer and History
(1992), and Greek Warfare: Myths and Realities (2004), as well as numerous articles
on aspects of war and peace in the ancient world. He has edited, among other
list of contributors xiii
things, a volume on War and Violence on Ancient Greece (2000) and the Cambridge
History of Greek and Roman Warfare (2007).
Anton Weiss-Wendt is Head of the Research Department at the Center for the
Study of the Holocaust and Religious Minorities in Oslo, Norway. His publications
include Murder without Hatred: Estonians and the Holocaust (2009); ‘Problems in
Comparative Genocide Scholarship’, in Dan Stone (ed.), The Historiography of
Genocide (2008); and ‘Extermination of the Gypsies in Estonia during World War
II’, Holocaust and Genocide Studies 17:1 (2003).
Nicolas Werth is a research director at the Centre National de la Recherche
Scientifique in France. He lived in Moscow from 1975 until 1993, and was one of
the first Western historians to access Soviet Archives. He is the co-author of The
Black Book of Communism (1999).
.............................................................................................
E D I TO R S ’
I N T RO D U C T I O N
CHANGING THEMES IN THE
STUDY OF GENOCIDE
.............................................................................................
donald bloxham
a. dirk moses
T H E S TAT E OF GENOCIDE STUDIES

................................................................................................................
An Oxford Handbook of Genocide Studies is easily justified. ‘Genocide’ is unfortu-
nately ubiquitous, all too often literally in the attempted destruction of human
groups, but also rhetorically in the form of a word that is at once universally known
and widely invoked—perhaps because it is frequently misunderstood. From its
introduction to the international public sphere with the United Nations General
Assembly resolution on genocide in 1946, the term was seized upon by all sides to
name the criminality of their persecution. Indians and Pakistanis made representa-
tions to the UN, accusing the other of genocide during the partition while, soon
thereafter, Baltic states likewise accused the Soviet Union, and African Americans
the United States for lynchings and discrimination. Later, during the Vietnam
War, leftist intellectuals assembled an unofficial court to indict the United States
for a genocidal campaign. In the late 1960s and early 1970s, secessionist peoples,
2 donald bloxham and a. dirk moses
such as the Ibo of Nige‘ria and the East Bengalis of East Pakistan, accused the state
of genocide in their ruthless suppressions. Now the Sudanese government stands
accused of committing genocide against its citizens in the Darfur region. This new
vocabulary of atrocity and destruction, coined by Raphael Lemkin only in 1943, is
unlikely to disappear. Now a discipline exists to study it scientifically.
The rise of academic genocide studies is illustrated by the emergence of new
scholarly journals since the 1990s: the Journal of Genocide Research and Genocide
Studies and Prevention, and the German Zeitschrift für Genozidforschung, which
complement the more Holocaust-centred journal Holocaust and Genocide Studies.
Encyclopaedias of genocide (printed and electronic), Internet discussion groups,
and a proliferation of comparative works and collections of case studies now make
scholarship increasingly accessible.1 Universities everywhere are expanding existing
Holocaust programmes or specialist research centres to include genocide, although
usually maintaining the distinction between the Holocaust and genocide. At the
level of public policy, ad hoc international and ‘hybrid’ criminal tribunals sit in
judgement of genocide and crimes against humanity in Africa, Europe, and Asia;
an International Criminal Court (ICC) seeks to do the same on a more permanent
and universal basis; the UN now has special advisors on the crime; and the USA has
established a task force with a view to creating a dedicated machinery to help
forestall imminent genocides or intercede in ongoing cases. These responses to
genocide have in turn fostered an extensive literature in case law, jurisprudence,
and international relations. The field has come a long way since Raphael Lemkin’s
struggles to gain acceptance for his neologism ‘geno-cide’ in the mid-1940s and, a
decade later, to publish his history of genocide.2
The comparative scholarship of genocide began with Raphael Lemkin himself
and was continued by a small group of dedicated scholars through the later Cold
War period.3 It increased exponentially in the 1990s when the Rwandan slaughter
hammered home to Western scholars that genocide was not a thing of the past,
while the implosion of Yugoslavia showed that the West could still host the crime.
A raft of books that dealt with these and other cases in an episodic manner
appeared; i.e., each chapter was devoted to a particular instance of genocide,
unrelated to each other except as examples of a generic definition or to highlight
the author’s thesis about the role of racism, democratization, modernization, and
so forth.4
1 E.g. Online Enyclopedia of Mass Violence: http://www.massviolence.org

2 Lemkin’s papers in the New York Public Library contain numerous letters from publishing houses
declining his manuscript, which was never completed or published.
3 Helen Fein, Accounting For Genocide: National Responses and Jewish Victimization during the
Holocaust (New York: Free Press, 1979); Leo Kuper, Genocide: Its Political Use in the Twentieth Century
(New Haven: Yale University Press, 1982).
4 George J. Andreopolous (ed.), Genocide: Conceptual and Historical Dimensions (Philadelphia:
University of Pennsylvania Press, 1994); Norman Naimark, Fires of Hatred: Ethnic Cleansing in
editor’s introduction 3
Another strand of scholarship pointed in a different direction. If one term

captured the political and therefore historical imagination after the end of the
bipolar Cold War, it was ‘globalization’. The idea of an increasingly interconnected
world centred on a Western political-economic core sent some students of geno-
cide in search of historical precursors and antecedents in the expansion of the West
before the twentieth century. Rather than simply compare discrete events, they
proposed a more contextual approach that places those events in relation to one
another. Above all, they wanted to tie discrete events and national histories to
transnational and international processes.5 A contemporary interest in the con-
nections between genocide in the European colonies and the Holocaust at Europe’s
core may be understood within this framework.6
Given these heady developments, it is easy to forget that genocide studies began
as a marginal field, part offspring of, part uneasy junior partner to, the longer
standing discipline Holocaust studies, itself a child of the 1970s. The relationship
between study of the Holocaust and study of genocide warrants reflection, because
it has been both negative and positive, characterized variously by synergies, pro-
cesses of self-definition by mutual exclusion, and occasional resentment. On one
Twentieth-Century Europe (Cambridge, MA: Harvard University Press, 2001); Robert Gellately and
Ben Kiernan (eds), The Specter of Genocide: Mass Murder in Historical Perspective (New York:
Cambridge University Press, 2003); Eric D. Weitz, A Century of Genocide: Utopias of Race and
Nation (Princeton, NJ: Princeton University Press, 2003); Samuel Totten, William S. Parsons, and
Israel W. Charny (eds), A Century of Genocide: Critical Essays and Eyewitness Accounts, 2nd edn (New
York: Routledge, 2004). This tendency of the literature is analyzed in various works over the past two
decades by Mark Levene, culminating in his Genocide in the Age of the Nation-State, 2 vols. (London:
Tauris, 2005). It is critiqued from a theoretical perspective in A. Dirk Moses, ‘Toward a Theory of
Critical Genocide Studies’, Online Encyclopedia of Mass Violence, 18 April 2008: http://www.
massviolence.org/Toward-a-Theory-of-Critical-Genocide-Studies. It is critiqued via a case study in
Donald Bloxham, ‘Three Imperialisms and a Turkish Nationalism: International Stresses, Imperial
Disintegration and the Armenian Genocide’, Patterns of Prejudice 36:4 (2002), 37–58 and Bloxham’s
subsequent The Great Game of Genocide: Imperialism, Nationalism and the Destruction of the Ottoman
Armenians (Oxford: Oxford University Press, 2005).
5 Mark Levene, ‘Why Is the Twentieth Century the Century of Genocide?’, Journal of World History
11:2 (2000), 305–36; A. Dirk Moses, ‘Conceptual Blockages and Definitional Dilemmas in the “Racial
Century”: Genocide of Indigenous Peoples and the Holocaust’, Patterns of Prejudice 36:4 (2002), 7–36;
Bloxham, ‘Three Imperialisms’; Bloxham, The Great Game.
6 Early initiatives in this direction, though rarely cited in the comparative genocide literature, were
Woodruff D. Smith, The German Colonial Empire (Chapel Hill, NC: University of North Carolina
Press, 1978); idem, The Ideological Origins of Nazi Imperialism (New York: Oxford University Press,
1989). For the first post Cold War study, Sven Lindquist, ‘Exterminate All the Brutes’ (London: Granta,
1992). For more up-to-date work, see A. Dirk Moses and Dan Stone (eds), Colonialism and Genocide
(London: Routledge, 2007); A. Dirk Moses (ed), Empire, Colony, Genocide: Conquest, Occupation and
Subaltern Resistance in World History (New York: Berghahn Books, 2008); Jürgen Zimmerer and
Joachim Zeller (eds), Genocide In German South-West Africa: The Colonial War (1904–1908) in
Namibia and Its Aftermath (London: Merlin Press, 2008). For a different perspective, see Robert
Gerwarth and Stephan Malinowski, ‘Der Holocaust als kolonialer Genozid? Europäische
Kolonialgewalt und nationalsozialistischer Vernichtungskrieg’, Geschichte and Gesellschaft 33 (2007),
439–66.
side of the ledger, the notion of the Holocaust’s ‘uniqueness’, based usually on the
totality of the Nazis’ murderous intent towards Jews, has worked to distinguish it
from the rest of the field—with a number of distorting effects.7 If the Holocaust is
taken as an ‘ideal type’ genocide, scholars and advocates of particular cases often
seek to fit theirs within a ‘Holocaust paradigm’ at the expense of careful context-
ualization.8 There are political consequences of this implicit hierarchy. The histo-
rian Peter Novick was one of many commentators to note this function of
Holocaust memory in the reluctance of Western policymakers to intervene in the
former Yugoslavia in the 1990s because its civil war did not resemble the Holo-
caust.9 Then there is the tendency in university syllabi, textbooks, and the mantras
of public commemoration of genocide to focus upon a few instances of genocide
that, for a variety of reasons, have qualified for the canon of general acceptance:
alongside the Holocaust, Armenia, Cambodia, Rwanda, and the former Yugoslavia,
and now Darfur tend to be included, but virtually no other cases.
A related consequence of the influence of the Holocaust model has been to
overemphasize the role of narrow political ideology in genocide. This tendency is
manifested by placing the Holocaust (along, perhaps, with one or two of the other
aforementioned select genocides) in a special class of ‘ideological genocides’, neatly
distinguished from more supposedly utilitarian genocides, such as those com-
mitted in the course of European imperial expansion, civil wars, and campaigns
of resource appropriation.10 Such a taxonomy ignores both the concessions that
the Nazis did make to practicality in their murder campaigns and, more impor-
tantly, the obviously ideological considerations that ultimately permitted, say,
Europeans to murder, say, native Americans as if they were simply practical
obstacles to be removed. This sort of specious dichotomy of motivation leads to
the implicit but nevertheless tangible sense that Holocaust and genocide studies
scholars exhibit the same Eurocentric insensitivities that have long characterized
Western attitudes towards the indigenous peoples crushed beneath the wheels of
occidental ‘progress’.11
7 Some scholars, such as Yehuda Bauer and Raimond Gaita, have at times sought to distinguish
Holocaust from genocide, the former connoting the intention of total destruction, the latter
something less absolute. See Bauer, ‘Comparison of Genocides’, in Levon Chorbajian and George
Shirinian (eds), Studies in Comparative Genocide (Basingstoke: Palgrave MacMillan, 1999), 31–43;
Gaita, ‘Refocusing Genocide: A Philosophical Responsibility’, in John K. Roth (ed.), Genocide and
Human Rights: A Philosophical Guide (London: Palgrave MacMillan, 2005), 153–66.
8 David Moshman, ‘Conceptual Constraints on Thinking about Genocide’, Journal of Genocide
Research 3 (2001), 432.
9 Peter Novick, The Holocaust in American Life (Boston: Houghton Mifflin, 1999), 245–55.
10 E.g. Saul Friedländer, ‘The Historical Significance of the Holocaust’, in Yehuda Bauer and
Nathan Rotenstreich (eds), The Holocaust as Historical Experience (London and New York: Holmes
and Meier, 1981), 4.
11 For a critique of this implicit theodicy, see Moses, ‘Conceptual Blockages and Definitional
Dilemmas in the “Racial Century” ’; Donald Bloxham, ‘Britain’s Holocaust Memorial Days:
Reshaping the Past in the Service of the Present’, Immigrants and Minorities 21 (2003), 41–62.
On the equally important other side of the ledger, initial interest in the Holo-
caust has often provided a springboard from which many students of genocide,
who were often Jewish, leapt to examine other cases in the 1980s, especially the
Armenian one.12 Moreover, the more advanced Holocaust historiography could
provide ready-made analytical questions for application in the different circum-
stances of other genocidal situations.13
An obvious function that a volume such as this can perform is to provide one of
the periodic updates that any field of enquiry needs. We seek to expand upon,
summarize, and help to analytically hone the mass of scholarship now being
conducted in the field across a variety of disciplines, and at the frontier interface
between the academic and the activist spheres—a ground that a number of our
authors occupy to the benefit of both sides. The diverse abundance of research and
reflection on genocide is a cause for satisfaction, but there is also the need to ask
what it all amounts to, and where it might be heading. If the present volume does
not presume to prescribe one future direction, it does seek to promote what its
editors see as some of the most fruitful avenues of analytic enquiry. Our interpre-
tation of what qualifies as fruitful is, inevitably, conditioned by our own disciplin-
ary presumptions, and this point brings us to a significant justification for the
volume.
H I S TO R I C I Z AT I O N AND C AU S AT I O N
................................................................................................................
For us as editors it is important that this handbook is published within Oxford
University Press’s history list. The ‘historicization’ of genocide is one of the central
goals of the volume. We argue for the importance of a historically based, interdis-
ciplinary method that embeds critical theories more firmly in empirical data. In
this way, we believe that the concept of genocide will become more useful to both
historians and social theorists, and more relevant to scholars in other fields who do
not currently regard the genocide concept as valuable for their particular area.
Our comments are categorically not a criticism of the existing collections of
essays on genocide. Indeed, a major section of this volume (part II) is devoted to
elucidating the ways in which the approaches of non-historians have illuminated
and in some cases pioneered the investigation of the subject. Lawyers, for instance,
12 This tendency culminated in Robert Melson, Revolution and Genocide: On the Origins of the
Armenian Genocide and the Holocaust (Chicago: University of Chicago Press, 1992).
13 Now, though, we are seeing how the quality of research among second and third generation
genocide scholars can begin to shed light back onto the Holocaust. Scott Straus, ‘Second Generation
Comparative Research on Genocide’, World Politics 59:3 (2007), 476–501.
have been at least as adept as historically oriented scholars in devising increasingly

sophisticated conceptualizations of intent, which remains the key concept in estab-
lishing the case for genocide. Besides, other disciplines can illuminate important
dimensions of genocide better than history as a result of their methodology: for
instance, social psychologists contribute comparative insights from cross-cultural
analysis of human behaviour in specific contexts, and anthropologists provide
insights into cultural particularities in ways that have influenced historians much
more than vice versa.14 Our intention is, as much as anything, to call for more
historians to bring their skills to bear in tandem with other traditions of scholarship:
genocide studies should be an interdisciplinary exercise par excellence.
Where historians can make a particular contribution to the interdisciplinary
exercise and where, therefore, the greater weight of analysis in this volume falls is in
the consideration of temporality, contingency, and particularly the matter of
causation. We understand causation not just in the terms of the immediate lead-
up to the crime, or the deeper background causes that are often referred to as
contexts, but in terms of the proposition that genocides are the outcome of
processes rather than ‘punctual’ events: they, and the people who enact them, are
constantly evolving phenomena, subject to a multiplicity of external influences as
well as internal volition. Structure and agency are inextricably intertwined.
Leaving aside the scholarship on the ‘final solution of the Jewish question’, which
has primarily been the preserve of historians, we find that, until recently, historians
have contributed relatively little to the study of genocide. More precisely, while
some historians have been involved in elucidating case studies, the shape and
assumptions of the field have been governed by disciplines more at ease with
making comparisons and contrasts across large tracts of time and space, and
with dealing with contemporary affairs. Even those historians who have written
wide-ranging studies on genocide and related matters have had perforce to write in
the style not of the idiographic, cultural tradition of history, which is concerned
with specificity and difference, but of the more nomothetic or law-creating tradi-
tion associated with the interface between history and the social sciences.15 Both
approaches have the obvious costs and benefits involved in the trade-off between
detailed knowledge of particularities and general knowledge of broader applicabil-
ity; and, in the event, every piece of historically oriented scholarship will inevitably
combine aspects of both. This volume seeks to establish a somewhat different
balance of the two than evinced in the existing scholarship. It attempts to blur the
sharp division that comparable collections, such as the excellent, recent ones edited
by Dan Stone and Ben Kiernan and Robert Gellately, still tend to exhibit between
14 E.g., Alexander Laban Hinton (ed.), Annihilation Difference: The Anthropology of Genocide
(Berkeley: University of California Press, 2002).
15 Moses, ‘Towards a Theory of Critical Genocide Studies’.
one-genocide-per-chapter empirical studies and broad thematic/conceptual stud-

ies, as mentioned above.16
The extant collections of commissioned essays are of varying coverage and
approach, but some commonalities are apparent. As noted above, as with single-
authored comparative works by the likes of Leo Kuper and Helen Fein, and latterly
Eric D. Weitz, Norman Naimark, and Ben Kiernan, the collections edited by Israel
Charny, Frank Chalk and Kurt Jonassohn, George J. Andreopoulos, Samuel Totten,
and William S. Parson, have similar approaches to their subject matter.17 They have
either focused on a few ostensibly comparable cases to try to establish general
similarities or patterns, and/or they have sought to categorize cases according
to broad typologies. Such homogenizing or taxonomic approaches reflect the
disciplinary grounding of most of the first genocide scholars in the social sciences,
above all, political science and sociology. As a result of this approach, relatively
little of the pioneering scholarship that established the sub-discipline of genocide
studies was based on original, empirical research. Unsurprisingly, given the youth
of the field that they were so important in shaping, these books also relied upon
limited acquaintance with the lesser known cases studies with which they deal.
Where specialists on these obscure instances were included, their enlistment
nevertheless tended not to result in sufficient space being allocated for the particu-
larity of the history to be fully revealed.
At the same time as being schematic in its approach, the earlier scholarship was
also too exclusive in its frame of reference. It created a conceptual trap for itself
because of a preoccupation with the definition and applicability of the term
genocide. This phenomenological approach provided an interesting replication
of the tired ‘uniqueness’ battle in Holocaust scholarship, in that it has served to
exclude issues from debate rather than stimulating constructive reflection on the
parameters of the subject. The approach has ramifications for historical under-
standing, with instances of outright genocide being accorded more attention than
other related phenomena.
The problem of ‘definitionalism’ has not left us, and it is unlikely to, given the
centrality of legal conceptualization at the ‘hard’ political end of genocide studies,
and the political capital that groups and states invest in claiming or denying the
applicability of the term to cases of particular concern to them. The inflated use of
the term genocide is the other side of the coin to overly restricted usage. The
question of definition inevitably recurs throughout this volume, but not to the end
16 Dan Stone (ed.), The Historiography of Genocide (London: Palgrave, 2008); Gellately and Kiernan
(eds), The Specter of Genocide.
17 Helen Fein, Genocide: A Sociological Perspective (London: Sage, 1993); Israel W. Charny (ed),
Genocide: A Critical Bibliographical Review (London: Mansell, 1988); Frank Chalk and Kurt Jonassohn
(eds), The History and Sociology of Genocide (New Haven: Yale University Press, 1990); Andreopoulos
(ed.), Genocide: Conceptual and Historical Dimensions; Ben Kiernan, Blood and Soil: A World History of
Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2008).
of some spuriously positivistic attempt to nail down ‘our’ version of it. As the
historian Charles Maier has noted, ‘Taxonomy in the social sciences is always
difficult, and often unfruitful.’ To attempt to corral cases into a single paradigm
is a Sisyphean conceit that ignores the inductive logic by which comparisons are
made. Definitive answers cannot be expected when we know that ‘an ideal type will
not fit any individual case exactly: It’s an abstraction from them all.’18
This collection is committed to probing the limits and the utility of the concept
of genocide for historical understanding, and placing the crime back in its
context(s) that may often include mass non-genocidal violence. We do this because
the focus on upper case Genocide often entails a focus on outcomes rather than
causes and processes that may or may not produce the mass killing which many
think is the substance of genocide. The focus on specific types of outcomes that
qualify as genocide is analogous to studying the peaks of mountains from above a
cloudline that only particularly tall mountains penetrate, when a glimpse beneath
the cloudline would illustrate that other mountains fell just short, and that the
tallest mountains were connected to others contiguously or via foothills and
valleys. Understanding the context in which genocide occurs is equivalent to
viewing the landscape in a wider perspective.19
The volume seeks not simply to examine particular cases or ideologies of geno-
cide, but to reflect in a more rounded way upon the relationship between genocide
and broader historical trends, periods, and structures. The approach entails going
beyond strictly comparative scholarship to something more consciously correlative
and contextual. This approach can be operationalized at a number of levels, whether
that of the individual group or polity, or of the supranational region, continent or
even world. Martin Shaw’s call for an international relations of genocide is an
attempt to associate different episodes within an overall developmental pattern of
interconnected state behaviour.20 Michael Mann’s The Dark Side of Democracy:
Explaining Ethnic Cleansing (2005) sheds light on how a number of individual
cases of genocide—occurring notably in ethnically plural societies passing through
key developmental phases when the ethnopolitical identity and territorial integrity
of the state is contested—might fit into such a framework; along the way, with
18 Charles S. Maier, ‘American among Empires? Imperial Analogues and Imperial Syndrome’,
German Historical Institute Bulletin 41 (2007), 21–2.
19 This is also the rationale for a project conceived in 2004 by Mark Levene and Donald Bloxham
and forthcoming as a series of ten monographs with Oxford University Press entitled Zones of
Violence. The attempt to combine critical conceptual approaches with detailed regional history and
empirical reconstruction is at the basis of Bloxham’s The Great Game and his volume on the Holocaust
in a continental context The Final Solution: a Genocide (Oxford University Press, 2009). Christian
Gerlach, Extremely Violent Societies (Cambridge: Cambridge University Press, 2010) also sets
individual cases into stimulating wider contexts.
20 See his chapter in this volume.
reference to European settler societies in colonial contexts, he debunks the spurious

notion that democracies do not engage in genocide, and thus the idea that the
spread of (capitalistic) democracy is the antidote to genocide as well as war.21
It is Mark Levene, however, who has gone furthest in attempting an overall
analysis of an international political economy that seems to have some inherent
tendencies towards encouraging murderous intergroup conflict, and in putting
that political economy in a distinct historical context, therefore fruitfully combin-
ing synchronic and diachronic elements. Levene’s Genocide in the Age of the Nation-
State (the first two of four volumes of which appeared in 2005) does not deny the
instance of genocide in premodern times, but is concerned with the particular
potentiality for genocide in a post-French revolution, post-industrial revolution
world in which states engaged in increasingly desperate and unrestrained competi-
tion for sovereign autonomy and thus resources. As a prerequisite for the struggle
they developed a heightened preoccupation with the loyalty and thus identity of
their populations, which were at once a major source of potential strength and
weakness. In this milieu, states reacted with increasing violence against heteroge-
neous elements perceived to be either holding back development because of
particular cultural patterns or threatening state integrity by their split loyalties.
Democratic and free market states contributed to the murderous dynamic quite as
much as authoritarian/totalitarian regimes and states with command or dirigiste
economies.
While we editors broadly concur with such structuralist and materialist inter-
pretations, and have expanded upon our views elsewhere, by no means all of our
authors would subscribe to the same views. We have included a number of scholars
whose interpretations cohere with a classically liberal understanding of genocide,
where the crime results above all from aberrant political ideologies and oppressive
political systems, and where the problem of genocide can be solved by the reasser-
tion of the healthy norms of international democratic society. Others of our
contributors may well have no strong view on the matter, or no general view at
all beyond their depiction of their own case(s), and all contributions can be read as
self-sufficient, independent essays. In no way do we seek to impose our own
editorial conception on our authors, and the fact that all can fit into the volume
suggests that distinctions between philosophies of genocide scholarship are often
more polarized in the abstract than in practice where, once again, the difference
between opposing depictions of the same historical episode are matters of degree
and nuance rather than anything else: in reality, all explanations eschew mono-
causality and embrace varying contexts and contingencies, just as even the most
contingent or, again, the most structural explanations cannot account for anything
without some reference to ideology and human agency.
21 Michael Mann, The Dark Side of Democracy: Explaining Ethnic Cleansing (Cambridge:
Cambridge University Press, 2005).
All we seek is to open up the debate about the relationship between individual
acts of genocide and the wider political economy and norms of the worlds in which
they occur, whether or not those worlds are coextensive with the actual globe (and
many of our studies do not require such a broad unit of analysis). We implicitly ask
to what extent common features of organized human life across large tracts of
space and equally importantly time, and amongst the widest varieties of peoples
and polities lend themselves to something like what we today call genocide. Such
features include competition for land and resources, imperial expansion, warfare,
subordination of populations along political or cultural lines, sovereignty disputes,
security fears, accelerated socio-economic change, and the re-casting of traditional
social hierarchies at moments of sudden flux such as revolution.
P L AC I N G G E N O C I D E IN H I S TO R I C A L ‘T I M E ’
................................................................................................................
Our part III on genocide in premodern and early modern times sheds light not just
on common contextual settings for genocide across the ages but also on some
continuities in the patterns of genocide’s enactment. Alongside the murder of elite
bearers of identity and the killing (or enslavement) of ordinary men capable of
resistance, such measures include the kidnapping (and rape) of women and
children for forced acculturation and the widespread destruction of cultural
artefacts. While not universal, there is clearly a set of widely deployed genocidal
pragmatics.
Contrary to the opinions of some scholars of the premodern world, and of states
like Turkey that have a vested interest in denying the applicability of the term
genocide to acts in their own past, it is therefore not anachronistic to discuss
genocide avant la lettre, no more than it is anachronistic to apply the modern
heuristic term ‘feudalism’ to the premodern past. Lemkin himself was convinced
that genocide had always been a part of the human experience, and the UN
Convention on the Prevention and Punishment of Genocide explicitly refers to
its transhistorical character. What certainly have changed over time are the social
cleavages on which genocide-like violence is perpetrated. It is in the enumeration
of potential victim groups on select grounds of communal identity that the
specifically mid-twentieth century context of the Genocide Convention is exposed,
whatever the historical allusions of the document. To what extent we can use the
Convention’s terms concerning ‘ethnic’, ‘national’, and ‘religious’ groups (not to
mention ‘racial’ [sic] groups) for different times in human history is open to a
contestation that varies in intensity depending upon the period in question. That is
true over and above the general recognition in today’s historical and social
scientific scholarship that all human communities of identity are constructed to

one degree or another rather than simply ‘given’.
If types of group vary over time, so too do ideas of how membership is
constituted and, thus, how the group might be dismembered. Accordingly, dis-
cussing potential cases of genocide from beyond the realm of a Western-created
modernity sometimes requires thinking along different lines of logic about group
destruction. For instance, the early medieval practice of strages gentium that James
Fraser describes in his chapter in our section on premodern and early modern
genocide illustrates precisely how the destruction of a proportionally small number
of certain elite signifiers of a group could be sufficient to represent the destruction
of the group ‘as such’, to use the Convention’s terms.
Given different ideologies, cleavages, and logics of genocide, does it therefore
make sense to delineate ‘modern’ from ‘premodern’ genocide in the same way one
might crudely delineate modern from premodern society? The answer depends
upon how one understands the protean concept ‘modernity’. An understanding
that leans particularly upon modernity’s material (economic and technical) aspects
would of course allow that the development of surveillance, bureaucracy, central
state strength, weaponry, etc., would create greater facility to pursue and murder
‘enemies’, and would equally allow that the increasing contact between different
peoples and the more intensive and extensive exploitation of resources might
provoke more and increasingly intense intergroup conflicts, but distinctions
along these lines between modern and premodern are of degree rather than
fundamental nature.
A comprehension of the peculiarities of modernity that more emphasizes cul-
tural, intellectual, and philosophical shifts is encapsulated in the work of Zygmunt
Bauman, which itself draws on deeper traditions in continental critical theory.
Bauman provided the scholarship of modern genocide with one of its most
powerful motifs when he invented the concept of the modern human society as
garden, the modern state as gardener.22 His conception of modernity incorporates
its economic and technical aspects but those are subordinate to a post-Enlighten-
ment cultural order embodied in growing secularism and a spirit of scientistic
problem-solving. For Bauman, modernity is characterized by man’s belief that he
can reshape humankind in an image of perfectibility that in a more religious age
was regarded as the sole preserve of God. Modern genocide is the radical applica-
tion of this doctrine of perfectibility by one particular section of mankind against
debilitating or imperfectible elements within and outside its collective body.
22 Zygmunt Bauman, Modernity and the Holocaust (Ithaca, NY: Cornell University Press, 1989). His
influence can be detected most clearly among the contributors to Amir Weiner’s collection
Landscaping the Human Garden: Twentieth-Century Population Management in a Comparative
Framework (Stanford: Stanford University Press, 2003).
We do not seek to arbitrate one way or another over the matter of whether a
Western-created modernity differs fundamentally, or only in degree, or only in
certain ways, from ‘premodernity’.23 What we would observe is that it tends to be
students of the modern, from Nietzsche to Foucault to Bauman, who invest most
in modernity’s putatively radical difference, whereas scholars of the premodern are
as interested in continuities as changes, partly because they see so many premodern
precursors of what modernists take to be quintessentially modern social, econom-
ic, and cultural developments. Medieval manifestations of genocide are, we believe,
particularly interesting because the medieval period is often held up by theorists of
the modern (and the postmodern) as modernity’s ‘other’ in terms of intellectual,
social, and political arrangements.24
On the matter of genocide, it is at least worthy of further discussion, then, that
Len Scales’s chapter adduces the example of a fourteenth-century English notary
in Dublin who advised his compatriots that ‘when [the Irish] fall into your hands
pluck them all up by the root, as the good gardener doth the nettle.’ Here is no
mere fleeting coincidence of vocabularies with Bauman’s: the notary’s injunction
was made in the idiom of root and branch destruction in the interests of order
and cleanliness, in a context where notions of filth and disease were equated not
just with sin but with threat, and where difference, even among members of
ostensibly the same religious faith, was by no means necessarily considered
reconcilable.
GENOCIDE AND THE W O R L D T O DAY

................................................................................................................
However one elucidates the relationship of genocide to the prevailing cultural-
philosophical order(s), it is demonstrably the case that in the contemporary world
and for the foreseeable future the perpetrator–victim relationship is complicated
by the existence of a distinctly modern ‘world order’. This order, as manifested
most obviously in occasional third party interventions, but more consistently
and characteristically by partisan political and economic pressures that can be
23 See the chapters by Ben Kiernan, Eric D. Weitz, Omer Bartov, and Marie Fleming in the section
‘Genocide and Modernity’ in Kiernan and Gellately (eds), The Specter of Genocide; the first volume of
Levene’s Genocide in the Age of the Nation State; A. Dirk Moses, ‘Genocide and Modernity’, in Stone
(ed.), The Historiography of the Holocaust, 156–93; Martin Shuster’s chapter in this volume; Donald
Bloxham, ‘Modernity and Genocide’, European History Quarterly 38:2 (2008), 294–311; and the final
chapter of Bloxham, The Final Solution: A Genocide.
24 Elizabeth Deeds Ermath suggests modernity might be called ‘postmedievalism’ (‘Ethics and
Method’, History and Theory 43 (2004), 69).
inflammatory as well as pacificatory, has been hailed as a potential panacea to

genocide.25 In assessing its prospects, we come to the final contribution of the
volume.
In considering the relationship between genocide and the way the world order is
set up, we seek to problematize the prospect of the ‘international community’ as
benevolent policeman, intervening in genocidal situations and punishing génoci-
daires. In this sense, we provide an implicit critique of the sort of picture painted by
Samantha Power in her Pulitzer prize-winning ‘A Problem from Hell’: America and
the Age of Genocide (2003). In Power’s vision, the USA, as political and cultural
leader of the international community (and, during the Cold War, of the ‘free
world’), exists in a solely reactive position vis-à-vis genocide. In her final analysis,
all that is really needed to combat genocide is for the USA to assert its values with
greater conviction and consistency, a greater determination to lead and an en-
hanced preparedness to assume the human and financial costs of interventionist
action. In a similar vein, the opening passage of the executive summary of
Preventing Genocide, the report of the US ‘Genocide Prevention Task Force’, chaired
by Madeleine Albright and William Cohen, declares that ‘genocide and mass
atrocities threaten American values and interests.’ Such approaches focus, as it
were, on America’s ‘sins of omission’ rather than its ‘sins of commission’. If we
allow that genocide does indeed run against American values, it has frequently
cohered with American interests, as we can see before the twentieth century in the
expansion of the United States of America at indigenous expense, and during the
twentieth century in American support for regimes engaged in genocide or ‘poli-
ticide’ in, inter alia, Latin America, Indonesia, and Iraq, some of which are
considered in this volume.
That which applies to American interests and the interests of so many other
states past and present also applies to the international institutions that now have a
greater potential than ever to intervene in and punish cases of genocide and related
atrocity. Our last section (V) examines some of these institutions and their pro-
spects. It examines them not simply in a responsive capacity, but in a way that
considers what forms of political violence they tacitly permit. This tacit permission
is distributed in two related ways. The first way is via the norms embedded in the
structures of international law and international custom. Those norms have
evolved partly in response to mass atrocity and partly in the interests of a state-
based international system in which, whatever the fashionable rhetoric of account-
ability, sovereign viability and extensive sovereign freedom of action within the
domestic realm still remain hugely important structural features. The second way is
through the power-political constellations—particularly in the form of great
powers and their alliances—that shape and restrict the reach of ostensibly
25 E.g. Gary Bass, Freedom’s Battle: The Origins of Humanitarian Intervention (New York: Knopf,
2008).
supranational organizations like the UN and the ICC in accordance with matters of
strategic and ideological interest. If genocide studies is to have any activist im-
plications, the critical attention paid to transgressive regimes must also be extended
to the states and organizations that create and (selectively) enforce the rules
themselves. Some of those states also provide livings for the vast majority of
genocide scholars.
The overall balance of the volume is as follows: beyond the present chapter, there
are fifteen broadly thematic chapters in parts I, II, and V and sixteen more spatially
and temporally delineated chapters in parts III and IV. All of the former are
grounded more or less extensively in concrete examples; almost all of the latter
reach further than individual instances and contain comparative and correlative
elements.
Our authors have taken advantage of their broad remits such that, for instance,
Christopher Browning writes not on ‘the final solution’ but on ‘the Nazi empire’;
Nicholas Werth writes on the murderous continuities as well as contrasts between
Tsarist and Soviet policy in managing a diverse and rapidly modernizing imperial
space; Daniel Feierstein considers a number of Latin American countries in the
politico-spatial context of a concerted transnational anti-communist policy;
Africa is divided up not according to individual genocides but according to
zones and particular time periods that allow for internal and external comparison;
genocides in Asia are divided into conceptual categories such as ‘secessionist’; and
so on. In this sense, there is not a clear divide between conceptualization and case
study, simply a variation in the relationship between the theoretical and the
empirical.
We have tried to be as balanced as possible in terms of chronological and
geographical coverage. One of the novelties of the volume is the extensive space
devoted to premodern cases, but in the modern period we have also sought to
establish a distribution that is neither Eurocentric, neither biased towards the era
of the two world wars, nor limited to the better known non-European cases.
Inevitably some candidates for inclusion are not present, but this is not necessarily
due to our myopia; a salutary lesson in putting the volume together was the
difficulty in locating suitably qualified authors who were interested in addressing
genocide in their area of expertise; we also suffered from the inevitable attrition
involved in compiling a large collective volume, as some of our authors withdrew
too late in the day to be replaced. We regret the absence of planned chapters on
genocide and empire, war, and the question of genocide in medieval Eurasia. With
the exception of the final one, most of the substantive issues that would have been
raised in those missing chapters have been covered to one degree or another
elsewhere in the book.
It is with particular regret that we had to forego a chapter on the relationship
between economic developmentalism and genocide, since that would have fur-
thered our interrogation of the structural violence inherent to the imperative of
material progress upon which the modern world is constructed.26 Nevertheless, the
book concludes with a reflection on one of the most deleterious impacts of the
ideology of developmentalism for homo sapiens: anthropogenic climate change
and its ramifications in coming conflicts of resource scarcity and mass refugee
movement.
We are grateful to all our authors for considering our often extensive suggestions
for expansion and cross-integration of their material, and for the rewriting that we
frequently requested. All responded with grace and professionalism. Each of their
essays includes a select bibliography of relevant further material in English; the
footnotes will provide additional specialist reading across the full range of relevant
languages.
26 Vinay Lal, ‘The Concentration Camps and Development: The Pasts and Future of Genocide’,
Patterns of Prejudice 39:2 (2005), 220–43; Gerlach, Extremely Violent Societies; Nafeez Mosaddeq
Ahmed, ‘Structural Violence as Form of Genocide: The Impact of the International Economic
Order’, Entelequia. Revista Interdisciplinar, [online]. Accessed January 2009; available at http://www.
eumed.net/entelequia
part i
.............................................................................................
CONCEPTS
.............................................................................................
chapter 1
.............................................................................................
RAPHAEL LEMKIN,
C U LT U R E , A N D T H E
CONCEPT OF
GENOCIDE
.............................................................................................
a. dirk moses
................................................................................................................
‘In the beginning was Raphael Lemkin’ is effectively how the conventional accounts
of the genocide concept begin. As the coiner of the word and ‘father of the genocide
convention’, his person is held to be coeval with the concept, so that biography
replaces intellectual history. That biography is written teleologically as a heroic
struggle against the odds, consummated in the ‘United Nations Convention on the
Punishment and Prevention of Genocide’ in 1948, and invested with poignancy
after his death, alone and exhausted in 1959, a martyr to the cause. His subsequent
obscurity intensified the hagiographical imperative in the recent accounts to revive
his memory and honor his achievement.1 Enthusiasts have now devoted plays to
Lemkin, and a book prize is given in his name. Reclaiming the lost son, the Polish
1 Samantha Power, ‘A Problem from Hell’: America and the Age of Genocide (New York: Basic Books,
2002); William Korey, An Epitaph for Raphael Lemkin (New York: Blaustein Institute for the
Advancement of Human Rights, 2002); John Cooper, Raphael Lemkin and the Struggle for the
Genocide Convention (Houndmills: Palgrave MacMillan, 2008).
20 a. dirk moses
state named a conference room at the Ministry of Foreign Affairs after him in 2005,
and then mounted a commemorative plaque on the house in which he had lived in
Warsaw.2
For all that, a critical literature is still in its infancy. Lemkin remains obscure in
the history of international law. The conferences dedicated to his memory invite
the usual suspects, and mainstream journals of history and international relations
generally eschew his term. To a great extent, ‘genocide studies’ has yet to break out
of its self-imposed isolation.3 Part of the problem is that Lemkin’s revealing
correspondence and invaluable unpublished manuscripts languish in archives in
New York and Cincinnati, despite assurances that they will be made available to the
public.4 This problem is compounded by the priorities of the self-proclaimed
‘pioneers of genocide studies’—those social scientists writing about genocide in
the 1980s and 1990s—who paid Lemkin lip service for ‘discovering’ genocide but
presumed to improve his definition without undertaking the necessary systematic
reconstruction and explication of his ideas. Conceiving of themselves as activist-
scholars on a mission to interdict genocide in the contemporary world, they were
more interested in, say, penning crusading letters to the New York Times than
embarking on the history of ideas.5
2 In his speech dedicating the conference room in 2005, the Polish academic representing the
Foreign Ministry identified Lemkin as a Pole and an American, but did not mention his Jewish
identity: http://poland.usembassy.gov/poland/rotfeld_hall.html; http://www.msz.gov.pl/Address, by,
Profesor,Adam,Daniel,Rotfeld,the,Ministry,of,Foreign,Affairs,in,honor,of,Raphael,Lemkin.,(Warsaw,,
October,18,,2005),2410.html. The plays are Catherine Filloux, Lemkin’s House (New York: Playscripts,
2005), and Robert Skloot, If the Whole Body Dies: Raphael Lemkin and the Treaty against Genocide
(Madison, WI: Parallel Press, 2006).
3 The only critical examinations of his work are Jürgen Zimmerer and Dominik Schaller (eds), The
Origins of Genocide: Raphael Lemkin as a Historian of Mass Violence (London: Routledge, 2009);
Bartolomé Clavero, Genocide or Ethnocide, 1933–2007: How to Mark, Unmake and Remake Law with
Words (Milan: Giuffré Editore, 2008); Ann Curthoys and John Docker, ‘Defining Genocide’, in Dan
Stone (ed.), The Historiography of Genocide (Houndmills: Palgrave MacMillan, 2008), 9–41; Martin
Shaw, What is Genocide? (Cambridge: Polity, 2007); Anson Rabinbach, ‘The Challenge of the
Unprecedented: Raphael Lemkin and the Concept of Genocide’, Simon Dubnow Institute Yearbook 4
(2005), 397–420. The German historian of Polish legal thought, Claudia Kraft, has also written lucidly
about Lemkin; see note 45 below. In terms of international law scholarship, there is no mention of
Lemkin in the much-cited Martti Koskenniemi, The Gentle Civilizer of Nations: The Rise and Fall of
International Law (Cambridge: Cambridge University Press, 2002).
4 Steven L. Jacobs of the University of Alabama has been cataloguing and editing 20,000 pages of
Lemkin’s papers for decades, but hardly any of it has been published. Most of Lemkin’s papers are
contained in three places: the Manuscripts and Archive Division of the New York Public Library
(NYPL), 42nd Street, New York; the American Jewish Historical Society (AHJS), 15 West 16th Street,
New York; and the Jacob Rader Marcus Center of the American Jewish Archives (JRMCAJA), 3101
Clifton Avenue, Cincinnati, Ohio. I have corrected his spelling in the quotations from the unpublished
manuscripts.
5 Samuel Totten and Steven L. Jacobs (eds), Pioneers of Genocide Studies: Confronting Mass Death in
the Century of Genocide (Westport, CT: Greenwood Press, 2002). An important exception is Leo
Kuper, who took Lemkin seriously in his Genocide: Its Political Use in the Twentieth Century (New
Haven: Yale University Press, 1981).
lemkin, culture, and the concept of genocide 21
Accordingly, Lemkin’s ideas are not always well understood, whether by those
who invoke him or by his critics. For some Holocaust specialists, his definition of
genocide is too broad, illegitimately associating the Holocaust with other crimes by
trivializing the former and miscategorizing the latter. For others, paradoxically,
Lemkin’s new word was modelled foursquare on the Holocaust, presuming, inac-
curately, that he must have been referring exclusively to the Nazi extermination of
Jews when he coined it during the Second World War.6 Either way, the—upon
reflection, extraordinary—assumption is that Lemkin did not properly understand
genocide, despite the fact that he invented the term and went to great trouble to
explain its meaning. Instead, most scholars presume to instruct Lemkin, retrospec-
tively, about his concept, although they are in fact proposing a different concept,
usually mass murder. To that end, even his texts have been bowdlerized to make
genocide mean mass killing and/or resemble the Holocaust. Thus a rising star in
the field quoted Lemkin as writing that the essence of genocide was the ‘aim of
annihilating the groups completely’, when Lemkin actually wrote ‘of annihilating the
groups themselves’.7 The mix-up was all the more inexplicable because, on the same
page as that from which this quotation is drawn, Lemkin made clear that total
extermination was not necessary for genocide to occur:
Genocide has two phases: one, destruction of the national pattern of the oppressed group;
the other, the imposition of the national pattern of the oppressor. This imposition, in turn,
may be made upon the oppressed population which is allowed to remain, or upon the
territory alone, after removal of the population and the colonization of the area by the
oppressor’s own nationals.8
For this reason, Lemkin tended to associate ‘destruction’—a word he preferred to

‘extermination’—with what he called ‘crippling’ a group: genocide, he wrote in
1946, is ‘the criminal intent to destroy or cripple permanently a human group’.9
The lesson to be drawn from this persistent misquotation and misinterpretation
of Lemkin is that his ideas, rather than solely his career, need to be studied
carefully. For the fact is that genocide is a curious anomaly in the post-war regime
6 Yehuda Bauer, ‘The Place of the Holocaust in History’, Holocaust and Genocide Studies 2 (1987),
211–15. According to Anton Weiss-Wendt, Lemkin ‘incorporated a great many of offences in his, rather
inclusive, interpretation of genocide’ (personal communication, 20 November 2007).
7 Scott Straus, ‘Contested Meanings and Conflicting Imperatives: A Conceptual Analysis of
Genocide’, Journal of Genocide Research 3:3 (2001), 360. Emphasis in original. I am not suggesting
such mistranscriptions are consciously committed. I am suggesting that they occur unconsciously
because Lemkin’s (mis)interpreters think that he must have intended genocide to mean total mass
murder of an ethnic group.
8 Raphael Lemkin, Axis Rule in Occupied Europe: Laws of Occupation, Analysis of Government,
Proposals of Redress (Washington, DC: Carnegie Endowment for International Peace, 1944), 79.
Emphasis added.
9 Raphael Lemkin, ‘Genocide as a Crime under International Law’, American Journal of
International Law 41:1 (1947), 147; cf. Caroline Fournet, The Crime of Destruction and the Law of
Genocide: Their Impact on Collective Memory (Aldershot: Ashgate, 2007).
22 a. dirk moses
of international humanitarian law, which is dominated by the discourse of human

rights with its emphasis on individuals, rather than the interwar focus on group
rights manifested in the politically contentious minority rights protection provi-
sions of the League of Nations.10 As we will see below, it embodies the social
ontology of ‘groupism’, because genocide is about the destruction of groups per se,
not individuals per se. We have, then, the uneasy coexistence of rival languages of
humanitarianism, though they are often conflated. To understand the language of
group rights, we need to reconstruct Lemkin’s thinking about genocide by placing
it in various historical contexts.
T WO C O N T E XT UA L O R I G I N S
................................................................................................................
As is well known, the Polish-Jewish jurist Raphael Lemkin (1900–59) invented the
term genocide in 1943 for his book on Nazi imperialism, Axis Rule in Occupied
Europe.11 Its origins, however, go back much further. Three discourses, I suggest,
were formative for the evolution of the concept. One was the social ontology of
‘groupism’ prevalent in the Eastern European context in which Lemkin was raised.
The second was the Western legal tradition of international law critical of conquest,
exploitative occupations, and aggressive wars that target civilians.
‘Groupism’
Lemkin was a proponent of what the sociologist Rogers Brubaker calls ‘groupism’:
‘the tendency to treat ethnic groups, nations, and races as substantial entities to
which interests and agency can be attributed’, that is, to regard them as ‘internally
homogeneous, external bounded groups, even unitary collective actors with com-
mon purposes’.12 Others might say that he was a ‘primordialist’ who reified groups
as ‘given entities that are held constant throughout the analysis’.13 This commit-
ment baffles American liberals who can see in Lemkin’s national cosmopolitanism
10 Mark Mazower, ‘The Strange Triumph of Human Rights, 1933–1950’, Historical Journal 47:2
(2004), 379–98.
11 Lemkin, Axis Rule. He coined the term in 1943, but the book was delayed for a year by contractual
negotiations with the publisher.
12 Rogers Brubaker, ‘Ethnicity without Groups’, in Andreas Wimmer et al (eds), Facing Ethnic
Conflicts: Towards a New Realism (Lanham, MD, 2004), 35.
13 Lars Cederman, ‘Nationalism and Ethnicity’, in Walte Carlnaes, Thomas Risse, and Beth A.
Simmons (eds), Handbook of International Relations (London: Sage, 2002), 412.
only an anachronistic return to ‘medieval organic imagery’ or fundamental confu-

sion.14 Closer inspection reveals a coherent worldview.
What is the source of this worldview? As a boy, Lemkin reports, he had been
first awakened to the persecution of human cultural groups by the story of the
attempted extermination of Christians by the Roman emperor Nero. By learning
about the travails of ethnic groups through the centuries—the Huguenots of
France, Catholics in Japan, Muslims in Spain—he concluded that ethnic destruc-
tion was a universal and enduring problem. The persecution of Jews was part of
this sorry tale, and he was well aware of their suffering; the Jews of his region near
Bialystok had suffered pogroms in 1906. But his sympathies were for people
everywhere.
Why did Lemkin’s sense of solidarity lead him to defend group rights as opposed
to individual or human rights? Growing up in the multinational world of Eastern
Europe, his cultural imaginary was irreducibly particular. Like the Polish romantic
nationalists of the nineteenth century, he shared the national cosmopolitanism of
Herder’s adherence to the individuality principle and Mazzini’s belief in the unique
role of each people in the ‘symphony of nations’:15
The philosophy of the Genocide Convention is based on the formula of the human cosmos.
This cosmos consists of four basic groups: national, racial, religious and ethnic. The groups
are protected not only by reason of human compassion but also to prevent draining the
spiritual resources of mankind.16
Undergirding the protection of group existence against extermination, then, is the

communitarian assumption that nations and nationhood are intrinsically valuable
because, unlike other human collectives such as political parties, they produce
culture, endow individual life with meaning, and comprise the building blocks of
human civilization.
It goes without saying that Lemkin’s upbringing as a religiously conversant Jew
flowed into his thinking. But how exactly? Did common Yiddish phrases form his
social imaginary? ‘May his name and memory be blotted out’ was the standard
saying about an enemy, itself derivative of the Biblical verse, ‘I will utterly blot out
the remembrance of Amalek’ (Exodus 17:14; cf. Deuteronomy 25:19), the Amelek
being the congenital enemy of ancient Jews. The Jewish festivals of Passover and
Purim commemorate escapes from slavery and genocide, respectively; during the
latter the name of the Persian king, Haman, a descendant of the Amelek, is met
14 Steven Holmes, ‘Looking Away’, London Review of Books, 14 November 2002, and Michael
Ignatieff, ‘The Danger of a World without Enemies: Lemkin’s Word’, The New Republic, 21 February
2001.
15 Andrzej Walicki, Philosophy and Romantic Nationalism: The Case of Poland (Oxford: Clarendon
Press, 1982).
16 Raphael Lemkin, ‘Description of the Project’, NYPL, Reel 3, Box 2, Folder 1.
24 a. dirk moses
with booing and other noise in order to ‘blot’ it out. We can only speculate exactly
how these rituals impacted on Lemkin, but this background cannot be ignored in
accounting for his worldview. The survival of Jews over the millennia, the mainte-
nance of their traditions, their cultural flourishing in the lands of the former
Polish-Lithuanian Commonwealth, where the vast majority of world Jewry lived
and, equally, the intense consciousness that peoples and their memories could be
entirely erased—these were the cultural milieu and drama in which Lemkin was
steeped.
Interestingly, though, he was never a Zionist. Lemkin was drawn to Bundist
notions of cultural autonomy because, like the Bundists, who were especially
strong in Poland, he believed in multiethnic states with minority protection rather
than monocultural states tied to specific plots of land that oppressed minorities. If
he was attracted to Herder’s romantic notion of cultural individuality, he was also
wary of integral nationalism. Lemkin was likely influenced by Karl Renner, the
non-Jewish Austro-Marxist, whom Lemkin wrote an effusive letter of praise as an
inspiration for his ideas. Bundism drew heavily on Renner’s thinking.17
Lemkin’s was an ecumenical cosmopolitanism. Being a Polish patriot and
advocate for all cultures never entailed renouncing his Jewish heritage or cultural
rooting. His Jewish identity was not structured like a zero sum game. He always
mentioned the genocidal persecution of the Jews by the Nazis in the same breath as
the mass murder of Polish Christians, Roma, and other victims. Central was his
attachment to the notion of ‘spiritual nationality’, a concept that most likely can be
traced to Jewish sources as well as to Herder. Here are possible connections with the
‘autonomism’ of Russian-Jewish historian Simon Dubnow, who wrote of Jewish
nationality that as ‘a spiritual or historical-cultural nation, deprived of any possi-
bility of aspiring to political triumph, of seizing territory by force or of subjecting
other nations to cultural domination, it is concerned only with one thing: protect-
ing its national individuality and safeguarding its autonomous development in all
states everywhere in the Diaspora.’18 Lemkin met the great historian during his
flight from Poland; unlike Lemkin, he did not escape the Nazis.
Why was culture so central to Lemkin’s conception of genocide? After the war,
Lemkin drew on the anthropology of Sir James Frazer and Bronislaw Malinowski to
flesh out his thinking. Malinowksi represented a special affinity. A fellow Pole, his
brand of functionalist anthropology, so revolutionary and influential in Britain,
was actually repackaging what Ernst Gellner calls ‘East European populist
17 Cooper thinks Lemkin was a Zionist, but the evidence he presents suggests he was a Bundist:
Cooper, Raphael Lemkin and the Struggle for the Genocide Convention, 93.
18 Simon Dubnow, Nationalism and History: Essays on Old and New Judaism, ed. and intro Koppel
S. Pinson (Philadelphia: Jewish Publication Society of America, 1958), 97; Simon Rabinovitch, ‘The
Dawn of a New Diaspora: Simon Dubnow’s Autonomism, from St. Petersburg to Berlin’, Leo Baeck
Institute Yearbook 50 (2005), 267–88.
ethnography invented in the service of nationalism, which had practised “going to

the people” more as a moral and political, rather than methodological, principle.’19
Malinoswki’s theory of culture allowed Lemkin to cast his Eastern European
primordialist intuitions in the language of modern social science.
From Frazer and Malinoswki, he took the proposition that culture derived from
the precultural needs of a biological life. He called it ‘derived needs’ or ‘cultural
imperatives’, but it was as constitutive for human group life as individual physical
well being (i.e., ‘basic needs’). Culture integrated society and enabled the fulfilment
of individual basic needs because it constituted the systematic totality of a variety
of interrelated institutions, practices, and beliefs. Culture ensured an internal
equilibrium and stability. These ‘so-called derived needs’, Lemkin wrote, ‘are just
as necessary to their existence as the basic physiological needs.’ He elaborated the
point thus:
These needs find expression in social institutions or, to use an anthropological term, the
culture ethos. If the culture of a group is violently undermined, the group itself disintegrates
and its members must either become absorbed in other cultures which is a wasteful and painful
process or succumb to personal disorganization and, perhaps, physical destruction.20
Consequently, he concluded, ‘the destruction of cultural symbols is genocide.’ To

destroy their function ‘menaces the existence of the social group which exists by
virtue of its common culture.’ This is pure Malinowski.21
Because culture incarnated the identity of peoples, Lemkin was a supporter of
the national minority treaties of the League of Nations. Minorities should not
be forcibly assimilated. As we will now see, the question of culture was also
central to how Lemkin related to the second influential context, colonialism and
imperialism.
Conquest, Occupation, and Cultural Change

The genocide concept is also the culmination of a long tradition of European legal
and political critique of imperialism and warfare against civilians. All of the
instances about which he wrote for his projected world history of genocide
occurred in imperial contexts or involved warfare against civilian populations.
Most of his case studies from the Eurasian land mass were taken from continental
empires: the Roman Empire, the Mongols, the Ottoman Empire, Charlemagne and
19 Ernst Gellner, Language and Solitude: Wittgenstein, Malinowski and the Habsburg Dilemma
(Cambridge: Cambridge University Press, 1998), 120.
20 Raphael Lemkin, ‘The Concept of Genocide in Anthropology’, NYPL, Box 2, Folder 2.
21 Ibid.; Bronislaw Malinowski, The Scientific Theory of Culture and Other Essays (Chapel Hill:
University of North Carolina Press, 1944), 36, 72–3.
26 a. dirk moses
the spread of German peoples eastwards since the Middle Ages.22 Here is a typical
statement from an article in the Christian Science Monitor in 1948:
The destruction of Carthage, the destruction of the Albigenses and Waldenses, the Crusades,
the march of the Teutonic Knights, the destruction of the Christians under the Ottoman
Empire, the massacres of the Herero in Africa, the extermination of the Armenians, the
slaughter of the Christian Assyrians in Iraq in 1933, the destruction of the Maronites, the
pogroms of Jews in Tsarist Russia and Romania—all these are classical genocide cases.23
Because genocide so often occurred in contexts of conquest and occupation,

Lemkin was naturally drawn to the jurisprudence on this question.
This jurisprudence had a long pedigree. As the historian Andrew Fitzmaurice has
shown, European theologians, philosophers, and lawyers have been debating the
morality of foreign occupation since the Spanish conquest of the Americas in the
sixteenth century. These Spanish intellectuals—above all, Bartolomé de Las Casas
and Francesco de Vitoria—based their case on natural law that invested rights in
Indigenous peoples. Hugo Grotius, Samuel Pufendorf, Emeric de Vattel, and
Christian Wolff continued this line of critique. Nineteenth- and twentieth-century
humanitarians who assailed the mistreatment of ‘native peoples’ by colonial
authorities and settlers stood in this tradition.24
Twentieth-century jurists who defended indigenous rights, like Charles Solomon
and Gaston Jèze, studied Vitoria carefully in making out their views. So did Lemkin,
who likely knew Jèze in the 1920s. But Las Casas was his hero: his ‘name has lived on
through the centuries as one of the most admirable and courageous crusaders for
humanity the world has ever known.’25 Lemkin explicitly appropriated Las Casas’
viewpoint in his study of the ‘Spanish Colonial Genocide’, a chapter in his projected
world history of genocide. He called his book on the Nazi empire Axis Rule in
Occupied Europe in order to place it in the tradition of criticizing brutal conquests.
Genocide for Lemkin, then, was a special form of foreign conquest, occupation, and
often warfare. It was necessarily imperial and colonial in nature. In particular,
genocide aimed to permanently tip the demographic balance in favour of the
occupier. In relation to the Nazi case, he wrote that ‘in this respect genocide is a
new technique of occupation aimed at winning the peace even though the war itself is
lost.’26 Any doubt that the roots of the genocide concept lie in the five-hundred-year
22 E.g. Raphael Lemkin, ‘Charlemagne’, AJHS, P-154, Box 8, Folder 6. On the Mongols: JRMCAJA,
Collection 60, Box 7, Folder 6. On Pan-German interest in colonizing Poland in the nineteenth
century: JRMCAJA, Collection 60, Box 6, Folder 13.
23 Raphael Lemkin, ‘War against Genocide’, Christian Science Monitor, 31 January 1948, 2. On the
relationship between genocide and warfare, see Shaw, What is Genocide?.
24 Andrew Fitzmaurice, ‘Anti-Colonialism in Western Political Thought: The Colonial Origins of
the Concept of Genocide’, in A. Dirk Moses (ed.), Empire, Colony, Genocide: Conquest, Occupation and
Subaltern Resistance in World History (New York: Berghahn Books, 2008), 55–80.
25 Raphael Lemkin, ‘Spain Colonial Genocide’, AJHS, P-154, Box 8, Folder 12.
26 Lemkin, Axis Rule, 81.
tradition of natural law-based critique of imperialism rather than solely in Lemkin’s

reaction to the Armenian genocide or Holocaust can be dispelled by his own words:
The history of genocide provides examples of the awakening of humanitarian feelings
which gradually have been crystallized in formulae of international law. The awakening of
the world conscience is traced to the times when the world community took an affirmative
stand to protect human groups from extinction. Bartolomé de las Casas, Vitoria, and
humanitarian interventions, are all links in one chain leading to the proclamation of
genocide as an international crime by the United Nations.27
Lemkin, like Las Casas, did not oppose colonization or empire as such. He was
typical of liberals in the first half of the twentieth century like J. A. Hobson and
supporters of the League of Nations mandate system. Empire could be supported on
humanitarian grounds if it served the interests of ‘civilization’. After all, imperial-
ism, however brutal at times, had also brought the spread of international law that
Lemkin regarded as the central civilizational instrument to combat genocide.
Malinowski was useful here, too, because he offered a theory of cultural change
that justified liberal imperial rule. Empires, humanely governed, contributed to
human progress through ‘diffusion’, he implied. Diffusion amounted to intercul-
tural exchange and was indentured to a theory of progress. It comprised
gradual changes occur[ing] by means of the continuous and slow adaptation of the culture
to new situations. The new situations arise from physical changes, creative energies within
the culture and the impact of outside influences. Without them the culture becomes static;
if they appear but are not met with adaptation of the whole culture pattern, the culture
becomes less integrated. In either case, it becomes weaker and may disintegrate entirely
when exposed to strong outside influences. The rise and fall of civilizations have been
explained on this general basis.28
Following Malinowski, Lemkin thought that cultural change was induced by exoge-
nous influences, as weaker societies adopt the institutions of more efficient ones or
become absorbed by them because they better fulfil basic needs.29 An empire that
promoted diffusion governed by ‘indirect rule’, Malinowski argued, because it suppo-
sedly enabled the autonomous indigenous acquisition of European institutions.30
27 Raphael Lemkin, ‘Proposal for Introduction to the Study of Genocide’, NYPL, Reel 3, Box 2,
Folder 1.
28 Lemkin, ‘The Concept of Genocide in Anthropology’. He cites Malinowski, A Scientific Theory of
Culture and Other Essays; Arthur Toynbee, A Study of History (London: Oxford University Press, 1947);
Ruth Benedict, Patterns of Culture (London: Routledge and Kegan Paul, 1935); Leo Louis Snyder, Race:
A History of Modern Ethnic Theories (New York: Longmans, Green, 1939); Herbert Seligmann, Race
against Man (New York: G. P. Putnam’s Sons, 1939).
29 Malinowski, A Scientific Theory of Culture and Other Essays, 61: ‘the conveyance of a cultural
reality from one culture to another’ means that ‘new needs are created’ in the subject society.
30 Paul T. Cocks, ‘The King and I: Bronislaw Malinowski, King Sobhuza II of Swaziland and the
Vision of Culture Change in Africa’, History of the Human Sciences 13:4 (2000), 25–47.
28 a. dirk moses
Diffusion was a theory of cultural learning processes that justified liberal imperial rule
by European powers.
How did he square this belief with his opposition to the heavy-handed assimila-
tion of minorities he opposed in the new central European nation-states between
the wars? ‘Diffusion is gradual and relatively spontaneous,’ Lemkin wrote, ‘al-
though it may lead to the eventual disintegration of a weak culture.’31 The question
was one of coercion. The absorption of ‘weaker’ cultures was not genocidal,
although he also thought all cultural disappearances were a tragedy of sorts:
Obviously throughout history we have witnessed decline of nations and races. We will meet
this phenomenon in the future too, but there is an entirely different situation when nations
or races fade away after having exhausted their spiritual and physical energies, and there is a
different contingency when they are murdered on the highway of world history. Dying of
age or disease is a disaster but genocide is a crime.32
Consequently, Lemkin was disturbed by occupations like German colonial rule

in Africa that ultimately culminated in genocide in German South West Africa and
German East Africa between 1904 and 1907. Their culture and members were
assaulted in a concerted attack rather than fading away.33 Plainly, Lemkin was as
concerned with the loss of culture as with the loss of life. Accordingly, he urged the
Nuremberg prosecutors not to confuse mass murder with genocide:
It appears in light of this evidence that the term genocide is a correct one since the
defendants aimed to destroy, cripple, or degrade entire nations, racial and religious groups.
The terms mass-murder or mass-extermination in the light of hitherto produced evidence
seems to be inadequate since they do not convey the racial and national motivation of the
crime. [M]ass-murder or extermination do not convey the elements of selection and do not
indicate the losses in terms of culture represented by the nation’s victims.34
Criticisms that Lemkin’s conception is imprecise or incoherent, and that therefore

mass murder should be the definitional core of genocide, miss his point entirely.35
Understanding Lemkin’s assumptions, however, should not blind us to their pro-
blems. These primordialist assumptions meant that he had difficulties in conceiving
of cultural hybridity and adaptation. The cultural options he envisaged in any
encounter seemed to have been either genocide or total assimilation.36 In keeping
31 Lemkin, ‘The Concept of Genocide in Anthropology’.

32 Raphael Lemkin, ‘The Principle of Diversity of Cultures’, JRMCAJA, Collection 60, Box 7, Folder
7/12. Part 1, Chapter 2, Sec. I. II, II, 3. Emphasis added.
33 Raphael Lemkin, ‘The Germans in Africa’, JRMCAJA, Collection 60, Box 6, Folder 9.
34 Memorandum from Raphael Lemkin to R. Kempner, 5 June 1946. United States Holocaust
Memorial Museum, R. Kempner Papers (RS 71.001).
35 E.g. Stuart D. Stein, ‘Conceptions and Terms: Templates for Analysis of Holocausts and
Genocide’, Journal of Genocide Research 7:2 (2005), 171.
36 A cognitive theory of ethnicity, by contrast, would show how that category is a perspective on the
world rather than a primordial, fixed, entity that engages in zero-sum relations with other ethnicities:
with this view, he tended to regard the encounter between European and Indigene as
grossly asymmetric, thereby playing down both indigenous agency and the often-
tenuous European grip on power, particularly in the initial stages of colonization. In
German South West Africa, for instance, he did not see that the German governor
was initially reliant on local chiefs. In fact, such reliance was most likely the norm,
because collaboration with indigenous elites made imperial rule both cheap and
efficient. In such cases, the imperial overlords cooperated with these elites rather
than trying to Europeanize local culture. In the event, indirect rule, far from being a
benign regime, often disrupted indigenous polities by promoting chiefly authority
at the expense of other social actors or by fetishizing ethnic differences (‘tribes’),
which programmed these societies for genocidal conflict after decolonization.37
Lemkin’s blindness to the question of survival and adaptation was rooted in his
particular concept of culture. Despite his Eastern European preoccupation with
peasant cultures (which Malinoswki had converted into anthropological ‘field-
work’ among the natives), he seems to have equated national culture with high
culture. Consider how he regarded the matter in this quotation:
All our cultural heritage is a product of the contribution of all nations. We can best
understand this when we realize how impoverished our culture would be if the people
doomed by Germany such as the Jews had not been permitted to create the Bible or give
birth to an Einstein, a Spinoza; if the Poles had not had the opportunity to give the world a
Copernicus, a Chopin, a Curie; the Greeks a Plato and a Socrates, the English a Shakespeare,
the Russians a Tolstoy and a Shostakovich, the Americans an Emerson and a Jefferson, the
Frenchmen a Renan and a Rodin.38
In this statement, the value of culture inhered in its elites who made contributions
valuable for humanity as a whole. Genocide could occur when they were extermi-
nated, and when libraries, houses of religious worship, and other elite institutions
of cultural transmission were destroyed, even if the mass of the population
survived and continued some hybrid popular culture. Here is what Lemkin wrote
about the Maya in twentieth-century Mexico, centuries after their ravaging at the
hands of the Spanish:
While the condition of the Indians has been improving since then, under a more progres-
sive Mexican administration, their lot is still hard and their cultural heritage has been
irrevocably lost. One million Indians still speak Maya dialect today. They still till the land
as their forefathers had done but they have lost their civilized habits, their remarkable skills
and knowledge long ago.39
Rogers Brubaker, Mara Loveman, and Peter Stamatov, ‘Ethnicity as Cognition’, Theory and Society 33:1
(2004), 31–64; Henry E. Hale, ‘Explaining Ethnicity’, Comparative Political Studies 37:4 (2004), 458–85.
37 Mahmood Mamdani, ‘Historicizing Power and Responses to Power: Indirect Rule and Its
Reform’, Social Research 66:3 (1999), 859–86.
38 Memorandum from Lemkin to Kempner, 5 June 1946. See fn. 34.
39 Raphael Lemkin, ‘Yucatan’, AJHS, P-154, Box 8, Folder 12. Emphasis added.
30 a. dirk moses
Clearly, this view is untenable today. Only white perceptions that ‘real’ Indians must
be ‘pure’ prevented Europeans seeing that ‘Indianness’ was retained even while
Indians adapted their culture and intermarried with others. Lemkin does not seem
to have considered the possibility that genocide could be attempted, that much
destruction could take place, and that cultural adaptation occurred nonetheless.
Formulating Genocide
Before he embarked on his world history of genocide after the Second World War,
Lemkin was a lecturer in comparative law at the Free University of Poland and the
Deputy Prosecutor of the District Court of Warsaw. In the late 1920s, he had
become involved in the Polish Commission for International Juridical Coopera-
tion, whose leading member, Emil S. Rappaport, was proposing that the League of
Nations criminalize aggressive wars. Through this senior colleague, Lemkin was
influenced by the proposal to make a certain class of crimes delicta juris gentium—
offences against the law of nations, meaning grave threats to public international
order that could be prosecuted anywhere under the principle of universal jurisdic-
tion. At the first international conference for the unification of international law in
Warsaw in 1927, these proposed crimes were piracy, counterfeiting of coins, pro-
ducing a public danger, trade in women and children, trade in narcotics, and traffic
in obscene publications. In subsequent years, the question of terrorism in relation
to endangering public order was debated, and Lemkin was included in a commis-
sion to consider the matter and report at the Madrid meeting in 1933. He did not
think the terrorism was a distinct crime, but rather comprised various criminal acts
that individually constituted a public danger. To that list, he proposed to add
barbarity, acts of vandalism, interrupting international communication, and pro-
pagating contagions.40 He also wanted to expand the remit of the law from mere
‘public danger’, which ‘threatens personally indeterminate individuals or an inde-
terminate quantity of the goods on a given territory’. In its stead, he suggested a
‘general (transnational) danger [that] threatens the interests of several States and
their inhabitants.’41 In future decades, he advocated the genocide concept in these
terms, namely that its elements were already crimes but that, taken together, it
constituted a transnational danger.
‘Barbarity’ and ‘Vandalism’ are of relevance for genocide because of their focus
on group protection. He had been indignant that the Turkish perpetrators of the
Armenian deportations and massacres were able largely to escape prosecution, and
40 Marc Segesser and Myriam Gessler, ‘Raphael Lemkin and the International Debate on the
Punishment of War Crimes, (1919–1948)’, Journal of Genocide Research 7:4 (2005), 456–7.
41 Raphael Lemkin, ‘Acts Constituting a General (Transnational) Danger Considered as Offences
against the Law of Nations’, 1933: http://www.preventgenocide.org/lemkin/madrid1933-english.htm
appalled by the massacres of the Assyrian Christians in Iraq. Now it was time to
criminalize them. ‘Barbarity’ combined acts against individuals and collectivities
and thereby exceeded the concept of human rights:
In particular, these are attacks carried out against an individual as a member of a collectivity.
The goal of the author [of the crime] is not only to harm an individual, but also to
cause damage to the collectivity to which the later belongs. Offenses of this type bring
harm not only to human rights, but also and most especially they undermine the
fundamental basis of the social order.42
Such acts comprised ‘massacres, pogroms, actions undertaken to ruin the econom-
ic existence of the members of a collectivity, etc.’ He added other acts that linked
the individual to the group, namely ‘all sorts of brutalities which attack the dignity
of the individual in cases where these acts of humiliation have their source in a
campaign of extermination directed against the collectivity in which the victim is a
member.’ Individually, they violated the criminal codes of civilized nations, but
taken together they endangered ‘the entire social order’ and therefore ‘shake the
very basis of harmony in social relations between particular collectivities’. For this
reason, they were a transnational danger.43
This reasoning was also deployed for the other suggested crime of ‘Acts of
Vandalism’. It too was an ‘attack targeting a collectivity’ in ‘the form of systematic
and organized destruction of the art and cultural heritage in which the unique
genius and achievement of a collectivity are revealed in fields of science, arts and
literature’. He made the suffering of a particular people a transational danger by the
postulate that ‘The contribution of any particular collectivity to world culture as a
whole forms the wealth of all of humanity,’ such that vandalism was tantamount to
an assault on ‘world culture’. Revealing his perspective on civilizational progress, he
noted that vandalism ‘throws the evolution of ideas back to the bleak period of the
Middle Ages’ and ‘shock[s] the conscience of all humanity’.44 Remarkably early in
his career, then, Lemkin highlighted the importance of culture to group life, but
always in relation to a cosmopolitan vision of world civilization.
Ultimately, his report was not even considered at the Madrid meeting, which was
preoccupied with terrorism, and his proposals were quickly forgotten.45 They
would become relevant ten years later when he adapted them in his famous
book, Axis Rule in Occupied Europe. Why did he write it? When in exile in the
United States as an academic and government advisor after 1941, he spread the
word among his colleagues and superiors about the Nazis’ exterminatory inten-
tions toward European Jewry. Receiving a scant hearing, he resolved to publish the
42 Ibid.
43 Ibid.
44 Ibid.
45 Claudia Kraft, ‘Völkermord als delictum iuris gentium: Raphael Lemkins Vorarbeiten für eine
Genozidkonvention’, Simon Dubnow Institute Yearbook 4 (2005), 79–98.
32 a. dirk moses
records of the German occupation he had been collecting, and devise a term for
what Winston Churchill, soon after the German invasion of the Soviet Union, had
called ‘a crime without a name’, namely, the ‘extermination’ of ‘whole districts’.
Like Lemkin, Churchill compared the Nazis ‘to the Mongol invasions of Europe in
the sixteenth century’. Nazism was a reversion to barbaric warfare.46 Contrary to
customary opinion, then, neither Lemkin nor Churchill were referring solely to the
Holocaust of European Jewry; they meant the totality of the German campaign.
Completed in 1943 but published in November 1944, Axis Rule is a massive, 674-
page book in which he first used and explained the meaning of genocide. What
precisely he meant, however, has been a subject of some controversy. Is mass killing
intrinsic to genocide? Indeed it is, many have asserted, and the Holocaust is
prototypical of genocide.47 Or is genocide a much broader term not conceptually
indentured to the Holocaust, as others insist?48 To elucidate Lemkin’s intentions,
we must consider this text as well as articles he wrote soon thereafter.
It is important to note that Lemkin devotes only one of twenty-six chapters in
Parts One and Two of Axis Rule to genocide. Part Three, which comprises more
than half the book, reproduces the German occupation decrees across Europe. The
nine chapters of Part One are each devoted to a technique of occupation: adminis-
tration, police, law, courts, property, finance, labour, legal status of the Jews, and
genocide. This structure suggests that the book is not an analysis of genocide per se,
but a study of German occupation in which genocide is a particular tool of
conquest. Indeed, he writes, ‘genocide is a new technique of occupation aimed at
winning the peace even though the war itself is lost’.49 Yet in the preface, he implies
that all of the techniques were aspects of genocide, such that it forms the concep-
tual core of his book:
The picture of coordinated German techniques of occupation must lead to the conclusion
that the German occupant has embarked upon a gigantic scheme to change, in favor of
Germany, the balance of biological forces between it and the captive nations for many years
to come. The objective of this scheme is to destroy or to cripple the subjugated people in
their development so that, even in the case of Germany’s military defeat, it will be in a
position to deal with other European nations from the vantage point of numerical, physical,
and economic superiority.50
A sentence later, however, he seems to restrict genocide to extermination,

thereby distinguishing it from other techniques.
46 Winston Churchill, Winston S. Churchill: His Complete Speeches, 1897–1963, ed. R. R. James, vol. 6
(New York/London: R. R. Bowker, 1974), 6474.
47 Steven T. Katz, Holocaust in Historical Context (Oxford: Oxford University Press, 1994), 129; Yves
Ternon, ‘Reflections on Genocide’, in Gerard Chaliand (ed.), Minority Peoples in the Age of Nation-
States (London: Pluto Press, 1989), 127.
48 Ward Churchill, A Little Matter of Genocide (San Francisco: City Lights Books, 1997), 67–75.
50 Ibid. xi. Emphasis added.
The practice of extermination of nations and ethnic groups as carried out by the invaders is
called by the author ‘genocide,’ a term deriving from the Greek word genos (tribe, race) and
the Latin cide (by way of analogy, see homocide [sic], fratricide) and is treated in a chapter
under the same name (Chapter IX).
So does genocide mean exterminating or ‘crippling’ a people? He begins Chapter

Nine by declaring that genocide is ‘the destruction of a nation or of an ethnic
group’. But what does destruction mean? We know that he did not think it is
consubstantial with the total disappearance of a people as a biological entity.
Destruction can mean crippling, an interpretation supported by the references
scattered throughout the book to non-murderous genocidal policies directed
towards other peoples occupied by the Nazis.51
Plainly, he combined his original formulations, barbarity and vandalism, to
form a new, more comprehensive concept. Vandalism—the destruction of cultural
works—was now a technique of group destruction.52 But is genocide a synonym
for the forced assimilation of the conquered people? Apparently not. Terms like
‘denationalization’ or ‘Germanization’—the imposition of the conqueror’s ‘nation-
al pattern’ on the conquered people—were unsatisfactory, he continued, because
‘they do not convey the common elements of one generic notion and they treat
mainly the cultural, economic, and social aspects of genocide, leaving out the
biological aspects, such as causing the physical decline and even destruction of the
population involved.’53 Was he hopelessly confused?
We need to recall Lemkin’s conception of nationhood. Nations comprise various
dimensions: political, social, cultural, linguistic, religious, economic, and physical/
biological. Genocide is a ‘coordinated plan of different actions’ that attacks them
‘with the aim of annihilating the groups themselves’. Annihilation cannot be
reduced to mass killing, however. ‘Generally speaking, genocide does not entail
the immediate destruction of a nation, except when accomplished by mass killings
of all members of a nation.’54 And yet an essential aspect of nationhood is the
physical/biological one. He thought the term ‘Germanization’ of the Poles inade-
quate, for example, because
it means that the Poles, as human beings, are preserved and that only the national pattern of
the Germans is imposed upon them. Such a term is much too restricted to apply to a
process in which the population is attacked, in a physical sense, and is removed and
supplanted by populations of the oppressor nations.55
We do not seem closer to a clear answer.
51 Ibid. 138–9, 196, 236–7.

52 He referred explicitly to his 1933 proposals in ibid. 91.
53 Ibid. 80. Emphasis added.
54 Ibid. 79.
55 Ibid. 80. Emphasis added.
34 a. dirk moses
Careful inspection of his writings reveals that, true to his concept of group life, he
did not consider cultural destruction in isolation from attacks on the physical and
biological elements of a group. Culture was inextricably interwoven with a broader
assault encompassing the totality of group existence: ‘Physical and biological
genocide are always preceded by cultural genocide or by an attack on the symbols
of the group or by violent interference with religious or cultural activities. In order
to deal effectively with the crime of Genocide one must intervene at the very
inception of the crime.’56 Nazi mass murder, for instance, could not be separated
from their attack on culture. ‘Side by side with the extermination of “undesirables”
went a systematic looting of artworks, books, the closing of universities and other
places of learning, the destruction of national monuments.’57
In Lemkin’s conception of it genocide affected all aspects of group life. ‘Like all
social phenomena,’ he wrote later, ‘it represents a complex synthesis of a diversity of
factors.’58 It was, therefore, ‘an organic concept of multiple influences and con-
sequences’.59 As a total social practice, genocide comprised various techniques of
group destruction. In Axis Rule, he outlined eight techniques used by the Nazis.
They warrant listing in full because they illustrate his holistic conception of geno-
cide, and demonstrate that mass killing was only one of a number of methods of
group destruction. They are discussed here briefly in the order given by Lemkin.60
Political techniques refer to the cessation of self-government and local rule, and
their replacement by that of the occupier. ‘Every reminder of former national
character was obliterated.’
Social techniques entail attacking the intelligentsia, ‘because this group largely
provides the national leadership and organizes resistance against Nazification.’ The
point of such attacks is to ‘weaken the national, spiritual resources’.
Cultural techniques ban the use of native language in education, and inculcate
youth with propaganda.
Economic techniques shift economic resources from the occupied to the occupier.
Peoples the Germans regarded as of ‘related blood’, like those of Luxembourg and
Alsace-Lorraine, were given incentives to recognize this kinship. There were also
disincentives: ‘If they do not take advantage of this “opportunity” their properties
are taken from them and given to others who are eager to promote Germanism.’
56 Raphael Lemkin, ‘Memorandum on the Genocide Convention’, AHJS, P-154, Box 6, Folder 5.
Because attacks on cultural symbols were embedded in a general attack ‘where cultural genocide
appears to be merely a step towards physical extermination, there will certainly be no difficulty in
distinguishing it from diffusion’: Lemkin, ‘The Concept of Genocide in Anthropology’.
57 Raphael Lemkin, Raphael Lemkin’s Thoughts on Nazi Genocide: Not Guilty?, ed. Steven L. Jacobs
(Lewiston, NY, 1990), 299, 303.
58 Raphael Lemkin, ‘The Concept of Genocide in Sociology’, JRMCAJA, Collection 60, Box 6,
Folder 13, 1.
59 Raphael Lemkin ‘Description of the Project’, NYPL, Reel 3, Box 2, Folder 1.
60 This discussion of the eight techniques is taken from Lemkin, Axis Rule, 82–90.
Biological techniques decrease the birth rate of occupied people. ‘Thus in

incorporated Poland marriages between Poles are forbidden without special
permission of the Governor . . . of the district; the latter, as a matter of principle,
does not permit marriages between Poles.’
Physical techniques mean the rationing of food, endangering of health, and mass
killing in order to accomplish the ‘physical debilitation and even annihilation of
national groups in occupied countries’.
Religious techniques try to disrupt the national and religious influences of the
occupied people. In Luxembourg, the method entailed enrolling children in ‘pro-
Nazi youth organizations’ so as to loosen the grip of Roman Catholic culture.
Alternatively, in Poland, where no such assimilation was possible, the Germans
conducted ‘the systematic pillage and destruction of church property and persecu-
tion of the clergy,’ in order to ‘destroy the religious leadership of the Polish nation’.
Moral techniques are policies ‘to weaken the spiritual resistance of the national
group’. This technique of moral debasement entails diverting the ‘mental energy of
the group’ from ‘moral and national thinking’ to ‘base instincts’. The aim is that
‘the desire for cheap individual pleasure be substituted for the desire for collective
feelings and ideals based upon a higher morality.’ Lemkin mentioned the encour-
agement of pornography and alcoholism in Poland as an example.61
What was the place of the persecution of Jews in this schema? This is an
important question, because some scholars contend that when Lemkin wrote his
book he ‘did not yet fully comprehend the total planned annihilation of the Jewish
people in Europe’.62 Consequently, they maintain, Lemkin conflated the fate of
Jews, whose total physical extermination the Nazis intended, with that of other
nationalities, who were subject to violent denationalization. The latter is genocide
but must be distinguished from the Jewish experience, which is a Holocaust.
Lemkin’s text reveals, however, that he was acutely conscious of the Nazis’ radical
plans for Jews. He devoted a specific chapter to Jews, outlining the ‘special status’
the occupiers created for them in every country they conquered. Nor was he
unaware of the extermination camps: ‘The Jewish population in the occupied
countries is undergoing a process of liquidation (1) by debilitation and starvation;
and (2) by massacres in the ghettos.’ ‘The Jews for the most part are liquidated
within the ghettos, or in special trains in which they are transported to a so-called
61 In a remarkable coincidence, Malinowski’s adumbration of culture approximates almost exactly

to Lemkin’s facets of national life in his Axis Rule, published in the same year, 1944. ‘From the dynamic
point of view . . . as regards the type of activity, culture can be analyzed into a number of aspects such
as education, social control, economics, systems of knowledge, belief, and morality, and also modes of
creative and artistic expression’ (The Scientific Theory of Culture and Other Essays, 150).
62 Yehuda Bauer, ‘The Place of the Holocaust in Contemporary History’, in Jonathan Frankel (ed.),
Studies in Contemporary Jewry, vol. 1 (Bloomington: University of Indiana Press, 1984), 204–5; Katz,
Holocaust in Historical Context, 129–30 n 15.
36 a. dirk moses
“unknown” destination.’ He was, in other words, well aware that the Jews were ‘to
be destroyed completely’.63
And yet, he included their experience in his ‘one generic notion’ of genocide.
Why did he not distinguish the Jewish case from that of other victims of the
Germans? Because he thought the various techniques of genocide issued in the
same catastrophic end: the destruction of nationhood or group culture, one way or
the other. Even if the Poles were not totally exterminated, Polish culture would be,
and that fact represented as grave a loss to humanity as the loss of Jewish culture.
That is what Lemkin meant by genocide.
F RO M THE N U R E M B E RG T R I A L S
TO T H E GENOCIDE CONVENTION
................................................................................................................
The moral shock of Nazi policies led to celebrated developments in international
law relevant to genocide, but the Nuremberg Trials were a diversion rather than a
stepping stone. In 1945, the Americans favoured prosecuting war crimes and
‘crimes against humanity’, which included ‘murder, extermination, enslavement,
deportation, and other inhumane acts committed against any civilian population’.
Thanks to the relentless lobbying of Lemkin, the indictment of the International
Military Tribunal included ‘deliberate and systematic genocide, viz., the extermi-
nation of racial and national groups . . . particularly Jews, Poles, and Gypsies.’ The
British prosecutor, Sir Hartley Shawcross, added, seemingly following Lemkin, that
‘Genocide was not restricted to extermination of the Jewish people or of the
gypsies. It was applied in different forms to Yugoslavia, to the non-German
inhabitants of Alsace-Lorraine, to the people of the Low Countries and of Nor-
way.’64 But, as Donald Bloxham has shown, the basic orientation of the Allies,
particularly the British, was to play down as much as possible the racially specific
dimension of Nazi crimes.65 The priority was prosecuting the German leaders for
waging an aggressive war (‘crimes against peace’); persecutions of their own
population were only salient insofar as they were connected to waging war.66
Genocide was deployed as rhetorical flourish, and consequently none of the

64 William A. Schabas, Genocide in International Law: The Crime of Crimes (Cambridge:
Cambridge University Press, 2000), 36–8.
65 Donald Bloxham, Genocide on Trial: War Crimes Trials and the Formation of Holocaust History
and Memory (Oxford: Oxford University Press, 2001).
66 Yves Beigbeder, Judging War Criminals: The Politics of International Justice (Houndmills: Palgrave
MacMillan, 1999), 48.
Nazis was convicted of genocide, let alone for pre-war crimes, as a dismayed
Lemkin noted.67 For this reason, the United Nations, which was meeting in its
first session when the first Nuremberg judgment was issued on 30 September/
1 October 1946, decided that its definition of genocide should cover crimes com-
mitted during peacetime, and rejected the British suggestion that genocide be
based on the ‘Nuremberg Principles’.68
In December 1946, the General Assembly of the UN adopted a resolution
affirming genocide as a crime denying ‘the right of existence of entire human
groups’ that issued in ‘great losses to humanity in the form of cultural and other
contributions’.69 This is pure Lemkin, and it comes as no surprise that he was a
tireless lobbyist of UN delegates, many of whom had contact with him and his
ideas. His reputation stood high. Britain’s representative reminded all that, had his
proposals been accepted in Madrid in 1933, the Nuremberg prosecutors would have
been in a legally stronger position. Significantly, the term genocide was preferred to
extermination so as to ensure that national destruction was not limited to mass
killing.70 He was then appointed as an expert to help formulate a draft convention
(the ‘Secretariat’s Draft’ of 1947). It defined genocide very broadly as acts com-
mitted with the ‘purpose of destroying [a human group] in whole or in part, or of
preventing its preservation or development’.71 Lemkin is recorded as supporting
the inclusion of this phase against objections that it was not an essential compo-
nent of genocide.72 He wrote: ‘Cultural Genocide is the most important part of the
Convention.’73 The term ‘cultural genocide’ was also included in the subsequent Ad
Hoc Committee’s draft genocide convention.74
Lemkin has been fundamentally misunderstood by scholars of genocide who
contend that he did not support the concept of cultural genocide.75 But he was also
a pragmatist and was prepared to compromise. In a letter to the New York Times
as early as November 1946, he saw that cultural genocide would encounter
strong objections from many UN delegates, for whom only mass murder ‘shocked
the conscience of mankind’, as the General Assembly resolution on genocide put
it a month later. Although he insisted that human groups ‘can be destroyed
through . . . disintegration of its spiritual resources’, he added that,
67 Power, ‘A Problem from Hell’, 49–50; Bloxham, Genocide on Trial, 67.

68 Schabas, Genocide in International Law, 42.
69 Lippman, ‘The Drafting of the 1928 Convention’, 7.
70 Power, ‘A Problem from Hell’, 51–4.
71 Neremiah Robinson, The Genocide Convention: A Commentary (New York: Institute of Jewish
Affairs, 1960), 123.
72 Schabas, Genocide in International Law, 179–80.
73 Lemkin, ‘Memorandum on the Genocide Convention’.
74 Matthew Lippman, ‘The Drafting of the 1948 Convention on the Prevention and Punishment of
the Crime of Genocide’, Boston University International Law Journal 3:1 (1985), 31.
75 Representative of this misunderstanding is Helen Fein, Genocide: A Sociological Perspective
(London: Sage, 1993), 9–11.
38 a. dirk moses
for the purposes of international legislation the definition must be limited to more basic
elements, such as killing, mayhem, and biological devices, as, for example, sterilization. One
should also limit oneself to such acts which are serious enough to be of international
concern. Only acts undertaken habitually and systematically and deriving from an
organized plan or conspiracy should be included.76
Legal assimilation was not cultural genocide, then, a conclusion that

advantaged states which sought to assimilate their indigenous populations and
other minorities after World War II. Lemkin’s residual faith in Western civiliza-
tion as the source of international humanitarian law may also have encouraged
this narrower reading of cultural genocide. But in the end, even this restriction of
cultural genocide’s meaning was unsatisfactory for most UN delegates, who
understood the Secretariat’s Draft convention as equating the closing of libraries
with mass murder. Many delegates were convinced by the Danish complaint that
it showed ‘a lack of logic and of a sense of proportion to include in the same
convention both mass murder in gas chambers and the closing of libraries.’77
Cultural genocide was eventually dropped from the final version of the conven-
tion.
Even so, the UN did not embrace mass murder as the primary mode of group
destruction. In fact, it largely retained Article II of the Ad Hoc Committee
Draft, which listed four genocidal techniques: (1) killing members of a group;
(2) impairing their physical integrity; (3) inflicting measures and conditions
aimed at causing their death; and (4) imposing measures intended to prevent
births within the group.78 The final convention includes a fifth element: the
forced transfer of children from one group to another, originally in the subsec-
tion on cultural policies in the Secretariat’s Draft, but now intended to comple-
ment the emphasis on the physical/biological consequences of genocidal
techniques.79 Mass murder, then, is only one of five techniques. Moreover, by
stipulating an intention to destroy a group ‘in whole or in part’, the General
Assembly affirmed Lemkin’s argument that permanently crippling a group was
genocidal. Clearly, what the UN defined as genocide was the first part of Lemkin’s
proposal in 1933, namely, barbarity. It excluded the equivalent of the second part,
vandalism.
76 Raphael Lemkin, Letter to the Editor, New York Times, 8 November 1946.
77 Lippman, ‘The Drafting of the 1948 Convention’, 45.
78 Robinson, The Genocide Convention, 132.
79 Matthew Lippman, ‘Genocide: The Crime of the Century: The Jurisprudence of Death at the
Dawn of the Millennium’, Houston Journal of International Law 23:3 (2001), 477.
C O N C LU S I O N
................................................................................................................
Lemkin thought that the Nazi policies were radically new, but only in the context of
modern civilization. Wars of extermination had marked human society from
antiquity until the religious conflagrations of early modern Europe, after which
the doctrine became normative that war is conducted against states rather than
populations.80 The Nazis, then, were at once an irruption of barbarism into
civilization and ‘the most striking and the most deliberate and thorough’ of
genocidal imperialists. ‘They almost achieved their goal in exterminating the
Jews and Gypsies in Europe.’81 The Jewish experience is both distinctive in its
extremity and part of a broader pattern. Given that forty-nine members of his
family died in the Holocaust, his ecumenical approach to human suffering was at
once astonishing and exemplary.
What of his legacy in relation to culture, so central to his concept of genocide
generally? Briefly surveying the post-war legal regime reveals an ambiguous legacy.
Although indigenous people often regard assimilation and development policies as
genocidal or at least culturally genocidal, we know that they have no legal protec-
tion from the UN Genocide Convention. ‘Cultural genocide’ is of rhetorical effect
only.82
Other legal instruments fill some of the gap. The International Labor Organiza-
tion ‘Convention (No. 169) concerning Indigenous and Tribal Peoples in Indepen-
dent Countries’ protects the individual and collective rights of such people. So does
Article 27 of International Covenant on Civil and Political Rights (1966), which
protects minority groups against assimilation:
In those States in which ethnic, religious or linguistic minorities exist, persons belonging to
such minorities shall not be denied the right, in community with the other members of
their group, to enjoy their own culture, to profess and practise their own religion, or to use
their own language.
And although the reference to ‘ethnocide’ in the ‘United Nations Declaration on

the Rights of Indigenous Peoples’ (2007) was removed, a number of articles in there
still offer protection against those policies that an explicit article on cultural
genocide would cover. Article 7 declares that
1) Indigenous individuals have the rights to life, physical and mental integrity,
liberty and security of person.

81 Raphael Lemkin, ‘Genocide’, American Scholar 15:2 (1946), 227–30.
82 Alexandra Xanthaki, Indigenous Rights and United Nations Standards: Self-Determination,
Culture and Land (Cambridge: Cambridge University Press, 2007), 114.
40 a. dirk moses
2) Indigenous peoples have the collective right to live in freedom, peace and
security as distinct peoples and shall not be subjected to any act of genocide
or any other act of violence, including forcibly removing children of the group
to another group.
Article 8 lists very specific acts that are prohibited, such as forced population
transfer, ethnic discrimination, forced assimilation, and land dispossession.83
And yet, while the UN described the new Declaration as ‘an important standard
for the treatment of indigenous peoples that will undoubtedly be a significant tool
towards eliminating human rights violations against the planet’s 370 million
indigenous people and assisting them in combating discrimination and marginali-
zation,’ it is not a legally binding instrument, and it was objected to by the USA,
New Zealand, Australia, and Canada, while many other countries abstained from
the vote.84 These classical settler societies with significant indigenous minorities
could not accede to the articles on self-determination, control of resources on
traditional land, including the right of veto against the state, the seeming uncertain
definition of ‘indigenous’, the status of indigenous customary law, and the princi-
ple of special provisions for indigenous sections of the population. The United
Kingdom spoke for many when it objected to the ‘groupism’ of the Declaration,
and foregrounded the individualism of human rights.
The United Kingdom fully supported the provisions in the Declaration which recognized
that indigenous individuals were entitled to the full protection of their human rights and
fundamental freedoms in international law, on an equal basis to all other individuals.
Human rights were universal and equal to all. The United Kingdom did not accept that
some groups in society should benefit from human rights that were not available to
others.85
What is more, the United Kingdom affirmed that it ‘had long provided political
and financial support to the socio-economic and political development of indige-
nous peoples around the world’.86 The modern state is by definition a develop-
mental one, and indigenous people can stand in the way of ‘progress’ and
‘modernity’. They may claim that development, at least in some circumstances, is
tantamount to cultural genocide and the state will deny it.
83 United Nations Declaration on the Rights of Indigenous Peoples, Adopted by the General
Assembly 13 September 2007: http://www.un.org/esa/socdev/unpfii/en/declaration.html
84 United Nations Permanent Forum on Indigenous People, Frequently Asked Questions, United
Nations Declaration on the Rights of Indigenous Peoples: http://www.un.org/esa/socdev/unpfii/
documents/dec_faq.pdf
85 United Nations General Assembly, GA/10612, press release on the Declaration: http://www.un.
org/News/Press/docs/2007/ga10612.doc.htm
86 Ibid.
It is unlikely that these legal instruments and declarations will protect indige-
nous people against development. Since the ‘gold standard’ of humanitarian
emergencies is genocide, anything that does not resemble it falls beneath the
radar of international public opinion. What is more, the general commitment to
‘development’, which appears in this declaration as well—of course, with the
qualification that indigenous people should codetermine it—is hardly likely to
‘shock the conscience of mankind’ as legally defined genocide does or is supposed
to. Lemkin might well consider indigenous people as weaker cultures who might be
‘absorbed’ by ‘cultural diffusion’. He was not opposed to the spread of Western
civilization; in fact, he saw the field of international law that he championed as the
antidote to genocide, which he coded as barbarism. If his language now seems
archaic, though, the tenacity of the indigenous identity claims and assertions shows
that the language of individual human rights does not suffice for many people(s)
since 1948.
SELECT BIBLIOGRAPHY
Clavero, Bartolomé, Genocide or Ethnocide, 1933–2007: How to Mark, Unmake and Remake
Law with Words (Milan: Giuffré Editore, 2008).
Cooper, John, Raphael Lemkin and the Struggle for the Genocide Convention (Houndmills:
Palgrave MacMillan, 2008).
Korey, William, An Epitaph for Raphael Lemkin (New York: Blaustein Institute for the
Advancement of Human Rights, 2002).
Kuper, Leo, Genocide: Its Political Use in the Twentieth Century (New Haven: Yale University
Press, 1981).
Lemkin, Raphael, Axis Rule in Occupied Europe: Laws of Occupation, Analysis of Govern-
ment, Proposals of Redress (Washington, DC: Carnegie Endowment for International
Peace, 1944).
—— ‘Key Writings of Raphael Lemkin on Genocide’, ed. James T. Fussell, http://www.
preventgenocide.org/lemkin/index.htm
Power, Samantha, ‘A Problem from Hell’: America and the Age of Genocide (New York: Basic
Books, 2002).
Schabas, William A., Genocide in International Law: The Crime of Crimes (Cambridge:
Shaw, Martin, What is Genocide? (Cambridge: Polity, 2007).
Zimmerer, Jürgen, and Dominik Schaller (eds), The Origins of Genocide: Raphael Lemkin as
a Historian of Mass Violence (London: Routledge, 2009).
chapter 2
.............................................................................................
‘ETHNIC
C L E A N S I N G’
VERSUS
GENOCIDE?
.............................................................................................
benjamin lieberman
Since the 1990s, ethnic cleansing has become one of the most widely known forms
of violence directed against groups. Ethnic cleansing is related to genocide, but
ethnic cleansing is focused more closely than genocide on geography and on forced
removal of ethnic or related groups from particular areas. The greatest overlap
between ethnic cleansing and genocide takes place when forced removal of popu-
lation leads to a group’s destruction. This relationship between genocide and ethnic
cleansing cannot be delineated in legal terms because there is no international
convention that defines ethnic cleansing. Given the absence of any such legal
definition, a historical approach is useful to outline the key elements of ethnic
cleansing. Antecedents of ethnic cleansing can be traced back to the ancient world
and to the Middle Ages. Ethnic cleansing emerged along multiple paths, both
through policies to control border zones in large land empires and through
practices employed by European settlers to push aside indigenous peoples. In the
nineteenth and twentieth centuries, the rise of modern ethnic cleansing took place
in several zones, including the old imperial lands of Central and Eastern Europe
and Western Asia. Multiple waves of ethnic cleansing remade the ethnic and
religious map of this region: the first ending shortly after the First World War,
‘ethnic cleansing’ versus genocide? 43
the second culminating and ending during and just after the Second World
War, and the third at the end of the Cold War. This history of ethnic cleansing
reveals the repeated use of coercion to forcibly remove particular groups from
specified areas, but also several paradoxes. Ethnic cleansing is often a policy carried
out by strong states to mould the population map, especially of border zones, but
the breakup of such states also generates power struggles that can lead to ethnic
cleansing. Thus, Norman Naimark refers to ethnic cleansing ‘as a product of the
most “advanced” stage in the development of the modern state’, while Michael
Mann argues new democracies are even more likely than ‘stabile authoritarian
regime’ to carry out ethnic cleansing.1 In yet another paradox ethnic cleansing
often derives at least in part from ethnic and religious conflicts or through
manipulation of such conflicts, but it is also sometimes advocated, at least in
modified form, as a remedy for such conflict.
ORIGINS AND DEFINITION OF THE

TERM ETHNIC CLEANSING
................................................................................................................
In the early 1990s, ‘ethnic cleansing’ entered the lexicon of terms closely associated
with genocide. Language referring to the idea of clearing away groups had been used
in previous conflicts, but the particular term ethnic cleansing only gained wide-
spread attention during the wars for the former Yugoslavia. Though now widely
condemned, the term ‘ethnic cleansing’ may actually have been coined by suppor-
ters of violent attacks designed to drive Bosnian Muslims out of mixed communities
in the spring of 1992. It has also been suggested that Serb sources adopted a term
previously employed by Serbs to describe Kosovar Albanian pressure against Serbs
in Kosovo in the 1980s.2 Almost from the start, the use of the term ‘ethnic cleansing’
caused controversy on the grounds that ethnic cleansing could function as a
euphemism to cover up violence or to render it more harmless. But despite its
provenance and potential for misinterpretation the term ethnic cleansing soon
gained widespread recognition as a major form of violence directed against groups.
1 Norman Naimark, Fires of Hatred: Ethnic Cleansing in Twentieth Century Europe (Cambridge,
MA: Harvard University Press, 2001), 4; and Michael Mann, The Dark Side of Democracy: Explaining
Ethnic Cleansing (Cambridge: Cambridge University Press, 2005), 4.
2 Klejda Mulaj, ‘Ethnic Cleansing in the Former Yugoslavia in the 1990s: A Euphemism for
Genocide?’, in Steven Béla Várdy and T. Hunt Tooley (eds), Ethnic Cleansing in Twentieth-Century
Europe (Boulder, CO: Social Science Monographs, 2003), 695; Drazen Petrovic ‘Ethnic Cleansing: An
Attempt at Methodology’, European Journal of International Law 5 (1994), 342–59; Naimark, Fires of
Hatred, 3.
44 benjamin lieberman
In October 1992, the UN Security Council requested that the Secretary General
appoint a Commission of Experts to investigate reports of expulsion, deportation,
and violence against individuals and property in the former Yugoslavia and
especially in Bosnia Herzegovina. The Commission in a first interim report in
February 1993 determined: ‘Considered in the context of the conflicts in the former
Yugoslavia, “ethnic cleansing” means rendering an area ethnically homogenous by
using force or intimidation to remove persons of given groups from the area.
“Ethnic cleansing” is contrary to international law.’ Methods for carrying out
ethnic cleansing included ‘murder, torture, arbitrary arrest and detention, extra-
judicial executions, rape and sexual assaults’ as well as deportation and military
assaults against civilians. In its final report in May 1994, the Commission described
ethnic cleansing as ‘a purposeful policy designed by one ethnic or religious group
to remove by violent and terror-inspiring means the civilian population of another
ethnic or religious group from certain geographic areas.’ Subsequent definitions
adopted similar language, referring to coerced and forced removal of a targeted
group from a specific territory.3
Like genocide, ethnic cleansing focuses on applying force to a group, but the
definition of ethnic cleansing is more closely related to geography. Ethnic cleansing
refers to removal of a group from a particular area. It is a means for forced
remaking of human landscape. Definitions of ethnic cleansing do not specify the
type of area from which a targeted group is to be removed, but in practice ethnic
cleansing often targets groups living in border areas with mixed populations. In the
very case that gave rise to the term ethnic cleansing, Serb forces attacked ethnic and
religious groups living next to Serbs along the borders of a prospective greater
Serbia. It is in such regions that those responsible for ethnic cleansing are most
likely to see their own group’s power as contested or even threatened.
The term can refer to the forced removal not only of ethnic groups but also of
similar related groups. Ethnicity typically denotes a group with an identity rooted
in common culture or history, but the term may also refer broadly to a group seen
as possessing a different and distinct identity from others. Ethnicity also overlaps
with other forms of identity, most notably religion; however, the precise combina-
tion of factors that defines the identity of groups targeted for removal is less
important than the relationship between perpetrators and victims. Typically,
perpetrators identify those they force out as an inherently threatening group.
Ethnic cleansing is closely related to forced migration and population transfers.
Forced migration is a broader term that includes ethnic cleansing, but that also
covers population displacement from other causes, such as public works projects.
3 Andrew Bell-Fialkoff, Ethnic Cleansing (New York: St. Martin’s Griffin, 1996), 3; Naimark, Fires of
Hatred, 3–5; Cathie Carmichael, Ethnic Cleansing in the Balkans: Nationalism and the Destruction of
Tradition (London: Routledge, 2002), 1–2; Mann, Dark Side of Democracy; and Benjamin Lieberman,
Terrible Fate: Ethnic Cleansing in the Making of Modern Europe (Chicago: Ivan R. Dee, 2006), xiii.
In organized population transfers, states agree on the transfer or exchange of

particular groups across international boundaries. Rules arranged for transfer
would in principle make population transfers less violent than other forms of
expulsion. However, there are two reasons to question any strong distinction
between population transfers and ethnic cleansing. First, transfers may fall short
of rules for orderly migration. Second, population transfers even if conducted with
concern for the security of civilians still rely on coercion. Entire groups are forced
to leave simply on the basis of their identity. The most carefully organized popula-
tion transfer would therefore amount to a form of ethnic cleansing.
ETHNIC CLEANSING, GENOCIDE,

A N D I N T E R NAT I O NA L L AW
................................................................................................................
Explaining the relationship between ethnic cleansing and genocide has caused
controversy. Ethnic cleansing shares with genocide the goal of achieving purity
but the two can differ in their ultimate aims: ethnic cleansing seeks the forced
removal of an undesired group or groups where genocide pursues the group’s
‘destruction’. Ethnic cleansing and genocide therefore fall along a spectrum of
violence against groups with genocide lying on the far end of the spectrum. The
two are distinct where ethnic cleansing does not lead to destruction of groups, but
extreme forms of ethnic cleansing overlap with genocide when the means
employed to carry out ethnic cleansing lead to genocide. When high mortality
through deportation or expulsion is predictable, intended, and expected it makes
sense to refer to genocide rather than to ethnic cleansing or to both ethnic
cleansing and genocide. The overlap between genocide and ethnic cleansing is
greatest when genocide is conceived of as Raphael Lemkin originally described the
term. Lemkin focused not only on actual mass killing, but also on what he
described as ‘destruction of the national pattern’, and forced removal of a popula-
tion from a given area can achieve that end.4
Ethnic cleansing and genocide are especially difficult to distinguish in the case of
acts of genocide that take place within a campaign of ethnic cleansing. The
Convention on the Prevention and Punishment of Genocide of 1948 lists acts of
genocide that include ‘killing members of the group’. Genocide most commonly
describes campaigns of mass extermination that target entire groups, but the term
4 Raphael Lemkin, Axis Rule in Occupied Europe: Laws of Occupation—Analysis of Government -

Proposals for Redress (Washington, DC: Carnegie Endowment for International Peace, 1944), 79–80;
and see the special issue of the Journal of Genocide Research 7 (December 2005), 441–559.
has also been used to refer to massacres that target all members of a particular
group in a given area. The International Criminal Tribunal for the former Yugo-
slavia (ICTY) has employed the term genocide in this manner. In August 2001, the
Trial Chamber in the case of Radislav Krstic determined that a genocide against
Bosnian Muslims took place in July 1995 during the Srebrenica massacre in which
women and children were deported and as many as eight thousand Bosnian
Muslim men and boys were murdered. It found General Krstic guilty of genocide
as well as other crimes, though an appeals chamber later determined that Krstic was
not a direct perpetrator.5
As for genocide, debate over identifying ethnic cleansing often focuses on intent.
Refugee movements or the flow of internally displaced people suggests the possi-
bility of ethnic cleansing, but applying the term ethnic cleansing requires
a judgment about organization and planning to encourage flight. Expulsion of
civilians in wartime is covered by crimes of war, but leaving a war zone does not
necessarily amount to ethnic cleansing if civilians flee because they fear insecurity
and the risk of suffering grave harm. Their movement suggests ethnic cleansing,
however, if they depart because they have good reason to fear that they will suffer
attack because of their identity. In practice, ethnic cleansing often takes place by
example: the sight of smoke rising from a burning village prompts flight from
surrounding communities.
The basic elements of ethnic cleansing are widely understood, but ethnic cleans-
ing, in contrast to genocide, has never been codified in international law. Instead
ethnic cleansing is seen as a form of previously defined crimes. The Commission of
Experts identified practices employed in ethnic cleansing as ‘crimes against
humanity’ that ‘can be assimilated to specific war crimes’ and added ‘that such
acts could also fall within the meaning of the Genocide Convention’. Warfare for the
former Yugoslavia therefore brought heightened scrutiny of ethnic cleansing with-
out action to anchor the term ethnic cleansing directly into international law. The
effect of this disjunction can be seen in the work of the ICTY. The ICTY is the legal
body most responsible for punishing crimes associated with ethnic cleansing, but its
proceedings have most often mentioned ethnic cleansing with the purpose of
providing background to a case or evidence of another related crime. In court
proceedings, ethnic cleansing has often appeared as a term within quotation marks.
Indeed, an attorney for a defendant at the ICTY sought to use the absence of an
international legal definition for ethnic cleansing as grounds to challenge the use of
the term. Specifically, he objected to the prosecutor’s use of the term, stating ‘It does
not exist in [the] Genocide Convention or in the international customary law.’6
5 International Criminal Tribunal for the Former Yugoslavia, Case Number IT-98–33, http://www.
un.org/icty/krstic/TrialC1/judgement/
6 International Criminal Tribunal for the Former Yugoslavia, Case number IT-97–24-PT, http://
www.un.org/icty/transe24Kovacevic/980227it.htm
H I S TO RY OF ETHNIC CLEANSING: ANTECEDENTS

................................................................................................................
In the absence of a definition of ethnic cleansing in international law, a historical
approach can outline the development and key characteristics of ethnic cleansing.
Like genocide, ethnic cleansing can be applied to events that occurred before any
definition of the term. Genocide took place before Raphael Lemkin coined the term
or the United Nations placed it into international law, and ethnic cleansing predated
the 1990s. Indeed, antecedents of ethnic cleansing can be traced as far back as the
ancient world. Deportations of civilians took place in Egypt, Mesopotamia, and the
Hittite Empire. In the first millennium BCE, the Neo-Assyrian Empire regularly
carried out mass deportations of people from conquered regions. Inscriptions tell,
for example, of removing more than 27,000 people from Samaria to Assyria, and
Sennacherib (705–681) was said to have deported 208,000 people from Babylon.
Assyrian texts may have exaggerated the numbers forcibly removed, but Assyrian
records recount hundreds of cases of deportations throughout the ancient Near East
during and after the reign of Tiglath-pileser III (745–727 BCE). Deportation was also
a major motif in Assyrian art. Motives for such deportations did not necessarily
focus on obtaining ethnic purity, but included punishing rebels, acquiring labour
and soldiers, resettling strategic areas, and urbanizing new and old cities.7
Antecedents of ethnic cleansing took several forms. During the Middle Ages and
into the early modern era Europe experienced multiple cases of religious or
confessional cleansing. Modern ethnic cleansing often targets victims on the
basis of their religious as well as ethnic or racial identity, but medieval and early
modern confessional cleansing focused still more strongly on religion. Both Eng-
land and France expelled Jews, and Catholic monarchs who conquered the last
Moorish realms of Spain in 1492 drove away Jews, or forced many Jews and
Muslims to discard or hide their religious identity. In 1685, King Louis XIV’s
revocation of toleration for Protestants led French Protestants or Huguenots to
scatter around the globe despite an official ban on their emigration.
PAT H S TOWA R D ETHNIC CLEANSING

................................................................................................................
Antecedents and early forms of ethnic cleansing emerged along different paths in
Eurasia and in regions taken over by European settlers. In Europe and Western Asia
7 Oded Bustenay, Mass Deportation and Deportees in the Neo-Assyrian Empire (Wiesbaden: Ludwig
Reichert Verlag, 1979), 2–4, 7, 16, 19–30, 33–4, 41–59.
controlling border areas served as a major cause for deportation and colonization
programmes. The Ottoman Empire employed a policy called surgun to exile
selected groups. From the Empire’s early years, Ottoman authorities periodically
moved Muslim populations, often semi-nomadic peoples, to bolster the border in
Turkey in Europe. These were primarily strategic initiatives aimed at converting
independent populations into loyal and useful subjects of the Sultan.8 The Otto-
man Empire also used forced migration to repopulate Constantinople after cap-
turing the city in 1453. Sultan Mehmed II moved Jews to Constantinople. Such
programmes differed from modern ethnic cleansing in motive. Ottoman rulers
who moved populations were much less guided by the pursuit of overall purity,
whether religious, ethnic, or racial, than by strategic and economic goals. The
Russian Empire also engaged in resettlement for strategic reasons. Thus, Russia in
the eighteenth century periodically moved Muslim Crimean Tatars away from the
front during times of war.9
In a very different path toward ethnic cleansing, European settlers displaced
non-Western populations. There was no general policy of either killing or expelling
every last non-Western population, but settler populations targeted specific
peoples for expulsion. Numerous processes combined to force out Native Ameri-
can or Indian populations from lands in much of North America: signing treaties
and buying land often functioned as the most effective ways to obtain native
lands.10 There was no single Indian relationship, either friendly or hostile, with
settler populations. Indians traded with newcomers and complex hybrid identities
emerged in border zones along the frontier. But force and the threat of force also
drove Indians from contested regions. The Indian Removals carried out by the
United States during the 1830s, which included the removal of Cherokees along the
Trail of Tears, relied on a mixture of legality and coercion.11 With most Indians
removed from the eastern United States military campaigns pushed Indians out of
areas on the edge of an expanding settler population along much of the Great
Plains, and the United States ultimately pressed surviving populations into reser-
vations.12 This type of ethnic cleansing helped to remake the ethnic map of large
areas of the United States.
18 Yonka Köksal, ‘Coercion and Mediation: Centralization and Sedentarization of Tribes in the
Ottoman Empire’, Middle Eastern Studies 42 (2006), 475.
19 Michael Khodarkovsky, Russia’s Steppe Frontier: The Making of a Colonial Empire, 1500–1800
(Bloomington, IN: Indiana University Press, 2002).
10 Stuart Banner, How the Indians Lost Their Land: Law and Power on the Frontier (Cambridge, MA:
Harvard University Press, 2005).
11 For a discussion of removals in a global context, see Patrick Wolfe, ‘Settler Colonialism and the
Elimination of the Native’, Journal of Genocide Research 8 (2006), 391–2, 396–7, 399–400.
12 Gary Clayton Anderson, The Conquest of Texas: Ethnic Cleansing in the Promised Land (Norman:
University of Oklahoma Press, 2005).
In this form, ethnic cleansing is rooted in settler conquest and colonization.

Treatment of native populations was less organized toward a single purpose than
modern forms of ethnic cleansing in that campaigns to move groups such as
Indians generally lacked the ideological goal of obtaining absolute purity present
in some modern cases of ethnic cleansing. But policies toward natives along
shifting frontiers shared the coercive core of modern ethnic cleansing.
Contests for land and power in border zones most often targeted native popula-
tions but on occasion Europeans also expelled other European settlers. During the
Seven Years War Great Britain drove Acadian, French settlers, from their lands in
Nova Scotia.13
Another form of ethnic cleansing emerged in the nineteenth and twentieth
centuries in Central and Eastern Europe and Western Asia. Large land empires,
Austria-Hungary, Russia, and the Ottoman Empire, dominated most of this region,
and despite traditions of strategic and economic forced migration these empires
experienced limited effects from early forms of cleansing under settler regimes. The
Russian Empire advanced slowly into the north Caucasus in the nineteenth century
and shifted populations in an effort to pacify Chechen areas, and Russia cleared
Circassians out of the northwest Caucasus at a time when Russia saw this region on
the Black Sea coast as susceptible to the influence of foreign powers.14 Russia
engaged in ethnic cleansing of Circassians, but neither Russia nor its fellow empires
pursued overall purity of population.
MODERN ETHNIC CLEANSING:

T H E F I R S T WAV E
................................................................................................................
From the late nineteenth into the twentieth centuries modern ethnic cleansing
remade the ethnic and religious map of much of Central and Eastern Europe and
Western Asia. Three massive waves of ethnic cleansing produced a shift toward
greater national homogeneity. By the First Balkan War of 1912–13 the Ottoman
Empire lost most of its possessions in Europe, and a combination of violent attack
13 John Mack Faragher, A Great and Noble Scheme: The Tragic Story of the Expulsion of the French
Acadians from their American Homeland (New York: Norton, 2005).
14 Peter Holquist, ‘“To Count, to Extract, and to Exterminate”: Population Statistics and
Population Politics in Late Imperial and Soviet Russia’, in Ronald Grigor Suny and Terry Martin
(eds), A State of Nations: Empire and Nation-Making in the Age of Lenin and Stalin (Oxford: Oxford
University Press, 2001), 116–19; and Brian Glyn Williams, ‘Hijra and Forced Migration from
Nineteenth-Century Russia to the Ottoman Empire: A Critical Analysis of the Great Crimean Tatar
Emigration of 1860–1861’, Cahiers du Monde Russe 41 (2000), 93.
and intimidation reduced the Muslim proportion of the population in several

regions, including parts of Bulgaria and Macedonia. This first long wave of violence
culminated during and immediately after the First World War. A second modern
European wave of ethnic cleansing peaked during and immediately after the
Second World War, and a third wave continued to remake the map of southeastern
Europe and of parts of Transcaucasia at the end of the Cold War.
Modern ethnic cleansing shares several common features. Perpetrators repeatedly
claim to be victimized by groups targeted for removal. The language used by
perpetrators to describe victims can often be traced in part to long-term ethnic or
religious tensions, but rapid radicalization often leads to ethnic cleansing. In some
cases such radicalization leads all the way to genocide, but in others ethnic cleansing
serves as the most ambitious method for removing supposedly threatening groups.
During the First World War varied authorities in both the Russian and Ottoman
Empires resorted to forced migration of groups accused of insufficient loyalty or of
outright betrayal. In the Russian Empire, the military command in the west moved
Jews away from the war zone, ordering forced evacuation of up to one-fifth of the
Empire’s Jews, and Russian authorities also moved ethnic Germans away from the
front. In the Ottoman Empire, leaders of the Committee of Union and Progress,
the political party that had seized power just before the war, adopted policies of
deportation, outright ethnic cleansing, and genocide. Many, though not all,
Greeks, were deported from selected coastal regions of the Turkey, and in 1915
Armenians were deported from most of Turkey in a campaign of ethnic cleansing
and genocide.
The persecution of Armenians during the First World War incorporated many of
the chief features of ethnic cleansing. The Ottoman Empire had already experienced
mass violence against Armenians, most notably with the massacres of Armenians
carried out under Sultan Abdul Hamid II that peaked in 1895, but the First World
War still brought unprecedented radicalization in anti-Armenian policy.
In this case ethnic cleansing led to genocide. Across Anatolia Armenians were
ordered out of their homes and for the most part driven south into the deserts of
Syria and Mesopotamia. As in many cases of ethnic cleansing, the campaign began
in border areas, but it soon spread across most of the Ottoman Empire. This was
a coordinated policy carried out against a group identified by both ethnicity
and religion. There were a few exceptions to this policy in that the Armenian
communities of the largest cities of the western empire, Smyrna (Izmir) and
Constantinople (Istanbul), were not destroyed at this time, though Armenians
living in Constantinople without their families and some political figures were
targeted, and more sweeping deportations of Armenians were considered.15 At the
same time Turkish deportations of Armenians also led to genocide. Frequent
15 Donald Bloxham, The Great Game of Genocide: Imperialism, Nationalism, and the Destruction of
the Ottoman Armenians (Oxford: Oxford University Press, 2005), 78, 121.
massacres, especially of Armenian men, repeated assaults along routes southward,

and the predictable lack of food and water in the desert heat caused the extermi-
nation of Armenians. This was ethnic cleansing so severe that it reached the level of
genocide.
Ethnic cleansing continued after the First World War when the Ottoman
Empire’s final collapse generated nationalist power struggles. Greece landed an
army in Turkey in 1919 starting a war that produced ethnic killing, deportations,
and ethnic cleansing. Turkish Nationalists in the east of Anatolia deported
Pontic Greeks, the Greeks who lived along the Black Sea Coast. In 1922, Turkish
Nationalist victory prompted large-scale Greek flight. The Peace Conference at
Lausanne reached agreement in 1923 for a Greek-Turkish population exchange
under which Turks were moved from Greece to Turkey, and most Greeks were
moved from Turkey to Greece, even though some of those moved ill fit the national
categories of Greeks or Turks because the criteria for expulsion referred to religion
rather than ethnic self-identification.16 Killings and arson were common during
the Greek-Turkish war, but the Lausanne population exchange was not genocide.
The new immigrants suffered many hardships, but the intent was a national
exchange. At the same time, the coercive nature of exchange meant that the
Greek-Turkish population exchange amounted to ethnic cleansing.

T H E S E C O N D WAV E
................................................................................................................
The next major wave of modern ethnic cleansing reached a peak during the Second
World War. Hitler spoke in the early days of World War II of creating a ‘New Order’.
He presented this as a way to better mark the boundaries between nationalities, but
Germany almost immediately turned to forced migration to remake the ethnic
map of much of Central and Eastern Europe as a zone under German and ‘Aryan’
domination. German authorities wished, for example, to make newly annexed
areas of western Poland more German and in pursuit of this goal deported
hundreds of thousands of Poles. With war against the Soviet Union, Himmler,
the Head of the SS and of the Reich Commission for the Strengthening of
Germandom (RKFDV), planned for even more ambitious programmes of depor-
tation. Experts employed in drawing up a General Plan East proposed deporting
16 Bruce Clark, Twice a Stranger: The Mass Expulsions that Forged Modern Greece and Turkey
(Cambridge, MA: Harvard University Press, 2006).
millions of Slavs, including Poles, Ukrainians, and White Russians, far towards the
east, but most plans were never carried out because of German defeat.
The Holocaust both incorporated and departed from ethnic cleansing. In terms
of ultimate goals, the Holocaust was a genocide distinct from ethnic cleansing: the
idea of exterminating all European Jews could not be satisfied by merely moving
them. At the same time discussion of ethnic cleansing as one option for ridding
Europe of Jews created one of several routes to genocide. During the early years of
the Second World War, Nazi obsession with finding a final solution to the Jewish
question in Europe, or the fact that Jews existed in Europe, generated a search for
plans, including resettlement. The idea of sending European Jews to the island of
Madagascar attracted attention, and Nazi racial experts also discussed forcibly
moving Jews elsewhere, including to the far reaches of the Soviet Union. Some of
these plans, if realized, would very probably have led to genocide, but the very
failure to move forward with such schemes also propelled Germany’s leaders
toward genocide.17
Once the Holocaust began, mass murder carried out in towns and villages of the
Soviet Union and in death camps in Poland soon superseded any policy of merely
removing Jews from areas under German control, but in at least one region ethnic
cleansing blended into genocide in a pattern reminiscent of the Armenian geno-
cide. Research into the Armenian genocide has prompted a search for connections
to the Holocaust both through Hitler’s much discussed comment on the fate of
Armenians and through research into the legacy of Germany’s wartime alliance
with Turkey, but the closest parallels to the Armenian genocide during the Holo-
caust can be found in areas where Romania initiated ethnic cleansing during the
Second World War: Bessarabia and Northern Bukovina.18 These were border
regions with mixed populations taken from Romania by the Soviet Union in
June 1940. Jews throughout Romania suffered pervasive anti-Semitism during the
war years, and the Soviet takeover further radicalized antagonism against Jews. In
Romania as in other regions along the eastern front, nationalists and government
authorities accused Jews of having aided and abetted in the Soviet occupation. In
June 1941, Romania under the leadership of Marshal Antonescu joined Nazi
Germany in the invasion of the Soviet Union, retook Bessarabia and Northern
Bukovina, and promptly carried out a policy of ethnic cleansing of Jews. Romanian
forces massacred Jews and also drove Jews out of their homes. Indeed,
Romanian officials used language almost identical to ethnic cleansing to refer to
17 Götz Aly, ‘Final Solution’: Nazi Population Policy and the Murder of the European Jews, trans.
Belinda N. Cooper and Allison Brown, (New York: Oxford University Press, 1999).
18 Mark Levene, ‘The Experience of Armenian and Romanian Genocide: 1915–16 and 1941–42’,
in Hans-Lukas Kieser and Dominik J. Schaller (eds), Der Völkermord an den Armeniern und die Shoah
(Zurich: Chronos Verlag, 2002), 423–62.
their goals: they called for a policy of ‘ground cleansing’.19 Romanian forces at first
forced surviving Jews into ghettos, but once Romania received a strip of territory
on the east Bank of the Dniester termed Transnistria, Romanian gendarmes began
to drive Jews east out of Bessarabia and across the Dniester. Some Jews were killed
on route but many also died from starvation and disease during their first winter in
Transnistria. As in the case of the Armenian genocide, extraordinarily violent
deportations led to genocide. Of the 125,000 to 150,000 Jews forced into Transnis-
tria only approximately 50,000 survived the war. Much as in the Armenian geno-
cide, massacres, exposure, starvation, and disease predictably caused large numbers
of deaths, though Jews forced into Transnistria in 1941 suffered from cold where
Armenians at least in 1915 more often suffered from extreme heat.20
In several regions of eastern and southeastern Europe, extremist nationalist
politics led to local wars of ethnic cleansing during the broader world war. In the
Balkans, the Ustasha, a Croatian fascist movement placed in power after the
German invasion of Yugoslavia in 1941, carried out a campaign of killings and
expulsions against Serbs. A bitter war of ethnic cleansing also took place between
Ukrainian and Polish armed bands in mixed Polish-Ukrainian border areas during
the latter years of the Second World War.21
On the other side of the eastern front the Soviet Union also resorted to ethnic
cleansing during the Second World War. The Soviet leadership built on pre-war
practices: during the 1930s the USSR under Stalin deported selected groups,
typically from ethnic groups that lived both within and outside of the borders of
the USSR.22 During the Second World War Soviet authorities swiftly deported
entire populations accused of insufficient loyalty. It took only days to round up
most Chechens and Ingush and Crimean Tatars in 1944. These were extraordinary
displays of power by Soviet security forces, which surrounded villages, collected
inhabitants, and sent them east to Central Asia and Siberia.23
The end and immediate aftermath of the Second World War produced a new
surge of ethnic cleansing. As Germany faced imminent defeat many German
civilians fled from areas including the Baltic and East Prussia rather than wait for
the arrival of the Soviet army. In spring and early summer 1945 Czech and Polish
19 Armin Heinen, ‘Gewalt-Kultur: Rümanien, der Krieg, und die Juden (Juni bis Oktober 1941)’,
in M. Hausleitner, B. Mihok, J. Wetzel (eds), Rumänien und der Holocaust: Zu den Massenverbrechen
in Transnistrien 1941–1944 (Berlin: Metropol Verlag, 2001), 33–52.
20 Raud Ioanid, The Holocaust in Romania: The Destruction of Jews and Gypsies under the
Antonescu Regime, 1940–1944, foreword by Elie Wiesel (Chicago: Ivan R, Dee, 2000).
21 Timothy Snyder, ‘The Causes of Ukrainian–Polish Ethnic Cleansing 1943’, Past and Present
179 (2003), 197–234.
22 Terry Martin, ‘The Origins of Soviet Ethnic Cleansing’, Journal of Modern History 70 (1998), 813–61.
23 Aleksandr M. Nekrich, The Punished Peoples: The Deportation and Fate of Soviet Minorities at
the End of the Second World War, trans. George Saunders (New York: Norton, 1978); and Nikolaı̌
Fedorovich Bougaı̌, The Deportation of Peoples in the Soviet Union (New York: Nova Science, 1996).
forces pushed Germans across the borders of Czechoslovakia and Poland. These
have been termed ‘wild expulsions’ and fully meet the key criteria of ethnic
cleansing. In August 1945, the United States, the United Kingdom, and the Soviet
Union agreed at Potsdam to an organized transfer of ethnic Germans out of
Poland, Czechoslovakia, and Hungary. According to the Potsdam agreement
transfers were supposed to take place ‘in an orderly and humane manner’, but
here again transfer amounted to ethnic cleansing: Germans were forced to leave
many lands or communities that had long been German simply because of their
identity. In all, 12 to 14 million ethnic Germans fled or were expelled or transferred.

T H E T H I R D WAV E
................................................................................................................
During the height of the Cold War the pace of ethnic cleansing slowed but never
stopped entirely. Recurrent violence between Greek and Turkish Cypriots created
distinct Greek and Turkish zones and culminated in 1974 in the island’s division
along ethnic lines after a coup and Turkish invasion. As the Cold War approached
its end, a new wave of ethnic cleansing gained international attention. In the
former Yugoslavia, Bosnian Muslims, Croats, Serbs, and Kosovar Albanians all
suffered ethnic cleansing. As war began in 1991 Serbs gained the upper hand in
disputed mixed border areas of Croatia, and in 1992 Serb forces drove Bosnian
Muslims and Croats out of contested areas of Bosnia-Herzegovina.24 Bosnian
Muslims and Bosnian Croat forces were nominally allies but fought against each
other, producing still more ethnic cleansing in 1993 in western Herzegovina and in
Central Bosnia. Bosnian Serb forces sought to complete the ethnic cleansing of
Bosnia in 1995, attacking enclaves such as Srebrenica, an eastern Bosnian town that
had been designated by the United Nations as a safe haven. But 1995 also saw a
sudden turn of fortune with offensives by Croatian and by Bosnian Muslim forces
and the ethnic cleansing of Serbs from regions including the Krajina border region
of Croatia. Ethnic cleansing again struck the former Yugoslavia in 1999 with
fighting for Kosovo, a region with a large Albanian majority. Serb forces at first
carried out ethnic cleansing of Kosovar Albanians, but with the retreat of Serb
forces in June 1999 Serbs left much of Kosovo.
At much the same time warfare in the Transcaucasus, the region south of the
highest Caucusus mountain chain in the Republics of Georgia, Armenia, and
24 Roy Gutman, A Witness to Genocide: The 1993 Pulitzer Prize-Winning Dispatches on the ‘Ethnic
Cleansing’ of Bosnia (New York: Macmillan, 1993).
Azerbaijan, also generated ethnic cleansing. Fighting over the Black Sea region of
Abkhazia ended with Georgian defeat in 1993 and the ethnic cleansing of Georgians.
In Armenia and Azerbaijan ethnic and religious conflict preceded the final Soviet
breakup, and with the Soviet Union’s end Armenian and Azerbaijani nationalist
aspirations clashed most intensely in Nagorno Karabkah, an enclave with an Arme-
nian majority within the borders of Azerbaijan. War for Nagorno Karabakh ended by
1994 with Armenian victory and the ethnic cleansing of Azerbaijanis.
Claims of victimization played a key role in generating ethnic cleansing in the
former Yugoslavia and in the Transcaucasus. Perpetrators and their defenders
presented ethnic cleansing as a defensive act against groups accused of presenting
a dire threat. Serb forces, for example, emerged from the breakup of Yugoslavia
with the greatest military might, but Serb nationalist ideology accused other
groups of threatening Serb survival in a society where multiple groups had
apparently lived in peace with each other for many years.
Such violence between groups that have long lived in close proximity presents
one of the most puzzling aspects of ethnic cleansing. In mixed societies perpetra-
tors of ethnic cleansing are likely to be familiar with the groups they seek to expel
from a coveted region. Such relationships belie the notion that violence stems from
ancient hatreds. In Yugoslavia, there was evidence of a past in which people of
different religions and ethnic backgrounds lived peacefully alongside each other. At
the same time, claims of past harmony also rested in part on myth. Intellectuals,
the media, and selected politicians sharpened ethnic antagonisms in Yugoslavia
during the 1980s, but they did not conjure up nationalist fears out of nothing.
Marriages and other relationships across ethnic and religious lines were common,
at least in parts of the country, but at the same time nationalist speeches and news
coverage played on intense fear of the very same groups that typically drew on
accounts of violence from the Second World War or from even further back in the
more distant past.
PA R A D OX E S OFE T H N I C C L E A N S I N G : S TAT E S
A N D S TAT E C O L L A P S E
................................................................................................................
This capacity of residents of societies affected by ethnic cleansing to both trust and
fear members of other ethnic and religious groups is one of several paradoxes of
ethnic cleansing. Modern ethnic cleansing, for example, is both a tool with which
states impose order and a result of the collapse of state power. Ethnic cleansing has
functioned as an extreme form of population politics in which states impose
simplified boundaries and ethnic categories on complex societies and pursue
rationality at a high human cost. The very goal of moving groups across boundaries
shows the imprint of states because states customarily draw up or at least maintain
boundaries. Furthermore, the elasticity of ethnic identity points to the role of states
in launching ethnic cleansing. In many cases, people moved through ethnic
cleansing or population transfers only imperfectly fit into the identity of the groups
targeted for removal. The population exchange carried out between Greece and
Turkey, for example, saw the movement of Muslims who spoke varieties of Greek to
Turkey, and conversely the movement of Christians who spoke Turkish to Greece.
Such complex, mixed hybrid identities did not fit into the idealized scheme of
nation-states.25 Similarly in the aftermath of World War II significant populations
of Central and Eastern Europe in areas such as Silesia could not be clearly identified
as possessing a single ethnic or national identity.
As a form of violent social engineering, ethnic cleansing is closely associated with
powerful dictatorships such as the Soviet Union and Nazi Germany. The Anfal
campaign conducted by Saddam Hussein’s Iraq against Kurds in 1989 also fits this
model. Anfal combined ethnic cleansing and genocide in that Saddam’s lieutenants
worked to clear Kurds out of particular areas and also carried out large massacres,
most notoriously with poison gas, that were identified by observers, including
Human Rights Watch, as an act of genocide.26
Powerful states have pursued and carried out ethnic cleansing, but imperial
breakup and collapse has also generated ethnic cleansing. A sequence of imperial
decline and nationalist conflict propelled major waves of modern ethnic cleansing.
Ottoman decline and breakup fed nationalist hopes of redeeming peoples con-
ceived of as having suffered under a ‘Turkish yoke’ and fuelled efforts to construct
nation-states out of mixed Ottoman lands. In the latter stages and at the end of the
Second World War the breakup of a very different kind of empire again fed ethnic
cleansing. The approach of German defeat encouraged ethnic cleansing by Polish
and Ukrainian armed bands, and the final collapse of the Nazi empire led to
expulsion, transfer, and the end of many German communities across Eastern
and Central Europe. Finally, in the 1990s the breakup of powerful states played a
key part in generating ethnic cleansing both in the former Yugoslavia and in parts
of the former Soviet Union.
The breakup of empires can spur power struggles that lead to ethnic cleansing
despite attempts to regulate the division of imperial lands. At a minimum partition
in British India in 1947 and the end of the British Mandate in Palestine in 1948 led
to large-scale forced migration. In South Asia, partition generated intense com-
munal violence and the flight of some 12 to 13 million people. Much of the
25 Renée Hirschon (ed.), Crossing the Aegean: An Appraisal of the 1923 Compulsory Population
Exchange between Greece and Turkey (New York: Berghahn Books, 2003).
26 Human Rights Watch, Iraq’s Crime of Genocide: The Anfal Campaign against the Kurds (New
Haven, 1995).
subcontinent remained relatively peaceful, but ethnic cleansing took place in

particular regions including parts of the Punjab where Sikh armed groups attacked
and expelled Muslims.27 At the same time, Muslim groups attacked Hindus and
Sikhs in West Punjab, though arson attacks and looting may well have proceeded
from mixed motives.28 In the case of Palestine, the British mandate ended in 1948
with the First Arab-Israeli War and the flight and expulsion of at least 700 thousand
Palestinian refugees, though an Arab minority remained in the state of Israel after
its independence. What remains disputed is the question of what produced this
flight and the extent to which Zionist and eventually Israeli forces planned to drive
out Palestinian Arabs. Debate in particular has centred on a plan created by the
Zionist fighting force, the Haganah, named Plan Dalet or Plan D. This plan has
been alternately described as a military operation carried out against Arab settle-
ments behind Jewish lines or as a plan for expulsion.29 The case for a centralized
plan of ethnic cleansing remains subject to debate, but this may also be a case where
Haganah forces discovered that they could carry out ethnic cleansing at the local
and regional level as their offensive drove out large numbers of Arabs.30
PA R A D OX E S O F E T H N I C C L E A N S I N G : E T H N I C
CLEANSING AND ETHNIC CONFLICT
................................................................................................................
The controversies over the partition reveal yet another paradox of ethnic cleansing.
Partition or division of ethnically or religiously mixed states has been identified
both as a cause of ethnic cleansing and as a possible remedy for ethnic cleansing. In
particular, the ‘security dilemmas’ created when a group in a mixed society takes
action that threatens another group’s security provides a case for partition.31
27 Ian Copland, ‘The Master and the Maharajas: The Sikh Princes and the East Punjab Massacres of
1947’, Modern Asian Studies 36 (2002), 657–704; idem, ‘The Farther Shores of Partition: Ethnic
Cleansing in Rajasthan 1947’, Past and Present 160 (August 1998), 203–39; Yasmin Khan, The Great
Partition: The Making of India and Pakistan (New Haven: Yale University Press, 2007), 130–1, 135–6.
28 Paul Brass, ‘The Partition of India and Retributive Genocide in the Punjab, 1946–1947: Means,
Methods, and Purpose’, Journal of Genocide Research 5 (2003), 82–3; and Khan, Great Partition, 159.
29 On the case for intentional ethnic cleansing see Ilan Pappe, The Ethnic Cleansing of Palestine
(Oxford: Oneworld Publications, 2006). For the most influential historian of the subject see Benny
Morris, The Birth of the Palestinian Refugee Problem Revisited, 2nd edn (Cambridge: Cambridge
University Press, 2004).
30 On the Haganah offensives see Meron Benvenisti, Sacred Landscape: The Buried History of the
Holy Land since 1948, trans. Maxine Kaufman-Lacosta (Berkeley, 2000), 117–18.
31 On the debate over partition theory see Nicholas Sambanis, ‘Partition as a Solution to Ethnic
War: An Empirical Critique of the Theoretical Literature’, World Politics 52 (2000), 437–83.
Finding boundaries in regions with mixed religious or ethnic groups has led to
violence, but the search for solutions to persistent ethnic and / or religious conflict
has also yielded the idea of separating hostile groups from one another. Whereas
genocide has generally only been supported by perpetrators and their close allies,
ethnic cleansing and closely related forms of population politics, especially
organized population transfers or exchanges, have also been seen as useful means
for resolving conflict.
Forced migration, including ethnic cleansing and related forms of coerced mi-
gration, emerged during the twentieth century as one of the chief solutions to
persistent ethnic and religious tension and conflict. Such cleansing was not neces-
sarily the first choice, but by the twentieth century it emerged as a widely used
method for restructuring the ethnic and religious map in mixed borderlands in
Europe and Western Asia. Between the First and Second World Wars democracies as
well as dictatorships displayed growing interest in population transfers as a possible
cure to salvage the nation-state. Thus, Czech President Eduard Benes, the last
democratically elected leader in Central and Eastern Europe in the 1930s, considered
population transfers as a way of reducing his country’s German population, though
only after years of Nazi manipulation of the Sudeten question.32 The Great Powers
also supported or at least tolerated forced migration as a means for removing the
potential instability created by the presence of minority populations.33 Democracies
approved of the large population transfers in Central and Eastern Europe at the end
of the Second World War. Meeting at Potsdam in the summer of 1945, Britain and
the United States along with the Soviet Union agreed to the forced migration of
most Germans residing outside the new boundaries of Germany. Churchill de-
scribed himself as resisting the most sweeping demands for transfer of 8 or 9 million
Germans, but still accepted the basic premise of mass population transfer. In 1944,
he notably told the House of Commons, ‘Expulsion is the method which, in so far as
we have been able to see, will be the most satisfactory and lasting. There will be no
mixture of populations to cause endless trouble . . . A clean sweep will be made.’
In 1945, at Postdam he voiced reservations about moving eight or nine million
Germans, but still spoke of accepting ‘transfer’ of 2–3 million Germans.34 By this
time, interest in ethnic cleansing crossed different types of states, political
32 Detlef Brandes, Der Weg zur Vertreibung, 1938–1945: Pläne und Entscheidungen zum ‘Transfer’ der
Deutschen aus der Tschechoslowakie und aus Polen (Munich: R. Oldenbourg, 2001), 5–6.
33 Donald Bloxham, ‘The Great Unweaving: The Removal of Peoples in Europe, 1875–1949’, in
Richard Bessel and Claudia Haake (eds), The Removing of Peoples in the Modern World (Oxford:
Oxford University Press, 2008).
34 Winston Churchill, Triumph and Tragedy (Boston: Houghton Mifflin, 1953), 658. For an account
that stresses Churchill’s doubts, see Alfred M. de Zayas, Nemesis at Potsdam: The Anglo-Americans and
the Expulsion of the Germans: Background, Execution, Consequences (London: Routledge and Kegan
Paul, 1977), 81–7.
systems, and ideologies. In Poland and Czechoslovakia groups with widely varied
political views wanted Germans gone.35
Faced with the similar dilemma of creating nation-states in ethnically and
religiously mixed regions, varied governments and political movements arrived
at similar conclusions on the need to move entire peoples or at least large
percentages of inconveniently placed ethnic or religious groups. Some embraced
cleansing as a positive ideal: undemocratic states carried out the most violent
ethnic cleansing. Ethnic cleansing was also likely to take place in new states and
in the early phases of democratization.36 There was then real disagreement about
means and methods and about whether to move populations as a first or last
option, but not about the usefulness of forcibly moving peoples across state
boundaries.
International opinion began to turn against separating populations just as ethnic
cleansing gained unprecedented attention with the breakup of Yugoslavia in the
1990s. The concept of ‘ethnic partition’ did not vanish in discussion of responses to
nationalist conflict, but Yugoslavia’s breakup brought both ethnic cleansing and
increasing resistance to the very goals of forced migration of particular ethnic or
religious groups.37 International efforts to curb violence often proved inadequate
with tragic consequences as at Srebrenica.38 But the Dayton Peace Accords that
ended hostilities in Bosnia-Herzegovina in 1995 represented a major move against
the principles of ethnic cleansing consensus. It is easy to overlook the importance
of what seems to be one of the most obvious elements of Dayton: the insistence on
a right of return. The Dayton Peace Accords asserted, ‘All refugees and displaced
persons have the right freely to return to their homes of origin.’ This was not always
the norm after previous waves of forced migration. In the case of Israel the United
Nations in 1949 in Resolution 194 called for the return of those refugees who wished
to ‘live at peace with their neighbours’. But with limited exceptions the Greek-
Turkish population exchange of 1923 was compulsory, and in similar fashion, the
Potsdam Protocol of 1945 saw the transfer of Germans as a necessity. Dayton, in
contrast, sought to preserve the principle of a mixed society.
Ethnic cleansing after Bosnia faced more condemnation than ever before, but as
genocide shows mere condemnation does not make mass violence impossible.
35 Sebastian Siebel-Achenbach, Lower Silesia from Nazi Germany to Communist Poland, 1942–1949
(New York: St Martin’s Press, 1994), 33, 50; T. David Curp, A Clean Sweep?: The Politics of Ethnic
Cleansing in Western Poland, 1945–1949 (Rochester: University of Rochester Press, 2006); and Emilia
Hrabovec, Vertreibung and Abschub: Deutsche in Mähren 1945–1947 (Frankfurt am Main: Peter Lang,
1995), 41, 43.
36 Mann, Dark Side of Democracy, 4.
37 On ‘ethnic partition,’ see Jack L. Snyder, From Voting to Violence: Democratization and
Nationalist Conflict (New York: Norton, 2000), 325.
38 David Rohde, Endgame: The Betrayal and Fall of Srebrenica, Europe’s Worst Massacre since World
War II (New York, 1997).
Ethnic cleansing is deeply rooted in key developments of modernity: the rise of

nations, the effort to categorize and define identities, the pursuit of purity, and
both the power of governments and the emergence of mass politics. As such it may
be a crime with both a past and a future.
SELECT BIBLIOGRAPHY
Bell-Fialkoff, Andrew, Ethnic Cleansing (New York: St Martin’s Griffin, 1996).
Carnegie Endowment for International Peace, The Other Balkan Wars: A 1913 Carnegie
Endowment Inquiry in Retrospect, with a new Introduction and Reflections on the Present
Conflict by George F. Kennan (Washington, DC: Carnegie Endowment, 1993).
Carmichael, Cathie, Ethnic Cleansing in the Balkans: Nationalism and the Destruction of
Tradition (London: Routledge, 2002).
Gutman, Roy, A Witness to Genocide: The 1993 Pulitzer Prize-Winning Dispatches on the
‘Ethnic Cleansing’ of Bosnia (New York: Macmillan, 1993).
Lieberman, Benjamin, Terrible Fate: Ethnic Cleansing in the Making of Modern Europe
(Chicago: Ivan R. Dee, 2006).
Mann, Michael, The Dark Side of Democracy: Explaining Ethnic Cleansing (Cambridge:
Naimark, Norman, Fires of Hatred: Ethnic Cleansing in Twentieth-Century Europe (Cam-
bridge, MA: Harvard University Press, 2001).
Ther, Philipp, and Ana Siljak (eds), Redrawing Nations: Ethnic Cleansing in East-Central
Europe, 1944–1948 (Lanham, MD: Rowman and Littlefield, 2001).
Várdy, Steven Béla, and T. Hunt Tooley, (eds), Ethnic Cleansing in Twentieth Century Europe
(Boulder: Social Science Monographs, 2003).
chapter 3
.............................................................................................
GENDER AND
GENOCIDE
.............................................................................................
elisa von joeden-forgey
The subject of gender in genocide is a relatively new research interest and still
remains peripheral to the field of genocide studies as a whole. While most compre-
hensive treatments of genocide do not take gender seriously into consideration,
significant new contributions by scholars such as Adam Jones and R. Charli
Carpenter are changing this state of affairs.1 Because genocide is a crime against
groups, in which individuals are targeted due to their group membership, the
assumption has long been that sex differentiation among the victims is of minor
importance to the process as a whole. The genocides in Bosnia-Herzegovina and in
Rwanda in the early 1990s, however, made clear the importance of gender con-
structs in genocide. Here the widespread and systematic rape of women and sexual
exploitation of men, as well as the obvious use of gendered patterns of attack, were
explicit parts of the perpetrators’ genocidal strategies. Thus began a new moment
in the study of genocide, one that has the potential to offer powerful tools for the
prediction, prevention, and prosecution of genocide.
1 R. Charli Carpenter, Born of War: Protecting Children Sexual Violence Survivors in Conflict Zones
(Bloomfield, CT: Kumarian Press, 2007); ‘Surfacing Children: Limitations of Genocidal Rape
Discourse’, Human Rights Quarterly 22:2 (2000), 428–77; ‘Forced Maternity, Children’s Rights, and
the Genocide Convention’, Journal of Genocide Research 2:2 (2000), 213–44; Adam Jones (ed.),
Gendercide and Genocide (Nashville: Vanderbilt University Press, 2004); Genocide: A Comprehensive
Introduction (London/New York: Routledge, 2006). See also Jones’ Gendercide Watch website: http://
www.gendercide.org
62 elisa von joeden-forgey
In this chapter I set out to demonstrate that a consideration of gender is crucial

to our understanding of the crime, because genocide is an historical process that is,
at its core, about group reproduction. As Helen Fein pointed out in a seminal essay
on the subject, ‘[r]eproduction serves to continue the group; genocide to destroy it.
Thus, perpetrators must either annul reproduction within the group or appro-
priate the progeny in order to destroy the group in the long run.’2 While the
perpetrators’ ultimate aim is the material destruction of the target group, the
means used to achieve this end tend to target men and women according to their
perceived and actual positions within the reproductive process. Genocides are
therefore characterized by highly symbolic and ritualized dramatizations of the
perpetrator’s obsession with demonstrating his or her destructive power over the
target group’s very life force. As part of the killing, then, one finds in all genocides a
shared set of tortures involving generative symbols and institutions (reproductive
organs, infants and small children, and the bonds that promote family coherence).
In many cases, these symbols can be destroyed in ways that do not require the
wholesale physical killing of all members of a group. In fact, it appears that the
‘total’ genocides, such as the Holocaust and Rwanda, are the exceptions; the norm
is rather the sex-selective killing of specific members of a group combined with a
host of strategies aimed at destroying the group’s ability to survive into the future.
I will engage ‘gender’ as both a marker of biological sex and as a set of cultural
practices and beliefs aimed at organizing relations of power between the sexes. Accord-
ingly, I will treat gender as both ‘a constitutive element of social relationships based on
the perceived differences between the sexes’ and ‘a primary way of signifying relation-
ships of power’.3 Considering the experiences of men and women simultaneously can
help us see genocide as a process that combines many different means of destruction in
order permanently to undermine the future of a group. Although direct killing is a
central part of the genocidal process, it is not the whole story. Studying gender in
genocide can help identify frequently overlooked long-term causes of genocide as well
as key problems faced by societies as they seek to rebuild in the wake of genocide.
MAKING GENDER VISIBLE

................................................................................................................
The study of gender in genocide opens up a new realm for the analytical discussion
of genocide—the so-called ‘private sphere’, what Zainab Salbi, the founder of
2 Helen Fein, ‘Genocide and Gender: The Uses of Women and Group Destiny’, Journal of Genocide
Research 1:1 (1999), 43.
3 Joan Wallach Scott, Gender and the Politics of History (New York: Columbia University Press,
1988), 42.
gender and genocide 63
Women for Women International, calls the ‘backline discussion’.4 She points out
that war and peace are usually understood solely according to the largely male
‘frontline discussion’ involving soldiers and politicians. But life is also, and perhaps
primarily, lived in the ‘backline discussion’ of feeding families, raising children, and
nurturing strong community bonds—things which, it must be said, are not
exclusive to women. Genocide, like war, cannot be properly addressed without a
serious and sustained investigation of the ‘backline discussion’ and the ways in
which it influences the lead-up to genocide, the perpetration of genocide, and the
search for justice and social healing after genocide. Indeed, the place of ordinary
life is the primary focus of genocidal violence.
The foundation for an investigation of the ‘backline discussion’ was laid by women
scholars of the Holocaust, who in the 1980s began to research the experiences of
women survivors. Until then Holocaust research had reflected the gendered assump-
tions of historical scholarship wherein the history of men stood in for the history of
humankind. Women’s experiences were considered to be derivative of and ancillary to
men’s, and consequently of little importance to history. It was the testimony of male
survivors that came to comprise the literary and historical canon of the Holocaust,
despite the fact that women were a majority in the Jewish population of Europe before
World War II.5 Indeed, women wrote the majority of memoirs and testimonials in
the first years after 1945,6 despite having a lower survival rate overall.7
The importance of women to a full understanding of the Holocaust was recog-
nized during the genocide itself by the noted Polish-Jewish historian Emmanuel
Ringelblum, who studied Jewish women and mothers in the Warsaw ghetto while
he too was incarcerated there with his family.8 But it was not until feminist scholars
began to investigate the lives of women in the Holocaust in the 1980s that gendered
differences in experience began to be recognized.9 Most scholars now accept that, in
4 ‘A Woman among Warlords: Interview with Zainab Salbi’, Wide Angle, August 31, 2007, Daljit
Dhaliwal, http://www-tc.pbs.org/wnet/wideangle/shows/warlords/interview/interview.pdf
5 Raul Hilberg, Perpetrators, Victims, Bystanders: The Jewish Catastrophe, 1933–1945 (New York:
Harper Perennial, 1993), 127.
6 Judith Tydor Baumel, Double Jeopardy: Gender and the Holocaust (London/Portland, OR:
Vallentine Mitchell, 1998), 41.
7 Hilberg, Perpetrators, 127, 130.
8 Baumel, Double Jeopardy, 40.
9 Lisa Pine, ‘Gender and the Family’, in Dan Stone (ed.), Historiography of the Holocaust (New York:
Palgrave Macmillan, 2004), 364–82; Elizabeth R. Baer and Myrna Goldenberg (eds), Experience and
Expression: Women, the Nazis, and the Holocaust (Detroit: Wayne State University Press, 2003); Atina
Grossmann, ‘Women and the Holocaust: Four Recent Titles’, Holocaust and Genocide Studies 16:1
(Spring 2002), 94–108; Esther Fuchs (ed.), Women and the Holocaust: Narrative and Representation
(Lanham, MD: University Press of America, 1999); Dalia Ofer and Lenore J. Weitzman (eds), Women
in the Holocaust (New Haven, CT: Yale University Press, 1998); John Roth and Carol Rittner (eds),
Different Voices: Women and the Holocaust (New York: Paragon House, 1993); Renate Bridenthal et al.
(eds), When Biology Became Destiny: Women in Weimar and Nazi Germany (New York: Monthly
Review Press, 1984).
Raul Hilberg’s words, ‘the road to annihilation was marked by events that specifi-
cally affected men as men and women as women.’10
The pioneering first two decades of work on women and the Holocaust focused
on the ways that gender affected Jewish experiences under Nazi domination, and
was strongly informed by the cultural feminist framework of the time. Cultural
feminism tended to essentialize gender difference, celebrating women’s ‘special
sphere’ rather than investigating how gender categories and norms intersect with
race, nationality, class to form identity and experience, as more recent treatments
have begun to do. The cultural feminist analysis of the Holocaust was criticized for
appearing to argue that women were more victimized than men because of their
‘double burden’ as Jews and as women (the assumption being that men carried
‘merely’ a single burden). It could also appear to ignore the sufferings of men by
casting them as aggressors. And, in celebrating women’s supposedly unique abil-
ities to find coping mechanisms in times of crisis, an assertion that many women
Holocaust survivors themselves made, cultural feminism at times suggested that
women somehow transcended the horrors of the camp experience.11
Critics of the cultural feminist model worried further that the focus on sexism
was overshadowing the core element of Nazi policy, which was its racist anti-
Semitism. Cynthia Ozick remarked in a letter to Joan Ringelheim, ‘[t]he Holocaust
happened to victims who were not seen as men, women, or children, but as Jews.’12
Other scholars, such as Anna Hardman and Zoë Waxman, protested that cultural
feminism ignored the variety and diversity of women’s lives as well as the moments
of antagonism and division between female victims.13 Pascale Rachel Bos has
argued that many of the gender differences that have been attributed to people’s
actual experiences are in fact differences in the way men and women construct
memory.14 Finally, Lawrence Langer has voiced scepticism about gender as an
important variable in the context of the unspeakable suffering brought about by
the Holocaust.15
What is clear from the research is that gender directly influenced people’s
experience of Nazi persecution at various moments within the overall pattern of
destruction. Gender norms shaped how Jews in Germany and elsewhere responded
to the Nazi threat.16 Gender also shaped the specific nature of people’s
10 Hilberg, Perpetrators, 126.

11 Joan Ringelheim, ‘Women and the Holocaust: A Reconsideration of Research’, in Roth and
Rittner (eds), Different Voices, 387.
12 Quoted in Baer and Goldenberg, Experience and Expression, xxviii.
13 Pine, ‘Gender and the Family’, 372.
14 Pascale Rachel Bos, ‘Women and the Holocaust: Analyzing Gender Difference’, in Baer and
Goldenberg (eds), Experience and Expression, 23–52.
15 Lawrence L. Langer, ‘Gendered Suffering? Women in Holocaust Testimonies’, in Dalia Ofer and
Lenore J. Weitzman (eds), Women in the Holocaust (New Haven, CT: Yale University Press, 1998), 355.
16 Pine, ‘Gender and the Family’; Baumel, Double Jeopardy, 15.
vulnerabilities, which were, in significant respects, different for Jewish men and
women. For example, rape and sexual exploitation are major themes in the
memoirs written by women survivors but are barely mentioned by men. Myrna
Goldenberg has well documented how women and girls faced the threat of sexual
exploitation at every step of the process of destruction.17 Although the Nazis
imposed strict laws against ‘race mixing’, Jewish women were sometimes raped
by German soldiers and SS men. They were raped in the camps by guards and
sometimes by other inmates as well.18 Girls who were placed into hiding also faced
possible exploitation from their care givers.19 The historian Nechama Tec found
cases of Jewish women partisans who were sexually exploited and also raped by
their comrades.20 After liberation, women faced the additional threat of being
raped by Soviet soldiers.21 The very real threat of sexual exploitation that Jewish
women faced from a variety of men alters the dominant image of the Holocaust as
a ‘closed’ historical event by demonstrating the multiple trajectories of violence
that coalesce in genocide and later feed back into post-genocide societies.
Apart from defining key differences in the way that men and women experienced
persecution, feminist study of the Holocaust has shown that in both ideology and
practice National Socialism was an expression of misogyny as well as racism.22
When the Nazis targeted Jewish women, they often did so in specific ways based on
women’s deep symbolic association with life-giving powers, a theme that runs
through the testimonials written by women survivors. Killing women, especially
pregnant women, was a microcosm of genocide for Nazi murderers, since it
allowed them to attack directly Jews’ spiritual and biological future. Thus,
women were killed at much higher rates than men upon arrival in the death
camps.23 In these camps pregnancy was treated with particular cruelty in accord
with its potent symbolism. The SS appear to have reserved special tortures for
pregnant women who were—in the case of Auschwitz—beaten ‘with clubs and
whips, torn by dogs, dragged around by the hair and kicked in the stomach with
heavy German boots. Then, when they collapsed, they were thrown into the
17 Myrna Goldenberg, ‘Lessons Learned from Gentle Heroism: Women’s Holocaust Narratives’,
Annals of the American Academy of Political and Social Science 548 (November 1996), 78–93.
18 Olga Lengyel, Five Chimneys: A Woman Survivor’s True Story of Auschwitz (Chicago: Academy
Chicago, 1995), 199.
19 Joan Ringelheim, ‘Genocide and Gender: A Split Memory’, in Ronit Lentin (ed.), Gender and
Catastrophe (London/New York: Zed Books, 1997), 26–8.
20 Nechama Tec, ‘The Fate of Women’, in Defiance: The Bielski Partisans (New York/London:
Oxford University Press, 1994), 154–69.
21 Isabella Leitner, ‘Book Two: Liberation’, Isabella: From Auschwitz to Freedom (New York/London:
Anchor Books, 1994), esp. 89–91, 108–11.
22 The classic studies of Nazi misogyny are Gisela Bock, Zwangssterilisation im Nationalsozialismus:
Studien zur Rassenpolitik und Frauenpolitik (Opladen: Westdeutscher Verlag, 1986), and Claudia
Koonz, Mothers in the Fatherland: Women, the Family, and Nazi Politics (New York: St Martin’s, 1986).
crematory—alive.’24 Women who gave birth in the camps were usually murdered
immediately along with their infants. If the pregnancy and birth escaped the
attention of the guards, other inmates were forced to kill the babies if they wished
to save the mother.25 Pregnant women caught up in Einsatzgruppen actions may
have elicited particularly sadistic treatment from their killers as well.26
In addition to demonstrating the importance of misogyny to genocidal ideology,
the study of gender and genocide also sheds light on the ways that perpetrators
instrumentalize gender in the killing process. Even in what have been called ‘gender
neutral’ genocides, such as the Holocaust and Cambodia, perpetrators treat the sexes
differently, though not in ways that significantly affect overall survival rates.27 In a
careful study of gender and the Holocaust in Veszprém, Hungary, Tim Cole shows
that while Jewish men aged 18 to 48 were much more likely to die as forced labourers
before deportations to the death camps began, by 1944 many of them were able to
avoid deportations, and almost certain death, precisely because of their labour
power.28 Most of the Jews deported were women, children, and the elderly. According
to Raul Hilberg, men in general died much more quickly than women in the early
phases of the Nazi occupation of Eastern Europe. This was largely because gender
norms facilitated the treatment of civilian men as enemy combatants, providing a
cover for their detention, forced labour, and massacre.29 Men died at much higher
rates in the ghettos, in part because of the hard labour they were forced to do.30 Men
may also have felt the responsibility to give up rations to their families or they may
have succumbed more quickly to the ravages of malnourishment and starvation due
to their body’s faster metabolism in comparison with women.31 Men also tended to
‘die first’ in the massacres committed by mobile killing squads in Poland, Russia, and
Serbia because it was easier for soldiers and police reservists to rationalize and justify
the killing of men, whom they identified as security threats.32 Even Heinrich Himm-
ler, who clearly had no qualms about killing Jewish men, needed an additional
rationalization for killing Jewish women and children.33
24 Gisela Perl, I Was a Doctor in Auschwitz (Salem, NH: Ayer, 1984), 80. Quoted from Goldenberg,
‘Lessons Learned’, 86.
25 Ibid. 86.
26 Raul Hilberg, The Destruction of the European Jews (New York: Holmes and Meier, 1985), 146.
27 Fein, ‘Genocide and Gender’, 43–63.
28 Tim Cole, ‘A Gendered Holocaust? The Experiences of “Jewish” Men and Women in Hungary,
1944’, in Randolph L. Braham and Brewster S. Chamberlin (eds), The Holocaust in Hungary: Sixty
Years Later (New York: Columbia University Press, 2006), 54.
29 Hilberg, Perpetrators, 128. He notes that there was a ‘reversal of fortunes’ after the development
of the gas vans and the death camps, which made killing women and children psychologically less
taxing on the killers.
30 Ibid. Hilberg notes that many women were also forced to do hard labour, though perhaps in
smaller numbers. See also Nechama Tec, Resilience and Courage: Women, Men, and the Holocaust
(New Haven, CT: Yale University Press, 2003), 11.
31 This latter possibility was suggested by Bos in ‘Women and the Holocaust’, 34.
33 Roth, ‘Equality, Neutrality, Particularity’, 11.
ADDING MEN
................................................................................................................
The focus on men’s gendered experience during genocide is very new. Up until
recently, gender studies were preoccupied primarily with the lives of women,
leaving men within the original framework of universal subjecthood and thereby
unintentionally reaffirming the assumption of the gender-neutrality of men’s lives.
Among historians and other genocide scholars the initial focus on women can be
explained very simply by the fact that gender studies were embraced by women
scholars far earlier than by men. These scholars were interested in including
women in historical narratives that had largely excluded them. It is therefore
hard to argue, as some have, that men have been discriminatorily excluded from
feminist scholarship on genocide.
The exclusion of male victims from international humanitarian attention is a
different story. Rather than a product of feminism, however, this exclusion is the
result of those patriarchal norms in international affairs that treat the term civilian
as coterminous with ‘women and children’.34 These same patriarchal norms have
ascribed to women a more peaceful nature, a theory that has had a measurable
impact on scholarship on gender and violence. These two related beliefs—that
women are by nature peaceful and that male victims are combatants—has ex-
ercised a direct, though often subtle, influence on the ways that observers measure
atrocities. Attacks on women and children frequently appear to generate greater
outrage than attacks on men, largely because attacks on men can be so easily
explained away with reference to their supposed ‘battle age’. Although public
outrage at atrocities against women and children is usually short-lived (it has
rarely translated itself into gender-sensitive priorities in war crimes tribunals or
gender-sensitive economic development efforts in postgenocidal societies), it is
nevertheless significant inasmuch as it can serve to bring a particular conflict to the
forefront of international media attention. Alternately, génocidaires can use patri-
archal traditions in international law semantically to hide their crimes behind
putative counterinsurgency efforts, as the Government of Sudan has done in
Darfur. The role played by gender constructs in genocide denial strategies is a
subject that has yet to be researched.
Recognition that civilian men are primary targets of genocide is therefore crucial
to any attempt to fashion an early warning system and end the impunity with
which génocidaires have committed mass murder up to the present day. Adam
Jones has shown that a policy of killing men first constitutes a ‘tripwire or
harbinger of fuller-scale root-and-branch genocides’, an insight that should be
34 R. Charli Carpenter, ‘Innocent Women and Children’: Gender, Norms and the Protection of
Civilians (Burlington, VT: Ashgate, 2006).
very useful to an early warning system.35 We have already discussed this pattern in
the Holocaust. Jones notes that this pattern is also evident to varying degrees in the
Armenian genocide, the genocides in Bosnia and Rwanda, in Kosovo, and in East
Timor. In these cases men were killed first to radicalize the killers and habituate
them to attacks on women and children. Sex-selective massacre may also be a
means of ‘decapitating’ the family basis of the religious and social structure, much
as the targeting of intellectuals is aimed at decapitating the institutional basis of the
public life of a group. The massacre of ‘battle-aged’ men can be used to expose and
render vulnerable the rest of the population.
For Jones, these sex-selective massacres, in addition to being harbingers of root-
and-branch genocides, are also instances of ‘gendercide’. Expanding upon the works of
feminist writers, Jones defines gendercide as ‘gender-selective mass killing’ and argues
that gendercide in and of itself is a form of genocide.36 The term gendercide was first
used by Mary Ann Warren, whose 1985 book of the same title examined instances in
which women and girls were the targets.37 Jones’ work on gendercide seeks to bring
to our attention the myriad ways in which men and boys are also vulnerable to
sex-selective violence. His interventions have been a necessary reminder of the extent
to which civilian men have been victims of mass murder throughout history.
There has been substantial debate about whether gendercide is itself a genocidal
process. In Jones’ edited volume Gendercide and Genocide, Stuart Stein and R. Charli
Carpenter argue that gendercide is not genocide because it is not committed with the
intent to destroy in whole or in part all members of a sex, as such.38 The central question
seems to be about the claims one would make about gendercide—is it a specific act
within an ongoing genocide or is it a specific sort of violence unto itself that can be at
times either genocidal in nature or used as a tool for genocide? Whatever one decides
about gendercide as genocide, it seems clear that sex-selective killing can be an early
warning sign of genocide as well as an act punishable by the Genocide Convention.
Two recent international court rulings on the Srebrenica massacre, where in July
1995 over 8,000 Bosnian Muslim men and boys were killed by Bosnian Serb forces
under the command of the indicted Gen. Radko Mladic, support the notion of
gendercide as genocide, though only in the limited sense that the gendercidal
massacre at Srebrenica was embedded in a wider ethnic conflict. The International
35 Adam Jones, ‘Gendercide and Genocide’, in Adam Jones (ed.), Gendercide and Genocide
(Nashville, TN: Vanderbildt University Press, 2004), 23.
36 For an overview of the concept of ‘gendercide’, see Jones, ‘Gendercide and Genocide’, 1–38. See
also his Gendercide Watch website: http://www.gendercide.org.
37 Mary Anne Warren, Gendercide: The Implications of Sex Selection (Totowa, NJ: Rowman and
Allanheld, 1985).
38 Stuart Stein, ‘Geno and Other Cides: A Cautionary Note on Knowledge Accumulation’, in Jones
(ed.), Gendercide and Genocide, 196–229; R. Charli Carpenter, ‘Beyond “Gendercide”: Operationalizing
Gender in Comparative Genocide Studies’, in Jones (ed.), Gendercide and Genocide, 230–56. See also
Jones’ response, ‘Problems of Gendercide: A Response to Stein and Carpenter’, in Gendercide and
Genocide, 257–71.
Criminal Tribunal for the Former Yugoslavia (ICTY) ruled in 2004 that the
Srebrenica massacre was genocide, a ruling upheld by the International Court of
Justice (ICJ) in 2007.39 The presiding judge of the ICTY reasoned that Serb actions
in Srebrenica constituted genocide for the following reasons: there was ample
evidence of intent to destroy the Bosnian Muslim group in Srebrenica, the men
and boys constituted a ‘substantial part’ of the group, and they were ‘emblematic’
of the group as a whole.40 Significant in this finding is the concept of emblematic
victims, which raises the importance of understanding gender constructs when
making determinations of genocide. The judgment noted:
In addition to the numeric size of the targeted portion [of the group], its prominence
within the group can be a useful consideration. If a specific part of the group is emblematic
of the overall group, or is essential to its survival, that may support a finding that the part
qualifies as substantial within the meaning of Article 4 [of the ICTY Statute].41
The court found that in a patriarchal society, the loss of 8,000 men within an
immediate population of 40,000 would seriously hinder the future procreation of
the group, potentially leading to its destruction.42
G E N O C I DA L R A P E
................................................................................................................
When Serbia started its war with Bosnia in 1992 one of the major news stories
coming out of the region was the systematic use of rape by the Serb forces to
enforce a policy of ‘ethnic cleansing’, as it was then routinely called. The journalist
Roy Gutman’s articles in Newsday brought international attention to the mass rape
of women during the war.43 This was not the first time that newspapers focused on
the brutality of rape during wartime, but it was the first time that women around
the world were successful in organizing an international movement to have rape
explicitly recognized and prosecuted as a war crime, a crime against humanity, and
a crime of genocide.44 The massive international effort to bring this about began
39 ICTY, Prosecutor v. Krstić, 19 April 2004, www.un.org/icty/krstic/Appeal/judgement/index.htm;

ICJ, Press Release, 26 February 2007.
40 ICTY, Prosecutor v. Krstić, Parts II.A & II.B. See also ICTY, Press Release, ‘Address by ICTY
President Theodor Meron, at Potocari Memorial Cemetery’, The Hague, 23 June 2004, http://www.
un.org/icty/pressreal/2004/p860-e.htm
41 ICTY, Prosecutor v. Krstić, 8.
42 Ibid. 14.
43 Roy Gutman, A Witness to Genocide (New York: Macmillan, 1993).
44 Louise Chappell, ‘Gender Mainstreaming in International Institutions: Developments at the UN
Ad Hoc Tribunals and the International Criminal Court’, Paper Presented at the Annual Meeting of the
with a ground-breaking article by Catherine MacKinnon, who argued early on in

the war that rape was being used by Serb forces as a tool of genocide.45 Her article
was followed by books by Alexandra Sitglmayer (1994) and Beverly Allen (1996),
each of which called attention to the particularly genocidal role that rape was
playing in the violence.
It is estimated that between 20,000 and 50,000 women and girls were raped
during the wars in the former Yugoslavia between 1991 and 1995. ‘While all sides in
the Bosnian conflict have committed rapes,’ notes Joana Daniel-Wrabetz, ‘Serbian
forces appear to have used rape on the largest scale, principally against Muslim
women.’46 Usually rape was accompanied by various tortures, including branding
with the Serbian cross, burning, slashing, beating, and threats of death against the
women and their family members, especially their children. Rape frequently was
used as a means of murder, but also served a policy of forced maternity to create
more ‘Serbian’ children. Women’s bodies were used to humiliate families and
communities as the perpetrators raped girls in front of their parents or forced
family members to rape each other.
Rape in Rwanda shared with rape in Bosnia these genocidal qualities, though here
it was much more widespread and usually used as a means of murder. The estimates
of the number of women raped reaches to 500,000, few of whom were allowed to
survive.47 While perpetrators used rape in this case primarily as part of a terrifying
and drawn-out ritual of killing, some Tutsi women were also subjected to forced
maternity under the logic that they would bear Hutu children, demonstrating the
multiple and self-contradictory levels on which perpetrators pursue the destruction
of the target group’s reproductive powers.48 Thousands of women survivors were
rendered permanently disabled from the brutality of the rapes, many having been
left incapable of bearing children. Furthermore, many assailants seem to have
International Studies Association, Hilton Hawaiian Village, Honolulu, Hawaii, 5 March 2005; Marsha
Freeman, ‘International Institutions and Gendered Justice’, Journal of International Affairs 52:2 (1999),
513; Kelly Dawn Askin, ‘Prosecuting Wartime Rape and Other Gender-Related Crimes under
International Law: Extraordinary Advances, Enduring Obstacles’, Berkeley Journal of International
Law 21:2 (2003), 317.
45 Catherine A. MacKinnon, ‘Turning Rape into Pornography: Postmodern Genocide’, MS 5 (July/
August 1993), 24–30. MacKinnon’s article was later published in Alexandra Stiglmayer (ed.), Mass
Rape: The War against Women in Bosnia-Herzegovina (Lincoln/London: University of Nebraska Press,
1994), 74–81.
46 Siobhan K. Fisher, ‘Occupation of the Womb: Forced Impregnation as Genocide’, Duke Law
Journal 46:1 (1996), 109; Joana Daniel-Wrabitz, ‘Children Born of War Rape in Bosnia-Herzegovina
and the Convention on the Rights of the Child’, in Carpenter (ed.), Born of War, 23.
47 SURF-Survivor’s Fund, ‘Statistics of the Genocide’, http://www.survivors-fund.org.uk/resources/
history/statistics.php
48 Human Rights Watch, Shattered Lives: Sexual Violence during the Rwandan Genocide and its
Aftermath (New York: HRW, 1996); Jones, ‘Gender and Genocide in Rwanda’, in Jones (ed.),
Gendercide and Genocide, 98–137; Marie Consolee Mukangendo, ‘Caring for Children Born of Rape
in Rwanda’, in Carpenter (ed.), Born of War, 40–52.
knowingly infected raped women with HIV, thereby ensuring their eventual and
untimely deaths even if they were to survive the genocide.49
In the face of heavy media attention on the use of rape in both the Bosnian and
the Rwandan genocides, questions were raised about the best way to characterize
these rapes. The controversy has revolved around the question of whether to
conceptualize ‘genocidal rape’ as a special category of rape. Catherine MacKinnon
sparked this debate when she argued that
rapes in the Serbian war of aggression against Bosnia-Herzegovina and Croatia are to
everyday rape what the Holocaust was to everyday anti-Semitism: both like it and not
like it at all, both continuous with it and a whole new departure, a unique atrocity yet also a
pinnacle moment in something that goes on all the time.50
Several feminists have voiced concern about the high-profile public attention that
has been focused on ‘genocidal rape’. Rhonda Copelon has argued that ‘[t]he elision
of genocide and rape in the focus on “genocidal rape” of Muslim women in Bosnia
is . . . dangerous,’ because ‘to emphasize as unparalleled the horror of genocidal rape is
factually dubious and risks rendering the rape invisible once again.’51 Susan Brown-
miller also prefers not to treat rape in Bosnia as a special category of rape, comment-
ing that ‘Serbian land advances have been accomplished in the age-old manner of
territorial aggression, with looting, pillage, and gratuitous violence that gets lumped
under the rubric of atrocity.’52 In a slightly different vein, former Executive Director of
Human Rights Watch Aryeh Neier criticized the focus on genocidal rape for elevating
the crime of rape in and of itself to genocidal proportions; in his words, unless the
rape is committed with the intent of forcing pregnancy, it is ‘inappropriate to single
out one element, rape, and assert that it, by itself, constituted genocide.’53
The case for rape as a crime of genocide was made most forcefully in the case of
forced pregnancy and maternity. The international law scholar Siobhan Fisher
characterized the Serbian policy of forced maternity as a genocidal ‘occupation
of the womb’.54 Writing on the Armenian genocide, Donald Bloxham has similarly
identified forced marriage and sexual slavery as the ‘colonization of the female
body’.55 Such policies are genocidal because the purpose is to force women to give
birth to children from the perpetrator’s group, thereby preventing them from
49 Mukangendo, ‘Caring for Children Born of Rape’, 45.

50 MacKinnon, ‘Turning Rape into Pornography’, 74.
51 Rhonda Copelon, ‘Surfacing Gender: Reconceptualizing Crimes against Women in Time of War’,
in Stiglmayer (ed.), Mass Rape, 197.
52 Susan Brownmiller, ‘Making Female Bodies the Battlefield’, in Stiglmayer (ed.), Mass Rape, 180.
53 Aryeh Neier, War Crimes: Brutality, Genocide, Terror and the Struggle for Justice (New York:
Times Books, 1998), 186. Quoted in Carpenter, ‘Surfacing Children’, 439.
54 Fisher, ‘Occupation of the Womb’, 124.
55 Donald Bloxham, ‘Internal Colonization, Inter-Imperial Conflict and the Armenian Genocide’,
in A. Dirk Moses (ed.), Empire, Colony, Genocide: Conquest, Occupation, and Subaltern Resistance in
World History (New York: Berghahn Books, 2008), 338.
carrying children from their own group.56 Many rapists in Bosnia and Rwanda
plainly stated this intent while committing the rapes. While such logic confounds
modern genetic understanding, it conforms to the highly patriarchal understand-
ing of reproduction in each society, where fathers determine the ethnic identity of
children. Fisher therefore argues that forced maternity conforms to subsection II
(b), (c), and (d) of the Genocide Convention: ‘Causing serious bodily or mental
harm to members of the group’, ‘Deliberately inflicting on the group conditions of
life calculated to bring about its physical destruction in whole or in part’, and
‘Imposing measures intended to prevent births within the group’.
Decisions by the international tribunals set up for Bosnia and Rwanda, established
in 1993 and 1994, respectively, have upheld much of the scholarly work on genocidal
rape. Fisher’s interpretation of forced maternity was confirmed by both the Interna-
tional Criminal Tribunal for Rwanda (ICTR) in Prosecutor v. Akayesu and the ICTY
in the Karadzic and Mladic decisions.57 In Prosecutor v. Akayesu, the ICTR further
found that rape and sexual violence ‘constitute genocide in the same way as any other
act as long as they were committed with the specific intent to destroy, in whole or in
part, a particular group, targeted as such.’58 While the Statute of the International
Criminal Court (ICC) does not list sexual violence or rape as specific elements of the
crime of genocide, the ICTY and ICTR decisions have set important precedents for
trying gender-based violence as genocide.59
R E L AT I O NA L V I O L E N C E
................................................................................................................
The debate about the status of rape in genocide and whether it itself is genocidal
neglects to consider the wider ‘relational’ context of much gender-based violence.60
Rape in genocide is frequently part of an elaborate and sustained ritual on the part
of perpetrators in which they focus not only on killing, raping, and expelling living
members of a group but also on the intensive targeting of symbols of the group’s
life force.61 Recognizing the wide-ranging targets of genocidal violence, Dirk Moses
56 Fisher, ‘Occupation of the Womb’, 93.

57 Mark Ellis, ‘Breaking the Silence: Rape as an International Crime’, Case Western Reserve Journal
of International Law 38 (2006/2007), 232–35.
58 ICTR, Prosecutor v. Akayesu, Case No. IT-96–4-T, 731.
59 Rape is specifically recognized as a crime of war and a crime against humanity. Ellis, ‘Breaking
the Silence’, 240.
60 Jones, ‘Gender and Genocide’, 25.
61 Elisa von Joeden-Forgey, ‘Devil in the Details: ‘Life Force Atrocity’ and the Assault on the Family
in Times of Conflict’, Genocide Studies and Prevention 5 (forthcoming 2010).
has recently called genocide ‘a “total social practice” that [affects] all aspects of
group life’.62 In such a context, rape can indeed be a crime of genocide. During
genocide, people are usually targeted in terms of their familial roles, that is, the
roles the perpetrators perceive them to play in the reproductive process of the
group. So women and girls are tortured specifically as mothers, daughters, and
sisters; similarly men and boys are targeted as fathers, sons, and brothers. In such
cases, rape usually involves both inversion rituals (forcing family members to
watch or participate in the torture and murder of loved ones) and ritual desecra-
tions of sacred symbols of the group’s generative force (such as sexual organs,
infants and small children, and family bonds).63 Common practices across geno-
cides include killing infants in front of their parents, forcing family members to
rape one another, destroying women’s reproductive capacity through rape and
mutilation, castrating men, eviscerating pregnant women, and otherwise engaging
in ritual cruelties aimed directly at the spiritually sacred, biologically generative,
and emotionally nurturing structures of family life.
The Armenian genocide is a key example of this genocidal pattern. Over and
over again, perpetrators followed a family-based pattern of destruction. When
villages were attacked, men were murdered and their surviving family members
were raped, expelled, and killed. Perpetrators frequently engaged in inversion
rituals and ritual desecrations in the process.64 As in other cases, rape during the
Armenian genocide served many purposes: it was part of the process of eliticide,
the destruction of a group’s leadership in order to sow confusion; it publicly
demonstrated the perpetrators’ mastery over the Armenian life force; it inflicted
‘total suffering’ on both the men and the women (and, presumably, the boys and
girls) who were tortured in two ways—through violent attacks on their own bodies
and by having to witness the immense suffering of their loved ones; and it
compromised the future integrity of the group by sowing the seeds of psychic
and familial dissolution.
Few scholars have recognized the central importance of relational violence to
genocide, or its terrifying efficacy, even though it is a consistent characteristic of
survivor testimony. The absence of a ‘relational framework’ in genocide scholar-
ship, to use Adam Jones’ phrase,65 has ensured that some of the crimes common to
genocide have languished in scholarly and legal obscurity. Most often these geno-
cidal ‘life force’ atrocities are categorized simply as ‘rape’ in the literature. So, for
example, atrocities listed as instances of sexual violence by the US State
62 A. Dirk Moses, ‘Empire, Colony, Genocide: Keywords and the Philosophy of History’, in
Moses (ed.), Empire, Colony, Genocide, 13.
63 Joeden-Forgey, ‘Devil in the Details’.
64 Katherine Derderian, ‘Common Fate, Different Experience: Gender-Specific Aspects of the
Armenian Genocide, 1915–1917’, Holocaust and Genocide Studies 19:1 (Spring 2005), 5.
65 Jones, ‘Gender and Genocide’, 25.
Department’s Atrocities Documentation Team in Darfur have included the follow-

ing acts:66
· The killing of young children;

· The use of infants as weapons against their parents;
· The slashing of pregnant women’s bellies and the murder of their babies; and
· The mutilation of women’s reproductive organs, including their breasts.
We see here that the acts that go recorded by governments and NGOs as ‘rape’
involve many kinds of tortures that are not synonymous with sexual violence.
An example from the Democratic Republic of Congo (DRC) demonstrates the
way that genocidal rape is part of an elaborate set of relational rituals aimed at the
total devastation of the life force of families and communities. A survivor named
Nadine told the American playwright Eve Ensler in 2007 about an attack on her
village that resulted in her gang rape and sexual enslavement. The unidentified
soldiers killed the village chief and his children, her parents, and her brother after
he refused to rape her. They then killed each of her three children—‘They flung my
baby’s body on the ground like she was garbage.’ Nadine was gang-raped and
suffered complete rupture of her vagina and anus. While enslaved by the soldiers,
she witnessed the evisceration of a pregnant woman, whose baby was cooked and
force-fed to Nadine and the other enslaved women.67 Although Nadine’s story is
framed by a magazine article about rape, the crimes and the victims far exceed the
word. Her case demonstrates how some instances of the current violence in the
DRC are clearly genocidal.
Focusing on relational violence and life force atrocities draws in many other
instances of gross violations of human rights that do not easily conform to the
common understanding of genocide. During the 1971 war in Bangladesh, for example,
many of the estimated 200,000 rapes were accompanied by relational violence similar
to that found in Bosnia, Rwanda, and the DRC, including the evisceration of pregnant
women and the mutilation of fetuses.68 Other instances include the Japanese Army’s
attack on Nanking in World War II, its ‘comfort women’ system, and the recent war in
Sierra Leone.69 Some of the election-related violence in the Rift Valley region in Kenya
66 Kelly Dawn Askin, ‘Prosecuting Gender Crimes Committed in Darfur’, in Samuel Totten and
Eric Markusen (eds), Genocide in Darfur: Investigating the Atrocities in the Sudan (New York:
Routledge, 2006), 146–8.
67 Eve Ensler, ‘Women Left for Dead—and the Man Who’s Saving Them’, Glamour Magazine,
http://www.glamour.com/news/articles/2007/08/reallifedrama
68 Yasmin Saikia, ‘Beyond the Archive of Silence: Narratives of Violence of the 1971 Liberation War
of Bangladesh’, History Workshop Journal 58 (2004), 275–87.
69 Masahiro Yamamoto, Nanking: Anatomy of an Atrocity (Westport, CT: Praeger, 2000); James Yin
and Shi Young, The Rape of Nanking: An Undeniable History in Photographs (Chicago: Innovative
Publishing Group, 1996); Iris Chang, The Rape of Nanking: The Forgotten Holocaust of WWII (New
York: Basic Books, 1997); Yuki Tanaka, Japan’s Comfort Women: Sexual Slavery and Prostitution during
World War II and the US Occupation (New York: Routledge, 2002); Kelly Dawn Askin, ‘Comfort
in 2008 also has shown a genocidal logic, especially as regards the treatment of
children.70 Each of these cases is marked by distinct patters of inversion rituals and
desecrations of symbols of the life force.
Determining what relationship such cases have to our understanding of genocide
will rely on how we understand perpetrator intent. In Nanking, Japanese Imperial
Army soldiers may have seen the city’s inhabitants as symbolic stand-ins for the
Chinese people as a whole. In the case of the Imperial Army’s sex slavery system, it
may be that some Japanese soldiers were opportunistically acting out their subjec-
tive genocidal fantasies against the Korean, Chinese, and Philippine peoples
through their torture, mutilation, and murder of young women in the dark ante-
rooms of the rape camps. In Sierra Leone, by contrast, it appears that perpetrators
saw civilians as such to be the group to be targeted with genocide, since there was no
clear ethnic logic to the attacks though life force atrocities were widespread. And in
the DRC, where many different militia groups are involved in committing genocidal
atrocities, it may be that genocidal atrocities have become a habitus, an unforeseen
long-term consequence of the world’s mishandling of the Rwandan genocide.
G E N O C I DA L M A S C U L I N I T I E S
................................................................................................................
The widespread nature of sexual violence against women and men during many
genocides indicates that genocide is a crime intimately connected to particular
concepts of masculinity.71 Though much research remains to be done before we
can determine exactly how this is the case, it seems clear that the ideologies and
practices associated with genocide are in large part the products of the historical
experience of men and their attempts to make meaning from this experience.
It is certainly true, as Adam Jones has suggested, that the field of masculinity
studies has the potential to yield insights specifically into the long-term cultural
Women: Shifting Shame and Stigma from Victims to Victimizers’, International Criminal Law Review
1 (2001), 5–32; The Women’s International War Crimes Tribunal 2000 for the Trial of Japanese Military
Sexual Slavery, Summary of Findings, 12 December 2000; Amnesty International, Sierra Leone: Rape
and other Sexual Crimes against Girls and Women (New York: Amnesty International, 2000); Human
Rights Watch, Sowing Terror: Atrocities against Civilians in Sierra Leone (New York: Human Rights
Watch, 1998); Amnesty International, Democratic Republic of Congo, Mass Rape: Time for Remedies
(New York: Amnesty International, 2004); Jan Goodwin, ‘Silence¼Rape’, The Nation, 8 March 2008.
70 Xan Rice, ‘Murder of the Children who Sought Sanctuary in Church’, The Guardian, 3 January
2008.
71 Ronit Lentin, ‘Introduction: (En)gendering Genocides’, in Ronit Lentin (ed.), Gender and
Catastrophe (London/New York: Zed Books, 1997), 7.
developments that facilitate genocide and the perpetrators’ intentions and

motivations.72
One area that would benefit from sustained scholarly attention is the relation-
ship between war and the development of a specifically genocidal form of mascu-
linity. Such ‘genocidal masculinity’ is characterized by a concept of power that
is dependent upon the destruction of those institutions and groups that the
perpetrators believe set limits upon and constitute threats to the full expression
of their masculine identity. Since genocide is a crime intimately linked with war,73
the ways that men make sense of war and seek to cope with it should shed light
on political, social, and cultural processes that can hasten the development and
spread of particularly violent forms of masculine identity. Promising work has
been done on this front regarding veterans of World War I in interwar Germany,
especially the noted writer Ernst Jünger, who called war the ‘male form of procre-
ation’.74 Andreas Huyssen has linked interwar fascist gender constructs like Jünger’s
to soldiers’ ‘traumatic experience of emasculation’ during the war.75 In his view,
interwar fascism was a means of ‘remasculinizing’ the self by rejecting the femin-
ized civilian and peacetime world and insisting on the liberatory and elevating
power of violence, a construct that took on genocidal dimensions within the
Nazi party.
Another promising line of enquiry is the relationship between institutions of
male domination and genocidal ideology. Christopher Taylor highlighted the
gendered nature of genocidal utopia when, writing on the Rwandan genocide, he
described Hutu Power ideology as one that sought ‘an imagined past condition of
patriarchy as well as the perpetuation of Hutu dominance.’76 The link between male
domination within the perpetrator group and genocide against ‘outside’ groups is a
common one, and usually expresses itself in terms of an ersatz patriarch (the leader,
the party) who is both god and father in that he exercises final power over
reproductive choices and determines who shall live and who shall die. This explains
why political leaders who oversee genocides also often promote authoritarian and
coercive reproductive policies within their own groups. Their efforts to erode
institutions of autonomous generation among ethnic or national insiders (by
reducing women to breeders, co-opting children, and forcibly separating family
72 Jones, ‘Gender and Genocide’, 26–7.

73 Martin Shaw, War and Genocide: Organized Killing in Modern Society (Cambridge: Polity Press,
2003).
74 Ernst Jünger, Der Kampf als inneres Erlebnis, Samtliche Werke 2.1, vol. 7 (Stuttgart: Klett-Cotta,
1980), 50.
75 Andreas Huyssen, ‘Fortifying the Heart–Totally Ernst Jünger’s Armored Texts’, New German
Critique 59 (Spring/Summer 1993), 9.
76 Jones, ‘Gender and Genocide in Rwanda’, 101–102; Christopher C. Taylor, Sacrifice as Terror: The
Rwandan Genocide of 1994 (Oxford/New York: Berg, 1999), 151–79.
members) are intimately intertwined with their plans to destroy outside groups, in
whole or in part.
The relationship between masculine identity and the behaviour of foot soldiers
in genocide also warrants more in-depth research. Euan Hague has shown how the
all-male rituals of genocidal rape in Bosnia-Herzegovina, for example, were a
means of performing the potency of their Serbian national identity.77 In his
interpretation, when Serb soldiers raped Muslim and Croat women and men,
girls and boys, they were exercising their masculinist domination over civilians
that they identified specifically as feminized ethnic enemies, and this drama was a
core feature of ‘hetero-masculinist’ constructions of Serbian national identity
under Milošević.78 Such an approach frames genocide as an expressive act that,
in large part because of its gendered nature, requires constant recapitulation. Such
an understanding helps explain why genocides tend to radicalize even further at the
peripheries and expand to new victim groups.79
Finally, the ways that women find agency in these masculinist projects needs to
be better explained. Roger W. Smith, discussing the high level of direct female
participation in the Cambodian genocide, notes that ‘[i]f there had been a question
about the capacity of women to participate in political murder and to exhibit
elements of will and cruelty that, in the common imagination, are restricted to
males engaged in warfare, Cambodia seems to have resolved it.’80 He suggests that
genocide in the Cambodian case ‘occurred where gender distinctions had been
eliminated’. However, despite the coercive gender neutrality of Khmer Rouge
ideology, it is nevertheless historically part of a highly masculinized revolutionary
tradition—Stalinism, to which the Khmer Rouge leaders had been introduced as
students in France.81 The Khmer Rouge’s radical attempt to destroy all family ties,
especially the bond between mothers and children, puts it squarely within the
norm of masculinist genocidal ideology. The high level of participation of women
does not change its hegemonic masculinity, though it shows it to be much more
complex a phenomenon than we may otherwise assume. While genocide may not
be an all-male crime, it remains a masculinist one.
77 Euan Hague, ‘Rape, Power and Masculinity: The Construction of Gender and National Identities
in the War in Bosnia-Herzegovina’, in Lentin (ed.), Gender and Catastrophe, 50–63.
78 Hague, ‘Rape, Power and Masculinity’, 55.
79 Robert Gellately, ‘The Third Reich, the Holocaust, and Visions of Serial Genocide’, in Robert
Gellately and Ben Kiernan (eds), The Specter of Genocide: Mass Murder in Historical Perspective
(Cambridge: Cambridge University Press, 2003), 241–63; Helen Fein, ‘Genocide, Terror, Life
Integrity, and War Crimes: The Case for Discrimination’, in George J. Andreopoulos (ed.),
Genocide: Conceptual and Historical Dimensions (Philadelphia: University of Pennsylvania Press,
1994).
80 Roger W. Smith, ‘Women and Genocide: Notes on an Unwritten History’, Holocaust and
Genocide Studies 8:3 (1994), 325–6.
81 Eric D. Weitz, A Century of Genocide: Utopias of Race and Nation (Princeton/Oxford: Princeton
University Press, 2003), 146–7.
GENDER AND THE DEFINITION OF GENOCIDE

................................................................................................................
When the Genocide Convention was drafted after World War II, international
law—like the social sciences—was concerned with the ‘public sphere’, which was
defined in gendered ways in opposition to the ‘private sphere’. The definition of
genocide reflects the public sphere emphasis of international law inasmuch as
many of the experiences most common to women in genocides were not explicitly
considered when crafting the convention’s language. This includes rape, but also
the host of other life force crimes common to genocidal destruction that occur in
contexts that generally have been assumed to be ‘private’ (such as the household,
the space of sexual intercourse and conception, and the relationships between
family members). Life force crimes can be prosecuted under one or another of
the elements of genocide listed by the convention, but to do so is artificially to tear
apart phenomena that belong together. Taken as a whole these particular crimes
demonstrate a patterned obsession with the destruction of the life force of a group
and are aimed at the most sacred aspects of communal life.
Looking at genocide through the lens of gender therefore can help us see aspects
of the crime that otherwise have remained ‘hidden’ by bringing together phenome-
na in a way that restores the internal logic of the original crime. Gender can help us
see that genocide is, in its most basic form, a crime against the generative power of a
group and the institutions that support it, especially the family. The perpetrators of
genocide organize the destruction of a group by targeting members in accordance
with the roles that they are perceived to play in the group’s biological and social
reproduction. Since the family is the basic unit of the reproduction of groups, and
since perpetrators so often find their victims in family situations, the family and the
roles that adhere to it are prime theatres for the enactment of genocidal intent.
When we look at victims in terms of their roles as members of families, we also
are able to identify genocidal intent very early on in a conflict. Atrocities against the
life force, which are so often focused on small groups like extended families, can be
used as evidence of genocidal intent for the purpose of early warning and inter-
vention. Furthermore, the family basis of much genocidal violence has the poten-
tial to offer us deeper insight into the longer term causes of genocide, particularly
in terms of genocidal ideologies and the creation of conditions under which people
are tempted to embrace and participate in genocidal killing.
Gender analysis problematizes definitions of genocide that rely on direct killing and
numbers of dead, since focusing only on group members killed outright—the majority
of whom are often men—can erase the genocidal intent behind the persecution of
women in cases where they are allowed to go on living. The genocides in Bosnia and
Rwanda, and the current violence in the DRC, point to the theoretical possibility of
committing and achieving genocide solely through the systematic use of sexual
violence against men and women alike. The immediate ideological precursors to the
Nazi party—the Pan Germans—recognized such a possibility in 1905 by proposing
mass sterilization programmes for unwanted groups that would result in their eventual
annihilation.82
If the study of gender in genocide is important to an early warning system, it is
equally important to rebuilding efforts in post-genocidal societies.83 Gender anal-
ysis brings attention to the intentional ways in which families are disrupted and
destroyed; therefore special emphasis will need to be placed on rebuilding families
and fostering cohesion. This will require intensive public policy efforts to recognize
and de-stigmatize male and female survivors of rape by giving due attention to the
concentric circles of suffering caused by the relational nature of genocidal violence.
Particular attention will need to be paid to women survivors who experience
enormous structural vulnerability in post-genocidal societies in the form of social
ostracism, impoverishment, and homelessness due to discriminatory customs of
inheritance and limited occupational options. Many are forced to raise children
alone, including children born of wartime rape and war orphans. Many are
suffering from disabilities and illnesses related to genocidal violence. Many
are unable to conceive or carry children, which can interfere with their ability
to marry and thereby condemn them to a lifetime of economic hardship. Further-
more, studies indicate that violence against women increases in postwar contexts.
In places like Bosnia, Rwanda, and Sierra Leone, where so many rapists have gone
unpunished, women continue to live in fear of reprisals.
Gender analysis demonstrates that genocidal violence is part of a continuum of
violence related to the mystery of life-giving and women’s central place within it. In
attempting the destruction of groups, which are, after all, microcosms of our
collective humanity, génocidaires are pursuing a scenario of power that is gendered
at its core. Recognizing the gendered nature of this crime holds out the promise of
helping us not only to understand it better and to define it more precisely, but also
to engage more successfully in genocide prevention—that aspect of the Genocide
Convention that has eluded us for over six decades.
SELECT BIBLIOGRAPHY
Allen, Beverly, Rape Warfare: The Hidden Genocide in Bosnia-Herzegovina and Croatia
(London/Minneapolis: University of Minnesota Press, 1996).
82 Josef Ludwig Reimer, Ein pangermanisches Deutschland. Versuch über die Konzequenzen der
gegenwärtigen wissenschaftlichen Rassenbetrachtung für unsere politischen und religiösen Probleme
(Berlin/Leipzig: Friedrich Luckhardt, 1905), 22–3; Elisa von Joeden-Forgey, ‘Race Power, Freedom,
and the Democracy of Terror in German Racialist Thought’, in Richard H. King and Dan Stone (eds),
Hannah Arendt and the Uses of History: Imperialism, Nation, Race, and Genocide (New York/Oxford:
Berghahn Books, 2007), 21–53.
83 These final comments are based on the essays included in Carpenter, Born of War.
Bridenthal, Renate, Atina Grossmann, and Marion Kaplan (eds), When Biology Became
Destiny: Women in Weimar and Nazi Germany (New York: Monthly Review Press, 1984).
Carpenter, R. Charli, Born of War: Protecting Children of Sexual Violence Survivors in
Conflict Zones (Bloomfield, CT: Kumarian Press, 2007).
Fein, Helen, ‘Genocide and Gender: The Uses of Woman and Group Destiny’, Journal of
Genocide Research 1:1 (1999).
Grossmann, Atina, ‘Women and the Holocaust: Four Recent Titles’, Holocaust and Genocide
Studies 16:1 (Spring 2002), 94–108.
Jones, Adam (ed.), Gendercide and Genocide (Nashville, TN: Vanderbilt University Press,
2004).
Koontz, Claudia, Mothers in the Fatherland: Women, the Family and Nazi Politics (New York:
St Martins Griffin, 1988).
Lentin, Ronit (ed.), Gender and Catastrophe (London/New York: Zed Books, 1997).
Pine, Lisa, ‘Gender and the Family’, in Dan Stone (ed.), Historiography of the Holocaust
(New York: Palgrave Macmillan, 2004), 364–82.
Smith, Roger, ‘Women and Genocide: Notes on an Unwritten History’, Holocaust and
Genocide Studies 8:3 (Winter 1994), 315–34.
Stiglmayer, Alexandra (ed.), Mass Rape: The War against Women in Bosnia-Herzegovina
(Lincoln/London: University of Nebraska Press, 1994).
Taylor, Christopher C., Sacrifice as Terror: The Rwandan Genocide of 1994 (Oxford/New
York: Berg, 1999).
chapter 4
.............................................................................................
T H E S TAT E A N D
GENOCIDE
.............................................................................................
anton weiss-wendt
................................................................................................................
This chapter explores the connection between the state and genocide. My
contention is that no form of mass violence, and least of all genocide, erupts
spontaneously. It requires premeditation, usually by a government with a record
of gross human rights violations. Indeed, I argue, genocide is intricately linked to
the idea of the modern state, despite a body of scholarship that questions that
link. Non-state agents such as radical political parties or armed militias are usually
incorporated into the governing structure and therefore rarely perform on their
own. The state may deliberately use them as proxies to obfuscate the decision-
making process and thus to shift responsibility for the crimes committed. Even
though the ruling body may not always emphasize the state interests in genocide,
the painstaking reconstruction of the chain of command, where possible, inevita-
bly points to the upper echelons of power as the original source of mass violence.
In some cases the subjects may not even be able to identify the leading individuals
who constitute the state. This, however, does not make the state less present at the
crime scene.
I use the conventional definition of state, as an organized political community
under one government. In case the forces in control of the government penetrate
through the entire state apparatus, including the civil service and military, this
political system becomes a ‘regime’. The discussion of an ‘ideal type of state’ in my
82 anton weiss-wendt
opinion is as fruitless as the construction of an ‘ideal type of genocide’. Neither the

state nor any of its constituencies possess certain innate characteristics that would
make it prone to violence. Like any other outcome of human activity, the crime of
genocide is developmental and can always be traced back to a particular set of
circumstances unique to a specific time and place.1
Until the early 1990s, the state had been considered as the prime, if not the only,
agent of genocide. The Yugoslav wars of secession and Rwanda genocide, however,
have bred dissent among scholars, some of whom began arguing that non-state
actors can at times perpetrate violence on a genocidal scale without the highest
authorities’ sanction. This article asserts the original view, by deconstructing the
arguments that emphasize the role of military units and the ambiguity of dictator-
ship. The analysis of the centre–periphery interaction further affirms the primacy
of the state.
While acknowledging the role of the state in drafting and implementing the
policies of mass murder, some scholars also consider auxiliary agents of genocide.
This is particularly true with scholars examining earlier cases of premeditated mass
death. When dealing with the phenomenon of genocide beyond Nazi Germany,
Lemkin barely spent any time discussing the perpetrators. In his pioneering book,
Axis Rule in Occupied Europe, he used general terms such as ‘occupant’, ‘oppressor’,
or ‘conqueror’. However, Lemkin’s ambition to write a global history of genocide
demonstrates his intention to consider, among others, church authorities, local
warlords, and civil rulers as agents of genocide.2 Thus his unpublished manuscript
includes a chapter of Spanish colonies in America abundant with references to
non-state agents of genocide. Lemkin held responsible for the crimes the local
administration, but also the colonists of New Spain, and sometimes their indigen-
ous collaborators. Simultaneously, he emphasized that the Spanish government
never authorized slavery as such and actually tried to ameliorate the conditions of
the indigenous people.3
It is true that the UN Genocide Convention does not consider genocide a crime
planned and executed necessarily by the state. During the debates on the draft
Genocide Convention in 1947 and 1948, the attention was deliberately shifted away
from the state. The United States and the Soviet Union successfully countered the
joint United Kingdom and Belgian amendment proclaiming a state or government
the most likely offender. In the end, the British were forced to concede that states
had to carry only civil, not criminal responsibility following a violation of the
1 Frank Chalk and Kurt Jonassohn (eds), The History and Sociology of Genocide: Analyses and Case
Studies (New Haven, CT: Yale University Press, 1990), 23.
Proposals for Redress (Washington, DC: Carnegie Endowment for International Peace, 1944), 79–93.
3 Michael McDonnell and Dirk Moses, ‘Raphael Lemkin as Historian of Genocide in the Americas’,
Journal of Genocide Research 7:4 (December 2005), 510, 512.
the state and genocide 83
convention.4 This comes as no surprise, taking into consideration that the UN

Genocide Convention was a result of a political compromise whose major goal was
to safeguard the interests of the signing parties rather than accurately reflect on
patterns of history.
When applying the concept of genocide to colonial Americas and Australia,
historians tend to interpret the Genocide Convention in the light of Lemkin’s
writings. Thus, Ben Kiernan writes that some English settlers committed genocide
in some parts of North America and later in Australia. He then adds that Virginia
Indians on occasion paid the white settlers in kind. He attempts to differentiate by
using the term ‘genocidal massacres’ instead of genocide. This enables him to argue
that the so-called ‘genocidal moments’ can in equal measure be referred back to
communal violence. Like Lemkin did fifty years earlier, Kiernan names particular
governors, trade companies, and splinter groups that carried out, wittingly or
unwittingly, partial genocide. At the same time, Kiernan does not let British
colonial authorities and American federal officials off the hook, suggesting the
crime of omission, and, indeed, legitimation by their general support of the
colonization project.5 Dirk Moses pursues a similar line of argumentation, discern-
ing criminal intent from the enormous cost of Australia’s colonization. The
dramatic decline of indigenous population from 750,000 in 1788 to 31,000 in 1911
makes him implicate both central and local authorities in pursuing policies
calculated to bring about their physical destruction, pursuant to Article II of the
Genocide Convention. The more negotiating power the local settlers community
had vis-à-vis the Colonial Office, the more radical the outcome, according to
Moses.6
Despite the complex classification elaborated by social scientists—who distin-
guish between retributive, institutional, utilitarian, monopolistic, developmental,
despotic, optimal, etc. genocide7—genocide is ultimately driven by ideology. In
view of the intent to destroy, in whole or in part, tactical and integral functions of
genocide become indistinguishable. Alternately, the act of physical or biological
destruction of an entire group, no matter to what ends, is innately ideological in its
4 William Schabas, Genocide in International Law: The Crimes of Crimes (Cambridge: Cambridge
5 Ben Kiernan, Blood and Soil: AWorld History of Genocide and Extermination from Sparta to Darfur
(New Haven, CT: Yale University Press, 2007), 7, 14, 16, 35, 221–5, 232, 244–7.
6 A. Dirk Moses, ‘Genocide and Settler Society in Australian History’, in Moses (ed.), Genocide and
Settler Society: Frontier Violence and Stolen Indigenous Children in Australian History (New York:
Berghahn Books, 2004), 24–36.
7 Roger Smith, ‘Human Destructiveness and Politics: The Twentieth Century as an Age of
Genocide’, in Isidor Wallimann and Michael Dobkowski (eds), Genocide and the Modern Age:
Etiology and Case Studies of Mass Death (Syracuse, NY: Syracuse University Press, 2000 [1987]),
21–39; Kurt Jonassohn and Frank Chalk, ‘A Typology of Genocide and Some Implications for the
Human Rights Agenda’, in ibid.; Helen Fein, Genocide: A Sociological Perspective (London: Sage
Publications, 1993), 32–50.
intent. The traditional understanding of ideology as a rigid system of ideas and

beliefs regarded as justifying action is usually augmented in the context of genocide
by a pseudoscientific dogma. Obviously, violent ideology is not confined to the
state and may exist on different levels down to radical splinter group, repressive
communities, or individual fanatics. In fact, an ideology is usually a product of
intellectuals. It does not mean though, contrary to what John Heidenrich has
argued, that ‘the bloodiest genocides of human history, especially in the twentieth
century, were orchestrated by intellectuals.’8 If it had not been for shrewd and
fanatical politicians, Charles Darwin’s postulate of survival of the fittest—to cite
just one, classical example—would have remained nothing more than what it was,
namely a theory. Nonetheless, accession to political power is a prerequisite for
implementing that ideology, however important other forms of less formal power
may be in shaping or stimulating genocide. In most cases, however, violent rhetoric
and/or terror are a means of attaining power rather than a goal. From the vantage
point of the rulers to be, resorting to genocide to hold sway is counterproductive.
Even the most brutal among the actors concede the necessity of sustaining popular
support. Therefore, genocide is rarely, if at all, used to build or take over the state.
The policies of exclusion, segregation, and mass murder usually emerge from
within the state that has already attained or striving to attain legitimacy by
whatever means.
T H E O R E T I C A L A P P ROAC H E S
................................................................................................................
Social and political scientists were among the first scholars of genocide. Unsurpris-
ingly, then, the scholars working in those fields have also developed a theoretical
discourse on state and genocide. Helen Fein postulated that in order to uncover the
origins of modern premeditated genocide we must first recognize it as organized
state murder. Centrally planned and purposeful, genocide is instrumental to the
perpetrators’ ends, Fein argued. From the viewpoint of a ruling elite, genocide has a
function that helps to legitimize the existence of the state. An ideology may justify
eradicating peoples that do not fit into the new nation, by assimilating, expelling,
or annihilating them. Fein put much emphasis on the rationality of the perpetra-
tors, who allegedly weigh opportunities, costs, and sanctions before setting on a
course of destruction. She established a straight connection between war and
genocide, contending that the former reduces the deterrence against the latter.
8 John Heidenrich, How to Prevent Genocide: A Guide for Policymakers, Scholars, and Concerned
Citizens (Westport, CT: Praeger, 2001), 34.
This awards the perpetrators both freedom or action and post facto justification.
Finally, she projected a higher risk of genocide in a situation of a crisis of national
identity caused by the defeat in war.9 Later on Fein broadened her theoretical
framework, restating that genocide may be both a premeditated and improvised
response to a problem or opportunity. Simultaneously, she began differentiating
between the functions of group destruction, one of which is to eliminate a
collectivity allegedly disloyal to the present regime and another to reinforce
cohesion by restructuring the population.10
Frank Chalk and Kurt Jonassohn agree with Fein that genocide has always
required a high degree of centralized authority and bureaucratic organization.
However, they provide a different explanation, by assuming that most people are
reluctant to slaughter innocent civilians en masse. Irving Horowitz, too, places the
state at the centre of his analysis of genocide. He views genocide as the ultimate
means of social control by a totalitarian state. Norman Cohn, counter Fein and
Horowitz, views genocide not as a result of calculated action but as an attempt to
realize fantasies of redemption, messianic and apocalyptic at the same time.11
Jonnasohn tends to side with Cohn, by arguing that ideological genocides have
been committed in the name of a fundamentalist religion, a millenarian political
theory, or racial purity.12 Barbara Harff, who has coined the term politicide, puts
revolutionary upheavals in the same category as the defeat in war as a likely cause of
genocide. For her, genocide is an instance of state terrorism.13 Following Fein and
Harff, Jack Nusan Porter emphasizes war or a defeat in war as one of the precondi-
tions for genocide. The perpetrator can stigmatize victims as traitors and conceal
mass murder as an extension of military warfare.14
During the past decade scholars have introduced several variables in the initial
debate on the role of the state in genocide. Among the most important qualifica-
tions is the fact that the regimes rarely exercise absolute authority and that the
ranks of genocide perpetrators include non-state actors claiming state authority.
Harff has provided probably the most nuanced account of state-generated violence.
A scenario leading to an ideological genocide involves a new elite coming to power,
usually through a coup or revolution, with a radical vision of a new society
9 Helen Fein, Accounting for Genocide: Victims and Survivors of the Holocaust (New York: Free
Press, 1979), 7–9; Fein, Genocide, 36.
10 Helen Fein, ‘Scenarios of Genocide: Models of Genocide and Critical Responses’, in Israel Charny
(ed.), Toward the Understanding and Prevention of Genocide: Proceedings of the International
Conference on the Holocaust and Genocide (Boulder, CO: Westview Press, 1989), 5.
11 Fein, Genocide, 37, 42, 49.
12 Kurt Jonassohn, Genocide and Gross Human Rights Violations in Comparative Perspective (New
Brunswick, NJ: Transaction Publishers, 1993), 23; idem, ‘What is Genocide?’, in Helen Fein (ed.),
Genocide Watch (New Haven, CT: Yale University Press, 1992), 24.
13 Fein, Genocide, 38–9, 50.
14 Jack Nusan Porter, Genocide and Human Rights: A Global Anthology (Lanham, MD: University
Press of America, 1982), 15.
purified of unwanted or threatening elements. Besides exclusionary ideology,

another factor contributing to escalation of conflict to genocidal levels is the
narrow ethnic base of a regime. A situation in which the elite disproportionately
represent one segment in a heterogeneous society may potentially lead to mass
violence or even genocide. The elites are likely to safeguard their interests
by designing policies of exclusion, prompting the under-represented groups to
challenge them, and thus perpetuating the authorities’ insecurity.
Harff notes that the Genocide Convention fails to take into account the possi-
bility that non-state actors can and do attempt to destroy rival ethnic and political
groups. However, mass murder is never accidental nor is it an act of individuals.
According to her, genocide and politicide is carried out with the explicit or tacit
approval of powers that be or those who claim state authority. Harff sums up her
arguments as follows: ‘any persistent, coherent pattern of action by the state and its
agents that brings about the destruction of a collectivity, in whole or part, is prima
facie evidence of authorities’ responsibility.’15
Scott Straus appears to agree with Harff. He argues that state involvement exposes
the causes and shapes the character of annihilation. It is essentially impossible to find a
case of modern genocide occurring without state participation. The state can provide
massive resources and coordinated planning required in a campaign of systematic
destruction. A state project involves ideology and institutions. The questions that
scholars should be asking are how do state officials decide on a policy of extermination,
how do they convince a subject population to commit or condone genocide, and
which institutional configurations induce the commission of the crime? At the same
time, Straus doubts if the state is a necessary attribute in the definition of genocide.
According to him, a campaign to eliminate a group of population launched by non-
state actors should be equally considered genocide.16 Among the first generation of
genocide scholars, Leo Kuper is one who has accounted for all historical eventualities
when he wrote that ‘genocide is generally, though not exclusively, a state crime,
committed by governments or with their knowledge and complicity.’17
The scholars mentioned above disagree on the degree of involvement of the state
in genocide, without negating the relationship between the two. One particular
historian, Christian Gerlach, however, has embarked on a mission to erase that link
completely. Gerlach, who is best known for his works on the Holocaust, shifted the
attention from the state to what he calls ‘extremely violent societies’. Gerlach
appears as the most radical critic of the theory of state, as it has developed in the
15 Barbara Harff, ‘No Lessons Learned from the Holocaust? Assessing Risks of Genocide and
Political Mass Murder Since 1955’, American Political Science Review 97:1 (February 2003), 57–70,
quotation at 59.
16 Scott Straus, ‘Contested Meanings and Conflicting Imperatives: A Conceptual Analysis of
Genocide’, Journal of Genocide Research 3:3 (2001), 365.
17 Leo Kuper, ‘The Genocidal State: An Overview’, in Pierre van den Berghe (ed.), State Violence
and Ethnicity (Niwot: University Press of Colorado, 1990), 19.
field of comparative genocide studies. He downplays the state policies by empha-

sizing the importance of multicausality and context on the one hand and by
referring to methodological problems with the term genocide on the other. The
focus on the state and its attributes such as ideology, bureaucracy, and elites—
according to him—is a residue of the totalitarian, Eurocentric model. In his quest
for a new terminology Gerlach effectively substitutes a top-down approach to the
study of genocide by a one-dimensional model of explanation from below. For
example, in his interpretation, state and its functionaries act more as representa-
tives of the larger population strata, making ‘intent’ irrelevant in establishing
whether genocide was perpetrated. By discarding intent to commit mass murder
and denying regimes the decision-making power, Gerlach shrugs off all the com-
plexities of the term genocide, including implications of state involvement,
through his criticism of the function of prevention.18 In the final analysis, the
phenomenon of participatory violence in no way excludes the intrinsic, structural
links between the authorities, various levels of state administration, and broader
masses. Indeed, a more sophisticated conception of intent and of the way it is
shaped at different levels could accommodate Gerlach’s case studies within more
conventional conceptions of genocide.
Recent studies by Michael Mann and Mark Levene make a significant contribu-
tion to our understanding of mass violence, including the role of the state. Written
by a sociologist and a historian, respectively, The Dark Side of Democracy and The
Meaning of Genocide are remarkable books with novel theses worth comparing.
Mann and Levene take a different perspective discussing one and the same phe-
nomenon, which they call by different names. Conspicuously, the subtitle of
Mann’s book is ‘Explaining Ethnic Cleansing’, while Levene put as an overall title
for his four-volume study (of which two have appeared so far), Genocide in the Age
of the Nation State. The massive investigation into the roots of human destructive-
ness rendered interesting results.
Levene argues that we should examine the broader context in which genocide
has arisen in the modern world rather than the particularities of each instance of
genocide. As he also argues in his chapter in this volume, he links the phenomenon
of genocide with the emergence of Western powers and their expansion outside
Europe. Although the rise of the West was not identical with a comprehensive
programme of annihilation, it did create a cultural discourse that had made such
policy possible. In pursuit of that goal, the state set to organize its human resources
accordingly.19 We are talking not about a totalizing state—Levene reminds the
18 Christian Gerlach, ‘Extremely Violent Societies: An Alternative to the Concept of Genocide’,

Journal of Genocide Research 8:4 (December 2006), 458–65.
19 Mark Levene, Genocide in the Age of the Nation State, vol i: The Meaning of Genocide (London:
I. B. Tauris, 2005), 155–6; idem, Genocide in the Age of the Nation State, vol ii: The Rise of the West and
the Coming of Genocide (London: I. B. Tauris, 2005), 103–19.
readers—but the homogenizing nation-state. Why nation-states target internal

minorities is explained by the resentment of nationalist elites at the perceived
cooperation of these minorities by external powers, especially former colonial
powers, as in the case of the new Iraqi state’s attack on the Assyrian population,
whose men had been used by the British to maintain their rule of the Arabs there.20
The term ‘pathological homogenization’ introduced by political scientist Heath-
er Rae may serve as a corrective to Levene’s longue durée analysis. Echoing Levene,
Rae contends that one of the methods of state-building throughout the modern
period had been cultivating identification through exclusion of minority groups,
often by violent means. Unlike the former, however, Rae states that pathological
means of homogenization used in the past by state-builders had led to forced
assimilation and expulsion, rather than outright genocide. The propensity for mass
murder and genocide increased in the twentieth century along with the bureau-
cratic and technological capacity of the state.21
Mann isolates causation in specific types of violence by differentiating between
different forms of mass violence and focusing more narrowly on the twentieth
century. In the absence of a precise definition of ethnic cleansing, Mann uses the
term inclusively, much like Norman Naimark before him. Whereas Levene explores
the causal link between modernity, nation-state, and genocide, Mann draws a
connection between modernity, democracy, and ethnic cleansing. His first argu-
ment is, that ‘murderous cleansing is modern, because it is the dark side of
democracy.’ The extreme forms of ethnic cleansing require state coherence and
capacity and therefore are usually directed by states. When it comes to a situation
when a state slides towards ethnic cleansing, according to Mann, radicalization and
factional split is more dangerous than disintegration. The novelty of his approach is
that he considers different levels of perpetrator rather than merely identifying
perpetrators. The top three levels include, respectively, radical elites operating
party-states, paramilitaries, and the popular power base. The elites, militants, and
core constituencies are interlinked and exercise power in three distinctive ways—
top-down, bottom-up, and sideways respectively. Mann contends that ethnic re-
fugees fleeing from threatened borderlands constitute one of the main core con-
stituencies, as they are more dependent on the state for their subsistence and values.
Unlike Levene, Mann differentiates between mass murder as a crime common
throughout human history and murderous ethnic cleansing which is distinctively
modern.22 What Mann and Harff have to say about the relationship between
modern state and genocide constitutes the core argument advanced in this chapter.
20 Mark Levene, ‘A Moving Target, the Usual Suspects and (Maybe) a Smoking Gun: The Problem
of Pinning Blame in Modern Genocide’, Patterns of Prejudice 33:4 (1999), 3–24.
21 Heather Rae, State Identities and the Homogenization of Peoples (Cambridge: Cambridge
University Press, 2002), 1–6, 14, 19, 212.
22 Michael Mann, The Dark Side of Democracy: Explaining Ethnic Cleansing (New York: Cambridge
University Press, 2005), 1–34, 70.
B U R E AU C R AC Y
................................................................................................................
In their thinking of genocide, several social and political scientists referred to the state
as the source of mass violence. Irving Horowitz went the farthest, proposing the
following definition of genocide: ‘a structural and systematic destruction of innocent
people by a state bureaucratic apparatus’.23 The idea of bureaucracy as a soulless
machine that may acquire a life of its own modifies the notion of the destruction
process. We do not any longer deal with the predictable situation in which the
authorities see their order passed down the chain of command to be executed.
Difficult to comprehend, the bureaucratic mood of operation makes the path to
destruction more convoluted and thus less apparent. Set in motion by humans, it runs
on autopilot. Although the outcome of this process rarely departs from the original
vision of the leadership, the genocidal intent gets reinterpreted in terms of produc-
tivity and expediency. Internal tensions and power struggle inform the decisions of
the officials on all levels of state bureaucracy much the same as they affect the clique.
However independent in its decisions it may appear, bureaucracy is an extension
of the state and therefore has only limited freedom of action. Regimes typically
control the administrative bureaucracy through patronage and enforced ideology.
Most extensively, the role of state bureaucracy in genocide has been elaborated on
the example of Holocaust. Raul Hilberg, who had come up with the concept of ‘desk
murderer’, presented the linear model of operation of German state and party
bureaucracy involved with the ‘Final Solution of the Jewish Question’. According to
Hilberg, a new agency was engaged every time its predecessor had failed the task. Karl
Schleunes described the bureaucratic endeavour as a stopgap process in which several
agencies worked on the ‘problem’ simultaneously. Neither Hilberg nor Schleunes
doubted the role of Hitler as a prime behind-the-scene mover. Zygmunt Bauman
moved the discussion into the realm of abstraction, talking about compartmentaliza-
tion of tasks that stifle moral judgment. Building upon the research mentioned above,
Götz Aly and Susanne Heim reached the farthest-reaching conclusion regarding the
capacity of bureaucratic apparatus to further the mass murder agenda. In their
interpretation, Nazi racial planers—or ‘architects of genocide’ as they called them—
enjoyed almost unlimited freedom of action while mapping the future of the occupied
East.24 One step further and the intricate link between state as ultimate authority and
bureaucracy as conduit of ascendancy would be severed. This is unlikely to happen,
23 Irving Horowitz, Genocide: State, Power and Mass Murder (New Brunswick, NJ: Transaction
Books, 1976), 18.
24 Raul Hilberg, The Destruction of the European Jews (Chicago: Quadrangle Books, 1961); Karl
Schleunes, The Twisted Road to Auschwitz: Nazi Policy toward German Jews, 1933–39 (London: Andre
Deutsch, 1970; Zygmunt Bauman, Modernity and the Holocaust (Ithaca, NY: Cornell University Press,
1989); Götz Aly and Susanne Heim, Architects of Annihilation: Auschwitz and the Logic of Destruction
(London: Phoenix, 2003).
though, because state bureaucracy is and remains an essential part of power structure,
also when it comes to genocide.
Donald Bloxham has arrived at a similar conclusion in his analysis of the role of
bureaucracy in mass murder. As many other scholars, though, he began with a
question: ‘to what extent can bureaucracies themselves show the genocidal way as a
result of their professional problem-solving abilities, leading rather than simply
enacting policies decided on from above?’ Bloxham observes that genocide, as
predominantly a crime of state, is usually executed by the administration in the
service of the state. He demonstrates on the example of Nazi Germany how an
extremist political group managed to penetrate and subvert existing state organs,
taking over the functions of the ‘normative state’. A new bureaucracy thus created
incorporates ideologically motivated young officials striving for rapid career ad-
vancement and older authoritarian conservatives acquiescing to radical policies.
Both groups perform in anticipation of rewards and at times personal security,
which they can only receive from the state that employs them. By encompassing
ideological engagement and systematic problem-solving state bureaucracies ulti-
mately reflect regime values. The state is important in regulating popular versus
organizational participation in mass violence, Bloxham concludes.25 The escalation
of political mass violence in the Soviet Union in the 1930s can be attributed in part
to the bureaucratic mood of operation characteristic of the first socialist society.
When it comes to bureaucracy, the most peculiar feature of Stalinist terror was
the quota system. The NKVD headquarters in Moscow provided local party
and security police branches with the figures of how many thousands or tens
of thousands of ‘enemies of the people’ or members of the ‘enemy nationalities’
should be deported from any given locality. In the process, local officials often
appealed to their superior to increase the quotas, which had a snowball effect,
particularly during the Great Terror of 1936–8. This phenomenon can be explained
through the nature of the Soviet system on the one hand and protuberant instinct of
survival on the other. Many historians see the planned economy (as opposed to
market economy) as the grounding principle of the socialist system. Mass collectiv-
ization and industrialization was the kernel of the First Five Year Plan introduced in
the Soviet Union in 1928. Mass deportations of peasants and the famine could be
viewed respectively as an intentional and unintentional consequence of social
engineering. The expression ‘fulfill and over-fulfill the Five-Year Plan’ has persisted
throughout the Soviet period. However, it was more than just material rewards and
career advancement that made the officials in charge to perpetuate the class and
ethnic cleansing. The erratic nature of Soviet terror—which puts it into the category
of revolutionary terror but sets it apart from Nazi terror, for example—made
no one safe, including the individuals who had administered it. An official
25 Donald Bloxham, ‘Bureaucracy and Organized Mass Murder: A Comparative Historical

Analysis’, Holocaust and Genocide Studies 22:1 (2008), 203–45.
who did not show enough zeal in exercising an assigned task, in this case population
management, could potentially join the ranks of the unfortunate he or she had sent
on paper to faraway destinations. Just consider the rotation within the NKVD
organization whose former heads, Genrikh Yagoda and Nikolai Ezhov, were exe-
cuted as ‘the enemies of the people’ in 1938 and 1940 respectively. Despite the
arguments advanced by revisionist historians,26 the element of fear was indeed
omnipresent in the Soviet society in the late 1930s and early 1940s.
M I L I TA RY
................................................................................................................
Some scholars have been subconsciously de-emphasizing the role of the state in
genocide by drawing attention to the military as a self-governing body capable of
generating violent impulses. The discourse usually centres on military leaders in
the field or the military establishment at large. Particular instances of mass murder
that have prompted the military-centred analysis range from colonial genocides
such as German South West Africa to political genocides such as Indonesia. Those
who consider the Gypsies, alongside with the Jews, the victims of Nazi genocide
point to the indiscriminate shooting of the Ukrainian and Russian Roma by the
Wehrmacht units in 1941.
Isabel Hull has argued in the case of German South West Africa that the exter-
mination of the native populations in 1904–6 developed out of imperial military
practices and was not ordered in Berlin. To ensure the unconditional implementation
of his ‘destruction order’ from October 1904, Gen. Lothar von Trotha had first to
establish total military control over the colony. By unseating the governor, who did
not subscribe to his brutal policies, von Trotha felt free to circumvent the normal
chain of command in the colonies. In his order to hunt down and starve the defeated
Herero, von Trotha followed the nineteenth-century German military doctrine that
prescribed the destruction of the enemy as a final goal of warfare. However, as Hull has
explained, it was Kaiser Wilhelm who ordered ‘to crush the uprising by all means’—a
standard formulation used with regard to colonial revolts. It has as much to do with
the personality of von Trotha, who had been known for a ruthless suppression of
native uprisings prior to his appointment in German South West Africa.27 In other
words, he had arrived in Windhoek ‘to do the job’, as he understood it. Moreover,
26 The strongest exponent of the ‘push from below’ theory is Robert Thurston, Life and Terror in
Stalin’s Russia, 1934–1941 (New Haven, CT: Yale University Press, 1996).
27 Isabel Hull, ‘Military Culture and the Production of “Final Solutions” in the Colonies: The Example
of Wilhelminian Germany’, in Robert Gellately and Ben Kiernan (eds), The Specter of Genocide: Mass
Murder in Historical Perspective (New York: Cambridge University Press, 2003), 144–62.
Governor Leutwein who had opposed von Trotha’s extermination policy was himself
a soldier experienced in putting down revolts. The General Staff that had installed von
Trotha in his position convinced the Kaiser to reverse the ‘destruction order’. As Hull
writes, Bismarck’s ultimate intention in masterminding a policy that had granted the
military a free hand was to safeguard the monarchy. Therefore, it would be incorrect
to present the German military, the General Staff, and specifically General von Trotha
as acting in opposition to the central authorities. Rather, they discharged duties in
accordance with contemporaneous norms and preconceptions and in anticipation of
sanction from above. By the same token, the parliamentary debates and the public
outcry caused by the inhuman treatment of colonial subjects in places like Tasmania
and German South West Africa testifies to the emergence of pluralistic society in
Europe rather than to a presumed goodwill of the European governments. Last but
not least, both cases can be deemed transitional and as such hardly fitting into the
pattern of modern ideologically driven genocide.
The emphasis on the military runs counter to a tendency of dissociating genocide
from warfare. Many activists hailed as groundbreaking the clause in the Genocide
Convention specifying that genocide can be committed not only in time of war but
also in time of peace. This may be regarded a personal victory for Raphael Lemkin,
who had failed to introduce the changes of genocide in the decision of the Interna-
tional Military Tribunal (IMT) in Nuremberg in 1946. The marginal treatment of the
Holocaust in IMT proceedings was partially due to the Allied decision to tie the Nazi
war crimes with the so-called crimes against the peace. Having taken the outbreak of
the Second World War as a starting point for the legal investigation, the prosecution
effectively excluded the 1930s policies of exclusion and discrimination—which had
paved the way to genocide—from consideration.28 The legal innovation that erased
the distinction between the maltreatment of civilians in time of war and peace has led
some social scientists to reject the link between war and genocide. According to Paul
Bartrop, for example, five out of fifteen major genocides of the twentieth century that
he has listed took place outside of a military conflict. He mentions specifically the
1932–3 famine in the Soviet Union, Indonesian massacres of 1965–6, Burundi killings
in 1972, and routine executions in Cambodia between 1975 and 1979.29 This and
similar analyses ingrained in the quantitative method tend to disregard context. To
the same extent to which the violent regimes invent enemies, they conjure up wars.
This is particularly true in the case of communist dictatorships, which perpetuate the
siege mentality.
The fear of losing control over the subject population, or even worse of being forced
from power, is a potent factor when it comes to unleashing ethnic violence. Common
and Memory (Oxford: Oxford University Press, 2001), 57–89.
29 Paul Bartrop, ‘The Relationship between War and Genocide in the Twentieth Century:
A Consideration’, Journal of Genocide Research 4:4 (2002), 519–32.
sense and history teach us that democracies are less prone to committing mass crimes
than dictatorships. Much like in human relationships, the shrewd ways of gaining
influence or misuse of office at the state level breed suspicion of those in power.
Although the outside threat to the ruling clique may be real, more often than not it is
imaginary. The regime perceives of a threat to its existence exaggerated. The statesmen
in the 1930s Soviet Union and the 1970s Cambodia acted out of fear of war. Both
countries were born into violence. Russia experienced a world war, a revolution, a civil
war, a foreign intervention, and famine before Stalin began implementing mass
collectivization and industrialization that conditioned the horrendous famine of
1932 and 1933, which struck Ukraine particularly hard. Confiscation of grain was part
and parcel of a defence programme that was meant to ensure the Soviet Union’s
survival. Stalin was one of many old Bolsheviks who had been obsessed with hostile
encirclement, border infiltration, and eventually full-scale war against the nascent
socialist state. In Stalin’s mind, whatever he did—including starving his own peo-
ple—made the country stronger in the face of an inevitable military confrontation with
the capitalist world. Magnified suspicion can at times border on paranoia. I mean
political rather than individual paranoia, for Stalin had been clinically diagnosed with
mental disorder to which his disastrous policies could have been attributed.
It was not substantially different in the case of the Khmer Rouge whose war rhetoric
had become a self-fulfilled prophecy. Cambodia went through a foreign occupation
during the Second World War, a struggle for independence, a civil war, and the
American bombing prior to the communist takeover in 1975. The alienation from
the Vietnamese patron soon escalated into the persecution of the Vietnamese minor-
ity, the incursion into the neighboring country’s territory, and eventually defeat of Pol
Pot’s Cambodia at the hands of the Vietnamese. The impending war with the ‘Western
imperialism’ and Vietnam served as an ultimate rationale for spreading terror within
Cambodia. Thus, there exists a direct connection between war scare and cumulative
radicalization. Simultaneously, it discredits the notion of total war as a conduit for
genocide. It may be true in some cases—for example Ottoman Turkey, as Jay Winter
has demonstrated,30 and Nazi Germany—but not in others.
PA R A M I L I TA R I E S
................................................................................................................
Mann considers the military as one of the four power networks—alongside with
ideological, political, and economic factors—that may produce the rationale for
30 Jay Winter, ‘Under Cover of War: The Armenian Genocide in the Context of Total War’, in
Gellately and Kiernan (eds), The Specter of Genocide, 189–213.
genocide (Mann continually uses the term ‘ethnic cleansing’). Armies, police
forces, and irregulars are the main agencies of military power.31 Whereas the
armed forces are expected to protect the interests of the state, the paramilitaries
often serve as an extension of the military, as it was the case in Indonesia, for
example. Other times, the initiative to build irregular troops comes directly from
the state authorities, without any mediation. Ottoman Turkey would be a good
example of that. The shock units formed by political parties in their quest for
power (though official justification is usually self-defence) represent the third type
of auxiliaries. This kind of militia sprang up in Rwanda in the wake of the Arusha
Accord and the Ndadaye assassination. Bands of armed men can also form
spontaneously—or so it may appear at first glance—in secessionist territories, as
it occurred in the Serb settled areas of Bosnia-Herzegovina. The most extensive use
of irregular forces is possible under condition of prolonged military occupation,
particularly in the territories that had experienced an enemy rule before. This
phenomenon is usually described as collaboration, however imperfect is the term.
Nazi Germany was fairly successful at raising local units amongst the Estonians,
Latvians, Lithuanians, Ukrainians, and some Muslim peoples in the occupied
Soviet territories. The state is eager to use militias, which it has clandestinely
trained and armed, to simulate the condition of civil war. That disposition can
afford two types of action, both beneficial to the state: denial or intervention. The
officials may denounce the ‘rumours’ of state involvement, letting the bloodshed
run its course. The role of militias in spearheading and/or meticulously carrying
out mass murder is difficult to gauge due to the multiple chain of command, which
can only be established with certainty post facto. The International Court’s of
Justice ruling in the case of Bosnia-Herzegovina against Serbia indirectly touched
upon this problem. Some observers commented that the judgment from February
2007, which pronounced Serbia not guilty of the crime of genocide, could have
been different if the government in Belgrade did not insist on withdrawing some
important documents that had allegedly exposed the subordinate position of the
Bosnian Serb administration vis-à-vis Milošević’s regime.32
In the modern period, the use of proxies has been associated with the colonial
conquest and the subsequent wars of liberation in Africa and Asia. During the
Cold War the superpowers fought each other in the name of a superior ideology
(at least nominally) by deploying indigenous troops that they had armed and
trained. What might have looked as a conventional civil war to the outside world
was planned and accounted for thousands of kilometres away. At times violence went
out of hand: the armed units that were supposed to fight ideological warfare had split
and regrouped along tribal lines, massacring each other instead of a designated
enemy and their patrons. In certain cases, magnified manifolds, this scenario
31 Mann, The Dark Side of Democracy, 32.

32 See New York Times, 9 April 2007.
played out also in the context of genocide. While the United States sponsored the
southerners in their fight against the Viet Cong, the Vietnamese communists
supported the Khmer Rouge guerrillas in neighbouring Cambodia. Once the
Khmer Rouge seized power in the country, they turned against their political backers
and the Vietnamese people at large. No matter what is the political setup, the type of
irregulars, or the extent of atrocities they commit, the traces inevitably go back to the
state or the shadowy forces acting in its name. No scholar has succeeded so far in
proving otherwise.
In Sudan and the former Yugoslavia the relationship between paramilitaries and
the state is rather unambiguous. Thanks to compulsory military service, by 1990
the Yugoslav People’s Army (JNA) numbered 185,000 men on active duty and half a
million reservists. Up to three million men were subject to conscription by the
Territorial Defence. Those who remained in the country when war broke out in
1991 had few choices. They could either join the armed forces or one of the militias
that operated in Croatia and Bosnia-Herzegovina. While the federal state was
falling apart, the nationalist authorities in Belgrade bet on paramilitaries, who
could control territory and thus strengthen the Serbian state. Out of 10,000 Serb
volunteers who fought in Bosnia, half had previously served in the Yugoslav
People’s Army. Unsurprisingly, then, the military controlled most of the militias,
with the rest run by the State Security Service. The latter group of paramilitary
units, including the notorious ‘Tigers’ under the command of Željko Ražnatović
(Arkan), emphasized their connection with the Security Service. In some cases,
though, the rank and file were unaware of that arrangement.33
Bosnian Serb forces and Serb paramilitaries account for most deaths in 1991–5
(with a high estimate put at 280,000). Evidence of central planning is in abun-
dance. The JNA provided arms to the local Serbs in Bosnia-Herzegovina, and on
several occasions participated in the atrocities. Paramilitary units were particularly
efficient at carrying out the policy of ethnic cleansing when attached directly to
regular Army units. Even though direct guidance for individual operations was
normally exercised at the commander’s level, the leadership set certain policy goals,
allowing their subordinates the latitude to achieve them. If it were otherwise,
Serbian political and military authorities in Bosnia and Belgrade would have
made an effort to suppress local gunmen, as did the legitimate Bosnian govern-
ment. The Belgrade leadership bears ultimate responsibility for the crimes com-
mitted, Norman Cigar insists. Milošević evidently provided financial and logistic
support for some of the most brutal Serb militias operating in Bosnia, channelling
it through Serbia’s Ministry of Defence. According to a militia leader Vojislav Šešelj,
33 Aleksandra Milićević, ‘Joining the War: Masculinity, Nationalism and War Participation in the
Balkans War of Secession, 1991–1995’, Nationalities Papers 34:3 (July 2006), 266; Milićević,
‘Paramilitaries and the State: The Case of Serbia’, paper delivered at the 7th Convention of the
International Association of Genocide Scholars, Sarajevo, 12 July 2007.
Milošević was in charge, even though he gave verbal rather than written orders. It is
true that at a later point Milošević reined in some militias, including that of Šešelj.
However, he did act only after he started viewing these groups as a political threat.
Along with other state agencies and nationalist organizations, the Serbian Ortho-
dox Church lent its hand in organizing, financing, and arming the infamous
Serbian Volunteer Guard, as its leader Arkan had subsequently acknowledged.
Arkan once stated that ‘every member of paramilitary units must in the first
place be responsible to the Serbian people and must respect the parliament and
the president of the Republic.’34
The Sudanese authorities have been using militias as a counterinsurgency
strategy since 1985. The government in Khartoum began mobilizing the Arab tribes
against the Sudanese People’s Liberation Army during the civil war in the south.
Baggara Arab militias received arms and military training to strike against the
Dinka and Nuba peoples suspected of supporting the rebels. The use of Arab
militias enabled Khartoum not only to conserve its own, overstretched resources,
but also to disguise its intervention as ‘age-old tribal conflict’. Between 1985 and
2003, militias supported by military intelligence and aerial bombardment carried
out a brutal policy of scorched earth, massacring, pillaging, and raping civilians. In
2003, Gen. Omar al-Bashir, who had seized power in the country fourteen years
earlier, began using the same strategy in the Darfur region. The Sudanese army,
untrained in desert warfare, was ineffective against the rebels, who used hit-and-
run tactics. Experiencing humiliating defeats, the government of Sudan set out to
crush rebellion in Darfur by arming militias from local Arab tribes, collectively
known as Janjaweed. The Janjaweed struck against the civilian populations of those
non-Arab tribes from which the rebels had largely drawn their recruits, that is, the
Fur, Zaghawa, Massalit, and others. Arab militias resorted to a brutal practice of
ethnic cleansing aimed at replacing the local population with Arab settlers, just as it
had done in oil-producing areas of the south and the Nuba Mountains.35
A loosely organized Arab militia force, on horseback and camel, comprised some
20,000 men. Many criminals were released on the promise of joining the militia.
This force was better trained, armed, and supplied than similar units in the past.
The recruits were paid a decent salary considering the economic situation of the
region. Many militiamen received regular army uniforms and insignias of ranks.
More important, Khartoum coordinated the activities of its regular forces and
those of Janjaweeds. Thus the government directed militia activities rather than
merely condoned them. The most powerful militia leader, Musa Hilal, does not
34 Norman Cigar, Genocide in Bosnia: The Policy of ‘Ethnic Cleansing’ (College Station: Texas A&M
University, 1995), 36, 48–55, 64–5, 104; Benjamin Lieberman, Terrible Fate: Ethnic Cleansing in the
Making of Modern Europe (Chicago: Ivan R. Dee, 2006), 307; Norman Naimark, Fires of Hatred: Ethnic
Cleansing in Twentieth-Century Europe (Cambridge, MA: Harvard University Press, 2001), 161.
35 Julie Flint and Alex de Waal, Darfur: A Short History of a Long War (London/New York: Zed
Books, 2005), 24–5, 55–61, 64, 68–9.
even try to deny his links to the government in Khartoum, from which he receives
marching orders. In fact, except for their sandals, turbans, and the emblem—an
armed man on camelback—the Janjaweed are undistinguishable from regular
troops. Hilal stated that he has raised a tribal militia at the request of the govern-
ment to fight the rebellion in Darfur. Since the beginning of the aggression in
Darfur, the Janjaweeds have become increasingly integrated into the Sudanese
military structure. Many militia members have been incorporated into the police,
security service, and various paramilitary organizations.36
D I C TATO R , O N E -PA RT Y R U L E , S TAT E

................................................................................................................
An intelligent doubt about the preponderance of the state in the genocide is rooted
in the continuous debate in the social sciences between those who emphasize the
role of individual and collective actors and those who accentuate institutional
structures.37 For some Holocaust historians it may sound like the old Intentional-
ist–Functionalist debate.
When it comes to initiating and sanctioning destruction, the visibility of leader
makes him a natural subject of inquiry. Unable to comprehend the complex
relationship between various levels of civil and military administration, survivors
instinctively search for answers in the personality of a dictator, who is believed, and
rightly so, to carry the burden of responsibility for committed atrocities. As with
the previous discussions of bureaucratic, military, and paramilitary structures, the
question is whether it is possible to dissociate the elites from the state. The evidence
suggests a negative answer.
Despite the significant body of literature that examines the world’s dictators from
a comparative perspective, personality characteristics of the leaders associated with
genocidal violence rarely match.38 What they all do have in common is the quest for
power. By explicating the synergy between the ruler and the ruled one can better
understand the motives behind the crime. Leaders and the ruling elites are often
motivated by self-interest. To achieve their goals, they may claim that the purpose
36 Ibid. 36–41, 86, 101–14; Gerard Prunier, Darfur: The Ambiguous Genocide (Ithaca, NY: Cornell
University Press, 2005), 97–109, 117, 134, 152–5; Joyce Apsel (ed.), Darfur: Genocide before Our Eyes
(New York: Institute for the Study of Genocide, 2005), 23, 26, 33, 40, 54–5, 62–3, 67.
37 George Andreopoulos, ‘Introduction: The Calculus of Genocide’, in Andreopoulos (ed.),
Genocide: Conceptual and Historical Dimensions (Philadelphia: University of Pennsylvania Press,
1994), 8.
38 See, for example, Ben Kiernan, ‘Pol Pot and Enver Pasha: A Comparison of the Cambodian and
Armenian Genocides’, in Levon Chorbajian and George Shirinian (eds), Studies in Comparative
Genocide (New York: St Martin’s Press, 1999), 165–78.
of destruction is to bolster the power of an entire people. The ultimate skill of an

authoritarian leader is to mobilize the population, by cultivating and channelling
violent impulses. They strive to move a culture down a path of mass violence; they
promote destructive ideologies, which can win them fanatical supporters; and they
create critical infrastructure in the form of a bureaucracy and/or a military.39
The perception of a leader who single-handedly decides on genocide runs the
danger of overlooking both structural factors and collective prejudices that may
fuel the machine of destruction. The one scholar who rejects this thesis is Benjamin
Valentino. Valentino argues that the violence is typically performed by a relatively
small group of people, usually members of military or paramilitary organizations.
Regime’s leaders do not need to seek the broader public support, as they are
capable of recruiting the few individuals to carry out genocide (Valentino uses a
substitute term, ‘mass killing’). Society at large comes to play a role in genocide
mainly as passive onlookers, compliant with authority and indifferent to the fate of
victims. A relatively small group of political or military leaders, according to
Valentino, can wring the acquiescence to their radical policies even in societies
actively opposed to them. He admits the factor of situational pressure only as far as
it relates to leaders’ strategic goals and beliefs. As an argument, he invites his critics
to imagine the Great Terror without Stalin or the Holocaust without Hitler.40
The worn out Hitler–Stalin comparison does not suggest a one-man dictator-
ship in each and every instance of genocide. A single standing figure of a tyrant is
not obvious in the case of Cambodia, is clearly missing in the case of Rwanda, and
should be multiplied by three in the case of Ottoman Turkey. The Khmer Rouge
genocide by all accounts lacked a charismatic leader. For quite some time, foreign
observers were unable to tell who was actually running Democratic Republic of
Kampuchea. Officially, Cambodia continued to be ruled by the government in exile
headed by Norodom Sihanouk, until his forced removal in March 1976. Important
decisions, however, were made by the mysterious body called Revolutionary
Organization (angkar padevat). The Khmer cadre pledged absolute loyalty to the
Organization, without necessarily knowing the leading personalities. The Khmer
Rouge denied the authoritarian nature of their regime by stressing the collective
nature of the leadership. For the same reason the names of the leaders of the
‘Organization’ were kept secret. Pol Pot went by the name Brother No. 1, Nuon
Chea Brother No. 2, and so on.41 The assassination of the Rwandan President,
whose plane was shot down on April 6, 1994, in circumstances still unknown,
39 Israel Charny, How Can We Commit the Unthinkable? Genocide: The Human Cancer (Boulder,
CO: Westview Press, 1982), 193–9; Linda Woolf and Michael Hulsizer, ‘Psychological Roots of
Genocide: Risk, Prevention, and Intervention’, Journal of Genocide Research 7:1 (2005), 106–8.
40 Benjamin Valentino, Final Solutions: Mass Killing and Genocide in the Twentieth Century (Ithaca,
NY: Cornell University Press, 2004), 2–7.
41 David Chandler, The Tragedy of Cambodian History: Politics, War, and Revolution since 1945
(New Haven, CT: Yale University Press, 1991), 246, 258.
propelled into power a group of relatives and close associates of Juvénal Habyar-
imana who had reasons to seek his death. The interim government subsequently
sworn in consisted of extremist Hutu politicians. However, it appears that those
individuals were mere puppets in the hands of the actual génocidaires. The
Committee of Union and Progress in Turkey was led by three men—Talat Pasha,
Enver Pasha, and Djemal Pasha—who had collectively devised a policy leading to
the demise of the Armenian minority.
The complex processes leading to genocide should not be reduced to a regime’s
leaders and their ‘insanity’. Whether economic gain or territorial acquisition,
revenge or security motivate a genocidal campaign against a certain group, the
leaders of the regime believe in the justice of their cause and therefore are incapable
of conceding that their policy was criminal. In that respect Talat Pasha, Hitler, Pol
Pot, and even al-Bashir, all resemble each other. The closest that Stalin came to
regret after the Great Terror in 1938 was his concession that some ‘mistakes’
were made.42 Pol Pot had fought to the last against the government in Phnom
Penh, however undemocratic it has been. Hitler preferred to go up in smoke but
not to surrender to his mortal enemies, and Talat Pasha died unrepentant, cut
down by a bullet of an Armenian assassin. Although genocide is almost always led
by state elites, this is the end process of state disintegration, reconstitution, and
radicalization.43
C O N C LU S I O N S
................................................................................................................
Scholars are reluctant to use qualifying adjectives such as ‘terrible’, ‘horrendous’,
and ‘inconceivable’ when talking about genocide. These words lack the precise
meaning and thus may impede the detached analysis. When all is said and done,
however, we are left with no other choice but to acknowledge that genocide occurs
when deadly calculus meets desperate and paranoid minds. Pogroms, massacres,
ethnic cleansing, or even mass killings—all these violent acts may have their origin
in popular culture, perpetuated and perpetrated by the masses. But not genocide!
The plan to wipe out an entire group—‘leaving none to tell the story’, in the words
of Alison des Forges—can only be born in the upper corridors of power, or
alternatively to crystallize on its way up through the existing hierarchies. In either
case, it requires the machine of state to implement the utopian vision of society.
42 Hiroaki Kuromiya, Stalin: Profiles in Power (Harlow: Pearson-Longman, 2005), 126.

Ideology, bureaucratic apparatus, political parties, the military, militias—these are

the constituents of a state. These structures are complimentary, not exclusionary.
From the vantage point of the perpetrators, genocide is contradictio in adjecto.
They believe that their nation is on the verge of collapse yet must regain its past
glory. They belittle and dread the victim group at the same time. They delegate the
delicate act of destruction to various agencies, while reserving the last word for
themselves. They want to implicate in murder as many people as possible, without
lessening their grip on power. They may be few, styled as many. The only element
that remains constant through all phases of genocide is the presence of the state.
Elites, political parties, bureaucracies, armed forces, and paramilitaries—all these
entities can enter agential state, in isolation or in aggregate. However weak may
appear the state, it serves as an invariable reference point in the case of genocide.
With all the variables, detours, ambiguities, and exceptions accounted for, genocide
is still primarily a crime of state.44 The authoritarian system of governance implies
that subordinate agencies and individuals can only carry out acts of genocide with
the active or tacit consent of the senior leadership. Whenever genocide may appear
to have been committed by individual actors in pursuit of their own goals, the
latter inevitably act in concert with the government seeking to expand state control.
In all circumstances, the central authorities remain clear stakeholders in the
outcome of the genocide.45
SELECT BIBLIOGRAPHY
Bloxham, Donald, ‘Bureaucracy and Organized Mass Murder: A Comparative Historical
Analysis’, Holocaust and Genocide Studies 22:1 (2008), 203–45.
—— Genocide on Trial: War Crimes Trials and the Formation of Holocaust History and
Memory (Oxford: Oxford University Press, 2001).
Chalk, Frank, and Kurt Jonassohn (eds), The History and Sociology of Genocide: Analyses
and Case Studies (New Haven, CT: Yale University Press, 1990).
Flint, Julie, and Alex de Waal, Darfur: A Short History of a Long War (London/New York:
Zed Books, 2005).
Horowitz, Irving, Genocide: State, Power and Mass Murder (New Brunswick, NJ: Transac-
tion Books, 1976).
Kiernan, Ben, Blood and Soil: A World History of Genocide and Extermination from Sparta to
Darfur (New Haven, CT: Yale University Press, 2007).
Levene, Mark, Genocide in the Age of the Nation State, vol i: The Meaning of Genocide
(London: I. B. Tauris, 2005).
44 Frank Chalk, ‘Redefining Genocide’, in Andreopoulos (ed.), Genocide 60.

45 Catherine Barnes, ‘The Functional Utility of Genocide: Towards a Framework for Understanding
the Connection between Genocide and Regime Consolidation, Expansion, and Maintenance’, Journal
of Genocide Research 7:3 (2005), 311, 313.
Mann, Michael, The Dark Side of Democracy: Explaining Ethnic Cleansing (New York:
Rae, Heather, State Identities and the Homogenization of Peoples (Cambridge: Cambridge
Valentino, Benjamin, Final Solutions: Mass Killing and Genocide in the Twentieth Century
(Ithaca, NY: Cornell University Press, 2004).
chapter 5
.............................................................................................
GENOCIDE AND
M E M O RY
.............................................................................................
dan stone
................................................................................................................
We live in a memory-obsessed age. Western culture is suffused with autobiogra-
phies, especially with traumatic life narratives about the legacies of abusive child-
hoods. Tourism consists to a large extent of the consumption of ‘heritage’ such as
castles and stately homes; memorials and museums increasingly dot the landscape;
and commemorative events seem to occur with increasing frequency. The history
of genocide is also affected by these broad cultural trends; indeed, in some respects
it exemplifies them. The perpetration of genocide requires the mobilization of
collective memories, as does the commemoration of it. For the individual victims
of genocide, traumatic memories cannot be escaped; for societies, genocide has
profound effects that are immediately felt and that people are exhorted (and
willingly choose) never to forget. ‘Dark tourism’—visits to death camps or other
sites of mass murder—is fully integrated into the tourist trail.1 Although thinkers
as diverse as Friedrich Nietzsche, Ernest Renan, Paul Ricoeur, and Marc Augé
might be right to suggest that forgetting is essential for the health of society,
genocide is less amenable to willed oblivion than most events because of the
deep wounds it creates; thus, in the memory politics that surround it, genocide
can scar societies long before and long after its actual occurrence. This chapter
1 J. John Lennon and Malcolm Foley, Dark Tourism: The Attraction of Death and Disaster (London:
Continuum, 2000).
genocide and memory 103
shows how genocide is bound up with memory, on an individual level of trauma

and on a collective level in terms of the creation of stereotypes, prejudice, and post-
genocide politics.
Before demonstrating the validity of these claims, it is necessary to say some-
thing about ‘memory studies’. The basic premise of the study of ‘collective memory’
is not a quasi-mystical belief in the existence of a social mind, or that societies can
be treated as organic wholes (in the manner supposed by many genocide perpe-
trators); rather, it is the basic claim that, in order to live meaningfully as a human
being, that is, in order to have memories (for, as neurologists increasingly show,
memory and selfhood are intrinsically linked), one must exist in a social setting.
This claim, which has its origin in the work of French sociologists Emile Durkheim
and Maurice Halbwachs, and perhaps reaches its zenith in Ricoeur’s last major
work, Memory, History, Forgetting (2000), overturns the intuitively appealing
‘methodological individualism’ of much twentieth-century thought, installing in
its stead a ‘methodological holism’. Whilst groups do not have memories in the
neurological sense and thus there is no organic basis to the term ‘collective
memory’, nevertheless, ‘Collective memories originate from shared communica-
tions about the meaning of the past that are anchored in the life-worlds of
individuals who partake in the communal life of the respective collective.’2
Thus collective memory becomes something that the historian or other scholar
can study; memory can be a subject for critical historiography in the same way as
gender or class. Historians can think theoretically about what collective memory is,
how it is constructed and what it excludes, and they can provide detailed case
studies, for example, in examining Italians’ memories of fascism or the ways in
which the My Lai massacre has been domesticated in American collective memory.
Most often historians have focused on what Pierre Nora calls ‘lieux de mémoire’,
sites such as memorials, museums, or significant buildings (like the Panthéon in
Paris, the Neue Wache in Berlin, or the Vietnam Veterans Memorial in Washington,
DC), showing how a group’s (usually a nation’s) self-identity is anchored in these
sites of memory. What such sites exclude becomes as relevant for understanding
collective memory as the narratives they promote.
More recently, some historians have criticized this model for studying collective
memory.3 It is too easy to do, they say, because it is focused on material objects or
aesthetic representations whose meaning can be shown to change over time as
people interact with them differently in changed circumstances. For example, the
meaning of Auschwitz to Catholic Poles living under communism before 1989 was
2 Wulf Kansteiner, ‘Finding Meaning in Memory: A Methodological Critique of Collective

Memory Studies’, History and Theory 41 (2002), 188.
3 Alon Confino and Peter Fritzsche (eds), The Work of Memory: New Directions in the Study of
German Society and Culture (Urbana: University of Illinois Press, 2002); Alon Confino, Germany as a
Culture of Remembrance: Promises and Limits of Writing History (Chapel Hill: University of North
Carolina Press, 2006); Kansteiner, ‘Finding Meaning’.
104 dan stone
different from the meanings that the camp acquired after the end of the Cold War
once the site became internationalized. From being a site that acted for Poles as a
metaphor for the evils of foreign occupation, Auschwitz became a key site in the
Europeanization of Holocaust consciousness when, after the collapse of commu-
nism, its overwhelmingly Jewish victims were increasingly recognized. Far more
meaningful than studying sites of memory, according to the critics, would be to
trace the ways in which conflicts over memory affect social relations. In other
words, we need to show how memory is linked with power. Doing so, argues Wulf
Kansteiner, requires scholars of memory to think more carefully about their
methodology. The scholarship, in his opinion, needs to delineate more clearly
the distinctions between individual and collective memory and to think more
about reception than about representation. It would benefit from adopting some
of the vocabulary and methodology of media studies, with the result that collective
memory would be understood as the result of the interaction of three ‘types of
historical factors: the intellectual and cultural traditions that frame all our repre-
sentations of the past, the memory makers who selectively adopt and manipulate
these traditions, and the memory consumers who use, ignore, or transform such
artefacts according to their own interests.’4
But for historians memory is more than just a research topic.5 Historians are also
part of the broader culture, one that already fifteen years ago was diagnosed as
suffering from a ‘surfeit of memory’.6 Critics of the memory culture argue that, like
‘heritage’, memory is exclusionary, reactionary, and nostalgic; at its worst, it can be
accused in its quest for authenticity and ‘re-enchantment’ of ‘projecting “psycho-
neurotic jargon” onto the memory of various national or (more often) ethnoracial
groups.’7 Memory is, in fine, one of the more dangerous tools of identity politics.
Thus, scholars need to consider their own investments in memory politics, espe-
cially when writing about subjects like genocide. Interventions in, for example,
debates about commemorative practices in Rwanda cannot be made on a whim.
But finally, memory is inseparable from history, so that even when the current
‘memory obsession’ has passed, when the piles of confessional literature have been
pulped and the commemorative ceremonies are unattended, still, as Ricoeur notes,
memory will be the ‘bedrock’ of history. The fact that people can say that ‘this has
happened’ remains the starting point for historiography.8 Studying the links
between genocide and memory means, then, examining the ways in which
4 Kansteiner, ‘Finding Meaning’, 180.

5 Kerwin Lee Klein, ‘On the Emergence of Memory in Historical Discourse’, Representations 69
(2000), 127–50.
6 Charles S. Maier, ‘A Surfeit of Memory? Reflections on History, Melancholy and Denial’, History
and Memory 5:2 (1993), 136–52.
7 Klein, ‘On the Emergence’.
8 Paul Ricoeur, Memory, History, Forgetting (Chicago: University of Chicago Press, 2004 [orig.
French 2000]). See also Dan Stone, ‘Beyond the Menmosyne Institute: The Future of Memory after
collective memories of past humiliations or victories are mobilized in the present,

showing how individuals and societies are traumatized by genocide, and analysing
the ways in which post-genocidal commemorative practices sustain collective
memories.
In 1950, Champetier de Ribes, the French Prosecutor, stated that Nazi crimes
‘were so monstrous, so undreamt of in history throughout the Christian era up to
the birth of Hitlerism, that the term “genocide” has had to be coined to define it.’
As the legal scholar Alexander Greenawalt, who cites de Ribes, notes, the United
Nations Genocide Convention (UNGC) was not merely a way of codifying indi-
vidual guilt. The concept of genocide ‘is as much about questions of history and
collective memory’.9 The background to the UNGC and questions of the definition
of genocide are explored elsewhere in this Handbook; here I wish only to develop
the point that genocide and memory are inseparable, for reasons of the cultural
freight that the term contains as well as, more obviously, the enormity of the crime
itself. In what follows, I will analyse the nature of this relationship.
M E M O RY A S M O B I L I Z AT I O N
................................................................................................................
It is tempting, when trying to understand perpetrators of genocide, to assume that
they are convinced of their own superiority, that they are the arrogant bearers of an
ideology that requires the merciless elimination of the weak. For example, one
interpretation of the Holocaust suggests that behind the murder of the Jews lay a
deeper desire to overthrow the moral law—represented by the Ten Command-
ments, the basis of Judeo-Christian civilization—and reinstate the right to commit
genocide, as in the virile, martial societies of ancient Greece.10 Such rhetoric is
not hard to find, especially in colonial settings where the social Darwinist notion
of superior races ‘superseding’ the inferior was common. Yet, in fact, most geno-
cides result from processes of worsening national or imperial crisis that give rise
to a feeling of massive insecurity or existential threat among the perpetrators.
A curious, paradoxical logic is at work: genocide perpetrators commit the most
horrific crimes in the belief—always exaggerated and sometimes outright fantasti-
cal—that they are defensive acts to ensure that they will not suffer the same fate. In
the Age of Commemoration’, in Richard Crownshaw, Jane Kilby, and Anthony Rowland (eds), The
Future of Memory (New York: Berghahn Books, 2010).
9 Alexander K. A. Greenawalt, ‘Rethinking Genocidal Intent: The Case for a Knowledge-Based
Interpretation’, Columbia Law Review 99:8 (1999), 2294.
10 Gunnar Heinsohn, ‘What Makes the Holocaust a Uniquely Unique Genocide?’, Journal of
Genocide Research 2:3 (2000), 411–30.
106 dan stone
other words, barbaric actions are justified for fear of being subjected to barbaric
actions. Germans in South West Africa (Namibia) ‘did not commit massacres in
the colonies because they were in a strong position and had the power to decide on
life or death of the indigenous population. On the contrary, German settlers felt
unsafe and were afraid to lose their existence.’11 In some cases, as in Rwanda, a
history of Hutu–Tutsi conflict from at least 1959 provided the background to
genocide. In the Ottoman Empire, small numbers of Armenians joined revolution-
ary movements that defied the state.12 Yet in none of these cases was it necessary for
the perpetrators to respond by seeking to slaughter the targeted population. What
mobilized them to do so, what exacerbated the sense of threat to the point at which
genocide became a viable and acceptable option, was fear underpinned by memo-
ry: of former oppression or supposed treason. Specifically, collective memories of
past suffering are almost always brought to bear on current crises, lending them
cultural meaning—the weight of dead ancestors weighing on the minds of the
living—and imbuing them with added ferocity. Memory fuels genocide.13
Stalin’s Soviet Union and Pol Pot’s Cambodia both illustrate the point. In the
former, the construction of the ‘Kulak’, which began with Stolypin’s reforms before
1917, revived fears of starvation and social conflict. Belief that peasants were
hoarding food, which would lead to death on a massive scale for urban dwellers,
then permitted massive oppression.14 And in the latter, Khmer Rouge support was
massively boosted by the effects of American bombing in the early 1970s. The
response to this attack does not explain the ferocity of the ‘auto-genocide’ between
1975 and 1979, but memories of French colonial wars, Prince Norodom Sihanouk’s
contempt for the majority rural population, and the age-old fear of the Vietnamese
certainly drove many ordinary Cambodians into the arms of the Khmer Rouge, as
did the regime’s revival of the grandeur of the Angkorian dynasty. As Ben Kiernan
notes, ‘The total reshaping of Cambodia under Pol Pot may be said to demonstrate
the power of a myth.’15
11 Dominik J. Schaller, ‘From Conquest to Genocide: Colonial Rule in German Southwest Africa
and German East Africa’, in A. Dirk Moses (ed.), Empire, Colony, Genocide: Conquest, Occupation, and
Subaltern Resistance in World History (New York: Berghahn Books, 2008), 311.
12 Donald Bloxham, The Great Game of Genocide: Imperialism, Nationalism and the Destruction of
the Ottoman Armenians (Oxford: Oxford University Press, 2005).
13 See Mark Levene, Genocide in the Age of the Nation State. vol i: The Meaning of Genocide
(London: I. B. Tauris, 2005), 196–202.
14 Terry Martin, ‘The Origins of Soviet Ethnic Cleansing’, Journal of Modern History 70:4 (1998),
813–61; Nicolas Werth, ‘The Crimes of the Stalin Regime: Outline for an Inventory and Classification’,
in Dan Stone (ed.), The Historiography of Genocide (Basingstoke: Palgrave Macmillan, 2008), 400–19;
Werth in this volume.
15 Ben Kiernan, ‘Myth, Nationalism and Genocide’, Journal of Genocide Research 3:2 (2001), 190. See
also idem, ‘Serial Colonialism and Genocide in Nineteenth-Century Cambodia’, in Moses (ed.),
Empire, Colony, Genocide, 205–28; idem, ‘Roots of Genocide: New Evidence on the US Bombardment
of Cambodia’, Cultural Survival Quarterly 14:3 (1990); David P. Chandler, ‘Seeing Red: Perceptions of
Cambodian History in Democratic Kampuchea’, in David Chandler and Ben Kiernan (eds), Revolution
The Rwandan example is equally full of such fears and fantasies, based on the
memory of Hutu–Tutsi conflict from at least the Hutu Revolution of 1959 if not from
the period of colonial rule (first German, then Belgian) from the late nineteenth
century. Tutsi refugees and their children actively kept alive the memory of the land
they had left (like Hutu refugees from Burundi in Tanzania),16 so that even those
young members of the RPF who had been born in Uganda and had never seen
Rwanda felt that they were ‘returning home’ in 1994. And the memory of the colonial
period, in which minority Tutsi domination was established according to the warped
racial logic of the colonizers, was mobilized by Hutu extremists in the run-up to the
genocide, especially as the framework for peace established by the Arusha Accords
started collapsing.17 Here the point about memory not as an organic phenomenon
but as a key component of political power is especially clear. For although there had
always been tensions between Hutus and Tutsis in Rwanda since the colonial period,
when the Belgian authorities institutionalized the distinction as ‘racial’,18 there was
nothing like a permanent state of war between the two ‘communities’, which were,
after the post-revolutionary violence of the early 1960s, in fact thoroughly mixed.
Only with the threat of war did Hutu extremists revitalize the memory of pre-1959
Rwandan society, dominated by the Tutsi minority, and whip up fear among the
Hutu population that they should eliminate the Tutsis because otherwise this same
fate would be reserved for them. Indeed, as recent research shows, the speed with
which certain parts of the country threw themselves into participating in genocide
was determined less by the reception of infamous propaganda such as the ‘Hutu Ten
Commandments’, Kangura magazine or Radio Télévision Libres des Milles Collines,
than affinity to the ruling MRND party, proximity to the front line and fear of the
approaching RPF.19 And, indeed, the RPF made equally effective use of collective
memories of expulsion and exile, with violent results both during and after the
genocide. Since the RPF took power, the government has come under increasing
and Its Aftermath in Kampuchea: Eight Essays (New Haven, CT: Yale University Southeast Asia Studies,
1983), 34–56; Karl D. Jackson, ‘Intellectual Origins of the Khmer Rouge’, in idem (ed.), Cambodia 1975–
1978: Rendezvous with Death (Princeton, NJ: Princeton University Press, 1989), 241–50.
16 Liisa Malkki, Purity and Exile: Violence, Memory and National Cosmology among Hutu Refugees
in Tanzania (Chicago: University of Chicago Press, 1995).
17 For example, Nigel Eltringham, ‘ “Invaders Who Have Stolen the Country”: The Hamitic
Hypothesis, Race and the Rwandan Genocide’, Social Identities 12:4 (2006), 425–46; René Lemarchand,
‘Exclusion, Marginalization and Political Mobilization: The Road to Hell in the Great Lakes’, University of
Copenhagen Centre of African Studies Occasional Paper (March 2000).
18 Mahmood Mamdani, When Victims Become Killers: Colonialism, Nativism, and the Genocide in
Rwanda (Princeton: Princeton University Press, 2001); Edith R. Sanders, ‘The Hamitic Hypothesis’,
Journal of African History 10:4 (1969), 512–32.
19 Scott Straus, The Order of Genocide: Race, Power, and War in Rwanda (Ithaca, NY: Cornell
108 dan stone
scrutiny by Western scholars who have grown suspicious of its ‘harmonising per-
spective on pre-colonial society and history’. The fear that Rwandan memories of
both the pre-colonial period and the 1994 genocide are being instrumentalized—for
example, by labelling all Hutu refugees as génocidaires or by employing guilt
discourses in the international arena—not only maintains RPF power but ‘perpe-
tuates violence in the Great Lakes.’20
Perhaps the most infamous example of such memory mobilization is the speech
given by Slobodan Milošević in 1989 at the site of the Battle of Kosovo Polje that took
place 600 years earlier on 28 June 1389. That battle (and that date—also the day of
Gavrilo Princip’s shooting of Archduke Franz Ferdinand in 1914) is ingrained into
Serbian memory as a moment of military defeat at the hands of the Turks, but a
moment of moral victory, on the basis of Knez Lazar choosing a heavenly instead of
an earthly kingdom for the Serbs. As well as confirming the Serb nation’s place in the
divine realm, the myth established the continuity of the Serb nation across the
centuries and confirmed Serbia’s right to its ancestral lands in Kosovo.21 It was also
the source of the ‘betrayal syndrome’—Serb allegations that Muslims in Yugoslavia
are ‘that part of themselves which betrayed the “faith of their forefathers.”’22 Milo-
šević’s speech is regularly cited as one of the key moments in his rise to power, and the
use of the legend of the battle a central component in his ethno-nationalist arsenal
and in the building of a nationalist consensus in Serbia. Although its significance can
be overstated, this manipulation of Serbian national memory—which of course
required grassroots activity to operationalize it, not Milošević alone—is key to
understanding the ‘ethnic cleansing’ that accompanied the Yugoslav wars of the
1990s and, especially, the violent efforts to expel ethnic Albanians from Kosovo at a
point when Serbia was already isolated as a pariah state in the eyes of the ‘interna-
tional community’. Extremists prevailed over moderates in Serbia because they
persuaded a large enough constituency that ‘the powerful can fear the weak.’23
More important even than the myth of Kosovo, which represents Serbian ‘deep
memory’, was the memory of what had happened in World War II. In the 1990s, the
20 Johan Pottier, Re-Imagining Rwanda: Conflict, Survival and Disinformation in the Late Twentieth
Century (Cambridge: Cambridge University Press, 2002), 130. See also Eric Stover and Harvey M.
Weinstein (eds), My Neighbor, My Enemy: Justice and Community in the Aftermath of Mass Atrocity
(Cambridge: Cambridge University Press, 2004); Gérard Prunier, Africa’s World War: Congo, the
Rwandan Genocide, and the Making of a Continental Catastrophe (New York: Oxford University
Press, 2008).
21 Florian Bieber, ‘Nationalist Mobilization and Stories of Serb Suffering: The Kosovo Myth from
600th Anniversary to the Present’, Rethinking History 6:1 (2002), 95–110; G. G. Raymond and S. Bajic-
Raymond, ‘Memory and History: The Discourse of Nation-Building in the Former Yugoslavia’,
Patterns of Prejudice 31:1 (1997), 21–30; Jasna Dragović-Soso, ‘Saviours of the Nation’: Serbia’s
Intellectual Opposition and the Revival of Nationalism (London: C. Hurst, 2002).
22 Milica Bakić-Hayden, ‘Nesting Orientalisms: The Case of Former Yugoslavia’, Slavic Review 54:4
(1995), 927.
23 Anthony Oberschall, ‘The Manipulation of Ethnicity: From Ethnic Cooperation to Violence and
War in Yugoslavia’, Ethnic and Racial Studies 23:6 (2000), 982–1001.
self-identification of Serbian and Croatian paramilitaries as Chetniks and Ustashe,

respectively, was a conscious echo of the war, when ‘Independent Croatia’—in
reality, a Nazi puppet state under the leadership of the clerico-fascist collaborator
Ante Pavelić—was responsible for the murder of tens of thousands of Serbs, Jews,
and Romanies. No serious historian doubts that Serbs were subjected to a genocidal
onslaught under the rule of Nazi-protected Croatia, but the manipulation of the
figures of the dead in the 1980s and 1990s was a major contributor to the worsening of
relations between the two major components of the Yugoslav federation. Croatia’s
neo-fascist president, Franjo Tudjman, was not only a Holocaust denier but a
belittler of Serb suffering during World War II, and Serbian historians and politicians
regularly exaggerated the numbers killed at Jasenovac and elsewhere in order to
spread fear throughout the Serbian population (especially outside of the borders of
Serbia) as Yugoslavia was breaking apart. A figure of 700,000 Serb deaths at Jasenovac
was commonly heard in the 1980s, when the true figure is likely to have been about
100,000. This strategy was highly effective, as fear of becoming victims of genocide
divided previously mixed communities into ethnically separate groups: ‘Everyone
was traumatized by all the talk of World War Two atrocities,’ wrote Bogdan Denitch,
‘even those who had seemed immune to nationalism.’24 Reliable figures of the dead
are still hard to come by, though the work of Tomislav Dulić, Robert M. Hayden, and
others has done much to bring clarity to this fraught issue—but scholarship alone is
of course insufficient to quell ultranationalist ideologies.25
The Holocaust can also to some extent be seen through this lens. Dirk Moses
argues that the Holocaust should be understood using a framework in which
genocide is seen as a combination of colonial expansion, security fears, and subal-
tern revenge. Hitler drew on the overseas colonial experience, especially in India and
North America, for inspiration for his own vision of a colonized Europe. The
treatment of Ukrainians, Poles, and other conquered nations certainly conforms
to this colonial pattern, in which the ‘natives’ were to become a reservoir of slave
labour. And the murder of the Jews, according to Moses, was in part a subaltern
genocide, through which Hitler aimed to ‘emancipate’ Germany from perceived
‘foreign occupation’, that is, Jewish rule. Thus, whilst Slavic populations were
24 Bogdan Denitch, Ethnic Nationalism (Minneapolis: University of Minnesota Press, 1996), 81,
cited in Oberschall, ‘The Manipulation’, 990.
25 Robert M. Hayden, ‘Mass Killings and Images of Genocide in Bosnia, 1941–5 and 1992–5’, in
Stone (ed.), The Historiography of Genocide, 487–516; idem, ‘Recounting the Dead: The Rediscovery
and Redefinition of Wartime Massacres in Late- and Post-Communist Yugoslavia’, in Ruby S. Watson
(ed.), Memory, Opposition and History under State Socialism (Santa Fe: School of American Research
Press, 1994), 167–84; Tomislav Dulić, Utopias of Nation: Local Mass Killings in Bosnia and Herzegovina,
1941–42 (Uppsala: Uppsala University Press, 2005). On the figures from the 1990s, see Ewa Tabeau and
Jakub Bijak, ‘War-Related Deaths in the 1992–1995 Armed Conflicts in Bosnia and Herzegovina: A
Critique of Previous Estimates and Recent Results’, European Journal of Population, 21 (2005), 187–215;
Research and Documentation Center, Sarajevo, Human Losses in Bosnia and Herzegovina 91–95
(CD-Rom, 2006).
110 dan stone
regarded as Untermenschen (subhumans), suitable for enslavement, the Jews were a

source of fear, for they sought to take over the world, and their elimination was a
project of ‘national liberation’.26 Genocide, in Moses’ formulation, ‘is as much an act
of security as it is racial hatred’.27 It is worth noting that this stress on Nazi fears of
Jews—as opposed to the standard narrative that stresses Nazi racial theory and the
need to rid the world of inferior ‘non-Aryans’—provides common ground between
scholars who incorporate the Holocaust into the new comparative genocide frame-
work, and those who argue that the racial paradigm at the heart of the Nazi
Weltanschauung ultimately owed less to race science than to a paranoid political
conspiracy theory. This view suggests that the Nazis were not so much driven by
their sense of superiority as by their fear of the power of ‘the Jew’. Hence the lengths
to which Goebbels went in his propaganda output to convince the German public
that ‘The Jews are guilty of everything!’28 The source of this sense of existential
threat was the ‘stab-in-the-back’ legend from 1918, the belief that Germany lost the
Great War because the Jews had betrayed the country. Michael Geyer notes that ‘The
rhetoric of Endkampf [final battle] found its most potent enemy in the figure of the
Jew.’29 Indeed, the feeding through of the memory of 1918 into Nazi ideology is a
textbook example of the power of traumatic memory, of what Mark Levene calls ‘the
perpetrator’s “never again” syndrome’. ‘They should not have staged 9 November
1918 with impunity’, fumed Hitler to the Czech foreign minister in 1939, ‘That day
shall be avenged . . . The Jews shall be annihilated in our land.’30
P O S T -G E N O C I DA L T R AU M AT I C M E M O RY
................................................................................................................
What happens after genocide? When communities are devastated, often all that is
left is memory, and that a ‘memory shot through with holes’.31 Thus survivors turn
26 A. Dirk Moses, ‘Empire, Colony, Genocide: Keywords and the Philosophy of History’, in Moses
(ed.), Empire, Colony, Genocide, 34–40.
27 A. Dirk Moses, ‘Moving the Genocide Debate beyond the History Wars’, Australian Journal of
Politics and History 54:2 (2008), 264.
28 See Jeffrey Herf, The Jewish Enemy: Nazi Propaganda During World War II and the Holocaust
(Cambridge, MA: Belknap Press of Harvard University Press, 2006). Goebbels cited at 209. See also
Doris L. Bergen, ‘Instrumentalization of Volksdeutschen in German Propaganda in 1939: Replacing/
Erasing Poles, Jews, and Other Victims’, German Studies Review 31:3 (2008), 447–70 for an example
of the manipulation of fears of German victimization at the hands of Poles.
29 Michael Geyer, ‘Endkampf 1918 and 1945: German Nationalism, Annihilation, and Self-
Destruction’, in Alf Lüdtke and Bernd Weisbrod (eds), No Man’s Land of Violence: Extreme Wars in
the Twentieth Century (Göttingen: Wallstein, 2006), 47.
30 Levene, The Meaning of Genocide, 197.
31 Henri Raczymow, ‘Memory Shot Through With Holes’, Yale French Studies 85 (1994), 98–105.
inwards, and focus on themselves and the need for familial and community repair.
This process is intrinsically related to memory, in the production of memorial
books and monuments and, in interacting with the wider world, in attempts to
bring what happened to general notice and to bring perpetrators to justice. If
collective memory is essential for mobilizing perpetrators, it also underpins
attempts to commemorate genocide in its immediate aftermath and to advocate
on behalf of survivors in their quest for justice.
A large literature now exists on reparations, compensation, restitution, war
crimes trials, truth commissions, and the developing international law on genocide
since the founding of the International Criminal Court in 1999. In numerous
contexts, from Guatemala to Poland, national commissions of inquiry have been
set up to inquire into genocidal pasts. Austria’s amnesia as regards its Nazi past was
only an extreme example of a common phenomenon, and most European states
have now ‘discovered’ the fact that Nazism and the Holocaust were part of their
histories too. Since the Stockholm Forum in 2000, many European states have
commissioned official investigations into their experience of and, often, collabora-
tion with Nazi occupation and genocide. For example, the question of the extent of
Nazi looting and of restitution for victims of the Holocaust has been an area of
remarkable scholarly activity since the end of the Cold War.32 Dan Diner has
highlighted the relationship between memory and restitution in the light of the
move to incorporate Holocaust Memorial Day into the European calendar and
European cultural identity: ‘a basic anthropological assumption’ exists, thinks
Diner, that presumes an ‘organic interconnection between restituted property
rights and the evocation of past memories, or vice versa: Restitution of property
as the result of recovered memory.’33 As he rightly notes, this link between memory
and property is both plausible and problematic.
The issue of property and restitution provides a link between issues of memory
that are victim community-focused and those aimed at the wider world. Perhaps
post-genocide trials represent the purest form of the latter. The image of the
twenty-two leading Nazis in the dock at Nuremberg is one of the most memo-
rable of the twentieth century, and the memory of Nuremberg informs the cur-
rently developing international law on genocide and human rights.34 Issues of
32 Martin Dean, Robbing the Jews: The Confiscation of Jewish Property in the Holocaust 1933–1945
(Cambridge: Cambridge University Press, 2008); Martin Dean, Constantin Goschler and Philipp Ther
(eds), Robbery and Restitution: The Conflict over Jewish Property in Europe (New York: Berghahn
Books, 2007).
33 Dan Diner, ‘Restitution and Memory: The Holocaust in European Political Cultures’, New
German Critique 90 (2003), 39–40. See also Dan Diner and Gotthard Wunberg (eds), Restitution
and Memory: Material Restoration in Europe (New York: Berghahn Books, 2007).
and Memory (Oxford: Oxford University Press, 2001); Mark Mazower, ‘An International Civilization?
Empire, Internationalism and the Crisis of the Mid-Twentieth Century’, International Affairs 82:3
(2006), 553–66.
112 dan stone
compensatory and/or corrective justice, as well as penal/retributive justice, are in

evidence in different sorts of trials, depending on whether these deal with repara-
tions or punishment. The Eichmann Trial exemplifies a deliberately orchestrated
attempt to bring Holocaust memory into the centre of Israeli public (as opposed to
private) consciousness, and the significance of post-genocide trials for memory
work is not to be underestimated. Even though it is widely acknowledged that the
punishment in such cases can never fit the crime—‘The Nazi crimes, it seems to
me, explode the limits of the law; and that is precisely what constitutes their
monstrousness’35—the impact of such trials explains why they have been avoided
in so many instances, from France to Cambodia, by the use of delaying tactics.
Numerous scholars identify shortcomings in the UNGC, and some assert that these
shortcomings have negative consequences for the establishment of collective mem-
ories of genocide;36 but there is a good reason why the authorities often resist and
place obstacles in the way of post-genocide trials.
When memory is the subject, the focus of attention is usually on commemora-
tive practices, monuments, and museums. An enormous body of research now
exists on Holocaust memorials and museums, of which there are many throughout
the world.37 But it is not only the Holocaust that provides material to test James E.
Young’s claim that monuments propagate an ‘illusion of common memory’. The
desire to memorialize traumatic events such as the Holocaust ‘may actually spring
from an opposite and equal desire to forget them’, since the assumption that the
monument is always there tends to encourage a lack of engagement with the
issues.38 A casual stroll through any major city, most of whose monuments remain
unnoticed and, for the inhabitants, unidentifiable, suggests that Young has a point.
Apart from the question of whether genocide memorials too readily take their
cue from representations of the Holocaust,39 it is worth considering what forms of
memory genocide memorials and museums are meant to encourage. One scholar
suggests that ‘fear of denial and scarcity of resources has resulted in the most
graphic genocide memorial in history: that of Murambi’ in Rwanda.40 At the
school where the massacre of several thousand Tutsis took place, the remains of
35 Hannah Arendt to Karl Jaspers, 17 August 1946, in Arendt/Jaspers Correspondence 1926–1969, ed.
Lotte Kohler and Hans Saner (San Diego: Harcourt Brace, 1992), 54.
36 For example, Caroline Fournet, The Crime of Destruction and the Law of Genocide: Their Impact
on Collective Memory (Aldershot: Ashgate, 2007); Gerry Simpson, Law, War and Crime: War Crimes
Trials and the Reinvention of International Law (Cambridge: Polity, 2007).
37 Dan Stone, ‘Memory, Memorials and Museums’, in idem (ed.), The Historiography of the
Holocaust (Basingstoke: Palgrave Macmillan, 2004), 508–32.
38 James E. Young, The Texture of Memory: Holocaust Memorials and Meaning (New Haven, CT:
Yale University Press, 1993), 6–7.
39 See David S. MacDonald, Identity Politics in the Age of Genocide: The Holocaust and Historical
Representation (London: Routledge, 2008).
40 William F. S. Miles, ‘Third World Views of the Holocaust’, Journal of Genocide Research 6:3
(2004), 388.
the dead were left as the monument, giving rise to a ‘traumatic silence’ amongst
visitors. The same is true of the bones that function as memorials at Nyamata,
Nyarabuye, and Ntarama, where ‘the function of the memorials is not to obtain
scientific evidence, but rather to produce an experience of memory.’41 In Cambo-
dia, the Tuol Sleng Museum of Genocidal Crimes and the Choeung Ek ‘killing
fields’ site serve a similar function. They also aim to preserve the memory of
genocide, but do so by shocking visitors (mostly Western tourists), partly by
deliberately borrowing a Holocaust-inspired form of representation, and partly
by instilling a new national narrative.42 And given that most of the Khmer Rouge
leaders have escaped the trials that belatedly began in November 2007 with the trial
of Kaing Guek Eav (the head of Tuol Sleng, known as Comrade Duch), their
memorial function is somewhat soured. In both countries, however, the signifi-
cance of the genocides means that the public display of body parts has been
permitted, contrary to usual custom, although one should bear in mind that ‘the
maintenance of a site to communicate its cursedness or ruination is itself a
sustained act of intervention.’43 The sheer mass of bones in these monuments
provokes the shock and horror that are appropriate responses to genocide, but
their anonymity means that they also recapitulate the logic of genocide: the
reduction of individual human beings to representatives of a (perpetrator-defined)
group. Hence the importance of local memorials and commemorative festivals in
Cambodia.44 And hence the great significance of naming in general, as seen in
many memorial practices, from the post-Holocaust yizker-bikher (memorial
books) to the recovery of names in Spain’s Todos los nombres project.45
Remembering genocide, however, is only one side of the coin of responding to
such traumatic events. The other is willed amnesia. The conscious turn to memory
tends—though this is not always true—to require the passage of time, for in the
immediate aftermath of genocide the scars are still too deep. Especially in instances
where former perpetrators and surviving victims must live together in close
41 Sara Guyer, ‘Rwanda’s Bones’, boundary 2 36:3 (2009).

42 Judy Ledgerwood, ‘The Cambodian Tuol Sleng Museum of Genocidal Crimes: National
Narrative’, in David E. Lorey and Willian H. Beezley (eds), Genocide, Collective Violence, and
Popular Memory: The Politics of Remembrance in the Twentieth Century (Wilmington, DE: Scholarly
Resources, 2002), 103–22. See also Burcu Münyas, ‘Genocide in the Minds of Cambodian Youth:
Transmitting (Hi)stories of Genocide to Second and Third Generations in Cambodia’, Journal of
Genocide Research 10:3 (2008), 413–39; David P. Chandler, ‘Cambodia Deals with its Past: Collective
Memory, Demonisation and Induced Amnesia’, Totalitarian Movements and Political Religions 9:2–3
(2008), 355–69.
43 Paul Williams, ‘Witnessing Genocide: Vigilance and Remembrance at Tuol Sleng and Choeung
Ek’, Holocaust and Genocide Studies 18:4 (2004), 242. On trials, see Jörg Menzel, ‘Justice Delayed or Too
Late for Justice? The Khmer Rouge Tribunal and the Cambodian “Genocide” 1975–79’, Journal of
44 Rachel Hughes, ‘Memory and Sovereignty in Post-1979 Cambodia: Choeung Ek and Local
Genocide Memorials’, in Susan E. Cook (ed.), Genocide in Cambodia and Rwanda: New Perspectives
(New Brunswick, NJ: Transaction, 2006), 257–80.
45 Guyer, ‘Rwanda’s Bones’.
114 dan stone
proximity, closing off memory, or at least trying to do so, is a meaningful way of

dealing with the past. In Rwanda, for example, what is striking about Susanne
Buckley-Zistel’s interviews with people from across the country’s diverse popula-
tion is that, whilst they often referred to the 1994 genocide, ‘the causes of the
genocide and the decades of tension between Hutu and Tutsi were ignored.’46
Precisely the years of tension from 1959 onwards that saw the mobilization of
memory in the early 1990s were the years that had to be ‘forgotten’ (that is to say,
left undiscussed), rather than the events of the genocide itself. Gacaca trials can
address issues of who did what in the context of the genocide, but leaves the
underlying causes unaddressed. Only time will tell whether the Rwandan govern-
ment’s attempt to switch the country from a Francophone to an Anglophone
position, to remove ethnic markers from ID cards, to rewrite Rwandan history,
and to advocate local as well as international forms of justice will help Rwandans to
overcome these conflict-ridden memories.
In Bosnia, Cornelia Sorabji shows that memories of traumatic events continue
‘to affect the social fabric’, possibly sustaining the sort of hostility that fuelled
conflict in the first place.47 Sorabji correctly notes that the risk of analysing
memory as a carrier of conflict is that it serves to perpetuate ‘ancient hatreds’
style arguments, which suggest that war in the Balkans is a more or less natural
condition. Thus, she proposes to situate individuals and their memories—‘real’ or
‘transmitted’—into the context of the politics of memory, that is, the broader
framework of competing narratives at group or state level that seek to ‘channel’
people’s memories in certain ways. For since ‘collective memory’ is not an organic
process (there is no group mind), it follows that the inter-relationship between
individuals (‘memory users’) and the group (‘memory makers’) needs to be
analysed. One should not assume ‘that human minds are endlessly manipulable
and that schooling or the broadcasting of nationalistic commemorative ceremonies
can fundamentally alter personal memories of strongly emotional, life-changing
events such as violent bereavement.’48
Of course, one of the characteristics of traumatic memory is that it cannot be
suppressed at will. It is by its very nature a memory that returns unexpectedly and
uncontrollably to haunt individual victims and post-genocide societies. There is no
need for memories of genocide to be ‘recovered’—in the dubious manner of
childhood abuse cases of the 1980s—since it has never gone away in the first
place. Many scholars are now rightly critical of the view, fashionable in the 1990s
especially in literary studies, that ‘traumatic memory’ is a widely applicable concept.
46 Susanne Buckley-Zistel, ‘Remembering to Forget: Chosen Amnesia as a Strategy for Local

Coexistence in Post-Genocide Rwanda’, Africa 76:2 (2006), 131.
47 Cornelia Sorabji, ‘Managing Memories in Post-War Sarajevo: Individuals, Bad Memories, and
New Wars’, Journal of the Royal Anthropological Institute, NS, 12 (2006), 1–18.
48 Sorabji, ‘Managing Memories’, 2.
The idea that whole societies can be traumatized has been subjected to serious
criticism, so that what we are generally left with is a more or less appropriate
metaphor, not a concept that carries any of the precise, clinical meaning that it does
when applied to individuals (when used carefully, and not just in the vernacular, as
in ‘what a traumatic day that was’). As Kansteiner notes, ‘none of the existing
concepts of Holocaust trauma is well suited to explain the effects of Holocaust
representations on individuals or collectives who encounter the Final Solution only
as a media event for educational or entertainment purposes.’49 Still, in the case of
societies that have experienced genocide, we are facing a situation where the
concept of traumatic memory, if it has any use at all, is about as applicable as
one can expect. This is why I noted at the outset that genocide is less amenable to
willed amnesia than other events. What one actually sees, for example, in the cases
of Bosnia or Rwanda mentioned above, is a form of repression, rather than a
‘healthy forgetting’ in the manner of Nietzsche. And what is repressed sooner or
later returns, as we currently see the memory of the post-Civil War ‘repression’—a
somewhat coy term for what some historians actually consider a genocidal on-
slaught—of the Nationalists’ enemies in Spain.50 The current tension in Bosnia and
the desperate situation in the Democratic Republic of Congo, where estimates are
that more than 5 million people have died in the post-1994 regional war, indicate
that the politics of post-genocidal memories are matters of life and death.
C O M M E M O R AT I O N AND M E M O RY C O N F L I C T S
................................................................................................................
In February 2008, Kevin Rudd, the new Australian Prime Minister, made a decisive
break with the politics of John Howard’s conservative administration by making a
public apology to the country’s indigenous people for the suffering endured by the
‘stolen children’ and their families. This policy, which began in the early twentieth
century and lasted until the 1960s, removed ‘half-caste’ children from Aboriginal
communities, bringing them up in separated institutions with the explicit aim
of assimilating Aborigines to ‘white’ culture. This was a change from the
49 Wulf Kansteiner, ‘Testing the Limits of Trauma: The Long-Term Psychological Effects of the
Holocaust on Individuals and Collectives’, History of the Human Sciences 17:2–3 (2004), 97; idem,
‘Genealogy of a Category Mistake: A Critical Intellectual History of the Cultural Trauma Metaphor,
Rethinking History 8:2 (2004), 193–221.
50 Helen Graham, ‘The Memory of Murder: Mass Killing, Incarceration and the Making of Francoism’,
in Alison Ribeiro de Menezes, Roberta Quance and Anne L. Walsh (eds), Guerra y memoria en la España
contemporánea/War and Memory in Contemporary Spain (Madrid: Verbum, 2009, 29–49).
116 dan stone
early-twentieth-century approach of biological absorption, or ‘breeding out the

black’, which aimed to prevent white Australia from being threatened—so the fear
went—by ‘a large black population which may drive out the white’.51 But whilst the
official programme of biological absorption came to an end around 1940, the
policy of child removal continued for several decades, devastating Aboriginal
communities and leading Sir Ronald Wilson to proclaim in his 1997 Bringing
Them Home report that the policy constituted genocide under article IIe of the
UNGC. Whether this was an appropriate designation is in this context not the
point (Rudd, incidentally, denies that it was genocide), so much as the fact that the
subsequent furore revealed the way in which controversy about genocidal origins
haunts ‘national memory’ generations after the cessation of frontier conflict.
The perpetration of genocide requires the mobilization of memory, as does its
punishment, though in the latter case there is a prima facie argument that ‘memory
mobilizes itself ’. Post-genocidal conflicts over memory, especially national memo-
ry, reveal another aspect of the question: memory can intervene in national politics
in unexpected ways and present challenges to long-held and cherished national
narratives. This is particularly true of settler societies, and is best illustrated by the
Australian case. With the emergence of what its opponents pejoratively called
‘black armband history’, debates over Australian history overshadowed contempo-
rary political debates concerning how best to deal with troubled Aboriginal com-
munities. Conservative historians, most notably Keith Windschuttle, charged
‘politically correct’ historians not only with failing to appreciate the true nature
of frontier conflict, in which mutual incomprehension rather than genocidal intent
was at work, but also with deliberately exaggerating the numbers of Aborigines
killed in massacres.52 Even official efforts at reconciliation were ‘framed in
nation-building language which implicitly refused to accommodate indigenous
aspirations of difference’.53 The ‘history wars’ that followed the publication of
Windschuttle’s revisionist book have been described as an ‘Australian Historiker-
streit’, a designation that is revealing, since the West German debate about the
uniqueness of the Holocaust that took place in the 1980s broke no new historical
ground but was fundamental to the self-image of the Federal Republic. So in
Australia, debates about how best to describe the past go to the heart of national
narratives. The challenge to the Australian story of mates pulling together to create
the ‘lucky country’ is one that did not sit well with the cultural politics of the
Howard government, which was not open to the fact that historians of early Australia
51 Quoted in Robert Manne, ‘Aboriginal Child Removal and the Question of Genocide, 1900–1940’,
in A. Dirk Moses (ed.), Genocide and Settler Society: Frontier Violence and Stolen Indigenous Children in
Australian History (New York: Berghahn Books, 2004), 229, 237; Pat O’Malley, ‘Gentle Genocide: The
Government of Aboriginal Peoples in Central Australia’, Social Justice 21:4 (1994), 46–65.
52 Keith Windschuttle, The Fabrication of Aboriginal History (Sydney: Macleay Press, 2002).
53 Damien Short, ‘Reconciliation, Assimilation, and the Indigenous Peoples of Australia’,
International Political Science Review 24:4 (2003), 506.
were not arguing that the colonization of Australia was the same as the Holocaust,
only that the similarities of the perpetrators’ discourses of race and security in both
cases ought to offer food for thought, particularly where current-day policies
towards Aborigines are concerned.54 But whilst debate rages in Australia—unlike
in Germany—as to whether the country should be understood as a ‘post-genocidal
society’, the fact that the colonization process was ‘objectively lethal’ for the Abor-
igines continues to be overlooked.55 Irrespective of the statistics and other facts being
debated by historians (and here the comparison with the Historikerstreit is uncon-
vincing, for in West Germany no historians questioned whether genocide had
occurred), the bigger point is that Australian collective memory was being decon-
structed and reconstructed anew or, for conservative historians, being undermined
by subversives bent on ridiculing national heritage.
Even long after genocide has taken place, memory wars can erupt when group
narratives are felt to be under threat. The history of nation-building is inseparable
from the ‘memories’ that nations create, in the shape of the narratives or monu-
ments they construct. Indeed, collective memory does not emerge after the process
has come to an end but is an essential part of the process whereby a group
constitutes itself as a group; as Jens Bartelson notes, ‘the coincidence of state and
nation that we normally take to be the very culmination of a successful process of
state formation had virtually been remembered into existence.’56 The motives of
memory, as James Young reminds us, are never pure.57
It is hardly surprising, then, that especially in societies founded on colonial
settlement, challenges to positive national narratives are considered problematic.
In Australia, whilst the official discourse has changed since the Rudd administration
took office, historians such as Tony Barta fear that the ‘public conversation’ will
remain dominated by a ‘decent disposal’ of the difficult questions. Nevertheless, by
comparison with the United States, where the genocide question is still almost
wholly ignored, even by prominent scholars of genocide, or Israel, where the
memory of the Holocaust continues to poison relations with the Palestinians with
devastating consequences, at least in Australia these memory conflicts are being
54 Moses, ‘Moving the Genocide Debate’, 254–5. See also Patrick Brantlinger, ‘ “Black Armband”
versus “White Blindfold” History in Australia’, Victorian Studies 46:4 (2004), 655–74; Neil Levi, ‘ “No
Sensible Comparison”? The Place of the Holocaust in Australia’s History Wars’, History & Memory 19:1
(2007), 124–56; Andrew G. Bonnell and Martin Crotty, ‘Australia’s History under Howard, 1996–2007’,
Annals of the American Academy of Political and Social Science 617 (2008), 149–65.
55 A. Dirk Moses, ‘An Antipodean Genocide? The Origins of the Genocidal Moment in the
Colonization of Australia’, Journal of Genocide Research 2:1 (2000), 89–106.
56 Jens Bartelson, ‘We Could Remember It for You Wholesale: Myths, Monuments and the
Constitution of National Memories’, in Duncan Bell (ed.), Memory, Trauma and World Politics:
Reflections on the Relationship between Past and Present (Basingstoke: Palgrave Macmillan, 2006), 51.
57 Young, The Texture of Memory, 2.
118 dan stone
articulated in the public sphere.58 ‘Memory wars’ have characterized the whole
world since the end of the Cold War, from Romania to Argentina, South Africa to
France. In post-genocidal societies, as we see in Bosnia or the DRC, such conflicts
are potentially destabilizing and certainly have the power not only to inspire a
cosmopolitan culture of human rights but also new outbursts of resentment and
revanchism. The only sure conclusion is that memory cannot be ignored and that
engagement with the issues—if not resolution of them—remains essential.
C O N C LU S I O N
................................................................................................................
In a key article on the historical study of memory, Alon Confino asks: ‘if the study
of memory focuses creatively on how people construct a past through a process of
appropriation and contestation, is the real problem not, perhaps, that people
construct the past by using the term “memory” at all?’59 There is, in other words,
a danger of studying a phenomenon (‘memory’) by taking it as its own explana-
tion. This problem, however, is not merely a methodological one of memory
studies but a reflection of the complex place that ‘memory’ holds in contemporary
societies. For memory is not simply synonymous with the way in which the past is
represented in the present; it is itself constitutive of the present. Memory and
identity go hand in hand.
Thus, irrespective of methodological problems, issues connected with memory
will continue to resonate. Exclusivist, exclusionary memories remain powerful in
many contexts; the generation of genocidal ideologies through the manipulation of
memory is as much a possibility as it ever was. Indeed, memory wars by no means
guarantee a peaceful resolution or mutually agreeable arbitration between compet-
ing versions of the past. As Peter Fritzsche notes, the reason that national memories
‘remain so resonant’ is ‘not because they are more true, but because the narratives
of collective guilt and collective victimization that they generate have the effect of
recognizing and commemorating individual suffering in socially meaningful, if
tendentious, ways.’60 ‘Memory studies’ is not an academic game, but an investiga-
tion into a phenomenon that can be as dangerous as playing with fire. For this
58 Tony Barta, ‘Decent Disposal: Australian Historians and the Recovery of Genocide’, in Stone
(ed.), The Historiography of Genocide, 296–322; Alfred A. Cave, ‘Genocide in the Americas’, in ibid.,
273–95; Avraham Burg, The Holocaust is Over, We Must Rise from its Ashes (Houndmills: Palgrave
Macmillan, 2008).
59 Alon Confino, ‘Collective Memory and Cultural History: Problems of Method’, American
Historical Review 102:5 (1997), 1403.
60 Peter Fritzsche, ‘The Case of Modern Memory’, Journal of Modern History 73:1 (2001), 117.
reason, memory cannot be avoided or swept aside. Despite the risks of perpetuat-
ing old divisions or reopening unhealed wounds, grappling with memory, espe-
cially after traumatic events like genocide, remains essential in order to remind the
victims that they are not the worthless or less than human beings that their
tormentors have portrayed them as. For nothing is more human, and thus more
geared towards the generation of meaning where meaning is otherwise absent (or
at least to ‘keeping watch over absent meaning’),61 than the broad spectrum of
practices that come under the heading of ‘memory’.
A C K N OW L E D G E M E N T
................................................................................................................
My thanks to Donald Bloxham, Becky Jinks, and Dirk Moses for their comments
on earlier versions of this chapter.
SELECT BIBLIOGRAPHY
Barayón, Ramón Sender, A Death in Zamora (Albuquerque: University of New Mexico
Press, 1989).
Bizot, François, The Gate (London: Harvill Press, 2003).
Confino, Alon, and Peter Fritzsche (eds), The Work of Memory: New Directions in the Study
of German Society and Culture (Urbana/Chicago: University of Illinois Press, 2002).
Eltringham, Nigel ed., ‘Identity, Justice and “Reconciliation” in Contemporary Rwanda’,
Special issue of the Journal of Genocide Research 11:1 (2009).
Gray, Peter, and Kendrick Oliver (eds), The Memory of Catastrophe (Manchester: Manche-
ster University Press, 2004).
MacDonald, David B., Identity Politics in the Age of Genocide: The Holocaust and Historical
Representation (London/New York: Routledge, 2008).
Ricoeur, Paul, Memory, History, Forgetting (Chicago/London: University of Chicago Press, 2004).
Rousso, Henry, The Vichy Syndrome: History and Memory in France since 1944 (Cambridge,
MA: Harvard University Press, 1991).
Stone, Dan (ed.), The Historiography of Genocide (Basingstoke/New York: Palgrave
Macmillan, 2008).
Suleiman, Susan Rubin, Crises of Memory and the Second World War (Cambridge. MA:
61 Maurice Blanchot, The Writing of the Disaster (Lincoln: University of Nebraska Press, 1986), 42.
part ii
.............................................................................................
INTERDISCIPLINARY
PERSPECTIVES
.............................................................................................
chapter 6
.............................................................................................
T H E L AW A N D
GENOCIDE
.............................................................................................
william a. schabas
Genocide is, first and foremost, a legal concept. Like many other terms—murder,
rape, theft—it is also used in other contexts and by other disciplines, where its
meaning may vary. Many historians and sociologists employ the term genocide to
describe a range of atrocities involving killing large numbers of people. But even in
law, it is imprecise to speak of a single, universally recognized meaning of genocide.
There is a widely accepted definition, first set out in article II of the 1948 Convention
for the Prevention and Punishment of the Crime of Genocide.1 Like most legal defini-
tions, its language is subject to various interpretations, and important controversies
remain about the scope of the concept even within the framework of what is a concise
and carefully worded definition. The crime of genocide has been incorporated
within the national legal systems of many countries, where national legislators
have imposed their own views on the term, some of them varying slightly or even
considerably from the established international definition. As a result, even in law,
one can speak of many definitions or interpretations of the concept of genocide.
The term itself was invented by a lawyer, Raphael Lemkin. He intended to fill a
gap in international law, as it then stood in the final days of the Second World War.
For more than two decades, Lemkin had been engaged at an international level in
attempts to codify new categories of international crimes involving atrocities
committed against vulnerable civilians. Even before Lemkin’s time, international
law recognized a limited number of so-called international crimes. As a general
1 Convention on the Prevention and Punishment of the Crime of Genocide (1951), 78 UNTS 277.
124 william a. schabas
rule, they were so designated not because of their shocking scale and extent, but for
more mundane reasons, namely because they escaped the territorial jurisdiction of
states. Piracy is the classic example, a crime committed on the high seas. Lemkin
and others argued from a different perspective, proposing the recognition of
international crimes where these represented serious human rights violations.
The beginnings of international prosecution for atrocities were already apparent
at the time of the First World War, when Britain, France, and Russia warned that
they would hold perpetrators to account for ‘these new crimes of Turkey against
humanity and civilization’. But the idea that a state could be held accountable for
atrocities committed against its own nationals remained extremely controversial,
and it was this gap in the law that Lemkin worked to fill. His initial proposal
evidenced a much broader concept of genocide than what was eventually agreed to
in the 1948 Convention. But Lemkin actively participated in the negotiations leading
to the Convention’s adoption, and while he would no doubt have hoped for a
somewhat different result, he cannot be detached from the Convention definition.
Indeed, following its adoption he campaigned aggressively for its ratification.
Lemkin’s famous proposal, contained in a chapter entitled ‘Genocide’ in his
book Axis Rule in Occupied Europe, called for the ‘prohibition of genocide in war
and peace’. Lemkin insisted upon the relationship between genocide and the
growing interest in the protection of peoples and minorities manifested in several
treaties and declarations adopted following the First World War. He noted the need
to revisit international legal instruments, pointing out particularly the inadequa-
cies of the Hague Convention of 1907, which he explained was ‘silent regarding the
preservation of the integrity of a people’. According to Lemkin,
the definition of genocide in the Hague Regulations thus amended should consist of two
essential parts: in the first should be included every action infringing upon the life, liberty,
health, corporal integrity, economic existence, and the honor of the inhabitants when
committed because they belong to a national, religious, or racial group; and in the second,
every policy aiming at the destruction or the aggrandizement of one of such groups to the
prejudice or detriment of another.2
GENOCIDE AND CRIMES AG A I N S T HUMANITY

................................................................................................................
The legal concept of genocide was forged in the crucible of post-Second World War
efforts to prosecute Nazi atrocities. Its development took place in conjunction with
Proposals for Redress (Washington, DC: Carnegie Endowment for World Peace, 1944), 90–3.
the law and genocide 125
that of other international crimes, especially crimes against humanity, with which
it bears a close but complex and difficult relationship. The development and
history of genocide as a legal concept cannot be properly understood without
considering the parallel existence of crimes against humanity.
Although the participants in the UN War Crimes Commission, established in
November 1943, and in the London Conference, which met from late June to early
August 1945 to prepare the Nuremberg trial of the major war criminals, opted to
use the term crimes against humanity in the prosecutions, they also employed the
word genocide as if it was more or less synonymous. In his ‘Planning Memoran-
dum distributed to Delegations at Beginning of London Conference, June 1945’,
where Justice Robert Jackson outlined the evidence to be adduced in the Nurem-
berg trial, he spoke of ‘[g]enocide or destruction of racial minorities and subju-
gated populations by such means and methods as (1) underfeeding; (2) sterilization
and castration; (3) depriving them of clothing, shelter, fuel, sanitation, medical
care; (4) deporting them for forced labor; (5) working them in inhumane condi-
tions.’3 The indictment of the International Military Tribunal charged the Nazi
defendants with ‘deliberate and systematic genocide, viz., the extermination of
racial and national groups, against the civilian populations of certain occupied
territories in order to destroy particular races and classes of people, and national,
racial or religious groups, particularly Jews, Poles, and Gypsies.’4 The term ‘geno-
cide’ was also used on several occasions by the prosecutors during the trial itself. Sir
David Maxwell-Fyfe, the British prosecutor, reminded one of the accused, von
Neurath, that he had been charged with genocide, ‘which we say is the extermina-
tion of racial and national groups, or, as it has been put in the well-known book of
Professor Lemkin, “a co-ordinated plan of different actions aiming at the destruc-
tion of essential foundations of the life of national groups with the aim of
annihilating the groups themselves.”’5 Lemkin later wrote that ‘[t]he evidence
produced at the Nuremberg trial gave full support to the concept of genocide.’6
Nevertheless, the Charter of the International Military Tribunal did not use the
word genocide, nor does it appear in the final judgment issued on 30 September
and 1 October 1946. The legal concept of crimes against humanity, as defined at
Nuremberg, suffered from a very serious limitation, in that it was confined to
atrocities committed in association with an aggressive war. This was quite inten-
tional on the part of those who drafted the legal provisions governing prosecutions,
especially the four great powers, the United States, the United Kingdom, France,
and the Soviet Union. Indeed, extending international law from classic war crimes
3 Report of Robert H. Jackson, United States Representative to the International Conference on

Military Trials (Washington, DC: US Government Printing Office, 1949), 6.
4 France et al. v. Goering et al., (1946) 22 International Military Tribunal (IMT) 45–6.
5 (1947) 17 IMT, 61. See also: (1947) 19 IMT 497, 498, 509, 514, 531.
6 Raphael Lemkin, ‘Genocide as a Crime in International Law’, American Journal of International
Law 41 (1947), 145, 147.
involving battlefield offences and various forms of persecution of civilians in an

occupied territory so that it would also cover atrocities committed by a govern-
ment against its own civilian population was not only novel and unprecedented, it
was also threatening to the very states who were organizing the prosecution. The
distinctions were set out quite candidly by the head of the United States delegation,
Robert Jackson, at a meeting of the London Conference on 23 July 1945:
It has been a general principle of foreign policy of our Government from time immemorial
that the internal affairs of another government are not ordinarily our business; that is to say,
the way Germany treats its inhabitants, or any other country treats its inhabitants is not our
affair any more than it is the affair of some other government to interpose itself in our
problems. The reason that this program of extermination of Jews and destruction of the
rights of minorities becomes an international concern is this: it was a part of a plan for
making an illegal war. Unless we have a war connection as a basis for reaching them, I would
think we have no basis for dealing with atrocities. They were a part of the preparation for
war or for the conduct of the war in so far as they occurred inside of Germany and that
makes them our concern.7
Speaking of the proposed crime of ‘atrocities, persecutions, and deportations on

political, racial or religious grounds’, which would shortly be renamed ‘crimes
against humanity’, Justice Jackson indicated the source of the lingering concerns
of his government:
[O]rdinarily we do not consider that the acts of a government toward its own citizens
warrant our interference. We have some regrettable circumstances at times in our own country
in which minorities are unfairly treated. We think it is justifiable that we interfere or attempt
to bring retribution to individuals or to states only because the concentration camps and
the deportations were in pursuance of a common plan or enterprise of making an unjust or
illegal war in which we became involved. We see no other basis on which we are justified in
reaching the atrocities which were committed inside Germany, under German law, or even
in violation of German law, by authorities of the German state.8
There is little doubt that the British, the French, and the Soviets had reasons of
their own to share these concerns. As a result, the definition of crimes against
humanity in article VI(c) of the Nuremberg Charter requires that atrocities be
committed ‘in furtherance of or in connection with any crime within the jurisdic-
tion of the International Tribunal’.9 In its final judgment, the International Military
Tribunal made a distinction between pre-war persecution of German Jews, which it
characterized as ‘severe and repressive’, and German policy during the war in the
occupied territories. Although the judgment frequently referred to events during
7 ‘Minutes of Conference Session of 23 July 1945’, in Report of Jackson, United States Representative,
331.
8 Ibid. 333 (emphasis added).
9 Agreement for the Prosecution and Punishment of Major War Criminals of the European Axis, and
Establishing the Charter of the IMT, annex, (1951) 82 UNTS 279.
the 1930s, none of the accused was found guilty of an act perpetrated prior to
1 September 1939, the day the war broke out.
Following the judgment, there was considerable outrage about the severe restric-
tion upon the concept of crimes against humanity. A member of the Nuremberg
prosecution team, Henry King, has described meeting Raphael Lemkin in the lobby
of the Grand Hotel in Nuremberg in October 1946, a few days after the Interna-
tional Military Tribunal completed its work:
When I saw him at Nuremberg, Lemkin was very upset. He was concerned that the decision
of the International Military Tribunal (IMT)—the Nuremberg Court—did not go far
enough in dealing with genocidal actions. This was because the IMT limited its judgment
to wartime genocide and did not include peacetime genocide. At that time, Lemkin was
very focussed on pushing his points. After he had buttonholed me several times, I had to tell
him that I was powerless to do anything about the limitation in the Court’s judgment.10
The disappointment soon manifested itself in the UN General Assembly, which was
meeting in New York. India, Cuba, and Panama proposed a resolution that they
said would address a shortcoming in the Nuremberg trial by which acts committed
prior to the war were left unpunished.11 One of the preambular paragraphs in the
draft resolution stated:
Whereas the punishment of the very serious crime of genocide when committed in time of
peace lies within the exclusive territorial jurisdiction of the judiciary of every State
concerned, while crimes of a relatively lesser importance such as piracy, trade in women,
children, drugs, obscene publications are declared as international crimes and have been
made matters of international concern.12
This paragraph never made it to the final version of Resolution 96(I), adopted in
December 1946, because the majority of the General Assembly was not prepared to
recognize universal jurisdiction for the crime of genocide. Nevertheless, the reso-
lution, somewhat toned down from the hopes of those who had proposed it,
launched a process that concluded two years later with the adoption of the
Convention for the Prevention and Punishment of the Crime of Genocide.13 Proposals
that the Genocide Convention make reference to crimes against humanity as a
related concept, or as some kind of broader umbrella under which the crime of
genocide was situated, were rejected by the drafters so as not to create any
confusion about the fact that genocide could be committed in time of peace as
well as in wartime. This could not be said with any certainty about crimes against
humanity at the time precisely because of the Nuremberg precedent.
10 ‘Remarks of Henry T. King, Jr., Case Western Reserve University School of Law, Genocide
Conference’, 27 September 2007, 1.
11 UN Doc. A/C.6/SR.22.
12 UN Doc. A/BUR/50.
13 Convention for the Prevention and Punishment of the Crime of Genocide, 277.
Thus, the recognition of genocide as an international crime by the General

Assembly of the United Nations in 1946, and its codification in the 1948 Convention,
can be understood as a reaction to the narrow approach to crimes against humani-
ty in the Nuremberg judgment of the International Military Tribunal. It was
Nuremberg’s failure to recognize the international criminality of atrocities com-
mitted in peacetime that prompted the first initiatives at recognizing and defining
the crime of genocide. Had Nuremberg affirmed the reach of international criminal
law into peacetime atrocities, the Genocide Convention might never have been
adopted. The term ‘genocide’ might then have remained a popular or colloquial
label used by journalists, historians, and social scientists but absent from legal
discourse.
T H E 1948 G E N O C I D E C O N V E N T I O N
................................................................................................................
The Convention for the Prevention and Punishment of the Crime of Genocide was
adopted unanimously by the UN General Assembly on 9 December 1948. It provides
the following definition of the crime of genocide:
In the present Convention, genocide means any of the following acts committed with intent
to destroy, in whole or in part, a national, ethnical, racial or religious group, as such:
(a) Killing members of the group;

(b) Causing serious bodily or mental harm to members of the group;
(c) Deliberately inflicting on the group conditions of life calculated to bring about its
physical destruction in whole or in part;
(d) Imposing measures intended to prevent births within the group;
(e) Forcibly transferring children of the group to another group.
In one sense, the definition is considerably narrower than that of crimes against
humanity, which can apply to a broad range of acts of persecution and other
atrocities committed against ‘any civilian population’. However, the definition is
manifestly broader because of the absence of any requirement of a link with
aggressive war.
Besides defining the crime, the Convention imposes several obligations upon
states that ratify it. They are required to enact legislation to provide for punishment
of persons guilty of genocide committed on their own territory. The legislation
must not allow offenders to invoke in defence that they were acting in an official
capacity. States are also required to cooperate in extradition when persons sus-
pected of committing genocide elsewhere find refuge on their territory. They may
not treat genocide as a political crime, which is an historic bar to extradition.
Disputes between states about genocide are automatically subject to the jurisdic-
tion of the International Court of Justice.
The title of the Convention speaks of prevention, but aside from a perfunctory
undertaking ‘to prevent’ genocide, there is nothing to suggest the scope of this
obligation. In 2007, in a case filed by Bosnia and Herzegovina against Serbia, the
International Court of Justice said there had been a breach of the Genocide
Convention because Serbia failed to intervene with its allies, the Bosnian Serbs, so
as to prevent the Srebrenica massacre of July 1995. The Court said that in view of
Serbia’s ‘undeniable influence’, the authorities should have ‘made the best efforts
within their power to try and prevent the tragic events then taking shape, whose
scale, though it could not have been foreseen with certainty, might at least have
been surmised’.14 The judgment clarifies that the obligation to prevent extends
beyond a country’s own borders. The principle it establishes should apply to other
States who take little or no action to respond when mass atrocity posing a risk of
genocide is threatened. This pronouncement is in the same spirit as an emerging
doctrine in international law expressed in a unanimous resolution of the UN
General Assembly, adopted in 2005, declaring that States have a ‘responsibility to
protect’ populations in cases of genocide, crimes against humanity, war crimes, and
ethnic cleansing.15
The Convention specifies that genocide is to be prosecuted by the courts of the
country where the crime took place or ‘by such international penal tribunal as may
have jurisdiction with respect to those Contracting Parties which shall have accept-
ed its jurisdiction’. The original General Assembly resolution proposed by Cuba,
India, and Panama called for recognition of universal jurisdiction over genocide.
This would mean that the courts of any state could punish the crime, no matter
where it was committed. The idea was rejected by the General Assembly in favour
of an approach combining territorial jurisdiction and an international institution.
The promised international court was not established for more than half a century,
when the Rome Statute of the International Criminal Court entered into force on
1 July 2002.16 Despite the Convention’s rejection of universal jurisdiction, in the
Eichmann prosecution the Israeli courts decided that it was accepted by customary
international law. Although no treaty authorizes universal jurisdiction over geno-
cide, and there is as yet no determination of its legitimacy by the International
Court of Justice, there now seems little doubt that it is permitted by international
law. In 2006 and 2007, the International Criminal Tribunal for Rwanda authorized
transfer of suspects for trial on the basis of universal jurisdiction with the approval
14 Case concerning Application of the Convention on the Prevention and Punishment of the Crime of
Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), Judgment, 26 February 2007, para. 438.
15 ‘2005 World Summit Outcome’, UN Doc. A/RES/60/1, para. 138.
16 Rome Statute of the International Criminal Court (2002), 2187 UNTS 90.
of the UN Security Council, further evidence of the broad acceptance of universal

jurisdiction over genocide.17
The Genocide Convention entered into force in 1951, after it had been ratified by
20 states. Approximately 140 states have now ratified the Convention. Several of
them have limited their commitments in the form of reservations. Most of these are
directed at excluding the jurisdiction of the International Criminal Court in the
event of disputes about the application of the Convention.
The definition of genocide set out in article II of the Convention has frequently
been criticized for its narrowness. For example, it applies to a limited number of
protected groups, and it requires an intent directed at the ultimate physical
destruction of the victimized group. There was disappointment when the Interna-
tional Court of Justice, in the Bosnia and Herzegovina case, dismissed attempts to
broaden the definition by interpreting the words ‘to destroy’ so as to encompass the
notion of ‘ethnic cleansing’. The Court said that ‘ethnic cleansing’, which it de-
scribed as the ‘deportation or displacement of the members of a group, even if
effected by force’, was not necessarily equivalent to destruction of that group, and
that destruction was not an automatic consequence of such displacement.18 The
relatively conservative approach to interpreting the definition, and a resistance to
broadening the scope through judicial action rather than amendment of the
Convention, is also reflected in judgments of the International Criminal Tribunal
for the former Yugoslavia19 and an authoritative report by a United Nations fact-
finding commission.20
Nor has there been any serious effort at the political level to amend or modify the
definition in Article II of the Convention. The ideal opportunity for such a
development would have been the adoption of the Rome Statute of the International
Criminal Court, when the definitions of the other core international crimes, crimes
against humanity and war crimes, were quite dramatically modernized. But when it
came to genocide, there were a few modest proposals, and these did not gain any
traction during the negotiations.21 At the Rome Conference, only Cuba argued for
17 Prosecutor v. Bagaragaza (Case ICTR-2005–86-R11bis), Decision on Prosecutor’s Request for

Referral of the Indictment to the Kingdom of the Netherlands, 13 April 2007. For Security Council
acquiescence, see UN Doc. S/PV.5697.
18 Case Concerning the Application of the Convention on the Prevention and Punishment of the Crime
of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), Judgment, 26 February 2007, para.
190. See also the chapter by Lieberman in this volume.
19 Prosecutor v. Krstić (Case IT-98–33-A), Judgment, 19 April 2004. Also Prosecutor v. Stakić (Case
IT-97–24-T), Judgment, 31 July 2003; Prosecutor v. Britanin (Case IT-99–36-T), Judgment, 1 September
2004; Prosecutor v. Blagojević et al. (Case IT-02–60-A), Judgment, 9 May 2007.
20 ‘Report of the International Commission of Inquiry on Darfur to the United Nations Secretary-
General, Pursuant to Security Council Resolution 1564 of 18 September 2004’, Geneva, 25 January 2005,
UN Doc. S/2005/60.
21 ‘Report of the Ad Hoc Committee on the Establishment of an International Criminal Court’,
UN Doc. A/50/22; para. 61; UN Doc. A/AC.249/1998/CRP.8, p. 2.; Herman von Hebel and Darryl
Robinson, ‘Crimes Within the Jurisdiction of the Court’, in Roy S. Lee (ed.), The International
amendment of the definition, proposing it be expanded to include social and

political groups.22
There is some evidence of innovation by national lawmakers when the provi-
sions of the Genocide Convention are translated into domestic criminal legislation.
The French Code pénal, for example, defines genocide as the destruction of any
group whose identification is based on arbitrary criteria.23 The Canadian imple-
menting legislation for the Rome Statute states that ‘“genocide” means an act or
omission committed with intent to destroy, in whole or in part, an identifiable
group of persons, as such, that, at the time and in the place of its commission,
constitutes genocide according to customary international law’, explaining that the
definition in the Rome Statute, which is identical to that of the Convention, is
deemed a crime according to customary international law. The legislation adds, in
anticipation: ‘This does not limit or prejudice in any way the application of existing
or developing rules of international law.’24
GENOCIDE AND C U S TO M A RY I N T E R NAT I O NA L L AW

................................................................................................................
Although written conventions or treaties are fundamental as a source of public
international law, binding norms may also be derived from custom, that is, a pattern
of behaviour or practice that exists because the parties believe it to be required as
a matter of legal obligation. Many rules and principles of international law exist in
this manner, without any treaty being required. Classic examples include diplomatic
immunities, the humane treatment of prisoners of war, and recognition that there
is a maritime perimeter surrounding a country that forms part of its sovereign
territory.
Shortly after the Genocide Convention was adopted, the International Court of
Justice issued an advisory opinion to clarify whether states that had ratified the
Convention but with reservations were actually party to the instrument. This was at
least theoretically relevant to a determination of when the treaty entered into force,
as this required twenty valid ratifications. The Court wrote that:
The origins of the Convention show that it was the intention of the United Nations to
condemn and punish genocide as ‘a crime under international law’ involving a denial of the
Criminal Court: The Making of the Rome Statute, Issues, Negotiations, Results (The Hague/London/
Boston: Kluwer Law, 1995), 89 n 37.
22 UN Doc. A/CONF.183/C.1/SR.3, para. 100.
23 Code Pénal (France), Journal officiel, 23 July 1992, art. 211–1.
24 Crimes against Humanity and War Crimes Act, 48–49 Elizabeth II, 1999–2000, C-19, s. 4.
right of existence of entire human groups, a denial which shocks the conscience of mankind
and results in great losses to humanity, and which is contrary to moral law and to the spirit and
aims of the United Nations. The first consequence arising from this conception is that the
principles underlying the Convention are principles which are recognized by civilized nations
as binding on States, even without any conventional obligation.25
This important statement is often cited as the judicial recognition of the prohibi-
tion of genocide as a customary legal norm, although the Court did not actually
refer to it expressly in this way. There is much subsequent authority for the
proposition that the prohibition of genocide, and the basic principles set out in
the Convention, form part of customary international law.26 According to the
International Court of Justice, in its 2007 ruling in the Bosnia and Herzegovina
case, the affirmation in article I of the Convention that genocide is a crime under
international law means it sets out ‘the existing requirements of customary inter-
national law, a matter emphasized by the Court in 1951.’27 In 2006, the International
Court of Justice said that the prohibition of genocide was ‘assuredly’ a peremptory
norm (jus cogens) of public international law, the first time it has ever made such a
declaration about any legal rule.28 A peremptory or jus cogens norm is so funda-
mental to customary international law that it cannot be subject to derogation.
P ROT E C T E D G RO U P S
................................................................................................................
The definition in the 1948 Convention applies to ‘national, ethnic, racial and religious
groups’. The concept is broadly analogous to what, at the time the Convention was
adopted, were considered as ‘national minorities’. This was clearly the perspective of
25 Reservations to the Convention on the Prevention and Punishment of the Crime of Genocide
(Advisory Opinion), [1951] ICJ Reports 16, p. 23. Quoted in Legality of the Threat or Use of Nuclear
Weapons (Advisory Opinion), [1996] ICJ Reports 226, para. 31; Case Concerning the Application of the
Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina
v. Serbia and Montenegro), Judgment, 26 February 2007, para. 161.
26 Prosecutor v. Sikirica et al. (Case IT-95–8-I), Judgment on Defense Motions to Acquit,
3 September 2001, para. 55; Prosecutor v. Musema (Case ICTR-96–13-T), Judgment, 27 January 2000,
para. 151; Prosecutor v. Bagilishema (Case ICTR-95–1A-T), Judgment, 7 June 2001, para. 54. The
Australian High Court has written that ‘[g]enocide was not [recognised as a crime under
customary international law] until 1948.’ Polyukhovich v. Commonwealth of Australia (1991), 101 ALR
545, at 598 (per Brennan J.).
27 Case Concerning the Application of the Convention on the Prevention and Punishment of the Crime
of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), Judgment, 26 February 2007, para. 161.
28 Case Concerning Armed Activities on the Territory of the Congo (New Application: 2002)
(Democratic Republic of the Congo v. Rwanda), Jurisdiction of the Court and Admissibility of the
Application, 3 February 2006, para. 64.
Raphael Lemkin and one of the other international experts who assisted the United
Nations in preparing the first draft of the Convention, Vespasian Pella.29
During the negotiations, there was an important debate about whether to
include political groups within the definition. Persecution on the grounds of
membership in a political group had been recognized at Nuremberg as a crime
against humanity. But the drafters of the Genocide Convention, Lemkin among
them, quite decisively rejected the inclusion of political groups. Some of the
subsequent literature on the subject has suggested that exclusion of political groups
was the result of pressure from the Soviet Union, but a careful reading of the
drafting history shows that opposition on this point was widespread.
In the first prosecution using a text derived from Article II of the Convention,
identification of the victim group did not raise any legal difficulties. Israeli law avoided
any discussion about the nature of ‘groups’ by simply reformulating the definition of
genocide so as to refer to ‘crimes against the Jewish people’,30 and nothing in the trial
record suggests that Eichmann ever challenged the fact that the victims of Nazi
atrocities were the ‘Jewish people’.31 The issue does not appear to have been particularly
controversial in litigation concerning the conflict in Bosnia and Herzegovina. A Trial
Chamber of the International Criminal Tribunal for the former Yugoslavia concluded
that ‘Bosnian Muslims’ were a ‘national group’,32 a finding that was not challenged on
appeal and that was accepted by the Appeals Chamber.33 After some initial uncertainty,
probably driven by discomfort with the contemporary legitimacy of the concept of
‘racial groups’, Trial Chambers of the International Criminal Tribunal for Rwanda have
taken judicial notice of the fact that the Tutsi, as well as the Hutu and the Twa, were
ethnic groups within Rwanda at the time of the 1994 genocide.34 In an innovative
interpretation, a Trial Chamber of the International Criminal Tribunal for Rwanda
held that the all ‘stable and permanent groups’ were protected by the Convention,35 but
its theory has had little resonance in subsequent case law.36
Generally, it is the perpetrator of genocide who defines the individual victim’s status
as a member of a group protected by the Convention. The Nazis, for example, had
detailed rules establishing, according to objective criteria, who was Jewish and who was
not. It made no difference if the individual, perhaps a non-observant Jew of mixed
parentage, denied belonging to the group. As Jean-Paul Sartre wrote: ‘Le juif est un
29 Vespasien V. Pella, La guerre-crime et les criminels de guerre, Réflexions sur la justice pénale
internationale, ce qu’elle est ce qu’elle devrait être (Neuchatel: Éditions de la baconnière, 1964), 80 n 1.
30 Nazi and Nazi Collaborators (Punishment) Law, 1950 (Law 5710/1950), s. I(a).
31 A-G Israel v. Eichmann, (1968) 36 ILR 5 (District Court, Jerusalem); A-G Israel v. Eichmann
(1968), 36 ILR 277 (Supreme Court of Israel).
32 Prosecutor v. Krstić (Case IT-98–33-T), Judgment, 2 August 2001, paras. 559–60.
33 Prosecutor v. Krstić (Case IT-98–33-A), Judgment, 19 April 2004, para. 6.
34 Prosecutor v. Kajelijeli (Case ICTR-98–44A-T), Judgment, 1 December 2003, para. 241.
35 Prosecutor v. Akayesu (Case ICTR-96–4-T), Judgment, 2 September 1998, para. 652.
36 ‘Report of the International Commission of Inquiry on violations of international humanitarian
law and human rights law in Darfur’, UN Doc. S/2005/60, para. 501.
homme que les autres hommes tiennent pour juif.’37 With considerable frustration,
lawyers and courts have searched for objective definitions of the protected groups. But
most of the judgments treat the identification of the protected group as an essentially
subjective matter. For example, Trial Chambers of the International Criminal Tribunal
for Rwanda have concluded that the Tutsi were an ethnic group based on the existence
of government-issued official identity cards describing them as such.38 A Trial Cham-
ber of the International Criminal Tribunal for the former Yugoslavia wrote that ‘the
relevant protected group may be identified by means of the subjective criterion of the
stigmatization of the group, notably by the perpetrators of the crime, on the basis of its
perceived national, ethnical, racial or religious characteristics. In some instances, the
victim may perceive himself or herself to belong to the aforesaid group.’39 The
prevailing view is that determination of the relevant protected group should be
made on a case-by-case, relying upon both objective and subjective criteria.40
ETHNIC CLEANSING AND C U LT U R A L G E N O C I D E

................................................................................................................
The Convention definition of genocide refers to the ‘intent to destroy’ without
further precision. The five punishable acts that follow consist of a combination of
physical, biological, and cultural attacks. For example, the fifth act of genocide in
the definition, forcibly transferring children from one group to another, quite
evidently does not involve their physical destruction. Rather, the elimination of a
group is contemplated by destroying the cultural memory and the national lan-
guage, through assimilation at a very young age. A literal reading of the definition
can therefore support an interpretation whereby acts of ‘ethnic cleansing’ or of
cultural genocide falling short of physical destruction would be punishable, a view
that some judgments appear to support.41
When the Convention was being drafted, the punishable acts were divided into three
categories, physical, biological, and cultural genocide. The UN General Assembly voted
37 Jean-Paul Sartre, Réflexions sur la question juive (Paris: Gallimard, 1954), 81–4.
38 Prosecutor v. Kayishema et al. (Case ICTR-95–1-T), Judgment, 21 May 1999, para. 98.
39 Prosecutor v. Brdanin (Case T-99–36-T), Judgment, 1 September 2004, para. 683.
40 Prosecutor v. Brdanin (Case IT-99–36-T), Judgment, 1 September 2004, para. 684. Also Prosecutor
v. Stakić (Case IT-97–24-A), Judgment, 22 March 2006, para. 25; Prosecutor v. Semanza (Case ICTR-97–
20-T), Judgment and Sentence, 15 May 2003, para. 317; Prosecutor v. Kajelijeli (Case ICTR-98–44A-T),
Judgment and Sentence, 1 December 2003, para. 811; Case Concerning the Application of the Convention on
the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and
Montenegro), Judgment, 26 February 2007, para. 191.
41 Prosecutor v. Krstić (Case IT-98–33-A), Partially Dissenting Opinion of Judge Shahabuddeen,
19 April 2004; Prosecutor v. Blagojević (Case IT-02–60-T) Judgment, 17 January 2005; Jorgić v. Germany
(App. 74613/01), 12 July 2007, para. 47.
quite deliberately to exclude cultural genocide from the Convention.42 It also rejected an
amendment from Syria to include as an act of genocide behaviour that today might be
called ‘ethnic cleansing’. The Syrian amendment read: ‘Imposing measures intended to
oblige members of a group to abandon their homes in order to escape the threat of
subsequent ill-treatment.’43 When the General Assembly agreed to include forcible
transfer of children, this was presented as an exception to the agreed upon exclusion
of cultural genocide.44 Consequently, a reading of the Convention definition that takes
into account the intent of its drafters will tend to reject inclusion of cultural genocide
and ethnic cleansing, and construe the words ‘to destroy’ as if they are modified
by ‘physically’ and ‘biologically’. On the other hand, a purely literal reading sustains
the view that cultural genocide is comprised within the words ‘to destroy’.
There are strong arguments for rejecting an approach to treaty interpretation
that puts too much emphasis on legislative intent, particularly in the field of
human rights law. Reliance upon the drafting history tends to freeze the provision,
preventing it from evolving so as to take into account historical developments and
changed attitudes. Be that as it may, courts to this day have shown great respect for
the relatively narrow perspective adopted by the General Assembly in 1948. This
tendency is only partially explained by an inherent conservatism, however. Just as
the crime of genocide emerged in international law as a reaction to the limitations
on crimes against humanity, more recently the law on crimes against humanity has
evolved to such an extent that it can now cover acts of ethnic cleansing and cultural
genocide, even when committed in peacetime. As a result, there is no ‘impunity
gap’, and there is little or no pressure in a legal sense for the expansion of the
definition of genocide by interpretation. Of course, there are important political
prerogatives and much symbolism associated with the label ‘genocide’, and many
victims are deeply disappointed when their own suffering is acknowledged as
‘mere’ crimes against humanity. They do not fully appreciate the importance of
the legal distinctions, which are the result of a complex historical debate. Thus,
while the distinction between genocide and crimes against humanity no longer has
significant legal consequences, it remains fundamental in other contexts.
NUMBERS AND GENOCIDE

................................................................................................................
The 1948 definition of genocide speaks of destruction of a group ‘in whole or in
part’. It was a noble attempt by the drafters to reach consensus on the numerical
42 UN Doc. A/C.6/SR.83.
43 UN Doc. A/C.6/234.
44 UN Doc. A/C.6/SR.82.
issue, but in reality the General Assembly used ambiguous terms and left their
clarification to judges in subsequent prosecutions. Several theories have emerged
with a view to circumscribing the notion of ‘in part’. Because the terms appear in
the preliminary paragraph of the definition, it is quite clear that they refer to the
genocidal intent. As a result, the fundamental question is not how many victims
were actually killed or injured, but rather how many victims the perpetrator
intended to attack. Even where there is a small number of victims, or none at
all—the Convention also criminalizes attempted genocide—the crime can be com-
mitted if the genocidal intent is present. The actual result, in terms of quantity, will
nevertheless be relevant in that it assists in assessing the perpetrator’s intent. The
greater the number of actual victims, the more plausible becomes the deduction
that the perpetrator intended to destroy the group, in whole or in part.
But there are other issues involved in construing the meaning of the term ‘in
part’. Could it be genocide to target only a few persons for murder because of their
membership in a particular ethnic group? A literal reading of the definition seems
to support such an interpretation. Nevertheless, this construction is rather too
extreme, and inconsistent with the drafting history, as well as with the context and
the object and purpose of the Genocide Convention. Two basic approaches to the
scope of the term ‘in part’ have emerged, each adding a modifying adjective,
‘substantial’ or ‘significant’, to the word ‘part’.
According to the Appeals Chamber of the International Criminal Tribunal for
the former Yugoslavia, ‘[i]t is well established that where a conviction for genocide
relies on the intent to destroy a protected group “in part,” the part must be a
substantial part of that group.’45 Noting that the Nazis did not realistically intend to
destroy all Jews, but only those in Europe, and that the Hutu extremists in Rwanda
sought to kill Tutsis within Rwanda, the Appeals Chamber said: ‘The intent to
destroy formed by a perpetrator of genocide will always be limited by the oppor-
tunity presented to him. While this factor alone will not indicate whether the
targeted group is substantial, it can—in combination with other factors—inform
the analysis.’46 In the factual context, the Appeals Chamber considered that the
Bosnian Muslim community in Srebrenica constituted a ‘substantial part’ of the
Bosnian Muslims as a whole, and that the attempt to destroy it amounted to
genocide.47
Another approach takes more of a qualitative than a quantitative perspective,
reading in the adjective ‘significant’. There is nothing to support this interpretation
in the drafting history of the Convention, and the idea seems to have been launched
by Benjamin Whitaker in a 1985 report to the UN Sub-commission for the
Protection and Promotion of Human Rights. He wrote that the term ‘in part’
45 Prosecutor v. Krstić (Case IT-98–33-A), Judgment, 18 August 2004, para. 8.

46 Ibid. para. 13.
47 Ibid. para 22.
denotes ‘a reasonably significant number, relative to the total of the group as a

whole, or else a significant section of a group such as its leadership’.48 Citing
Whitaker’s report, an expert body established by the UN Security Council in 1992
to investigate violations of international humanitarian law in the former Yugosla-
via held that ‘in part’ had not only a quantitative but also a qualitative dimension.
According to the Commission’s chair, Professor M. Cherif Bassiouni, the definition
in the Genocide Convention was deemed ‘sufficiently pliable to encompass not only
the targeting of an entire group, as stated in the convention, but also the targeting
of certain segments of a given group, such as the Muslim elite or Muslim women.’49
This approach was adopted by the Prosecutor of the International Criminal Tribu-
nal for the former Yugoslavia, in some of the initial indictments,50 and was subse-
quently accepted by trial judges.51 Although not explicitly endorsing the ‘significant
part’ gloss on the Convention, the Appeals Chamber of the Tribunal considered the
relevance to the Srebrenica Muslim community of the destruction of approximately
7,000 men. It referred to an observation of the Trial Chamber about the patriarchal
character of Bosnian Muslim society in Srebrenica, and the consequent impact upon
the future of the community that would result from the killing of its adult male
population. ‘Evidence introduced at trial supported this finding, by showing that, with
the majority of the men killed officially listed as missing, their spouses are unable to
remarry and, consequently, to have new children. The physical destruction of the men
therefore had severe procreative implications for the Srebrenica Muslim community,
potentially consigning the community to extinction.’52 In other words, the adult males
were a ‘significant part’ of a community, the Srebrenica Muslims, that was itself a
‘substantial part’ of the group as a whole, namely, Bosnian Muslims.
G E N O C I DA L I N T E N T
................................................................................................................
In principle, what sets criminal law apart from other areas of legal liability is its
insistence upon establishing that the punishable act was committed intentionally.
At best, inadvertent or negligent behaviour lies at the fringes of criminal law, and
will certainly not apply when the most serious crimes, including genocide, are
48 Benjamin Whitaker, ‘Revised and Updated Report on the Question of the Prevention and
Punishment of the Crime of Genocide’, UN Doc. E/CN.4/Sub.2/1985/6, para. 29.
49 ‘Final Report of the Commission of Experts’, UN Doc. S/1994/674, para. 94.
50 Prosecutor v. Karadžić et al. (Case IT-95–18-R61, IT-95–5-R61), Transcript of hearing of 27 June
1996, p. 15. Also Prosecutor v. Jelisić et al. (Case IT-95–10-I), Indictment, 21 July 1995, para. 17.
51 Prosecutor v. Jelisić (Case IT-95–10-T), Judgment, 14 December 1999, paras. 82, 93; Prosecutor v.
Sikirica et al. (Case IT-95–8-T), Judgment on Defense Motions to Acquit, 3 September 2001, para. 80.
52 Prosecutor v. Krstić (Case IT-98–33-A), Judgment, 18 August 2004, para. 28.
concerned. As a rule, criminal legislation does not spell out a requirement of intent,
as this stipulation is considered to be implicit. Exceptionally, the definition in the
Convention refers to the intent of the perpetrator, which must be to destroy the
protected group in whole or in part. There are actually two distinct intents
involved, because the underlying genocidal act, for example killing or causing
serious bodily or mental harm to a member of the group, must also be carried
out intentionally.
Courts often refer to the ‘specific intent’ of genocide, or the dolus specialis, so as
to distinguish it from non-genocidal killing. In principle, all crimes must be
committed intentionally, in the sense that they are the result of an active mind
operating consciously. The definitions of some crimes go beyond this general
presumption, and state expressly that they must be committed with a special or
specific intent. The assertion that genocide is committed when one of the punish-
able acts, killing for example, is perpetrated ‘with intent to destroy’ a protected
group, leads to the observation that it is a crime of ‘specific intent’ or, according to
jurists trained in continental law, one of dolus specialis. Application of this classic
criminal law paradigm to genocide has resulted in what may be an exaggerated
focus by some judges on the individual perpetrator, taken in isolation. The
International Criminal Tribunal for the former Yugoslavia has adopted the view
that an individual, acting alone, can commit an act of genocide to the extent that he
or she engages in killing with a genocidal intent.53 The problem with such analysis
is that it loses sight of the importance of the plan or policy of a state or analogous
entity. In practice, genocide within the framework of international law is not the
crime of a lone deviant but the act of a state. The importance of a state policy
becomes more apparent when the context shifts from individual prosecution to a
broader and more political determination.
For example, in September 2005, the UN Security Council commissioned a study
to determine whether genocide was being committed in Darfur. The resulting
expert report did not seriously attempt to determine whether any single individual
within Sudan had killed with genocidal intent. Rather, it examined the policy of the
Sudanese government, stating: ‘The Commission concludes that the Government
of Sudan has not pursued a policy of genocide.’54 The Commission said that there
was evidence of two elements of the crime of genocide. The first was the presence of
material acts corresponding to paragraphs in the definition of the crime set out in
article II of the 1948 Convention for the Prevention and Punishment of the Crime of
Genocide. It observed that ‘the gross violations of human rights perpetrated by
Government forces and the militias under their control’ included reports of killing,
causing serious bodily or mental harm, and deliberate infliction of conditions of
53 Prosecutor v. Jelisić (Case IT-95–10-T), Judgment, 14 December 1999, para. 100.

54 ‘Report of the International Commission of Inquiry on Violations of International Humanitarian
Law and Human Rights Law in Darfur’, UN Doc. S/2005/60, para. 518.
life likely to bring about physical destruction. The second was the subjective
perception that the victims and perpetrators, African and Arab tribes respectively,
made up two distinct ethnic groups. But, said the Commission,
one central element appears to be missing, at least as far as the central Government authorities
are concerned: genocidal intent. Generally speaking the policy of attacking, killing and forcibly
displacing members of some tribes does not evince a specific intent to annihilate, in whole or
in part, a group distinguished on racial, ethnic, national or religious grounds.55
Article III of the Genocide Convention establishes that in addition to criminal

liability for the actual perpetrators of the crime, accomplices are also punishable.
The transposition of concepts of complicity drawn from ordinary criminal law to
the international setting of mass atrocity lacks some precision. In reality, it is the
organizers and instigators of genocide who bear the greatest responsibility; the
physical acts themselves are committed by individuals who are low in the hierarchy,
and who may well be ignorant of the genocidal intent.
The statutes of the international criminal tribunals make provision for prosecution
of the commander or superior where the acts themselves are committed by subordi-
nates, even in the absence of evidence that actual orders or directions were given. This
approach to liability, drawn from a notorious post-Second World War case,56 has
proven to be of only theoretical interest. The scenario whereby a superior is convicted
for failing to prevent subordinates from committing genocide is implausible once it is
understood that this is a crime that stems from a state or organizational plan or policy.
Many contemporary international criminal prosecutions are based upon
a theory known as ‘joint criminal enterprise’. It recognizes that atrocities that
qualify as international crimes, including genocide, are committed by groups and
organizations, acting with a common purpose. In practice, it means that the leaders
or organizers will be held responsible for the crimes committed by their associates,
even those that they did not specifically intend, to the extent that these were a
reasonable and foreseeable outcome of the common purpose or joint enterprise.57
S TAT E R E S P O N S I B I L I T Y
................................................................................................................
Although the definition of genocide is framed as a crime, implying that it applies
only to individuals, the 1948 Genocide Convention imposes duties upon states to
55 Ibid.
56 United States of America v. Yamashita (1948), 4 LRTWC 1, pp. 36–7; In re Yamashita, 327 U.S.
1 (1945).
57 Prosecutor v. Britanin (Case IT-99–36-A), Judgment, 3 April 2007, paras. 420–5.
prevent genocide and clearly envisages their liability before the International Court
of Justice. Any doubts on this point were resolved in the February 2007 judgment of
the International Court. There remains an ongoing debate among international
lawyers as to whether states actually commit crimes. The Court avoided the
question when it ruled that Serbia was liable for failing to prevent genocide,
whether qualified as a crime or as an internationally wrongful act.
The Court also held that where charges of genocide are made, they must be
established by proof ‘at a high level of certainty appropriate to the seriousness of
the allegation’.58 This is a considerably more demanding standard than what would
normally be applied in ordinary cases involving State responsibility before the
International Court of Justice, and it appears to approximate the norm applied in
criminal prosecutions. For example, the Rome Statute of the International Criminal
Court says that ‘[i]n order to convict the accused, the Court must be convinced of
the guilt of the accused beyond reasonable doubt.’59 In adopting this approach, the
International Court of Justice greatly reduced the likelihood of a result inconsistent
with that of the international criminal tribunals. Its exigent standard of proof with
respect to genocide virtually assured that the International Court of Justice, dealing
with state responsibility, and the International Criminal Tribunal for the former
Yugoslavia, dealing with individual responsibility, would remain very much on the
same wavelength.
C O N C LU D I N G REMARKS
................................................................................................................
First proposed by Raphael Lemkin in 1944 to fill a gap in the existing law that would
adequately address Nazi crimes against minorities, especially European Jews, the
crime of genocide was subsequently codified in the first important UN human
rights treaty, the Convention on the Prevention and Punishment of the Crime of
Genocide, adopted in December 1948. To a large extent, the Convention arose as a
response to the inadequate codification of crimes against humanity at Nuremberg,
which failed to address atrocities committed during peacetime. The Genocide
Convention offered a definition of the crime that covered peacetime acts, but the
General Assembly would only agree to this if the crime itself was defined rather
narrowly.
58 Case Concerning Application of the Convention on the Prevention and Punishment of the Crime of
Genocide (Bosnia and Herzegovina v. Yugoslavia), Judgment, 26 February 2007, para. 210.
59 Rome Statute of the International Criminal Court (2002), 2187 UNTS 90, art. 66(3).
Over the years, the limited definition of genocide in the 1948 Convention has
provoked much criticism and many proposals for reform. But by the 1990s, when
international criminal law went through a period of stunning developments, it was
the atrophied concept of crimes against humanity that emerged as the best legal
tool to address atrocities. A gap in the law needed to be filled. Instead of enlarging
the definition of genocide in order to accomplish this, the international commu-
nity opted for an expanded view of crimes against humanity instead. As a result,
genocide as a legal concept remains essentially reserved for the clearest of cases of
physical destruction of national, ethnic, racial, or religious groups. An important
ruling of the International Court of Justice of February 2007 confirms these
observations about the stable and relatively conservative approach to genocide
that is likely to prevail in the case law for many years to come.
SELECT BIBLIOGRAPHY
Boot, Machteld, Genocide, Crimes Against Humanity, War Crimes: Nullum Crimen Sine Lege
and the Subject Matter Jurisdiction of the International Criminal Court (Antwerp/Oxford/
New York: Intersentia, 2002).
Drost, Pieter Nicolaas, Genocide, United Nations Legislation on International Criminal Law
(Leyden: A. W. Sythoff, 1959).
Kress, Claus, ‘The Crime of Genocide under International Law’, International Criminal Law
Review 6 (2006), 461–502.
Lemkin, Raphael, Axis Rule in Occupied Europe, Laws of Occupation, Analysis of Govern-
ment, Proposals for Redress (Washington, DC: Carnegie Endowment for International
Peace, 1944).
Mettraux, Guénaël, International Crimes and the ad hoc Tribunals (Oxford: Oxford Univer-
sity Press, 2005).
Robinson, Nehemiah, The Genocide Convention: A Commentary (New York: Institute of
Jewish Affairs, 1960).
Schabas, William A., ‘Genocide and the International Court of Justice: Finally, a Duty to
Prevent the Crime of Crimes’, Genocide Studies and Prevention 2:2 (2007), 101–22.
—— Genocide in International Law, 2nd edn (Cambridge: Cambridge University Press,
2009).
chapter 7
.............................................................................................
S O C I O LO G Y A N D
GEN O CIDE
.............................................................................................
martin shaw
Genocide is a crime of social classification, in which power-holders target partic-

ular populations for social and often physical destruction. Hence it can be
described as a peculiarly sociological crime, in which the activity of social
classification is perverted by pseudoscience. Not for nothing did the journalist
William L. Shirer describe the forever classifying National Socialists, whom he
observed first hand, as ‘sociologists’.1 Yet it was not sociologists who invented the
terminology for this crime, nor were they particularly open to the idea of ‘geno-
cide’ when Raphael Lemkin first outlined it in Axis Rule in Occupied Europe. It was
the American Journal of Sociology that printed its ‘harshest review’, in which
Melchior Palyi astonishingly ‘blamed Lemkin for his failure to explore the “exten-
uating circumstances” of Nazi behaviour.’2
It is true that sociology’s history does not reveal quite the same depths as that of
its sister discipline, anthropology. ‘Physical anthropologists, eugenicists, ethno-
graphers, and social anthropologists were equally busy during the first half of the
1940s,’ Gretchen E. Schafft records, ‘in “racial” studies, in Mendelian genetics, in
ethnographic studies of prisoners of war, and in sorting people by psychological
and physical characteristics. In these and in so many different ways they helped to
1 William L. Shirer, Berlin Diary: The Journal of a Foreign Correspondent 1934–1941 (London, 1941).
2 Samantha Power, ‘A Problem from Hell’: America and the Age of Genocide (London: Flamingo,
2003), 21.
sociology and genocide 143
determine the outcomes of the lives of their subjects.’3 In occupied Poland,

researchers carried out ethnographic research ‘in conjunction with the SS, who
provided protection for the scientists and ensured the compliance of the subjects.
People were taken at gunpoint to collection places where they were measured,
interviewed, and sometimes fingerprinted.’4 But this involvement in genocide was
not an aberration for social science. In the business of racial classification, ‘one
could move so easily from a study of differences to the conviction that differences
could be gradated into a hierarchical value system.’5 As Alexander Hinton has
argued, there was something inherent in the concept of social classification that
lent itself to this approach: ‘Diverse ways of life were compressed into relatively
stable categories, a homogenizing tendency that was paralleled by the anthropo-
logical typologies of race.’6 This was a danger for sociology as well as anthropology.
It is true that during the Nazi era, according to Wolfgang Glatzer’s history of the
German Sociological Association, ‘just about all reputable sociologists emigrated,
especially those of Jewish origin’, while some ‘attempted to struggle through the
years of the Third Reich without giving in to the Nazis’; only ‘a third group adhered
more or less openly to Nazi ideology, defining themselves as Volkish sociologists.’7
Moreover Max Weber (who died in 1920) had dismissed race as a social category,
pointing out that ‘the possession of a common biological inheritance by virtue of
which persons are classified as belonging to the same “race,” naturally implies no
sort of communal social relationship between them.’8 And yet Weber’s own earlier
history demonstrated that sociologists were hardly immune to the degrading
effects of the racial classifications that led to the anthropologists’ collaboration.
In the 1890s, he had advocated that the estates of the Prussian aristocracy, increas-
ingly manned by cheaper Polish labour, should be recolonized with native German
farmers to save the Reich from ‘Polonization’. Weber even claimed that the danger
of a Polish invasion lay in their ‘physiological cleft’ from the Germans. So when
nearly half a century later Hitler expelled Poles (and Jews) in order to Germanize
the areas of conquered Poland adjacent to Germany, he advanced an idea similar to
one proclaimed by the man who would become known as the greatest modern
sociologist.9
3 Gretchen E. Schafft, ‘Scientific Racism in the Third Reich: German Anthropologists in the Nazi
Era’, in Alexander Laban Hinton (ed.), Annihilating Difference: The Anthropology of Genocide
(Berkeley: University of California Press, 2002), 119.
4 Ibid. 128.
5 Ibid. 124.
6 Alexander Laban Hinton, ‘The Dark Side of Modernity: Toward an Anthropology of Genocide’, in
idem (ed.), Annihilating Difference, 14.
7 Wolfgang Glatzer, ‘German Sociological Association’, available at http://www.soziologie.de/#3.%
20Die%20DGS%20und%20der%20NS
8 Max Weber, The Theory of Social and Economic Organization, ed. Talcott Parsons (New York: Free
Press of Glencoe, 1964), 138.
9 Peter Thomas, ‘Being Max Weber’, New Left Review, 41 (Sept.–Oct. 2006).
144 martin shaw
Thus sociology has more to redeem than Palyi’s review. And yet the consensus
among the sociologists who have written about genocide is that until recently the
discipline’s subsequent record was astonishingly poor. Mostly genocide ‘was largely
overlooked or suppressed by social scientists until the 1970s’.10 As Irving Louis
Horowitz suggested, ‘Many sociologists exhibit a studied embarrassment about
these issues, a feeling that intellectual issues posed in such a manner are melodra-
matic and unfit for scientific discourse.’11 Michael Mann, who has produced
perhaps the most important sociological contribution to the study of genocide,
confessed that his earlier work ‘had neglected the extremes of human behaviour’.
He ‘had not thought much about good and evil’.12 Textbooks continue to ignore or
marginalize genocide even today. Zygmunt Bauman claimed: ‘When measured
against the work done by historians or theologians, the bulk of academic sociology
looks more like a collective exercise in forgetting and eye-closing.’13 Not surpris-
ingly, he memorably concluded that ‘the Holocaust has more to say about the state
of sociology than sociology in its present shape is able to add to our knowledge of
the Holocaust.’14 His attack was fundamental:
The nature and style of sociology has been attuned to the selfsame modern society it
theorized and investigated; sociology has been engaged since its birth in a mimetic
relationship with its object—or, rather, with the imagery of that object which it constructed
and accepted as the frame for its own discourse. And so sociology promoted, as its own
criteria of propriety, the same principles of rational action it visualized as constitutive of its
object. It also promoted, as binding rules of [its] own discourse, the inadmissibility of
ethical problematics in any other form but that of a communally-sustained ideology and
thus heterogenous to sociological (scientific, rational) discourse. Phrases like ‘the sanctity of
human life’ or ‘moral duty’ sound as alien in a sociology seminar as they do in the smoke-free,
sanitized rooms of a bureaucratic office.15
The historian Herbert Hirsch has seen these limitations as intrinsic to social
science: ‘It is unfortunate that Holocaust and genocide studies are being pressured
into a phase of social science rationality . . . only to become bogged down in the
elusive variable and definition, as everyday life becomes almost entirely eliminated
from their concern.’16 Nevertheless Bauman’s work proved a starting point for
sociological research, informed by the very humane values that Hirsch advocated,
without abandoning his discipline’s generalizing concerns. His agenda, ‘to open up
10 Helen Fein, ‘Genocide: A Sociological Perspective’, Current Sociology 38:1 (1990), 5.

11 Irving Louis Horowitz, quoted ibid. 6.
12 Michael Mann, The Dark Side of Democracy: Explaining Ethnic Cleansing (Cambridge:
Cambridge University Press), ix.
13 Zygmunt Bauman, Modernity and the Holocaust (Cambridge: Polity, 1989), 9–10.
14 Ibid. 3.
15 Ibid. 29. Emphasis in original.
16 Herbert Hirsch, Genocide and the Politics of Memory: Studying Death to Preserve Life (Chapel
Hill: University of North Carolina Press, 1995), 81.
the findings of the specialists to the general use of social science, to interpret them
in a way that shows their relevance to the main themes of sociological inquiry, to
feed them back into the mainstream of our discipline,’17 has guided some signifi-
cant scholars. Weber’s own ‘principles of rational action’ could be utilized in
critical analysis, explaining contextually the development of murderous intentions
and violent means and their realization in political conflict.
R EV I S I T I N G C O N C E P T S AND ‘C L A S S I F I C AT I O N ’
................................................................................................................
Sociologists have contributed particularly to the debate on the meaning of geno-
cide. Some say we can have too much ‘definitionalism’, and that—whatever we call
things—the important thing is concrete understanding. Nevertheless, words mat-
ter. As Weber put it: ‘The apparently gratuitous tediousness involved in the
elaborate definition of . . . concepts is an example of the fact that we often neglect
to think out clearly what seems to be “obvious,” because it is intuitively familiar.’18
Lemkin invented ‘genocide’ because he wanted to describe—and highlight for
countervailing action—a general class of violent actions. We do not have to
adopt his terminology or definitions; indeed we cannot avoid modifying them.
Yet if it was important that Lemkin introduced ‘genocide’, it is also important we
are aware of how we change its meaning. If we use it in new ways, or introduce new
terms to describe some of the phenomena it originally designated, we need to
explain why. In any case, all serious concepts must be used coherently—with
internal coherence of meaning as well as valid reference—and must be capable of
explanation. For genocide studies to be coherent, the field must forge adequate
concepts.
The answer to the misuse of classification is not, therefore, to abandon classifi-
cation. We simply cannot do this: classification is an inescapable part of human
cognition. Social scientists’ classifications—like those of genocidists—are particu-
lar versions of this general human activity. Classification’s danger is always, as Nigel
Eltringham has suggested, that ‘we “misplace concreteness” and set out to “prove”
that our abstract concepts . . . really do correspond to reality, rather than being
contingent approximations.’19 Genocidists go a big step further in trying to enforce
their social classifications, making reality correspond at the cost of lives. But
victims, especially resisters, also advance their own categories. They assert their
17 Bauman, Modernity and the Holocaust, xiii. Emphasis in original.

18 Weber, Theory of Social and Economic Organization, 140.
19 Nigel Eltringham, Accounting for Genocide (London: Pluto, 2004), 7.
146 martin shaw
understandings of their identities; they assert their status as ‘victims’, and as

‘civilians.’ And at the same time, resisters impose classifications on those who
would classify them. They describe their persecutors’ actions as ‘genocide’ and
classify them as ‘crimes’; they call their attackers ‘perpetrators’, ‘criminals’, and
génocidaires. Social scientists mostly develop existing meanings rather than in-
venting new terms; we cannot avoid referring to categories developed by active
participants. We need to make sense of perpetrators’ intentions, but ultimately we
should reject their absolutist and euphemistic categories—like ‘ethnic cleansing’—
and develop categories that articulate the experiences of victims, resisters, and
bystanders.
L E M K I N ’ S O W N S O C I O LO G I C A L F R A M EWO R K
................................................................................................................
That Hirsch’s concerns were in the end misplaced can be grasped even before we
approach the writings of sociologists themselves. For violence against civilian
populations always displays a ‘social logic’, which in turn demands to be critiqued
or deconstructed by a critical ‘sociology’. Although the founder of genocide studies
was a lawyer by trade, his fundamental understanding of the phenomenon was
sociological rather than legal or simply historical. Moreover, although there is
much lip service to his achievements, there is little appreciation of how significant
his distinctive understanding remains. Indeed the appreciation of Lemkin’s work
has suffered from his political success. It is to the Convention, rather than Lemkin
himself, that even social scientists often refer in defining genocide. This tendency is
unfortunate because, although Lemkin’s was far from the last word, he offered a
much more adequate sociological understanding. Moreover many authors, trying
to improve on the Convention, have actually moved even further from Lemkin in
ways that militate against sociological knowledge. Recovering the meaning of
genocide for Lemkin is a necessary beginning for the sociology of genocide.
As is well known, from Lemkin’s first formulations, when in 1933 he proposed a
draft law banning ‘barbarity’, he was looking for a term and a law that brought
together a whole class of violent and humiliating actions against members of
collectivities. Genocide was not a specific type of violence, but a general charge
that highlighted the common elements of many acts that ‘taken separately’ con-
stituted specific crimes. Lemkin saw it as including not only organized violence but
also economic destruction and persecution. What concerned him was precisely the
‘common feature’ of these types of action: their threat to the existence of a
collectivity. When Lemkin finally introduced the term ‘genocide’ in 1944, he
again warned against a narrow interpretation:
Generally speaking, genocide does not necessarily mean the immediate destruction of a
nation, except when accomplished by mass killings of all members of a nation. It is intended
rather to signify a coordinated plan of different actions aiming at the destruction of
essential foundations of the life of national groups, with aim of annihilating the groups
themselves.20
The nuances of the key word ‘destruction’ were indicated here by the difference
between ‘immediate destruction’ of a nation and ‘destruction of essential founda-
tions’ of its life. Lemkin was clear that genocide refers generally to the latter;
‘immediate’ destruction in the sense of ‘mass killings of all members of a nation’
was a specific type but did not define genocide. Lemkin’s definition was exemplified
in the substance of his book, where he wrote that ‘the Nazi genocide was effected
through a synchronized attack on different aspects of life of the captive peoples,’
politically, socially, culturally, economically, biologically, religiously, physically, and
morally.21 Genocide, for the term’s inventor, was a comprehensive process in which
a power ‘attacked’ and ‘destroyed’ the way of life and institutions of peoples.
‘Physical’ genocide—especially mass killing—was only one dimension of the
comprehensive ‘attack’.
In contrast to Lemkin’s original view, the Convention on the Prevention and
Punishment of the Crime of Genocide—while maintaining something of his broad
approach by defining genocide as a range of ‘acts committed with intent to destroy,
in whole or in part, a national, ethnical, racial or religious group, as such’—in
practice narrowed it down by giving a restricted list of acts (killing, bodily and
mental harm, conditions leading to physical destruction, restricting births, trans-
ferring children) with greater emphasis on physical destruction. This was particu-
larly evident in the third clause, where ‘inflicting on the group conditions of life’
was genocidal only in so far as it was ‘calculated to bring about its physical
destruction’. For Lemkin in contrast it was quite clear that ‘a synchronized attack
on different aspects of life’ was genocidal in itself. The Convention laid stronger
emphasis than Lemkin on physical and biological destruction, and less on broader
social destruction. In contrast Lemkin—however much he wished to establish and
enforce genocide law—offered a historical-sociological account. He was surely
right that in order to understand genocide, we should see killing and physical
harm as elements of the broader process of social destruction. Even the Nazis did
not aim simply to kill subject peoples, not even the Jews: they aimed to destroy
their ways of life and social institutions. It is implausible to reduce this aim to a
‘means’ of physical destruction, as the Convention implied. Indeed it was the other
way round: when physical destruction came to be a distinct, eventually overriding
end this was an extreme development of pre-existing Nazi policies of social
Proposals for Redress, Axis (Washington, DC: Carnegie Endowment for International Peace, 1944), 79.
21 Ibid. xi-xii.
148 martin shaw
destruction. Lemkin was correct to stress the integrated, multidimensional, nature

of the attack, and not to fall (as later writers have) into the trap of separating
physical violence from social destruction. In this sense his work, rather than the
Convention, remains the essential starting point for sociological understanding.
Nevertheless Lemkin had not presented a fully plausible account of the relations
of socially destructive ends and violent or murderous means. His listing of ‘the field
of physical existence’ as just one aspect of Nazism’s coordinated attack was too
mechanical. He failed to clarify that while genocide involved much more than
killing, violence in a broad sense and its threat lay behind all genocidal policies.
Although genocide could not be defined by a specific violent method like killing, the
idea of social destruction necessarily entailed generally violent methods. What else
could social ‘destruction’ mean? The deficiency of Lemkin’s listing approach meant
that this relationship between violence and social destruction remained to be fully
grasped.
T H E S O C I O LO G I C A L C O N T R I B U T I O N
ON THE CONCEPT OF GENOCIDE
................................................................................................................
After Lemkin, the social science of genocide was mostly neglected until the 1980s.
The first major advance was Leo Kuper’s wide-ranging study, which criticized the
Convention, emphasized the importance of colonial genocide and introduced the
crucial idea of ‘genocidal massacre’ to refer to localized mass killings.22 After Kuper,
however, many social scientists worked from the legal definition, which was an
unusual way of defining a sociological concept. For example, Helen Fein closely
followed the Convention when she proposed: ‘Genocide is sustained purposeful
action by a perpetrator to physically destroy a collectivity directly or through
interdiction of the biological and social reproduction of group members, sustained
regardless of the surrender or lack of threat offered by the victim.’23 The idea that
physical destruction of a group could either be ‘direct’ or be carried out through
interdiction of reproduction faithfully reflected the Convention’s specification of
means. Likewise ‘sustained purposeful action’ resumed the Convention idea of
intentional destruction. Yet Fein noted the difficulties of the latter, and that to
‘avoid the whole question of inference of intent, [some authors] propose that we
simply eliminate intent as a criterion.’24 She disagreed with Tony Barta’s alternative
22 Leo Kuper, Genocide: Its Uses in the Twentieth Century (Harmondsworth: Penguin, 1981), 32.
23 Fein, ‘Genocide’, 24.
24 Ibid.
proposal of ‘a conception of genocide which embraces relations of destruction’.25

Accepting that the meaning of genocide was firmly lodged in the idea of destruction
as the intentional, organized policy of particular actors, she argued that we cannot
‘remove from the word the emphasis on policy and intention which brought it into
being.’ However, she insisted on a more usual understanding of ‘intention’:
Intent or purposeful action—or inaction—is not the same in law or everyday language as
either motive or function. An actor performs an act, we say, with intent if there are
foreseeable ends or consequences: for what purpose is different from why or for what
motive is the act designed.26
We need, she proposed, ‘the sociological concept of purposeful action’ as a ‘bridge’

between the legal concept of intent and a broader understanding.27
Fein modified the Convention concept in three additional respects. First, she
referred only to collectivities and groups in general, and did not reproduce the
Convention’s restriction to ‘national, ethnical, racial and religious groups’. She saw
the protected groups as ‘basic kinds, classes, or sub-families of humanity, persisting
units of society’, and argued cogently that ‘the specification of groups should be
consistent with our sociological knowledge of both the persistence and construc-
tion of group identities in society’ and ‘should conform to the implicit universalis-
tic norm and a sense of justice, embracing the right of all non-violent groups to
co-exist.’28 Second, by inserting the word ‘physically’ in her definition, she went even
further than the Convention in narrowing the scope of the crime from Lemkin’s
original idea that genocide is ‘a synchronized attack on different aspects of life of
(captive) peoples’ towards an exclusive emphasis on killing and other measures of
‘biological’ rather than social destruction. Third, Fein’s phrase, ‘sustained regardless
of the surrender or lack of threat offered by the victim’, emphasized genocide’s
separation from war: victims are destroyed even if they are not military threats.
A more distinctive sociological contribution came from Frank Chalk and Kurt
Jonassohn, whose widely quoted definition stated: ‘Genocide is a form of one-
sided mass killing in which a state or other authority intends to destroy a group,
as that group and members in it are defined by the perpetrators.’29 In defining
genocide simply as ‘mass killing’, Chalk and Jonassohn departed more radically
from the Convention, identifying genocide with physical destruction even more
narrowly than Fein, and moved still further from Lemkin’s broad concept. Fein
rightly criticized them for failing to allow for ‘other forms of intentional
biological destruction’ and pointed out that specifying states as perpetrators
25 Tony Barta quoted ibid. 16. Emphasis in original.

26 Ibid. 19.
27 Ibid. 20.
28 Ibid. 24.
29 Frank Chalk and Kurt Jonassohn, The History and Sociology of Genocide: Analyses and Case
Studies (New Haven, CT: Yale University Press, 1990), 23.
150 martin shaw
was unnecessarily restrictive.30 Even if states are commonly organizing centres,

parties, settlers, paramilitaries, and others have also been responsible, thus it
seems perverse to define genocide by the state (even with the let-out of ‘or
other authority’).
However, it was Chalk and Jonassohn’s definition of the victim group ‘as that
group and members in it are defined by the perpetrators’ that was their most
distinctive contribution. This was important for emphasizing that perpetrators
work according to their own, often fantastical, ideas of ‘enemy’ groups. This idea
brought into question the assumption made by Lemkin, the Convention, and Fein
that groups necessarily exist ‘objectively’ or are defined by their own consciousness
of their identity, and this is what genocidists aim to destroy. Since all agree that the
genocidists’ intention to destroy a group is the starting-point in defining genocidal
action, it follows that it is their idea of that group which (first of all) counts in the
process. As Fein suggested, Chalk and Jonassohn’s definition involved the opposite
danger of according too much significance to the ideas of the perpetrators:31
it needs to be corrected to acknowledge the contending role of victims’ self-
perceptions. However, by recognizing the role of ideas in genocide, Chalk and
Jonassohn opened up an important, indeed an unavoidable dimension.
They also gave new emphasis to the general separation of genocide from war.
Their idea of ‘one-sidedness’ could be taken as meaning that genocide is something
one party does to another without conflict or resistance. Clearly there is a funda-
mental sense in which one-sidedness is what genocide is about: everyone under-
stands that its violence is targeted by an organized, armed power against a largely
unarmed population. However, such violence has often been imposed in the
context of conflict between organized forces (war) and often provokes resistance,
and hence new conflict, so that the genocidal element of one-sided killing is often
part of a situation of two- or many-sided conflict and violence. To define genocide
as ‘one-sided’ killing was potentially misleading if it led scholars to miss these
connections.
B I O P OW E R , M O D E R N I T Y, AND THE H O LO C AU S T
................................................................................................................
If writers like Fein, Chalk, and Jonassohn made distinctively sociological argu-
ments central to the emerging field of genocide studies, their work did little to
overcome the general sociological neglect. Only when major sociological writers
30 Fein, ‘Genocide’, 13.

31 Fein criticizes this aspect of Chalk and Jonassohn’s definition: ‘Genocide’, 13.
turned their attention to genocide did the wider discipline began to take notice. No
less a figure than Michel Foucault had implicated genocide in his arguments about
the nature of power in modernity, when he commented: ‘If genocide is indeed the
dream of modern powers, this is not because of a recent return of the ancient right
to kill; it is because power is situated and exercized at the level of life, the species,
the race, and the large-scale phenomena of population.’32 He clearly linked geno-
cide to modern war,33 and both to profound transformations of the mechanisms of
power in modern times. Sovereign power was now
power bent on generating forces, making them grow, and ordering them, rather than one
dedicated to impeding them, making them submit, or destroying them. There has been a
parallel shift in the right of death, or at least a tendency to align itself with the exigencies of a
life-administering power and to define itself accordingly. This death that was based on the
right of the sovereign is now manifested as simply the reverse of the right of the social body
to ensure, maintain or develop its life.34
For Foucault therefore genocide represented a manifestation of modern ‘bio-

power’, reflecting the fact that ‘life and its mechanisms’ had been brought ‘into
the realm of explicit calculations and made knowledge-power an agent of transfor-
mation of human life.’35 The corollary of the state’s management of life forces was a
new management of death: ‘One might say that the ancient right to take life or let
live was replaced by a power to foster life or disallow it to the point of death.’36
However, if genocide is a manifestation of the general ‘power over life’, by which
the state takes upon itself the task of regulating and maintaining its subjects’ minds
and bodies, we still need to ask why this is exercised in a genocidal manner.
Foucault’s explanation that genocide is the other side of the modern state’s ‘func-
tion of administering life’ seems insufficient. His relatively unexamined idea of
genocide appears overinfluenced by particular pseudoscientific, eugenic ideologies
and the exceptionally rationalized murder of the extermination camps. Although
totalitarian rulers did sometimes appear to be mobilizing power to raise and cull
entire populations like livestock, scholars have shown that even their genocidal
decisions were pragmatic, launched in contexts of political conflict and war.
Elsewhere genocide has been influenced less by ambitious ideological concepts
and even more by specific political aims. Genocide is practiced by regimes and
armed groups that hardly have totalitarian ambitions or capabilities; conversely,
today’s Western states certainly ‘manage life’ but they do not practice genocide.
32 Michel Foucault, The Will to Knowledge: The History of Sexuality, vol. 1, trans. R. Hurley
(Harmondsworth: Penguin, 1998), 136–7.
33 Ibid. 136–7.
34 Ibid. 136.
35 Ibid. 143.
36 Ibid. 138.
152 martin shaw
Foucault’s linkages were suggestive but they didn’t explain the rise of genocide or
why genocides are initiated.
Nevertheless, Foucault’s formulations captured a widespread belief in the
enabling implication of modernity. This view of modern society was seminally
explored by Bauman, for whom ‘the Holocaust was an outcome of a unique
encounter between factors by themselves quite ordinary and common’37 and ‘a
rare, yet significant and reliable, test of the hidden possibilities of modern society’.38
He contended ‘that every “ingredient” of the Holocaust—all those many things
that rendered it possible—was normal . . . in the sense of being fully in keeping with
everything that we know about our civilization, its guiding spirit, its priorities, its
immanent vision of the world.’39 Bauman specified this ‘modern’ focus in terms of
mentality, technology, and organization, building on Raul Hilberg’s conclusion
that the machinery of destruction ‘was structurally no different from organized
German society as a whole. [It] was the organized community in one of its special
roles.’40 The Holocaust was a textbook case of ‘scientific management’, ‘a paradigm
of modern bureaucratic rationality’.41 The department in the SS headquarters in
charge of the destruction of European Jews entirely fitted Weber’s description of
modern administration:
in the last resort—the choice of physical extermination as the right means to the task of
Entfernung [removal, elimination] was a product of routine bureaucratic procedures: means-
end calculus, budget balancing, universal rule application. To make the point sharper still—
the choice was an effect of the earnest effort to find rational solutions to successive
‘problems,’ as they arose in the changing circumstances. . . . it arose out of a genuinely
rational concern, and it was generated by bureaucracy true to its form and purpose.42
Moreover, bureaucracy provided the ‘moral sleeping pills’ that made possible the
Holocaust’s ‘technical-administrative success’.43
Yet like Foucault’s idea, Bauman’s concept of the Holocaust was over-rationalized.
Although he recognized that only ‘[t]he possibility of the Holocaust was
rooted in certain universal features of modern civilization; its implementation on
the other hand, was connected with a specific and not at all universal relationship
between state and society,’44 his account narrowed down its meaning and made it
difficult to compare with other equally modern genocides. When he wrote that ‘the
Holocaust left behind and put to shame all its alleged pre-modern equivalents,
exposing them as primitive, wasteful and ineffective by comparison. . . . It towers
37 Bauman, Modernity and the Holocaust, xiii.

38 Ibid. 12.
39 Ibid. 8.
40 Raul Hilberg, quoted ibid.
41 Ibid. 149.
43 Ibid. 26.
high over the past genocidal episodes in the same way as the modern industrial plant
towers above the craftsman’s cottage workshop,’45 he focused entirely on the indus-
trialized extermination of the Final Solution. We got little sense of the activities of
the Einsatzgruppen, shooting huge numbers alongside improvised ditches and
burning them alive in their villages. Yet this direct slaughter was the first phase of
extensive mass murder, the decisive shift of gear that paved the way for Auschwitz.
The activities of the Einsatzgruppen were not only closer to premodern para-
digms. They proved a more enduring model than the extermination camps for later
génocidaires. In Rwanda, notoriously, machine guns and machetes proved quite as
murderous as the gas chambers, without the need for bureaucracy on the German
scale (although the organizers did employ modern political organization and mass
media). In Bosnia, Serbian nationalists destroyed Bosniak and Croatian society
while slaughtering only a minority, through intimidation, expulsion, torture, rape,
and killing in improvised concentration camps. In contemporary destructions of
indigenous peoples, the methods and organization have been much more basic
than the Final Solution’s. Not surprisingly, a student of these genocides has argued
that Bauman’s ‘thesis linking genocide to a specific level of state formation,
technological efficiency, rationality, and subjectivity is belied’ by other examples.46
Such critiques have led to an alternative version of the modernity thesis.
According to the social anthropologist Nancy Scheper-Hughes, genocide is not
an exceptional, high-bureaucratic realization but an endemic feature of modernity.
She suggested ‘a genocide continuum’ composed of ‘small wars and invisible
genocides’ conducted ‘in the normative social spaces of public schools, clinics,
emergency rooms, hospital wards, nursing homes, court rooms, prisons, detention
centers, and public morgues. The continuum refers to the human capacity to
reduce others to nonpersons, to monsters, or to things, that gives structure,
meaning and rationale to everyday practices of violence.’ The human species had
developed ‘a genocidal capacity’ seen in ‘all expressions of social exclusion, dehu-
manization, depersonalization, pseudo-speciation that normalize atrocious behav-
ior and violence toward others’.47 The kernel of truth in this argument is that once
we recognize the possibility of genocidal outcomes it is alarmingly easy to imagine
how everyday social relations could facilitate them. But once again these can only
be enabling conditions. The key question is, in what circumstances can devaluing
and dehumanization mutate into systematic, organized violence? Since the former are
common and the latter relatively exceptional, this kind of argument from moder-
nity to genocide shares the over-abstraction of Bauman’s version that it criticizes.
45 Ibid. 89.
46 Nancy Scheper-Hughes, ‘Coming to Our Senses: Anthropology and Genocide’, in Hinton (ed.),
Annihilating Difference, 366.
47 Ibid. 369.
154 martin shaw
A S O C I O LO G Y OF ‘E T H N I C C L E A N S I N G ’
................................................................................................................
In order to overcome this problem, a closer specification of causality is necessary,
and this in turn implies a sociological engagement with the historical literature.
There is, of course, a substantial literature of comparative genocide studies. Until
recently sociology’s contribution was very limited, but this has changed with
Mann’s ambitious sociology of ‘ethnic cleansing’. Mann is an unrivalled practition-
er of synthetic and comparative methods in historical sociology, and his richly
sourced study provides—in reality for the first time—a comprehensive sociological
explanation of the darkest side of modernity. He outlines eight theses that frame
detailed case studies of New World genocides, Armenia, Nazism, Communism,
Yugoslavia, and Rwanda.48
Mann’s key argument is that ‘murderous ethnic cleansing’ is the dark side not
simply of modernity but specifically of democracy. While all the four ‘sources of
social power’—economic, ideological, political, and military49—are involved, he
offers ‘essentially a political explanation’. His main case is that ‘cleansing is a hazard
of the age of democracy since amid multiethnicity the ideal of rule by the people
began to intertwine the demos with the dominant ethnos, generating organic
conceptions of the nation and the state that encouraged the cleansing of mino-
rities.’ Certainly ‘cleansing’ results less from democracy itself than from its perver-
sion: ‘Regimes that are actually perpetrating murderous cleansing are never
democratic, since that would be a contradiction in terms. . . . Indeed as escalation
proceeds, all perpetrating regimes become less and less democratic.’ And ‘regimes
newly embarked upon democratization are more likely to commit murderous
ethnic cleansing than are stable authoritarian regimes.’ In fact, ‘stabilized institu-
tionalized democracies’ are the least likely to commit ‘cleansing’, although they
often have it in their pasts, the more violently where ‘settler democracy’ took hold:
‘The more settlers controlled colonial institutions, the more murderous the cleans-
ing. . . . It is the most direct relationship I have found between democratic regimes
and mass murder.’ Although Mann labels ‘cleansing’ a product of ‘inter-group
relations’, he is clear that it is no simple product of ethnic differences: to become
murderous these need political expression.
Yet although this is a political account, Mann shows that genocide is not simply
statist. There are ‘three main levels of perpetrator’: radical elites running party-
states, violent paramilitary bands, and ‘core constituencies providing mass though
not majority popular support’. Ordinary people ‘are brought by normal social
48 Mann, The Dark Side of Democracy, 2–9; all quotations in the following paragraphs are taken
from these pages unless otherwise specified.
49 Mann developed this fourfold typology of power in The Sources of Social Power, 2 vols
(Cambridge: Cambridge University Press, 1986 and 1993).
structures into committing murderous ethnic cleansing’, and they have many
different motives. However, Mann recognizes that ‘[e]thnic cleansings are in
their murderous phases usually directed by states, and this requires some state
coherence and capacity.’ Radical party elites are crucial because they homogenize
diverse social forces, welding them into more cohesive blocs, and centralize nor-
mally fragmented state structures in the pursuit of exceptional goals. This happens
through political struggle, and typically through a series of escalations. The danger
zone exists ‘when (a) movements claiming to represent two fairly old ethnic groups
both lay claim to their own state over all or part of the same territory and (b) this
claim seems to them to have substantial legitimacy and some plausible chance of
being implemented.’ Escalation to the brink happens
when one of two alternative scenarios plays out. [Either] The less powerful side is
bolstered to fight rather than to submit . . . by believing that aid will be forthcoming from
outside . . . [Or] The stronger side believes it has such overwhelming military power and
ideological legitimacy that it can force through its own cleansed state at little physical or
moral risk to itself.
Finally, going ‘over the brink’ into actual perpetration occurs ‘where the state
exercising sovereignty over the contested territory has been factionalized and
radicalized amid an unstable political environment that usually leads to war.’
Mann’s account challenges the absolutist concept of ‘intention’, echoing Fein’s
insistence on a flexible ‘sociological concept of purposeful action’. He demonstrates
convincingly that an entire historical episode of murderous politics, over many
months or years, cannot be explained by singular intentionality: ‘Murderous
cleansing is rarely the original intent of perpetrators. . . . [It] typically emerges as
a kind of Plan C, developed only after the first two responses to a perceived ethnic
threat fail . . . To understand the outcome, we must analyze the unintended con-
sequences of a series of interactions yielding escalation.’ Instead of interpreting
murder as the direct consequence of longstanding intentions, he emphasizes the
contingency of violent outcomes: ‘Out of . . . political and geopolitical crises radi-
cals emerge calling for tougher treatment of perceived ethnic enemies. In fact,
where ethnic conflict between rival groups is quite old, it is usually somewhat
ritualized, cyclical and manageable. Truly murderous cleansing, in contrast, is
unexpected, originally unintended, emerging out of unrelated crises like war.’
Although in this richness and complexity Mann’s is a highly plausible sociological
framework, problems remain. He disarms criticism by acknowledging that ‘[g]iven
the messiness and uniqueness of societies, my theses cannot be scientific laws.
They do not even fit perfectly all my case studies.’ Yet the coherence of his detailed
explanations too frequently appears to be, at least partially, at odds with his general
theses. His ‘ethnic competition’ framework works well for Rwanda and former
Yugoslavia, but it does not fit as well the major cases of twentieth-century political
mass murder, such as the Holocaust. These very darkest episodes appear to have
156 martin shaw
only tenuous connections to democracy. They are products, Mann acknowledges,

of ‘a few highly authoritarian regimes’ that ‘deviate’ from the norm of stable
authoritarianism, and ‘mobilize majoritarian groups into a mass party-state mo-
bilizing the people against “enemy” minorities’. Although Hitler had indeed per-
verted democracy, it is difficult to fit the Holocaust, Mann recognizes, into his
pattern of ethnic competition. Germans and Jews were ‘two fairly old ethnic
groups’, but it was hardly the case that they ‘both laid claim to their own state
over all or part of the same territory.’ Likewise the extensive mass murders of
Communist states are problematic. Mann plausibly describes many of them as
classicide, because their targets were social classes, and explains them as ‘mistaken
revolutionary projects’ rather than as ethnic conflict. But he also offers the perver-
sion of socialism as a class variant of the more common ethnic perversion of
democracy: ‘socialist ideals of democracy also became perverted as the demos
became entwined with the term proletariat, the working class, creating pressures
to cleanse other classes.’ Along with the perversion of national democracy, this was
then a second ‘general way in which democratic ideals were transmuted into
murderous cleansing’. However, we may question whether the idea of the proletar-
iat (working class) was really a moving force in creating ‘pressures to cleanse other
classes’.
A more fundamental question is whether Mann correctly identifies the main
explanatory locus of murderous policies among his four types of power. He had
earlier provided the clearest argument that political and military are two separate
forms of power.50 That murderous power is primarily military does not of course
mean that its occurrence should be explained simply by military developments. Yet
Mann’s mainly political explanation is questionable. As he acknowledges, his theses
connecting ‘cleansing’ to democracy ‘apply beforehand, to the earlier phases of
escalation of ethnic conflict’. It is military power that ‘proves decisive in the later
stages of the worst cases of ethnic cleansing. Armies, police forces, and irregular
extrastate paramilitaries are the main agencies.’ Indeed, he provides detailed argu-
ments that should lead to a heavily ‘military’ explanation:
Most 20th-century cases of ethnic cleansing occurred during wars or during the chaotic
transfer from war to peace. . . . Ideologically tinged wars reduce shared rules and convert
civilians into enemies. . . . Civil wars and wars of secession with a strong ethnic component
are dangerous for ethnic groups trapped behind enemy lines. The lure towards murderous
ethnic cleansing increases when it can be accomplished at low military cost, with little fear
of retaliation. . . . Military campaigns may generate tactical lure towards atrocities against
civilians that were not originally intended. . . . Guerrilla warfare lures guerrillas to kill
civilians. . . . These [and others] are all features of military power that may produce mur-
derous cleansing.51
50 Ibid. i.10–11.
Yet at the most general level of Mann’s analysis, in formulating his theses and
locating his explanation in political power, this role of war is strangely neglected.
So we need to go further: genocide generally (although not always) arises from and
is carried out through war. Most genocide is actually part of genocidal war.52
Murderous politics do not arise primarily from political power relations, but out
of political power that is fundamentally conditioned by military power: out of the
interactions of political and military power.
T H E P RO B L E M OF C O N C E P T UA L P RO L I F E R AT I O N
................................................................................................................
The limitations of Mann’s framework are conceptual as well as analytical. Although
‘ethnic cleansing’ and ‘genocide’ are highly contested terms, he does not fully
justify the adoption of ‘cleansing’ as a master concept. This is unfortunate, since
the term is widely rejected as a perpetrator euphemism unsuitable for social-
scientific use. As Norman Naimark, a historian who uses the term, remarks:
‘There is nothing “clean” about ethnic cleansing. It is shot through with violence
and brutality in the most extreme form.’53 Mann defines it as ‘the removal by
members of one [ethnic] group of another such group from a locality they define as
their own.’ His insistence that ‘murderous cleansing’ is a subtype gives credence to
the notion that removal can be non-murderous or non-violent—a manifestation of
political but not military power. Yet, the wholesale removal of a population group
from their homeland is generally involuntary, resisted, and enforced through
extreme coercion. The forms and extents of violence vary greatly, but ethnic
removal generally falls under Lemkin’s original concept of genocide as ‘the destruc-
tion of a nation or of an ethnic group’.54
Mann has adopted the 1990s reinvention of genocide as ‘ethnic cleansing’ and
the corresponding narrowing of genocide itself to the intentional murder of all the
members of a group. His work in this sense fits into a wider pattern of scholarship,
as well as political and legal commentary, that adopts a narrow concept of genocide
as exclusively defined by premeditated mass killing, and therefore uses other
terms—not only ‘ethnic cleansing’ and ‘classicide’ but also ‘ethnocide’, ‘politicide’,
‘gendercide’, ‘cultural genocide’, ‘urbicide’, etc.—to refer to aspects of what Lemkin
considered under a broad definition of genocide. ‘Genocide’ then refers to a
52 For more support for these assertions, see my War and Genocide (Cambridge: Polity 2003),
Chapter 2.
53 Norman Naimark, Fires of Hatred: Ethnic Cleansing in Twentieth-Century Europe (Cambridge,
MA: Harvard University Press, 2001), 193.
158 martin shaw
specific type of political violence, rather than a broad class of violent actions, and
another term is needed to act as the master concept that Lemkin originally
intended ‘genocide’ to be. In Mann’s case, ‘ethnic cleansing’, refined as ‘murderous
cleansing’, assumes this broad role; but non-sociological writers have produced
other proposals, such as Rudolph Rummel’s idea of ‘democide’.55 From the point of
view of genocide studies, there is a strong case that the many new ‘-cide’ terms are
unimaginative imitations of Lemkin’s original word, and that these ‘-cides’ should
be understood as ways of referring to the many sides of genocide. Certainly, the
rapid conceptual proliferation has been accompanied by confusion over the mean-
ings and relationships of terms, rather than improved understanding. Understood
broadly and sociologically, suitably refined and expanded, ‘genocide’ itself can
provide a coherent framework for understanding the range of organized violence
targeted specifically at civilian populations.56
T OWA R D S A S T RU C T U R A L C O N C E P T OF GENOCIDE
................................................................................................................
However, if the idea is to function in this fashion, genocide studies need a new
phase of conceptual discussion to overcome the present state of contestation and
confusion. While many case studies, historical and contemporary, adopt
conflicting definitions according to the circumstances of particular cases, and
others either implicitly or explicitly use the Holocaust as a standard (although
one case could never act as an adequate definition of the general concept), there is a
strong need for greater conceptual clarity. This is a task to which sociology can
make a crucial contribution. As Weber suggested, there is a crucial difference
between explanations of individual events and those of general phenomena,
which defines the respective roles of the interrelated subjects of history and
sociology. Sociology is ‘a generalizing, regularity-seeking discipline, rather than
an “idiographic” one, and its method [is] essentially typological,’ although it is also
‘interdefined’ with history, and facilitating ‘the causal analysis of singular historical
phenomena’.57 Therefore as Thomas Burger has put it, the difference between
sociology and history ‘is not just one of degree. . . . For the task of sociology is
55 Rudolph Rummel, Death by Government (Brunswick, NJ: Transaction Publishers, 1997).

56 The argument in this and the preceding paragraph is developed in my What is Genocide?,
chs 4 and 5.
57 Weber, Theory of Social and Economic Organization, 161. Thus many argue that history and
sociology need to be combined in a ‘sociological history’ or a ‘historical sociology’. See C. Wright
Mills, The Sociological Imagination (London: Oxford University Press, 1959); Philip Abrams, Historical
Sociology (Shepton Mallet: Open Books, 1982).
the construction of a special kind of general concept whereas the goal of history is the
formation of individual concepts.’58 Moreover, without clear definitions, words
would be ‘vague thought-images created to meet the unconsciously felt need for
adequate expression whose meaning is only concretely felt but not clearly thought
out.’59 Clearly, defined terms enabled ‘the conceptually clear description of indi-
vidual historical developments, whenever this is necessary and the merely felt
mental picture is too unprecise [sic] for a particular purpose in question.’60
From the point of view of sociological concept formation, genocide studies are
mostly stuck at the first stage, of defining the phenomenon by the subjective
meaning of the action for the actors themselves (and of course for a particular
group of actors, the ‘perpetrators’). Moreover, we have seen that there has been
a tendency, heavily influenced by the legal discourse, to identify this subjective
meaning with a rarefied concept of a singular, absolute ‘intention’. Sociologists like
Fein and Mann have provided a more satisfactory way of understanding intention,
but this is only a beginning. As Weber explained, understanding the subjective
meaning of action (in his well-known term Verstehen) is not in itself an analytical
method, but a starting point from which it is necessary to proceed to causal
explanation through the development of structural concepts of the relationships
between actors. So we need not just a way of representing the subjective meaning of
actions, but models, ‘mental constructs, consisting of a number of elements
standing in particular relationships to each other, which are designed to represent,
for descriptive purposes, a specific segment of empirical reality and the interde-
pendencies existing in it.’61 Genocide should be defined not only by the ‘intentions’
of ‘perpetrators’, but by the social relationships between them and the ‘victim’
group, including the relations within each of these categories of actor, and relations
to other actors, conventionally described as ‘bystanders’. The patterns of these
relationships define genocide not just as a type of social action (‘committed’ by
‘perpetrators’), but as a structure of conflict between actors. This approach will lead
us, of course, to transcend ideas of genocide as ‘one-sided’ action, for although
genocide is certainly a very asymmetrical type of conflict, the responses of the
targeted group, for example resistance, and of other actors, for example in armed
intervention, are also of crucial importance to understanding it. Only if genocide is
generally understood as a structural form will it be possible to develop coherent
explanations. Grasping genocide itself as a structure will enable it to be seen in
relation to other structures—cultural, economic, political, and military—within
which genocidal relations develop.
58 Thomas Burger, Max Weber’s Theory of Concept Formation: History, Laws and Ideal Types
(Durham, NC: Duke University Press, 1987), 138. Emphasis added.
59 Ibid. 136.
60 Ibid.
160 martin shaw
I have therefore proposed that we need to define genocide as a type of unequal

social conflict between two sets of actors, which is conditioned in the first
instance by the type of action carried out by the more powerful side. Thus
genocide as a structure should be understood as ‘a form of violent social conflict
or war, between armed power organizations that aim to destroy civilian social
groups and those groups and other actors who resist this destruction’. Genocidal
action (or genocide as action, a sense closer to previous understandings) can be
defined as ‘action in which armed power organizations treat civilian social groups
as enemies and aim to destroy their real or putative social power, by means of
killing, violence and coercion against individuals whom they regard as members
of the groups.’62
T H E D E S T RU C T I O N OF ‘G RO U P S ’
................................................................................................................
Sociology has also helped clarify the vexed question of the ‘group’ element of
genocide. Although some have proposed that any mass killing should be seen as
genocide, the idea of group destruction has been central to the concept. Indeed it is
difficult to see how intentional harm against a large number of people could not be
informed by a ‘group’ idea. In legal and other discourse, there has been a great deal
of discussion of different types of group—whether the ‘ethnic, national, racial, and
religious’ groups of the Convention refer to different categories, or are different
ways of describing the same thing; and whether ‘political’, ‘social’ (‘class’) and other
groups should be added to the protected list. Sociologists have, as we have seen,
short-circuited this discussion by proposing a generic concept of group (Fein) and
adding that attacked groups are subjectively defined by perpetrators (Chalk and
Jonassohn). So we can say that genocide involves the attempt to destroy a group of
people, regardless of how far groups defined by perpetrators correspond to ‘real’
groups that are intersubjectively recognized by their members or objectively
identifiable by observers. I have argued that the one commonality of all target
groups in genocide is that they are all essentially civilian populations, largely
unarmed in the face of armed violence.63
However, the ‘group’ element leads to another key issue, the relationship
between group ‘destruction’ and harm towards individuals. Lemkin’s use of the
common suffix ‘-cide’ suggested that the destruction of groups was similar to the
killing of individuals denoted by established terms like infanticide and patricide.
62 Shaw, What is Genocide?, 154.

63 Ibid. Chapter 8.
Genocide involved the ‘killing’ of a social group, as the UN General Assembly

made clear when it asserted, ‘Genocide is a denial of the right of existence of
entire human groups, as homicide is the denial of the right to live of individual
human beings.’64 Yet this powerful analogy, like all analogies, had a potential to
mislead. At that time, there was a greater propensity than today to consider
societies as organisms, in which the functioning of the whole determined the
activities of the parts, and the well being of the individual was subordinated to the
whole. This view was formulated most starkly in fascist, Stalinist, and indeed
genocidal ideologies, but there was a more general belief in the holistic nature of
national societies and individuals’ subordination to them. This holistic tendency
was represented in sociology, too, but in reality human groups or societies were
not so like biological organisms. Human societies were constituted through social
relations between individual human beings: ‘societies’ and ‘groups’ are ways of
thinking about how these relations bind individuals together. As Weber had
written,
For other cognitive purposes . . . it may . . . be convenient or even indispensable to treat
social collectivities . . . as if they were individual persons. . . . But for the subjective inter-
pretation of action in sociological work these collectivities must be treated as solely the
resultants and modes of organization of the particular acts of individual persons, since
these alone can be treated as agents in a course of subjectively understandable action.65
Since social groups are not like individuals, it follows that ‘destroying’ them is not
like killing embodied human beings, either. We have seen that there is a powerful
trend in genocide theory to reduce ‘destruction’ to its ‘physical’ and ‘biological’
dimensions. The logical conclusion is that genocide is seen simply as mass killing.
However, this position is sociologically incoherent even in its own terms. Because
groups are social constructions, they can be neither constituted nor destroyed
simply through the bodies of their individual members. Destroying groups must
involve a lot more than simply killing, although killing and other physical harm are
rightly considered important to it. The discussion of group ‘destruction’ is obliged,
then, to take seriously Lemkin’s ‘large view of this concept’66—discarded in geno-
cide’s reduction to body counts—which centred on social destruction. A more
appropriate way of interpreting group destruction is therefore to see it as involving
a nexus between the destruction of collective ways of life and institutions and
bodily and other harm to individuals.
64 UN General Assembly Resolution 96(I), 11 December 1946, quoted by William A. Schabas,

Genocide in International Law (Cambridge: Cambridge University Press), 45. Emphasis added.
65 Weber, Theory of Social and Economic Organization, 101.
66 Schabas, Genocide in International Law, 228.
162 martin shaw
C O N C LU S I O N
................................................................................................................
This chapter has argued that sociology can make central contributions to genocide
studies. As a generalizing, concept-forming discipline it can help clarify the still-
contested meaning of genocide and clarify some of its theoretical difficulties.
Through historical-sociological work it can contribute to the comparative study
of genocides. So far its contributions have been modest, but some important work
has begun to overcome the sharply critical judgements of the discipline made by
Zygmunt Bauman and others two decades ago. There is reason to believe that the
terrible crimes of social classification that have marred modernity are at last
beginning to receive an appropriate response from those whose professional
concern is the critical analysis of social forms.
SELECT BIBLIOGRAPHY
Bauman, Zygmunt, Modernity and the Holocaust (Cambridge: Polity, 1989).
Chalk, Frank, and Kurt Jonassohn, The History and Sociology of Genocide: Analyses and Case
Studies (New Haven, CT: Yale University Press, 1990).
Fein, Helen, ‘Genocide: A Sociological Perspective’, Current Sociology 38:1 (1990).
Horowitz, Irving Louis, Taking Lives (New Brunswick, NJ: Transaction Books, 1979).
Kuper, Leo, Genocide: Its Political Uses in the Twentieth Century (Harmondsworth: Penguin,
1981).
Mann, Michael, The Dark Side of Democracy: Explaining Ethnic Cleansing (Cambridge:
Shaw, Martin, War and Genocide (Cambridge: Polity, 2003).
—— What is Genocide? (Cambridge: Polity, 2007).
chapter 8
.............................................................................................
POLITICAL
SCIENCE AND
GENOCIDE
.............................................................................................
scott straus
................................................................................................................
The discipline of political science has been an important arena for scholarly
innovation about the study of genocide. Political scientists penned several land-
mark qualitative and quantitative comparative studies of genocide; political scien-
tists have proposed several influential theories of genocide; and political scientists
have been instrumental in broadening the concept of genocide. More generally,
genocide is almost always a political phenomenon. The occurrence of genocide is
usually closely tied to the state, either directly when national governments act as the
chief instigator and perpetrator of violence or indirectly when national govern-
ments enable private actors or local officials to commit mass violence. Genocide is
also political in that the origins and dynamics of the phenomenon generally entail
struggles for power, the containment of threat, and control over the distribution of
goods and property. These are political processes, and thus a discipline devoted to
the study of politics has a set of natural advantages when approaching the study of
genocide.
At the same time, the actual fit between political science and genocide studies
has hardly been seamless. For many years, the study of genocide remained firmly at
164 scott straus
the margins of the discipline, as some of the first political scientists to study
genocide attest.1 That began to change in the 1990s, with the onset of several
high-profile cases and a general geopolitical shift after the end of the Cold War.
The net result is that analysis of genocide as a political phenomenon—and related
topics such as genocide prevention, intervention to stop genocide, and justice after
genocide—have received considerably more visibility in the discipline.
Nonetheless, despite this increased attention, there is a not a standard political
science paradigm when it comes to studying genocide. As a discipline, political
science is theoretically and methodologically diverse, and what can be said of the
discipline may also be said of political scientists’ studies of genocide. More specifi-
cally, no theoretical consensus with regard to the structural conditions, triggers, or
causal dynamics that drive, facilitate, or accelerate genocidal violence has emerged.
There is also not a standard research design that emerges from political science;
rather, political scientists have developed multiple different approaches to the
study of genocide, ranging from single-country studies, to comparisons of a
limited number of countries, to cross-national quantitative studies. Some political
scientists focus their explanations on macro-social conditions and states; others
focus on micro-level dynamics and units of analysis. In short, the diversity that
characterizes the discipline characterizes political scientists’ approaches to the
study of genocide.
Given the absence of theoretical and methodological convergence in the disci-
pline, this chapter focuses attention on three major themes in considering the
relationship between political science and genocide studies. In the first substantive
section, the chapter discusses the evolution of genocide studies within the disci-
pline and expands on the points raised above. In the second section, the chapter
identifies seminal contributions that have emerged from some four decades of
political science studies of genocide. In particular, I isolate five major areas of
innovation and concentration:
1) A methodological emphasis on the comparative method, including both quan-
titative and qualitative studies;
2) A move to broaden the concept of genocide using related but different terms;
3) A theoretical emphasis on regime type;
4) A theoretical emphasis on political leaders’ decision-making calculus—more
specifically, political scientists have been in the forefront of developing ratio-
nalist explanations of genocide; and
5) A theoretical emphasis on the connections between warfare and genocide.
1 Herbert Hirsch, ‘Studying Genocide to Protect Life’, in Samuel Totten and Steven Jacobs (eds),
Pioneers of Genocide Studies (New Brunswick, NJ: Transaction Publishers, 2002), 113–27; Roger Smith,
‘Who Is My Neighbor?’, in ibid. 179–93.
political science and genocide 165
In the third substantive section, I present some general critiques of political

science approaches and suggest avenues for future research in the discipline.
Before going forward, a few caveats are in order. The first is that the chapter is
not comprehensive. The main focus of the chapter is on political science explana-
tions of the origins and dynamics of genocide, with a bias towards political
scientists who consider genocide as a transhistorical and cross-national phenome-
non. Moreover, the focus is on aggregate contributions from political scientists and
thematic areas of emphasis—I do not seek to review every political science theory
of genocide. That means certain studies will not be discussed. It also means that
whole topics covered by political scientists will not receive sustained attention. One
major analytic emphasis among political scientists not discussed in the chapter
concerns the opportunities and obstacles that attend the prevention of genocide
and efforts to stop genocidal processes once they have begun. Often this literature
goes under the heading of ‘humanitarian intervention’, and in addition to scholars
of international law political scientists have been in the forefront of studies of
humanitarian intervention.2 Another major area of research concerns the after-
math of genocide, in particular the study of different judicial mechanisms of
accountability for genocide crimes. The literature generally operates under the
heading of ‘transitional justice’, and political scientists have been instrumental in
this area as well.3 I do not focus on either the humanitarian intervention literature
or the transitional justice literature for two main reasons. First, to keep the chapter
manageable, I concentrate on methodological and theoretical approaches to
2 Some of the most important work by political scientists on humanitarian intervention include
Michael Barnett, Eyewitness to a Genocide: The United Nations and Rwanda (Ithaca, NY: Cornell
University Press, 2002); Timothy Crawford and Alan Kuperman (eds), Gambling on Humanitarian
Intervention: Moral Hazard, Rebellion and Civil War (New York: Routledge, 2006); Herbert Hirsch,
Anti-Genocide: Building an American Movement to Prevent Genocide (Westport, CT: Praeger, 2002);
J. L. Holzgrefe and Robert Keohane (eds), Humanitarian Intervention: Ethical, Legal, and Political
Dilemmas (New York: Cambridge University Press, 2003); Roger Smith (ed), Genocide: Essays toward
Understanding, Early-Warning, and Prevention (Williamsburg, VA: College of William and Mary/
Association of Genocide Scholars, 1999); Benjamin Valentino, ‘Still Standing By: Why America and
the International Community Fail to Prevent Genocide and Mass Killing’, Perspectives on Politics 1:3
(2003), 565–76; Thomas Weiss, ‘Halting Genocide: Rhetoric vs. Reality’, Genocide Studies and
Prevention 2:1 (2007), 7–30; Nicholas Wheeler, Saving Strangers: Humanitarian Intervention in
International Society (Oxford: Oxford University Press, 2003).
3 For some examples of political scientists who study justice mechanisms, see Gary Bass, Stay the
Hand of Vengeance: The Politics of War Crimes Tribunals (Princeton: Princeton University Press, 2000);
Leigh Payne, Unsettling Accounts: Neither Truth Nor Reconciliation in Confessions of State Violence
(Durham, NC: Duke University Press, 2008); Victor Peskin, Trials and Tribulations: The Politics of War
Crimes Tribunals in the Balkans and Rwanda (New York: Cambridge University Press, 2008); Kathryn
Sikkink and Ellen Lutz, ‘The Justice Cascade: The Evolution and Impact of Foreign Human Rights
Trials in Latin America’, Chicago Journal of International Law 2:1 (2001), 1–33; Kathryn Sikkink and
Carrie Booth Walling, ‘The Impact of Human Rights Trials in Latin America’, Journal of Peace Research
44:4 (2007), 427–45; and Jack Snyder and Leslie Vinjamuri, ‘Trials and Errors: Principle and
Pragmatism in Strategies of International Justice’, International Security 28:3 (2003), 5–44.
166 scott straus
explaining why genocide occurs, which is a major focus in the political science
literature. Second, the now-extensive literatures on humanitarian intervention and
transitional justice are not limited to the prevention of genocide, but pertain to
sometimes different phenomena, such as humanitarian crises that emerge from
state collapse or apartheid.
A second caveat is that political science is not an exclusive domain for the themes
and contributions discussed in the chapter. The general rise in prominence of
genocide studies in political science is analogous to other kin disciplines, including
history, sociology, and anthropology. Moreover, there is considerable overlap of
thematic and methodological concerns in sociology and history. Sociologists and
historians, for example, have developed comparative approaches to the study of
genocide. Nonetheless, if not exclusive domains, the areas highlighted in the
chapter are intellectual zones where political scientists have made distinctive and
repeated contributions. These are the areas where political scientists have left and
are likely to continue to leave their mark on the study of genocide.
T H E S T U DY OF GENOCIDE AND THE DISCIPLINE

OF POLITICAL SCIENCE
................................................................................................................
During the past decade, there has been a general growth of interest in genocide, and
the political science discipline is not an exception to that trend.4 Papers on different
aspects of genocide are now common at the discipline’s major annual conferences
(such as the American Political Science Association and the International Studies
Association). In the past decade, studies of genocide have been published in the
major university presses for political scientists, as well as in some of the discipline’s
flagship journals, notably the American Political Science Review, International
Organization, and World Politics. I explore below some reasons for the upsurge of
scholarly interest among political scientists. However, it is important to note that
as genocide studies first developed in the 1970s and 1980s, the topic remained
largely marginal to the discipline. The point is not that political scientists were
absent from the early stages of the development of genocide studies—many were
quite instrumental—but rather that the study of genocide had very limited visibil-
ity within the discipline.
The reasons why the study of genocide was initially marginal are instructive.
First, however rigorous and systematic scholars may wish to make the study of
4 Scott Straus, ‘Second-Generation Comparative Research on Genocide’, World Politics 59:3 (2007),
476–501.
genocide, concern with the topic has almost inevitable normative dimensions.
While not alien to political science, normatively driven inquiries are not always
welcome within a discipline that prides itself on value-neutral, social scientific
scholarship. Second, until the mid-1990s, the topic of genocide was not prom-
inent in foreign policy or major media outlets, and political scientists often
take their cue from topics of general policy and public interest. Moreover,
genocide as an outcome remained for many political scientists strongly asso-
ciated with the Holocaust. As a perceived singular, single, and historical case,
the Holocaust did not attract substantial interest as representative of a broader
phenomenon.5 Third, as an international crime codified in a UN Convention
and as originally coined by a jurist (Raphael Lemkin), the concept of ‘genocide’
had strong legal origins and connotations, which in turn inhibited political
science scholarship on genocide. Finally, genocide studies emerged as (and
remains) an interdisciplinary area of research. Political scientists are not uni-
formly disinterested in interdisciplinary areas of inquiry, but the interdisciplin-
ary character of genocide studies kept the topic at some distance from the
centre of the discipline.
A marked change became apparent in the mid-1990s for a number of reasons.
First, the broader geopolitical context changed with the end of the Cold War.
A number of topics that were once somewhat marginal to political science became
topics of increased interest and analysis, including international human rights,
humanitarianism, civil war, and political violence. The interest in genocide is part
of this broader scholarly trend. Second, several high-profile conflagrations, which
the press, policymakers, and courts frequently labelled as ‘genocide’, were promi-
nent in the mid-1990s and later in the 2000s. The primary cases were the Balkans
series of wars in the mid- to late 1990s; the 1994 violence and civil war in Rwanda;
and Darfur in the mid-2000s. Each of these cases commanded significant public
and policy interest, and the result is that the topic of genocide was brought to the
attention of a wide array of scholars. The cases in turn generated political-science-
type questions concerning the causal dynamics that drive genocide and the
attendant search for commonalities among disparate cases. Finally, attention to
genocide became increasingly prominent in international politics. To take a few
examples: in 2004, the UN Security Council established a special advisor on the
prevention of genocide; in 2006, the United States incorporated genocide preven-
tion into the country’s National Security Strategy; and in 2008 a coalition of
prominent US-based think tanks and non-governmental organizations produced
5 Alan Zuckerman, ‘Political Science and the Jews: A Review Essay on the Holocaust, the State of
Israel, and the Comparative Analysis of Jewish Communities’, American Political Science Review 3:4
(1999), 935–45.
168 scott straus
a high-level report on genocide prevention chaired by Madeleine Albright

(a former Secretary of State) and William Cohen (a former Secretary of Defense).6
The newfound attention to genocide should not be overstated. Genocide is not a
top foreign policy concern or a top concern in international forums like the UN
Security Council or General Assembly. Genocide remains one of dozens of topics
analysed by political scientists who study international politics. And the public and
press concern with specific cases is episodic. Nonetheless, the topic commands
considerably more visibility and legitimacy than it ever has, and that has contrib-
uted to a large upsurge of research and political science publications.
T H E M AT I C A R E A S OF FOCUS AND I N N OVAT I O N

................................................................................................................
In this section, I discuss four-and-a half major areas of focus and innovation in
how political scientists approach the study of genocide. I also raise questions and
concerns about each of the areas. In order, the areas of focus are:
1) The comparative method (including attendant conceptual creativity—the ‘half ’
alluded to previously);
2) A theoretical interest in regime type;
3) A theoretical interest in rationalist explanations; and
4) A theoretical interest in the relationship between warfare and genocide.
The Comparative Method (and Conceptual Innovation)

Political scientists often seek to make generalizations about the causal factors that
drive a particular outcome, and a principal tool by which political scientists
identify generalizable causal patterns is the comparative method. The approach is
foundational for one of the main subfields within the discipline, that of compara-
tive politics, and indeed many political science studies of genocide are the product
of scholars within the subfield. The comparative method is also common and
important for scholars of international relations, which represents the other pillar
of political science studies of genocide.
Political scientists’ use of the comparative method to examine genocide takes
many forms. One is to examine a particular case and to make comparative general-
izations on the basis of that case. The method is evident in analysis of some seminal
cases, in particular Rwanda, the former Yugoslavia, and to an extent the
6 The report may be downloaded at http://www.usip.org/genocide_taskforce/index.html

Holocaust.7 A second comparative approach receiving increased use in the disci-

pline is to make within-country comparisons, in particular by comparing villages
or regions.8 The approach has the advantage of holding macro-level, national
factors constant while examining variation across regions, time periods, and
individuals. However, the most common comparative approach for political scien-
tists who study genocide and arguably within the broader discipline is to make
comparisons for which countries are the units of analysis. Even this approach is
varied: some political scientists conduct qualitative studies of a handful of
countries; other political scientists engage in qualitative research, but where a
dozen or more cases are compared; and finally some political scientists treat
genocide as a quantifiable dependent variable and use statistical methods to test
hypotheses.
Examples of each of these different approaches are evident in the political
science scholarship on genocide, ranging from the earliest studies to the most
contemporary. Leo Kuper’s 1981 book, Genocide: Its Political Use in the Twentieth
Century, is a founding work of scholarship in the genocide studies field.9 Kuper, a
political scientist, makes a number of wide-ranging observations in the book,
including durable insights about the politics of ratifying the UN Genocide Con-
vention as well as observations about obstacles to intervention. Kuper also
advances an argument that the structural condition that underlies different cases
across the twentieth century is the presence of deep divisions in plural societies.
The claim is that persistent social cleavages are the wellspring of the enmity,
dehumanization, and distrust that fuel genocide. To substantiate the claim,
Kuper uses the comparative method of the second type described above: he
presents a number of short case histories at a general level. He then observes
empirical and causal patterns and draws out his argument about deep divisions.
Another seminal work in the genocide studies field is a focused comparison of
the Holocaust and the Armenian genocide by political scientist Robert Melson.10
In analysing two cases, rather than multiple ones at a general level, Melson is able
to examine the cases in greater depth and to identify dynamic processes within
each case. Melson concludes that two major structural factors underlie the cases.
07 For examples, see V. P. Gagnon, The Myth of Ethnic War: Serbia and Croatia in the 1990s (Ithaca,
NY: Cornell University Press, 2005); Daniel Jonah Goldhagen, Hitler’s Willing Executioners: Ordinary
Germans and the Holocaust (New York: Random House, 1996); and Mahmood Mamdani, When
Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda (Princeton: Princeton
08 For example, see Lee Ann Fujii, Killing Neighbors: Webs of Violence in Rwanda (Ithaca, NY:
Cornell University Press, 2009), and Scott Straus, The Order of Genocide: Race, Power, and War in
Rwanda (Ithaca, NY: Cornell University Press, 2006).
09 Leo Kuper, Genocide: Its Political Use in the Twentieth Century (New Haven, CT: Yale University
Press, 1981).
10 Robert Melson, Revolution and Genocide: On the Origins of the Armenian Genocide and the
Holocaust (Chicago: University of Chicago Press, 1992).
170 scott straus
First, he argues for the importance of a prior revolution; revolutions have

multiple effects, but principally they redefine and divide political communities
into authentic and undeserving populations. Second, Melson argues that war is
significant. In wartime, states link enemies of the revolution to external wartime
enemies, thereby increasing the risk that the domestic ‘enemies’ will be targeted
for elimination. Melson’s argument about war remains influential and echoes
more contemporary research on genocide. Focused comparison is also found in
recent studies by political scientists Jacques Sémelin and Manus Midlarsky, each
of whom analyses three cases in depth and generates an argument on the basis of
the comparison.11
While political scientists have been in the forefront of qualitative comparative
research on genocide, the method is evident in kin disciplines. In their seminal
book for the genocide studies field, for example, historian Frank Chalk and
sociologist Kurt Jonassohn employ a methodological approach similar to that of
Kuper. They present a fairly general account of multiple different cases, which, in
their case, date to antiquity.12 Melson’s use of focused comparison of a small
number of cases is the methodological choice in important works by historian
Eric Weitz, psychologist Erwin Staub, and sociologist Michael Mann.13
The final comparative method discussed here, quantitative cross-national stud-
ies of genocide, is more distinctive among political scientists. Sociologist Helen
Fein has done quantitative research, comparing Jewish victimization rates across
Nazi-occupied territories during the Holocaust.14 However, the method is fairly
rare outside political science. Among the most influential political science quanti-
tative studies is that of Barbara Harff. In her quantitative research, Harff concep-
tualizes genocide as analogous to ‘politicide’ (systematic mass murder of a political
group). In an influential study in the American Political Science Review, Harff
analysed more than 30 cases of genocide and politicide between 1955 and 2001.
She concludes that six main risk factors help explain the onset of genocide and
politicide. The factors include not only upheaval, but also the existence of prior
genocides or politicides; elite ‘exclusionary ideology’; autocratic regimes; ethnic
11 Manus Midlarsky, The Killing Trap: Genocide in the Twentieth Century (New York: Cambridge
University Press, 2005); Jacques Sémelin, Purifier et détruire: Les usages politiques des massacres
et génocide (Paris: Le Seuil, 2005) (translated as Purify and Destroy: The Political Uses of Massacre
and Genocide, trans. Cynthia Schoch (New York: Columbia University Press, 2007)).
12 Frank Chalk and Kurt Jonassohn, The History and Sociology of Genocide: Analyses and Case
Studies (New Haven, CT: Yale University Press, 1990).
13 Michael Mann, The Dark Side of Democracy: Explaining Ethnic Cleansing (New York: Cambridge
University Press, 2005); Ervin Staub, The Roots of Evil: The Origins of Genocide and Other Group
Violence (New York: Cambridge University Press, 1989); Eric Weitz, Century of Genocide: Utopias of
Race and Nation (Princeton: Princeton University Press, 2003).
14 Helen Fein, Accounting for Genocide: National Responses and Jewish Victimization during the
minority rule; and low trade openness.15 A similar quantitative approach is also
found in the work of political scientist Matthew Krain, who bases his analysis on an
earlier data set of genocide and politicide put together by Harff and Ted Gurr.
Krain refers to his dependent variable as ‘state-sponsored mass murder’, and
he argues that ‘political opportunity structure’, especially civil war, is the key
determinant.16
A different approach is found in the work of political scientist Benjamin
Valentino. Like Harff and Krain, Valentino moves away from a strict focus on
genocide and anchors his research on the concept of ‘mass killing’, which refers to
events in which 50,000 civilians or more died in a five-year period.17 In co-
authored quantitative work (discussed in greater depth below), Valentino examines
mass killing events in the twentieth century, arguing that mass killing is most
commonly associated with counterinsurgency.18 Yet another approach is that of
Rudolph Rummel, who quantifies what he calls ‘democide’, which includes geno-
cide and other forms of state-orchestrated violence against civilians.19 In his book,
Rummel generates statistical counts of the numbers of dead across numerous
instances of democide in the twentieth century. In his study, quantification is less
a method for testing hypotheses cross-nationally, and more for determining the
volume of civilian death across different cases.
Two related but more sophisticated data sets on large-scale violence against
civilians have since been created. The first is the ‘One-Sided Violence’ data set
coming out of the Uppsala Conflict Data Program. Authors Kristine Eck and Lisa
Hultman define ‘one-sided violence’ as ‘civilians that are deliberately and directly
targeted by governments or non-state groups’.20 For data collection, the authors
employ a computer program that collects articles on individuals killed or injured
from five major international news media sources from 1989 to 2004. A similar data
collection project comes out of the University of Kansas, led by political scientist
Philip Schrodt. The Political Instability Task Force Worldwide Atrocities Dataset
captures any incident reported in four major media outlets in which at least five
civilian deaths were killed for a nominally political purpose.21
15 Barbara Harff, ‘No Lessons Learned from the Holocaust? Assessing Risks of Genocide and
Political Mass Murder since 1955’, American Political Science Review 97:1 (2003), 57–73.
16 Matthew Krain, ‘State-Sponsored Mass Murder: The Onset and Severity of Genocides and
Politicides’, Journal of Conflict Resolution 41:3 (1997), 331–60.
17 Benjamin Valentino, Final Solutions: Mass Killing and Genocide in the Twentieth Century (Ithaca,
NY: Cornell University Press, 2004).
18 Benjamin Valentino, Paul Huth, and Dylan BalCH-Lindsay, ‘ “Draining the Sea”: Mass Killing
and Guerrilla Warfare’, International Organization 58 (2004), 375–407.
19 Rudolph Rummel, Death by Government (New Brunswick, NJ: Transaction Publishers, 1994).
20 Kristine Eck and Lisa Hultman, ‘One-Sided Violence against Civilians in War: Insights from New
Fatality Data’, Journal of Peace Research 44:2 (2007), 235.
21 The media outlets are the Agence France-Press, the Associated Press, the New York Times, and
Reuters. For more information, please visit http://web.ku.edu/keds/data.dir/atrocities.html
172 scott straus
As these examples make clear, quantitative approaches to the study of genocide

generally entail collecting evidence on more than just genocide cases. One reason is
that there are not enough genocide cases to support a quantitative study of
genocide only. Genocide is thus paired with another outcome or treated as one
example of a broader phenomenon. That necessity has meant political scientists
have engaged in important conceptual innovation around the term genocide, and
here again we see a distinctive contribution from the discipline. Harff, for example,
pairs genocide with the concept of politicide; Krain refers to ‘state-sponsored mass
murder’; Valentino treats genocide as one form of mass killing; Rummel considers
genocide as one type of democide; Eck and Hultman conceptualize genocide
within the context of one-sided violence; and Schrodt and his collaborators
embed genocide within the study of ‘atrocity’. The move in this direction—to
think of genocide within a broad class of violence against civilians—is sensible, as
I argue in the Conclusion.
As promising as qualitative and quantitative comparative research is, several
problems are apparent.22 One common problem is that cases treated as similar are
in fact quite different. This is particularly evident in quantitative studies that code
heterogeneous events—ones that vary considerably in duration, magnitude of
violence, character of violence, socio-economic context, and historical period—
as having the same value. On the quantitative side, the Eck and Hultman study and
the Worldwide Atrocities data set present particularly attractive options for obviat-
ing some of the problems associated with unit heterogeneity. These data sets allow
genocide to be conceptualized along a continuous, rather than a dichotomous,
scale;23 they link the study of genocide to the broader phenomenon of organized
violence against civilians; they make observations on many more cases than other
studies do; and they identify a transparent and consistent method of data collec-
tion. In short, for quantitative approaches to the study of genocide—which is one
of the distinctive contributions of the political science literature—there is an
advantage to disaggregating the outcome as a form of deliberate violence against
civilians and collecting data along those lines.
For qualitative comparative studies, an area that has not received enough
attention is the study of ‘non-genocide’ or ‘negative’ cases. The modal qualitative
comparison is to examine cases of genocide that happen in different periods and in
different regions to discover commonalities among them. However, not enough
attention has been paid to the many cases that existing theory would suggest could
result in genocide but do not. Comparing genocide cases to non-genocide cases is
also a method for pairing cases that have fewer macro-level differences between
22 The points made in this section are similar to those made in Straus, ‘Second-Generation
Research’.
23 Krain also measures the ‘severity’ of state-sponsored mass murder; however, severity is measured
in terms of the duration of the violence, not the magnitude of violence in terms of numbers killed.
them. In other words, political scientists who study genocide qualitatively would
do well to compare more similar countries and enabling circumstances, but
countries and circumstances that have variation on the outcome of interest.
Doing so will help ideally to isolate the factors that drive genocide.
Regime Type
A trademark of political science is the study of the state, and a natural extension
among political scientists who study genocide (and related outcomes) concerns
how regime type does and does not influence the occurrence of genocide. In one of
the earliest general books on genocide, political scientist Irving Louis Horowitz
argues that the concentration of power is central to why genocide occurs. ‘Geno-
cide is the operational handmaiden of a particular social system, the totalitarian
system,’ writes Horowitz.24 Horowitz’s claim is deepened in Rummel’s work, in
which he argues that ‘absolute power’ constitutes the structural origin of demo-
cide. For both Horowitz and Rummel, a democratic system where there are checks
and balances on power serves as one of the best antidotes to genocide. Harff
similarly identifies authoritarian regimes as one of the six key risk factors that
can lead to genocide.25
The relationship between regime type and violence outcomes is a long-standing
concern in political science. Horowitz’s reference to the link between genocide and
totalitarianism is clearly indebted to the work of Hannah Arendt, a political
philosopher who famously wrote about totalitarianism in Nazi Germany and the
Soviet Union.26 Like the idea that deep divisions and enmity between groups is
a wellspring for genocide, the idea that genocide is more likely in dictatorial states
has a certain common-sense resonance. It seems intuitive mass killing would be
more likely where power is concentrated among the few and where citizens do not
enjoy broad democratic rights.
Nonetheless—and despite regime type being a distinctive concern among politi-
cal scientists—there are a number of concerns with the argument. Several authors
publishing in the past decade or so, in particular Krain, Valentino, and Mann, have
raised specific concerns with the argument. One concern is simply a problem of
timing: authoritarian regimes are often long-standing, yet genocides occur in
particular moments or periods of time. The causal factor is static, yet genocide is
a dynamic outcome. Another concern is that there are many more authoritarian
states than genocides. While one could argue that authoritarian political structures
24 Irving Louis Horowitz, Genocide: State Power and Mass Murder (New Brunswick, NJ:
Transaction, 1976), 36.
25 Harff, ‘No Lessons Learned’.
26 Hannah Arendt, The Origins of Totalitarianism (New York: Meridien, 1959).
174 scott straus
are a necessary condition for genocide to occur, they are certainly not sufficient.
A third concern is empirical. In his statistical tests, Krain does not find support for
the hypothesis that the concentration of power affects the onset of state-sponsored
mass murder. Mann argues that non-authoritarian states also commit what he calls
‘murderous ethnic cleansing’, and they do so in the name of democracy. Mann
identifies transitional regimes—ones moving between authoritarian and demo-
cratic—as the most likely political conditions to trigger genocide. Indeed, the most
recent cases of genocide are an uneven fit with the authoritarian/totalitarian
argument: in the Balkans and in Rwanda, the regimes that committed genocide
and mass violence were in the midst of democratic transitions and made ethnic
nationalist claims in the name of majoritarian rule.
Rationalist Explanations
A third area of distinctive contribution by political scientists is what might be called
a ‘rationalist’ approach to the study of genocide. Rationalist paradigms are com-
mon in the discipline. That is, rather than being the product of irrational, megalo-
maniacal leaders or the product of deeply prejudiced masses, the major claim is that
genocide is the outcome of deliberate, strategic actions of elites. This type of
rationalist argument is especially pronounced and well articulated in the contem-
porary work of Valentino. However, the framework is not new to Valentino. In a
piece published in the late 1980s, political scientist Roger Smith, for example, talks
of how genocide is ‘calculated to achieve the ends of the perpetrator’.27 In 1990s
work, political scientist George Andreopolos advances the notion of a ‘calculus of
genocide’; René Lemarchand talks of the ‘rationality of genocide’; and Krain
discusses how changing political contexts shape strategies that elites use against
challengers, in some cases facilitating state-sponsored mass murder.28 In other
words, a rationalist framework—albeit with different inflections and different levels
of emphasis—is a recurring theme among political scientists who study genocide.
The rationalist framework is, as noted above, most influentially presented in
Valentino’s work. In a 2004 book, Final Solutions, Valentino’s central theoretical
claim is that genocide and mass killing are the products of ‘strategic’ decisions by
leaders.29 He argues that analysis should focus at the elite level; mass publics, he
27 Roger Smith, ‘Human Destructiveness and Politics: The Twentieth Century as an Age of
Genocide’, in Isidor Walliman and Michael Dobkowski (eds), Genocide and the Modern Age:
Etiology and Case Studies of Mass Death (Westport, CT: Greenwood Press, 1987), 23.
28 George Andreopoulos, ‘Introduction: The Calculus of Genocide’, in idem (ed.), Genocide:
Conceptual and Historical Dimensions (Philadelphia: University of Pennsylvania Press, 1994), 1–28;
René Lemarchand, ‘Rwanda: The Rationality of Genocide’, Issue: A Journal of Opinion 23:2 (1995), 8–11;
Krain, ‘State-Sponsored Mass Murder’, 333.
29 Valentino, Final Solutions.
submits, are not necessary to perpetrate large-scale violence. Genocide and mass
killing are chosen as the best available means to achieve leaders’ most important
political and military goals. In subsequent co-authored work, Valentino pushes the
analysis further through collecting data and quantifying mass killing events in the
second-half of the twentieth century. Valentino and his collaborators argue that
mass killing is most common as counterinsurgency—as a response to guerrilla
warfare. In their words, mass violence is an effort to ‘drain the sea’, depriving rebels
the cover and support they need to challenge a state militarily.30
There are multiple implications of the rationalist approach. One is that the
remedy—genocide prevention—is not to reshape society but rather to change elite
incentives (or to change elites). Another implication is that a strongly emphasized
strategic framework moves analysis away from seeing genocide as the product of
deep divisions, enmity, pathology, and social stress. Valentino’s work, in which
strategic calculation is the primary prism by which to understand genocide, is thus
quite different from the work of Leo Kuper. A third implication is the ‘normaliza-
tion’ of genocide studies in that genocide and mass killing are not outside the ambit
of strategic, rational calculations that analysts perceive in other circumstances.
A related claim is made by Martin Shaw, who, as I discuss below, similarly wants to
‘normalize’ the study of genocide by embedding it in the study of war.31 Outside
political science, the theme is found elsewhere in contemporary scholarship,
notably the work of historian Donald Bloxham.32
Despite the welcome normalization of the study of genocide, there are important
tensions and problems with rationalist approaches. Genocide and mass killing are,
at a minimum, risky, extreme, and expensive responses to threats. They cost money
and reputation; in the contemporary world, they invite prosecution; and, more-
over, genocidal and mass killing policies often fail: the architects of mass violence in
the late Ottoman Empire, Germany, Serbia, and Rwanda all lost power immediately
after their genocidal policies. Genocide is also deliberate—by definition, genocide
is in the realm of intentional, calculated, usually state-led violence. That being the
case, identifying the conditions that push political elites to promulgate highly risky
policies and pinpointing the kinds of objectives that push elites to extremes is a
research priority. Valentino emphasizes ‘final solutions’: elites choose mass vio-
lence after other policies have failed. He also points to different scenarios and
conditions, in which elites choose mass violence such as guerrilla warfare but also
ethnic conflict, and communist revolution. Rationalist approaches thus help make
genocide understandable and analysable, but rationalist approaches need to specify
30 Valentino et al, ‘ “Draining the Sea” ’.

31 Martin Shaw, What is Genocide? (Cambridge: Polity, 2007).
the Ottoman Armenians (Oxford: Oxford University Press, 2005).
176 scott straus
why political leaders opt for quite risky action and when the promotion of
genocide seems rational.
Warfare and Genocide

Valentino’s research points towards another recurring theme in the political science
literature on genocide. Indeed, if there was a common environment that political
scientists emphasize as conducive to genocide, that environment would be armed
conflict. The argument is most apparent in the work of Martin Shaw, a scholar of
international relations (whose orientation and writing clearly overlap with sociol-
ogy). Other political scientists reference the importance of armed conflict, even if
armed conflict is not the dominant focus. The emphasis in the work of French
political scientist Jacques Sémelin, for example, is on ideological constructs that
lead to mass violence and genocide. However, Sémelin argues that armed conflict is
a condition that creates anxiety and uncertainty, prompting political elites to
redefine political communities in ideological terms that can lead to genocide.
A linkage between war and genocide is found explicitly, as noted above, in the
work of both Melson and Krain. In Harff ’s early research, war is a type of social
upheaval that can lead to genocide.33 In my research on Rwanda, I argue that war
was a critical factor that shaped the origins and perpetration of the genocide; I also
argue warfare shaped why political elites chose mass violence in Darfur.34 Outside
political science, other scholars such as Mann and Weitz reference armed conflict as
a condition that has a causal relationship to genocide, even if, again, armed conflict
is not their focus. In sum, the emphasis on a connection between armed
conflict and genocide is a recurrent and distinct, though not exclusive, emphasis
in the political science literature on genocide.
The claim that war and genocide are interrelated is most pronounced in Shaw’s
work. In two distinct books, Shaw makes a strong and explicit connection between
war and genocide.35 In the first, War and Genocide, Shaw argues that genocide is a
distinct form of war and that the logic of warfare is intimately connected to the
logic of genocide. He contends that genocides are committed when an organized
armed force considers civilians to be enemies in war, and he links the prevalence of
genocide to a broader trend of ‘degenerate’ warfare. In a second book, Shaw
extends some of his observations about the connection between war and genocide,
33 Barbara Harff, ‘The Etiology of Genocides’, in Isidor Walliman and Michael Dobkowski (eds),
Genocide and the Modern Age: Etiology and Case Studies of Mass Death (Westport, CT: Greenwood
Press, 1987), 41–59.
34 Straus, The Order of Genocide; Scott Straus, ‘Rwanda and Darfur: A Comparative Analysis’,
Genocide Studies and Prevention 1:1 (2006), 41–56.
35 Martin Shaw, War and Genocide: Organized Killing in Modern Society (Cambridge: Polity, 2003);
Shaw, What Is Genocide?
but also examines and rethinks both the foundations of the concept of genocide
and how genocide should be studied. Shaw concludes that genocide is a ‘form of
violent social conflict’ in which armed groups seek to destroy civilian social
groups.36 In so doing, Shaw argues that genocide should be treated as a generic
concept, and he contends that genocide and ‘genocidal action’ should not be
considered exceptional, but rather connected to the more common phenomena
of war and conflict.
The theme of war and genocide is also evident in Midlarsky’s work, also a
specialist in international relations. However, Midlarsky sees genocide as excep-
tional, or at least highly infrequent. In the book, Midlarsky analyses three cases of
genocide in the twentieth century: the Armenian genocide, the Holocaust, and the
Rwandan genocide. Like other political scientists, his primary theoretical focus is
on the decision-making calculus of leaders, arguing that genocide is the product of
‘imprudent realpolitik’ brought on specifically by loss, usually territorial, in war-
time. By ‘imprudent’, Midlarsky emphasizes that the response to loss is ‘dispropor-
tionate’ to a perceived provocation or threat.37 Thus, Midlarsky adds a dimension
of irrationality or at least unwise decision-making in the name of state interests. In
that there is an important distinction from Valentino and others who stress the
deliberate, strategic origins of genocide. Midlarsky also brings in an international
dimension, arguing that leaders of states that commit genocide are facilitated by
allies abroad. In sum, Midlarsky’s precise theoretical innovations vary from
Valentino and Shaw, as does his methodology and case selection, but the connec-
tion between war and genocide is a running theme for all three contemporary
scholars—and thereby echoing the earlier research of political scientists Robert
Melson and Barbara Harff.
As robust as the connection between warfare and genocide is, the mechanisms
and reasons why armed conflict and genocide are connected are subject to different
interpretation. In the work of Valentino and his co-authors, the emphasis is on the
strategic use of mass violence to destroy insurgents’ civilian support bases. Mass
killing is a deliberate military strategy to defeat guerrillas who have strong civilian
support or who pose a major threat to a regime. Shaw argues in general that war
and genocide are entwined, such that genocide cannot be separated conceptually or
in explanatory terms from war. He argues more specifically that in war a ‘power
organization’ is more likely to define civilian populations as ‘enemies’ and that
‘power organizations’ tend to be militarized or at least that political power and
military power are closely connected.38 In Sémelin’s hands, war contributes to
defining some groups as internal enemies, and war increases uncertainty and
36 Shaw, What Is Genocide?, 154.

37 Midlarsky, The Killing Trap, 94.
38 Shaw, What Is Genocide?, 147.
178 scott straus
vulnerability, which can lead to the use of violence.39 For Midlarsky, war creates
conditions of state insecurity, and loss in war can lead to disproportionate
responses. In my work on Rwanda, I argue that war shaped the logic of genocide,
legitimized violence, created uncertainty and insecurity, thereby empowering hard-
liners over moderates, and led specialists in violence (soldiers, paramilitaries, and
militias) to enter the domestic political arena.
In sum, the political science literature indicates that war matters, but the
questions of how and why war matters remain an area for future research. Existing
theory highlights, variously, war type and dominant strategy (guerrilla versus
conventional warfare), the entwining of military and political arenas in war, the
effects of uncertainty and insecurity produced in war, and the labelling of civilian
populations as ‘enemies’ in war. However, an empirical problem in the war–
genocide linkage is that most civil and international wars do not lead to genocide.
Shaw argues that a theory of genocide should explain a range of organized violence
against civilians in war. Shaw wants to embed theories of genocide within a broader
universe of cases of what he calls ‘genocidal action’. At the same time, it is
important to recognize differences in outcomes—that is, there is an important
empirical difference between ‘genocidal action’ that leaves several hundred civilians
dead and ‘genocidal action’ in which hundreds of thousands are killed. Thus,
understanding why some armed conflicts lead to limited violence against civilians
while others lead to large-scale violence against civilians and genocide is a critical
question, but one that remains underdeveloped.
C O N C LU S I O N S AND N EW D I R E C T I O N S
................................................................................................................
In the development and evolution of genocide studies, political science has made
significant and lasting contributions. As (most often) a crime of state and
a function of policy choices and social mobilization, the fit between the study of
genocide and the discipline should come as no surprise. Nonetheless, the existing
political science literature is far from theoretical, conceptual, or methodological
convergence with regard to both the origins of genocide and how to study it. Given
the complexity of the outcome, as well as the growing yet still comparatively small
amount of political science research on genocide, the absence of consensus is not
necessarily a weakness. Rather, the lack of consensus presents an opportunity for
future research that seeks to build on the insights of past work. In the conclusion,
I take a step in that direction by making some observations about gaps in the
39 Sémelin, Purifier et détruire, 172, 178.

existing literature and by suggesting some avenues for future political science
research on genocide.
The first observation is that the study of genocide (and mass killing) has been
strangely cloistered from studies of kin phenomena, in particular the use of
violence in war. As noted above, the empirical and theoretical intersections
between genocide and armed conflict are strong. There exists a very large and
rapidly growing political science literature on civil war and the use of violence in
wars, both international and domestic.40 That literature has major theoretical
implications for the study of genocide, but in large respect the connections remain
underexplored. For example, in his seminal research on violence in civil war,
Stathis Kalyvas distinguishes between ‘discriminate’, targeted violence and ‘indis-
criminate’, widespread violence.41 Kalyvas’ attention to information and relative
control in the context of armed conflict carries direct potential insight into the
dynamics of genocide. More generally, his use of the concept of ‘indiscriminate’ is
certainly akin to how some scholars conceptualize ‘genocide’ and related terms,
such as mass killing and politicide. Political scientists who study genocide thus
need to be in greater theoretical engagement with this work, and vice versa.
Similarly, in quite different research on international conflict, Alexander Downes
analyses the determinants of ‘barbarism’—intentional targeting and violence
against civilians in wartime.42 Like ‘indiscriminate violence’, large-scale, intentional
violence against civilians is not far from the concept of genocide. Yet there remains
little theoretical engagement between the research on intentional targeting of
civilians in international war and genocide. Another example is that of ‘asymmet-
ric’ warfare, which designates the overwhelming power of one side in an armed
conflict.43 Given the connections between war and genocide and that victims
in genocide often have disproportionately less resources than perpetrators, exam-
ining the two phenomena would be worthwhile. Researchers who examine these
phenomena might underscore differences between genocide and ‘asymmetric’
war, barbarism, or indiscriminate violence in war, but doing so would sharpen
theory on genocide and initiate a scholarly conversation with students of related
phenomena.
The second observation is the need for greater attention to methodological
concerns. ‘Methodology’ is meant broadly, including attention to concepts, hy-
pothesis generation, and research design. While there remain quite different
concepts in use (from genocide, to mass killing, to politicide, to democide) within
40 Stathis Kalyvas, ‘Civil Wars’, in Carles Boix and Susan Stokes (eds), Handbook of Political Science
(New York: Oxford University Press, 2007), 416–34.
41 Stathis Kalyvas, The Logic of Violence in Civil Wars (New York: Cambridge University Press,
2006).
42 Alexander Downes, Targeting Civilians in War (Ithaca, NY: Cornell University Press, 2008).
43 Ivan Arreguı́n-Toft, How the Weak Win Wars: A Theory of Asymmetric Conflict (New York:
180 scott straus
the political science literature, most scholars examine a large-scale outcome. But a
relatively underexplored avenue would be to disaggregate the concept, identifying
researchable dimensions of genocide (or related terms). Doing so would connect
research to the literature on violence, but doing so also could lead to a focus on
escalation, mobilization, or the coordination of power.44 In short, disaggregating
genocide into smaller, researchable, but salient questions offers an avenue for
potential future research. Research design broadly conceived is another obvious
area of greater thinking. Given that genocide is relatively infrequent, what is the
best way to study the phenomena? How should negative cases be incorporated
into the analysis? How should (positive and negative) cases be selected? In
what ways might a multi-method (quantitative/qualitative) research design be
employed? How might insights from game theory apply to the study of genocide?
Could genocide be formally modelled? What types of data are needed to test
hypotheses? What would an empirical research on genocide look like beyond the
existing cross-national data on occurrence or level of violence in cases? These and
other methodological questions are essential for advancing political science
explanations of genocide, but to date they have received fairly little self-conscious
reflection in the political science literature.
The third observation is the relative absence of critical thinking about explana-
tion and causation. The most common causal model in political science is to look
for macro-level structural conditions that give rise to genocide. However, consis-
tent with the suggestion to disaggregate, theory might focus on a specific dimen-
sion of genocide, such as escalation or explaining the difference between low
violence and high violence. Similarly, the question of specifying connections
between top- and bottom-level policy is an area worthy of greater research. How
do national and local levels interact? How do domestic and international dimen-
sions interact? What explains the specific timing of when leaders decide to imple-
ment a policy of genocide? Another under-represented area concerns theorizing
constraint. That is, much theory focuses on conditions and incentives that would
push leaders to promulgate genocide. Much less attention is on the conditions that
restrain leaders from the use of mass violence.45 In short, political scientists might
consider the trade-offs in explaining genocide as a macro phenomenon or in
explaining certain dimensions of genocide.
44 In their works, Mann and Sémelin move in this direction as does Donald Bloxham in ‘Organized
Mass Murder: Structure, Participation, and Motivation in Comparative Perspective’, Holocaust
Genocide Studies 22 (2008), 203–45.
45 This is work that I have begun: see Scott Straus and Daniel Knudsen, ‘The Costs of Genocide:
Theorizing Restraint and Negative Cases in the Study of Mass Violence’, unpublished manuscript,
Department of Political Science, University of Wisconsin; available upon request.
SELECT BIBLIOGRAPHY
Harff, Barbara, ‘No Lessons Learned from the Holocaust? Assessing Risks of Genocide and
Political Mass Murder since 1955’, American Political Science Review 97:1 (2003), 57–73.
Horowitz, Irving Louis, Genocide: State Power and Mass Murder (New Brunswick, NJ:
Transaction, 1976).
Krain, Matthew. ‘State-Sponsored Mass Murder: The Onset and Severity of Genocides and
Politicides’, Journal of Conflict Resolution 41:3 (1997), 331–60.
Kuper, Leo, Genocide: Its Political Use in the Twentieth Century (New Haven, CT: Yale
Melson, Robert, Revolution and Genocide: On the Origins of the Armenian Genocide and the
Midlarsky, Manus, The Killing Trap: Genocide in the Twentieth Century (New York: Cam-
bridge University Press, 2005).
Rummel, Rudolph, Death by Government (New Brunswick, NJ: Transaction Publishers,
1994).
Sémelin, Jacques, Purifier et détruire: Les usages politiques des massacres et génocide (Paris: Le
Seuil, 2005).
Shaw, Martin. War and Genocide: Organized Killing in Modern Society (Cambridge: Polity,
2003).
—— What is Genocide? (Cambridge: Polity, 2007).
Straus, Scott, ‘Second-Generation Comparative Research on Genocide’, World Politics 59:3
(2007), 476–501.
Valentino, Benjamin. Final Solutions: Mass Killing and Genocide in the Twentieth Century
——, Paul Huth, and Dylan Balch-Lindsay, ‘ “Draining the Sea”: Mass Killing and Guerrilla
Warfare’, International Organization 58 (2004), 375–407.
chapter 9
.............................................................................................
A N T H RO P O LO G Y
AND GENOCIDE
.............................................................................................
kevin lewis o’neill
In the broadest of terms, anthropology is the study of humanity, but more

practically (although perhaps stated in a more obtuse way), anthropology is the
study of culture—the attitudes, behaviours, and practices that constitute a given
community. Anthropology’s method of choice has long been ethnography, which is
a qualitative research technique that places anthropologists amidst particular
cultures for extended periods of time to conduct formal and informal interviews,
to observe everyday life, and to examine archival materials when they are available.
Traditionally, Western anthropologists have pursued ethnographic research in
distant, far-off places to make the argument that ‘those people over there’ are
either quite different or remarkably similar to ‘us over here’. With an eye to the
colonial subplots that these assumptions carry, anthropology has changed consid-
erably over the past few decades, applying its deft ethnographic touch to such
slippery objects of study as globalization, nationalism, and technology.
Amidst this new wave of anthropology, an emerging object of study continues
to come into focus: genocide. While anthropology has long been silent on the
topic, recent scholarship has proven that a tremendous amount can be learned
from a comparative perspective when researchers interview and observe survi-
vors, perpetrators, aid workers, and government officials in genocidal and post-
genocidal contexts. The result has been an anthropology of genocide that has
been slow to start but is now quickly emerging as a particularly evocative subfield
of not just anthropology but also genocide studies. The anthropology of geno-
cide, in fact, lends analytical clarity and empirical rigour (as well as emotional
anthropology and genocide 183
texture) to a range of issues, including truth, memory, and representation in

post-genocidal spaces.
Anthropology’s growing interest in genocide has a number of roots, including a
continued interest in both modernity and globalization as well as violence and
terror; a shift from small village studies to research that examine the state-level
dynamics in situations of upheaval, flux, and violence; and a greater commitment
to reflexivity, historicity, and engaged anthropology. The questions that now
constitute anthropology’s growing interest in genocide include: What happens to
people and the societies in which they live during and after genocide? How are the
devastating events experienced and remembered on the individual and collective
levels? How do these memories intersect and diverge as governments in post-
genocidal states attempt to produce a more monolithic ‘truth’ about the past?
Another important cluster of questions reads: How are representations of a violent
past structured by one’s positioning as a survivor, perpetrator, journalist, or
ethnographer? And what are the epistemological, ethical, and empirical entangle-
ments in which researchers find themselves enmeshed in post-genocidal contexts?
The formation of these anthropological questions builds from several other intel-
lectual developments such as critical assessments of ethnography, nationalism,
violence, and refugees, but nonetheless continues to extend far beyond these issues
in rather creative and thought-provoking ways.
E T H N O G R A P H Y, N AT I O NA L I S M ,
VIOLENCE, REFUGEES
................................................................................................................
The anthropology of genocide has been able to develop as rapidly as it has because
of at least four vibrant anthropological literatures. They are critical reflections on
ethnography, nationalism, violence, and refugees. With regard to ethnography,
anthropologists have become keenly aware that their method of analysis does not
represent knowledge in an objective sense, but rather produces knowledge from a
subjective perspective.1 Ethnography is not only limited in its scope but is also
a powerful tool of knowledge production that too often carries an overextended
weight of authority. In critical response, contemporary ethnographic practices in
cultural and social anthropology stress that every representation is historical and
contestable, that anthropological observations are not the result of either an
1 Renato Rosaldo, Culture and Truth: The Remaking of Social Analysis (Boston: Beacon Press, 1989).
184 kevin lewis o’neill
omniscient perspective or absolute scientific clarity.2 While it would be possible to

see this epistemological anxiety as a threat to any study of genocide (because of its
affinity to cultural relativism),3 it is important to stress that attention to these
issues actually strengthens the anthropology of genocide with a unique sensitivity
to questions of representation, allowing the anthropologist of genocide to situate a
survivor or perpetrator’s testimony not only historically but also along lines of
ethnicity, class, and gender, for example.4
The anthropology of genocide has also gained a great deal from anthropological
considerations of the nation-state. Given the nation-state’s centrality in genocidal
processes, anthropologists have found it productive not to assume that the nation-
state exists with as much density as other social scientific perspectives might
suggest. While political scientists, for example, might tend to focus on the role of
the Guatemalan nation-state during acts of genocide against indigenous popula-
tions, the anthropologist might also ask how committing acts of genocide con-
tributed to the formation of a national identity—how acts of extraordinary
violence construct the cultural boundaries that determine who is and who is not
a Guatemalan. The anthropological contribution here is the idea that acts of
genocide are cultural practices that contribute to the formation of the nation-
state as an ‘imagined community’, in the words of Benedict Anderson.5 Anderson
famously argues that a nation is a socially constructed community constituted by
those who see themselves as a part of that group, and that a range of cultural
developments and practices contribute to the formation of the modern nation-
state as an imagined community. While Anderson focuses on print capitalism’s
ability to foster a common vernacular (and, in turn, a common national identity),
anthropologists of genocide explore genocidal acts as an equally potent (if not
more dramatic and especially heinous) means to imagine the nation. It is impor-
tant to note, however, that this process of identity construction can only continue
during the very process of genocide. Radicalization develops further among the
perpetrators even as they commit the act. Genocide can sometimes be a way of
testing limits and accidentally discovering how far perpetrators can go.6
2 James Clifford and George Marcus (eds), Writing Culture: The Politics and Poetics of Ethnography
3 Thomas Cushman, ‘Anthropology and Genocide in the Balkans: An Analysis of Conceptual
Practices of Power’, Anthropological Theory 4:1 (March 2004).
4 Beatriz Manz, Paradise in Ashes: A Guatemalan Journey of Courage, Terror, and Hope (Berkeley:
University of California Press, 2004).
5 Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism
(London: Verso, 1983).
6 Bette Denich, ‘Dismembering Yugoslavia: Nationalist Ideologies and the Symbolic Revival of
Genocide’, American Ethnologist 21:4 (May 1994), 367–90; Robert M. Hayden, ‘Imagined Communities
and Real Victims: Self-Determination and Ethnic Cleansing in Yugoslavia’, American Ethnologist 23:4
(November 1996), 783–801.
Third, the anthropology of genocide builds on the field’s increasingly sophisti-

cated approach to violence, which continues to explore the topic on at least two
different scales. The first is a routine, even everyday, kind of violence, which
includes identity politics.7 The violence of racial or class-based slurs, even assaults,
allows anthropologists to understand how individuals slog through the kind of
habitual violence that ultimately ‘primes’, in the words of Alex Hinton, genocidal
events.8 The second is an increasing attention to more large-scale events, such as
state-sponsored violence, and the cultural practices and occasions that constitute
these large events.9 Implied in these two levels of analysis is the anthropological
effort to situate violence historically, noting that violence neither emerges nor is
experienced ex nihilo. Rather, violence is imagined and performed amidst thick
historical and cultural contexts. Anthropology’s emphasis on human experience as
well as how violence becomes situated within a culture (and situated differently by
different people within that culture) allows anthropologists to gain purchase on
not just genocide’s cultural complexity but also genocide’s emotional depth.
Violence, for the anthropologist, exists as an odious form of cultural expression.10
Finally, anthropology’s growing attention to refugees has allowed the field to
gain an added (and rather productive) perspective on genocide studies. Anthro-
pology, it has been noted, long assumed a tight relationship between culture and
space.11 For quite some time, anthropologists could assume (in seemingly uncom-
plicated ways) that Rwandans, for example, live in Rwanda and amidst Rwandan
culture. This perspective has historically resulted in studies that elide the reality of
displacement, diaspora, and deterritorialization. Yet, genocidal events displaced
tens of thousands of Rwandans, for example, making ‘Rwandan culture’ a rather
complicated object of ethnographic study. Rwandans as refugees literally shifts the
content and context of what might be understood as Rwandan culture.12 Anthro-
pological approaches to genocide have since adjusted, building on an ethnographic
attention to those people who are in constant motion—who live at the margins and
amidst the borders of transnational contexts. A vivid example of this approach
comes from Liisa Malkki’s ethnographic research with Hutu refugees who lived in
7 Arthur Kleinman, ‘The Violences of Everyday Life: The Multiple Forms and Dynamics of Social
Violence’, in Veena Das, Arthur Kleinman, Mamphela Ramphele, and Pamela Reynolds (eds), Violence
and Subjectivity (Berkeley: University of California Press, 1997), 226–41.
8 Alex Laban Hinton, Why Did They Kill?: Cambodia in the Shadow of Genocide (Berkeley:
University of California Press, 2005); Veena Das, ‘Language and Body: Transactions in the
Construction of Pain’, in Arthur Kleinman, Veena Das, and Margaret Lock (eds), Social Suffering
(Berkeley: University of California Press, 1997), 67–91.
9 Nancy Scheper-Hughes and Philippe Bourgeois (eds), Violence in War and Peace: an Anthology,
(London: Basil Blackwell Press, 2004).
10 Neil Whitehead (ed.), Violence (Santa Fe, NM: School of Advanced Research Press, 2004).
11 Akhil Gupta and James Ferguson, ‘Beyond “Culture”: Space, Identity, and the Politics of
Difference’, Cultural Anthropology 7:1 (February 1992), 6–23.
12 Rogers Brubaker, Ethnicity without Groups (Cambridge, MA: Harvard University Press, 2004).
Tanzania after the 1972 genocide in Burundi.13 Malkki’s work details how Hutu
refugees understood the refugee category from their own perspective. As persons
who had long owned (in an emotional sense) the category of the refugee by living
in refugee camps and by self-identifying as refugees, Malkki demonstrates how the
Hutu saw themselves as a nation in exile.14 As will be detailed below, those persons
displaced by genocide fit squarely within today’s conceptualization of an anthro-
pology of genocide.
A N T H RO P O LO G Y A N D G E N O C I D E
................................................................................................................
It is difficult but important to recount the fact that before an anthropology of
genocide can be outlined in any detail, one needs to recognize the historical
relationships that have existed between anthropology and genocide. Given that
acts of genocide themselves gain legitimacy from ideologies that mix nationalism
with ethnicity to produce a threatening kind of difference, it may come as no
surprise that anthropology—a field of study that has been historically interested in
not only origin myths but also nationalism and ethnicity—has contributed to
genocidal activities. Anthropology’s participation in the Holocaust is one well-
documented case in point.
Bettina Arnold, for example, explains how archaeological research contributed
to the justification of genocide in Nazi Germany through the construction of
difference.15 Arnold makes the argument that archaeology’s cultural capital—its
authoritative voice as well as its ability to contribute to the ideological under-
pinnings of political systems—placed the practice of archaeology at the centre of
German National Socialism. National Socialists, she reports, employed archaeo-
logical research to define and reify nationalist and ethnic identities in Germany.
National Socialists, for example, saw race as the defining quality for one’s
membership in the Germanic community and this construction of race depended
most heavily upon the material record: ‘In the nineteenth and early twentieth
centuries, Germany was where Germans were or could be shown to have been.
13 Liisa Malkki, ‘Speechless Emissaries: Refugees, Humanitarianism, and Dehistoricization’,

Cultural Anthropology 11:3 (August 1996), 377–404.
14 Liisa Malkki, Purity and Exile: Violence, Memory, and National Cosmology among Hutu Refugees
in Tanzania (Chicago: University of Chicago Press, 1995).
15 Bettina Arnold, ‘Justifying Genocide: Archaeology and the Construction of Difference’, in Alex
Laban Hinton (ed.), Annihilating Difference: The Anthropology of Genocide (Berkeley: University of
California Press, 2002), 95–116. See also Bettina Arnold and Henning Hassmann, ‘Archaeology in Nazi
Germany: The Legacy of the Faustian Bargain’, in Phillip Kohl and Clare Fawcett (eds), Nationalism,
Politics and the Practice of Archaeology (Cambridge: Cambridge University Press).
Germans established territory by occupying it and leaving a distinctive material

record of their presence.’16 Archaeological research under the Third Reich was thus
able to construct its own imagined community and sphere of moral obligation
through the material record, while at the same time justifying the Holocaust through
the mapping of ethnicity and racial superiority. Through the creation and mainte-
nance of origin myths, archaeology facilitated both nationalism and genocide.
As for cultural anthropology, Gretchen Schafft’s From Racism to Genocide details
how anthropologists contributed to the needs of the Nazi state by allowing their
research on race to inform Nazi political views.17 While some argue that Schafft’s
argument is overstated,18 she insists that this kind of collaboration took place on
several different levels. Anthropologists, much like the archaeologists of Arnold’s
work, helped to solidify the German Reich, both through the creation of an
imagined community rooted in the past and with the production of a future,
even purer, gene pool. This kind of genocidal purification project took place by way
of anthropologists—scientists of race—who participated in sterilization cam-
paigns. Much of this research on race, Schafft argues, was meaningless, given that
‘there were no standards by which to judge Jewishness, so no conclusions ever
could have been drawn from the morphological measurements.’19 Yet, anthropol-
ogists did not act alone, or at least with their own funding. Schafft’s work also
establishes the transatlantic connections vis-à-vis international philanthropic
agencies, such as the Rockefeller Foundation, that supported anthropological
research on race during the Third Reich—even when it was obvious what kind of
violence this research supported.
This kind of anthropological rationalization of genocide, sadly, does not pertain
only to the Holocaust. Nancy Scheper-Hughes, for example, points towards the
anthropological justification of apartheid and its genocidal dimensions as well as
Alfred Kroeber’s silence in the face of the genocide that took the lives of Californian
Indians.20 In a dramatic sentence, Scheper-Hughes writes, ‘Modern anthropology
was built up in the face of colonial genocides, ethnocides, mass killings, population
die-outs, and other forms of mass destruction visited on marginalized peo-
ples whose lives, suffering and deaths have provided us [anthropologists] with a
livelihood.’21 The argument, if not a plea, is for anthropologists to become not only
16 Arnold, ‘Justifying Genocide’, 97.

17 Gretchen E. Schafft, From Racism to Genocide: Anthropology in the Third Reich (Urbana:
University of Illinois Press, 2004).
18 Dan Stone, ‘Review: From Racism to Genocide: Anthropology in the Third Reich’, H-Net
Reviews in the Humanities & Social Sciences (March 2005).
19 Schafft, From Racism to Genocide, 32.
20 Nancy Scheper-Hughes, ‘Coming to our Senses: Anthropology and Genocide’, in Hinton (ed.),
Annihilating Difference, 348–81.
21 Scheper-Hughes, ‘Coming to our Senses’, 348.
more cognizant of the violence that surrounds them while in the field, which is a
challenge that present-day anthropologists have continued to meet to an increasing
extent, but also to engage genocide head-on as an object of study. This final call to
research and action constitutes what is now understood as the anthropology of
genocide.
F O R E N S I C , A C T I V I S T, C R I T I C A L
................................................................................................................
Central to the anthropology of genocide is the notion that socially and historically
located cultural practices construct genocidal and post-genocidal contexts and that
these practices are best observed through qualitative research methods. Indeed, one
of the major ways in which anthropology contributes to the field of genocide
studies is through the ethnographic method of data collection. Extended, face-to-
face engagements with communities yield textured, thick descriptions that more
macroanalyses of genocide overlook. To this end and through these means, con-
temporary formations of the anthropology of genocide have taken three distinct
forms. They are forensic anthropology, activist anthropology, and anthropology
as cultural critique. Each approaches a relatively common question from three
distinct methodological approaches. The overarching question is: What do cultu-
ral practices and lived experiences look like in genocidal and post-genocidal
contests—spaces where trauma, grief, and fear, as well as power and state control,
frame and at times define the contours of everyday life for survivors and perpe-
trators alike?
Attending to this question, many (if not all) anthropologists of genocide main-
tain that their work is politically engaged. The research often aims to contribute to
those who work on and in post-genocidal contexts, ranging from scholars in the
field of genocide studies and human rights to staff workers at non-governmental
organizations, aid workers, and development officers. Because of anthropologists’
field experience and attunement to local understandings, they have a great deal to
offer interdisciplinary discussions about genocide. This is to say that while scholars
in other fields have conducted excellent analyses of the macro-level factors facil-
itating genocide, fewer have been able to approach genocide from this type of a
local perspective.
Forensic anthropology, for example, is a branch of applied anthropology that
assists in the identification of deceased individuals whose remains are unrecogniz-
able because of extensive burns, mutilation, or decomposition. Traditionally used
in legal and academic settings, forensic anthropology has become increasingly
involved with not only issues of human rights but also genocide.22 Forensic
anthropologists now work within post-genocidal contexts to discover, document,
and detail clandestine gravesites. They often work with reconciliation commission
groups and transitional justice teams, documenting past abuses.
The work of Victoria Sanford, for example, details the role of the Guatemalan
Forensic Anthropology Foundation (GFAF) in the exhumation of clandestine
graves.23 The GFAF’s forensic work documents a state-sponsored genocide that
left some 200,000 Mayan dead and discarded—buried throughout the Guatemalan
highlands in mass gravesites. Denied not only their lives but also a culturally
appropriate burial, murder victims of the Guatemalan genocide continue to be
exhumed by forensic anthropologists so that they can be properly buried by their
community. Alongside this project, forensic anthropologists are also committed to
the careful documentation of each body and its markings, for the purposes of
formal reconciliation and litigation.
Typifying the kind of political role that forensic anthropologists now play in post-
genocidal contexts, Sanford’s work makes a series of larger theoretical arguments that
are themselves rooted in the persistent but nonetheless muddied work of forensic
anthropology. Through participant-observation with forensic anthropologists, San-
ford suggests that ‘the transformation of a private memory creates a public space,
however small, where survivors learn to speak; it breaks down to externally imposed
understandings and chips away at the power structures imposed through silent
negotiation of life-shattering events.’24 Sanford articulates the promise of forensic
anthropology—the hope that clarifying the past will foster a just future.
Forensic anthropology’s hope and practice builds from and is related to a second
anthropological approach to genocide: activist anthropology. At its most basic,
activist anthropology, to quote Charles R. Hale, means ‘a method through which
we affirm a political alignment with an organized group of people in struggle and
allow dialogue with them to shape each phase of the process, from conception of
the research topic to data collection to verification and dissemination of the
results.’25 Activist anthropology’s call to action prompts anthropologists to use
their data to assist a particular community or organization—to be an engaged
observer.26 With regard to the anthropology of genocide from an activist’s
22 Christopher Joyce and Eric Stove, Witnesses from the Grave: The Stories Bones Tell (London:
Little, Brown, 1991); William Maples and Michael Browns, Dead Men Do Tell Tales: The Strange and
Fascinating Case of a Forensic Anthropologist (Pella, IA: Main Street Books, 1995).
23 Victoria Sanford, Violencia y Genocidio en Guatemala (Guatemala: F&G, 2003); eadem, Buried
Secrets: Truth and Human Rights in Guatemala (New York: Palgrave Macmillan, 2003).
24 Sanford, Buried Secrets, 12.
25 Charles R. Hale, ‘Activist Research v. Cultural Critique: Indigenous Land Rights and the
Contradictions of Politically Engaged Anthropology’, Cultural Anthropology 21:1 (2006), 97.
26 Victoria Sanford and Asale Angel-ajani (eds), Engaged Observer: Anthropology, Advocacy, and
Activism (New Brunswick, NJ: Rutgers University Press, 2006).
perspective, practising activist anthropology has meant giving witness to genocidal

and post-genocidal events and making these events public. ‘Giving a voice to the
voiceless’ is one approximation of activist anthropology in genocidal and post-
genocidal contexts.
Human rights activism, for example, largely rests on the power of testimony—
one of the most central elements of anthropological data. Anthropologists, for
example, often leave their respective field sites with hundreds of hours of recorded
testimonies that provide a local perspective and, possibly, some emotional texture
to a particular object of study. Testimony, simply put, is a convincing genre of
anthropological data. The same is true when building a legal case for prosecuting
genocide. Ever since the Holocaust, human rights activists have used testimony as a
powerful tool for both litigation and reconciliation.27 And, more recently, activist
anthropologists have worked with truth commissions to interview survivors and
perpetrators, documenting with ethnographic detail genocide’s lived experience—
in Cambodia, Rwanda, Guatemala, and Bosnia, for example. Their work makes
and re-makes a commitment to applying ethnographic data for communities who
otherwise would remain either silent or forever under-represented.
Finally, the anthropology of genocide’s third manifestation fits squarely within
what is commonly known as cultural critique, ‘an approach to research and writing
in which political alignment is manifested through the content of the knowledge
produced, not through the relationship established with an organized group of
people in struggle.’28 The anthropology of genocide as cultural critique oftentimes
deconstructs genocidal events by way of discourse theory, historical analysis, and
critical theory. As part of this growing engagement with the anthropology of
genocide, a number of anthropologists began writing about genocide, particularly
after the genocides of the 1990s in the former Yugoslavia and Rwanda. However, to
date, the nascent anthropological literature on genocide (as critique) has been
published in diffuse forums—articles, book chapters, and a handful of ethnogra-
phies. Two edited volumes by Alex Hinton, however, provide a more systematic
effort at constructing an anthropology of genocide.29
Hinton’s edited volumes, in fact, provide the clearest and most sophisticated
examples of what an anthropology of genocide looks like when imagined from the
perspective of cultural critique. In the introduction to one of Hinton’s volumes, he
argues that ‘Within genocide studies, much work has focused on the macro-level
processes, exploring how genocide is linked to historical, political, economic, and
structural factors. Similarly, journalists have published powerful accounts of the
27 Shoshana Felman and Dori Laub, Testimony: Crises of Witnessing in Literature, Psychoanalysis,
and History (London: Routledge Press, 1992).
28 Hale, ‘Activist Research v. Cultural Critique’, 98.
29 Hinton (ed.), Annihilating Difference; idem (ed.), Genocide: An Anthropological Reader (Malden:
Blackwell, 2002).
origins and experience of genocide.’30 An anthropology of genocide, Hinton then

suggests, does both tasks at one and the same time; an anthropology of genocide
shuttles between the macro (the relationship, for example, between notions of
modern progress and acts of genocide from a historical perspective) and the micro
(genocide as a local process, as an ethnographic problematic). Hinton’s ultimate
point is that anthropology’s strengths—for example, its ability to link the con-
struction of binaries, otherness, and meta-narratives to the texture of lived experi-
ences—allow scholars to see genocide anew in challenging and critically insightful
ways.
Hinton argues with much reference to anthropological literature that modernity
has motivated genocidal violence. Modern notions of progress combined with the
means to engage in systematic attempts at annihilation have resulted in genocidal
violence. And, closely related to the question of modernity for Hinton is the issue
of difference-making. He writes, ‘Group identities are premised on the existence of
an “other” from which “we” is distinguished, often in terms of an ethnocentric set
of binary oppositions.’31 Modernity’s meta-narratives, Hinton continues, supply
those in power with the cultural grids from which they can distinguish between
civilized and savage as well as between good/evil, moral/immoral, and order/
disorder. Invoking the work of famed anthropologist Mary Douglas,32 Hinton
speaks about the cultural manufacturing of difference and purification, noting
that difference is necessary but not sufficient to produce genocide.
A growing number of book-length manuscripts reflect Hinton’s vision. One
example worth reviewing, even if only briefly, is Christopher C. Taylor’s Sacrifice
as Terror.33 The book provides a symbolic analysis of violence and terror during the
years prior to the 1994 Rwandan genocide. In a semi-autobiographical tone, Taylor
historicizes the Rwandan genocide, demonstrating that the long-held cultural
distinction between Hutu and Tutsi had been manipulated and warped by Euro-
centric stereotypes introduced during colonialism. Alongside a strong plea to
anthropologists to be critical of such cultural constructions, Taylor then details
how these kinds of divisions patterned Rwandan violence. He makes the argument
that the violence itself was structured by cultural assumptions and legitimated
by them also. Gender, for example, played a significant role. Taylor makes the claim
that, during the genocide, women were important agents and symbols of violence.
As agents, women played important roles on both sides during the conflict. As
symbols, Taylor notes how gender contributed to the construction of ethnic differ-
ence. Amidst the genocide, women existed as symbols of potential purification but
30 Hinton (ed.), Genocide, 3.

31 Ibid. 9.
32 Mary Douglas, Purity and Danger: An Analysis of the Concepts of Pollution and Taboo (New York:
Routledge, 1991).
33 Christopher C. Taylor, Sacrifice as Terror: The Rwandan Genocide of 1994 (Oxford: Berg, 1999).
also tools manipulated to reassess patriarchy. Taylor’s argument has been tempered
by Scott Straus’s The Order of Genocide, which makes the point that there is a
historicity to when difference is rendered toxic in a genocidal sense.34
Taylor’s commitment to structural, symbolic, and discursive analyses typifies the
anthropology of genocide as cultural critique—the kind of analysis that champions
genocide’s victims and deconstructs the cultural and historical relationships that
make genocide possible. The same can be said of Alex Hinton’s own book-length
manuscript, Why Did They Kill? 35 Hinton investigates the Khmer Rouge’s rise to
power and the genocidal outcome of the Democratic Kampuchea regime. He
provides an ethnographic analysis of the perpetrators’ motivation to kill—to
participate in genocide—in a way that does not reduce agency to ideology. The
central observation is that local, cultural frameworks melded with state ideologies,
forming a world vision that allowed perpetrators to kill. One of Hinton’s contribu-
tions, then, is to demonstrate that genocide is not understood as a mere event from
an ethnographic perspective, but as a process that complicates more traditional
understandings of state-sanctioned genocidal violence.
T RU T H , M E M O RY, R E P R E S E N TAT I O N
................................................................................................................
In the end, it is safe to say that the anthropology of genocide is only in its beginning
stages. Important now is how this young field will begin to read across the lines that
scholars have already drawn, and also beyond the analytic categories that the field
of anthropology uses to understand genocide. One provocative line of analysis is
anthropological research that addresses truth, memory, and representation in post-
genocidal contexts. The focus here is on discursive privileging and moments of
silencing that complicate and constitute issues of truth and falsity in post-genocid-
al contexts. One of the clearest examples of this muddied new frontier comes from
the construction of genocide’s definition.36
Polish jurist Raphael Lemkin sought to create new international law. Writing in the
shadow of the Holocaust, Lemkin wanted to give a name to acts of mass murder.
He ultimately took the Greek genos (race, tribe) and the Latin cide (‘killing’) to coin
34 Scott Straus, The Order of Genocide: Race, Power, and War in Rwanda (Ithaca, NY: Cornell
35 Alex Laban Hinton, Why Did They Kill?: Cambodia in the Shadow of Genocide (Berkeley:
36 Kevin Lewis O’Neill and Alex Laban Hinton, ‘Representation and/in Post-Genocidal Contexts:
An Introduction’, in idem (eds), Genocide: Truth, Memory, and Representation (Durham: Duke
the term ‘genocide’. As Lemkin’s own writings evidence, the word intended to denote
a crime that involved ‘a coordinated plan of different actions aiming at the destruction
of essential foundations of life of national groups, with the aim of annihilating the
groups themselves.’37 Of interest here are the social processes that adjusted the
definition, both its meaning and scope. In 1948, for example, Resolution 96-I became
international law, with genocide referring to ‘acts committed with intent to destroy, in
whole or in part, a national, ethnical, racial or religious group, as such.’ As the text
makes clear, these acts include both outright ‘killing’ and ‘causing serious bodily or
mental harm’ or creating ‘conditions of life’ intended to physically destroy the
members of a group. The definition of genocide in Resolution 96-I also encompassed
attempts to eliminate a group’s survival by ‘preventing births’ or ‘forcibly transferring
[their] children’ to another group. Yet, it is important to note that several key elements
of Lemkin’s original conceptualization of genocide were left out of Resolution 96-I.38
How? United Nations delegates debated the wording of the Convention for two
years.39 Certain types of groups (‘national, ethnical, racial, or religious’) were
included. Others, such as economic and political groups, were eventually excluded
for a variety of reasons ranging from the conceptual (i.e., some argued that political
and economic groups were not ‘enduring’) to the pragmatic (i.e., many states
feared interference in their internal political affairs and some, such as the Soviet
Union, feared accusations they had committed genocide against such groups).40
Similarly, in Axis Rule in Occupied Europe, Lemkin argues that there was a range of
other factors omitted. Those include political, social, cultural, economic, biological,
physical, religious, and moral ‘techniques of genocide’.41 While physical, biological,
and cultural acts appeared in the initial draft of the Convention, the broader sense of
‘cultural genocide’ was eliminated from the final text.42 Cultural genocide, for
example, includes curtailing or banning a language, traditional socialization practices,
artistic endeavours, ritual practices, social institutions, and so forth. The framers of
the 1948 Convention on Genocide erased these aspects from the text for conceptual/
legal reasons (some argued that cultural genocide was already prohibited in interna-
tional law) and practical reasons (colonial powers, for example, likely feared accusa-
tions of cultural genocide). There were similar debates over issues of motivation/
intent, the scale of destruction, and punishment/enforcement.43
37 Raphael Lemkin, Axis Rule in Occupied Europe; Laws of Occupation, Analysis of Government,
Proposals for Redress (Washington, DC: Carnegie Endowment for International Peace, 1944), 79.
38 O’Neill and Hinton, ‘Representation and/in Post-Genocidal Contexts’.
39 Leo Kuper, Genocide: Its Political Use in the Twentieth Century (New Haven, CT: Yale University
Press, 1981), 23.
41 Lemkin, Axis Rule in Occupied Europe, 82.
42 Kuper, Genocide.
The very conception and legal definition of genocide was forged in a highly
politicized atmosphere, one that resulted in inclusions and exclusions and a moral
gradation of atrocity. The destruction of political groups, while abhorrent, was
written out of the Genocide Convention and became something else, an implicitly
lesser crime; cultural genocide similarly dropped from sight, eventually reemerging
as ‘ethnocide’. Yet, it is precisely these sorts of processes of inclusion and exclusion
that the anthropology of genocide now turns.
One forward-looking example comes from Alex Hinton and Kevin Lewis
O’Neill’s edited volume on truth, memory, and representation.44 The volume
represents anthropology of genocide’s continued interest in post-genocidal con-
texts. While some of the chapters grapple with the formal definition of genocide
directly, most carry these themes far and wide to consider how genocide is itself
represented and remembered in a variety of contexts. Along these lines, the book
examines several cases commonly regarded as genocide and other cases in which
there is contestation over the term, ranging from the North–South conflict in
Sudan to Guatemala, where debate continues over whether civil conflict reached
genocidal proportions. The Guatemalan case also illustrates how such views change
over time, as many more people came to view the violence against indigenous
Mayans as genocide following the findings of the Historical Clarification Commis-
sion in 1999. These more contested cases are revealing about the ways in which
horrific violence is represented, remembered, and linked to truth claims about
genocide.
Accordingly, the essays in the volume do not just explore the relationship of
genocide to truth, memory, and representation, but also are concerned with how
discourses about genocide, truth, memory, and representation are interlinked.
Conerly Casey’s chapter, for example, addresses the talk of genocide in Nigeria.45
While some scholars suggest that Nigeria has been the site of genocide or genocidal
massacres, others would argue genocide has not taken place there. It is a struggle
over definitions, criteria, and key terms. What is genocide? And how does a grieving
community define it? In response, Casey’s chapter explores the ways in which
discourses about genocide in Nigeria are bound up with issues of representation,
memory, and truth claims, particularly with regard to the media. Central to Casey’s
chapter is the idea that the media disrupts the temporality of history, creating
entangled representations of social reality. In media-saturated places such as North-
ern Nigeria, Casey argues, youths’ uncertain experiences of local conflicts become
placed next to regional and global media images of political unrest. The result is an
affective citizenship that becomes constituted by historical ruptures, such as vio-
lence. Casey’s intent is to make clear that media representations play an especially
44 Hinton and O’Neill, Truth/Memory/Representation and Genocide.

45 Conerly Casey, ‘Mediated Hostility: Media, “Affective Citizenship”, and Genocide in Northern
Nigeria’, in Hinton and O’Neill (eds), Genocide.
important role in the construction of political belonging among Nigerian youths

who are (or who may become) perpetrators in genocidal massacres.
At the same time, the anthropology of genocide is directly concerned with issues
of truth, memory, and representation in the aftermath of what most people would
agree is genocide. Central to the growing anthropological study of genocide, in fact,
is the idea that the ‘truth’ of genocide often becomes a power-laden tool over which
politicians, activists, and the international community wrestle by asserting and
contesting representations cobbled together from the often fragmented and clash-
ing memories of survivors, perpetrators, witnesses, and bystanders. While scholars
can make distinctions between truth, memory, and representation for the sake of
analytical clarity, these divisions frequently become problematic on the ground.
This conceptual unsteadiness can be troubling for those who want to definitively
explain how, why, and when mass murder takes place.
The stakes are high in this area, as various individuals, groups, governments, and
institutions vie to map out a narrative of the past that legitimates their agendas or
desire for justice, to assert or reject the right to legal redress for and moral outrage
about ‘the crime of all crimes’, and to acknowledge or disavow memories, experi-
ences, suffering, and losses linked to mass murder. These often fiery debates rage
around us in the world today, ranging from the Turkish government’s continuing
denials of the Armenian genocide to the debate over whether genocide is taking
place in Darfur. There is also the case of denial in which various Western govern-
ments seek a ‘balanced’ or qualified view about their complicities in colonial
projects.46 As noted earlier, another example is the debate over whether Guatema-
la’s civil war (1960–96) was genocidal. What the field aims to appreciate, however, is
not whether Guatemala’s civil conflict was genocidal itself, but how Guatemala’s
two truth commission reports represent genocide. Moreover, the question that the
anthropology of genocide now addresses is how such truth commission reports
function as machines of documentation—how they collect, analyse, edit, and
publish extraordinary amounts of data that form seemingly comprehensive his-
tories of human rights abuses. A close textual analysis, for example, demonstrates
that Christian imaginations motivate representations of genocide in Guatemala.47
While acknowledging that such tribunals and truth commissions are often very
important to people who have been the victims of mass atrocity, the anthropology
of genocide suggests a new trajectory—one that demonstrates that there are
inevitably ‘grey zones’ and silences that are erased from these ‘official’ accounts.
This new line of analysis lingers on the disorderliness of genocide as well as the
inevitable incompleteness of any attempt to remember and represent the truth of
46 Stanley Cohen, States of Denial: Knowing about Atrocities and Suffering of Others (Malden, MA:
Blackwell, 2001).
47 Kevin Lewis O’Neill, ‘Writing Guatemala’s Genocide: Christianity and Truth and Reconciliation
Commissions’, Journal for Genocide Research 7:3 (September 2005), 310–31.
mass killings, systematic torture, and the deliberate attempt to annihilate a single
group of people from the face of the earth. This line of analysis takes into account
genocide’s ‘psychic imprints of terror’, in the words of Leslie Dwyer, that destabilize
everyday life for many living after genocide.48
C O N C LU S I O N
................................................................................................................
Rather than seeking complete truths, full memories, and factual representations,
the anthropology of genocide encourages scholars of genocide to focus on the
cultural work that practices of truth, memory, and representation do in genocidal
and post-genocidal contexts and how that work differs from one social and
historical space to another. This, again, is not to say anthropologists are not
concerned with the lived reality of genocide, or that anthropologists are somehow
dismissive of debates over whether genocide occurred in a given region. Given
ethnography’s insistence on an engaged research that places anthropologists amidst
zones of terror and abandonment, an argument could be made that anthropolo-
gists are some of the most empathetic observers and critics of genocide. By
definition, anthropologists of genocide witness; they listen. Anthropologists also
demonstrate that scholars of genocide must rely on their imaginations to under-
stand the incomprehensible and in a way that uncovers how and why truth,
memory, and representation exist as entangled efforts in spaces of such traumatic
and horrific acts of mass violence.
SELECT BIBLIOGRAPHY
Das, Veena, Arthur Kleinman, Mamphela Ramphele, and Pamela Reynolds (eds), Violence
and Subjectivity (Berkeley: University of California Press, 1997).
Hinton, Alex Laban, Why Did They Kill?: Cambodia in the Shadow of Genocide (Berkeley:
—— (ed.), Annihilating Difference: The Anthropology of Genocide (Berkeley: University of
California Press, 2002).
—— (ed.), Genocide: An Anthropological Reader (Malden: Blackwell, 2002).
—— and Kevin Lewis O’Neill (eds), Genocide: Truth, Memory, and Representation
(Durham, NC: Duke University Press, 2009).
48 Leslie Dwyer, ‘The Intimacy of Terror: Gender and the Violence of 1965–66 in Bali’, Intersections:
Gender, History, and Culture in the Asian Context 10 (August 2004).
Malkki, Liisa, Purity and Exile: Violence, Memory, and National Cosmology among Hutu
Refugees in Tanzania (Chicago: University of Chicago Press, 1995).
Sanford, Victoria, Buried Secrets: Truth and Human Rights in Guatemala (New York:
Palgrave Macmillan, 2003).
Scheper-Hughes, Nancy, and Philippe Bourgeois (eds), Violence in War and Peace: An
Anthology (London: Basil Blackwell Press, 2004).
Taylor, Christopher C., Sacrifice as Terror: The Rwandan Genocide of 1994 (Oxford: Berg,
1999).
Whitehead, Neil (ed.), Violence (Santa Fe, NM: School of Advanced Research Press, 2004).
chapter 10
.............................................................................................
SOCIAL
P S YC H O LO G Y A N D
G EN OCIDE
.............................................................................................
paul a. roth
Hier ist kein Warum.

Primo Levi1
[T]he International Military Tribunal (IMT) at Nuremberg introduced

conspiracy and membership in a criminal organization as charges that
transcended individual involvement by what could be seen as ‘guilt by
association.’ In the minds of their American authors, one of the main
purposes of these charges was to find a legal basis for . . . ‘a ghoulish
embarras de richesse’: the large number of perpetrators—estimated at the
time at hundreds of thousands—in numerous branches of the German
executive, bureaucratic and economic apparatus.
Jürgen Matthäus2
Understanding perpetrator and perpetration is the essential element to

understanding genocide. Other ‘lessons’ are ancillary.
Donald Bloxham3
1 Primo Levi, Survival in Auschwitz.

2 Jürgen Matthäus, ‘Historiography and the Perpetrators of the Holocaust’, in Dan Stone (ed.), The
Historiography of the Holocaust (New York: Palgrave Macmillan, 2004), 199.
3 Donald Bloxham, ‘From Streicher to Sawoniuk: the Holocaust in the Courtroom’, ibid. 414.
social psychology and genocide 199
We will not have come to terms with the past until the causes of what
happened then are no longer active. Only because these causes live on
does the spell of the past remain, to this very day, unbroken.
Theodor W. Adorno4
This chapter examines what purports to be a core standing problem in the

explanation of genocide, viz. how to account for the large number of people willing
to participate in mass murders. Yet this core status notwithstanding, I contend that
research in social psychology has already answered the question of ‘perpetrator
production’. Recruiting people to be perpetrators proves to be alarmingly easy. In
addition, the application of social psychology to genocide has also become en-
tangled in an ongoing moral debate, a debate that focuses on whether an
emphasis on the extrinsic predictors of behaviour fits at all well or comfortably
with a sense that people should be held morally (and legally) responsible for the
choices they make.5 This chapter argues as well that social psychology neither
casts a pall of inevitability over such events nor provides moral exculpation for
those involved.
In what follows, I use the phrase ‘situationist social psychology’, ‘situationism’,
and cognate terms to designate a research tradition that emphasizes how situa-
tional variables most often prove determinative of individual and group beha-
viour. ‘Situational’ contrasts with ‘dispositional’, i.e., an emphasis on factors
specific to the psychology of an individual. The paradigm takes a person’s
immediate context—the ‘situation’ in which one finds oneself—as a highly
reliable predictor of behaviour. The key factor in terms of ‘defining the situation’
concerns the group or social norms that implicitly or explicitly govern expected
behaviour in the situation. Experiments in this tradition place people in contexts
where the usual norms have been changed or expectations must be challenged. The
emphasis on situation implies that social stability (the following of certain
standards of behaviour) should be understood as a function of the ‘normative
stability’ of the contexts in which a person happens to be. Changes in normative
expectations change behaviour. More generally, the paradigm teaches how people
have a powerful tendency to conform to stated or implied norms in social and
institutional contexts.
4 Theodor W. Adorno, ‘What Does Coming to Terms with the Past Mean?’, in Geoffrey H. Hartman
(ed.), Bitburg in Moral and Political Perspective (Bloomington, IN: Indiana University Press, 1986) 129.
This essay was originally published in German in 1959.
5 For a somewhat equivocal view of the relation between the social psychology and moral theory,
see John M. Doris and Dominic Murphy, ‘From My Lai to Abu Ghraib: The Moral Psychology of
Atrocity’, Midwestern Studies in Philosophy 31 (2007), 25–55. See also the special issue of the journal
Metaphilosophy devoted to the topic, ‘Genocide’s Aftermath: Responsibility and Repair’,
Metaphilosophy 37 (2006), 299–543.
200 paul a. roth
Experimental work in situationist social psychology has taught troubling yet

important lessons.6 Although the key research has been much discussed for close to
four decades, its implications remain underappreciated. A reason for the lack of
uptake of this research might be the unpleasantly shallow picture it suggests of
what actually guides human behaviour, for the experiments suggest that people
simply adapt to the norms present in the situation in which they find themselves
and do not ‘carry over’ previous standards. But the lessons of situationism prove
too valuable to let lie idle.7
The ‘situationist’ paradigm in social psychology pioneered by Kurt Lewin and
developed experimentally by Solomon Asch, Stanley Milgram, and Philip Zimbardo
can be shown to have broad and somewhat surprising application to historical cases
of genocide.8 This serendipitous marriage of social science and historical research
represents a type of model for how the details of historical research (often thought to
resist extrapolation to other cases) fits well with the generalizing proclivities of social
science. Yet the relevant social psychological parameters fit a wide range of historical
cases. Surprise (or horror) arises insofar as the experimental data demonstrate how
relatively easily a substantial number of people can be co-opted for the purpose of
assisting in mass murder.
Definitional issues regarding what does or does not count as genocide will not be
examined in this essay. By focusing instead on how people transform into perpe-
trators, the definitional or conceptual debates surrounding genocide assume no
more than legalistic status. As a legal type, definition may matter for purposes of,
e.g., mobilizing international support. But some suggestion that an event called
into existence by definitional fiat can then serve as the basis for comparative
analysis is a thin reed on which to rest any claim to science: ‘Concepts determine
case selection, which in turn shapes causal inference.’9 Yet for concepts such as that
of genocide, definitions create the cases. And for cases so stipulatively created, no
reason exists for assuming that they then fit into any causal order.10
6 For an excellent overview and characterization of work in this tradition, see Lee Ross and
Richard E. Nisbett, The Person and the Situation (Boston: McGraw Hill, 1991). A recent updating of
research in this area can be found in Philip Zimbardo, The Lucifer Effect: Understanding How Good
People Turn Evil (New York: Random House, 2008).
7 Works such as, e.g., Ervin Staub, The Roots of Evil (New York: Cambridge University Press, 1989)
overtheorize genocide and yet add nothing of predictive value.
8 Christopher R. Browning’s work provides the compelling model. See his now classic book,
Ordinary Men: Reserve Police Battalion 101 and the Final Solution in Poland (New York:
HarperPerrenial, 1993). For a defence of Browning’s use of the social psychological research, see my
‘Beyond Understanding: The Career of the Concept of Understanding in the Human Sciences’, in
Stephen P. Turner and Paul A. Roth, eds., The Blackwell Guide to the Philosophy of the Social Sciences
(Malden, MA: Blackwell, 2003) and ‘Hearts of Darkness’, History of the Human Sciences 17 (2004),
211–51.
9 See Scott Straus, ‘Contested Meanings and Conflicting Imperatives: A Conceptual Analysis of
Genocide’, Journal of Genocide Research 3 (2001), 359.
10 See my ‘Historical Explanation’, Journal of the Philosophy of History 2 (2008), 214–26.
The social psychological experiments suggest a focus instead on comprehending

the construction of a corps of perpetrators. The victims may be almost randomly
chosen once one understands how perpetrators can be readily recruited. What
makes conditions conducive to mass-producing perpetrators of acts such as geno-
cides?11 Posing the problem this way helps highlight two distinct philosophical
components inseparable from the social psychological task of explaining participa-
tion in mass murder. One concerns the motivational question—what could ac-
count for a choice to behave in the way that genocidal killers and mass murderers
do? The second concerns the relation of any proposed explanation and that of
responsibility. Since explanations typically speak to causes, and causes can readily
be understood to mitigate agency—freedom of or responsibility for choice—it
might appear that the more one explains, the less people can be held responsible.
However, properly understood, situationist accounts explain and yet require no
diminution of attribution of responsibility.
The first section sets a philosophical frame for discussions of explanations of
genocide. The second section examines how certain results from social psychology
nicely accord with and support this frame. The third and final section looks at Ian
Hacking’s accounts of ‘making up people’ as enriching and supporting the theo-
retical lessons informed by the empirical work.
GENOCIDE AND ‘T H O U G H T L E S S ’ B E H AV I O U R
................................................................................................................
The social trauma of the Second World War and its political aftermath were never
far from the minds of European philosophers of that time. What seem to be
philosophical abstractions turn out to connect to grim political concerns. For
example, take Jean-Paul Sartre’s declaration that ‘Existence precedes essence.’
Essences constitute a thing’s nature, what it must be. Existence without a prior
essence permits a type of freedom. One’s own choices can then determine who or
what one will be. But if a lack of an essence makes existential freedom possible, it
also makes people at core unreliable. For nothing need determine how one acts.
‘Tomorrow . . . some may decide to set up Fascism, and the others may be cowardly
11 Regarding how definitional concerns have had some interesting ramifications in moral, legal,
and historical debates, see Ann Curthhoys and John Docker, ‘Defining Genocide’, in Stone (ed.), The
Historiography of Genocide, 9–41. As with so much else, Arendt anticipates these definitional concerns.
Genocide, she argued, represents a new break in the moral order. ‘Nothing is more pernicious to an
understanding of these new crimes . . . than the common illusion that the crime of murder and the
crime of genocide are essentially the same’ (Hannah Arendt, Eichmann in Jerusalem, rev and enlg edn
(New York: Penguin Books, 1977), 272).
202 paul a. roth
and muddled enough to let them do it. Fascism would then be the human reality,
and so much the worse for us.’12 War revealed the ambivalent legacy of postulating
human freedom subject to no moral imperatives.
Writing about a decade and a half after Sartre, Adorno too ponders what the
experience of the Second World War reveals with regard to a ‘social weakening of
personal autonomy’.13 Both worry in light of that experience about who will choose
to resist oppression. Revelations surrounding perpetrator behaviour and the death
camps in the intervening decade and a half sharpens this worry about whether or
how people will in fact exercise autonomy. In this context, Adorno famously asks,
‘What Does Coming to Terms with the Past Mean?’ Do enduring but unaddressed
social conditions continue to exist that incline people to political oppression and
mass murder? Has post-war society comprehended the social conditions that
caused or abetted the horrific acts? Insofar as these conditions persist, they leave
in place the forces that ultimately produce, for example, genocides. Only by
bringing these processes to collective self-awareness might their hold be broken.
The proliferation of genocides through the second half of the twentieth century
suggests that the social determinants remain in place.
Adorno’s essay touches on two themes that remain relevant to any attempt to
explain why genocide-like events recur. The first concerns his emphasis on ‘objec-
tive conditions’—the social situations in which people find themselves—as having
explanatory priority over individual ‘subjective’ factors. This suggested focus on
objective situation (however cashed out) shall be termed the ‘choice problem’—
what determines how people act? The choice problem typically seeks answers
regarding motivation—why people would choose to do what they did. The second
concerns the affective character—a need to explain the enthusiasm often mani-
fested by those inflicting the pain. This is the ‘smile problem’.14 The very questions
suggest how social psychology might inform a philosophical anthropology. Answers
to these problems would presumably indicate whether conditions could be altered so
as to make genocidal behaviour less likely and if so how.
The concerns of Sartre and Adorno connect to yet a third figure whose philo-
sophical writings attempt to comprehend how people behaved under conditions of
12 Jean-Paul Sartre, ‘Existentialism is a Humanism’, in George Sher (ed.), Moral Philosophy

(New York: Harcourt Brace Jovanovich, 1987), 480–1.
13 Adorno, ‘What Does Coming to Terms with the Past Mean?’, 117.
14 Adorno echoes both worries in the following remark. ‘[T]he past one wishes to evade is still so
intensely alive. National Socialism lives on, and to this day we don’t know whether it is only the ghost
of what was so monstrous that it didn’t even die off with its own death, or whether . . . the readiness for
unspeakable actions survives in people, as in the social conditions that hem them in’ (Adorno, ‘What
Does Coming to Terms with the Past Mean?’, 115). Social conditions are one thing; people’s ‘readiness
for unspeakable actions’ quite another. See also Roth, ‘Beyond Understanding’, 322 and ‘Hearts of
Darkness’, 226–36. Daniel Jonah Goldhagen takes the smile problem as key to explaining the
Holocaust. See his Hitler’s Willing Executioners: Ordinary Germans and the Holocaust (New York:
Vintage, 1997).
occupation and oppression. Hannah Arendt too takes the choices made by perpe-
trators not as reflecting their individual quirks, but as revelations about us all.
Schooled in the same philosophical tradition as were Sartre and Adorno, her oft-
quoted and yet widely misunderstood remark about the ‘banality of evil’ with
respect to Eichmann reflects a shared concern about the nature and conditions
of choice.
How could Eichmann deny recognition to others of a common humanity, and so
of moral worth? Her answer—recognizable to anyone in the Kantian tradition
from which she comes—identifies Eichmann’s failure as one of refusing responsi-
bility for his choices by letting others determine the ends of action. Sartre and
Adorno and Arendt all share with Kant the view that our ability to will—to
choose—separates us from beasts in the field. Other animals cannot but behave
as instinct dictates; only humans can will to act against instinct. Eichmann fails this
essential test of one’s humanity, for he uses reason only to follow rules others have
made for him.
It was sheer thoughtlessness—something by no means identical with stupidity—that
predisposed him [Eichmann] to become one of the greatest criminals of that period. And
if this is ‘banal’ and even funny, if with the best will in the world one cannot extract any
diabolical or demonic profundity from Eichmann, that is still far from calling it common-
place. . . . That such remoteness forms reality and such thoughtlessness can wreak more
havoc than all the evil instincts taken together which, perhaps are inherent in man—that was,
in fact, the lesson one could learn in Jerusalem. But it was a lesson, neither an explanation of
the phenomenon nor a theory about it.15
A literal failure to ‘think for himself ’ marks him as thoughtless, as behaving as if a

being who could not recognize and follow the moral law. Human will allows us
only this choice, i.e., to follow the dictates of reason over that of animal nature. If
one holds their will in abeyance, behaviour cannot be original—something other
than a product of instinct or animal nature. If not original, then banal. The
consequences of any failure of will thus can provide a lesson, an example of the
consequences of indifference to aligning action to our human nature. But nothing
can compel humans to act against instinct, i.e., freely.
Evil becomes banal once the actions that produce it lack just this type of Kantian
thoughtfulness, i.e., becomes a mere following of ends given by others and not by
reason. Arendt specifically comments upon Eichmann as manifesting a ‘strange
interdependence of thoughtlessness and evil’. Indeed, when she speaks explicitly of
Eichmann as personifying the ‘banality of evil’, it is in the context of ‘forgetting’ he
was at his own death sentence, i.e., faced with his own mortality. This proves to be a
defining moment. For in the philosophical tradition to which Arendt belongs, such
a ‘forgetting’ comes significantly freighted. Recognizing oneself as human involves
15 Arendt, Eichmann in Jerusalem, 288.

204 paul a. roth
always being confronted with the fact of one’s own mortality. Losing sight of this
just is losing sight of what defines us as human—our common fate and our
capacity to choose. Eichmann’s crime, Arendt contends, consists precisely in this
special thoughtlessness.16
The substitution of technical reasoning—bureaucratic, economic, technologi-
cal—for considerations about the best conditions for humans characterizes con-
temporary thoughtlessness about the ends of action. Arendt worries theses issues
most thoroughly. For her, the disconnect that arises between thinking and willing
signifies the area of most concern in contemporary life. By letting ends set by things
and institutions go proxy for those freely chosen, banality ensues. Insofar as
modern technology makes masses of people economically unnecessary and to the
extent technology allows for the ready mobilization of killers, the world stage has
been set for a ‘new’ crime to emerge.
So what begins as an extremely abstract pronouncement about humankind’s
lack of essence and the nature of free choice becomes, when filtered through the
related musings of Adorno and Arendt, a question about the material conditions
that engender ‘thoughtless’ action. Arendt, in particular, suggests that the Holo-
caust results from this type of thoughtless behaviour. In this regard, the oft-
discussed connection between Arendt’s invocation of the ‘banality of evil’ and
work in social psychology must be seen in light of how experimental settings can
readily induce just such ‘thoughtless’ behaviour.
I N D U C I N G ‘T H O U G H T L E S S B E H AV I O U R ’
A N D C R E AT I N G P E R P E T R ATO R S
................................................................................................................
Can one experimentally create situations so as to induce such ‘thoughtless’ beha-
viour? Note that what calls for explanation concerns not only the production of
perpetrators, but also their ‘disappearance’. That is, characteristics of genocides in
the twentieth century include the large number of perpetrators, their wantonness,
and their later return to live undisturbed among the populace. If one considers the
political furor accompanying, e.g., the residential locations of sex offenders, the
reassimilation of perpetrators appears strikingly anomalous.17
In emphasizing the importance of situational factors, the salient feature con-
cerns how the experiments effectively construct scenarios that require people to
challenge norms (decide against a group or an authority figure) or forge normative
16 Arendt, Eichmann in Jerusalem, 252, 279.

17 This point struck Arendt as well. Ibid. 16–19.
limits (e.g., how to modulate or enact the authority one has over others).18 So, for
example, people will identify a shorter line as the longest in a set if a majority of
others in their group do or choose to inflict pain despite having no prior history of
a willingness to do so.19 The core point demonstrated by these experiments
concerns the power of the ‘conformity effect’—the amazing willingness of people
to simply assimilate the norms of the situations in which they find themselves.
Experiments in the ‘Asch paradigm’ provides replicable demonstrations of how this
powerful ‘conformity effect’ trumps other factors as a predictor of behaviour.20
An interesting and important feature of the classic experiments in the Asch–
Milgram–Zimbardo line is that in each case the outcomes proved contrary to the
prior expectations of the experimenters.21 In Asch’s experiments, pressure to
conform presumably arises merely from the implied challenge of contravening
those who have already announced a decision. Milgram’s justly celebrated exten-
sion of this experimental paradigm explored whether this proclivity to conformity
to a real or imagined norm would be manifest if the choice involved inflicting pain
or possible harm on others.
In earlier writings, I emphasized how these experiments could be brought to
bear on two key explanatory problems noted in the introduction—the ‘choice
problem’ and the ‘smile problem’. The choice problem can be characterized as
follows: why do so many people with no prior history of brutalization or murder
participate, at one level or another, in the killing operations involved in genocides
and mass murders?22 The smile problem connects to the choice problem. For if the
choice problem asks why people did these acts voluntarily, the smile problem
18 Material in this and the subsequent three paragraphs borrow from my discussion in ‘Hearts
of Darkness’, 217–20.
19 For a detailed account of Asch’s experiments in this vein and important related work, see
Zimbardo, The Lucifer Effect, esp. Chapter 12. An extensive overview and summary presentation of
the relevant research can also be found in Philip Zimbardo, ‘Transforming Good People into
Perpetrators of Evil: Can We Reverse the Process?’, Ricerche di psicologia 28 (2005), 1–52.
20 See the helpful and historically broad accounts of this research tradition and its influence on
social psychology in Ross and Nisbett, The Person and the Situation.
21 In The Lucifer Effect, Zimbardo devotes over 200 pages to elaborating the details of the Stanford
Prison Experiment (SPE). From his discussion one learns that there is a web site devoted just to
the SPE, http://www.prisonexp.org, one devoted to Zimbardo’s book featuring his extensive
retrospective analysis of the material, http://www.LuciferEffect.com (‘Lucifer Effect’ t-shirts can be
purchased from this site), and one devoted to Zimbardo himself, http://www.zimbardo.com. Given
that the SPE (unlike, e.g., the Milgram experiment) was effectively a one-time occurrence that occurred
37 years ago, it yet remains a focus of much attention. Much more so than the Milgram experiment, the
SPE poses ethical obstacles to its replication. But see S. H. Lovibond, X. Mithiran, and W. G. Adams,
‘The Effects of Three Experimental Prison Environments on the Behaviour of Non-convict Volunteer
Subjects’, Australian Psychologist 14 (1979), 273–87. Their results strikingly confirm those of the SPE. ‘It is
clear that our Standard Custodial regime induced ordinary people with little knowledge and no
experience of prisons, to behave in much the same way as prisoners and officers in real prisons’ (283;
see also 278). See discussions of the SPE and replications in Zimbardo, The Lucifer Effect, 250–5.
22 See my discussion in ‘Beyond Understanding’, 320.
206 paul a. roth
points to the disconcerting fact that many do so with apparent enjoyment. Simply
put, those inflicting harm appeared to enjoy it.23
A close analysis of Milgram’s work and Zimbardo’s establishes that the situa-
tional analysis accounts in all essentials for the number of perpetrators and their
otherwise incomprehensible brutality. The experiments compellingly demonstrate
that a ‘conformity effect’ suffices to predict both the extent of participation (and
thus obviate any apparent problem of choice) and the sadistic enthusiasm that
individuals invest in roles to which accidents of experimental design (or of fate)
have assigned them (and thus solve the smile problem).
Milgram’s work bears primarily on questions of choice; Zimbardo’s addresses
issues related to the smile problem.24 To briefly summarize, the ‘Milgram experi-
ment’ (and its variants) involves subjects who believe that they are administering
electric shocks to someone as part of a learning experiment. The core of this
experiment concerns the extent to which subjects continue to inflict the electric
shocks despite pleas, screams, cries, etc. from the supposed victim. In some
variations, in excess of 60% of subjects regularly deliver up to maximum shocks
(450 volts). No coercion of the subject exists other than the norm requiring someone to
‘obey’ the experiment’s protocols.
Zimbardo was the chief architect of the Stanford Prison Experiment (SPE). This
involved arbitrarily assigning college students (pre-screened for personality ab-
normalities) to roles as prisoners or as guards. The experiment had to be termi-
nated after only a few days because of the brutality invested by guards in their roles
and the debilitating psychological effects on those assigned to be prisoners. The
chief theoretical point was to establish how structures, roles, or situations trump
dispositional factors as determinants of behaviour. This the experiment did in
compelling fashion. The famously unexpected result concerned the pleasure people
took in exercising arbitrary power over others. For by experimental design in the
SPE the roles were left underspecified just for the purpose of seeing how those
assigned the role ‘filled them in’. As Zimbardo states in that essay, ‘The third feature
23 Emphasis on the smile problem constitutes the core of Goldhagen’s critique of Browning’s use of
social psychology to explain perpetrator behaviour. In the view of some social psychologists, it
remains a standing objection to the explanatory reach of Milgram’s work. See, e.g., Arthur G.
Miller, ‘What Can the Milgram Obedience Experiments Tell Us about the Holocaust?: Generalizing
from the Social Psychology Laboratory’, in idem (ed.), The Social Psychology of Good and Evil (New
York: Guilford Press, 2004), 193–239, but esp. 212–16. Note how the chief criticisms of the explanatory
power of Milgram emphasize what I term the smile problem. Quite inexplicably, although Zimbardo
has an essay in the Miller anthology, his work receives no discussion in the other essays, and Miller
himself does not connect or emphasize the joint importance of using Milgram and Zimbardo together
for purposes of explanation. For my introduction and discussion of the smile problem, see ‘Beyond
Understanding’, 322. See also ‘Hearts of Darkness’, 226–36.
24 For an excellent summary of work done by Milgram and the many replications of his results, see
Thomas Blass, ‘The Milgram Paradigm after 35 Years: Some Things We Now Know about Obedience
to Authority’, in idem (ed.), Obedience to Authority: Current Perspectives on the Milgram Paradigm
(Mahwah, NJ: Lawrence Erlbaum Associates, 2000), 35–59. See also Zimbardo, The Lucifer Effect.
of the study was the novelty of the prisoner and guard roles: Participants had no
prior training in how to play the randomly assigned roles. Each subject’s prior
societal learning of the meaning of prisons and the behavioural scripts associated
with the oppositional roles of prisoner and guard was the sole source of guid-
ance.’25 Zimbardo’s design strongly suggests that cruelty will simply emerge; it does
not have to be planned for. The smile problem asks how people can take pleasure in
inflicting pain, especially when accidents of circumstance account for their posi-
tion to do so. The SPE and related research indicate that the mere ability to do this
engenders pleasure for many in doing so.26
Taken together, these experiments indicate that it requires little incentive to
recruit people for unsavory purposes. What appear as explanatory puzzles arise
only because such behaviour does not appear to be the norm. But confusion arises
if one takes a choice of norms as explaining the stability of the situations, as
opposed to seeing stability as a product of mere conformity to norms. When for
whatever reason stability disappears, people will simply adapt to what they take to
be expected of them. In short, the work of Milgram and Zimbardo retrodictively
account for general behaviour of people under the Nazis and make (or should have
made) what happened at Abu Ghraib predictable.27
A third puzzle also can be solved by the situationist account—the ‘nasty neigh-
bour problem’. The nasty neighbour puzzle arises insofar as one might assume that
a general populace would not desire known mass murderers to continue to live
undisturbed in their midst. But, as Arendt notes for the German case and has been
seen in later cases of genocides, the general populace reabsorbs perpetrators with
seeming complacency.28 Indeed, I am unaware that this has been recognized as an
explanatory problem by social psychologists or historians examining perpetrator
behaviour.
In tracing different accounts of perpetrator behaviour that have marked the
historical literature about the Holocaust, Mark Roseman notes that one model
25 See Philip G. Zimbardo, ‘A Situationist Perspective on the Psychology of Evil: Understanding

How Good People Are Transformed into Perpetrators’, in Miller (ed.), The Social Psychology of Good
and Evil, 39.
26 For a detailed analysis of just why this solves the choice problem, see Roth, ‘Hearts of Darkness’,
232–3. Regarding the smile problem, see 233–6. Important here is related work by Zimbardo on
deindividuation. See account in The Lucifer Effect.
27 The social psychology explains the events recounted in Victor Klemperer, I Will Bear Witness,
2 vols (New York: Random House, 1998).
28 Arendt, Eichmann in Jerusalem, 16. For a detailed study which brings particular empirical force
to Arendt’s observation, see Dick de Mildt, In the Name of the People: Perpetrators of Genocide in the
Reflection of Their Post-War Prosecution in West Germany (Boston: Martinus Nijhoff. 1996). For a
personal perspective, see Katharina von Kellenback, ‘Vanishing Acts: Perpetrators in Postwar
Germany’, Holocaust and Genocide Studies 17 (2003), 305–29. Mark Roseman also raises this issue in
‘Beyond Conviction? Perpetrators, Idea, and Action in the Holocaust in Historiographical
Perspective’, in Frank Biess, Mark Roseman, and Hanna Schissler (eds), Conflict, Catastrophe and
Continuity: Essays on Modern German History (New York: Berghahn Books, 2007), 83–103.
208 paul a. roth
emphasizes just a form of the nasty neighbour puzzle.29 But while Roseman does not
embrace this specific model, he acknowledges it as one of the standing explanatory
challenges regarding perpetrator behaviour:
Yet Arendt and others had grasped an essential problem that continues to be posed to us by
the Nazi regime, and which lies at the core of [historian Hans] Mommsen’s analysis too—
namely, how a body of men could operate with such comprehensiveness and relentlessness
right up to the last minute, and then let go of the program, like that, as if it had never been
theirs.30
As with the choice and smile problems, situational analyses offers a resolution of
the nasty neighbour puzzle. That is, the situation explains why people readily
become killers, indeed even killers who relish their work. Altering it then also
accounts for the otherwise counter-intuitive result that these very same people not
only reintegrate peaceably in their respective societies, but also that they can be
accepted as if they posed no threat.
Although the nasty neighbour puzzle has not figured prominently in debates
regarding explanations of perpetrator behaviour, it should. For too often those
committing the crimes simply melt back into the general populace. In this regard,
recent work indicates that there appears nothing remarkable about the particular
backgrounds from which the perpetrators of the Holocaust, for example, were
drawn.31 Somewhat oddly, given that the number of perpetrators in the German
case has been estimated at up to 250,000, the investigator focuses on about 1,500
cases, drawing only from those tried for a war crime. Not too surprisingly, he finds
a disproportionate number of ‘real Nazis’—committed party members with a long
history of Nazi association—in this group. But what light does this shed on how
the mass of perpetrators were co-opted?
More relevantly, Scott Straus attempts to find a way to accurately estimate the
number of perpetrators in the Rwandan genocide.32 The numbers are large, both in
absolute terms and in terms of the relative proportion of the population. This
detailed effort at quantifying the number of perpetrators and bystanders appears
consistent with numbers developed for the Holocaust case. I have elsewhere
provided a detailed analysis of the Milgram experiments that show why the
percentages of obedient subjects suffices to provide the needed number of perpe-
trators just by virtue of a tendency to obey authority. The question of choice ceases
29 Roseman, ‘Beyond Conviction?’, 91.

30 Ibid. 99. Roseman and I differ sharply on Arendt on banality.
31 Michael Mann, ‘Were the Perpetrators of Genocide “Ordinary Men” or “Real Nazis”? Results
from Fifteen Hundred Biographies’, Holocaust and Genocide Studies 14 (2000), 334–5. For an account at
variance both with Mann’s and my own, see Donald Bloxham, ‘Organized Mass Murder: Structure,
Participation and Motivation in Comparative Perspective’, Holocaust and Genocide Studies 22:2 (2008),
203–45.
32 Scott Straus, ‘How Many Perpetrators Were there in the Rwandan Genocide? An Estimate’,
Journal of Genocide Research 6 (2004), 85–98.
to matter. By demonstrating what people will do simply sidesteps any question of

whether they could have chosen otherwise.33 Put another way, situations do not
necessitate behaviours. But the tendency to conform makes this freedom to choose
moot. A type of herd mentality effectively trumps any actual ability to behave
otherwise.
There remains in the social psychological literature an unfortunate tendency to
equate situationist explanations with ‘judgmental dopes’, as caricatures of people
blindly shaping themselves to situations. For example, a recent article discusses
straight-faced and with endorsement a reading of the Arendtian notion of banality
and thoughtlessness as equivalent to asserting the ‘Nazi killers were unwitting
minions’,34 and that the experiments were meant to show that the people involved
were ‘amoral automatons’.35 Thus does the situationist paradigm come to be labelled
‘reductionist’.
Arendt, Milgram, and Zimbardo played a critical part in taking us beyond reductionist
explanations of tyranny as a simple product of pathological individuals. But now, their
reductionist explanations of tyranny as a simple product of pathological situations—the
banality-of-evil hypothesis—seem equally untenable. Instead, . . . an interactionist under-
standing that sees the social psychology of individual tyrants and collective tyranny as inter-
dependent and mutually reinforcing.36
But the foregoing only reflects a fundamental misreading of these texts. Nothing
in the Milgram experiments or in the SPE makes the situation inherently patho-
logical. The ‘pathology’ consists in the thoughtless investment people make in roles
handed them by an accident of fate. Arendt’s Kantian-inflected notion of banality
resonates deeply with what Milgram and Zimbardo demonstrate empirically.
In an effort to reject what he characterizes as ‘strong situationism’, Leonard
Newman maintains that controversies featuring disputes between situational and
dispositional variables are passé. Rather, ‘It has long been recognized that people
and their traits and the situations in which they find themselves interact.’37 Yet the
problem with distinguishing this model from the situationist also becomes imme-
diately if unwittingly apparent. For while interactionists maintain that traits and
situations mutually effect one another, Newman adds that ‘Traits will express
33 Roth, ‘Hearts of Darkness’, 232 ff; idem, ‘Beyond Understanding’, 319–26.

34 S. Alexander Haslam and Stephen Reicher, ‘Beyond the Banality of Evil: Three Dynamics of
an Interactionist Social Psychology of Tyranny’, Personality and Social Psychology Bulletin 33 (2007),
615–22, esp. 617–19.
35 Ibid. 619. Donald Bloxham discusses some of these readings of Arendt in ‘Organized Mass
Murder’.
36 Haslam and Reicher, ‘Beyond the Banality of Evil’, 621.
37 Leonard S. Newman, ‘Beyond Situationism: The Social Psychology of Genocide and Mass
Killing’, in Helgard Kramer (ed.), NS-Täter aus interdisziplinärer Perspektive (München: Martin
Meidenbauer, 2006), 110.
210 paul a. roth
themselves in some situations and not others.’38 Indeed, one does not even need to
do any new experiments to prove this. The old experiments suffice for that
purpose. ‘In fact, if you look carefully at the results of those studies [Asch,
Milgram, Zimbardo], the evidence is right there. Not everyone obeys, not everyone
conforms, and not everyone gets swallowed up into a role.’39 True enough. But
what then could be Newman’s point regarding how to distinguish between situa-
tionism and interactionism? The data from the classic situational experiments only
made claims about how a certain percentage of people would respond. ‘Interac-
tionism’ proves indistinguishable from what it proposes to replace.40
Surprisingly, Newman appears to concede the point, for he drops the idea that
specific traits interact in particular ways, and instead emphasizes that situations
can later alter the traits that people initially have. He cites literature that indicates
that even though individuals might be aware that they create a certain situation,
e.g., a stench caused by unsanitary conditions that perpetrators established, people
will nonetheless ‘blame the victim’. He sums this up as follows: ‘People are not only
affected by situations; they also change those situations. The process is bi-directional.
In addition, people are not always aware of how they have changed situations, and
when they are, they do not always take their influence into account.’41 Indeed,
Christopher Browning’s now classic Ordinary Men offers clear and graphic de-
monstrations of this bidirectional process of influence. But this simply shows the
extent to which the situationist paradigm already incorporates the dynamic aspect
on which interactionism insists. Likewise, Newman explicitly raises the smile
problem as if it were a difficulty for situationism, whereas in fact the ability to
deal also with this issue proves to be a situationist strength.42
Mass murder has become a feature of the twentieth century. Arendt’s concern
that genocides represent a new break in the moral order implicates the newly
evolved technologies. Insofar as certain technologies—e.g., those of mass commu-
nication—help to engineer changes in the perceived normative framework, tech-
nology abets thoughtlessness in the Kantian sense. Even in cases where the actual
means of killing remain relatively primitive, e.g., Rwanda, the use of such technol-
ogies allows for the ready mobilization and manipulation of would-be perpetra-
tors. Technology makes practical obstacles to mass murders for those in the
38 Leonard S. Newman, ‘Beyond Situationism: The Social Psychology of Genocide and Mass
Killing’, in Helgard Kramer (ed.), NS-Täter aus interdisziplinärer Perspektive (München: Martin
Meidenbauer, 2006), 110.
39 Ibid.
40 Newman does appreciate how Arendt has been misread by social psychologists, See Leonard
S. Newman, ‘The Banality of Secondary Sources: Why Social Psychologists have Misinterpreted
Arendt’s Thesis’ (2001, unpublished talk).
41 Newman, ‘Beyond Situationism’, 115.
42 Ibid., 111–13; see Roth, ‘Beyond Understanding’, and ‘Hearts of Darkness’.
changed situation disappear.43 Research and history show that prior norms will not
provide sufficient constraint.
‘T H O U G H T L E S S N E S S ’ I N (F U RT H E R )
THEORETICAL PERSPECTIVE
................................................................................................................
The work in social psychology has profound implications, in turn, for philosophi-
cal anthropology—our understanding of the type of beings that we are. And, in
fact, these empirical and theoretical considerations have been further extended
through some of the important and innovative work done by Ian Hacking on
‘making up people’.44 Hacking explores how categories of people or medical
classifications of them come into being (e.g., suicides, the normal, child abuse,
multiple personality disorder). By what processes, he asks, do human kinds
become sorted into recognized types that bureaucrats and others then use as
bases for how to treat people?
In raising these questions, he goes to the heart of how social science might be
possible, for scientific generalizations typically utilize categories (kinds) of entities.
The generalizations, that is, apply to like items, and the relevant likenesses mark out
the properties by which to sort things into kinds. The stability of generalizations thus
presumes a stability of the kinds. If one seeks generalizations about human behaviour,
the question to be answered concerns what makes for stable human kinds.
Hacking develops examples from a fascinating range of areas—homosexuality
considered both as a medical category and as a ‘life style’, multiple personality
disorder, suicides. In each case, he examines the technological, social, and scientific
factors surrounding the emergence of each of these ways of categorizing individuals.
So, for example, he shows how innovations in classifying and counting made possible
by the development of statistical analysis abets efforts to measure the ‘health’ of
43 For related reflections, see A. Dirk Moses, ‘Genocide and Modernity’, in Stone, The
Historiography of Genocide (ed.), 166–70.
44 Key essays by Hacking in this regard include ‘Making Up People’, in Thomas Heller, Morton
Sosna, and David Wellberry (eds), Reconstructing Individualism (Stanford: Stanford University Press,
1986), 222–36; ‘Multiple Personality Disorder and its Hosts’, History of the Human Sciences 2 (1992),
3–31; ‘World Making by Kind-Making: Child Abuse for Example’, in Mary Douglas and David Hull
(eds), How Classification Works (Edinburgh: Edinburgh University Press, 1992), 180–238; ‘The Looping
Effects of Human Kinds’, in Dan Sperber, David Premack, and Ann J. Premack (eds), Causal Cognition
(Oxford: Clarendon Press, 1995), 251–83; Rewriting the Soul: Multiple Personality and the Sciences of
Memory (Princeton: Princeton University Press, 1995); ‘Between Michel Foucault and Erving
Goffman: Between Discourse in the Abstract and Face-to-Face Interaction’, Economy and Society 33
(2004), 277–302.
212 paul a. roth
societies, and so ultimately to classify and count suicides. The emergence and growth
of multiple personality disorder as a diagnostic category likewise links with the
evolution of the definition of child abuse from physical to primarily sexual abuse.
Childhood sexual abuse becomes by definition a key part of the etiology of multiple
personality disorder. So as one is found pervasive, so is the other. Likewise, the
homosexual considered as a distinct medical and social type also proves to be of
recent historical origin. In all cases, a drive to medicalize problems (and so make them
into individual and not social problems) or create scientifically certified categories of
social misfits (as with multiple personality disorder) suggests how individuals can be
at the mercy of others for ways to think about themselves, and how bureaucratic or
social imperatives trump what the available evidence actually establishes.
But the ways in which people become classified by existing institutional arrange-
ments also manifest what Hacking calls ‘looping effects’. These involve cases where
people come to inhabit a classificatory category that has been created. Hacking’s
looping effect, I suggest, should be seen as a manifestation of what has been discussed
above as the conformity effect. Hacking’s work points to the presence of institutionally
pervasive pressures to conform to the ways in which one is diagnosed or sorted as a
type of person. The categories do not need a prior reality, so to speak. Once they have
an institutional context, people will make them ‘real’ and populate the categories.
As a philosopher of science, Hacking has a keen awareness of the stakes here in
exploring how human kinds come to be, change, and perhaps fade away. ‘I could
develop the argument that what I call human kinds are at the historical root of
sociology—the science of normality and deviance.’45 This looping effect becomes
another way to theoretically account for how people come to ‘make real’ roles they
never envisioned themselves as inhabiting. In this regard, the implicit concern
involves whether one takes the designation of a certain type of behaviour—‘perpetra-
tor’, for example—as the name of a kind. Taken as a kind, one assumes common
properties, and looks for elements all the individuals share. But the distinction
between the notions of looping and that of situations indicates that the common
elements may be largely or wholly extrinsic to the individuals involved. People readily
become perpetrators by falling prey to situational factors. Looping effects indicate
that the behaviour in question involves more than the conformity effect, but also the
self-definition of the actor. The conformity effect does not imply looping, while the
looping effect indicates how categories can be taken up by those so categorized.46
In short, looping effects point to a power of categorization that goes beyond
what the conformity effect demonstrates. Categories matter existentially, so to
speak. People come to inhabit and so perpetuate and make real the categories by
which they are classified. But their ultimate significance as historical or social
45 Hacking, ‘Looping Effects’, 360.

46 Hacking readily acknowledges the immense influence that the work of Michel Foucault has had
on his approach to these issues.
artefacts remains distinct from the power of the conformity effect. If perpetrators
adopt and adapt a categorization of themselves as perpetrators, they will internalize
kind-like attributes.
Thus one way in which some human kinds differ from some kinds of thing is that
classifying people works on people, changes them, and can even change their past. The
process does not stop there. The people of a kind themselves are changed. Hence ‘we’, the
experts, are forced to rethink our classifications. Moreover, causal relationships between
kinds are changed. . . . This is not because we have found out more about the natural
disorder, but because people who see themselves as having this human disorder now find
in themselves memories of trauma, often traumas of a kind that they could not even have
conceptualized twenty years ago. . . . This in turn generates a looping effect, because people
of the kind behave differently and so are different. That is to say the kind changes, and so
there is new causal knowledge to be gained.47
Otherwise, what makes them perpetrators is only conformity.

A dramatic illustration of Hacking’s looping effect can be found in Zimbardo’s
description of the well-known case where a schoolteacher, Jane Elliott, discriminated
among her students by an arbitrarily chosen characteristic, e.g., eye color. The students
quickly realize that discrimination exists and determine the group to which they belong.
They then behave as either those entitled to privilege or those who expect the worst.
A key feature of this incident concerns how the children came to occupy categories of a
kind of person who was being discriminated against. Though simple in structure, the
case indicates the power of categorization and how adaption to contextual norms
induces a looping effect, i.e., creates and sustains certain patterns of behaviour.48
As Hacking insists, his account of looping effects goes beyond what, e.g., labelling
theory suggests insofar as the individual may shape the category in which placed,
and in doing so alters both the kind and what there is to know about the kind in
question. In labelling theory a certain normative tag (e.g., ‘delinquent’) comes to be
applied to certain behaviours. In Hacking’s case, however, the emphasis concerns
how people ‘fill out’ institutionally specified but otherwise underdetermined social
roles. In many interesting cases, behaviours become tagged as representing a
category in advance of any determination that they reflect common properties or
causes. But this calls into existence institutionally designated but vaguely articulat-
ed kinds. Actual individuals then come to adapt to and inhabit these categories,
making them real after the fact, so to speak.
Hacking suggestively juxtaposes in a complementary fashion the philosophical
approaches of Michel Foucault (concerned with structures of power as structures of
knowledge) and sociologist Erving Goffman, who in his turn did classic studies of
self-construction within institutional frameworks and in face-to-face interactions.49
47 Hacking, ‘Looping Effects’, 369.

48 The case is described by Zimbardo in The Lucifer Effect, 283–4.
49 Hacking, ‘Between Michel Foucault and Erving Goffman’.
214 paul a. roth
By combining Foucauldian macroanalyses of how power influences structures of

everyday life and Goffman’s microanalyses of the environment of ‘closed institu-
tions’—asylums, hospitals, boarding schools—Hacking provides a compelling pic-
ture of how certain types of social norms become institutionalized and sustained even
in the absence of conscious decisions by people to accept these norms. The net effect
sketches how the constraints of a physical space and the assigned but indeterminate
roles that people must work out within them determine who people can be.50
The emphasis on how people come to inhabit the roles offered them emphasizes
the existential insight that constraints allow for the discovery of who we are. For in
the process of filling out a role, one learns and defines in a literal sense who one is.
Does one feel different, has one a different experience of oneself, if one is led to see oneself
as a certain type of person? Does the availability of a classification, a label, a word or phrase,
open certain possibilities, or perhaps close off others? . . . It seemed to me that a new way of
describing people does not only create new ways to be, but also new ways to choose—in the
existentialist philosophy, new ways to choose who one is.51
Marx famously observed that ‘Men make their own history, but they do not make it
just as they please; they do not make it under circumstances chosen by themselves,
but under circumstances directly encountered, given, and transmitted from the
past.’52 Hacking looks to update just this view by supplementing Marx’s broad
sweep with the finer grained analyses of Foucault and Goffman.
I want instead to draw on both the archaeological and the sociological approach to better
understand the ways in which the actual and possible lives of individuals are constituted. . . .
This is not because how we are can be freely chosen, but because the choices that are open to
us are made possible by the intersection of the immediate social settings . . . and the history
of that present.53
The unifying point linking the thinkers discussed concerns the importance of
choices. People create themselves within institutional spaces and categories not
of their making. In this regard, Hacking’s work extends our understanding of how
50 Erving Goffman, Asylums: Essays on the Social Situation of Mental Patients and other Inmates
(Garden City, NY: Anchor Books, 1961). Hacking puts the point of comparison this way: ‘Goffman
analyzed, by a series of ideal types, the ways in which human roles are constituted in the face-to-face
interactions within an institution setting, and how patterns of normality and deviance work on
individual agents—and how the agents change those norms, by a sort of feedback effect. Foucault’s
archaeologies establish the preconditions for and the mutations between successive institutional
forms. His later genealogies are closer to how the historical settings work on people to form their
potentialities, but never indicate how this happens in daily life. Goffman does that in rich detail, but
gives no hint of how the surrounding structures themselves were constituted’ (‘Between Michel
Foucault and Erving Goffman’, 288).
51 Ibid. 285.
52 Karl Marx, ‘The Eighteenth Brumaire of Louis Bonaparte’, in Lewis S. Feuer (ed.), Marx and
Engels: Basic Writings on Politics and Philosophy (New York: Anchor Books, 1959), 320.
53 Hacking, ‘Between Michel Foucault and Erving Goffman’, 288.
the conformity effect ranges over medical categories and social institutions. It helps
replicate the categories created by which to count, classify, and treat people.
Perpetrators may be unwittingly made by circumstance, but the same can be said
for many other types of people. My argument has been to establish that the ability of
the situationist approach to deal successfully with the choice problem and the smile
problem implies that the affective dimensions of genocidal behaviours—the alarming
enthusiasm people bring to these roles—needs (alas) no special, or especially deep,
explanation. The nasty neighbour problem and its proposed solution, i.e., that people
implicitly recognized that role-determined behaviour does not constitute a general
threat once the general circumstances have changed, adds I suggest an additional
measure of credence to the results. To the best of my knowledge, no other explanatory
proposal put forward so neatly resolves all three of these problems. These activities,
rather, reflect all too well the sort of elements involved in ‘making up people’.
That these problems all find explicit anticipation in Arendt’s Eichmann in
Jerusalem underscores the ways in which philosophy and social science remain
curiously tied together, at least on this issue. For on the one hand, the social
scientific resources exist for explaining why people would suddenly become mur-
derers on a mass scale, pursue this with grisly abandon, and then return to ‘normal’
lives when circumstances change. On the other hand, the empirical work appears to
generate a type of moral revulsion or psychological resistance to the idea that people
can just be so malleable. But perhaps a key to lessening the likelihood of creating
perpetrators resides precisely in coming to terms with acknowledging the fact that
though people can choose against conformity effects, relatively few will.
A C K N OW L E D G E M E N T S
................................................................................................................
I thank Donald Bloxham, Yves Gingras, Colin Koopman, Dirk Moses, and Renee C.
Winter for their comments on earlier drafts of this essay.
SELECT BIBLIOGRAPHY
Adorno, Theodor W., ‘What Does Coming to Terms with the Past Mean?’, in Geoffrey H.
Hartman (ed.), Bitburg in Moral and Political Perspective (Bloomington, IN: Indiana
Browning, Christopher R., Ordinary Men: Reserve Police Battalion 101 and the Final Solution
in Poland (New York: Harper Perennial, 1993).
216 paul a. roth
Hacking, Ian, Rewriting the Soul: Multiple Personality and the Sciences of Memory (Prince-
ton: Princeton University Press, 1995).
—— ‘Between Michel Foucault and Erving Goffman: Between Discourse in the Abstract
and Face-to-Face Interaction’, Economy and Society 33 (2004), 277–302.
Moses, A. D., ‘Structure and Agency in the Holocaust: The Case of Daniel Jonah
Goldhagen’, History and Theory 37:2 (1998), 194–219.
Roth, Paul A., ‘Beyond Understanding: The Career of the Concept of Understanding in the
Human Sciences’, in Stephen P. Turner and Paul A. Roth (eds), The Blackwell Guide to the
Philosophy of the Social Sciences (Malden, MA: Blackwell, 2003).
—— ‘Hearts of Darkness,’ History of the Human Sciences 17:2–3 (2004), 211–51.
Stone, Dan (ed.), The Historiography of the Holocaust (New York: Palgrave Macmillan,
2004).
Zimbardo, Philip, The Lucifer Effect: Understanding How Good People Turn Evil (New York:
Random House, 2008).
chapter 11
.............................................................................................
P H I LO S O P H Y A N D
GENOCIDE
.............................................................................................
martin shuster
}1
................................................................................................................
Surprisingly few philosophers have written about genocide, leading to the
question of what it would mean to pose genocide as a philosophical problem. In
the course of this chapter, I will review some of the ways philosophers have
addressed genocide and then, by using the work of Max Horkheimer and Theodor
W. Adorno, suggest an alternative way in which they can proceed further. The first
question to pose, then, is why have philosophers been prone not to discussing
genocide?1 Answering this question will go a long way in helping to understand
what philosophy can and cannot do in analysing genocide.
Many philosophers fall into two broad camps: those who feel uncomfortable
addressing genocide, and those who do not see any particular philosophical
problem in the occurrence of genocide. The former tend to feel as if a philosophical
intervention into genocide would be unfruitful owing to the complexities involved:
either that they lack the requisite empirical knowledge or that they need to gain
clarity on a variety of ‘smaller’ issues involving perhaps ethics, social/political
philosophy, and/or the philosophy of action (just to name a few areas); all of
1 Kenneth Seeskin raised this question in 1992 about the Nazi genocide and philosophy, but the
point still stands today and with respect to the study of genocide as such. See his review of Berel Lang’s
Act and Idea in the Nazi Genocide: Kenneth Seeskin, ‘Lang, Act and Idea in the Nazi Genocide’, The
Jewish Quarterly Review 88, no. 1–2 (1992), 273.
218 martin shuster
these issues would need to be properly worked through before tackling the ‘big’
issue of genocide.
Alternatively, or coextensively, it may be the case that philosophers simply do
not see a particular philosophical problem in the occurrence of genocide. For
example, one of the most notable monographs to come out on evil in recent
years treats genocide as an example of atrocity,2 different in scale, scope, and/or
degree, but not in kind from other evils.3 On such a view, an understanding of evil
implicitly addresses itself to understanding genocide. One could repeat this proce-
dure with other issues such as racism, murder, and so forth: genocide is implicitly
addressed by considering these other issues. Drawing a connection between phi-
losophy and genocide can then be either too easy or too difficult.
Another way of addressing genocide, but not dealing with it as a particular
problem, is to see genocide as representative of some deeper problem, for example,
nihilism or relativism. Certain readings of Martin Heidegger take this to be his
position: that the Holocaust (and presumably other genocides) is simply a mani-
festation of a deeper nihilism,4 which itself is simply a stage in the ‘destiny of Being’
(Seingeschick).5 Alternatively, genocide may be seen as exemplifying or illustrating
some more basic philosophical problem. For example, one may ask about the exact
parameters of moral responsibility and then adduce a particular instance within a
specific genocide as revealing some feature about moral responsibility.6 With these
sorts of approaches, it is often irrelevant which particular genocide is being
discussed or whether the perpetrator in question is from the Nazi Waffen-SS or
Hutu interahamwe. While this generality may or may not be problematic for the
study of genocide as such (see my discussion of the Nazi genocide in the next
section), the chief issue with the approaches just outlined is that they are not really
insights into genocide, at least no more than discussions of murder generate
insights into suicide—that is, such discussions are and are not insightful (they
are insofar as suicide is a species of murder, but they are not insofar as the
differences between the two seems to be precisely the point). Furthermore, it
should be noted that defining philosophy is as problematic as defining genocide
2 See Adam Morton, On Evil (New York: Routledge, 2004).

3 Emil Fackenheim criticized this sort of approach to the Nazi genocide in To Mend the World
(Bloomington: Indiana University Press, 1994), 11.
4 For a concise overview of Heidegger’s thought vis-à-vis nihilism, see Robert B. Pippin,
Modernism as a Philosophical Problem: On the Dissatisfaction of European High Culture, 2nd edn
(Oxford: Blackwell, 1999), 114–44.
5 See Richard Wolin, The Terms of Cultural Criticism: The Frankfurt School, Existentialism,
Poststructuralism (New York: Columbia University, 1995), 142. For a compelling rebuttal of Wolin
and readings akin to his, see Philippe Lacoue-Labarthe, Heidegger, Art, and Politics: The Fiction of the
Political, trans. Chris Turner (Oxford: Basil Blackwell, 1990). There is here an added question of
Heidegger’s relation to the Nazi genocide in the context of Nazism. See }6.
6 See, for example, Geoffrey Scarre’s ‘Moral Responsibility and the Holocaust’, in Eve Garrard and
Geoffrey Scarre (eds), Moral Philosophy and the Holocaust (Burlington: Ashgate, 2003), 103–16.
philosophy and genocide 219
(see below): the philosophical enterprise itself is often a contested or multifaceted

mode of inquiry,7 spanning a variety of approaches and orientations.
}2
................................................................................................................
Before suggesting an alternative way in which philosophers can contribute to the
study of genocide (}}7–10), I would like, in the several sections that follow, to
review the various species of ways in which philosophers have specifically ad-
dressed genocide. I should mention that when discussing genocide, philosophers
like many scholars have often focused on the Nazi genocide. This attention is not
unjustifiable,8 but I do think that it can be dangerous by giving one the false sense
that whatever one says about the Nazi genocide can thereby be said of genocide as
such. That is one issue; another is that the uniqueness of the Nazi genocide has
itself become a topic of inquiry, with some stressing its uniqueness.9 I do not enter
this debate in any fashion here, except only to say both that I view the Nazi
genocide as an act of genocide and that concerns about the applicability of
philosophical insights gleaned from an analysis of the Nazi genocide must be
addressed on a case by case basis in order to determine their applicability to
genocide as such.
This point about the status of Nazi actions vis-à-vis genocide and their unique-
ness or lack thereof naturally suggests one of the first areas of inquiry pursued by
philosophers: the definition of genocide itself. There are different ways of framing
this question: one may inquire about the definition itself—what does it mean to
commit an act of genocide? What other terms does it presuppose for its efficacy?
How does it differ from ethnic cleansing or mass murder? Or, alternatively, one
may ask about the nature of the groups targeted by perpetrators—what sorts of
7 For some of the issues, see Richard Rorty, ‘Analytic and Conversational Philosophy’, in idem (ed.),
Philosophy as Cultural Politics: Philosophical Papers, vol. 4 (Cambridge: Cambridge University Press,
2007), 120–31.
8 Particularly, since historically the event often is taken to be a watershed event. See Christopher
Browning, Nazi Policy, Jewish Workers, German Killers (Cambridge: Cambridge University Press,
2000), 32. The question here, of course, is whether the same is true philosophically.
9 See Alan Rosenberg, ‘Was the Holocaust Unique? A Peculiar Question?’, in Isidor Walliman and
Michael N. Dobkowski (eds), Genocide and the Modern Age (New York: Greenwood, 1987), 145–61.
Also, Kenneth Seeskin, ‘What Philosophy Can and Cannot Say about Evil?’, in Alan Rosenberg and
Gerald E. Myers (eds), Echoes from the Holocaust (Philadelphia: Temple University, 1988). More
recently, Norman Geras also takes up this issue in ‘In a Class of its Own?’, in Garrard and Scarre
(eds), Moral Philosophy and the Holocaust, 25–57.
220 martin shuster
groups can be the victims of genocide?10 Obviously, these questions are related:
one’s definition informs what sorts of groups fall under its purview and one’s
understanding of victim groups informs one’s definition. This is furthermore
complicated by the fact that genocide is largely a legal term,11 and so already
some legal framework is implicitly presupposed. I mention this here at the outset
because it is not at first obvious why philosophers should be any more qualified to
define or comment on genocide in this definitional capacity than jurists or social
scientists. One way to refine this point is to distinguish between two approaches to
genocide definition. On one hand, one can be doing work with genocide definition
in order to find ways to link particular events together or as a precursor to a
typology of genocide. In proceeding this way one must be able to cogently
distinguish genocide from ethnocide (or ‘cultural’ genocide), take stock of various
proposed definitions, perhaps account for its misuse, and so forth. On the other
hand, however, one could be working with the definition in order to make sense of
what it would mean for an agent to have the intention of genocide. These two
approaches are not mutually exclusive, but the former is largely a task best left to
jurists, historians, and social scientists, since they are more equipped to proceed in
such a fashion,12 while the latter is eminently ripe for philosophical discussion.
Indeed, the notion of intention—a philosophical mainstay since philosophy’s
inception—underpins the two parameters (act and group membership) men-
tioned above.
}3
................................................................................................................
Several distinct questions emerge with a consideration of intention in the context
of genocide. First, is the question of whether perpetrators can intend to do evil: that
is, do we take it that genocide perpetration involves wanting to do evil or are the
perpetrators acting on (ultimately incoherent or mistaken) utilitarian grounds (i.e.
are they ultimately assuming that they are acting for the good)?13 Related to this
10 Compare to Berel Lang, Act and Idea in the Nazi Genocide (Chicago: University of Chicago Press,
1990), 6.
11 For two differing positions on whether genocide is fundamentally a legal or a sociological term,
see William A. Schabas, ‘The Law and Genocide’ and Martin Shaw, ‘Sociology and Genocide’, both in
this volume.
12 Two overviews of the issues involved in such an approach are Mark Levene, Genocide in the Age
of the Nation State, 4 vols, vol i: The Meaning of Genocide (London: I. B. Tauris, 2005); and Scott
Straus, ‘Contested Meanings and Conflicting Imperatives: A Conceptual Analysis of Genocide’,
Journal of Genocide Research 3:3 (2001), 349–75.
13 Lang, Act and Idea in the Nazi Genocide, 26.
question is the question of what perpetrators must ultimately think of victims. Is it

the case that it is truly possible to treat victims as not human? Or is it the case that
any treatment, no matter how inhuman, fundamentally presupposes a victim’s
human agency?14
A not wholly unrelated, even broader question centres on the very nature of
intention itself. First, one may ask whether and to what extent it is possible to
separate—prior to the act—intentions from actions.15 Second, and related, the way
an action is understood is dependent on the way in which it is described: the same
action may be susceptible to more than one correct description.16 It is possible,
then, that the same action could turn out to be intentional or unintentional,
depending on the way in which it is described; furthermore, descriptions them-
selves then become a site for philosophical discussion, particularly in the context
of genocide. It may be that the whole view of selves as discrete monads expressing
‘inner’ states via ‘outer’ acts is itself problematic and that the self is itself a
fundamental social achievement, its inner and outer components being inseparable
from its existence in a social space.17
From a certain perspective, then, debates like those that took place between
functionalists and intentionalists seem unusually contrived, with both sides start-
ing with not only unconvincing, but also unexamined views of individuals and
acts.18 Further, the relative reconciliation between the two groups seems almost
accidental, with neither having examined the expansive philosophical literature on
the topic.19
Likewise, read with these issues in mind, Hannah Arendt’s work then becomes
the obverse side of this discussion.20 The banality of evil is an affirmation not only
of the fundamentally social nature of selfhood, but also a powerful comment on all
of the issues raised here. In addition to the trivial point that genocide requires a
particular social backdrop to its execution, I take Arendt to argue that ‘thought-
lessness’ is a particular, modern way of comporting oneself towards or navigating
14 Berel Lang, ‘The Concept of Genocide’, Philosophical Forum 16:1–2 (1984/5), 10–13. Also see
Stanley Cavell, The Claim of Reason: Wittgenstein, Skepticism, Morality, and Tragedy (Oxford: Oxford
15 Lang, Act and Idea in the Nazi Genocide, 24.
16 See Donald Davidson, ‘Actions, Reasons, and Causes’, in Essays on Actions and Events (Oxford:
17 Terry Pinkard, Hegel’s Phenomenology: The Sociality of Reason (Cambridge: Cambridge
University, 1996), 142.
18 Berel Lang makes an analogous point in ‘Philosophy’s Contribution to Holocaust Studies’, in
Garrard and Scarre (eds), Moral Philosophy and the Holocaust, 6.
19 Two excellent overviews of the philosophical issues involved in analysing intentions are G. E. M.
Anscombe, Intention (Ithaca, NY: Cornell University, 1969); and Alfred Mele (ed.), The Philosophy of
Action (Oxford: Oxford University Press, 1997).
20 For a more elaborate treatment of Arendt from social psychological perspective, see Paul A.
Roth, ‘Hier ist kein Warum: Philosophical Reflections on the Social Psychology of Genocide’, in this
volume.
222 martin shuster
the world, one that involves seeing people and things in a distinct light and one that
in no way abrogates responsibility, at least no more than the almost automatic
actions of a virtuous person preclude them from moral praise.21 Read in this light,
then, the earlier Origins of Totalitarianism has a strong link to Eichmann in
Jerusalem: the former provides the theoretical backdrop for the latter, laying out
the social and ethical structures and conditions for the destruction of human
spontaneity, while the latter explores the moral psychology of a perpetrator in
such a state of affairs.22 Furthermore, it is precisely at this point that Michel
Foucault’s and Giorgio Agamben’s work on biopolitics—the idea that the modern
sovereign has power not just over subjects but power of administration over the
biological process as such (i.e., power over life, not just death)—connects to our
discussion.23 If we find Arendt’s reading of totalitarianism as the project of the
destruction of human spontaneity compelling,24 then biopolitics becomes the (or
arguably, a) means of completing the account, of sketching its stakes and logic.25
Through biopower, the involvement of a state in the regulation of ever more levels
of human life down to bodily control, reproduction, and ultimately consciousness,
a state strives to radically alter the nature(s) of its subjects, thereby transforming
the very makeup of the state from the ‘ground up’.
}4
................................................................................................................
As I mentioned earlier, however, intentions are only one side of the coin; scholars
have focused as well on elaborating the notion of victim groups as a way of linking
act/intention and genocide. One way to do this is to elaborate a thesis of ‘social
death’, wherein what is produced is a ‘meaninglessness of one’s life and even of its
termination’.26 Social death is a state of affairs where one is ‘no longer able to pass
21 Compare to John McDowell’s position on virtue: John McDowell, Mind, Value, and Reality
(Cambridge: Harvard University Press, 1998), 77–221.
22 Richard J. Bernstein, Hannah Arendt and the Jewish Question (Cambridge, MA: MIT Press, 1996),
137–53. For an opposing view, see Dana Villa, ‘Conscience, the Banality of Evil, and Idea of a
Representative Perpetrator’, in Politics, Philosophy, Terror: Essays on the Thought of Hannah Arendt
(Princeton: Princeton University Press, 1999), 39–61.
23 See Giorgio Agamben, Homo Sacer: Sovereign Power and Bare Life (Stanford: Stanford
University, 1998), Michel Foucault, The History of Sexuality, vol. 1 (New York: Vintage, 1978), 135–59.
24 See Hannah Arendt, The Origins of Totalitarianism (London André Deutsch, 1986), 437–60.
25 For more on this, see J. M. Bernstein, ‘Intact and Fragmented Bodies: Version of Ethics “after
Auschwitz”’, New German Critique 33:1 (2006), 40.
26 Claudia Card, ‘Genocide and “Social Death”’, Hypatia 18:1 (2003), 73.
along build upon . . . traditions, cultural developments . . . and projects of earlier

generations.’27 By focusing on the consequences for victim groups, the social death
view is able to account for a meaningful distinction between mass murder and
genocide, but only at the expense of collapsing the distinction between ethnic
cleansing and genocide (already fragile and partial as it is).28 On this view, what is
important is the ‘social health’ of a particular group.29 Stress is taken off ‘mere’
physical well being (although this too is important) and put onto the various items
that feed into providing a meaningful life for agents (such social goods as relation-
ships, goals, and traditions). Furthermore, it is precisely the assault on these social
components that makes social death intolerable, ultimately distinguishing geno-
cide from mass murder of individuals as individuals. The chief issue with this view
is that although perverse, it may nonetheless be true that genocidal suffering may
contribute to the flourishing of social health (indeed, one can imagine robust
communities created solely due to such suffering). If one finds this depraved and
points to the great bodily harm (and mass death) involved in such suffering, then
one is invoking a standard above and beyond that of social death, and so the
original intention of the thesis loses its potency. All of this points to the funda-
mentally normative character of what is used to anchor this view: that social death
is an evil precisely because it is intolerable—but ‘intolerable’ is itself a contested
term that pushes those invoking it to instance some other justification or justifica-
tory standard.30
}5
................................................................................................................
The focus on victim groups above goes hand in hand with three other broad areas
of inquiry that philosophers have pursued: trauma, memory, and the limits and
status of representation. Often drawing on the connections between these three,
one can ask about the experience of and relationship between trauma and memory;
and in doing so, one may ask about the limits of representation.
27 Ibid. Card is drawing on the work of Orlando Patterson; see Orlando Patterson, Slavery and
Social Death (Cambridge, MA: Harvard University Press, 1982).
28 In defending Card, Mohammed Abed embraces this conclusion. See Mohammed Abed,
‘Clarifying the Concept of Genocide’, Metaphilosophy 37:3–4 (2006), 308–30.
29 This phrase is mine.
30 For more on this, see Adam Morton, ‘Inequity/Iniquity: Card on Balancing Injustice and Evil’,
Hypatia 19:4 (2004), 199.
224 martin shuster
Generally, discussion of this point can be mapped onto two opposing poles,31
consisting of realists and antirealists.32 In discussing the Nazi genocide, for exam-
ple, Zygmunt Bauman is able, via a realist account (an account that purports to
provide a sufficient theoretical explanation that posits a more or less ascertainable
correspondence between its terms and the empirical reality to which its terms
allude), to make the genocide intelligible using a sociological framework.33 Alter-
natively, Elie Wiesel has argued that the Nazi genocide simply cannot be explained,
that it ultimately ‘transcends history’.34 In turn, Giorgio Agamben has attempted to
draw the implications of looking at the relationship as a fundamentally contra-
dictory or aporetic one: where the notion of testimony is precisely the relation
between being compelled to bear witness to an event that fundamentally cannot be
witnessed.35
Attempting to bypass this debate, some have tried to ignore this broader
question of the relationship between these two poles and instead have focused on
analysing issues that transcend or saturate these poles. For example, philosophers
since Augustine have inquired about the nature of memory, and it is quite possible
that genocide either introduces some sort of particular nuance on the topic or
raises fundamental issues about or in relation to it.36 Or the experience of trauma
or suffering as such may fundamentally alter one’s constitution of experience and/
or the self (whether experientially, phenomenologically, or otherwise).37 Indeed, in
exploring this area, one quickly notices that the three topics mentioned in the
beginning of this section are difficult to unweave: they point towards one anoth-
er.38 Although a difficulty, this is also precisely where more ‘genocide’-themed
philosophy precisely pushes back towards ‘mainstream’ philosophical concerns
with the nature of the self, the philosophy of mind, and so forth.
31 For an overview, see Michael Rothberg, ‘Introduction’, to Traumatic Realism: The Demands of
Holocaust Representation (Minneapolis: University of Minnesota Press, 2000).
32 For an example of the former, see the aforementioned Hannah Arendt, Eichmann in Jerusalem
(New York: Viking Press, 1963). For an example of the latter, see Arthur A. Cohen, The Tremendum
(New York: Continuum, 1981).
33 Zygmunt Bauman, Modernity and the Holocaust (Ithaca, NY: Cornell University, 2000).
34 Irving Abramson (ed.), Against Silence: The Voice of and the Vision of Elie Wiesel (New York:
Holocaust Library, 1985), 158, Elie Wiesel, ‘Trivializing the Holocaust: Semi-Fact and Semi-Fiction’,
New York Times, 16 April 1978.
35 Giorgio Agamben, Remants of Auschwitz (New York: Zone, 1999), 13. See also Sara Kofman,
Smothered Words, trans. Madeleine Dobie (Evanston: Northwestern University Press, 1998), 36.
36 For an overview, see Robert Eaglestone, The Holocaust and the Postmodern (Oxford: Oxford
University Press, 2004), especially 72–101.
37 Adorno has been read as developing this thread. For more on this point, see Raymond Geuss,
‘Suffering and Knowledge in Adorno’, Constellations 12:1 (2005), 3–20. The aforementioned Rothberg
book, Traumatic Realism, attempts to think through a realist account under the aspect of trauma.
38 Ruth Leys’s genealogical analysis of trauma superlatively illustrates precisely this point; see Ruth
Leys, Trauma: A Genealogy (Chicago: University of Chicago, 2000).
}6
................................................................................................................
The approaches sketched in }}3–5 manage to tease philosophical threads out
of genocide, but it may be possible that the enterprise of philosophy itself—as
such—is causally connected to genocide. There are many ways to establish this
connection: one can argue that certain thinkers (e.g. Heidegger or Nietzsche)
endorsed philosophical positions that directly led to, encouraged, or justified
genocide.39 Or one may argue that the whole of the philosophical tradition is
analogously related to genocide (whether as direct input or as intellectual pre-
cursor); that is, that the philosophical tradition itself carries a kernel of thinking
or proposes an orientation to the world that is somehow explicitly or implicitly
genocidal.40 However one draws this connection, under this rubric, some sort of,
generally causal, connection is established between a particular philosophical
position and genocide.
There are many problems with this sort of approach. First is that it commonly
elides the distinction between complicity and causality (and, furthermore, does not
even distinguish between degrees of complicity).41 Second, there is the danger of
mere reductio ad Hitlerum, that is, of mere guilt by association. Third, and most
important, however, is that crucial differences on both ends of discussion, on the
pole of genocide or philosophy, are simply overlooked or elided. On the pole of
genocide, this approach can fundamentally miss the specificity of genocide and
perpetrators, lumping very diverse groups of perpetrators together, ignoring their
individual differences in motivation for, role in, and history amidst genocide.42 In
order to draw analogues between X ideology and Y philosophy, one must posit a
coherent ideology X, but what genocide study has revealed is that genocides often
only have vague (or confused or diverse or multifaceted) ideologies behind them
39 See, e.g., Hubert Kiesewetter, Von Hegel zu Hitler: Die politische Verwirklichung einer totalitären
Machtstaatstheorie in Deutschland, 1815–1945 (Frankfurt am Main: Peter Lang, 1974); Abir Taha,
Nietzsche, Prophet of Nazism: The Cult of the Superman—Unveiling the Nazi Secret Doctrine
(Bloomington: AuthorHouse, 2005), Richard Wolin, The Politics of Being (New York: Columbia
University Press, 1990); idem, The Seduction of Unreason: The Intellectual Romance with Fascism
from Nietzsche to Postmodernism (Princeton: Princeton University Press, 2006). See also fn 5 above.
40 See, e.g., David Patterson, ‘The Complicity of Modern Philosophy in the Extermination of the
Jews’, in Dennis B. Klein, et al. (eds), The Genocidal Mind: Selected Papers from the 32nd Annual
Scholars’ Conference on the Holocaust and the Churches (St Paul: Paragon House, 2005), 32–63, idem,
Open Wounds: The Crisis of Jewish Thought in the Aftermath of Auschwitz (Seattle: University of
Washington Press, 2006).
41 For more on this, see David E. Cooper, ‘Ideology, Moral Complicity and the Holocaustm’, in
Garrard and Scarre (eds), Moral Philosophy and the Holocaust, 9–24.
42 See Peter Haas’s critique of David Patterson in John K. Roth (ed.), Ethics after the Holocaust:
Perspectives, Critiques, and Responses (St. Paul: Paragon House, 1999), 177–80.
226 martin shuster
and that the variety of agents involved in genocide are often operating at different
levels of influence vis-à-vis ideology (if at all).43
Analogously, scholars may perform the same selective procedure at the philosoph-
ical end of the discussion: they may cast their philosophical figures in hackneyed or
uncritical terms, ignoring certain sources in favour of others, or simply approaching
the sources they do pick with a lack of subtlety.44 For scholars who urge a relationship
between not just a particular philosopher, but the whole philosophical tradition,
these two dangers are compounded by a third. Often, scholars who posit the case
must oppose the philosophical tradition to something—in some cases it is religion,45
or in others it is an alternate or revised conception of philosophy.46 The danger here
is that oppositions of this sort are difficult to delineate let alone maintain, particu-
larly as the scholars in question are—presumably—incapable of fully transcending
the philosophical traditions which they implicate in genocide; at the very least, it is
not so easy to think through such an opposition, as opposed to merely posit it (such
oppositions perhaps may be maintained, but they are generally marshalled with
insufficient resources).47
In addition, the analogy to the approach sketched in }1 should by now be
apparent. This approach proceeds largely by taking genocide as an instantiation
of a deeper difficulty: whether it is a problem with the philosophical tradition as
such or a problem with a particular philosopher who makes, by association or
appropriation, certain segments or types of society susceptible to genocide perpe-
tration. These issues, however, are insufficiently contextualized—historically and
philosophically—and thereby genocide is seen to arise—at least philosophically—
out of thin air. If, for example, we take Nietzsche’s will to power to be the force that
43 For more on this point with respect to the Nazi genocide, see Donald Bloxham and Tony
Kushner, The Holocaust: Critical Historical Approaches (Manchester: Manchester University Press,
2005), 61–176. For Rwanda, see Scott Straus, The Order of Genocide: Race, Power, and War in Rwanda
(Ithaca, NY: Cornell University Press, 2008), 65–153. Analogous procedures can be performed with all
of the major genocides of the twentieth century.
44 Figures like Kant, Hegel, and Nietzsche are notoriously misrepresented in this literature. All
three figures had a very complex relationship, e.g., to Judaism. For a corrective on Hegel and
Nietzsche, see Yirmiyahu Yovel, Dark Riddle: Hegel, Nietzsche, and the Jews (Cambridge: Polity,
1998). As well, for Nietzsche, see Weaver Santaniello, Nietzsche, God, and the Jews: His Critique of
Judeo-Christianity in Relation to the Nazi Myth (Buffalo: State University of New York Press, 1994). In
order to deal with Kant’s relationship to Judaism, one would need to take seriously Kant’s
developments in the Opus Postumum, as I intend to in future work.
45 E.g., for David Patterson, it is a particular gloss of Judaism. See Patterson, Open Wounds, 32–63.
46 I will come to this shortly, but I have in mind thinkers like Adorno and Levinas.
47 For more on some of the issues with this and the opposition between Judaism/Greek thought (of
which David Patterson makes so much of ), see Leora Batnitzky, Leo Strauss and Emmanuel Levinas:
Philosophy and the Politics of Revelation (Cambridge: Cambridge University Press, 2006), 5–8. See also
the classic piece by Derrida, ‘Violence and Metaphysics’, in Writing and Difference (Chicago:
University of Chicago, 1978), 79–154. On the point of philosophy and religion, see Hent de Vries,
Philosophy and the Turn to Religion (Baltimore: Johns Hopkins University Press, 1999).
drives genocide, then it remains entirely unanswered not only why the will to power
must manifest itself in this fashion, as opposed to, e.g., ‘mere’ mass murder or
pogromic violence or disciplinary force or whatever, but also why the will to power
became a compelling philosophical position in such a context in the first place. Just
as historical analyses of genocide must be situated in a broader matrix of Western
history and civilization, so too must philosophical analyses be deployed in a similar
expansive but nonetheless historicized context.48
}7
................................................................................................................
One way to respond to the above problems of contextualization is to attempt to
connect the project of Enlightenment with genocide. There are others, but I focus
on the Enlightenment precisely for its suggestiveness and its scope.49 On such a
move, genocide becomes a structural fixture within the evolution of the broader
landscape of our political, economic, and social systems and commitments. This
proposal, of course, is a very complicated procedure. First, ‘Enlightenment’ is a
contested term. Second, making sense of the relation—if any—between Enlighten-
ment and genocide is notoriously difficult.50 In part, this has to do with the
relationship between ideas and events: in a certain sense, it is obviously true that
the Enlightenment had a hand in what came after it—but in this sense, it is trivial,
as trivial as pointing out that metallurgy or the principle of non-contradiction had
a hand in genocide. In another sense, it is patently unbelievable to assert that the
Enlightenment as a series of events had anything to do with genocide.
The key, then, is in how we gloss our terms (‘genocide’ included, see }2). The
concerns here can be separated most broadly into historical and philosophical
ones. The historical issues include locating the Enlightenment: was there a coherent
48 For an argument for the necessity of such a historical contextualization as well as an example of
it, see Levene, Genocide in the Age of the Nation State.
49 A related approach would be to connect genocide to modernity. For an overview, see A. Dirk
Moses, ‘Genocide and Modernity’, in Dan Stone (ed.), The Historiography of Genocide (Houndmills:
Palgrave Macmillan, 2008), 156–92. See also Martin Shaw’s discussion of Zygmunt Bauman in his
‘Sociology and Genocide’, in this volume. As well, one could focus on capitalism; see Moishe Postone’s
‘Nationalsozialismus und Antisemitismus: ein theoretischer Versuch’, in Dan Diner (ed.),
Zivilisationsbruch: Denken nach Auschwitz (Frankfurt am Main: Fischer, 1988), 242–254; Moishe
Postone, Time, Labor, and Domination: A Reinterpretation of Marx’s Critical Theory (Cambridge:
50 For an overview, see John Docker, ‘The Enlightenment, Genocide, Postmodernity’, Journal of
228 martin shuster
Enlightenment programme? Who were its chief figures, time periods, and locales?51
The philosophical issues include, implicitly, these historical issues as well as the
problem of accounting for the scope of the enlightenment: if enlightenment thinking
is taken in a broad enough sense, then it includes the very project of philosophy
itself, and if that is the case, then it becomes difficult—if not impossible—to use
philosophy to criticize its own shortcomings. Indeed, there arise then crucial and
complex issues centring on autonomy.52 Ultimately, if a mode of thinking itself is
at stake—as opposed to any particular historical period—then questions of
method must be explicitly addressed. Horkheimer and Adorno’s Dialectic of
Enlightenment (DE) is fruitful in this context; I turn to it in the following
sections.53
}8
................................................................................................................
When one looks at DE, however, it is not at first obvious that the account therein
has anything at all to do with genocide. Indeed, Horkheimer and Adorno write in
the Preface that they ‘had set out to . . . explain why humanity, instead of arriving at
a truly human condition, is sinking into a new type of barbarism’ (DE xiv/16).54 Of
course, this question is so broad as to border on vacuous, so Horkheimer and
Adorno quickly revise their intentions as ‘merely’ inquiring about the ‘self-destruc-
tion of enlightenment’ (DE xvi/18). As per }7, this question brings to the fore the
problem of how to understand ‘enlightenment’ in DE. Adorno and Horkheimer
urge the reader not to understand the enlightenment in DE as designating a
51 For more on these issues and others surrounding the Enlightenment, see Peter Gay, The
Enlightenment: The Rise of Modern Paganism (New York: W. W. Norton, 1995), idem, The
Enlightenment: The Science of Freedom (New York: W. W. Norton, 1996), Jonathan I. Israel,
Radical Enlightenment: Philosophy and the Making of Modernity (Oxford: Oxford University Press,
2002), idem, Enlightenment Contested: Philosophy, Modernity, and the Emancipation of Man (Oxford:
52 For an elaboration of this point, see Pippin, Modernism as a Philosophical Problem.
53 The English edition cited is Max Horkheimer and Theodor W. Adorno, Dialectic of
Enlightenment, Philosophical Fragments, ed. Mieke Bal and Hent de Vries, trans. Edmund Jephcott
(Stanford: Stanford University Press, 2002). The German cited is Max Horkheimer, Gesammelte
Schriften, 19 vols (Frankfurt am Main: S. Fischer, 1987), vol. v. Henceforth, DE. All citations in this
piece will follow the format of English/German.
54 Translation modified. Indeed, this opening claim seems precisely to countermand the advice of
their friend Benjamin. In his eighth thesis, Benjamin wrote: ‘The current amazement that the things
we are experiencing are “still” possible in the twentieth century is not philosophical. This amazement
is not the beginning of knowledge—unless it is the knowledge that the view of history which gives rise
to it is untenable.’ See Walter Benjamin, ‘Theses on the Philosophy of History’, in Illuminations
(New York: Shocken Books, 1968), 253–65.
historical period, but rather a process (DE 1/25).55 In order to make sense of the
continual reference to the entirety of the Western philosophical and extra-philo-
sophical tradition, we must understand this ‘process’ maximally as involving not a
critique of culture detailing the various failings of some ill-defined ‘Enlightenment
Project’,56 but rather as involving a philosophical interrogation of the most basic
elements of thinking which aims to enlighten, where such thinking is taken in the
broadest possible terms. As Horkheimer and Adorno write: ‘Enlightenment un-
derstood in the widest sense as the advance of thought, has always aimed at
liberating human beings from fear and installing them as masters’ (DE 1/25).
This notion of enlightenment designates an outlook that spans a variety of tempo-
ral and spatial boundaries.
If we take enlightenment in this sense, then, we need to sketch, at least in broad
terms, some of the claims that Horkheimer and Adorno do make about enlighten-
ment, particularly in the specific relationship that they propose between enlight-
enment and myth. Adorno and Horkheimer write:
Enlightenment has always regarded anthropomorphism, the projection of subjective prop-
erties onto nature, as the basis of myth. The supernatural, spirits and demons, are taken to
be reflections of human beings who allow themselves to be frightened by natural phenom-
ena. According to enlightened thinking, the multiplicity of mythical figures can be reduced
to a single common denominator, the subject. (DE 4/28)
Furthermore, they propose that ultimately ‘enlightenment . . . wanted to dispel

myths, to overthrow fantasy with knowledge’ (DE 1/25). In this sense, the idea is
that ‘the mind, conquering superstition, is to rule over disenchanted nature’ (DE 2/
26). Knowledge, then, is power. What is more, it is a power that ‘knows no limits,
either in its enslavement of creation or in its deference to worldly masters’ (DE 2/
26). So much so that if it is allowed to develop unchecked by anything external,
enlightenment turns on everything, including the things that it professed to hold
dear. All ideas, but most notoriously professedly universalistic ideas like human
rights, become suspect and take on a mythical light (DE 3/27). Every ‘definite
theoretical view is subjected to the annihilating criticism that it is only a belief,
until even the concepts of mind, truth, and, indeed, enlightenment itself have been
reduced to animistic magic’ (DE 7/33). As Horkheimer and Adorno point out, ‘in
the authority of universal concepts enlightenment detected a fear of the demons
through whose effigies human beings had tried to influence nature in magic rituals’
(DE 3/27).57 Indeed, ‘no difference is said to exist between the totemic animal, the
dreams of the spirit-seer, and the absolute Idea’ (DE 3/27). Ultimately, ‘anything
55 For more on this, see Simon Jarvis, Adorno: A Critical Introduction (Cambridge: Polity Press,
1998), 24–6.
56 James Schmidt, ‘What Enlightenment Project?,’ Political Theory 28:6 (2000), 734–57.
57 Translation modified.
230 martin shuster
which does not conform to the standard of calculability and utility must be viewed
with suspicion’ (DE 3/28). In short, ‘enlightenment is totalitarian’ (DE 3/28).
Continuing with the political metaphor, Horkheimer and Adorno write, ‘en-
lightenment stands in the same relationship to things as the dictator to human
beings. He knows them to the extent that he can manipulate them’ (DE 6/31). Since
enlightenment is merely the removal of fear of the unknown, then humans only
‘believe themselves free of fear when there is no longer anything unknown’ (DE 11/
38). To achieve this goal, the self ultimately equates truth in general with classifying
thought (DE 10/36). Fate, then, which enlightenment sought to combat in its battle
against mythology, returns, with a vengeance (DE 8–9/34–5). As Horkheimer and
Adorno point out, ‘abstraction, the instrument of enlightenment, stands in the
same relationship to its objects as fate, whose concept it eradicates: as liquidation’
(DE 9/36). Fate returns in the form of science: where fate once explained everything
as pre-ordained, science performs the same procedure via lawful repetition within
universal laws (DE 8/35). Furthermore, they write, again invoking a political
metaphor, that ‘the distance of subject from object, the presupposition of abstrac-
tion, is founded on the distance from things which the ruler attains by means of the
ruled’ (DE 9/36). Through distance from the particular object to abstraction in the
form of universality, enlightenment reverts to the very mythology that it sought to
overcome. Indeed, we have returned precisely to where we have started: everything
is ‘reduced to a single common denominator’, namely, ‘the subject’ (DE 4/29). Not
only that, but Horkheimer and Adorno further claim that myth itself is already
enlightenment since myth as well sought to ‘report, to name, to tell of origins . . .
therefore to narrate, record, explain’ (DE 5/30). Moreover, since myths are contin-
ually collected and recorded, they themselves in the end become proper teachings
and rituals. For this reason, myth ultimately ‘becomes enlightenment’ (DE 6/31).
Indeed, mythology actually ‘set in motion the endless process of enlightenment’,
since ‘magic like science is concerned with ends’ (DE 7/33). In short, then, ‘just as
myths already entail enlightenment, with every step enlightenment entangles itself
more deeply in mythology’ (DE 8/34).
According to Horkheimer and Adorno, this dialectic then precisely presupposes
genocide—as an institution—as its teleological conclusion. As Adorno writes later
in Negative Dialectics (ND): ‘Genocide is the absolute integration. It is on its way
wherever men are leveled off—“polished off,” as the German military called it—
until one exterminates them literally, as deviations from the concept of their total
nullity. Auschwitz confirmed the philosopheme of pure identity as death’ (ND 362/
355).58 Anything that is unable to be integrated is eliminated.
58 The English edition cited is Theodor W. Adorno, Negative Dialectics, trans. E. B. Ashton (New
York: Continuum, 1973). The German is Adorno, Gesammelte Schriften, 20 vols. (Frankfurt am Main:
Suhrkamp Verlag, 1984), vol v. All citations in this piece will follow the format of English/German.
There is, however, a very deep question here about the necessity of this dialectic.
Even if we grant Horkheimer and Adorno the myriad premises involved in their
argument, it is still not apparent that all rationality—whether in the form of myth or
enlightenment—must be so totalizing. Why precisely is genocide the teleological
conclusion to this process? Even if we grant Adorno the premise ‘that objects do not
go into their concepts without leaving a remainder’ (ND 5/15), that is, that there will
always be something that cannot and will not be integrated, why precisely must this
something be eliminated as opposed to, e.g., marginalized (apartheid) or disempow-
ered (slavery) or spatially rather than ‘existentially’ eliminated (ethnic cleansing or
exile)? While I do think that Horkheimer and Adorno have answers to these questions,
elaborating them is beyond the scope of this chapter; rather, I would like to suggest an
alternative way of instantiating the relationship between philosophy and genocide.
}9
................................................................................................................
It is imperative that philosophers start with the fact of genocide (itself already a
mediated term denoting a variety of connected data). From this we should work
backwards, not with the aspiration of being able to determine causally what prompted
genocide, but rather with the intention of developing, at least provisionally, some
account that can make sense of genocide within a broader philosophical picture, only
in order then—and this is the crucial step—to be able to suggest some deeper or more
fundamental area of philosophical inquiry as following from this account.59 As an
example, I would point out that a neglected significance of DE is that Horkheimer and
Adorno mean to cast suspicion on the inviolability of any sort of philosophical
justification: any ultimate philosophical ground or justification becomes suspect in
light of their claims about reason via the elaboration of their dialectic.60 The philo-
sophical problem of genocide—if there is one—is not, then, that it occurs, but rather
that we can make no ultimately justifiable argument for why it should not occur.
Richard Rorty raises an analogous point when he writes the following:
all vocabularies, even those which contain the words which we take most seriously, the ones
most essential to our self-descriptions—are human creations . . . this would mean giving up
the idea that liberalism could be justified, and Nazi or Marxist enemies of liberalism
refuted, by driving the latter up against an argumentative wall—forcing them to admit
59 Hannah Arendt takes this approach in Origins of Totalitarianism. For more on this, see Robert
Pippin’s ‘Hannah Arendt and the Bourgeois Origins of Totalitarian Evil’, in Alan D. Schrift (ed.),
Modernity and the Problem of Evil (Bloomington: Indiana University, 2005), 149.
60 A notable exception to this neglect is J. M. Bernstein, Adorno: Disenchantment and Ethics
(Cambridge: Cambridge University Press, 2001).
232 martin shuster
that liberal freedom has a ‘moral privilege’ which their own values lacked . . . any attempt to
drive one’s opponent up against a wall in this way fails when the wall against which he is
driven comes to be seen as one more vocabulary, one more way of describing things.61
Indeed, to push Rorty’s point further,62 our liberal vocabulary (vocabulary used
here in the aforementioned Rorty-ian gloss) has as much dissimilarity as it does
similarity to, e.g., the vocabulary of the Third Reich.63 For Horkheimer and
Adorno, the disturbing implication of DE is that not merely totalitarian regimes,
but rather the whole gamut of Western thinking is perhaps complicit, or at the very
least, intimately close to the practice of genocide. Indeed, their account is funda-
mentally opposed to any reading of history that sees genocide as a barbaric or
regressive tendency, as a deviation from modern Western civilization, or as some-
thing opposed to proper, progressive, enlightened thought. Ultimately, for them,
not only is the dualism between the two naı̈ve, but halting genocide is not simply
getting people to be ‘more Western’, or ‘more enlightened’ or ‘more like us’—these
dualisms may themselves precisely be part of the problem.64 Ultimately, then, DE
contextualizes the practice of genocide in a broader Western framework. Where
scholars are performing this procedure in a contextualized historical framework, by
studying the complex historical web of global interests, narratives, and material
conditions that contributes to, props up, or perhaps encourages genocide,65 we
should see Horkheimer and Adorno as initiating a fundamentally similar proce-
dure, only in a philosophical register. It is in this vein that we should understand
Adorno’s 1959 claim, a claim he made after having returned to Germany, ‘I consider
the survival of National Socialism within democracy to be potentially more men-
acing than the survival of fascist tendencies against democracy.’66 The problem,
though, is not relativism or that we precisely have these competing vocabularies,67
61 See Richard Rorty, Contingency, Irony, and Solidarity (Cambridge: Cambridge University,
1989), 53.
62 It may be the case that there is something to be said about how, when, and why vocabularies
change and that Rorty’s position that it is all contingency is dogmatic: see Pippin, Modernism as a
Philosophical Problem, 69.
63 It is in this tenor that we should read Peter J. Haas, Morality after Auschwitz (Philadelphia:
Fortress Press, 1988), Claudia Koonz, The Nazi Conscience (Cambridge, MA: Harvard University Press,
2003).
64 In this sense, we can see it as the philosophical counterpart to historical studies such as Levene,
Genocide in the Age of the Nation State; and Michael Mann, The Dark Side of Democracy: Explaining
Ethnic Cleansing (Cambridge: Cambridge University Press, 2005).
65 See ibid.
66 Theodor W. Adorno, Critical Models: Interventions and Catchwords, trans. Henry W. Pickford
(New York: Columbia University Press, 1998), 90.
67 For an analysis of the connection between relativism and genocide, see Michael Freeman,
‘Speaking about the Unspeakable: Genocide and Philosophy’, Journal of Applied Philosophy 8:1
(1991), 3–17.
but rather that we have these competing vocabularies and they are in such close
proximity to each other.68
Ultimately, given that these competing vocabularies are bound to rub against each
other and given that such friction and disagreement is ultimately a constitutive part of
any community,69 I want to suggest that the question becomes: given this proximity
and inherent friction of vocabularies, how does or can disagreement lead to cohesion
or unity without silencing certain segments of society or without manifest conflict
with others? Ultimately, how do we achieve community? This question, I want to
urge, though is not a question explicitly about how to structure or organize our
communities, but rather is a question about the very possibility of community. Given
the failure of any cognitive model to show conclusively the necessity of a global
community,70 how is it that we can achieve a solidarity that would countenance the
claims of any segment of the global population, not just those segments closest to us
(whether spatially or otherwise)? Again, this is not, at this point, a question about the
empirical specifics of any particular community, or of community as such. (Indeed, it
may be the case that the notion of a ‘global community’ is itself problematic and that
we should be after ‘communities’ in the plural.) Such empirical matters are deeply
important, particularly given the omnipresence, historically, of the notion of com-
munity to the execution of genocide, but answers to these empirical concerns,
however, can only be properly justified after the theoretical ground, centring on
how to gloss the idea of duties or commitments to one another, has been cleared.
}10
................................................................................................................
It is in this tenor that we should understand the later Adorno’s stress on education
‘after Auschwitz’ (Adorno’s stress on the Holocaust should be taken with the caveat
vis-à-vis this point in }4) and his insistence on a new categorical imperative.71 As
68 This issue could be framed as drawing out certain stakes of Davidson’s argument that there can
be no differing conceptual schemes. See Donald Davidson, ‘On the Very Idea of a Conceptual Scheme’,
in Inquiries into Truth and Interpretation, 2nd edn (Oxford: Clarendon, 2001), 183–99. Also, see idem,
‘A Coherence Theory of Truth and Knowledge’, in Subjective, Intersubjective, Objective (Oxford:
Clarendon, 2001), 137–54. Of course, there are deep differences between Rorty and Davidson, see
Robert Brandom (ed.), Rorty and His Critics (Oxford: Blackwell, 2000), 65–81.
69 This is a broad claim that I cannot properly contextualize and justify here, but Stanley Cavell has
explored this issue in various texts. In particular, see his The Claim of Reason: Wittgenstein, Skepticism,
Morality, and Tragedy (New York: Oxford University Press, 1979); and Conditions Handsome and
Unhandsome (Chicago: University of Chicago, 1990).
70 I take this to be a consequence of the argument of DE, although I have not made this argument
explicitly.
71 See Theodor Adorno, ‘Education after Auschwitz’, in Critical Models, 191–204.
234 martin shuster
he writes, ‘Hitler has forced (aufgezwungen) a new categorical imperative upon

humans in the condition of their unfreedom: to arrange their thoughts and actions
so that Auschwitz will not repeat itself, so that nothing similar will occur’ (ND 362/
358).72 Adorno’s’ new imperative poses many questions, not the least of which is
what one should think of such a new imperative after ‘a century of genocide’.73
What is at least as interesting, if not more so, is that Adorno continues with the
following:
This imperative is as rebellious towards its justifications as the given one of Kant’s. To deal
with it discursively would be an outrage [Frevel]: for it causes us to feel, bodily, the moment
of the moral addendum [Hinzutretenden am Sittlichen]. Bodily, because it is now the
practical abhorrence of the unbearable physical agony to which individuals are exposed
even with individuality about to vanish as a form of mental reflection. It is only in the
unvarnished materialistic motive that that morality [Moral] lives on.’ (ND 362/358)74
Adorno’s stress seems precisely to be on an ‘a-cognitive’ (as opposed to non-

cognitive) model of solidarity,75 one that works along some sort of material axis,
with the experience of raw bodily suffering prompting a response. This is what
I take Adorno to mean by the ‘unvarnished materialistic motive’: the somatic
experience of bodily suffering is to call forth an immediate response from an
agent. To attempt to anchor this response in grounds or logical forms (as philoso-
phy does) is to ‘mock’ the ‘urgency’ of this call (ND 285–6/280–3). Of course, it is
notoriously difficult to determine how to ‘cash out’ this proposal of Adorno’s,76
since it precisely brings to the fore all of the methodological issues touched on
earlier. Furthermore and most difficult, if we opt for this a-cognitive approach,
how do we make sense of it philosophically?
To conclude, philosophers can ultimately contribute to genocide studies by
trying their hands at these sorts of questions, formulating others in this vein, and
all the while keeping the context (philosophical and historical) in which such
questions arise in mind. The question of genocide—the philosophical problem
of genocide—is really a cluster of problems that serves to highlight the systemic
and structural nature of genocide. Indeed, if we see genocide as a wretched, but all
too common means of resolving or incorporating conflicts,77 then it becomes
73 Samuel Totten and William S. Parsons (eds), A Century of Genocide: Critical Essays and
Eyewitness Accounts, 3rd edn (New York: Routledge, 2008).
75 There is an affinity between Adorno’s proposal and Rorty’s position in Contingency, Irony, and
Solidarity. Where the two differ is that Rorty flatly says, ‘there is no answer to the question, “Why not
be cruel”—no noncircular theoretical backup for the belief that cruelty is horrible’ (xv). Adorno
would agree with Rorty, but would argue that there is such a backup along a material axis.
Understanding this axis is precisely the challenge.
76 For an attempt, see Bernstein, Adorno, 263–457.
77 See Stanley Cavell’s account of morality in The Claim of Reason, 269.
crucial to explore the deepest, most structural and systematic causes for conflict.
Genocide, then, re-enters the philosophical frame as a distinctly modern, but
nonetheless ultimately human possibility—one that has a particular genealogy
and one that can be explored via its underpinnings in a complex network of
philosophical commitments and positions.
................................................................................................................
Parts of this chapter were written under the auspices of a Diane and Howard Wohl
Fellowship at the Center for Advanced Holocaust Studies, US Holocaust Memorial
Museum, 2007–8. Thanks to the Center staff and fellows, as well, thanks to editors
of this volume for their comments and suggestions.
SELECT BIBLIOGRAPHY
Agamben, Giorgio, Homo Sacer: Sovereign Power and Bare Life (Stanford: Stanford Univer-
sity, 1998).
Arendt, Hannah, Eichmann in Jerusalem (New York: Viking Press, 1963).
—— The Origins of Totalitarianism (London André Deutsch, 1986).
Card, Claudia, ‘Genocide and Social Death,’ Hypatia 18:1 (2003), 63–79.
Eaglestone, Robert, The Holocaust and the Postmodern (Oxford: Oxford University Press,
2004).
Garrard, Eve, and Geoffrey Scarre (eds), Moral Philosophy and the Holocaust (Burlington:
Ashgate, 2003).
Horkheimer, Max, and Theodor W. Adorno. Dialectic of Enlightenment, Philosophical
Fragments, trans. Edmund Jephcott, ed. Mieke Bal and Hent de Vries (Stanford: Stanford
Lang, Berel, Act and Idea in the Nazi Genocide (Chicago: University of Chicago Press, 1990).
Levene, Mark, Genocide in the Age of the Nation State. 4 vols (London: I. B. Tauris, 2005).
Pippin, Robert B., Modernism as a Philosophical Problem: On the Dissatisfaction of European
High Culture, 2nd edn (Oxford: Blackwell, 1999).
Roth, John K. (ed.), Ethics after the Holocaust: Perspectives, Critiques, and Responses
(St. Paul: Paragon House, 1999).
Rothberg, Michael, Traumatic Realism: The Demands of Holocaust Representation
(Minneapolis: University of Minnesota Press, 2000).
part iii
.............................................................................................
PREMODERN AND
EARLY MODERN
GENOCIDE
.............................................................................................
chapter 12
.............................................................................................
GENOCIDE IN THE
A N CIEN T WORLD
.............................................................................................
hans van wees
European literature begins with a story of genocide: the destruction of a city and
its entire population in the legendary Trojan War. In a chilling episode of the Iliad,
the commander-in-chief of the Greek army, Agamemnon, notices his brother
hesitating to kill a Trojan captive, so:
He ran up to him, shouting his orders:
‘Menelaos! . . . Not a single one of them must escape sheer destruction at our hands. Not
even if a mother carries one in her belly and he is male, not even he should escape. All
together they must be exterminated from Troy, their bodies untended and invisible.’
With these words the hero swayed his brother’s mind, since he gave seemly advice.1
Modern readers find it hard to accept that genocide could be called ‘seemly’
(aisima), but although the Iliad sympathizes deeply with the sad fate of individual
Trojans, it also places the blame for the war squarely on the Trojans’ collective
shoulders and there can be no reasonable doubt that for Homer the annihilation of
the Trojans was a legitimate goal of war.2
1 Homer, Iliad 6.54–65. (Translations from Greek and Latin are author’s own; the Bible is cited
from the Revised Standard Version; the sources of translations from other ancient languages are
indicated in the notes.) This paper has benefited from the valuable comments and suggestions of
Amélie Kuhrt, Karen Radner, and the editors.
2 Hans van Wees, Status Warriors (Amsterdam: Gieben, 1992), 176–81; contra e.g. Simon Goldhill,
‘Supplication and Authorial Comment in the Iliad: Iliad Z 61–2’, Hermes 118 (1990), 373–6; Naoko
Yamagata, ‘Aisima pareipon. A Moral Judgement by the Poet?’, La Parola del Passato 45 (1990), 420–30.
240 hans van wees
The massacre of all Troy’s male inhabitants and the enslavement of its women
and children was fictional, but it had many counterparts in ancient history. It was
almost the normative form of genocide—the use of that term will be justified
shortly—in ancient Greece and some other parts of the ancient world, although
mass enslavements and mass executions which made no distinctions of gender or
age are also widely attested. The Greeks’ reasons for treating the Trojans so brutally
were typical of the motivations for genocide in antiquity: it was usually an act of
‘conspicuous destruction’, a display of force designed to assert the power and status
of the perpetrator in the face of a perceived challenge. Ancient genocide sometimes
had a religious dimension, but ‘ethnic’ or racial antagonism rarely if ever played a
part.
Ancient history spans four millennia, from about 3500 bc to ad 500, a geograph-
ical range from Spain to Afghanistan, and a vast number of distinct societies and
cultures, many of which changed dramatically over the centuries. Even if one
concentrates on the first millennium bc, as this chapter does, largely omitting
the civilizations of the Bronze Age and later Rome, to speak of ‘genocide in
antiquity’ therefore entails extreme generalization. Yet broad patterns do emerge,
and it is both possible and worthwhile to analyse in general terms the ideology,
practice, and historical development of genocide in the ancient world.
‘C U T T I N G O P E N P R E G NA N T W O M E N ’:
T H E R H E TO R I C O F G E N O C I D E
................................................................................................................
The evidence for genocide in antiquity ranges from highly rhetorical celebrations
or condemnations of the annihilation of an enemy to laconic notices about the
destruction of cities, and its value is often hard to assess. Is a claim that the enemy
was ‘utterly destroyed’ a record of genocide or a hyperbolic boast of overwhelming
victory? What does it mean when cities are said to be ‘razed to the ground’, when so
many of these places reappear in the sources only a few years later, as if nothing had
happened? We do not always have enough evidence to answer such questions. Yet
even when we cannot tell what reality lay behind the words, the rhetoric is valuable
because it reveals ancient ideologies of genocide.
A case in point is the horrific image of victorious besiegers who ‘cut open
pregnant women’, which is occasionally found in Assyrian and Babylonian
poems, repeatedly conjured up in the Bible, and implied in Agamemnon’s exhor-
tation to his brother. Its earliest appearance is in an Assyrian poem which probably
glorifies a military victory of Tiglat-Pileser I, c.1100 bc:
genocide in the ancient world 241
He slits the wombs of pregnant women; he blinds the infants.

He cuts the throats of their strong ones. . . .
Whoever offends the god Asshur will be turned into a ruin.3
The message here and in the biblical parallels is that enemies of all ages will be
eliminated: adults killed by the sword, children blinded so as to render them
helpless (or, elsewhere, beaten to death), and foetuses ripped from their mother’s
wombs. The sentiment is precisely that of Agamemnon: not even the unborn
should be allowed to escape. In the Iliad, it is specifically male foetuses which are
to be destroyed, and the Assyrian and Hebrew texts, too, seem to allude to the
massacre of men (‘the strong ones’).4
The extermination of males of all ages is presented as legitimate punishment for
offending Asshur or Yahweh. By contrast, the prophet Amos proclaims that God
will punish the Ammonites who cut open pregnant women in the town of Gilead in
Israel merely in order to ‘enlarge their border’. Used in the naked pursuit of power,
without religious justification, this form of genocide seemed unacceptable to a
number of biblical authors.5
Did this brutal rhetoric correspond to an equally brutal practice? One could not,
of course, selectively kill only male foetuses; the pregnant women and their unborn
girls, too, would die. But conceivably this consequence was seen as acceptable
‘collateral damage’ of the symbolically important extermination of all males.
Annalistic accounts of the events sung by poets and predicted by prophets,
however, do not suggest that the entire male population was annihilated, even
when it might have suited their purposes to make this claim.6 As for the Iliad, the
motif of killing unborn boys appears once only, and never again in the whole of
Greek literature, which suggests that it was a piece of extreme rhetoric rather than a
Greek genocidal practice. It may in fact have been a borrowing from the Near East
by Greek raiders and mercenaries, which even as a rhetorical conceit proved too
strong for later Greek tastes. This form of genocide thus served as the ultimate
threat, boast, and accusation for centuries in the ancient world, but may never have
been perpetrated.
An even more extreme example is the ideology of extermination set out in the
laws of war formulated in Deuteronomy, and allegedly put into practice by the
3 VAT 13833, rev. 3–6: text and translation in Mordechai Cogan, ‘ “Ripping Open Pregnant Women”
in the Light of an Assyrian Analogue’, Journal of the American Oriental Society 103 (1983), 755–8.
4 The closest biblical parallels are Hosea 14.1 on Samaria and 2 Kings 8.11–13 on Israel; elsewhere the
foetus-killing motif appears on its own, but is presumably to be understood as the culmination of the
whole genocidal programme, not as an isolated atrocity (contra Paul Kern, Ancient Siege Warfare
(Bloomington: Souvenir Press, 1999), 84; cf. Amélie Kuhrt, ‘Women and War’, Nin 2:1 (2001), 7–8).
5 Amos 1.13; cf. the implied criticism of the usurper Menahem’s genocidal war in 2 Kings 15.16, with
Cogan, ‘ “Ripping Open Pregnant Women” ’, and Kern, Ancient Siege Warfare, 7–8.
6 See e.g. Cogan, ‘ “Ripping Open Pregnant Women” ’, 756, for Tiglat-Pileser; 2 Kings 13.1–25;
17.5–21, for the fall of Samaria/Israel.
242 hans van wees
Israelites in their occupation of the Promised Land. The law demands that any city
of non-believers within this God-given territory should be struck with a ‘ban’
(herem), which entails that:
You shall save alive nothing that breathes, but you shall utterly destroy them, the Hittites
and the Amorites, the Canaanites and the Perizzites, the Hivites and the Jebusites, as the
Lord your God has commanded.7
The meaning of ‘everything that breathes’ fluctuated. When the army killed all the
men of the Midianites but took the women, children, and animals as booty, Moses
angrily told the soldiers to execute all the boys and adult women as well, sparing
only virgin girls and livestock. Usually, the ban did not spare virgin girls either, and
in some instances, such as Joshua’s sack of Jericho or Saul’s destruction of the
Amalekites, the ban required even cattle, sheep, donkeys, and camels to be slaugh-
tered. The law did, however, categorically exempt trees.8 The stories of Moses and
Saul both spelled out the message that the killing of livestock was an unpopular
measure but a religious necessity to be enforced regardless.9
The genocidal campaigns claimed for the early Israelites, however, were largely
fictional: the intrinsic improbability and internal inconsistencies of the account in
Joshua and its incompatibility with the stories of Judges leave little doubt about
this. Much of the biblical ideology of the ban was in fact formulated later, in the
seventh century bc,10 yet it was neither unique nor entirely a later literary inven-
tion. In an inscription of c.810 bc, Mesha, king of Moab, Israel’s old enemy, boasts
that he massacred the population of a town, 7,000 men, women, and children: ‘for
[the god] Ashtar-Kemosh I had it put to the ban.’ We need to be wary of taking this
claim at face value, since Mesha also made the quite unwarranted boast that, as a
result of his campaigns, ‘Israel utterly perished forever.’11 But the similarity be-
tween his words and those found in biblical accounts does show that even if the
events were not historical, the ideology of legitimate genocide which underlies
these stories did exist, and may occasionally have been put into practice.
At the other extreme of the rhetorical range, we have one-word notices that so-
and-so ‘sacked’ or ‘destroyed’ a city, without any elaboration on what happened to
its people, or bald statements that a defeated people were ‘sold into slavery’.
7 Deuteronomy 20.16–17; cf. 13.12–16.

8 Numbers 31.14–18 (Midianites); Joshua 6.20–21 (Jericho); 1 Samuel 15.3 (Amalekites). Livestock
spared: Joshua 8.2, 27; 11.14. Trees exempt: Deuteronomy 20.19–20.
9 Detailed analysis in André Lemaire, ‘Le hérem dans le monde ouest-sémitique’, in Laila Nehmé
(ed.), Guerre et conquête dans le Proche-Orient ancien (Paris: J. Maisonneuve, 1999), 79–92; Susan
Niditch, War in the Hebrew Bible (Oxford: Oxford University Press, 1993).
10 For introductions to these source problems, see n. 9 above, and Amélie Kuhrt, The Ancient Near
East, vol ii (London: Routledge, 1995), 417–37; Robert Carroll, ‘War in the Hebrew Bible’, in John Rich
and Graham Shipley (eds), War and Society in the Greek World (London: Routledge, 1993), 25–44.
11 ‘The Moabite Stone’: Mark Chavalas, The Ancient Near East: Historical Sources in Translation
(Malden/Oxford: Blackwell, 2006), no. 126; cf. Niditch, War in the Hebrew Bible, 31.
Sometimes such elliptic statements simply reflect the brevity of our sources, but
even otherwise detailed accounts often say little more. The surviving Greek and
Roman historians mostly belong to a school of thought which did not like to
elaborate on the suffering caused by war, partly as a matter of general historio-
graphical principle, but also because they often felt that a destroyed or enslaved city
deserved its fate and did not deserve the sympathy which a description of its misery
would evoke.12 Such attitudes are almost the opposite of the celebrations of force in
some Near Eastern texts, and leave us with a record that may understate rather than
exaggerate the frequency and extent of destruction. But this record is in its own way
a no less telling expression of the common ancient view that in certain circum-
stances the annihilation of a community could be a legitimate, desirable, and even
necessary course of action.
T H E C I T I E S O F M E N : TA RG E T S
F O R A N N I H I L AT I O N
................................................................................................................
The victim of a campaign of destruction was typically the population of a single
city, and the question arises whether this counts as genocide. We should certainly
exclude from this category the countless sieges which caused huge damage and loss
of life but did not end with the elimination of the besieged city’s entire population.
Also to be excluded are massacres committed by soldiers running amok and wiping
out entire towns, without being ordered to do so, or indeed in defiance of orders to
stop killing. But where the population of a city was executed or permanently
dispersed by the design of military or political authorities, the label ‘genocide’
surely is appropriate even if a city does not easily fit into the categories of ‘national,
ethnical, racial or religious group’ defined by the United Nations as potential
victims of genocide. In the ancient world, towns and cities—most ranging in size
from a few thousand to several tens of thousands of inhabitants and the largest few
reaching six-figure populations—were by far the most significant social, economic,
and political entities. City-states predominated in large parts of the Mediterranean
and the Near East; many ancient empires took the form of hierarchical alliance
systems in which one city-state set itself above others. Where city-states prevailed,
the inhabitants of a town constituted in effect a ‘national’ group, the destruction of
which can reasonably be described as genocide.
12 Both factors play a part in Polybius’ classic discussion of the sack of Mantinea in 223 bc (History
2.56.7–12; cf. Victor d’Huys, ‘How to Describe Violence in Historical Narrative’, Ancient Society 18
(1987), 209–50).
244 hans van wees
City-states were often part of wider ‘ethnic’ groups—unlike in the modern world
where ethnic groups are usually a subordinate part of nation-states. The citizens
of Sparta were not only Spartiates, but also shared with the other inhabitants of
the region the name ‘Lacedaemonians’; they claimed a shared origin with many
other cities as ‘Dorians’; and at the highest level of ethnic identity they counted
themselves as ‘Hellenes’, Greeks. Rome shared with neighboring city-states an
identity as ‘Latins’; the city-states of central Italy which we know as Etruscan called
themselves collectively ‘Rasenna’; and so forth. These identities emerged and faded
over time, but were meaningful to those who claimed to share them, and generally
entailed a sense that they had a language, customs, and cults in common.13 The
destruction of an ethnic group above the level of the city-state was seldom if ever
sought in the ancient world, though the alleged wars of extermination between
Israelites, Moabites, and other groups of ethnically affiliated cities in the Levant,
noted above, may be an exception.
Where city-states were not the predominant form of political organization, broader
ethnic identities were of primary importance, as they were, for example, among the
Persians, Thracians, and Macedonians, with their ‘tribal’ subdivisions, and in Egypt.
One might say that destroying a single settlement could not count as ‘genocide’ in a
region where the main group affiliations were to a larger ethnic unit, but I would argue
that towns, and even villages, everywhere had sufficiently distinctive identities for
their annihilation to constitute a form of genocide. The extinction of larger ethnic
groups as such was probably never envisaged, although the armies of Alexander the
Great in India, and the legions of Caesar in Gaul, massacred on such a large scale that
they must have come close to wiping out entire tribes.14
The massacre of all inhabitants of a city was quite rare, though not unknown.
More common was the killing of all men, or all men of military age, while the
women and children were led away to be sold as slaves. This is described as
standard procedure for dealing with a city captured by force in a range of texts
from Deuteronomy and the Iliad to Livy’s history of Rome.15 Some scholars have
13 Ethnicity: Jonathan Hall, Ethnicity in Greek Antiquity (Cambridge: Cambridge University Press,
1997); idem, Hellenicity: Between Ethnicity and Culture (Chicago: University of Chicago Press, 2002);
Irad Malkin (ed.), Ancient Perceptions of Greek Ethnicity (Cambridge, MA.: Harvard University Press,
2001); and relevant chapters in Tim Cornell and Kathryn Lomas (eds), Gender and Ethnicity in Roman
Italy (London: Accordia, 1997).
14 See A. B. Bosworth, Alexander and the East (Oxford: Oxford University Press, 1996), 133–65.
Appian, Gallic History 2, claimed that Caesar killed 1 million and enslaved 1 million Gauls, out of a
total population of 4 million; Caesar himself reports, e.g., the sale into slavery of 53,000 Atuatuci
(Gallic War 2.33.6–7).
15 Deuteronomy 20.12–15 (cities outside the Promised Land); Iliad e.g. 6.448–65; 9.591–5; 22.62–8;
Livy 21.13.9. See W. Kendrick Pritchett, The Greek State at War, vol v (Berkeley: University of California
Press, 1991); Pierre Ducrey, Le traitement de prisonniers de guerre dans la Grèce antique (Paris: Boccard,
1968; new edn 1999); Mars M. Westington, ‘Atrocities in Roman Warfare to 133 bc’, PhD dissertation,
University of Chicago, 1938, 68.
imagined that men were killed because they were ‘more difficult to control than
women and children’, so that enslavement was not feasible.16 This is hardly a
tenable view, given that countless men were in fact reduced to chattel slavery and
other forms of forced labour throughout antiquity and beyond. Conversely, the
youngest children and the elderly were spared despite being no use as slaves: they
would often find no buyer and be left to die of exposure, hunger, or attack by wild
animals. On one occasion, a Roman army actually butchered the young and the old
but made slaves of the adult men.17 Instead, if a distinction by gender and age was
made, it was because communities were conceived of as consisting essentially of
adult men. Women, children, and the aged were mere dependants. The death of the
men amounted to the annihilation of the community; killing the others was not
necessary.
We have good contemporary evidence for instances of genocide which took this
sharply gendered form, as we shall see, but the impression created by literary
sources that it was standard practice is probably misleading, since quite a range
of different treatments of sacked cities is attested. A common Roman practice, for
instance, appears to have been to engage initially in a period of indiscriminate
slaughter—‘one may often see not only the corpses of human beings, but dogs cut
in half and the dismembered limbs of other animals’—until the commander gave
the signal to stop; then survivors of both sexes and all ages were systematically
rounded up to be sold, which in one notorious campaign produced a total of
150,000 slaves from 72 settlements in Epirus. Similar mass enslavements occurred
with some frequency in the Greek world.18 Not even nuclear families would be kept
together when they were sold into slavery, so the result was complete dispersal of a
community. The irrevocable destruction of a group in this way could be counted as
a form of genocide, even if few were killed.
Rape did not play the ancillary part in genocidal campaigns which it has played
in more recent history. A few brief and rather euphemistic references leave no
doubt that many women, and indeed young men, were raped during the sack of
cities, but our sources tend to treat this as an incidental and distasteful aspect of
siege warfare. Unlike killing and enslavement it was not a matter of public policy.
A concerted campaign of rape may just be hinted at in the Iliad when the Greeks are
16 Kern, Ancient Siege Warfare, 23, citing I. J. Gelb, ‘Prisoners of War in Early Mesopotamia’, Journal
of Near Eastern Studies 32 (1972), 72.
17 Left to die: Xenophon, Agesilaos 1.21–2; cf. Tacitus, Histories 3.33.1. Killed: Livy, Ab urbe condita
42.63.10–11.
18 Roman practice: Polybius, History 10.15.4–16.9, who speaks only of ‘booty’, but this clearly
included people (see Adam Ziolkowski, ‘Urbs Direpta, or How the Romans Sacked Cities’, in Rich
and Shipley (eds), War and Society in the Greek World, 69–91); Epirus: ibid. 30.16; Livy, Ab urbe condita
45.34.5; Pliny, Natural History 4.39. For mass enslavement in general, see Hans Volkmann, Die
Massenversklavungen der Einwohner eroberter Städte in der hellenistisch-römischer Zeit, 2nd edn
(Stuttgart: Franz Steiner, 1990).
246 hans van wees
told: ‘Let no one be in a hurry to go home until he has slept with the wife of a
Trojan, in revenge for the shocks and sobs of Helen’ (2.354–6). But this could be
explained instead as a reference, not to rape as it was understood then, but to
soldiers having sex with Trojan women who were allocated to them as slaves—a
form of sexual coercion which was regarded as legitimate, and which allowed men
to establish a form of control over women which was far more lasting and
comprehensive than a brief and limited exercise of power through ad hoc sexual
violence. Near Eastern sources say even less on the subject than Greek and Roman
sources do.19
The physical site of the city was another target for destruction. Towns are forever
being ‘burnt down’ and ‘razed to the ground’ in ancient records, and in many cases
this may be no more than a formulaic reference to the damage done in the course of
the siege and haphazard vandalism by pillaging soldiers. A common theme in Near
Eastern texts is the turning of a defeated city’s territory into an uninhabitable
wasteland by ‘sowing it with salt’, planting it with weeds, or covering it with
brimstone and pitch,20 but this could surely have been done only symbolically,
on a small scale, if at all. Deliberate flooding as a punishment is also attested,
though not very reliably. In the Greek and Roman world cultivation of the territory
of a destroyed city might be forbidden, sometimes by dedicating it to a god, so that
the land ‘reverted to a sheep-walk’.21
The purpose of the physical destruction of buildings and land is made clear in an
inscription which claims that Sennacherib of Assyria flooded Babylon in 689 bc, ‘so
that it might be impossible in future days to recognize the site of that city and its
temples.’ The destruction of a Spanish town, Ilurgia, in 206 bc, is described in
similar terms: the Romans wanted to ‘erase the memory’ of their enemy. Converse-
ly, to leave buildings standing was regarded as leaving a ‘tomb and monument’ to
the former inhabitants, and the Romans are said to have left the walls and houses of
another Spanish town, Astapa, intact as a mark of respect for a community that
19 For the Greek and Roman evidence, see Elisabeth Vikman, ‘Sexual Violence in Warfare, Part I:
Ancient Origins’, Anthropology and Medicine 12:1 (2005), 21–31; Kern, Ancient Siege Warfare, 154–62,
235–6, 345–7; ancient attitudes to rape generally: Susan Deacy and Karen Pierce (eds), Rape in
Antiquity (London/Swansea: Classical Press of Wales, 1997). Omission of women from classical
historiography of war: Simon Hornblower, ‘Warfare in Ancient Literature: The Paradox of War’, in
Philip Sabin et al. (eds), The Cambridge History of Greek and Roman Warfare, vol i (Cambridge:
Cambridge University Press, 2007), 42–7. Near-absence from Near Eastern evidence: Kern, Ancient
Siege Warfare, 80–3.
20 Sowing with salt: e.g. Judges 9.45; other evidence collected in Moshe Weinfeld, Deuteronomy and
the Deuteronomic School (Winona Lake: Eisenbrauns, 1992), 110–16; R. T. Ridley, ‘To Be Taken with a
Pinch of Salt: The Destruction of Carthage’, Classical Philology 81 (1986), 140–6.
21 Flooding: see below. ‘Sheep-walk’: Isocrates, Orations 14.31 (Kirrha, c.590 bc); Diodorus of Sicily,
The Library 15.63.1; Strabo, Geography 7.5 (315); Dionysius of Halicarnassus, Roman Antiquities 2.16.1.
The story of the ‘cursing’ of Carthage’s territory is discussed by Ridley, ‘To Be Taken with a Pinch
of Salt’.
had committed mass suicide rather than surrender.22 The idea that the victims
should vanish without a trace was extended in Greek literature even to the non-
burial of the dead: hence Agamemnon’s demand in the Iliad that the Trojans’
corpses should be left ‘untended and invisible’, that is, without tombs to keep their
memory alive. Extermination of the people was not enough: ideally, even the
evidence of their existence had to be obliterated.
A LT E R NAT I V E S TO A N N I H I L AT I O N :
G E N O C I D E I N C O N T E XT
................................................................................................................
The prominence of warfare in the historical record of the ancient world, the brutal
rhetoric of many of the sources, and the suggestion found in authors such as Thucy-
dides and Plato that states were engaged in a permanent war of all against all conspire
to create the impression that ancient cities and states were always out to destroy one
another. They were not, of course. International relations in the ancient world were
complex, with a wide range of recognized relationships between states, including
kinship and friendship, as well as formal treaty obligations, and sophisticated diplo-
matic mechanisms, from inviolable envoys and ambassadors, via exchanges of letters
and gifts, to international arbitration. Despite some modern claims to the contrary,
peace was always considered the norm and ideal, even if it was often interrupted by the
necessary evil of war.23 When war did break out, its goals usually stopped well short of
annihilating the enemy, and we need to consider the full range of options which an
ancient state had in dealing with a hostile city before we can try to determine why
sometimes a city-state, kingdom, or empire chose to resort to genocide.
From as early as the third millennium bc, wars were often concluded with pacts
of non-aggression or treaties of equal alliance, sometimes on terms which
22 Babylon: J. A. Brinkman, ‘Sennacherib’s Babylonian problem: an interpretation’, Journal of

Cuneiform Studies 25:2 (1973), 89–95 (translation, 94); Kuhrt, The Ancient Near East, 582–7. Ilurgia:
Livy, Ab urbe condita 28.20.5–7. ‘Monument and tomb’: Livy 31.29.11; cf. 26.16.11–12. Astapa (206 bc):
Appian, Punic Wars 6.33 (a different account in Livy 28.22–3).
23 For a brief survey of ancient international relations, see Hans van Wees, ‘Peace and the Society of
States in Antiquity’, in Jost Dülffer and Robert Frank (eds), War, Peace and Gender (Essen: Klartext,
2009); David Bedermann, International Law in Antiquity (Cambridge: Cambridge University Press,
2001). On Greece and Rome, see also the relevant chapters in Anja Hartmann and Beatrice Heuser
(eds), War, Peace and World Orders in European History (London: Routledge, 2001), and in both
volumes of Sabin et al. (eds), The Cambridge History of Greek and Roman Warfare. On the Near East,
see especially Mario Liverani, International Relations in the Ancient Near East, 1600–1100 bc
(Basingstoke: Palgrave, 2001); on Assyria, see Bustenay Oded, War, Peace, and Empire: Justifications
for War in Assyrian Royal Inscriptions (Wiesbaden: Ludwig Reichert Verlag, 1992).
248 hans van wees
required close cooperation. Alternatively, war might end with an unequal treaty,
imposing military obligations on the defeated side; the Romans in addition
confiscated some of the land of their new allies.24 The most powerful states
went beyond alliance-making to impose outright subjection. Egyptian, Assyrian,
and Persian kings demanded tribute as a token of submission; regular taxes and
other duties replaced, or were added to, such obligations as conquests were
consolidated. The Romans similarly created subject ‘provinces’ rather than allies
when they expanded beyond Italy.25 In all such cases, communities were left
wholly intact, apart of course from any casualties suffered in combat, and often
retained a high degree of local autonomy as well, losing little more than their
international standing.
More violent measures begin with the elimination of a city’s leadership, which
was particularly common in dealing with ‘rebels’, i.e. cities which in some way
offended against the terms of their subordination, or sometimes simply refused to
be subordinated. Tension between a ruling class and the rest of the population was
a constant feature of political and social life in Greek cities, and in order to
strengthen or regain control over their subject allies, classical Athenians and
Spartans exploited such divisions by executing or exiling hostile or rebellious
political elites.26 They had precursors in the Near East, such as Sennacherib’s
treatment of rebels in Ekron and Jerusalem, a few years before his demolition of
Babylon. In Ekron, the elite were theatrically executed, some commoners enslaved,
and the rest spared. Biblical accounts of the siege of Jerusalem emphasize that
Sennacherib’s representatives refused to negotiate with the elite alone, and insisted
on announcing the terms of surrender in Hebrew so that the common people
would understand—evidently an attempt to exploit internal divisions. Exiling or
executing the ruling elite alone later became the Romans’ standard procedure in
dealing with rebellious cities.27
24 See previous note, and in addition: Polly Low, Interstate Relations in Classical Greece: Morality
and Power (Cambridge: Cambridge University Press, 2007); Hans van Wees, Greek Warfare: Myths and
Realities (London: Duckworth, 2004), 12–15; Stephen Oakley, ‘The Roman Conquest of Italy’, in Rich
and Shipley (eds), War and Society in the Greek World, 9–37; and a comparative study by Barry Strauss,
‘The Art of Alliance and the Peloponnesian War’, in Charles D. Hamilton and Peter Krentz (eds), Polis
and Polemos (Claremont: Regina Books, 1997), 127–40.
25 Subjection: e.g. Oded, War, Peace and Empire; tribute: e.g. Karen Radner, ‘Abgaben an den König
von Assyrien aus dem In- und Ausland’, in H. Klinkott et al. (eds), Geschenke und Steuern, Zölle und
Tribute (Leiden/Boston: Brill, 2007), 213–30. Roman provinces: e.g. Andrew Lintott, Imperium
Romanum (London/New York: Routledge 1993), 16–42.
26 For the classical period, see Hans-Joachim Gehrke, Stasis (Munich: Beck, 1985); Andrew Lintott,
Violence, Civil Strife and Revolution in the Classical Greek City (London: Routledge, 1982); for archaic
Greece, see Hans van Wees, ‘Stasis, Destroyer of Men’, in Cedric Brélaz and Pierre Ducrey (eds), Sécurité
collective et ordre public dans les sociétés anciennes (Vandoeuvres: Fondation Hardt, 2008), 1–48.
27 Ekron: Daniel Luckenbill, Ancient Records of Assyria and Babylonia, vol ii (Chicago: University of
Chicago Press, 1927; reprinted 1989), nos. 211–12. Jerusalem: 2 Kings 18.26–8; 2 Chronicles 32.18; Isaiah
36.11–13. Rome: e.g. Westington, ‘Atrocities in Roman Warfare to 133 bc’, 86–107.
In a famous passage, Thucydides has the Athenian assembly debate the pros and
cons of killing only the responsible members of the ruling class versus massacring the
entire population of Mytilene, an allied city which had revolted. The argument in
favour of the former is essentially that it will in the long run be more profitable not to
destroy the whole community, and the view that material advantages, rather than
moral qualms, militated most strongly against genocide, may have been widely shared.
The Athenians decided in the end to kill ‘only’ a thousand leading Mytileneans.28
Elsewhere, elites were driven into exile or deported. The Athenians in 506 bc
expelled ‘the Horsemen’ who ruled Chalcis, and in 338 bc the Romans forced the
senators of Velitrae to go and live ‘on the other side of the Tiber’. Deportation and
resettlement of local elites was common practice in the empires of the Near East:
the ‘Babylonian exile’ of the priests and ruling class of Judah from 586 to 537 bc is
only the best known instance of many. Settlers from other parts of the empire
usually replaced the deportees.29
Not just elites, but whole communities might be driven from their homes. The victors
sometimes let a defeated enemy vacate their city unmolested when they surrendered,
typically allowing them to bring literally only the clothes on their backs.30 Large-scale
deportations are well attested in the Near East, where Assyrian kings in particular moved
tens of thousands of people from one end of the empire to the other. Communities were
usually not deported in their entirety: some families were selected for deportation while
others stayed behind. This relocation did much to weaken local communities, of course,
but they were not destroyed and the deportees were kept together as a group as much as
possible. They were ‘counted as Assyrians’ in their new homes rather than reduced to
slavery. The Persians, judging by Greek accounts, did deport entire communities—
where possible, as on islands, soldiers marched in a linked chain across the territory to
ensure that not a single person avoided capture—but again kept them intact. The people
of Eretria, deported a few days before the battle of Marathon and put to work scooping
petroleum from a well, still formed a distinct, Greek-speaking community deep in
Persian imperial territory two generations later.31
Genocide, in short, was far from the only or normal outcome of hostile inter-
group relations in the ancient world. Quantifying its relative frequency compared
with other ways of dealing with enemy cities is not possible, but we do have enough
information to show that mass executions were far rarer than mass enslavement. In
28 Thucydides, History 3.36–50, with Simon Hornblower, A Commentary on Thucydides, vol. i,

books I-III (Oxford: Oxford University Press, 1991), 420–41.
29 Chalcis: Herodotus, Histories 5.77.2; Velitrae: Livy, Ab urbe condita 8.14. Assyria: Bustenay Oded,
Mass Deportations and Deportees in the Neo-Assyrian Empire (Wiesbaden: Reichert, 1979), 43–5.
30 See van Wees, Greek Warfare: Myths and Realities, 261 n 44.
31 Assyrians: analysis in Oded, Mass Deportations and Deportees in the Neo-Assyrian Empire, and
J. Nicholas Postgate, ‘Ancient Assyria: a Multi-Racial State’, Aram 1 (1989), 1–20 (reprinted in Postgate,
The Land of Assur and the Yoke of Assur (Oxford: Oxbow, 2007). Persians: Herodotus, Histories 3.149;
6.31, 119.
250 hans van wees
his first campaign, Sennacherib claimed to have sacked eighty-nine fortified towns
and 820 villages, yet he boasted about the extermination of the population of only
one town, Hirimme.32 The pattern for Roman wars of expansion in the third and
second centuries bc is similar. In the Third Samnite War (298–290 bc), at least
eleven cities and 40,000 people were enslaved by the Romans, but not a single
massacre is recorded. The First Punic War (264–241 bc) saw the enslavement of five
cities in Sicily, and of 27,000 North Africans from an alleged 200 or 300 villages, but
only one mass killing (at Lipara, 252) and one near-massacre at Myttistratos (257).
After surviving Hannibal’s invasion of Italy, the Romans enslaved many settlements
during their counter-offensive in North Africa, but perpetrated only one mass
killing (at Locha, 203). The statistics for their wars of conquest in Spain, Greece,
and the Near East are comparable. Numerous other campaigns enslaved people in
their tens of thousands but at no point attempted to exterminate the inhabitants of
any settlement.33
The relative rarity of mass murder suggests that it was not just a by-product of
randomly brutal warfare but an option deliberately chosen in particular circum-
stances. This brings us back to the question of what motivated genocidal actions by
ancient states and armies.
C O N S P I C U O U S D E S T RU C T I O N : P RO F I T,
P OW E R , A N D G E N O C I D E
................................................................................................................
On 19 December 51 bc, Cicero, then governor of Cilicia, wrote a letter to his close
friend Atticus in which he gave a jocular account of enslaving a small town in his
remote province:
At the time of the Saturnalia festival, the Pindenissetans surrendered themselves into my
power, on the fifty-seventh day after we began our siege of them. ‘What the devil? The
Pindenissetans? Who are they?,’ you may say. ‘I have never even heard the name.’ Well, what
am I supposed to do? Can I turn Cilicia into Aetolia or Macedonia? . . . We went to
Pindenissus, a well-defended settlement of the Free Cilicians which has been under arms
32 Luckenbill, Ancient Records of Assyria and Babylonia, nos. 234 (with different numbers), 261, 267,
272, 274; Hirimme: ibid. nos. 235, 266, 276, 303. Cf. the analysis of Assyrian records in Kern, Ancient
Siege Warfare, 69–75.
33 Spain: at least eight fortified towns and numerous villages enslaved; a single massacre at Ilurgia,
206; Greece and the Near East: nine cities enslaved, all men massacred at Antipatreia, 200. The above is
based on evidence collated by Volkmann, Die Massenversklavungen der Einwohner eroberter Städte in
der hellenistisch-römischer Zeit, and Ziolkowski, ‘Urbs Direpta’, 83–4. Note that the ancient record is far
from complete.
as long as anyone can remember. The people are savage and fierce and in every way
prepared to defend themselves. [With great effort but without loss of life,] we finished
the business. A jolly Saturnalia—also for the soldiers, to whom I handed over the booty
except the captives. The slaves were sold on the third day of the Saturnalia. At the time of
writing, the proceeds on my dais stand at about 120,000 sesterces.34
From the mock-solemn opening sentence to the modest sum recorded at the end,
Cicero self-deprecatingly stressed the trivial nature of the event (which left no trace
in the historical record outside his letters). He felt no need to explain his reasons
for selling the population into slavery or for attacking the town, except to hint at
their refusal to submit to Roman authority.
Cicero was aware that his victory was insignificant, but he wanted public
recognition all the same, and even set his sights on a triumphal procession, the
highest military honour. He therefore recorded details of his campaign in a second
letter, which he sent to Cato, one of the most powerful men in Rome, asking him to
propose public honours on his behalf:
I led the army to Pindenissus . . . Because this place was located in a very elevated and well-
defended spot, and was inhabited by people who had never submitted to any king, and
because they had also taken in refugees [from an earlier battle] and were eagerly awaiting
the arrival of the Parthians, I came to the decision that it was a matter of upholding the
reputation (existimatio) of the empire to crush their impertinence, so that the spirit of
others who are hostile to our empire might be more easily broken.
He added that ‘they came into my power by coercion when every part of their city
had been either demolished or set on fire.’ That enslavement was their fate is not
even mentioned.35 Here, Cicero offers three reasons for sacking the city: to punish
the people of Pindenissus for taking sides with Rome’s enemies, the Parthians, who
were rumoured to be planning an invasion; to punish them for more generally
showing disrespect towards the Roman empire; and to deter others from acting
with similar disloyalty and disrespect. The very difficulty of capturing the place, in
its mountainous location, becomes a reason for its destruction: it serves as dem-
onstration of Rome’s power to coerce any community, however inaccessible and
defiant. These letters are worth quoting at length because it is rare to have a
perpetrator’s private account of an act of genocide, and because Cicero alludes to
many of the reasons for mass enslavement and mass killing which we commonly
encounter in ancient texts.
The material profit to be made from selling people into slavery, in Cicero’s case
evidently at best a secondary reason, seems the primary motivation of several
earlier Roman campaigns. Operations in Sardinia and Istria in 177 flooded the
market with up to 85,000 slaves and gave rise to the expression ‘Sardinians for sale!’
34 Cicero, Letters to Atticus 5.20.1 and 4.

35 Cicero, Letters to his Friends 15.4.10.
252 hans van wees
to describe an abundance of cheap goods, and neither in these campaigns nor in

the mass enslavement of 150,000 Molossians from Epirus only ten years later do we
know of any particular justification for such severity. It is significant that the
victims inhabited regions where levels of urbanization were low, which made the
population ‘savages’, in Roman eyes, as well as vulnerable to attack. The Romans
seem to have regarded such people as legitimate targets for slave-raiding on a
massive scale. The Greeks similarly developed a notion that all non-Greeks were
‘barbarians’ whose enslavement was a kind of hunting, and as such in no need of
special justification.36 In some cases, then, the annihilation of communities was not
a goal in itself, but merely an incidental consequence of a ruthless pursuit of profit.
The question of intent, however, is academic: slave-raiding states could not fail to
realize that the ultimate result of their actions would be genocide.
One suspects that profit was always a significant motive for mass enslavement,
even when quite different formal reasons were adduced. At the very least, when a
state considered how to retaliate against an enemy or punish a subject city, the
prospect of material gain must often have swung the balance in favour of enslave-
ment rather than massacre—which is why the former was so much more common
than the latter. By the same token, however, the decision to massacre was inhibited
by economic, as well as moral and political, concerns. The choice was not between
killing an enemy and letting him or her go, but between selling captives at a profit
or executing them and going empty-handed. When Menelaus hesitated to kill a
Trojan, this was not due to any humanitarian feelings but because it would cost
him the ‘infinite ransom’ which a living prisoner might bring. The Israelites needed
reminding repeatedly that putting an enemy to the ban meant forgoing all booty of
slaves, livestock, and valuables. Conversely, we are told, the only way to stop
furious Roman soldiers massacring the people of Myttistratus was to promise
that they could keep as personal booty anyone whom they caught and did not kill.37
If the profit motive encouraged enslavement but discouraged killing, only when
it was countered by even more powerful motivations did states resort to genocidal
massacres. An endless variety of such motivations is mentioned in ancient sources,
but the majority have at least one feature in common: those perpetrating the
massacre saw themselves as inflicting revenge or punishment for what one may
call an ‘aggravated’ challenge to their power and status as a community and/or to
the power of a god whose cause they champion. A few typical scenarios will have
to suffice to illustrate this point.
36 For Sardinia and Istria, see Volkmann, Die Massenversklavungen der Einwohner eroberter Städte
in der hellenistisch-römischer Zeit, 27, 54–5. For Epirus, see n. 18, above. For ancient attitudes to
enslavement, see Peter Garnsey, Ideas of Slavery from Aristotle to Augustine (Cambridge: Cambridge
37 Menelaus: Homer, Iliad 6.43–53. Israelites: above, at n. 9–10. Myttistratus: Zonaras, Historical
Epitomes 8.11.10; cf. Polybius, History 1.24.11; Diodorus of Sicily, The Library 23.9.4.
First, imperial powers sometimes destroyed small independent towns on the

grounds that their very independence constituted an affront which made the
empire look weak. This is the main reason adduced by Cicero for his enslavement
of Pindenissus, and elsewhere the same principle is cited to justify a massacre. In
Thucydides’ ‘Melian Dialogue’, the Athenians explain bluntly why they must
incorporate the small, neutral island of Melos into their naval empire: ‘Since you
are islanders, and weaker than the others, you cannot be allowed to escape from
those who rule the sea.’ When the Melians refused to submit, the Athenians killed
the men, and enslaved the women and children.38 The same reasoning probably lay
behind Sennacherib’s destruction of Hirimme, another small town which had so
far refused to submit to any king.39 Resistance to imperial power is evidently a
worse offence for a small town than for a large city. Refusal to submit implies that
the opponent regards himself as somehow the ‘equal’ of the empire, which is more
insulting when it comes from an insignificant little town than from a serious rival.
The destruction of defiant small towns was therefore not just a matter of con-
solidating imperial power, but of upholding status. The demands of power politics
might have been satisfied by merely forcing Pindenissus, Melos, Hirimme, and
their like to submit to imperial authority, but the demands of status required that
they be destroyed to wipe out the insult.
A second scenario involves the destruction of a roughly equal opponent who is
regarded as in some way too persistent in his hostility. For instance, while relations
between Athens and most of its rivals alternated between hostility and alliance, the
Athenians saw the neighbouring Aeginetans as implacable enemies, who had
started hostilities in the dim past and kept attacking without provocation. In 431
bc, the Athenians drove the Aeginetans out of their island, forcing the refugees to
find new homes all over Greece; a large group settled in Thyrea. Not content with
this result, in 424 bc the Athenians sent a fleet to attack Thyrea, which they
captured, looted, and burned down. All Aeginetans captured alive were taken to
Athens, where a formal decision was made to execute every last one ‘on account of
the hostility which they had always shown in the past.’40 In much the same way, the
Romans came to regard the Carthaginians as an exceptionally bitter and relentless
enemy, which may explain their eventual destruction of Carthage and enslavement
of its people in 146 bc. Power-political motivations clearly played a prominent role,
but it is striking that the Aeginetans and Carthaginians were destroyed at a time
when they no longer posed a serious threat to their old rivals, however dangerous
38 Thucydides, History 5.97, 101, 116.4 (cf. 4.122.5–6), with Simon Hornblower, A Commentary on
Thucydides, vol iii, books V.25–VIII.109 (Oxford: Oxford University Press, 2008), 216–25.
39 Their defiance is noted: Luckenbill, Ancient Records of Assyria and Babylonia, no. 266; cf. 276.
40 Thucydides, History 2.27; 4.57; for the Athenian tradition of an age-old feud with Aegina, see
Herodotus, Histories 5.81–9, written around the time of the events described above.
254 hans van wees
they might once have been. Their annihilation was strongly symbolic: a demon-
stration by Athens and Rome of how great a superiority they now enjoyed.41
Also common was a third scenario: the destruction of a formerly friendly or allied
city which was deemed to have committed a particularly heinous act of treachery,
which seriously endangered the city betrayed. This situation features in another of
Thucydides’ set-pieces, the ‘Mytilenean Debate’, concerning a proposal to execute all
the men and enslave the women and children of Mytilene. This city had changed sides
in the middle of major war, raising the spectre of a general defection to Sparta of all
Athens’ allies; its destruction, it was argued, would act as a deterrent to other would-be
rebels. For similar reasons, the Romans massacred or enslaved more Italian cities than
at any other time during the wars in which they came closest to defeat. In the Second
Samnite War (326–304 bc), they killed the inhabitants of four towns, killed all the men
in a fifth, and enslaved the population of a further two; during Hannibal’s invasion of
Italy, they enslaved at least 14 Italian and Sicilian towns, and massacred the people of
two more. In each case, the victims had either changed sides or were suspected of being
about to do so.42 Here we have genocide carried out for calculated political effect, as a
means of deterrence and intimidation. It was used in much the same way in campaigns
of conquest, as in Cicero’s mini-war against Pindenissus: harsh treatment of the first
town to offer any resistance served to frighten others into submission.
Yet political pragmatism was not always the whole story even in these cases. A major
argument used in favour of the massacre of all men of Mytilene, according to Thucy-
dides, was that this ally’s betrayal did not just create a serious threat to security but
entailed a serious breach of trust, because relations between Athens and Mytilene had
been unusually close and privileged. ‘They were not ruled by Athens, like the others,’ but
retained autonomy ‘and were treated with the highest respect by us.’ Their betrayal was
all the more culpable because there was evidence that they had been planning to defect
for a long time. ‘We should never have treated them with more respect than anyone else,’
he adds, ‘then they would not have become so arrogant; for it is human nature to
despise those who show deference and to admire those who concede nothing.’43 Once
again, emotive matters of status creep into the picture alongside political calculation:
a failure to reciprocate respect and friendship calls for violent retaliation.
A final scenario is the annihilation of a community as punishment for a religious
offence. This principle is widely attested, but seems to have been more often an
additional justification for genocide than a major motivation in its own right. In
Greece, the so-called Amphictyons, a league of states in charge of the oracle of Apollo
41 The Athenians did not occupy the Aeginetans’ land in Thyrea, and the Romans left the territory
of the Carthaginians untouched for more than a decade, so their destruction brought no material
benefit.
42 Mytilene: Thucydides, History 3.37.2, 39.7–8, 40.4–7. Rome: the evidence is again collated by
Volkmann and Ziolkowski, as cited in fn 33, above; the latter, however, questions the evidence for the
Second Samnite War.
43 Special status of Mytilene: Thucydides, History 3.36.2, 39.2 and 5.
at Delphi, swore an oath that they would ‘uproot’ any city which committed offences
against the sanctuary, and tradition claimed that in 591 bc they destroyed the town of
Kirrha for acts of aggression against the oracle.44 Even earlier, the inscription which
recounts Sennacherib’s destruction and flooding of Babylon in 689 also states that the
victorious king retrieved cult images which had been taken from Assyria by the
Babylonians 418 years earlier. The implication is that Babylon was being punished for
this ancient act of sacrilege. But none of the other records of Sennacherib’s campaigns
against Babylon mention either this religious offence or the punishment of destruction,
although they hint at plenty of other political and personal reasons why the king might
have dealt harshly with the city.45 At the other end of antiquity, when the Persian king
Sapor II was shot at during his siege of Amida in ad 359 and an arrow damaged his
cloak, he responded ‘as if the enemy had committed sacrilege’ and set about trying to
destroy the city.46 The sarcastic tone of the Roman eyewitness who reports this shows
that he did not take this episode seriously—but it also shows how widely accepted was
the idea that genocide was suitable punishment for an act of collective sacrilege.
Cities might in theory also be destroyed for breaking divinely sanctioned rules,
rather than direct offences against gods or god-like rulers. The sack of Corinth and the
flooding of Sybaris are said to have been provoked by these cities’ violation of
respectively Romans and Crotoniat diplomatic envoys, regarded as sacrosanct.47
The Iliad implicitly justifies the annihilation of Troy by showing the Trojans accu-
mulating offences with a religious dimension: they threaten to kill Menelaus when he
visits Troy on a diplomatic mission, they break the divinely sanctioned bond of
hospitality in abducting his wife, and finally they break a truce which they had
sworn to uphold on pain of destruction by the gods.48 Since all treaties and truces
and many other aspects of international relations in the ancient world were under the
protection of the gods, it was rarely difficult to accuse an opponent of a religious
offence. Near Eastern empires, moreover, generally saw themselves as implementing
the will of their chief gods, such as Asshur for the Assyrians or Auramazda for the
Persians, so that any opposition to their power was by definition an offence against
these divinities.49 Religious offences alone, however, without other aggravating
circumstances, were not often adduced as a reason for genocide.
44 The oath and Kirrha: Aeschines Orations 2.115; 3.107–9; and n. 25, above (doubts about the
historicity of the oath seem to me unjustified: Hans van Wees, ‘The Oath of the Sworn Bands’, in
Andreas Luther et al. eds., Das frühe Sparta (Stuttgart: Steiner Verlag, 2006), 139–42).
45 Babylon’s religious offence: Luckenbill, Ancient Records of Assyria and Babylonia, no. 341; other
accounts: see Brinkman (n. 22, above).
46 Ammianus Marcellinus, Histories 19.1.5, with Josh Levithan, ‘Emperors, Sieges and Intentional
Exposure’, in Edward Bragg et al. (eds), Beyond the Battlefields (Newcastle: Cambridge Scholars
Publishing, 2008), 37–8.
47 Corinth: Livy, Epitome 52; Sybaris: Strabo, Geography 6.1.13.
48 Breaking oath: Homer, Iliad 3.67–4.222; quotation: 3.298–301; cf. 4.235–9; 7.351–3. Threats to
ambassador: 11.123–5, 139–42; cf. 3.205–24. Discussion: van Wees, Status Warriors, 176–81.
49 Assyrians: Oded, War, Peace and Empire; Persians: Kuhrt, The Ancient Near East, 676–82.
256 hans van wees
The exception, as noted above, were the wars of extermination under the ban
developed in the Levant. Here the idea that the actions of a king and his people
represented the will of their supreme god was pushed further, to the point where
any opposition to their power could be treated as a serious religious offence which
demanded the annihilation of the enemy as a sacrifice to the offended god, be he
Yahweh of Israel or Ashtar-Kemosh of Moab.
Extraordinary as the concept of the ban was, it rested on the same fundamental
premise as more common legitimations of genocide in the ancient world: a chal-
lenge to the power of a community and/or its gods must be answered with a display
of force in proportion to the seriousness of the challenge and may require an act of
‘conspicuous destruction’ which completely eliminates the challenger. What deter-
mined the seriousness of a challenge was partly its impact on security and power
politics, but often its symbolic significance and impact on status counted at least as
heavily. A challenge from a treacherous friend, an inveterate enemy, or a low-status
opponent required a harsher response than an otherwise equally dangerous chal-
lenge from an open and equal rival. The more powerful a community, the more it
might be inclined to demand respect in proportion to its status and to respond
violently to even the slightest, least dangerous, challenge. The same was true of gods,
who were imagined as responding to offences with greater force in proportion to
their vastly greater power. A god as uniquely powerful as Yahweh and a people as
closely identified with their god as the Israelites could therefore almost be expected
to exterminate anyone who stood in their way: they pushed common ancient
attitudes to their more-or-less logical and most brutal extreme.
C O N C LU S I O N S : G E N O C I D E AND ITS LIMITS

................................................................................................................
Intentional genocide was perpetrated, and legitimate, according to ancient sources,
when a community had committed a serious offence which called for the ultimate
punishment. The form and extent of the destruction wrought is often difficult to
determine because the ancient rhetoric of genocide variously over- or understated the
damage done, and the explanations for genocide offered in ancient texts equally need
to be treated with caution, since they are unlikely to tell the full story. Some might
argue that the sources’ emphasis on the symbolic, religious, and status-related nature
of offences does not tell the true story at all, and that the extermination of commu-
nities must in reality have been motivated by a search for more power, resources, and
territory. In support of that view, one might point out that the ideology of genocide
was most widely accepted, and genocidal campaigns most common, among cities and
states engaged in rapid military expansion. Macedonian armies, for instance,
destroyed many cities during their campaigns in Greece and in the Persian Empire
under Philip II and Alexander III the Great, but once these conquests had been
consolidated into three fairly stable new kingdoms, a century went by without a
single genocidal campaign being recorded.50 The Romans enslaved and massacred on
a huge scale while they extended their power across the Mediterranean, but once their
control was established, very few further acts of genocide are attested. Conversely,
scruples about genocidal warfare began to be much more forcefully articulated in
Greece only when continuing warfare no longer led to significant expansion, as a de
facto balance of power was achieved in the fifth and especially fourth century bc.51 Yet
there are enough instances of genocide which did not result in any political advantage
or material gain for the perpetrators—the Athenian massacre of the Aeginetans in
Thyrea is a striking example—to show that it was not always merely an excuse for
expansionism.
I would argue that we must accept at least the basic premise of the sources:
whatever its political or economic motivations, genocide was always also, and
sometimes mainly, an act of ‘conspicuous destruction’ which served to display the
power of the perpetrators and to restore or enhance their status. And a concern with
status might not only lead to the destruction of a city which brought its destroyers
no additional power or wealth, but also, just occasionally, save a city when purely
political and economic reasons might have favoured its destruction, as when the
Spartans after a long and bitter war spared Athens in deference to the city’s eminent
international status, or when Caesar decided not to wipe out Massilia, ‘more on
account of the reputation (nomen) and antiquity of their state than because it had
done anything to deserve this from him’.52 Genocide was most commonly perpe-
trated by the most aggressively expansionistic states not only because such states
pursued their own material interests with more ruthless calculation, but also
because they pursued prestige more competitively and with greater emotional
intensity than others. Those who aimed for the highest possible status in the
world order were least able to tolerate any challenge to their honour and most
willing to eliminate without a trace those who seemed to show insufficient respect.53
50 Admittedly, the history of this century is not particularly well attested.

51 Ethical debate: e.g. Ducrey, Le traitement de prisonniers de guerre dans la Grèce antique, 313–32;
Volkmann, Die Massenversklavungen der Einwohner eroberter Städte in der hellenistisch-römischer Zeit,
71–91.
52 Caesar, Civil War 2.22. Athens: Xenophon, Hellenica 2.2.19–20.
53 Role of honour in international relations: Greece: J. E. Lendon, ‘Homeric Vengeance and the
Outbreak of Greek War’, in Hans van Wees (ed.), War and Violence in Ancient Greece (London/
Swansea: Classical Press of Wales, 2000), 1–30; van Wees, Greek Warfare, 19–33. Rome: J. E. Lendon,
Empire of Honour (Oxford: Oxford University Press, 1997), 74–7 (cities within empire); Susan Mattern,
Rome and the Enemy (Berkeley: University of California Press, 1999), 171–222.
258 hans van wees
SELECT BIBLIOGRAPHY
Deacy, Susan, and Karen Pierce (eds), Rape in Antiquity (London/Swansea: Classical Press
of Wales, 1997).
Ducrey, Pierre, Le traitement de prisonniers de guerre dans la Grèce antique (Paris: Boccard,
1968; new edn 1999).
Kern, Paul, Ancient Siege Warfare (Bloomington: Souvenir Press, 1999).
Niditch, Susan, War in the Hebrew Bible (Oxford: Oxford University Press, 1993).
Oded, Bustenay, Mass Deportations and Deportees in the Neo-Assyrian Empire (Wiesbaden:
Reichert, 1979).
—— War, Peace and Empire: Justifications for War in Assyrian Royal Inscriptions (Wiesba-
den: Ludwig Reichert Verlag, 1992).
Sabin, Philip, Hans van Wees, and Michael Whitby (eds), The Cambridge History of Greek
and Roman Warfare, vols i–ii (Cambridge: Cambridge University Pess, 2007).
van Wees, Hans, Greek Warfare: Myths and Realities (London: Duckworth 2004).
Vikman, Elisabeth, ‘Sexual Violence in Warfare, Part I: Ancient Origins’, Anthropology and
Medicine 12:1 (2005), 21–31.
Volkmann, Hans, Die Massenversklavungen der Einwohner eroberter Städte in der hellenistisch-
römischer Zeit, 2nd edn (Stuttgart: Franz Steiner, 1990).
chapter 13
.............................................................................................
E A R LY ME D I EVA L
E U RO P E
T H E C A S E O F B R I TA I N
AND IRELAND
.............................................................................................
james e. fraser
As far as the influential sixth-century historian Gregory of Tours was concerned,

the history of fifth-century Gaul (roughly modern France) boiled down to tales of
saints and their powers (virtutes sanctorum) and tales of genocides (strages gen-
tium), literally ‘the slaughter of peoples’. The Venerable Bede might have said the
same thing, two centuries later, about seventh-century Britain. Flourishing in
Northumbria, the kingdom of the northern English, the great scholar wrote that,
three generations before his own, ‘a great strages was conducted of both the Church
and the nation (gens) of the Northumbrians.’ The perpetrator of this atrocitas
against the men, women, and children of Bede’s nation was a British (which is to
say Welsh-speaking) king, Caedualla, who ‘for a long time raged through all their
kingdoms, meaning to wipe out (eradere) the whole English race (genus)’. His
allegation reads like unambiguous genocide, and Bede, like Gregory, would have
scoffed at any suggestion that his society was incapable of strages gentium. This
chapter assesses his assumptions. Its focus is early medieval Britain and Ireland.
From time to time, Insular attitudes are considered against a broader backdrop, but
a British case study is sufficient to answer the central research question relating to
early medieval genocide. Although Len Scales has taken important steps towards
260 james e. fraser
introducing medievalists to genocide studies (and vice versa), that question re-
mains, at present, rather basic and fundamental: did strages gentium happen?1
Britain is an ideal place to confront this question. It retained next to nothing of
the state apparatus inherited in part by the barbarian kingdoms that supplanted
Roman administration in the Latin West, and perpetuated in the Greek East. As
such, there can be no question of applying modern, or indeed classical templates
for understanding state-sponsored genocide, which might be applicable to the
interesting Byzantine evidence. The island was also sharply segregated along
ethno-linguistic lines, which tended to be rather blurrier elsewhere in Europe,
and less likely to become flashpoints of violence. Moreover, students of early
medieval Britain have been talking about genocide for many years, or at least
wondering about it, in attempting to understand the coming of the Anglo-Saxons,
the growth and expansion of their culture and political pre-eminence, and, latterly,
a similar Gaelic expansion in the Scottish north. The lack of any decisive resolution
to these debates suggests that concrete and unambiguous evidence for early
medieval acts meeting the criteria for genocide outlined in the UN Convention
on the Prevention and Punishment of the Crime of Genocide will always elude us,
wherever we might seek them. For the most part we find ourselves stumbling back
into the laps of Bede and his fellow commentators, seeking to understand their
descriptions of strages gentium for what they are and are not; and also for what they
reveal about early medieval conceptions surrounding the destructions of peoples.2
I D E N T I F Y I N G G E N O C I D E : T H E P RO B L E M
OF BIBLICAL MODELS
................................................................................................................
Our problems in establishing whether genocides actually took place in Western
Europe in this epoch are legion, as scholarship surrounding the adventus Saxonum
has shown. That can hardly occasion surprise: it is difficult enough to amass
decisive proof of genocides only years or generations old, never mind strages
gentium from early medieval Britain. Certainly Bede’s description of the ethnic
1 Gregory: Historiarum libri X, ii.prol. (Bruno Krusch and Wilhelm Levison (eds), Gregorii
Turonensis Opera 1: Libri historiarum X, Monumenta Germaniae Historica: Scriptores Rerum
Merovingicarum, 2nd edn (Hanover: Hahn, 1937–51)). Bede: Historia ecclesiastica gentis Anglorum,
ii.20 (Bertram Colgrave and R. A. B. Mynors (eds), Bede’s Ecclesiastical History of the English People
(Oxford: Oxford University Press, 1991)). Scales: Len Scales, ‘Bread, Cheese and Genocide: Imagining
the Destruction of Peoples in Medieval Western Europe’, History 92 (2007), 284–300.
2 Ethnic segregation: Alex Woolf, ‘Apartheid and Economics in Anglo-Saxon England’, in N. J.
Higham (ed.), The Britons in Anglo-Saxon England (Woodbridge: Boydell and Brewer, 2007), 116.
early medieval europe 261
carnage wrought by Caedualla is nothing like as clear-cut as it first appears. He

wrote one hundred years later, with all of the accusatory, one-sided indignation
that investigators have come to expect of victims of alleged atrocities and their
descendants. In the previous chapter of his Ecclesiastical History, he had been
jubilant in the face of genocide. The perpetrator that time was Aeðilfrith, the
‘very brave’ pagan king of Bede’s own Northumbrian people, whom he reports
had ‘laid waste (uastauit) the British gens’, subduing and occupying its districts,
‘having exterminated or subjugated the natives (exterminatis uel subiugatis indi-
genis)’. Translating Bede’s Latin verb exterminare as ‘to exterminate’ may not,
however, be accurate. It implies mass murder to us, when the early medieval
Latinist probably understood exterminare to mean ‘to drive [something] beyond
(ex) the boundaries (termini) [of something else]’. In a later quotation, we shall see
Bede associating acts of exterminatio with bloodbaths, but its usage here is ambig-
uous, more likely to refer to population displacement than massacres.3
Now, for our purposes it may matter little whether Aeðilfrith massacred his
British victims in the course of depopulating their devastated home districts. Either
outcome amounts to the violent physical removal of a people from its territory, and
deserves consideration as an act of genocide, as the UN Genocide Convention
defines that crime as intended destruction in whole or in part. However, we must
accept that translation problems and the hindsight of witnesses introduce layers of
complexity and difficulty in interrogating our sources about genocide. Another
problem is that nothing would be more inappropriate than to evaluate Bede’s
morality using our own inadmissible moral compass. His condemnation of geno-
cide as atrocitas when visited upon his own people by Britons looks hypocritical
and chauvinistic to our eyes, if perhaps predictable, alongside his triumphalist
estimation of the reverse case. His moral position was more complex, however,
than accusations of hypocrisy allow. Bede reflected on Aeðilfrith’s achievement
with recourse to allusions to the Bible, the source of his morality as a devout monk
and one of the foremost experts of the age in biblical studies. The king, he wrote,
‘might be compared with Saul, who was once king of Israel’, on account of his
treatment of the Britons; and he also felt that Jacob’s blessing of Benjamin his son
was suitable for application to Aeðilfrith’s brutalization of them: ‘Benjamin shall
ravin as a wolf; in the morning, he shall devour the prey, and at night he shall divide
the spoil.’ For modern non-specialists, increasingly, such biblical allusions repre-
sent obscure asides, even unwelcome interruptions of a lively narrative. However,
they are often vital clues, revealing to the specialist a great deal about what a writer
like Bede was thinking as he attempted to evaluate an historical episode.4
Why did this particular scripture spring to Bede’s mind in contemplating
Aeðilfrith’s British genocide? The Israelite tribe of Benjamin was regarded by
3 Bede: Hist. eccl., i.34.

4 Ibid. Jacob: Genesis 49:27.
262 james e. fraser
Christian scholars at the time as emblematic of St Paul, a wolf who had ravined in
the morning (by persecuting Christians early in his career), but who had divided
the spoil in the evening (by evangelizing among the gentiles after his conversion).
Bede’s underlying point, in making this connection, was thus to imply that the
British genocide perpetrated by Aeðilfrith had been, ultimately, for the greater
good, like the victimization of those Christians whom Paul had persecuted on his
road to conversion. The eighth-century reader was encouraged to see extermina-
tion of Britons as part of the trajectory that saw the Northumbrians convert
eventually to the Christian religion of their victims (like Paul did), and then
outdo them as exemplary believers (like Paul did). Perversely, from the standpoint
of modern ethics, it seems that Bede was looking upon the Britons as bit-players in
their own genocide: his biblical allusion makes the story all about the Northum-
brian English, whose brutality prior to hearing the gospel heralded their later
redemption and greatness as a Christian nation.
Even that is not the whole story. There was a deeper association between the
Benjamites, genocide, and redemption that influenced the formulation of Bede’s
doctrine relating to Aeðilfrith. There is a biblical account of a Benjamite genocide
perpetrated by their fellow Israelites, who banded against them on account of their
wickedness. The result was a great battle, on the first and second days of which the
Israelites were ‘massacred’ (occiderunt) by the Benjamites: these successes were the
‘ravining in the morning’ which Jacob had foreseen. On the third day, however,
having called upon God for aid, the Israelites turned the tables on the 26,000
Benjamites, and ‘the Lord smote (percussit) them before the eyes of the sons of
Israel, and they destroyed 25,100 men of them.’ The Israelites then smote (percus-
serunt) all the Benjamite cities with such slaughter that the tribe would have died
out, had not maidens been forcibly removed from a neighbouring people in order
to provide them with wives. Like Paul, the Benjamites had required to be disem-
powered in order to see the light. From Bede’s perspective, his own Northumbrian
gens had shared that same requirement, as his linking it to Benjamin shows.5
That conclusion brings us back to the first genocide outlined in this essay—the
Northumbrian one perpetrated by a cruel Briton. This strages Bede clearly reck-
oned to be another stop on his people’s road to their Damascus. For him, like other
thinkers in early medieval Europe, genocide was a key divine method of dealing
with wayward gentes. As a cleanser, it was surpassed only by the Flood—the
ultimate biblical case of strages gentium. The vulgate Bible refers no less than
seventy-four times to groups that, like the Benjamites, experienced percutio, ‘smit-
ing’. God typically plays the role of advocate, and even instigator of the carnage,
5 Bede, Benjamin, and St Paul: J. M. Wallace-Hadrill, Bede’s Ecclesiastical History of the English
People: A Historical Commentary (Oxford: Oxford University Press, 1988), 48. Benjamite genocide:
Judges 20–1.
intending it as a just punishment. Since Bede likened Aeðilfrith with Saul, we might
as well consider an example from that king’s story:
So Saul took the kingship over Israel, and he fought against all his enemies . . . and wherever
he turned, he kept on conquering (superabat). And he gathered a host and smote (percussit)
the Amalekites, delivering Israel from the hands of those who had plundered them . . .
Samuel said to Saul . . . ‘This is what the Lord of Hosts says . . . Go now, smite (percute) the
Amalekites and totally destroy everything that belongs to them. Do not spare them. Put
to death men and women, children and infants, cattle and sheep’ . . . Then Saul smote
(percussit) the Amalekites . . . Agag, king of the Amalekites, he took alive, and all his people
he totally destroyed (interfecit) with the edge of the sword. But Saul and the people spared
Agag, and the best of the sheep and the cattle . . . Then the word of the Lord came to Samuel,
saying, ‘I am sorry that I have made Saul king, because he has turned away from me, and has
not carried out my instructions.’
Here genocide is God’s revenge upon the Amalekites for victimizing the Israelites,
satisfaction eluding him because the carnage falls short of utter extermination.
The underlying morality here of feud or vendetta—Guy Halsall’s ‘customary
vengeance’—was fully embraced by early medieval societies: injury and insult
were to be avenged, violently if necessary, and irredeemable malefactors invited
obliteration. A closer look at Bede’s testimony reveals that the newly converted
Northumbrians whom Caedualla destroyed had, as Bede believed, apostatized
against Christianity. That kind of abject treachery was punishable by violent
retribution in ordinary society, and by genocide where God was concerned.6
As regards the central question of this essay, we must note that social and
theological truths arguably encouraged Bede to ascribe appropriate genocidal
ambitions to Caedualla’s devastating campaigns, whether or not the British king
had himself espoused them. This problem haunts the early medievalist at every
turn where genocide is concerned. As Scales has established for the medieval epoch
entire, ‘tales of inter-ethnic bloodshed’ could ‘[project] clear-cut ethnic purges
onto outbreaks of violence which must often have been both more limited in scope
and more complex in character,’ particularly where a biblical parallel waited in the
wings, to which events could be likened. By introducing such a biblical dimension
to his history, Bede was conveying to his readers a multifaceted understanding of
why the Northumbrians, chastened like the Benjamites, had quickly returned to the
Christian fold after apostasy, with greater resolve and commitment. We have
already seen that he gives the game away in his account of Aeðilfrith’s British
genocide, showing that he had concluded that such persecution had been a
6 Savagery as cleanser: e.g. J. M. Wallace-Hadrill, The Long-Haired Kings (London: Methuen, 1962),
61–2. Saul and the Amalekites: 1 Samuel 14:47–15:11. Customary vengeance: Guy Halsall, ‘Violence and
Society in the Early Medieval West: An Introductory Survey’, in idem (ed.), Violence and Society in the
Early Medieval West (Woodbridge: Boydell Press, 1998), 22–6; Wallace-Hadrill, Long-Haired Kings,
122–8. Bede: Hist. eccl., iii.1. Biblical dimension of medieval notions of genocide: Scales, ‘Bread, Cheese
and Genocide’, 294–5.
264 james e. fraser
necessary Pauline step towards conversion and outstripping these Christians sub-
sequently in the enterprises of evangelization and reformation. Elsewhere, Bede
confirms such a reading of this account by levelling explicit accusations of wicked-
ness at the Britons: they had unjustly withheld Christianity from the Anglo-Saxons
(in contrast to the later Anglo-Saxon thirst for mission); and they were being more
obstinate against reform than any other Insular people (in contrast with the leading
reformist role played by the Anglo-Saxons). Here, again, genocide was theologically
appropriate. Bede may therefore be suspected of concluding that it had actually
taken place on the conjoined understanding that, on the one hand, God ought to
have willed it, no less than he had willed the percutio of the Amalekites, and, on the
other, that Aeðilfrith had been just the Saul for the job.7
I D E N T I F Y I N G G E N O C I D E : T H E P RO B L E M
O F H I S TO R I O G R A P H I C A L M O D E L S
................................................................................................................
The important general points for genocide historians in all of this are twofold. On
the one hand, early medieval accounts of genocide can be problematic. On the
other, and as Scales has argued, despite modern assumptions about the intrinsic
modernity of genocide, the medieval assumption was that strages gentium was
intrinsic to the age. The foregoing discussion is emphatically not to say that
Aeðilfrith and Caedualla had not, in their turn, dealt harshly or brutally with
their victims. It is not even to say that neither of them wished or succeeded in ‘acts
committed with intent to destroy, in whole or in part, a national, ethnical, racial or
religious group,’ as the UN Genocide Convention defines genocide. Admitting the
possibility is one way to explain the Anglo-British hostility characteristic of both
Northumbrian and North British early medieval writings. However, these texts
tend to date from many generations after ‘first contact’, and on the whole are
probably a far better guide to contemporary ‘historicized’ antagonisms, the roots
of which were being sought in older times (and indeed projected into them), than
to the ethnic attitudes of those older times themselves. Scholars find themselves in
the unsatisfying position of being unable to rule out the possibility of real acts of
genocide, having no particular reason to do so, but unable as well to place a great
deal of weight on Bede’s descriptions as proofs.8
7 Scales, ‘Bread, Cheese and Genocide’, 286–7. Bede’s accusations: Hist. eccl., i.22, iii.1, v.23.
8 Anglo-British relations: e.g. David Rollason, Northumbria, 500–1100: Creation and Destruction of
a Kingdom (Cambridge: Cambridge University Press, 2003), 57–109, especially 93 ff (Northumbrian
focus).
In addition to influences exerted upon his historiographical methodology by his

biblical learning, Bede was following in an established historiographical tradition
which had long since concluded that the coming of the Anglo-Saxons to Britain
involved genocide visited upon a wicked British nation for its sins against God. The
architects of that tradition were not Anglo-Saxon, but British. Gildas, who flour-
ished in the middle third of the sixth century, argued in a text used by Bede that the
Britons, having alienated Rome through treason, left themselves vulnerable to
barbarian ‘assaults and massacres’. God had granted them a respite, but because
they embraced wickedness, wishing ‘to purge his family, and to cleanse it from such
an infection of evil’, God brought the Anglo-Saxons to Britain, ‘in just punishment’.
Here again all of the same caveats raised against Bede apply: for the same reasons, it
was a natural assumption that a wicked nation (once identified) should justly be
pulverized at God’s pleasure. Early medieval commentators were, as Paul Fouracre
has observed, attuned to ‘the manifestly vigorous execution of divine judgement’.
Gildas was explicitly marshalling his apocalyptic account as a stern warning to his
fellow Britons, with the implication that, by mending their ways, they would find
the strength from God to rid Britain of the Anglo-Saxon menace in a great counter-
genocide. It would be difficult to find any era of Christian civilization in which
such talk did not resonate, from Gildas’s time to our own. His influence upon
Bede’s notion 200 years later of Anglo-British genocide and counter-genocide at
the opening of Northumbrian history is obvious.9
WA S G E N O C I D E A L L I N T H E M I N D
I N T H E E A R LY M I D D L E A G E S ?
................................................................................................................
Given the uncertainties surrounding what seemed, at the outset, to be pretty
unambiguous allegations of genocide by a careful historian like Bede, whose
work continues to be admired today, as well as the open question of what genocide
might look like in the archaeological record, students of early medieval genocide
find themselves in a difficult position. Not without reason, genocidal conventional
models for understanding early Insular history, arising from naı̈ve readings of such
textual evidence, have been abandoned, and there are numerous Continental
9 Gildas: De excidio Britanniae, i.15, i.19, i.22, i.24 (Michael Winterbottom (ed.), Gildas: The Ruin of
Britain and Other Documents (London/Chichester: Phillimore, 1978), 87–142). Later medieval
accounts with some of the same imagery: Scales, ‘Bread, Cheese and Genocide’, 288–9. Paul
Fouracre, ‘Attitudes towards Violence in Seventh- and Eighth-Century Francia’, in Halsall (ed.),
Violence and Society, 60–75, at 63.
266 james e. fraser
parallels. However, the findings of genetic research have recently seen genocide, or
at least mass population displacement, return from the cold as a subject for
scholarly deliberation in early British history. At the same time, medievalists,
early and late, are increasingly willing to contemplate genocide as an aspect of
medieval life and society, as well as of medieval thought.10
Apart from what they thought, then, did early medieval people actually commit
acts ‘with intent to destroy, in whole or in part, a national, ethnical, racial or
religious group’? This question returns us to the problem of Latin terminology. It is
clear that the biblical percutiones of the Amalekites and Benjamites constitute
genocide by anybody’s reckoning, and it need not be doubted that Bede had
precisely the same horrors in mind when he wrote about Anglo-British massacres
(strages), brutality (atrocitas), and devastation (uastatio). At the very least, then,
early medieval commentators can be said both to have been familiar with the
concept of genocide, and to have perceived strages gentium as quite possible, both
in thought and in deed, within the sociological and technological confines of the
times. Such perceptions must carry greater weight than modern preconceptions
about what such societies, lacking advanced technological weapons of mass de-
struction, were capable of.
Bede and Gildas expected to horrify and to edify through their lurid descriptions
of ethnic carnage, economic devastation, and depopulation. They expected these to
be accepted as true accounts of the past, chilling proofs of God’s earnest will that all
people should live according to his word and laws. There is nothing to suggest that
they expected readers to find them far-fetched; indeed, attempts to horrify and
edify depend on an ability to play on genuine and present fears. Neither were their
audiences exclusively erudite biblical scholars fully versed in contemporary forms
of text-criticism, reading on a multiplicity of levels, so that surface features paled
into insignificance. Bede dedicated his work to his king, and a draft was sent to the
royal court in advance ‘for perusal and criticism’. Gildas names and shames a series
of kings in the second person, as if the sermon was intended for their eyes: ‘what are
you doing, Aurelius Caninus,’ he wrote for example; ‘are you not being engulfed by
the same slime as the man I have just spoken about?’ Although one would not wish
to press the point too far, nor a long way down the social scale, notions of secular
illiteracy in the early medieval epoch can be exaggerated, even in Britain and
Ireland. The Carolingian historian Nithard was active as a military leader, with a
strong sense of God’s role in affecting outcomes on battlefields. We may take it that
in general terms the military elites who ruled and dominated early medieval realms
10 Continental historiographical parallels: e.g. Bryan Ward-Perkins, The Fall of Rome and the End of
Civilization (Oxford: Oxford University Press, 2005). Genetics: Mark G. Thomas et al., ‘Evidence for
an Apartheid-Like Social Structure in Early Anglo-Saxon England’, Proceedings of the Royal Society B
(doi: 10.1098/rspb.2006.3527 (2006)); Martin Richards et al., ‘Genetics and the Origins of the British
Population’, Encyclopedia of Life Sciences (forthcoming), provides some welcome caveats.
were no strangers to genocidal rhetoric and historical models, biblical or otherwise,

whether or not they understood them on all of the levels that the clerical intelli-
gentsia did. As a result, we cannot dismiss the possibility that such familiarity
encouraged genocidal ambitions, or for that matter ambitions calculated to pre-
vent genocide being inflicted upon their people.11
With this in mind, let us turn to the eighth-century annalist, connected with the
great monastery in the Hebridean island of Iona in Atlantic Scotland, who was
familiar as a monk with the biblical models known to Bede and Gildas, to guide
him in his choice of terminology. This man recorded a percutio in 741, inflicted
upon the Dál Riata, the leading Gaelic people in the West Highlands of Scotland, by
a Pictish king from the north-east lowlands, only about ten years after Bede wrote.
Similarly, a percussio of the district of Dyfed in south-west Wales a century before
was recorded by a British annalist. These records could not be more terse or
dispassionate, and could scarcely be less informative. At the same time, they are
precious for their contemporaneity. Their writers had neither the time nor the
inclination to assimilate these elusive events within the grander narratives of a Bede
or a Gildas. This situation may be contrasted with the more problematic hindsight
of the Anglo-Saxon Chronicle, compiled more than a century after Bede, which
envisioned the early wars between Anglo-Saxon and Briton as race wars peppered
with genocidal episodes. Thus at Anderitum in the 490s, that chronicle records, a
successful siege by two Anglo-Saxon kings was followed by a massacre, in which
they ‘killed all who lived in there; there was not even one Briton left there.’ The
chronicle is not by any means a contemporary witness to this alleged carnage, and
Asser’s description, based on it, of an Anglo-Saxon genocide of Britons on the Isle
of Wight is even less reliable. Our eighth-century chronicler, however, was a
contemporary witness to his percutio Dáil Riata.12
Early medieval Latinists had a range of terms available to them to describe
devastating military campaigns, and percutio, although it is very common in the
Bible, is very rare in Insular usage as applied to such campaigns. All that really
prevents us from concluding that these percutiones are contemporary records of
acts of genocide or ethnic cleansing like the biblical examples are our own pre-
conceptions, though of course it cannot ever be proven that these annalists equated
11 Bede: Hist. eccl., preface. Gildas: De excid., i.30.

12 Insular percutiones: Annals of Ulster, 741.10 (Seán Mac Airt and Gearóid Mac Niocaill (eds), The
Annals of Ulster (to A.D. 1131) (Dublin: Dublin Institute for Advanced Studies, 1983)); Annales
Cambriae, 645.1 (Edmond Faral (ed.), La Légende Arthurienne: études et documents, vol iii (Paris: H.
Champion, 1929), 44–50). Anglo-Saxon Chronicle: Anglo-Saxon Chronicle, s. a. 491 (e.g. Susan Irvine
(ed.), The Anglo-Saxon Chronicle: a Collaborative Edition 7 (MS E) (Cambridge: D. S. Brewer, 2004)).
Asser: ‘Life of King Alfred’, }2; W. H. Stevenson (ed.), Asser’s Life of King Alfred, Together with the
Annals of St. Neots, Erroneously Ascribed to Asser (Oxford: Clarendon Press, 1904); for discussion, see
Simon Keynes and Michael Lapidge (eds.), Alfred the Great: Asser’s Life of King Alfred and Other
Contemporary Sources (London: Penguin, 1983), 230.
268 james e. fraser
percutio with genocide. Similarly, Bede had a further tale of genocide to tell,
involving a different Caedwalla, an Anglo-Saxon king of Wessex this time, which
may be regarded as more compelling evidence of genocide than the examples
already discussed. ‘After Caedwalla had gained possession of the kingdom,’ he
wrote, ‘he also captured the Isle of Wight . . . ; and he endeavoured to exterminate
(exterminare) all the natives by merciless slaughter (ac stragica caede), and to
replace them by men from his own kingdom.’ Here, as promised above, we find
Bede linking exterminatio with strages. He was describing events that had taken
place when he was thirteen years old, not generations earlier. Moreover, involving
as they do two groups of Anglo-Saxon pagans, they are not bedevilled by the
Anglo-British dimension which problematizes our accounts above. Indeed, unlike
in those cases, Bede supplied no explanation or justification of Caedwalla’s aggres-
sive ethnic purge of the Isle of Wight, presumably because it was self-explanatory
and insufficiently remarkable to invite further comment.13
Early medieval populations enjoyed no intrinsic rights to exist or flourish: they
were entitled to expect instead, at God’s pleasure, to be harvested by enemies bent
on devastation, pillage, killing, rapine, and enslavement. All indications are that
secular society accepted that might was right, not least because it was a condition
bestowed by God. The Irish wisdom text Audacht Moraind concerned to outline
princely virtue, idealized princes who were just, merciful, and supportive of their
own people, and who kept them safe. Its author accordingly saw bloodshed at
home as ‘a vain destruction of all rule’. In contrast, he urged the ‘true’ prince to
‘remove the shame of his cheeks by arms in battle against other lands’. In that
context, far from requiring moral justification, aggression and bloodshed were
moral imperatives whereby ‘the prince’s truth’ could be tested and proven.14
GENOCIDE AND THE QUESTION OF IDENTITY

................................................................................................................
These factors encourage acceptance that, in these instances, we have accounts of
real acts tantamount to genocide in early medieval Britain. Can we draw the same
conclusion about the Hebrides and Orkney, obliterated (deletae sunt) in 671 and 681
respectively, possibly by the same Pictish king each time? Among these different
examples of contemporary records, only in the Dalriadic one in Scotland can we be
confident that the victims were a particular ‘national, ethnical, racial or religious
13 Bede: Hist. eccl., iv.16. Self-evident nature of ethnic violence: Scales, ‘Bread, Cheese and
Genocide’, 285–6.
14 Might was right: e.g. Fouracre, ‘Attitudes Towards Violence’, 67. Audacht Moraind: Fergus Kelly
(ed.), Audacht Moraind (Dublin: Dublin Institute for Advanced Studies, 1976).
group’, distinct from the perpetrators. This problem is compounded by the fact
that it is not always clear how groups (and thus potential perpetrators and victims
of genocide) went about distinguishing one gens from another. This question has
been the subject of a very considerable body of research, the size and scope of which
can come as something of a surprise to non-specialists. Nationalist historiography
of the nineteenth and twentieth centuries understandably regarded the collapse of
the Roman Empire and the rise of the earliest medieval realms of Western Europe
as a pivotal moment in the genesis of the nations which became the great powers of
Europe and the colonial world. Unsurprisingly perhaps, given the uses to which
such historiography could ultimately be put in informing the nationalist and racist
ideologies which shaped the great and terrible wars and colonial projects of the age,
the evidence relating to the formation, function, and perception of ethnicity in the
early Middle Ages is fraught with sometimes acute controversy which must be
passed over here.15
The Latin terminology associated with ethnicity (gens and natio in particular)
envisions ethnic identity as ‘kinship “writ large”’, marking early medieval people
out, in human sciences terms, as ‘primordialists’ who believed that ethnicity was
largely a function of ‘the givens of birth’—ties of blood (real and imagined) to kin
and gens, homeland, language, and culture. Yet individual identities were also
influenced by situational concerns, as any number of examples show. Early medie-
val communities thus possessed flexible means of ascertaining their sameness with,
or difference from other groups, and could shift the goalposts to suit their
immediate situation. Thus the obliteration of pagan Anglo-Saxon communities
on the Isle of Wight by the forces of a pagan Anglo-Saxon king of Wessex was
capable of being envisioned as a struggle between two quite distinct groups,
however much they might, in other situations, invite consideration as constituents
of a single race. The potential for such situationality in early medieval ethnicity,
including the role of kings in identifying otherness, raises the particular issue of
‘ethnogenesis’, and the sometimes bitter scholarly dispute surrounding the para-
digmatic theoretical model formulated by Reinhard Wenskus to explain the crea-
tion of peoples in early medieval times. Wenskus envisioned new post-Roman
ethnic groups as coalescing around a coherent group of aristocrats, which he called
a Traditionskern, a ‘core of tradition’, and concluded that its sacred traditions
preserved ethnic traits of great antiquity that were transferred outwards to the
entire gens which it had come to dominate. On this model, the behaviour of kings
and other social elites is the principal engine which shaped, stabilized, and over-
turned early medieval ethnic solidarities.16
15 Hebridean episodes: Annals of Ulster, 672.2, 682.5.

16 Early medieval ethnicity: Patrick J. Geary, ‘Ethic Identity as a Situational Construct in the Early
Middle Ages’, Mitteilungen der Anthropologischen Gesellschaft in Wien 113 (1983), 15–26; for a more
general statement on medieval ethnicity, see Scales: ‘Bread, Cheese and Genocide’, 290. Primordialism:
270 james e. fraser
‘Traditionskern ethnogenesis theory’ has not gone unchallenged, and criticism

has revolved in particular around the notion of the ‘core of tradition’. The critique
has considerably more to commend it than the paradigm, with the result that we
are not required to accept that ethnic consciousness within the early medieval gens
emanated from an elite core. It is an important point for the study of genocide.
Had Wenskus’s theory held true, the destruction of the Traditionskern would have
emerged as sufficient ipso facto to bring about ‘ethnonemesis’, the eradication of the
ethnos. Such a conclusion does not seem to be justified. Our texts do indicate, none
the less, that the leaders of a community enjoyed an intimate relationship with its
ethnic identity. A famous passage from an Irish legal tract from the period observes
that ‘a people (túath) is no people, without scholar, church, poet, and king.’ The
sentiment chimes with that aspect of early medieval thought that encouraged
Wenskus, and those who have since embraced his theory, to adopt their elitist
notion that the roots of ethnic consciousness were planted in an elite Traditions-
kern. Fortunately, the Irish sentiment also chimes with Scales’s observations from a
later medieval perspective that, in medieval thought
to strip a people of its identity, in the form of its rights and liberties, was to render it
incapable of its own defence, and thus expose it to the gravest of perils . . . [A people] was
conceivable as a community of a shared identity, rooted in law and privilege. To forfeit these
was to stand helpless before an implacably hostile world. From disenfranchisement and loss
of common identity to physical annihilation . . . was but a short step.
On this showing, that early medieval Irish lawyers regarded communities as

non-peoples without a scholar, a church, a poet, and a king had nothing (neces-
sarily) to do with any sense that elites were the engine that drove ethnic conscious-
ness, Traditionskern-fashion. The point was instead that a túath or people was
regarded as something ‘rooted in law and privilege’, which, if it lacked scholar,
church, poet, and king, was in no position to protect and uphold the rights and
liberties involved, and so could not be recognized as possessing such entitlements
in the first place.17
Siân Jones, The Archaeology of Ethnicity: Constructing Identities in the Past and Present (London/New
York: Routledge, 1997), 56–83. Insular examples of situational identity: Adomnán, Vita sancti
Columbae, i.2 (Alan O. Anderson and Marjorie O. Anderson (eds), Adomnán’s Life of Columba, 2nd
edn (Oxford: Oxford University Press, 1991)); Bede: Hist. eccl., iv.22 (see also James Campbell, Essays in
Anglo-Saxon History (London/Ronceverte: Hambledon Press, 1986), 86–7). Wenskus: Reinhard
Wenskus, Stammesbildung und Verfassung: das Werden der frühmittelalterlichen gentes (Cologne:
Böhlau, 1961). Synopses and evaluations of ‘ethnogenesis’ paradigm: Andrew Gillett (ed.), On
Barbarian Identity: Critical Approaches to Ethnicity in the Early Middle Ages (Turnhout: Brepols,
2002), esp. 1–18, 39–68, 221–39.
17 Irish legal tract: Fergus Kelly, A Guide to Early Irish Law (Dublin: Dublin Institute for Advanced
Studies, 1988), 4. Scales, ‘Bread, Cheese and Genocide’, 292.
GENOCIDE AND THE P U RG AT I O N OF ELITES

................................................................................................................
It follows naturally from such thinking that, Traditionskern or no Traditionskern,
elite purges in the early Middle Ages could, in many cases, double as ethnic ones in
the imagination, since disempowerment could threaten a community’s capacity to
uphold and defend the common customs, laws, and privileges which it believed set
it off from other groups. In other words (and although it was certainly not
intended as such), the Irish sentiment could, if it was widely appreciated, have
provided something of a formula for genocide in early medieval Europe. The
ninth-century Frankish biographer Einhard would seem to have been thinking
along similar lines to the Irish jurists, when he envisioned the destruction of Avar
elites by Charlemagne as equating to the depopulation of Avar-held Pannonia. This
link between ethnicity and protection of laws and privileges offers something of an
explanation, too, of instances where the gens seems to be equated with the army in
the early Middle Ages, with the corresponding point that massacres of warriors
could have been intended, and understood at some level, as genocidal acts. It may
be that we have an Insular example. After a battle in Ireland in 738 between the
powerful Uı́ Néill kingdoms and the Laigin of Leinster, the Laigin king ‘was
beheaded by a battle-sword’, and the Uı́ Néill ‘in unaccustomed fashion, routed,
trampled, crushed, overthrew and destroyed their Laigin adversaries, so much so
that almost the entire enemy was well nigh annihilated.’18
The biblical percutiones discussed above bear no relation either to elite purges or
to massacres confined to military men. Early medieval commentators could at least
envisage genocide as mass murder on the kind of horrific scale that we tend to
associate with it today, and could also justify it with the help of moral theology. It
may therefore be that we would recognize as genocidal acts the rare percutiones on
Insular record in this period. In contrast, the purging of elites probably fails to
register today as genocide, but did not fail to do so then, and seems to have been
fairly common in early medieval Britain. Bede’s account of the genocide visited
upon the denizens of the Isle of Wight by Caedwalla of Wessex conveniently weds
the purgation of elites with extermination:
I think that I must not pass over in silence the fact that among the first fruits of the island
who believed [i.e. converted to Christianity] and were saved were two boy-princes (regii
pueri) . . . When the enemy was approaching the island they escaped by flight and crossed
over [to the mainland] . . . They thought they could remain concealed from the victorious
king; but they were betrayed and condemned to death.
18 Armies as nations: Michael Kulikowski, ‘Nation versus Army: A Necessary Contrast?’, in Gillett
(ed.), On Barbarian Identity, 69–84. Avars: Einhard, Vita Karoli magni, }13 (Oswald Holder-Egger
(ed.), Einhardi Vita Karoli Magni, Monumenta Germaniae Historica: Scriptores Rerum
Germanicarum 25 (Hannover: Hahn, 1911)). Uı́ Néill and Laigin: Annals of Ulster, 738.4.
272 james e. fraser
The implications of the full account are that Caedwalla succeeded in his genocidal
agenda, which featured as its centrepiece the killings of these young princes. His
account of the other Caedualla, the British king who strove to eradicate the
Northumbrians, is also characterized by the flights of princes, adult and child, to
the courts of foreign kings whence none returned, as well as by the executions of
those who had failed to elude capture. Princely exile is a recurrent feature of regime
change in Bede’s Britain, suggesting that purges were an anticipated aspect of
political change. Men like Aeðilfrith, Caedualla, and his later namesake may not,
in fact, have visited upon their victims levels of elite carnage which differed greatly
from the normal amounts of blood shed when dynasty replaced dynasty through
the usual course of succession custom. It may have been their otherness, rather
than bloodthirstiness in excess of the norm, which encouraged the sense that
genocide had taken place.19
GENOCIDE AND RELIGIOUS OTHERNESS

................................................................................................................
In the cases of both Aeðilfrith, a pagan exterminating Christians, and Caedualla, a
Christian wiping out apostates, religion was a prominent element of their other-
ness, at least for Bede. In his own day, Bede continued to see the religious otherness
of the Britons as a key factor in their ongoing troubles with the Anglo-Saxons,
observing that ‘they oppose . . . the whole state of the catholic church by their
incorrect Easter and their evil customs,’ and so were ‘opposed by the power of
God and man alike’, with the result that ‘they have been brought partly under
English mastery.’ Scholars have debated the reasons for Bede’s infatuation with the
seventh-century Insular schism over observances like the correct calculation of the
date of Easter, but here he indicated that it helped to explain why God endorsed
Anglo-Saxon subjugation of British communities. Similar thinking underlies the
problematic note, in a later medieval chronicle relating to the extinction of the
Picts in northern Britain, that they had been ‘annihilated’ (deleuit) because ‘God,
by reason of their wickedness, deigned to make them alien from and void of their
heritage.’20
If the early medieval Insular world was capable of looking upon genocidal
suppressions of religious groups as righteous acts against the wicked, the most
19 Bede: Hist. eccl., ii.20; iii.1; iv.16; iv.22.

20 Bede: Hist. eccl., v.23. Picts: Chronicle of the Kings of Alba, }2 (Marjorie O. Anderson (ed.),
Kings and Kingship in Early Scotland, 2nd edn (Edinburgh/London: Scottish Academic Press, 1980),
149–53).
infamous attempt to destroy one in the age was perpetrated by Charlemagne on

the Continent in and after the 770s. In 775, according to a contemporary
chronicle, three years after destroying and pillaging one of their shrines, he
began all-out war on ‘the treacherous and treaty-breaking gens of the Saxons’,
intending that it should be ‘either defeated and forced to accept the Christian
religion, or entirely exterminated (tollerentur)’. Charlemagne’s biographer Ein-
hard wrote that this war was atrocius, full of atrocities, lasting over thirty years
until the victorious king ‘transported some 10,000 men . . . and dispersed them
in small groups, with their wives and children, into diverse parts of Gaul and
Germany.’ This depopulation, attested as well in a contemporary chronicle, was
intended to ensure that the Saxons would ‘be united with the Franks and
become one people with them’. As news of Charlemagne’s war spread, an
Anglo-Saxon abbot wrote to him, entreating him to ‘increase [his] righteous
zeal for their conversion, suppress the worship of idols, [and] cast down the
buildings of their temples’, that he might ‘find a rewarder in heaven in him
whose name and knowledge [Charlemagne] will have spread on earth’. Thus was
the moral rectitude of the Saxon genocide confirmed on the grounds of religious
persecution, as Bede had envisioned for his Northumbrian examples, where
motives of border security, imperialism, persecution, and strages gentium were
similarly inextricably intertwined. The same mix of factors was surely involved
in Charlemagne’s campaigns against the Avars, which according to Einhard left
their lands ‘completely uninhabited’; in the suppression of a Saxon rebellion ‘in
a great bloodbath (nimia caedes)’ two generations later; and in Ireland in 902,
when the pagan Scandinavian community which had captured and settled at
Dublin fifty years earlier was ‘driven from Ireland’ by a conspiracy of Uı́ Néill
and Laigin kings, ‘and escaped half dead after they had been wounded and
broken’. The famous expulsion of Irish settlers from Dyfed by a dynasty of
North Britons soon after the end of Roman Britain, recorded in a ninth-century
text, is, however, almost certainly pseudo-historical.21
21 Charlemagne and the Saxons: Annales regni Francorum, s. a. 772, 775–6, 804 (Friedrich Kurze
(ed.), Annales regni Francorum (741–829), Monumenta Germaniae Historica: Scriptores Rerum
Germanicarum 6 (Hanover: Hahn, 1895)); Einhard, V. Karoli, }7; see also e.g. Henry Mayr-Harting,
‘Charlemagne, the Saxons, and the Imperial Coronation of 800 ’, English Historical Review 111 (1996),
1113–33. Anglo-Saxon approbation: Dorothy Whitelock (ed.), English Historical Documents c.500–1042,
2nd edn (London/New York: Methuen, 1979), 832–3. Avars: Einhard, V. Karoli, }13. Saxon caedes:
Nithard, Historiae, iv.6 (Ernst Müller (ed.), Nithardi Historiarum Libri IIII, Monumenta Germaniae
Historica: Scriptores Rerum Germanicarum 44 (Hanover: Hahn, 1907)). Scandinavians: Annals of
Ulster, 902.2. Dyfed: Alex Woolf, ‘The Expulsion of the Irish from Dyfed’, in Karen Jankulak and
Jonathan M. Wooding (eds.), Ireland and Wales in the Middle Ages (Dublin: Four Courts Press, 2007),
102–15.
274 james e. fraser
B RU TA L I T Y A N D D E H U M A N I Z AT I O N
A S M O R A L G R EY A R E A S
................................................................................................................
Clerical commentators were of course capable of looking upon brutality as a
hallmark of wickedness. Bede certainly did so with regard to Caedualla. Yet for
early medieval clergy there was nothing intrinsically immoral about horrific
violence, up to and including genocide. Context was everything. Gildas celebrated
how Roman soldiers, arriving in Britain to deliver the provincials from barbarian
invaders, ‘laid low a great number’, and ‘planted in their enemies’ necks the claws of
their sword-points’, so that they ‘caused among them a slaughter like the fall of
leaves’. A psalm which his hagiographer said was sung by the Irish St Columba
outside the gates of a royal Pictish fortress exhorts kings to ‘ride forth victoriously
on behalf of truth, humility and righteousness’, that ‘peoples (populi) shall fall
beneath your feet.’ Gildas’s lurid details relating to the ‘dreadful and devastating
onslaughts’ endured by the Britons at the hands of barbarians, in which ‘a quick
end saved [men] from the miserable fate which awaited their brothers and chil-
dren,’ were intentionally apocalyptic. Yet the apocalypse reflects, as well as exagger-
ating, warfare as it was practised in Britain when Gildas wrote. From a modern
perspective war crimes drip from the pages of early medieval chronicles, but our
witnesses, as often as not, present such details as the products of divine or natural
justice. In that respect, their treatment of strages gentium forms a subgroup of
clerical treatment of war more generally, in which notions of war crimes were
rather different than our own. Nithard provides particularly revealing ninth-
century insight, recording that after a battle in which he had fought on the winning
side, a convocation of bishops found that the victors ‘had fought for justice and
equity alone, as God’s judgement had made clear’ in the granting of victory. Each
man among them was found to be ‘an instrument of God, free from responsibility’,
save any who ‘knew that he had either counseled or committed anything . . . from
wrath or hatred or vainglory or any passion’. In the early Middle Ages, war crimes
were thought crimes.22
The UN Genocide Convention is not solely concerned, of course, with attempts
to obliterate groups directly through violent atrocities. There are also questions of
inflicting ‘mental harm’ on them, and creating ‘conditions of life’ calculated to
eliminate them. It is too difficult to know the minds of early medieval people well
enough to comment on the psychological effects of ethnic victimization. It may
be worth raising, however, the subject of dehumanization. In his biography of
Charlemagne, Notker related stories in which Franks spoke of Scandinavians as
22 Attitudes towards violence: e.g. Fouracre, ‘Attitudes towards Violence’, 62–3. Gildas: De excid.,
i.15–19. Adomnán: V. sanct. Columbae, i.37; Psalm 45 (Vulg. lxiv). Nithard, Hist., iii.1.
‘dog-heads’ (cynocephales), of Slavs and Avars as ‘tiny birds’ (aviculae), and of one
Slavonic group as ‘tadpoles’ (ranunculi) and ‘worms’ (vermiculi) who died ‘squeal-
ing their incomprehensible tongue’. Non-Latinate readers will have missed the
additional belittling implications of the -culi/-culae endings of some of these
words, emphasizing the wretchedness of these dehumanized enemies. Gildas pro-
vides the most picturesque Insular examples, with recourse to ugly imagery all too
familiar to us today, speaking of the Britons’ aggressive Insular neighbours as ‘dark
throngs of worms (vermiculi) which wriggle out of narrow fissures in the rock when
the sun is high,’ or as ‘greedy wolves (lupi), rabid with extreme hunger, which, dry-
mouthed, leap over into the sheepfold when the shepherd is away.’ Much later,
Bede’s older contemporary Stephen of Ripon referred to Pictish rebels against
Northumbrian domination as ‘bestial peoples’ (populi bestiales) and as ‘swarms
of ants’ (formicarum greges).23
That such descriptions may be thought offensive seems to be confirmed by Bede,
whose discretion prevented him from adopting the dehumanizing epithets fa-
voured by Gildas and Stephen, despite having otherwise made use of their works
as sources of historical information. Dehumanizing rhetoric is not genocide, but is
regarded by modern commentators as intrinsic to the process, or at least as
preparation for the righteous extermination of such inhuman pests. Something
more sinister may have been on the cards in a West Saxon law code dating from
Bede’s lifetime. Here, as Alex Woolf has argued, the fact that the honour prices
compensating British nobles and normal freemen were set respectively at just 50
and 60 per cent of those of their Saxon counterparts may have had a much greater
and more devastating effect even than the classification of Britons as second-class
citizens in Wessex. As Woolf theorizes, ‘the imbalance . . . may hold within it the
very key to the disappearance of the Britons’ in much of lowland Britain:
If, for example, a hypothetical English and British nobleman . . . got into a series of disputes
with one another and were dealt with fairly by the courts . . . then all compensations paid by
the Briton to the Englishman would be twice the value of those paid to him by his
opponent. The end result would be that the property and finally the land [held by the
Briton] would pass to the Englishman.
If Woolf ’s theory correctly encapsulates the long-term effects of such stark legal
segregation, which he terms ‘apartheid’ unreservedly, we are looking here at
something of a blueprint for the eradication of the Britons. However, it is unlikely
that such eradication was envisaged or intended, so that it cannot really be
23 Notker, Gesta Karoli, ii.12–13 (Hans F. Haefele (ed.), Notker der Stammler: Die Taten Karls des
Großen, Monumenta Germaniae Historica: Scriptores Rerum Germanicarum, Nova Series 12
(Hanover: Hahn, 1959)); for discussion, see Matthew Innes, ‘Memory, Orality and Literacy in an
Early Medieval Society’, Past and Present 158 (1998), 3–36, at 27–8. Gildas: De excid., i.16, i.19. Stephen,
Vita sancti Wilfrithi, }19 (Bertram Colgrave (ed.), The Life of Bishop Wilfrid by Eddius Stephanus
(Cambridge: Cambridge University Press, 1927)).
276 james e. fraser
regarded as genocide through the creation of conditions of life calculated to destroy

a group. It is more likely to have been the natural result of a system devised in order
to emphasize the political superiority of Anglo-Saxon groups within a polity which
included British ones, while at the same time recognizing the wisdom of preserving
the legal rights of the latter in order to stave off resistance. As such, this scenario
may stand as an example of the kinds of indirect damage which could, in the early
Middle Ages, be inflicted by dominant groups upon submissive ones, as a result of
a tendency to devalue and dehumanize those regarded as outsiders.24
C A P T I V E S , S L AV E S , AND D E P O P U L AT I O N
................................................................................................................
Similar indirect damage could be inflicted upon early medieval groups through the
forcible transfer of women and children into other groups, a genocidal crime
recognized by the UN Genocide Convention. In 697, Adomnán of Iona formulated
and promulgated a Law of Innocents, embraced in Ireland and northern Britain. Its
purpose was to discourage the victimization of innocentes or non-combatants—
principally women, children, and clerics—by the armed classes of Insular society
whose arms were emblematic of their free status. The effects of the law are
unknowable. What is important for our purposes is that it anticipates that victimi-
zation of women and children was rampant. Indeed, the text of the law provides
some chilling details about men, women, and children being butchered, mutilated,
and degraded by warriors in horrific ways. These details are much more convincing
than Notker’s fanciful tale of Charlemagne rounding up the children of one
unnamed people and ‘shortening by a head’ all who exceeded the height of a
sword. Taking innocentes captive seems to have been fairly common. Sometimes,
the intention was to ransom them back to their home communities. Early Irish law
cast no moral aspersions upon the killing of unransomed captives; and the Church
(which did) was inclined to intervene and ransom them.25
An alternative to murder was enslavement. Bede tells the tale of a warrior
nobleman who tries but fails to pass himself off to his captors as a country
24 Dehumanizing rhetoric and genocide: Rowan Savage, ‘ “Disease Incarnate”: Biopolitical

Discourse and Genocidal Dehumanisation in the Age of Modernity’, Journal of Historical Sociology
20:3 (2007), 404–40. Woolf, ‘Apartheid and Economics’, 127–9 (for genetic support, see Thomas et al.,
‘Evidence for an Apartheid-Like Social Structure’).
25 Adomnán, Cáin Adomnáin, }}33–6, 40, 44, 46, 50 (Máirı́n Nı́ Dhonnchadha, ‘The Law of
Adomnán: a Translation’, in Thomas O’Loughlin (ed.), Adomnán at Birr, AD 697: Essays in
Commemoration of the Law of the Innocents (Dublin: Four Courts Press, 2001)), 53–68. Notker, Gest.
Karoli, ii.12. Kelly, Guide to Early Irish Law, 128–9.
bumpkin, receives a stay of execution for his honest admission of his true identity,
and is sold into slavery to foreign buyers. People could be harvested for enslavement
more actively. The famous St Patrick described a slave raid in which newly converted
Irish women and men ‘were butchered and slaughtered with the sword’ by the
retinue of a Christian British king, and others ‘were removed and carried off to
faraway lands’. There they were sold into slavery, ‘who were freeborn’, although the
raiders had evidently enslaved some of the women themselves. Patrick’s indignant
reaction was roused, not by the raid as such, but by the fact that he had not been
given an opportunity to redeem the captives, who had been enslaved rather than
being kept safe for ransoming. As a youth in his native Britain, Patrick had himself
been a victim of a similar raid and was removed to Ireland along with youths of both
sexes, where he was enslaved. Communities denuded of their women and children in
this way (to say nothing of bodies of normal freemen) could face a bleak future: the
Benjamites discussed above avoided annihilation only because their Israelite con-
querors replaced their massacred women with others captured from abroad. Here
again, however, whether the intention was ever to threaten a group’s existence is an
open question. In neither of the cases involving himself is there any hint in Patrick’s
testimony of specific genocidal motives behind, or consequences from, the transpor-
tation of women and children. That being said, cross-cultural anthropological
analysis of this question suggests that ‘non-state’ societies like these recognized at
some level its genocidal potential, if only on a local scale. All that can really be said
with confidence is that, where genocidal motives existed in early medieval Europe,
tactics like these were well known and available for scattering a population among
several captors, and undermining its reproductive future.26
F I NA L T H O U G H T S : G E N O C I D E A N D
T H E E A R LY M E D I EVA L M I N D
................................................................................................................
Scholars have good reason to baulk at the application of a term like ‘genocide,’ with
all its twentieth-century moral and legal baggage, to early medieval episodes of
violence, depopulation, and displacement which would otherwise seem to meet the
criteria for genocide enumerated in the UN Convention on the Prevention and
Punishment of the Crime of Genocide. Yet that Convention observes in its
26 Bede: Hist. eccl., iv.22. Patrick, Epistola, }}3, 7, 10, 13, 15, 16, 22 (A. B. E. Hood (ed.), St. Patrick: His
Writings and Muirchu’s Life (London/Chichester: Phillimore, 1978), 35–8). Patrick, Confessio, }}1, 16
(Hood (ed.), St. Patrick, 23–34). Anthropology: Lawrence H. Keeley, War before Civilization (Oxford:
Oxford University Press, 1996), 48.
278 james e. fraser
Preamble ‘that at all periods of history genocide has inflicted great losses on
humanity,’ and our examination of the early medieval evidence from Britain and
Ireland suggests that this period at least need not be excluded from such an
assertion. Historians having long since abandoned judgemental evaluations of
past societies according to anachronistic ethics; no early medieval person or people
was guilty of the crime of genocide. It is an alien modern concept which early
medieval people would have struggled to comprehend. For them, strages gentium
were an integral part of the natural order of things ordained by God, the ultimate
judge of humanity, whose sentences could include the smiting of individuals and
whole peoples. Even had they any conception of crimes against humanity, to have
questioned God’s sentence of percutio was to commit Saul’s crime against him; and
a crime against God was always going to outweigh other considerations in early
medieval Europe.
Whether or not we absolve them from guilt, early medieval Europeans may have
something to contribute to genocide studies today. The evidence shows that
concepts of ethnic obliteration, and what constituted it, were variable, depending
upon class and social standing, and extended well beyond the comparatively
narrow confines of actual physical slaughter. It may be particularly important
that disempowerment and defencelessness could be regarded as annihilation, the
social theory of the age envisioning an intimate and mutually reinforcing relation-
ship between elites and ethnicity. If modern social theories disagree with early Irish
law that a group’s scholars, churches, poets, and king represent the pillars of its
peoplehood, worthy of strenuous protection, they may routinely identify other
entities and institutions whose endangerment is liable to occasion indignation for
seeming to be a threat against a community’s very existence. A consideration of
genocide in the early Middle Ages thus reinforces for us that genocide and social
theory are inextricably linked in every age, and that genocide may take on a
number of guises over time and space, the shape and conception of a given society
shaping the formula for its obliteration. Is existing legislation sufficiently flexible to
take adequate account of the variation in social theories which characterizes
humanity today? If not, we may be failing to provide adequate protection against
annihilation to threatened groups and communities, especially those whose social
theories may have more in common with early medieval thought than the West.
SELECT BIBLIOGRAPHY
Colgrave, Bertram, and R. A. B. Mynors (eds), Bede’s Ecclesiastical History of the English
People (Oxford: Oxford University Press, 1991).
Geary, Patrick J., ‘Ethnic Identity as a Situational Construct in the Early Middle Ages’,
Mitteilungen der Anthropologischen Gesellschaft in Wien 113 (1983), 15–26.
Gillett, Andrew (ed.), On Barbarian Identity: Critical Approaches to Ethnicity in the Early
Middle Ages (Turnhout: Brepols, 2002).
Halsall, Guy (ed.), Violence and Society in the Early Medieval West (Woodbridge: Boydell
Press, 1998).
Keeley, Lawrence H., War before Civilization (Oxford: Oxford University Press, 1996).
Kelly, Fergus, A Guide to Early Irish Law (Dublin: Dublin Institute for Advanced Studies,
1988).
Nı́ Dhonnchadha, Máirı́n, ‘The Law of Adomnán: A Translation’, in Thomas O’Loughlin
(ed.), Adomnán at Birr, AD 697: Essays in Commemoration of the Law of the Innocents
(Dublin: Four Courts Press, 2001), 53–68.
Richards, Martin et al., ‘Genetics and the Origins of the British Population’, Encyclopedia of
Life Sciences (forthcoming).
Scales, Len, ‘Bread, Cheese and Genocide: Imagining the Destruction of Peoples in
Medieval Western Europe’, History 92 (2007), 284–300.
Thomas, Mark G. et al., ‘Evidence for an Apartheid-Like Social Structure in Early
Anglo-Saxon England’, Proceedings of the Royal Society B (doi: 10.1098/rspb.2006.3527
(2006)).
Wallace-Hadrill, J. M., The Long-Haired Kings (London: Methuen, 1962).
Ward-Perkins, Bryan, The Fall of Rome and the End of Civilization (Oxford: Oxford
Winterbottom, Michael (ed.), Gildas: The Ruin of Britain and Other Documents (London/
Chichester: Phillimore, 1978).
Woolf, Alex, ‘Apartheid and Economics in Anglo-Saxon England’, in N. J. Higham (ed.),
The Britons in Anglo-Saxon England (Woodbridge: Boydell and Brewer, 2007), 115–29.
chapter 14
.............................................................................................
CENTRAL AND
L AT E M E D I EVA L
E U RO P E
.............................................................................................
len scales
The wicked shall be turned into hell, and all the nations that forget God
Psalm 9:17
N EW O R D E R : G E N O C I DA L F A N TA S I E S ,
E U RO P E A N T R A N S F O R M AT I O N S
................................................................................................................
For some, genocide is Europe’s peculiar gift to the world. Others insist that, far
from being atavistic, genocide is a crime of—as well as against—civilization.1
1 For a radical claim for the (premodern) European origins of genocide see Ward Churchill, A Little
Matter of Genocide: Holocaust and Denial in the Americas 1492 to the Present (San Francisco: City Lights
Books, 1997); for genocide and civilization: Michael Mann, The Dark Side of Democracy: Explaining
Ethnic Cleansing (Cambridge: Cambridge University Press, 2005), Ch. 1; Eric D. Weitz, ‘The Modernity
of Genocide: War, Race, and Revolution in the Twentieth Century’, in Robert Gellately and Ben
Kiernan (eds), The Specter of Genocide: Mass Murder in Historical Perspective (Cambridge: Cambridge
central and late medieval europe 281
Where, then, to place those centuries—between, approximately, the years 1000 and
1500—in which, if some very distinguished medieval historians are to be believed,
European civilization itself was formed? On the whole, sunny vistas still prevail
here—especially on that formative period between the eleventh and thirteenth
centuries, which is sometimes called the ‘high’ Middle Ages. Here was a Europe
of cathedrals and universities, of mind and spirit, in which the peaceful arts
prospered in growing towns, new orders of monks sought a foretaste of heaven
on earth, and the values of chivalry began to soften the violence of the Dark Age
warlord. ‘Humanism’ found a home already in the twelfth no less than the fifteenth
century. A ‘Europe of sensibility’ was being born, which ‘brooked no internal
boundaries’.2 Where to look for the dark side? Was there, from the present volume’s
perspective, a ‘dark side’ at all? The history of genocidal thought and action in
medieval Europe remains unwritten. Perhaps that is because there is none to write.
There is indeed a history to recount, and the formation of European literate
culture, or civilization, must be given a central part in it. So, too, however, must the
developing structures of power upon which civilizations rest; and the history of
power in medieval Europe is one of new divisions as well as emergent unities. An
eleventh-century illustrated gospel book, now in the Pierpont Morgan Library in
New York, shows Christ cleansing the Temple with a whip.3 It might stand as
emblematic for much of what was to follow in the period—and what follows in
these pages. The book was made for Countess Matilda of Tuscany, close ally of Pope
Gregory VII (1073–85), whose radical vision of a society transformed under papal
headship did so much to set Latin Europe on new foundations. The image signals
some salient characteristics of the dawning era: a preoccupation with purity and a
mission to purge the impure; a new decisiveness in setting boundaries and ruling
both in and out; a growing facility in discrimination—understood both as discern-
ment and disfavour; a quest for perfectible Christian communities; and a willing-
ness to seek these ends by force. In each of these developments, culture and power
were bound up together. They were elements in a ‘Europe of sensibility’ that
induced its inhabitants to build up, not tear down, walls. People grew increasingly
sensible of difference—of the threats which it seemed to pose, but also of their own
capacity to meet those perceived threats with violence. The earthly paradise was
not for all, and if some were to be gathered in, others must needs be cast out.
The Catholic Church under its reforming popes led the way in separating sheep
from goats, while also introducing into European life a new stress on the general,
the total, the all-encompassing. These same principles are also to be observed at
2 Jacques Le Goff, The Birth of Europe 400–1500, trans. Janet Lloyd (Oxford: Blackwell, 2005), 42;
and for an influential older account, R.W. Southern, The Making of the Middle Ages (London:
Hutchinson, 1953).
3 Malcolm Barber, The Two Cities: Medieval Europe 1050–1320 (London: Routledge, 1992), facing
192.
282 len scales
work in the development of political communities of various kinds in the centuries

after 1000—among them the European kingdoms and the Christian peoples
(gentes, nationes) to which it was widely believed these kingdoms gave constitu-
tional form.4 Peoples claimed a central part in the political assumptions of medi-
eval Western Europeans—just as they did in the Bible, from which the political
vocabulary of the Latin Middle Ages was to a large extent drawn.5 Medieval writers
believed that humanity was fundamentally divided on ethnic lines. Its social
building blocks were communities held together by ties of blood and shared origin,
though also displaying common cultural attributes, ranging from language and law
to elements such as dress, hairstyle, food, and manner of war. Common blood also
took visible form in stature, physique, and skin colour. That the ethnic unities
which medieval observers claimed to perceive were in fact cultural constructs
hardly needs stating—but then, they always are. There seems no reason, therefore,
to shrink from ascribing to medieval people fully fledged notions of ethnicity (we
might also say ‘race’) and—when these were linked to political titles—nation-
hood.6 For their articulation, medieval writers commanded a substantial and
flexible terminology, in Latin and the vernacular tongues, and the manner in
which they applied this is often strikingly close to modern usage. If it is also
frequently vague, question-begging, and contradictory, that, too, will scarcely
surprise the student of modern nationalist discourse. The period after the first
millennium is marked by the emergence in embryonic form of a Europe of
sovereign ‘nation-states’—political communities which combined claims to con-
stitutional autonomy and territorial integrity with a population base conceived as
ethnically homogeneous.7 By 1300 this process was, as we will see, entering a crucial
and in some ways ominous phase.
The world of Latin Europe in the central and late Middle Ages was an increas-
ingly interconnected one, shaped by the development of communications channels
and media more intensive and sophisticated than before. The size of surviving
document archives and the number of known manuscript books show a marked
increase for the centuries after the first millennium. To a heightened degree, this
was a textual, a black-and-white world, in which words themselves became instru-
ments of power—instruments which, with the expanding study of law, philosophy,
4 Susan Reynolds, Kingdoms and Communities in Western Europe 900–1300, 2nd edn (Oxford:
Oxford University Press, 1997), ch. 8.
5 For the biblical rhetoric of nation see Regina M. Schwartz, The Curse of Cain: The Violent Legacy
of Monotheism (Chicago: University of Chicago Press, 1997), esp. ch. 4.
6 For the medieval terminology see Robert Bartlett, ‘Medieval and Modern Concepts of Race and
Ethnicity’, Journal of Medieval and Early Modern Studies 31 (2001), 39–56, as well as Reynolds,
Kingdoms and Communities, 256–61.
7 For a single example: Colette Beaune, The Birth of an Ideology: Myths and Symbols of Nation in
Late-Medieval France, trans. Susan Ross Huston (Berkeley: University of California Press, 1991).
and theology, literate Europeans were wielding with a new confidence, precision,
and transformative effect.
These various changes, in religion, politics, and culture, were in their turn made
possible by the transformation of economic life and social organization in the
centuries following the first millennium.8 Europe’s population rose steadily: on
some estimates, between two- and threefold overall between 1000 and the Black
Death (1348–50). The population of France may have grown from around five to
perhaps fifteen million over the same period. All such figures carry a large margin
of uncertainty. Not in doubt, however, are the consequences of demographic
growth. Vast areas of land were cleared for agriculture to sustain the burgeoning
population, forests were felled, marshes drained, and thousands of new villages
established. Towns grew in size and number, proliferating across landscapes in
which urban life was previously largely unknown. In Westphalia, east of the Rhine
in northern Germany, only six towns were to be found before 1180: by 1350 there
were 138 (though mostly small). Trade routes, bearing a swelling traffic of people,
news, and rumour, as well as goods, now linked towns to their rural hinterlands,
and entire regions to each other, to the farthest ends of Europe, and to the world
beyond. Money, unfamiliar to many at the millennium, became ubiquitous in the
following centuries. And with its increased circulation came the powerful, disturb-
ing responses that, underpinned by the Church’s teachings and sanctions, it
evoked: fascination at its mysteriously transformative effect; and deep fear of that
same evident power to change (and taint?) all that it touched.
By the late Middle Ages, Europe’s most economically advanced regions were
heavily urbanized. In 1300, northern Italy boasted cities with populations in excess
of 100,000. By the same date, between fifteen and twenty per cent of English men
and women were living in towns. In the fifteenth century, the urban element in the
population of Flanders was over a third. Urban centres, and the values and
mentalities which they nurtured, now offered a potent challenge to the landed,
aristocratic order which had developed across medieval Europe—and a strong
attraction to those rural peasantries whose landlords aspired to hold them in
subjection. Tensions and conflicts between town and country, but also within the
towns themselves, were heightened after the mid-fourteenth century by the sharp
and protracted population fall which recurrent bouts of plague visited on much of
Europe. Despite these changes, however, the human landscape across the Conti-
nent was marked by sharp regional contrasts. While late medieval Brabant (in the
8 For what follows see: William Chester Jordan, Europe in the High Middle Ages (London: Alan
Lane, 2001), Ch. 1; Wim Blockmans and Peter Hoppenbrouwers, Introduction to Medieval Europe, 300–
1550, trans. Isola van den Hoven (London: Routledge, 2007), Ch. 5; N. J. G. Pounds, An Economic
History of Medieval Europe, 2nd edn (London: Longman, 1994), esp. 248–9; R. H. Britnell, Britain and
Ireland 1050–1530: Economy and Society (Oxford: Oxford University Press, 2004), 138; Peter Moraw,
Von offener Verfassung zu gestalteter Verdichtung: Das Reich im späteren Mittelalter 1250 bis 1490 (Berlin:
Propyläen, 1985), 49, 50.
284 len scales
north-west, bordering Flanders) sustained around forty-five inhabitants to every

square kilometre, an equivalent space in the Polish bishopric of Poznań was still
home to just two people. Europe’s margins long remained relatively empty—and
continued to beckon incomers, of diverse status and of more and less pacific intent,
from the more densely settled heartlands.
Medieval people thought genocidally. This was partly a consequence of their
disposition radically to simplify their world, its past and imagined future, into a
story of peoples. Because it was a dynamic story, in which some peoples rose and
prospered, others necessarily fell, and even disappeared. Vanishing ‘like the Avars’
was a familiar enough motif for it to be applied proverbially by a twelfth-century
Russian chronicler.9 And because earthly life in medieval accounts was filled with
violence, the fall of peoples was also violent. Illustration of this view may be found
in the numerous legendary accounts of the origins of European nations.10 Al-
though a handful of these date from the early Middle Ages (and the origin of the
genre itself is Roman), their number increases sharply from the eleventh and
twelfth centuries. It was not invariably through violence that, in these legends,
Europe’s peoples attained their medieval homelands; but bloodshed is a common
enough motif to indicate a characteristic mode of thought. From Bavaria to
Brittany and beyond, the forebears of medieval populations were repeatedly por-
trayed as immigrant warrior bands that had won their territories in a remote past
by destroying or expelling the indigenous peoples whom they discovered upon
arrival. The tale of the settlement of Britain, recounted by Geoffrey of Monmouth
in the twelfth century and repeated and elaborated thereafter, has the Trojan émigré
Brutus and his companions purging the rich isle of Albion of its aboriginal giants.
In the settlement myth of the continental Saxons, the Thuringians were present in
the Saxons’ land first and were removed, through a combination of killings and
expulsions, to make way for the newcomers. The Scottish origin legend, as it is
encountered in the early fourteenth century, has the ancient Scots occupying their
homeland ‘having first driven out the Britons and altogether destroyed the Picts’.11
Thereafter, the victors commonly renamed the conquered land from their own
leader or some reputed ancestor, in this way remaking its identity and obliterating
what had gone before. In names were concentrated a people’s status, titles, and
claims, and therefore, in one important sense, its very existence.
9 Quoted in Timothy Reuter, ‘Whose Race, Whose Ethnicity? Recent Medievalists’ Discussions of
Identity’, in idem, Medieval Polities and Modern Mentalities, ed. Janet L. Nelson (Cambridge:
Cambridge University Press, 2006), 102. The Avars were an obscure but historical central European
people, which disappeared for largely unexplained reasons at some time late in the first millennium.
10 Susan Reynolds, ‘Medieval Origines Gentium and the Community of the Realm’, History 68
(1983), 375–90; for ethnic destruction in the legends, Len Scales, ‘Bread, Cheese and Genocide:
Imagining the Destruction of Peoples in Medieval Western Europe’, History 92 (2007), 295.
11 A. A. M. Duncan, The Nation of Scots and the Declaration of Arbroath (1320) (London: Historical
Association, 1970), 34.
Such tales are revealing in a number of ways. For one thing, they reflect the
earlier roots of medieval beliefs about the relations between different peoples. One
of these was in the authority of Antiquity. The legend of Troy offered an example of
how, through ethnic destruction and dispersal, new peoples might come into
being. More pervasively, there was the template for ethnic history set out in the
Old Testament. This precedent had a special importance in portraying the migra-
tion and settlement of one particular, divinely favoured, people and the slaughter
and eviction of others as according with God’s plan. It was therefore not only
inevitable but right that some peoples should prevail and others face oblivion. The
idea had already received famous and influential formulation by Bede in the eighth
century, in his account of the triumph of the English over the native peoples of
Britain,12 but it was to attain new prominence and applicability after the millen-
nium, in an age which emphasized in novel ways the organically Christian quality
of political communities. Underlying all was a vision, inherited from the early
Middle Ages and perpetuated through epic and heroic tales, of the feuding and
mutual undoing of kindreds, clans, and, by natural extension, peoples in obscure
yet vividly evoked indigenous pasts. The aestheticization of violence found in
heroic literature, where the most elaborate metaphors were reserved for warriors,
their weapons, and the harm they wrought, was to live on in a new and heightened
ideological framework in the age of the Crusades.
It is no accident that legends recounting the settlement and overthrow of peoples
proliferated in Western Europe when they did. They belong to an age which
brought a new urgency and confidence to the tasks of explaining, labelling,
classifying, and distinguishing. Between the eleventh and the fourteenth centuries,
a great deal of broadly ethnographic lore, on subjects ranging from the reputed
origins of Islam to the nature and habitation of the fabled ‘monstrous races’ of
mankind, passed into circulation among literate Europeans: primarily, though not
exclusively, members of the clergy.13 The revived interest in classical learning
during the twelfth century both extended the quantity of knowledge available
about the world and its peoples and enriched the conceptual vocabulary for their
analysis. The fruits of these developments are evident in ethnographic writings
such as those of Gerald of Wales in the late twelfth century on the Celtic peoples of
the British Isles.14 The impulse to give account (though not necessarily seek
understanding) of other peoples was quickened by the Crusades against Turks
and other Muslims and against the pagan peoples of northern Europe. The rise of
the Mongols in the thirteenth century not only confronted Westerners with the
12 Bede’s Ecclesiastical History of the English People, ed. Bertram Colgrave and R. A. B. Mynors
(Oxford: Clarendon Press, 1969), 202–5.
13 John Block Friedman, The Monstrous Races in Medieval Art and Thought (Cambridge, MA:
Harvard University Press, 1981); John Tolan, Saracens: Islam in the Medieval European Imagination
(New York: Columbia University Press, 2002).
14 Robert Bartlett, Gerald of Wales, 1146–1223 (Oxford: Clarendon Press, 1982).
286 len scales
shock of a strange and terrifying ethnic Other but also, before long, opened up
routes through Asia along which some Europeans were able to attain directly an
expanded vision of a multifariously peopled world.
Because medieval people regarded ethnicity as an active and fundamental
historical presence, they characteristically simplified and exaggerated, sometimes
to the point of fictionality, its role in the events and developments with which
they linked it. Genocidal thought therefore occupies a more substantial place in
European history in this period than do genocidal acts, however broadly con-
ceived. And medieval ideas of what constituted a people’s destruction were very
broad. Writers seldom had in mind only systematic mass killing—though killing
was nearly always, directly or indirectly, a part of their picture, and they were
quite capable of imagining it being done systematically. In many cases, the
actions which they invoked are more akin to modern notions of ethnic cleansing
than the organized mass murder which some would now regard as the main
object of the term genocide. These actions, real or imagined, characteristically
envisaged the forcible creation of ethnically homogeneous landscapes through
processes involving various combinations of compulsory resettlement, eviction,
exile, and actual slaughter. Often, however, they clearly meant by a people’s
‘extermination’ or ‘elimination’—terms by no means rare in medieval writ-
ings—little more than its political disenfranchisement, the lopping-off (by
means more or less bloody) of its native ruling elite, or the suppression of its
means of common defence. Used rhetorically, the language of ethnic destruction
might refer to developments in themselves no more violent than subjection to
foreign rulers, the dominance of aliens at court, or the violation of alleged native
privileges, customs, or laws. None of this, however, should be interpreted as
meaning that medieval people were not serious in using such language, or that its
use need not be taken seriously by modern readers. To act in ways which
undermined a people’s standing within a competitive economy of peoples or
(much the same thing) to threaten the continuation of its name was potentially
to act genocidally. Such acts portended real and grave consequences. By stripping
a people of the tangible lineaments of common being—indeed, of its historical
and constitutional charters to be—they marked the start of a path which might
well lead quickly to enslavement, murder, and oblivion. This was not an unreal-
istic way of thinking; and the trajectory of destruction which it anticipates was
indeed mapped out by more than one ethnic group in this period.
It was not that medieval Europeans imagined the history of relations between
different peoples only as violent, or thought that good could never come of
their interaction. Some origin myths depict immigrant bands intermarrying
with indigenous populations, thereby bringing forth offspring who represented
a happy blending of the supposedly distinctive qualities of each people.
A collection of maxims from twelfth-century Hungary contends that a multieth-
nic kingdom is stronger than one resting only upon a single language and
law.15 Characteristically, however, medieval writers’ judgements on the conse-

quences of ethnic mixing were less favourable. For the fourteenth-century
chronicler Thomas Gray, the political fickleness of the English was a regrettable
consequence of their being ‘a mixture of diverse nations’. The mutual enmity
which seemed readily to arise when different peoples came together likewise
argued for their being kept apart. Such antipathies appeared for some to have an
inveterate quality. For the German chronicler Ekkehard, the coming-together of
German and Frankish knights on the First Crusade only brought to light their
‘natural enmity’.16
Ancient natural barriers between different European peoples were indeed being
undermined, in a more mobile and interconnected age, of which the Crusading
movement is a prime manifestation. But this seems only to have encouraged in
some quarters a heightened insistence on their naturalness and the need for their
maintenance. The thirteenth-century chronicler and cartographer Matthew Paris,
on a celebrated map of Britain, gave much prominence to the Hadrianic and
Antonine walls, with labels denoting the peoples they had ‘once’ kept apart.
Other ethnic barriers, meanwhile, had a more urgent and continuing relevance.
High-medieval world maps (mappae mundi) took pains to depict the great wall
which it was believed Alexander the Great had built in the Caucasus, to pen up the
cannibalistic peoples of Gog and Magog.17 Heroic and salutary feats of ethnic
engineering were for medieval observers a hallmark of the great ruler. Yet the
prophetic scheme of Christian history also disclosed to them that such barriers
would not stand for ever, for it was necessary that the genocidal scourge of the
‘unclean peoples’ at some time be unleashed upon Christendom.
R U L E R S , T E R R I TO R I E S , A N D T H E C H A N G I N G
MAP OF PEOPLES
................................................................................................................
The ethnographic turn and the quest for common origins were impelled by shifts
in the political map involving both the formation of political communities
within Europe and the dealings of European powers with their neighbours in
a wider world. In this climate, origin legends had an importance beyond mere
15 Benedykt Zientara, ‘Foreigners in Poland in the 10th–15th centuries: Their Role in the Opinion of
the Polish Medieval Community’, Acta Poloniae Historica 29 (1974), 5–6.
16 Sir Thomas Gray, Scalacronica 1272–1363, ed. and trans. Andy King, Publications of the Surtees
Society 209 (Woodbridge: Boydell, 2005), 95; Ludwig Schmugge, ‘Über “nationale” Vorurteile im
Mittelalter’, Deutsches Archiv für Erforschung des Mittelalters 38 (1982), 446 n 28.
17 P.D.A. Harvey, Medieval Maps (London: British Library, 1991), 27, 74.
288 len scales
antiquarianism. They were claims to power and constitutional independence:

manifestos, in short, for ethno-political survival where this seemed imperilled. In
an age of political expansion and consolidation, there were some who found in the
brutally dog-eat-dog world which they evoked, where the genocidally ruthless
prevailed, tracts for their own times. The Scottish origin myth was recounted in
response to English pretensions, in the age of Edward I (1272–1307), to suppress and
swallow up the hitherto-distinct kingdom of Scots. An imaginative vision of the
actual fate of those who went down before the onset of conquering kings was
unfolded early in the thirteenth century by Gervase of Tilbury, an Englishman
writing for the German emperor Otto IV. Gervase told of how Henry II (1154–89)
had intervened in Ireland, reordering social relations there, ‘though not without
the shedding of much blood of English and Britons’. A new civilization had thereby
dawned in the island, ‘once the foul Irish race had been expelled’.18 In Scotland, too,
‘a succession of holy kings’ had introduced beneficial changes, but only after ‘the
Scots, men of a foul way of life’ were driven out. Of course, no such mass expulsion
had occurred in either land. Yet, mistaken as they are, Gervase’s claims are highly
significant, indicating how developments even in the very recent past might be
understood—or rather, radically misunderstood—in terms of the self-same model
of violent ethnic replacement found in the origin myths. Socio-political reordering
meant also a new ethnic order. Re-shaking the kaleidoscope of peoples, moreover,
pertained especially to kings.
Far-reaching developments were indeed afoot; and at Europe’s expanding high-
medieval margins these took on a particular character. If contemporaries simplified
the role of ethnicity within them and exaggerated its importance and conse-
quences, their reactions nevertheless illuminate the experience of social and politi-
cal change for those caught up in it—and the material responses, which for some,
those changed facts of life seemed to invite. Between the eleventh and fourteenth
centuries, population growth, migration, and the transplantation to Europe’s
frontier zones of legal systems, technologies, and social and political institutions
previously developed at its Western continental core enabled the establishment of
societies which some historians have termed colonial.19 At Europe’s western and
eastern extremities, incoming warrior elites founded their dominance over indige-
nous communities both upon new, overmastering technologies of violence (nota-
bly those associated with the armoured heavy cavalryman) and upon self-justifying
doctrines of cultural superiority. New power relationships were therefore bound up
18 Robert Bartlett, England under the Norman and Angevin Kings 1075–1225 (Oxford: Clarendon
Press, 2000), 101.
19 For what follows see generally Robert Bartlett, The Making of Europe: Conquest, Colonization and
Cultural Change 950–1350 (Harmondsworth: Penguin, 1993). These developments have been placed
within a longer term view of the development of interethnic violence by Mark Levene, Genocide in the
Age of the Nation State, vol ii: The Rise of the West and the Coming of Genocide (London: I. B. Tauris,
2005), 29–34.
with changed relations between peoples and cultures—though the interconnec-

tions were in reality more complex and varied than the remarks of medieval writers
mostly allow us to suspect. Nevertheless, the expansion of certain peoples was
indeed in some respects at the expense of others. And it accorded with their
familiar habits of thought when contemporary and later observers diagnosed a
harsh fate for the losers. Such commentators were not always remote or ignorant.
Helmold of Bosau (d. after 1177), who chronicled the conversion of the Baltic Slavs
and the settlement of Germans and Flemings in their lands, could scarcely have
been better informed. Himself a priest at work on the Slav-German frontier when
the movement was at its height, he knew personally many of the main participants.
For Helmold too, immigration and social change inevitably meant ethnic replace-
ment:
Now . . . because God gave plentiful aid and victory to [Henry the Lion, duke of Saxony]
and to the other princes, the Slavs have been everywhere crushed and driven out. A people
strong and without number have come from the bounds of the ocean, and taken possession
of the territories of the Slavs. They have built cities and churches and have grown in riches
beyond all estimation.20
Helmold’s account of his times is not (here in contrast to Gervase of Tilbury’s)

fundamentally wrong: not only was the ethnic composition of his region changing:
coercion and, at least locally, forced displacements were indeed part of the story.
But in its bald ethnic essentialism it is certainly misleading. Nevertheless, it was the
first draft of a history that would harden to orthodoxy in the years that followed. In
the fifteenth century, the Church reformer and historiographer Dietrich of Niem,
himself a native Saxon, would shift part of the by-then mythologized story further
back in time. For Dietrich, the Frankish conqueror Charlemagne (768–814) had
already destroyed and expelled the Slavs from lower Saxony, ‘apart from a handful,
who down to the present day dwell in certain rural hamlets, mostly in marshy
places, though under perpetual servitude to the Saxons’.21
Yet nothing was fixed for ever. Since all medieval peoples tended in their own
estimations to rate as doughty warriors, there was no remaking of the ethnic map
that might not in its turn be undone by some heroic future act of collective
violence. As early as the ninth century, the History of the Britons (perhaps wrongly)
associated with the name of Nennius was holding out the prophetic hope that the
red British serpent would chase from the island the white of the Saxon interloper.
The same hope was still alive at the end of the twelfth when, according to Gerald of
Wales, the Welsh were drawing from the prophecies of Merlin the expectation
‘that both the nation (natio) and name of the [English] foreigners shall
20 The Chronicle of the Slavs by Helmold, Priest of Bosau, ed. and trans. Francis Joseph Tschan
(New York: Columbia University Press, 1935), 235–6.
21 Hermann Heimpel, Dietrich von Niem (c. 1340–1418) (Münster: Regensbergsche, 1932), 261.
290 len scales
be expunged’ from their land.22 Those who moved between the lines on Europe’s
colonial frontiers could turn to their advantage the hopes and fears of both camps.
As a fourteenth-century Irish poet explained to the first earl of Desmond:
In the [English] foreigners’ poems we promise that the Irish shall be driven from Ireland; in
the Irishmen’s poems we promise that the foreigner shall be routed across the sea.23
By the troubled fifteenth century, such fears, resting on wildly mythologized

recollections of high-medieval migrations and ethno-demographic shifts of the
kind we have glimpsed already, might become a basis for explicit agitation.
A manifesto of the Czech Hussites (1420) rallied its audience to arms against
their German neighbours with a warning that ‘just as they did to our tongue on
the Rhine, in Misnia, in Prussia, and drove it out, the same they intend to do to us
and to occupy the places of the banished.’24
The two or three centuries which follow the millennium can accurately be called
an age of kings. The number of European kingdoms itself grew substantially as also
did the capabilities of royal government and the claims advanced in the names of
rulers and peoples. The empire-building pretensions of kings, and of quasi-regal
figures like Henry the Lion of Saxony (d. 1195), lay at the heart of many of those
premonitions and allegations of violent ethnic change which recur in writings from
the time. The medieval habit of mapping ethnicity onto constitutional formations
meant not only the emergence of new ‘peoples’ within new realms but also, in a
climate of political and dynastic flux, the spectre of obliteration for others. The
viewpoint which conceived of political revolution as ethnic destruction was given
classic formulation by the chronicler Henry of Huntingdon, for whom the Norman
Conquest of England was an implementation of God’s judgement on the sinful
English, namely that they should cease to exist as a people. In twelfth-century
imagination, a people’s undoing did not need to involve mass killing. It might do,
however: another chronicler, Orderic Vitalis, tells of a plot which came to the
notice of King Stephen ‘to kill all the Normans on a fixed day and hand the
government of the kingdom over to the Scots.’25 Not all were prepared to accept
22 Nennius, British History and the Welsh Annals, ed. and trans. John Morris (London: Phillimore,
1980), 30–1; R. R. Davies, ‘The Peoples of Britain and Ireland 1100–1400 II: Names, Boundaries and
Regnal Solidarities’, TRHS 6th ser. 5 (1995), 4.
23 Katharine Simms, ‘Bards and Barons: The Anglo-Irish Aristocracy and the Native Culture’,
in Robert Bartlett and Angus MacKay (eds), Medieval Frontier Societies (Oxford: Clarendon Press,
1989), 181.
24 František Šmahel, ‘The Idea of the “Nation” in Hussite Bohemia’, Historica 16/17 (1968/9), 222.
25 Henry, Archdeacon of Huntingdon, Historia Anglorum: The History of the English People, ed. and
trans. Diana Greenway (Oxford: Clarendon Press, 1996), 412–13; and for Orderic, John Gillingham,
‘Henry of Huntingdon and the Twelfth-Century Revival of the English Nation’, in Simon Forde, Lesley
Johnson and Alan V. Murray (eds), Concepts of National Identity in the Middle Ages (Leeds: University
of Leeds, 1995), 76.
divine judgement as binding or view the settlement of 1066 as irreversible: inter-

ethnic violence might write its own, new histories.
At Europe’s margins, changing topographies of peoplehood and power came
together with the expanding resources, claims, and possibilities of kingship. By the
thirteenth century, rulers of the more highly developed kingdoms were possessed
of both the means and the will to reorder in some detail the ethnic landscapes of
their realms. Between the 1220s and 1240s, Frederick II of Hohenstaufen (d. 1250)
relocated the entire Muslim population of his Sicilian kingdom—perhaps num-
bering between 15,000 and 30,000 people—to the mainland town of Lucera. We
have not yet quite heard the last of the Muslims of Lucera. A more modest example
of ethno-social engineering illuminates what the king stood to gain. In 1295,
Edward I founded an English settler-borough beside his castle at Beaumaris,
deep in newly conquered Wales. The population of a native township on the site
was forcibly resettled twelve miles away. The pattern was repeated elsewhere in
Wales around the same time, amounting to a not insignificant re-shaking of the
pattern of peoples there.26 Economic advantage, defence and security, and the
visible display of dominance by a quasi-imperial conqueror might all alike recom-
mend a royal policy of local ethnic displacement.
When the king’s actions benefited his peers and native-born followers, all might
be well. But matters were not always so simple and the position of kings themselves
in relation to the settlement movements of the twelfth and thirteenth centuries
could prove troublingly ambivalent. On the one hand, it had been customary since
the earliest times for rulers to draw useful or prestigious foreigners to their realms
and courts—a practice whose extension the economic opportunities of the age
strongly favoured. On the other, the idea of kings as fathers to their (ideally,
ethnically cohesive) peoples was at this time finding increasingly powerful expres-
sion. There was nothing new about foreign favourites at court drawing the resent-
ment of native elites; but now, in European frontier regions, those high-status
interlopers were just one element within a larger, socially diverse immigration
process. In this climate, alien courtiers seemed in the eyes of some to be malevolent
harbingers of a radically new ethnic order. A Polish chronicler of the early four-
teenth century believed that Germans brought in to advise the young princes of
Głogów had incited them ‘to exterminate the entire Polish nation, both clergy and
laity, and especially the knights’.27 Such fancies took wing the more readily when a
ruler harnessed aliens to local development projects like Edward I’s in Wales.
Certain kings of Bohemia earned a dark name among the Czech-speaking political
classes for using Germans in this way. Přemysl Otakar II (1253–78) had resettled
26 Julie Anne Taylor, ‘Lucera Sarracenorum: A Muslim Colony in Medieval Christian Europe’,
Nottingham Medieval Studies 43 (1999), 110–25; R. R. Davies, ‘Colonial Wales’, Past and Present
65 (1974), 3–23.
27 Quoted in Zientara, ‘Foreigners in Poland’, 20.
292 len scales
with Germans the suburb beneath Prague castle, at the myth-laden heart of his
realm, on one account by expelling the native residents. Within little more than a
generation of his death, indigenous myth-making had turned the king’s unpatriot-
ic act into a full-scale plan to hand his kingdom over to the Germans. In the febrile
atmosphere of the Hussite agitations of the fifteenth century, the same and worse
was being reported of the cosmopolitian Charles IV of Luxembourg (1347–78).
Charles had ‘thought to settle the Czech land with the German race, of which he
himself came, and gradually root out the Czechs from it.’28
Such spectres drew substance both from an awareness of what kings by the late
Middle Ages were capable of doing and from some strikingly ethnocentric ideas
about what they might and should do. In the more governmentally sophisticated
realms, such perceptions and assumptions came together relatively early. A twelfth-
century estimation of the powers (and intentions) of the English monarchy is
glimpsed in Henry of Huntingdon’s depiction of the St Brice’s Day massacre of
1002, in which, according to the Anglo-Saxon Chronicle, ‘all the Danish men in
England’ had been killed at the behest of King Æthelred II.29 The sparseness of
contemporary evidence makes this instance of genocide by royal command hard to
judge; but historians have argued for its likely modest extent. Henry, however,
presents it as a coordinated act of government, underpinned by the systematic
dispatch of royal letters to every town in the realm. His picture, while unlikely to be
accurate, was certainly prophetic, and by the end of the thirteenth century some
royal bureaucracies were capable of targeting unpopular or controversial groups
with police actions of chilling scope and suddenness. ‘You have achieved in one day
what the Pharaohs of ancient Egypt failed to do,’ was one chronicler’s response to
Edward I’s expulsion of the Jews from England. For the poet Geoffrey of Paris, the
actions instituted by Philip IV of France (1285–1313) against Jews, Templars, and
others merged into a single vision of royal purge: ‘Jews, Templars and Christians j
Were caught and put in bonds, j And driven from one country to another.’30 The
Capetian imitators of Christ were cleansing the regnal Temple with a whip.
The late medieval wars that those formidable royal bureaucracies came to sustain
led some to perceive in the kingdoms of Europe, however unrealistically, instru-
ments of organized mass violence portending the outright ethnic obliteration of
their neighbours. Edward I’s celebrated claim of 1295, that the French king was
28 For Otakar see Jörg K. Hoensch, Přemysl Otakar II. von Böhmen: Der goldene König (Graz: Styria,
1989), 103; Kronika Neplachova, ed. J. Emler, Fontes rerum Bohemicarum 3 (Prague: Spolku
Historického v Praze, 1882), 476; for Charles IV, Šmahel, ‘The Idea of the “Nation” ’, 126 n 118.
29 Henry, Archdeacon of Huntingdon, Historia Anglorum, ed. and trans. Greenway, 340–1; The Anglo-
Saxon Chronicle, ed. Dorothy Whitelock (London: Eyre and Spottiswoode, 1961), 86.
30 Robin R. Mundill, England’s Jewish Solution: Experiment and Expulsion, 1262–1290 (Cambridge:
Cambridge University Press, 1998), 1; E. A. R. Brown, ‘Persona et Gesta: The Case of Philip the Fair’,
Viator 19 (1988), 237: ‘Christians’ is here doubtless being used as a catch-all term for the king’s political
opponents.
preparing an invasion in order to ‘delete’ the English tongue, was to be the first of
several such pieces of rhetorical scaremongering set down in the name of late
medieval English kings.31 In France, where some regions suffered the protracted
ravages of war, the late medieval vision of destruction attained Biblical propor-
tions: a late fourteenth-century tapestry portrays the English kings as crowned and
mounted Apocalyptic locusts, emerging from the bottomless pit to devour the land
(Revelation 9.1–11).32 As will shortly become clear, the totality of this vision of
destruction was characteristic of an important strand in medieval ethno-religious
thought, as was also its polarizing quality, pitting holy kings and chosen peoples
against personified evil—dehumanized, bestial, and unclean.
The notes of extremism, and the harnessing to violent acts of a violently ethno-
centric rhetoric, also infused other elements of late medieval life. The richness and
the character of the surviving sources may admittedly make the fourteenth and
fifteenth centuries appear to us more distinct from what had gone before than they
actually were. It is clear, however, that reports and predictions of large-scale
bloodshed were also underlain by new currents and tensions in social and religious
life. The prospect of organized mass killing was periodically in the air in various
contexts, not always related to ethnicity. Some radical religious groups believed a
general slaughter of the clergy to be imminent, while both lords and peasant
communities in various parts of Europe confronted from time to time the seemingly
impending prospect of fundamental and bloody social upheaval. Where members
of different ethnic groups were involved, movements and conflicts with complex
origins were easily seen as struggles between implacable rival peoples. Such percep-
tions came to the fore especially in the towns, whose spectacular growth in number
and size across much of Europe was one of the most enduring legacies of the post-
millennium period. Towns, with their complex and periodically acute social ten-
sions and power rivalries, penned up natives and aliens together in closely confined
spaces under conditions apt to render the outsider both visible and vulnerable.
The hothouse urban environment encouraged the invocation of ethnic divisions
to explain conflicts stemming in part from other discontents: economic, political,
devotional, local, or even professional. Alien university masters and students might
be forced out, as several hundred Germans were from Prague in 1409. Foreign
garrisons were massacred—as in Palermo in 1282 or Bruges in 1302.33 In various
31 William Stubbs, Select Charters and other Illustrations of English Constitutional History, 9th edn
(Oxford: Clarendon Press, 1913), 480.
32 See René Planchenault, L’Apocalypse d’Angers (Paris: Caisse Nationale des Monuments
Historiques, 1966).
33 Peter Moraw, ‘Das Mittelalter’, in F. Prinz (ed.), Deutsche Geschichte im Osten Europas: Böhmen
und Mähren (Berlin: Siedler, 1993), 154–5; Steven Runciman, The Sicilian Vespers: A History of the
Mediterranean World in the Later Thirteenth Century (Harmondsworth: Penguin, 1960), ch. 13;
J. F. Verbruggen, The Battle of the Golden Spurs (Courtrai, 11 July 1302: A Contribution to the
History of Flanders’ War of Liberation, 1297–1305, trans. David Richard Ferguson (Woodbridge:
Boydell, 2002), 25.
294 len scales
parts of Europe, barriers were now being raised, with towns making it harder for
foreigners to settle or practise crafts, or excluding them altogether. Bouts of
economic hardship and political instability were liable to be attended by outbreaks
of violence against resident foreign merchants, seamen, or artisans. Occasionally,
these attained major proportions and left substantial numbers dead, as did the
attacks on ‘Flemings’ in the south and east of England in 1381.34 The ‘great rising’ in
June of that year was a response to the perceived corruption and incapacity of
English royal government, particularly in its handling of the war with France. Its
background, however, lay in the disintegration of social hierarchies and the legal
relationships that underpinned them in the circumstances of sharp demographic
downturn, falling rents, and rising labour costs which followed the Black Death.
The anti-alien violence which accompanied the rising, particularly the killing of
many foreign artisans and merchants in London, illustrates clearly how urban
spaces might act to draw social and political discontents to a focus in interethnic
bloodshed.
Often, admittedly, such late medieval bloodshed was the outcome of mere riots
or drunken brawls. Increasingly, however, if contemporary reports are to be
believed, the voice of the people was to be heard on such occasions, clamouring
for a general destruction of the foreigner. ‘Death to the French!’ ‘Kill all Flemings!’
Those same reports tell also of the homely watchwords that conspirators coined to
affirm their solidarity and to tongue-tie and expose the alien in their midst.
Without question, our sources commonly exaggerate the destructiveness of such
disorders and simplify the role of ethnicity in them; but this exaggeration itself
reveals much about contemporary moods. And where popular disturbances at-
tracted more than local support, modest revisions to the ethnic map bequeathed by
high-medieval migration might indeed result. The revolt of Owain Glyn Dŵr in
early fifteenth-century Wales induced some inhabitants of the English boroughs to
take flight for England. Local ethnic displacement was more substantial where
social and political tensions were combined with religion, as they were in fifteenth-
century Bohemia, resulting in the purging of German communities from a number
of towns, including the capital.35
By this time, contending political and social groups were able to hurl at each
other stereotyped vocabularies of ethnocentric abuse of much radicalism and
menace. Compiling lists of the supposed good and (particularly) bad qualities of
different peoples was nothing new: as a literary exercise it went back to Antiquity.
The classification of peoples began, however, to shed its dry schoolbook quality
during the twelfth century, in context of a broader interest among educated
34 Rodney Hilton, Bond Men Made Free: Medieval Peasant Movements and the English Rising of 1381
(London: Temple Smith, 1973), 195–8.
35 R. R. Davies, The Revolt of Owain Glyn Dŵr (Oxford: Oxford University Press, 1995), 275; Moraw,
‘Das Mittelalter’, 163–4.
Europeans in the nature of humanity, coupled with a revived application of

Antique notions of the barbarian.36 These developments were taking place at a
time when colonization and state-building were lending such cultural pursuits
a keen political edge, and developments in the Church underpinning them with
a new facility in judging and condemning. Against this background, certain
peoples—particularly those which found themselves locked in violent competition
with their neighbours—came to be subjected to negative portrayals more eloquent
and absolute than before, backed by a new doctrinal authority, and bearing an
underlying call to destruction. Definitions of humanity emphasizing reason, culti-
vation, and order were deployed to portray enemies and subject populations as less
than fully human. The Irish, in English accounts, were ‘wild’, the Scots ‘bestial men’.
The Germans were ‘dog-heads’—semi-humans, evoking the monstrous beings
supposed to inhabit the earth’s arid margins. The lives of such ones were surely
cheaper than those of the fully human. The English in fourteenth-century Ireland
were accused of saying that it was no more a sin to kill an Irishman than a dog.37 To
monstrosity and sub-humanity were added associations of filth, pollution, and
parasitism. The Czechs were a ‘putrid odour’ to the Bohemian king Sigismund
(1419–37), while for a fourteenth-century Czech pamphleteer their German rivals
were ‘wolves in the fold, flies on the food, serpents in the bosom, harlots in the
house’.38
That a land should be cleansed of such pollutants appeared axiomatic; and a
luxuriating imagery of weeds and vermin indicated not only problems, but solu-
tions. ‘When [the Irish] fall into your hands pluck them all up by the root, as the
good gardener doth the nettle,’ urged a fourteenth-century Dublin notary.39 Talk of
beasts—the Irish as hares, for example—led on naturally to talk of hunting. The
task could seem the more urgent since beastliness and hybridity were no neutral
states. Monsters (as their supposed Latin cognates, monstro, moneo, made clear)
were signs and warnings: they spoke of sin. For Gerald of Wales, it was the
propensity of the Irish for bestial, incestuous, and other illicit sex that explained
the proliferation of malformed people in their land. But to some, the image of their
36 W. R. Jones, ‘The Image of the Barbarian in Medieval Europe’, Comparative Studies in Society and
History 13 (1971), 376–407.
37 Die Königsaaler Geschichts-Quellen mit den Zusätzen und der Fortsetzung des Domherrn Franz von
Prag, ed. J. Loserth, Fontes rerum Austriacarum: Oesterreichische Geschichtsquellen I.8 (Vienna:
Oesterreichische Akademie der Wissenschaften, 1875), 164 (dog-heads); Irish Historical Documents
1172–1922, ed. Edmund Curtis and R. B. McDowell (London: Methuen, 1943), 43.
38 Thomas A. Fudge, The Magnificent Ride: The First Reformation in Hussite Bohemia (Aldershot:
Ashgate, 1998), 269; Wilhelm Wostry, ‘Ein deutschfeindliches Pamphlet aus Böhmen aus dem 14.
Jahrhundert’, Mitteilungen des Vereins für Geschichte der Deutschen in Böhmen 53 (1915), 229.
39 James F. Lydon, ‘Nation and Race in Medieval Ireland’, in Simon Forde, Lesley Johnson, and
Alan V. Murray (eds.), Concepts of National Identity in the Middle Ages (Leeds: University of Leeds,
1995), 107. For the imagery of vermin and pollution in modern racist rhetoric see Zygmunt Bauman,
Modernity and the Holocaust (Ithaca, NY: Cornell University Press, 1989), 66–72.
296 len scales
neighbours spoke of things beyond sin: active malevolence, imminent danger. The
vocabulary of interethnic defamation thus merged at its extreme end with one of
absolute evil. Heretics were also routinely compared to parasites; the encroaching
Mongols too were a monster people, their wickedness encoded in misshapen
bodies. Bohemia’s Germans were for one Czech author an ‘accursed tribe’, as utterly
outside the fold as Muslims or Jews.40 To speak in this way of wayward and unclean
races was to invoke an Old Testament template—one which pointed towards the
extirpation of the wicked by the chosen under their divinely blessed leaders.41
P E O P L E S AC C U R S E D O F G O D : V I O L E N C E
A N D T H E C H R I S T I A N F RO N T I E R
................................................................................................................
It was when the language of ethnic distinction became overlaid with that of divine
favour and disfavour that fantasies of the destruction of peoples most often found a
measure of material fulfilment. The Reform movement in the eleventh-century
Church offered western Europeans not only the vision of a purified Christian
community and distinct criteria for inclusion and exclusion, but a strong impera-
tive to act for its creation. And action was from the outset conceived in part as a
cleansing struggle between peoples. A chronicle account of Pope Urban II’s speech
of 1095 inaugurating the First Crusade has him appeal to the Franks as a people
divinely chosen. The Muslim occupiers of the Holy Places, on the other hand, were
‘an accursed race, a race utterly alienated from God’.42 While the mass slaughter of
Muslims was not a specific aim of the Crusaders, a degree of territorial purification
was. Not only Muslims and Jews, but all who were not Latin Christians were
initially purged from Jerusalem following the city’s fall in 1099 and forbidden to
dwell there. While some native Christians were subsequently readmitted, the bar to
Jews and Muslims remained. And if outright mass killing proved to be but a partial
and temporary phenomenon, it certainly occurred. Not only were many of Jeru-
salem’s Muslims and Jews put to the sword when the city fell; large-scale slaughter
of non-Christians continued in other Palestinian cities upon their capture by the
40 Asa Simon Mittman, ‘The Other Close at Hand: Gerald of Wales and the “Marvels of the West” ’,
in Bettina Bildhauer and Robert Mills (eds), The Monstrous Middle Ages (Cardiff: University of Wales
Press, 2003), 97–112; Alfred Thomas, ‘Czech-German Relations as reflected in Old Czech literature’, in
Bartlett and MacKay (eds), Medieval Frontier Societies, 202.
41 See Schwarz, Curse of Cain, esp. ch. 2.
42 Robert of Reims, cited in Edward Peters (ed.), The First Crusade: The Chronicle of Fulcher
of Chartres and Other Source Materials, 2nd edn (Philadelphia: University of Pennsylvania Press,
1998), 27.
Latins throughout the first decade of the twelfth century. Moreover, in a manner
which was to be characteristic of crusading warfare, Western commentators cele-
brated the killing of non-Christians, and talked up the body count. When Jerusa-
lem fell, men rode through blood up to their knees and the bridle bits of their
horses. ‘Has anyone ever seen or heard of such a slaughter of the infidel race?’,
wondered one chronicler. ‘God alone knows the number for no one else does.’43
The same large and general acts, the same absolute distinctions between damned
and saved, were transferable to other Christian frontiers. Henry of Livonia tells how in
1227 ‘all the people of both sexes’ were baptized on Ösel (Saaremaa) in the Gulf of Riga,
following the island’s conquest by a crusading army. The Christian priests ‘watered the
nations by the font, and their faces with tears’. It was not to last. Another chronicler
explains how in 1260 the island’s inhabitants ‘broke away and left not a single Christian
alive in all their territories’. ‘Later’, he adds, ‘many of them were destroyed for doing
this.’44 Bloodshed followed quickly on the watering of the gentiles—and each was
conceived as a general act. The manner of thought was characteristic; so too the scale of
violence. ‘Kill them all; Truly, God will know his own!’ The words, attributed to a papal
legate at the massacre of the population of Béziers (1209) during the Albigensian
crusade, may never have been uttered; but the writer who set them down just a few
years afterwards captured the spirit of time and context.45 Crusaders shed blood with
hopeful hearts and easy consciences, and the chivalric norms curbing some of the
extremities of war at Europe’s core counted for much less on a religious frontier. The
population in some parts of Prussia may have fallen for a time by between twenty and
fifty per cent during the Teutonic Order’s destructive conquest—part holy war, part
strategic land-grab—in the thirteenth century.46 Crusading doctrine, to its propo-
nents, stiffened the soldier’s will against pragmatism and backsliding. It made of him
also an ethnic warrior. St Bernard of Clairvaux sought to rouse up the German nobility
for a crusade against the Baltic Slavs with the injunction that they should ‘take
vengeance on the [pagan] peoples and exterminate them from the land of our
Christian name’. There was to be no peace ‘until, with God’s aid, either the [heathen]
rite itself or the population [natio] has been destroyed.’47
43 Hans Eberhard Mayer, ‘Latins, Muslims and Greeks in the Latin Kingdom of Jerusalem’, History
63 (1978), 175–92; Petrus Tudebodus, Historia de Hierosolymitano Itinere, ed. John Hugh Hill and
Laurita L. Hill (Paris: Paul Geuthner, 1977), 142.
44 The Chronicle of Henry of Livonia, trans. and intro. by James A. Brundage (Madison: University
of Wisconsin Press, 1961), 245; Richard Fletcher, The Conversion of Europe: From Paganism to
Christianity 371–1386 AD (London: HarperCollins, 1997), 501.
45 Mark Gregory Pegg, A Most Holy War: The Albigensian Crusade and the Battle for Christendom
(Oxford: Oxford University Press, 2008), 77.
46 Hartmut Boockmann, Deutsche Geschichte im Osten Europas: Ostpreußen und Westpreußen
(Berlin: Siedler, 1992), 138.
47 Hans-Dietrich Kahl, ‘Crusading Eschatology as Seen by St Bernard in the Years 1146 to 1148’,
in Michael Gervers (ed.), The Second Crusade and the Cistercians (New York: St Martin’s Press, 1992),
42–3.
298 len scales
The resulting expedition—the ‘Wendish Crusade’ of 1147—predictably took a

different and far more limited course than that urged by the zealot Bernard.
Nevertheless, his conception of the Crusade as a radical contest of peoples is
affirmed by others. Helmold of Bosau writes as often of Slavs and Saxons as he
does of pagans and Christians. Most religious frontiers were also cultural and
ethnic ones, and religion itself readily conceived as an attribute of peoplehood,
reflecting shared character and identity. A story survives of how a group of
Livonians in the late twelfth century decided to wash off their recent baptism
with the waters of the River Dvina and thus send their Christianity back to the land
of the Saxons, whence it came. It was not therefore only for Bernard that religious
non-conformity dictated ethnic destruction. It was as a race of heretics that, for the
German Dominican Johann Falkenberg (writing c.1412), all Poles merited extermi-
nation.48 The doctrines of Church reformers and Crusade preachers were by the
late Middle Ages furnishing western Europeans with a powerful conception of
Christian community, within which the presence of alien groups looked increas-
ingly unacceptable.
The power of this doctrine to reshape the ethnic landscape became particularly
apparent with its growing assimilation to the political sphere. By the thirteenth
century, princes and their learned apologists were increasingly harnessing invoca-
tions of sacred community to notions of shared political allegiance and legal
doctrines exalting the power of monarchs within their consolidating realms.
Their exclusionary potential was realized earliest on the frontier. The fall of Seville
to the king of Castile in 1248 was followed by the wholesale expulsion of its Muslim
population. In thinly populated frontier zones, such changes were apt to prove only
temporary. At Europe’s core, however, they would be more secure. The last
Crusader strongholds in the Latin East fell in 1291; but at exactly that time rulers
were acting with a new resolve to build their own purified Christian holy lands on
European soil. It marked a significant new departure when the entire Muslim
population of Minorca was enslaved following the island’s fall to the Aragonese
in 1287. In 1300, the deeply devout Charles II of Naples (nephew of one royal saint
and father to another) sold into slavery the Muslims of Lucera. The city was
henceforth to be a Christian space, protected by the Virgin Mary.49 The Promised
Land had come home; but this only served to highlight the obligation upon those
48 Chronicle of Henry of Livonia, trans. and intro. Brundage, 34; Hartmut Boockmann, Johannes
Falkenberg, der Deutsche Orden und die polnische Politik (Göttingen: Vandenhoeck and Ruprecht,
1975), ch. 2.
49 Richard Fletcher, The Cross and the Crescent: Christianity and Islam from Muhammad to the
Reformation (London: Alan Lane, 2003), 112 (Seville); for Minorca and Lucera see David Abulafia,
‘Monarchs and Minorities in the Christian Western Mediterranean around 1300: Lucera and Its
Analogues’, in Scott Waugh and Peter Diehl (eds), Christendom and Its Discontents: Exclusion,
Persecution and Rebellion 1000–1500 (Cambridge: Cambridge University Press, 1996), 234–63.
(Christian) Israelites to whom it rightly belonged to destroy the unclean Canaanite

peoples whose presence still befouled it.
T H E T E M P L E A N D T H E W H I P : T H E J EWS
I N A C H R I S T I A N E U RO P E
................................................................................................................
A thirteenth-century mappa mundi portrays the figure of Christ as physically merged
with a created and peopled world.50 The centuries after the first millennium saw Christ
drawn closer to humanity, which in turn became conceivable as a Christian body—or,
politically, as a community of Christian peoples under their kings. Christ’s body and
blood lent legitimacy and affirmed common bonds; but they also became an increas-
ingly extensive charter for shedding the blood of those not of that body (-politic). The
communities of Jews, which by the eleventh century were already numerous and
widely scattered through Europe, would in the period which followed become the
object of the most virulently genocidal rhetoric and the most extensive and radical
violence to be suffered by any European ethnic group. And yet, unique though these
were in scope, intensity, and consequences, they also resemble in certain ways the
patterns of abuse and coercion which we have discovered underlying other medieval
interethnic conflicts. Latin Christians wrote about Jews in a (highly abusive) language
of race. For Peter the Venerable (d. 1156) they were a ‘wretched people’. The chronicler
Guibert of Nogent portrays Crusaders in 1096 wondering why they were making an
arduous journey to the East when ‘the Jews, of all races the worst foes of God, are before
our eyes.’51 Jews became the subject of a repertoire of defamatory stereotypes which
mirrored and extended those which Christian peoples applied to each other. Jews too
were likened to vermin. They were less than fully human, assimilated by physiognomy
(as also were Muslims) to the ‘monstrous races’; they were associated with demons and
with the baser animals: in bestiaries, the hyena was a Jew prototype. Jews acted as a
community—to Christian polemicists, an encompassing, malevolent, and conspira-
torial community. As a people, they were linked by blood to other accursed races: for
Matthew Paris, the Mongols were the ten Lost Tribes, whose destructive onset was in
concert with their kinsmen within Christendom.52
50 Harvey, Medieval Maps, 28.

51 Jeremy Cohen, ‘Christian Theology and Anti-Jewish Violence in the Middle Ages: Connections
and Disjunctions’, in Anna Sapir Abulafia (ed.), Religious Violence between Christians and Jews:
Medieval Roots, Modern Perspectives (Basingstoke: Palgrave, 2002), 47, 49.
52 Debra Higgs Strickland, Saracens, Demons and Jews: Making Monsters in Medieval Art
(Princeton: Princeton University Press, 2003), 133–6, 147–8; Sophia Menache, ‘Tartars, Jews,
Saracens and the Jewish-Mongol “Plot” of 1241 ’, History 81 (1996), 319–42.
300 len scales
They came, moreover, increasingly to be associated, by a variety of commenta-

tors, with general schemes of pollution and violence, inviting in turn violent
general responses. In 1321, wild rumours associated the Jews of Languedoc in a
well-poisoning conspiracy with lepers (who wished all people to be leprous) and
with outside Muslim powers.53 With the high-medieval humanization of the figure
of Christ and the Christianization of human communities went a growing tenden-
cy to depict Jews as not only deicidal but genocidal. Their supposed practice of
mistreating consecrated communion hosts was not only a re-enactment of the
Crucifixion on Christ’s miraculous body; it was an attack on the unity of Christ
and his people.
There were still some within Christian Europe who came to the Jews’ defence:
distinguished churchmen, popes among them, who reiterated the traditional
defence of the Jews’ presence in Christendom as divinely willed and spoke against
the wilder anti-Judaic fantasies. Over time, however, their arguments proved less
and less able to prevail against rival currents, also drawing on Christian justifica-
tions, which urged the Jews’ exclusion or destruction. It was from fear of their
‘extermination’, explained Pope Innocent IV in 1247, that Jews in Germany had
sought his aid.54 The fear was by this date an increasingly realistic one.
The earliest major attacks on Jewish communities in Europe had been during the
First Crusade, and subsequent expeditions were also attended by localized bouts of
bloodletting. However, it was from the later thirteenth century that anti-Jewish
violence in Europe took on new proportions. It is estimated that several thousand
may have perished in the agitations which convulsed parts of Germany in 1298,
inspired by host-desecration charges. However, bloodshed of a quite new extent
and thoroughness was attained in the massacres which heralded the arrival of the
Black Death in central Europe during the years 1348–50. In town after town, well-
poisoning rumours became a pretext for the systematic killing of entire Jewish
communities. Nearly a thousand may have died in the German town of Erfurt
alone. In some places—Basel, Strasbourg, Constance—the Jews were forced into
specially constructed houses to be burned. A contemporary chronicler claimed that
it took six days to burn Strasbourg’s Jews on account of their number. ‘And I could
believe’, he mused, ‘that the end of the Hebrews had come.’55
Elsewhere in Europe, princes had already by this date taken steps of their own to
reaffirm the Christian character of their realms, by means of the mass expulsion of
53 Malcolm Barber, ‘Lepers, Jews and Moslems: The Plot to Overthrow Christendom in 1321’,
History 66 (1981), 1–17.
54 Edward A. Synan, The Popes and the Jews in the Middle Ages (New York: Collier-Macmillan,
1965), 114–15.
55 Michael Toch, Die Juden im mittelalterlichen Reich (Munich: Oldenbourg, 1998), 60 (for 1298);
František Graus, Pest, Geißler, Judenmorde: Das 14. Jahrhundert als Krisenzeit (Göttingen:
Vandenhoeck and Ruprecht, 1987), 168–214; Rosemary Horrox (ed. and trans.), The Black Death
(Manchester: Manchester University Press, 1993), 208–10.
their Jewish populations. Whereas Church reformers and heretics had pursued
their rival visions of a purged and purified Christian heaven on earth, Catholic
monarchs now laid more limited plans for building ‘heaven in one country’.
England’s Jews were forced out by Edward I in 1290. The Jews of France, who
had suffered temporary expulsions and other oppressions under previous kings,
were systematically driven from the realm by Philip IV in 1306. As many as 100,000
people may have been compelled to leave.56 The pattern was repeated in other late
medieval realms, down to the mass expulsion of the large Jewish populations of the
Iberian kingdoms at the end of the fifteenth century. In part, princes now acted in
this way because, as we have seen, they had attained the governmental means to do
so. More importantly, however, some rulers had come by this time to view the
purification of their realms as a sacred duty. There was henceforth to be just one—
organically Christian—Chosen People, rightfully occupying the holy soil of its own
sovereign kingdom under its rightful and anointed king. Read as a constitutional
text, the Bible itself had come to represent, for the peoples of medieval Europe
and for their rulers, the most powerful, and fatally empowering, collective origin
myth of all.
R U L E D O U T : E XC LU S I O NA RY I M P U L S E S
A N D T H E M A K I N G O F E U RO P E
................................................................................................................
The existence of peoples in Europe in the central and later Middle Ages reflected
the facts of power: for contemporaries, ethnic communities were axiomatically
political ones. To imperil a people’s political status, as embodied in its privileges,
laws, common institutions, in the power of its members to act politically, and more
numinously in their sense of shared prestige and distinctiveness within a world of
peoples, was to act genocidally. In a period which saw extensive changes to the
European political map, the spectre of such acts never seemed far remote. Where
the interactions of different peoples were most intensive, stress-laden, and ideo-
logically and politically charged—on the frontier, at the courts of princes, or in the
great towns—acts of ethnic destruction were anticipated, and in some quarters
sought, most keenly. Medieval people were prone to simplify and exaggerate the
role of ethnicity in the conflicts of their day. Consequently, the destruction of
peoples—even in their own broad understanding of that phenomenon—did not
occur nearly as often as they expected. Despite this, their diagnosis of its causality
56 William C. Jordan, The French Monarchy and the Jews: From Philip Augustus to the Last Capetians
(Philadelphia: University of Pennsylvania Press, 1989), 202.
302 len scales
and likely course was fundamentally correct. When entire populations were indeed
subjected to systematic violence, enslavement, or eviction, those acts were usually
preceded—often over a protracted period—by other seemingly lesser ones, which
served to strip the group of its political and legal independence: its autonomous
right to be. ‘The king’s Jews’ necessarily waited on the king’s will; and his will was
by the late Middle Ages taking, across much of Europe, an increasingly ethnocen-
tric turn.
The pattern was not bound to be repeated everywhere, of course. Political and
legal marginalization did not have to lead to collective oblivion: it did not do so,
for example, in late medieval Ireland or post-conquest Wales. Outright ethnic
destruction was most likely to occur where political subjugation was reinforced by
fundamental religious difference. Pagans, Muslims, and Jews, but also, in an age
of sharpened conceptions of religious orthodoxy, adherents of (for their oppo-
nents) false forms of Christianity, were singled out for extreme solutions. For the
rest, the history of this long period is partly one of how, through more intensive
and precisely defined interactions, different imagined ethnic groups evolved
forms of coexistence and mutual accommodation. Nevertheless, Europeans were
also by the end of the Middle Ages more practised and accomplished ethnic
discriminators and excluders, in thought and deed, than they had been in earlier
times. Their world was one of more sharply defined and cohesive communities, in
which the alien was more conspicuous, more readily and vehemently named, and
less easily accommodated: a world which sought unities and was keenly sensible
of the problems of coping with multiplicity. Their power fundamentally to rule
out, and conviction that such ruling-out was needful, ultimately derived in large
part from those same currents of social change, religious reform, and intellectual
renewal which had also provided a foundation for the period’s most startling
cultural achievements and advances—as some would say, for European civiliza-
tion itself.
SELECT BIBLIOGRAPHY
Bartlett, Robert, Gerald of Wales, 1146–1223 (Oxford: Clarendon Press, 1982).
—— The Making of Europe: Conquest, Colonization and Cultural Change 950–1350 (Har-
mondsworth: Penguin, 1993).
—— and Angus MacKay (eds), Medieval Frontier Societies (Oxford: Clarendon Press, 1989).
Fletcher, Richard, The Conversion of Europe: From Paganism to Christianity 371–1386 ad
(London: HarperCollins, 1997).
—— The Cross and the Crescent: Christianity and Islam from Muhammad to the Reformation
(London: Alan Lane, 2003).
Forde, Simon, Lesley Johnson, and Alan V. Murray (eds), Concepts of National Identity in
the Middle Ages (Leeds: University of Leeds, 1995).
Friedman, John Block, The Monstrous Races in Medieval Art and Thought (Cambridge MA:
Gillingham, John, The English in the Twelfth Century: Imperialism, National Identity and
Political Values (Woodbridge: Boydell, 2000).
Moore, R. I., The Formation of a Persecuting Society: Power and Deviance in Western Europe,
950–1250 (Oxford: Blackwell, 1987).
Reynolds, Susan, Kingdoms and Communities in Western Europe 900–1300, 2nd edn (Oxford:
Strickland, Debra Higgs, Saracens, Demons and Jews: Making Monsters in Medieval Art
(Princeton: Princeton University Press, 2003).
chapter 15
.............................................................................................
C O LO N I A L L AT I N
AMERICA
.............................................................................................
nicholas a. robins
................................................................................................................
The conquest of Latin America resulted in the deaths of tens of millions of
individuals, primarily as a result of disease and forced relocation into more con-
centrated settlements, as well as through exterminatory attacks on those who
resisted Iberian domination. Severe exploitation aggravated the process through
overwork, nutritional deficits, and reduced resistance to illnesses generally. Parallel-
ing this process were concerted efforts to destroy the religious and cultural fabric of
native societies through the systematic destruction of sacred objects, the death of
indigenous religious leaders, and the prohibition of native rites. Indian rebellions
against colonial rule could also take the form of subaltern genocides in which the
oppressed seek the extermination of their overlords. The study of genocide in
colonial Latin America challenges many prevailing conceptualizations in genocide
studies. Among these are the construction of intent in defining genocidal outcomes,
the role of the state in genocides, and the relative military capacities of victim
groups.
Understanding the incidence and breadth of genocide in colonial Latin America
is conditioned by several factors, the two most salient being controversies over the
size of pre-conquest populations and the role of intent. Related to these are the
impact of diseases on the native population, relative to other factors such as forced
colonial latin america 305
labour and migration, and conflicting conceptualizations of genocide. These fac-

tors are explored below within the context of colonial Latin American genocides.1
The work of Raphael Lemkin, who coined the term genocide in Axis Rule in
Occupied Europe, is generally associated with the Jewish Holocaust. What is less
widely recognized, however, is that Lemkin developed many of his ideas in the
wider context of colonialism. Indeed, colonialism was inherent to the concept as
developed by Lemkin, who wrote that ‘Genocide has two phases: one, destruction
of the national pattern of the oppressed group: the other, the imposition of the
national pattern of the oppressor’ that may be done through the ‘colonization of
the area by the oppressor’s own nationals’.2 Lemkin recognized the vital role of
culture in the maintenance and perpetuation of individual and collective identity,
and viewed deliberate cultural destruction as a form of genocide.’3
In referring to the ‘methods’ of what he called ‘cultural’ genocide, Lemkin
included ‘desecration and destruction of cultural symbols, destruction of cultural
leadership, destruction of cultural centers, prohibition of cultural activities, [and]
forceful conversion’. Other means included ‘physical’ genocide, such as ‘massacre
and mutilation, deprivation of livelihood, slavery’, and ‘biological’ genocide which
included ‘separation of families, sterilization, destruction of foetus [sic]’.4 Like
many who wrote after him, Lemkin’s understanding of genocide in Latin America
was informed by romantic notions of pre-Hispanic civilizations (such as the Aztecs
having a ‘fairly democratic government’) and heavy reliance on the writings of the
Dominican friar Bartolomé de Las Casas.5
D E M O G R A P H I C I M P LO S I O N AND GENOCIDE
................................................................................................................
The conquest and ensuing domination of Latin America was intimately linked to
the spread of pathogens to which the native populations in the region initially had
little or no resistance. Beginning with the arrival of Columbus in the New World in
1492, the Spanish conquistadors established their dominance first in Hispaniola
1 For a wider framing of the definitional debate and their relation to colonial and modern
genocides, see A. Dirk Moses, ‘Conceptual Blockages and Definitional Dilemmas in the “Racial
Century”: Genocides of Indigenous Peoples and the Holocaust’, Patterns of Prejudice 36:4 (2002), 7–36.
2 Michael A. McDonnell and A. Dirk Moses, ‘Raphael Lemkin as Historian of Genocide in the
Americas’, Journal of Genocide Research 7:4 (2005), 501; Raphael Lemkin, Axis Rule in Occupied Europe
(Washington, DC: Carnegie Endowment for International Peace, 1944), 79.
3 Raphael Lemkin, ‘The Concept of Genocide in Anthropology’, cited in McDonnell and Moses,
‘Raphael Lemkin as Historian’, 514.
4 McDonnell and Moses, ‘Raphael Lemkin as Historian’, 504.
5 Ibid. 505, 515.
306 nicholas a. robins
and Cuba before successively extending their control to what is today Panama in
1510, to Mexico in 1521 and subsequently to the rest of Central America. From there,
they advanced southward to present day Peru in 1532, and thence to contemporary
Bolivia, Ecuador, and much of South America.
Once released in the region, disease often spread in advance of the conquerors,
and facilitated their conquest. In the case of Peru, it appears that smallpox had
spread south from Panama in 1524, eight years before the arrival of Pizarro, killing
the Inca Huayna Capac and much of the population and setting off the civil war
which facilitated the Spanish conquest of the region. In the case of Brazil, after the
Portuguese claimed the region in 1500 and an initial focus on brazilwood extrac-
tion, sugar production relying heavily on slaves imported from Africa came to
dominate.6
Estimates of the pre-conquest population in Latin America tend to vary widely, a
result of a virtual absence of pre-contact baseline population data, inaccurate
censuses in the colonial period, and differing methodologies. In such situations,
errors can quickly become compounded. To the extent that there were native
population records on the eve of conquest, with very few exceptions they did not
survive, either being lost, or as in many cases, deliberately and systematically
destroyed by the Spaniards.
Once established in the region, the colonists did conduct censuses to determine
tribute levies; however, challenges in logistics and coordination took their toll on
accuracy. In addition, most surveys were limited to individuals of tribute age
(males between eighteen and fifty years old), and as a result any extrapolation
requires an estimate of average family size. Royal authorities were reluctant to
conduct censuses, not only due to their cost and complexities, but because they
recognized that, until the eighteenth century, the native population had been
declining. Hence, an accurate census would in many cases result in a reduction
of tribute and other levies on native communities, and this colonial officials clearly
wanted to avoid.7
Further complicating modern efforts to determine pre-conquest populations are
methodological differences that can yield widely varying results. Approaches that
examine the ‘carrying capacity’ of the land, the native crops grown on it, and the
caloric intake of the indigenes are especially useful at a micro level, but, as with
6 Henry F. Dobyns, ‘An Outline of Andean Epidemic History to 1720’, Bulletin of the History of
Medicine 37:6 (1963), 494, 497; Heraclio Bonilla, ‘1492 y la población indı́gena de los Andes’, in
Heraclio Bonilla, Robin Blackburn, et al. (eds), Los conquistados: 1492 y la población indı́gena de las
Americas (Bogotá: Tercer Mundo Editores/ Facultad Latinoamericana de Ciencias Sociales, 1992), 106;
Nicolás Sánchez-Albornoz, The Population of Latin America: A History, trans. W. A. R. Richardson
(Berkeley: University of California Press, 1974), 61.
7 Nicholas A. Robins, Priest-Indian Conflict in Upper Peru: The Generation of Rebellion, 1750–1780
(Syracuse: Syracuse University Press, 2007), 17.
other approaches, extrapolation can lead to significant inaccuracies.8 Archaeolog-

ical approaches face many challenges, not the least of which is data literally
concealed beneath the ground.9 When there is pre-conquest population data for
an area, and a subsequent colonial census, one can generate a depopulation ratio.
While valuable, such an approach can be limited by differing or uncertain age
ranges for pre-conquest tribute payers and generally small sample sizes.10
More widely applicable are approaches that examine mortality from disease in a
known context, and then apply that to the area under study. For example, if
smallpox is known to kill thirty per cent of a population without immunity in
one area, this ratio can be applied elsewhere. Altitude, and the cooler climate
associated with it, as well as population densities play important roles in the spread
of diseases and are factors which must be taken into account. The picture is further
clouded in the not uncommon situation where there is uncertainty concerning
exactly what disease was afflicting the indigenes. Despite these shortcomings, this
approach yields more reliable results than those based on carrying capacity or
archeology.11 The final and perhaps most accurate approach is that based on post-
conquest census projections, where there are two comparable censuses and a rate of
depopulation can be calculated and retroactively applied.12
As the preceding indicates, one must approach estimates of pre-contact popula-
tions with considerable caution. Despite the differing approaches and their inher-
ent limitations, there is a general convergence among historical demographers
indicating that between 1492 and 1600, approximately ninety per cent or more of
the native population in Latin America perished.13 In central Mexico, for example,
utilizing the carrying capacity approach and depopulation ratios, Woodrow Borah
and Sherburne Cook estimated that the population was 25,200,000 in 1518, but by
1605 it had fallen to 1,075,000, a drop of approximately ninety-six per cent.14 In a
critique of their research and utilizing the carrying capacity approach, Brooks
8 Noble David Cook, Demographic Collapse: Indian Peru, 1520–1620 (Cambridge: Cambridge
University Press, 1981), 18, 28.
9 For a discussion of the broader limitations of this approach, see ibid. 30–40.
10 Ibid. 41–2, 54.
11 Ibid. 59, 62, 64.
12 Ibid. 107.
13 Noble David Cook, Born to Die: Disease and New World Conquest, 1492–1650 (Cambridge:
Cambridge University Press, 1998), 206; C. T. Smith, ‘Depopulation of the Central Andes in the
16th Century’, Current Anthropology 11:4–5 (1970), 459; Sherburne F. Cook and Woodrow Borah, Essays
in Population History: Mexico and California, vol iii (Berkeley: University of California Press, 1979), 1.
This discussion focuses on Mexico and Peru; however, for other regions in Latin America, see Cook,
Born to Die; Sánchez-Albornoz, Population of Latin America; Dobyns, ‘Outline of Andean Epidemic
History’; and Alfred Crosby, The Columbian Exchange: Biological and Cultural Consequences of 1492
(Westport, CT: Greenwood Press, 1972).
14 Woodrow Borah and Sherburne Cook, The Aboriginal Population of Central Mexico on the Eve of
the Spanish Conquest (Berkeley: University of California Press, 1963), 4–5, 88–90; Cook and Borah,
Essays in Population History, 1.
argues that the pre-contact population was 5,000,000, which yields a decline of a
not much less catastrophic eighty per cent.15
In the Andes, smallpox may have killed one-half of the population before the
Spanish conquest of the region.16 In the area of present day Peru, N. D. Cook
utilizes census projections to conclude that the population in 1530 was approxi-
mately 9,000,000; however, by 1580 only between 600,000 and 1,000,000 indigenes
remained, a collapse of approximately ninety-three per cent. As in Mexico, he notes
that disease and mortality rates were higher on the coast, where populations were
more densely concentrated and temperatures higher. Indeed, on the Peruvian coast
the Indians were ‘almost completely wiped out’.17
Despite differing methodologies, their respective limitations, and the differences
they yield, there is little debate over the scale of the demographic implosion which
resulted from the conquest of what came to be called Latin America. It does,
however, raise questions over the relationship between native exploitation and
the population collapse. Furthermore, and of significant importance for the field of
genocide studies, it raises the issue of intent. To what varying extents did the
colonists want to dominate and exploit the indigenes or exterminate them?
F O RC E D L A B O U R , I N T E N T, A N D
THE DEMOGRAPHIC DISASTER
................................................................................................................
While historical demographers have focused on the agency of disease in the
massive death associated with the conquest, others, while recognizing the role of
disease, stress the brutality and dislocation associated with various forced labour
systems, and their concomitant impact on the native population, impacts which
were exacerbated by hopelessness and suicide.18 Disease and exploitation were
clearly interrelated. Deliberate, state-mandated policies forcing natives to live in
concentrated settlements, such as missions and Spanish-style towns, and other
policies that subjected them to harsh and abusive labour conditions had a tragically
synergistic effect in promoting depopulation. Overwork reduced the ability of the
15 Francis Brooks, ‘Revising the Conquest of Mexico: Smallpox, Sources and Populations’, Journal
of Interdisciplinary History 24:1 (1993), 1–2, 5, 7.
16 Dobyns, ‘Outline of Andean Epidemic History’, 494, 497; Bonilla, ‘1492 y la población indı́gena
de los Andes’, 106; Sánchez-Albornoz, Population of Latin America, 61.
17 Cook, Demographic Collapse, 114.
18 See, for example, David Stannard, American Holocaust: Columbus and the Conquest of the New
World (New York: Oxford University Press, 1992), and Ward Churchill, A Little Matter of Genocide:
Holocaust and Denial in the Americas 1492 to the Present (San Francisco: City Lights Books, 1997).
body to fight infections, lowered life expectancy, and sometimes killed on the spot,
and diseases spread much more rapidly in close quarters. Forcing indigenes to
labour in climates to which they were not accustomed, especially in the hot lowlands
where disease was more prevalent, was often a death sentence. Although mortality
among populations with no resistance to new diseases would have been extremely
high in non-exploitative conditions, exploitation unquestionably aggravated the
situation. In so weakening the native population, and with it their ability to resist
colonial rule, it also served the imperial objective perpetuating domination.
A brief review of colonial labour systems illustrates these dynamics. Although
Indians could only be technically enslaved, and thus sold as such, when they
disavowed the authority of the Spanish king, in practice, however, there emerged
an array of forced labour systems which resulted in de facto Indian slavery. Among
such systems was the encomienda system, in which a Spaniard, usually but not
always a conquistador or his descendant, was granted rights to Indian tribute in a
specific area. This was not a grant of land, but of labour, and the beneficiary, or
encomendero, undertook the obligation of ‘Christianizing’ the Indians. Often this
Christianizing was nothing more than a quick, mass baptism devoid of any previous
catechism. Fearing the emergence of a New World nobility, the Spanish crown
sought to eliminate this system with the implementation of the New Laws of 1542.19
Replacing the encomienda system was that of the repartimiento, in which royal
authorities became responsible for the assignment of Indian labour to the colo-
nists. Other systems included bonding the native to an agricultural estate through
inheritable debt. The Crown also instituted systems of corvée labour in which
rotating groups of Indians were forced to work for a limited term on projects of
public benefit under specific conditions, which were, however, consistently ignored
by the colonists. Other systems included penal labour and wage labour, the latter of
which often became a form of debt peonage.20
There is no dispute concerning the coercive nature of Indian labour systems in
colonial Latin America, and the fact that many died as a result of them. They were
designed to induce the natives to produce a surplus and to forcibly extract it from
them. Whatever the specific system, it was rigorously reinforced by recourse to the
lash, pillory, prison, and taking family members as hostages for debts, among other
means. The larger question, however, is the impact that such systems had on the
total native population.21
Despite the clear role of disease in the demographic disaster, some authors seek
to establish the intention of the colonists to exterminate native populations by
19 C. H. Haring, The Spanish Empire in America (New York: Harbinger, 1947), 40–41, 51.
20 Peter Bakewell, Miners of the Red Mountain: Indian Labor in Potosı́, 1545–1650 (Albuquerque:
University of New Mexico Press, 1984), 69; Donald Wiedner, ‘Forced Labor in Colonial Peru’, The
Americas 16:4 (1960), 377; Haring, The Spanish Empire in America, 15.
21 Robins, Priest-Indian Conflict, 69–76.
focusing on exploitation or purposeful extermination as the leading cause of

Indian depopulation. More polemical than scientific, their approach borders on
dismissing the effects of disease, averring for example that labour abuses in the
Andes ‘precipitated the demographic crisis’ or confuse mercury mining with silver
mining.22 Others, while recognizing that disease played a role, focus on the abuses
committed against natives to argue that the Spanish engaged in a ‘deliberate racist
purge’ or policy of ‘systematic extermination’ against them.23 There is no question
as to the brutality of the system implemented by the Spanish, but by minimizing
the role of disease one gets a distorted picture of what actually happened.
The debate concerning the relative weight of disease versus deliberate abuses in
the depopulation equation is important as it concerns intent. In the case of the
Spanish conquest of the Americas, the invaders sought riches and the means to
extract them. The domination of the native population was central to this strategy,
black slaves generally only being introduced in areas where the native population
had been exterminated, such as Hispaniola and Cuba, or could not otherwise
survive. The survival of the empire and its extractive systems depended heavily
upon the survival of the indigenes, as surplus producers, and paradoxically, upon
their continuing weakness so as to forestall uprisings. While the colonial enterprise
demanded natives for labour, the multifaceted extractive systems that coerced this
labour also maintained them in a position of physical weakness. Although there
were numerous colonial policies that were genocidal and will be discussed below,
the Spanish sovereigns sought the domination of the natives, not their physical
extinction, and issued a host of edicts, routinely ignored in the colony, in an effort
to preserve the labour force. As Latin American population history has made clear,
however, an absence of intention to kill does not mean an absence of massive death.
Although genocidal intent, and even knowledge, were lacking in terms of the
spread of pathogens, the same cannot be said concerning armed resistance to
conquest. Such resistance occurred at two levels, the establishment of the Spanish
in the region, and the expansion of the frontier, both of which in Lemkin’s view
would be considered ‘physical genocide’.24 In the former case, the invaders were
vastly outnumbered, and were quick to resort to massacres both as a means to
achieve immediate military victory and to instill terror and panic in the wider
population.25 Such efforts led to the deliberate and systematic slaughter of non-
combatants, and the enslavement and usually relocation of those who survived.
Once established in the region, early colonial authorities had sizeable numbers of
soldiers who were often idle and the cause of disturbances. Opening new frontiers
22 Teresa Cañedo-Arguelles Fábrega, Potosi: La version aymara de un mito europeo. La minerı́a y sus
efectos en las sociedades andinas del siglo XVII (La Provincia de Pacajes) (Madrid: Editorial Catriel,
1993), 45; Stannard, American Holocaust, 74, 89, 91; Churchill, A Little Matter of Genocide, 87–8.
23 Stannard, American Holocaust, xii; Churchill, A Little Matter of Genocide, 86.
25 Ibid. 506–7.
was a way to rid Spanish settlements of disruptive and potentially dangerous

elements, and at the same time offered the possibility of bringing new territory,
and subjects, into the realm. As an Indian group legally became rebels once they
had ignored a formal call for submission, and rebels against the Crown could
legally be enslaved, expansion of the frontier, and the suppression of Indian
rebellions, provided a continuing pretext for both genocidal massacres and en-
slavement. Those who resisted were often subject to onslaughts ‘by fire and blood’
which had as their objective the physical elimination of the rebel group and the
enslavement of any survivors. Thus, for Indians the choices were few indeed. One
alternative was integration into the Spanish realm, and subjection to ethnocidal
policies which sought the cultural destruction of the indigenes. The other option
was resistance, which elicited a deliberate response from the Spaniards which
sought the elimination and/or enslavement of their adversaries.26 Migration from
conquered areas often entailed a change in climate which could be deadly in
lowland areas, and also could leave the refugees at the mercy of established groups
in the region.
E T H N O C I D E A N D T H E E XT I R PAT I O N
O F N AT I V E R E L I G I O U S P R AC T I C E S
................................................................................................................
To the physical genocide that resulted from deliberate extermination of specific
groups which resisted Iberian domination was added a host of other genocidal
policies systematically inflicted upon the indigenes. In this context, clerics, both
secular and regular, became a vortex of ethnocide, or the destruction of culture and
its repositories. Many policies were patently ethnocidal and concerned the intro-
duction of Catholicism to the region in the sixteenth and seventeenth centuries.
Although Indians were exempt from the loathed Holy Office of the Inquisition,
they were no less subject to religious persecution. Native religions and their cultural
underpinnings were the focus of well-organized campaigns of extirpation, often
involving killing native religious authorities and the systematic destruction of
native places of worship and sacred documents.27
As Duviols notes in the Andean context, the goal was to ‘destroy every vestige of
the pagan religion, both in objects and in the spirit of the Indians’, and to Lemkin,
26 Charlotte Gradie, The Tepehuan Revolt of 1616: Militarism, Evangelism, and Colonialism in
Seventeenth-Century Nueva Vizcaya (Salt Lake City: University of Utah Press, 2000), 33.
such acts constituted ‘cultural genocide’.28 Extirpation campaigns were formal and
systematic affairs, and mixed violent and peaceful means. Often, Indians were first
given an opportunity to confess concerning their beliefs, practices, and the loca-
tions of shrines. Those who were less forthcoming, or unconvincing, were sub-
jected to torture. Such techniques were often effective, and resulted in the gathering
up and destruction of shrines and other items associated with native worship, and
the concealment of their ashes or shards. Those found guilty of idolatry were often
flogged and/or had their head shaved as a form of humiliation, and were forced to
attend catechism to complete the process.
Apart from the perceived spiritual dividends of Catholicism, extirpation cam-
paigns could yield silver and gold, further inflaming the zeal with which such
campaigns were undertaken.29 Indeed, such was the fervour with which such
operations were conducted that occasionally the residents of a village would flee
before the arrival of the extirpation committee. Other times, those leading the
campaign were charged with excesses, such as unlawfully seizing Indian lands and
destroying houses.30 To give an idea of the yield of such efforts, by 1559 in the region
of Huaylas, Peru, Augustinian extirpators reported destroying over 5,000 native
religious artefacts.31
Despite such attacks, native religions often survived, although in adaptive,
syncretic forms. Part of this was due to the fact that a fragment of a native religious
object was believed to retain its complete power, and some natural religious sites,
such as hills and mountains, simply could not be destroyed.32 While some tradi-
tional practices continued furtively, others melded into Catholicism, and the
polytheistic nature of native belief systems imbued it with an important degree
of adaptability.33 Ironically, in their efforts to destroy native religions, idolatry
campaigns could help preserve information concerning native rites for future
generations. For example, books that essentially were training manuals for extir-
pators often contain extensive information concerning native beliefs and practices.
Gleaned through hostile eyes, such vestiges are often all that remain of pre-
conquest beliefs.34
Although such campaigns were not always successful in eradicating native
religious beliefs and practices, they did help frame the terms of a modus vivendi
28 Pierre Duviols, La destrucción de las religiones andinas (Conquista y colonia) (Mexico City:
UNAM, 1977), 423; McDonnell and Moses, ‘Raphael Lemkin as Historian’, 508, 511.
29 Duviols, La destrucción de las religiones andinas, 9, 249–59, 373–7, 424.
30 Ibid. 405–18, 422.
31 Ibid. 430.
32 Ibid. 436; Nicholas Griffiths, The Cross and the Serpent: Religious Repression and Resurgence in
Colonial Peru (Norman: University of Oklahoma Press, 1996), 8, 26.
33 Duviols, La destrucción de las religiones andinas, 437.
34 See, for example, Pablo Joseph de Arriaga, The Extirpation of Idolatry In Peru, trans. L. Clark
Keating (Lexington: University of Kentucky Press, 1968).
between the two spiritual, and cultural, worlds. Once established in a region, some
priests were willing to tolerate a degree of discreet idolatry. For example, saint’s
days celebrations often served as a cover for continuing traditional rites. Part of this
tolerance reflected an acceptance of reality; in other cases it was a tacit agreement of
mutual tolerance of forbidden practices, such as priestly promiscuity or their
engagement in commercial enterprises. Other times, it reflected a fear among
clerics that a lack of tolerance could result in their own death, especially through
poisoning as natives often were forced to serve as cooks for the priest, among other
duties.35 Nevertheless, in Indian villages the priest embodied spiritual, economic,
and political power, with the desirability of a parish often being a function of its
economic potential. Any overt challenge to clerics was dealt with harshly, through
flogging, incarceration, banishment, seizure of goods, or accusations of idolatry.36
Beyond extirpation campaigns, ethnocide often took more discreet but not less
destructive forms. Rural clergy were especially important in this regard due to their
close contact with the natives. Coerced conversions and participation in Catholic
rituals, involuntary marriages, and massive forced relocation into new settlements
and missions all had among their objectives the destruction of native religions.
Although efforts to impose monogamy inhibited native reproduction, forced
relocation was especially detrimental to native populations, as higher population
concentrations greatly facilitated the spread of disease.37 The exploitation and
abuse suffered by Indians at the hands of their priests, as well as their frequent
hypocrisy regarding sexual activity, alienated many from the Catholic faith, even
when they ascribed supernatural powers to it.38
To the genocidal massacres of conquest, and the ethnocide of forced religious
conversion, was added forced migration which would be considered by Lemkin as
physical genocide through the ‘deprivation of livelihood’.39 Forced migration
occurred at three levels: migration into new, state-mandated settlements, migration
to fulfill state-mandated labour obligations, such as the mita in Peru, and migra-
tion to avoid such demands. In the cases of mercury production in Huancavelica in
Peru and silver mining in Potosı́ (in present-day Bolivia), certain provinces were
assessed a quota of labourers to provide for a fixed length of time. Because their
token wage, if paid, was below a subsistence level, many travelled with their families
who also worked in the area as ore sorters, vendors, and domestic servants. While
many in the mines succumbed to cave-ins, falls, toxic gasses, and other accidents,
others were so weakened that they never fully recovered or maimed such that they
35 Duviols, La destrucción de las religiones andinas, 401–3.

36 Griffiths, The Cross and the Serpent, 161, 245, see also Robins, Priest-Indian Conflict, 69–76.
37 Cook, Demographic Collapse, 89, 253.
38 Kenneth Mills, Idolatry and Its Enemies: Colonial Andean Religion and Extirpation, 1640–1750
(Princeton: Princeton University Press, 1997), 254–6, 284–5.
39 McDonnell and Moses, ‘Raphael Lemkin as Historian’, 506, 507–8.
could not earn an income or work the land. Those sent by overlords to the hot
climes of the coca plantations often quickly perished from disease.
Those that did not die under conditions of forced labour often had little reason to
return home, as many who did found they had lost access to their lands. Worse, as the
labour levies were based on grossly outdated censuses, many Indians in Peru would
also find that they would be forced to return for another labour turn in the mines in
only two or three years, as opposed to the theoretical seven for the mita. As a result,
many, ironically, stayed in the silver mining city Potosı́ where wages and tasks for non-
conscripts were better than many other places. Others migrated to provinces that were
not subject to the labour draft, and where they also paid less tribute.40
Other factors affecting native populations included widespread miscegenation,
often forced and sometimes not, which reduced the native population and led to
the emergence of a new group, mestizos (those of Indian and Spanish heritage),
who often assumed positions of authority over the natives. Furthermore, children
were routinely seized in lieu of debts. If the debt, now converted in effect into a
ransom, was not paid, the child was forced to work it off. This effort to pay off the
debt could easily become an open-ended process, as charges for their upkeep could
be applied to the debt, and they could remain in permanent bondage. Such
practices, due to their effects on the family unit and reproduction, were considered
by Lemkin to be examples of ‘biological’ genocide.41
It is clear that the indigenes in Latin America did not simply suffer genocide,
they suffered genocides: physical, biological, and cultural.42 The unintentional
spread of disease was part and parcel of this process as it facilitated the conquest
and the human and cultural destruction which ensued. Concerning the exploita-
tion of the natives, at the level of Crown policy in Spain, there were a plethora of
laws specifically designed to protect the Indians from a wide variety of abuses
which the Crown was aware were routinely committed. Such laws, however, were of
little effect, not only because of the exploitative nature of the colonial system and
the flexibility given to local officials by the Crown, but also because the laws
concerning the Indies were riddled with contradictions and ambiguities. Just as
there were laws that sought to temper the degree of exploitation, others permitted
forced migration, familial dissolution, coerced relocation, and efforts to extermi-
nate native religions, practices, and culture.43
40 Peter Bakewell, Miners of the Red Mountain: Indian Labor in Potosı́, 1545–1650 (Albuquerque:
University of New Mexico Press, 1984), 69; Jeffrey A. Cole, The Potosı́ Mita, 1573–1700. Compulsory
Indian Labor in the Andes (Stanford: University Press, 1985), 26–28.
42 Ibid. 504.
43 On this topic, see for example, Silvio Zavala, El servicio personal de los indios en el Peru, vol i
(Mexico City: El Colegio de Mexico, 1978); Roberto Levillier, Don Francisco Toledo, supremo
organizdor del Perú (Buenos Aires: Espasa-Calpe, S.A., 1940), and Ruben Vargas Ugarte, Pareceres
juridicos en asuntos de indias (1601–1718) (Lima: CIP, 1951).
Policies concerning religion had multiplicative effects that went well beyond
forcible conversions. As with private individuals, religious orders owned land and
often encroached on that of native communities, reducing their ability to provide for
themselves. Local clergymen also routinely demanded unremunerated services from
their parishioners, and drained them of any surplus they may have through forced
participation in religious services and confraternities. Furthermore, forced marriages
and the imposition of monogamy eliminated choice and restricted population
growth. Finally, the forced relocation of Indians to Spanish-style towns not only
uprooted them from places which had religious significance to them, but also
facilitated the spread of disease and further spurred depopulation. The clergy had a
disproportionate role in native ethnocide not only due to systematic efforts to destroy
native belief systems and the culture that surrounded them, but also because they were
the only non-Indian legally allowed to reside in native communities.
N AT I V E R E S I S TA N C E A N D S U BA LT E R N
GENOCIDE
................................................................................................................
Given the nature of the system under which the natives lived, and died, it should
come as little surprise that rebellions against colonial rule could assume a genocidal
character. Such cases of ‘subaltern genocide’, or genocides from below where
subject populations seek the extermination of their oppressors, were the exception
rather than the rule in the continuum of resistance. Most rebellions were short,
local affairs against a specific individual, such as a priest or governor, or against a
policy, such as tribute assessments, forced purchase of goods, or religious fees. On
occasion, however, such uprisings were broader in scope and objectives.44
One such subaltern genocidal rebellion was the Tepehuan uprising in present-
day Mexico, which lasted from late 1616 until it was finally suppressed in 1620.
Hundreds of Hispanic settlers and their slaves perished in the rebellion, in addition
to several clerics.45 This rebellion erupted in a context of severe depopulation from
disease, which was exacerbated by forcing populations to live at missions.46 In 1615,
44 For detailed case studies of such subaltern genocides, see Nicholas Robins, Native Insurgencies
and the Genocidal Impulse in the Americas (Bloomington: Indiana University Press, 2005) and
Genocide and Millennialism in Upper Peru: The Great Rebellion of 1780–1782 (Westport, CT: Praeger,
2002). Most recently, see Nicholas Robins and Adam Jones (eds), Genocides by the Oppressed:
Subaltern Genocide in Theory and Practice (Bloomington: Indiana University Press, 2009).
45 Daniel Reff, ‘The “Predicament of Culture” and the Spanish Missionary Accounts of the
Tepehuan and Pueblo Revolts’, Ethnohistory 42:1 (1995), 63; Gradie, The Tepehuan Revolt of 1616.
46 Reff, ‘The “Predicament of Culture” ’, 71.
a dire situation for the Tepehuans was made worse as a result of a drought and
famine.47 As elsewhere, native religious leaders were powerless to change the course
of events and saw their prestige decline. As the missionaries were equally unable to
stop the spread of disease, many natives saw little reason to remain Christians.48
Indeed, over time, many Indian groups began to associate Catholic rituals with ‘evil
spells that resulted in death and disease.’49 The leader of the Tepehuan revolt,
Quautlatas, commanded his followers not only to kill those of Iberian origin, but
anyone who had converted to Christianity. To inspire his followers, he assured
them that those who fell in battle would be resurrected within a week of their
victory, adding that the elderly would be young again.50 In addition, he promised
that reascendant native deities would forever prevent the Spanish from returning
by sending storms to sink their ships.51
Another subaltern genocidal uprising was the 1680 Pueblo Revolt in present-day
New Mexico. The Franciscan order, backed by military force, was instrumental in
advancing the Spanish presence northward in Mexico. In the Pueblo region, while
nominally subject to civilian authorities, they wielded considerable autonomy and
power. As in other regions of Latin America, by the mid-seventeenth century, the
sedentary Indians of the region had experienced a population collapse from disease
of well over ninety per cent, and suffered consistent and systematic attacks against
traditional religion and customs. A long-planned and well-coordinated revolt
erupted in 1680, led by a medicine man known only as Popé.52
On the night of 10 August 1680, the natives of the region rose up, killing almost
all non-Indians they captured. Those who could fled to Santa Fe, the district
capital, where they took refuge with the governor. Soon besieged, they were vastly
outnumbered by their adversaries, who cut off their water supply. With surrender
as the only other option, the Hispanics made a bold attack on the Indians, killing
many and creating enough disarray among the rebels to allow them to begin an
exodus southward. Eventually, they made it south of the Rio Grande, and tallied
their losses. Of approximately 2000 settlers in the region, 401 people had been
killed, all but ninety-five of them being women and children. In addition, the rebels
had destroyed thirty-four villages along with many ranches and agricultural estates.
For the next twelve years the Pueblo Indians would live free of Spanish control.53
47 Gradie, The Tepehuan Revolt of 1616, 150.

48 Reff, ‘The “Predicament of Culture” ’,71; idem, Disease, Depopulation and Culture Change in
Northwestern New Spain, 1518–1764 (Salt Lake City: University of Utah Press, 1991), 277.
49 Reff, Disease, Depopulation and Culture Change, 273.
50 Reff, ‘The “Predicament of Culture” ’,70; idem, Disease, Depopulation and Culture Change, 273;
Gradie, The Tepehuan Revolt of 1616, 149.
51 Gradie, The Tepehuan Revolt of 1616, 149.
52 Andrew Knault, The Pueblo Revolt of 1680: Conquest and Resistance in Seventeenth-Century New
Mexico (Norman: University of Oklahoma Press, 1995), 153–5, Robins, Native Insurgencies, 27–9.
53 Robins, Native Insurgencies, 30–4.
The objectives of the rebels were plainly stated by many.54 For example, the
insurgent leader Alonso Catiti called upon followers ‘to assemble in order to go to
the Villa to kill the governor and all who were with him’. Others were inspired by
the call ‘to kill the friars, the alcalde mayor and the other persons’ in Santa Fe, while
others simply did ‘not want religious or Spaniards’.55 These were not idle observa-
tions; only women were among the few prisoners taken by the rebels. Not only did
the Indians seek to exterminate the interlopers, they sought to systematically
remove all vestiges of Spanish rule. Popé and other leaders prohibited anything
of Spanish origin, including crops, Christian names, and the practice of Catholi-
cism. Those who had been baptized were commanded to engage in ritual bathing
to remove the perceived taint of baptismal waters, and all those who had been
forced by the friars into marriages were declared free from matrimony. Native
objectives were also expressed at a symbolic level, such as by the destruction of
church bells, which had commanded them to work and worship, the piling of dead
friars on altars, and the scalping of statues of Christ.56
Although ultimately defeated, the Great Rebellion of Peru and Upper Peru
(present-day Bolivia) of 1780–2 was no less genocidal and in fact was the largest
regional challenge to Spanish rule in the Americas before the independence wars.
In the years leading up to the rebellion, the Indian exploitation had been growing
more severe. Not only were new taxes on Indian staples levied and aggressively
collected, but Spanish provincial governors had increasingly expanded their system
of forcing the Indians to purchase goods at highly inflated prices under the
repartimiento de mercancı́as system. The clergy, seeing their economic and political
power under attack by the Bourbon monarchy, increased their own efforts to strip
the Indians of any surplus through increasing the number of saint’s days celebra-
tions, and the fees associated with them. Priests also issued more frequent demands
for unremunerated labour, routinely charged extortionate fees for services, and
utilized funeral charges as a means of taking from the deceased any property they
may have had. Finding no effective redress in colonial courts and having little hope
of a better future other than that offered by millennial prophecies, in 1780 a storm
of genocidal violence swept the region.57
The insurgency was a loosely confederated series of uprisings, the first of which
erupted in Chayanta, in what is today south central Bolivia. Initially led by Tomás
Catari, a hereditary leader denied the leadership of his village, his brothers Dámaso
and Nicolás led an even more violent phase after his death. Calling for the
54 For a discussion of the intentionality of rebel actions and their relation to their broader
objectives in this and the Great Rebellion of Peru and Upper Peru, see ch. 4 of Robins, Native
Insurgencies, 68–95.
55 Ibid. 72.
56 Ibid. 142–5.
57 Scarlett O’Phelan Godoy, Rebellions and Revolts in Eighteenth Century Peru and Upper Peru
(Colonge: Bohlau Verlag, 1985), 109, 119, 148, 207, 260, 270–80.
elimination of civil taxes, religious fees, forced labour, and coerced purchases, the
insurgents killed those who opposed them. Initially, mestizo village overlords were
targeted; however, as the rebellion progressed and radicalized, those of lighter skin
in general were systematically killed. Usually, the rebels would isolate a village,
while the defenders took refuge in the local church. They would then attack the
town, storm the church, and kill those inside. On some occasions, women would be
spared to serve the rebels, as were some men who could serve as scribes or operate
or repair weapons.
Similar grievances led to similar events in the region of Cuzco, present-day Peru,
where the rebellion was led by José Gabriel Túpac Amaru, the hereditary leader of the
village of Tungasuca. Villages were attacked and people systematically killed on the
basis of race and ethnicity before the tide turned against them following a brief siege
on Cuzco, the ancient Inca capital, in January 1781. The city of La Paz, capital of
present-day Bolivia, was also besieged between March and October 1781 by another
rebel, Túpac Catari. Approximately 10,000 people, or one-third of the city’s popula-
tion, perished, mostly from starvation, before royalist forces finally broke the siege.58
As in the Pueblo Revolt, the exterminatory nature of this insurgency was evident
not only in the systematic slaughter of whites and those of lighter skin, but in the
rebels’ frequent efforts to destroy objects associated with the Catholic Church.
Statues of Christ were denuded and thrown into pyres, monstrances deliberately
shattered, and many priests killed who did not flee their parishes in time. Overall,
the ferocity and exterminatory nature of the insurgency was highest in Upper Peru
and increased with time and distance from nominal, and more conservative,
leaders such as Túpac Amaru and Tomás Catari. Prophecies promising the return
of native rule were frequently reinforced by a belief that not only had native deities
finally risen to expel the invaders, but that those who fell in battle would resurrect
and enjoy the utopia they were creating.59
Subaltern genocides offer important insights into how native peoples perceived
ethnicity and race, as opposed to how it was viewed by the colonial elite. Unlike
colonial legislation that utilized a detailed progenitor-based system to categorize
and regulate racial categories, rebels in these movements employed a more ‘empir-
ical’ approach in which race and colour were considered in a wider context.60 While
the light skin of Spaniards and Creoles clearly identified them as ‘white’, mestizos
and even Indians were also marked for death based on such factors as their dress,
occupation, religious orientation, and the primary language they spoke. Cultural
markers such as these generally reflected one’s position in the hierarchy of colonial
58 Robins, Native Insurgencies, 38–50.

59 Nicholas A. Robins, Genocide and Millennialism in Upper Peru: The Great Rebellion of 1780–1782
(Westport, CT: Praeger, 2002), 202–5.
60 For an example of such categorization in Brazil, see Patrick Wolfe, ‘Land, Labor, and Difference:
Elementary Structures of Race’, American Historical Review 106 (2001), 896.
exploitation. As a result, for example, a mestizo who served in a position of

authority as an overseer or tax collector and who wore Spanish-style clothing
would not be considered a native in rebel eyes. But simply being mestizo was in
and of itself often insufficient to mark one for death, a tendency epitomized by the
fact that Túpac Amaru was a mestizo.61 Cultural orientations, and their physical
expression, were central in defining indigenous identity, and often who lived and
who died.62
Such cases of subaltern genocides call into question widely held views of genocide.
For example, such uprisings were the means to establishing an Indian state, not an
expression of state policy. Furthermore, genocidal movements are not bound to any
specific time period, and subaltern genocides demonstrate that genocides are not
always from above or about the strong over the weak. Although in many genocides
the victim group does not possess the military or organizational means of defeating
their adversaries, in cases of subaltern genocide such as the Great Rebellion (or the
‘Sepoy Mutiny’ in British-ruled India), the insurgents were confronting a formally
organized and better armed military force. Unlike genocides from above, subaltern
perpetrators of genocide often were defeated by their enemies.
Subaltern genocide also challenges prevailing concepts of leadership. While that
of the Pueblo revolt was highly centralized, that of the Great Rebellion (and later
the Caste War of Yucatan) was highly fragmented, confederational, and far from
bureaucratic. In some cases, such as with Tomás Catari in Upper Peru and Túpac
Amaru in Peru, despite their charismatic appeal, the objectives of the formal
leadership were considerably more limited in scope and generally more conserva-
tive than that of those who operated in their name. This led to them having
minimal operational control in areas outside of their immediate command, and
as a result they became largely nominal to the rebellions that they are credited with
leading. In contrast, local level leaders had prominent roles in these uprisings and
were central in the perpetration of genocide.63
C O N C LU S I O N
................................................................................................................
Exploring the genocides of conquest and colonization in Latin America highlights
the shortcomings of conventional definitions of genocide. According to some
interpretations of the 1948 UN Convention on genocide, it is possible to have a
61 Robins, Genocide and Millennialism, 59, 195.

62 Robins, Native Insurgencies, 4, 8–9, 76–7.
63 Ibid. 118–24.
‘genocide’ free of death. Actions ‘causing serious bodily or mental harm to mem-
bers of the group’ are legally considered genocide, yet can be interpreted as not
necessarily involving mass killing even when the object is the destruction of a
group. Likewise, although in a broader intellectual context, deliberate cultural
destruction, or ethnocide, and the deliberate elimination of languages, or lingui-
cide, are also often considered genocide.64 On the other hand, the unintended
extinction or near extinction of a people from disease, a literal genocide and what
could also be termed ‘collateral genocide’, is not considered genocide according to
the UN Convention.
While the debate over the role of intentionality in genocide will no doubt
continue, the empirical reality of the Latin American case supports an approach
to genocide in which intent, whether specific or general, is only one component of
a complex, interrelated, and often impersonal process.65 Intentional policies and
events, such as genocidal massacres to aid conquest, forced relocation of subjects,
deprivation of livelihood, separation of families, and systematic cultural destruc-
tion, were facilitated by an unintended, demographic collapse of monumental
proportions. That genocidal colonial policies and practices reflected the norms of
their times did not mitigate their effect on the population, nor does it make them
any less genocidal. Furthermore, the emphasis on intent, and the UN Genocide
Convention generally, is focused on establishing a judicial basis to punish present
and future perpetrators. Applying such modern standards to historical processes
only serves to limit, not deepen, our understanding of them.66
SELECT BIBLIOGAPHY
Borah, Woodrow, and Sherburne Cook, The Aboriginal Population of Central Mexico on the
Eve of the Spanish Conquest (Berkeley: University of California Press, 1963).
Churchill, Ward, A Little Matter of Genocide: Holocaust and Denial in the Americas 1492 to
the Present (San Francisco: City Lights Books, 1997).
64 Frank Chalk and Kurt Jonassohn, ‘The Conceptual Framework’, in eidem (eds), The History and
Sociology of Genocide: Analyses and Case Studies (New Haven: Yale University Press, 1990), 9–10.
65 Concerning such divisions within genocide studies and the primacy ascribed to the Holocaust,
see Moses, ‘Conceptual Blockages’, and on the role of impersonal processes in genocides, see Vinay Lal,
‘The Concentration Camp and Development: The Pasts and Future of Genocide,’ Patterns of Prejudice
39:2 (2005), 220–43.
66 Concerning the debate on this topic, see the following three articles in the Journal of Genocide
Research 10:1 (2008): Tony Barta, ‘With Intent to Destroy: On Colonial Intentions and Genocide
Denial,’; Norbert Finzsch, ‘If it Looks Like a Duck, If It Walks Like a Duck, If It Quacks Like a Duck’,
and David Stannard, ‘Déjà Vu All Over Again’. See also Isidor Walliman and Michael Dobkowski
(eds), Genocide and the Modern Age: Etiology and Case Studies of Mass Death (Westport, CT:
Greenwood Press, 1987).
Cook, Noble David, Demographic Collapse: Indian Peru, 1520–1620 (Cambridge: Cambridge
Duviols, Pierre, La destrucción de las religiones andinas (Conquista y colonia) (Mexico City:
UNAM, 1977).
Gradie, Charlotte, The Tepehuan Revolt of 1616; Militarism, Evangelism, and Colonialism in
Seventeenth-Century Nueva Vizcaya (Salt Lake City: University of Utah Press, 2000).
Griffiths, Nicholas, The Cross and the Serpent: Religious Repression and Resurgence in
Colonial Peru (Norman: University of Oklahoma Press, 1996).
Las Casas, Bartolomé de, Brevissima relación de la destrucción de las Indias. (Bayamón,
Puerto Rico: Universidad Central de Bayamón, 2000).
Moses, A. Dirk, ‘Conceptual Blockages and Definitional Dilemmas in the “Racial Century”:
Genocides of Indigenous Peoples and the Holocaust’, Patterns of Prejudice 36:4 (2002), 7–36.
Robins, Nicholas, Native Insurgencies and the Genocidal Impulse in the Americas (Bloo-
mington: Indiana University Press, 2005).
Stannard, David, American Holocaust: Columbus and the Conquest of the New World
(New York: Oxford University Press, 1992).
chapter 16
.............................................................................................
RETHINKING
GENOCIDE IN
N O RT H A M E R I C A
.............................................................................................
gregory d. smithers
Since the 1940s, American historians have been reluctant to incorporate Raphael
Lemkin’s concept of genocide into the narrative of the United States’ history. This
reluctance is in large measure a reflection of the context in which Lemkin’s concept
of ‘genocide’ entered popular and political discourse in America. The word geno-
cide became part of the American vernacular with the onset of Cold War politics in
the late 1940s and early 1950s.1 As the world grappled with the magnitude of the
atrocities visited upon the European Jewish population by the Nazi regime and
contended with the emergence of the ‘Communist bloc’ under the leadership of the
Soviet Union, most Americans found it difficult to see how Lemkin’s neologism
had anything to do with their own nation’s history.2 And yet, Lemkin’s concept of
genocide, like the colonial origins of the United States, was rooted in the often-
violent encounters between European colonists and Native Americans. In the wake
of Axis Rule in Occupied Europe (1944), Lemkin arrived at his broad-reaching
definition of genocide after studying the historical impact of colonialism on
1 In the United States, lawmakers worried that ratification of the UN Convention on Genocide
would expose the federal government to charges of genocide against Native Americans and African-
Americans. Lawrence J. LeBlanc, The United States and the Genocide Convention (Durham: Duke
University Press, 1991), 26, 64.
2 Ibid. 6–7, 50.
rethinking genocide in north america 323
indigenous peoples throughout the Americas. In North America, English settler

colonialism left indigenous Americans struggling to survive European diseases for
which they lacked immunity, and grappling to understand the significance of
cultural, social, environmental, and economic changes that settler colonialism
brought to the east coast of North America. As recent scholarship has recognized,
the legacy of settler colonialism in North America continued to impact Native
Americans long after the Founding Fathers drafted and passed a national constitu-
tion.3 It is in the context of settler colonialism that the concept of genocide needs to
be re-examined in relation to North American history.
Lemkin defined genocide as the ‘destruction of a nation or of an ethnic group’.
He coined the term from the ancient Greek word genos, meaning ‘race’ or ‘tribe’,
and the Latin cide, or ‘killing’. In Lemkin’s mind, the genocide of a people did not
have to occur in one instance, for the most important factor in defining a genocidal
event was intent, or the existence of ‘a coordinated plan of different actions aiming
at the destruction of essential foundations of the life of national groups, with the
aim of annihilating the groups themselves’. The issue of intent has divided Ameri-
can historians. Some, as we will see, define intent broadly, and refer to both the
cultural and biological destruction of indigenous people. Other scholars argue that
American history lacks evidence of coordinated government efforts to commit
genocide on any group of people.4 During the early years of the Cold War, the idea
that English colonial authorities or the US Government had ever acted with
genocidal intentions toward Native Americans undermined the notion of Ameri-
can exceptionalism, and struck at the heart of America’s self-identification as a
beacon for liberty and freedom in the face of Communist oppression.5
During the civil rights era of the 1960s, historians began to challenge the
traditional historical interpretations of American exceptionalism. New insights
emerged that reassessed racial, ethnic, and gender relations in US history.6 These
studies focused on the violent and often genocidal aspects of colonial encounters
between the British and American Indians. Historians such as Francis Jennings,
3 Michael A. McDonnell and A. Dirk Moses, ‘Raphael Lemkin as Historian of Genocide in the
Americas’, Journal of Genocide Research 7:4 (2005), 501–29. See also Ann Laura Stoler, ‘Tense and
Tender Ties: The Politics of Comparison in North American History and (Post) Colonial Studies’,
Journal of American History 88:3 (2001), 829–65.
Proposals for Redress (1944; New York: Howard Fertig, 1973), 79. See also Samantha Power, ‘A Problem
from Hell’: America and the Age of Genocide (New York: Harper Perennial, 2002), 42.
5 John K. White, Still Seeing Red: How the Cold War Shapes the New American Politics (Boulder, CO:
Westview Press, 1997), 5; William E. Odom and Robert Dujarric, America’s Inadvertent Empire (New
Haven: Yale University Press, 2004), 226.
6 Seymour M. Lipset, American Exceptionalism: A Double-Edged Sword (New York: W. W. Norton,
1996); Ian Tyrell, ‘Making Nations/Making States: American Historians in the Context of Empire’,
Journal of American History 84:3 (1999), 1015–43; David W. Noble, Death of a Nation: American Culture
and the End of Exceptionalism (Minneapolis: University of Minnesota Press, 2002).
324 gregory d. smithers
Roy Harvey Pearce, Richard Drinnon, Winthrop Jordan, George Fredrickson, and
Roger Daniels enriched our understanding of the various aspects of Western racism
and the impact of racially motivated violence that targeted Native American,
African-American, and Asian American peoples. Like settler societies in Australasia
and southern Africa, the colonies that became the United States in 1789 (the year
that George Washington became the first President of the United States) were a
product of complex sociocultural, economic, and geopolitical motives that helped
European and American colonizers structure settler colonial society in North
America. As Richard Cole Harris informs us, ‘Colonialism has more than one
voice’; understanding these complex and sometimes contradictory voices goes a
long way to explaining ‘genocidal episodes’ in the settler colonial history of North
America, and particularly the United States.7
S E T T L E R C O LO N I A L I S M I N E N G L I S H
N O RT H A M E R I C A
................................................................................................................
The sixteenth- and seventeenth-century motives for English colonization along the
eastern seaboard of North America were multifaceted and complex. Historically
specific and geographically contingent factors, such as the sociopolitical context of
early modern Europe, the hope of a New World ripe with socioeconomic oppor-
tunities, and changing European perceptions of Native American peoples, all
played a role in shaping settler colonialism in English North America. Puritans
and Separatists, for example, fled religious persecution in England and Europe, an
experience that shaped their collective ambition to establish economically profit-
able settlements characterized by political self-rule in the New England frontier.8
With the rapid decline of New England’s Native American population due to the
ravages of disease, thousands of acres of forests, meadows, and marshlands were
left ‘relatively unoccupied’.9 Puritan missionaries hoped to convert Native Amer-
icans, or ‘praying Indians’ as the missionary John Eliot called them, to Christianity
(a conversion that Lemkin labelled ‘cultural genocide’), while settlers aimed to fill
these ‘empty’ spaces with their version of settler colonial civilization, carving out of
7 Richard Cole Harris, Making Native Space: Colonialism, Resistance, and Reserves in British
Columbia (Vancouver: University of British Columbia Press, 2002), 46; A. Dirk Moses, ‘Conceptual
Blockages and Definitional Dilemmas in the “Racial Century”: Genocides of Indigenous Peoples and
the Holocaust’, Patterns of Prejudice 36:4 (2002), 28.
8 Ray Allen Billington, American History before 1877 (Totowa, NJ: Rowman and Littlefield, 1965),
16–17.
9 Daniel Vickers, A Companion to Colonial America (Malden, MA: Blackwell, 2003), 55.
the New England wilderness the towns, villages, and farms that comprised the
material foundation for a Christian society in the New World.10 Settler colonialism
in New England therefore involved what Alfred Crosby has called the ‘demographic
takeover’ of New England, the reorganization of land use, and for the surviving
indigenous inhabitants, the gradual restructuring of traditional social, economic,
and political folkways.11
English settlements and plantations also emerged in the mid-Atlantic, the
Carolinas, and Chesapeake Bay during the seventeenth and eighteenth centuries.
In the Chesapeake, small farms and plantations dotted the landscape along the
James River to the mouth of the Chesapeake Bay. The Chesapeake experienced slow
and uneven economic development for much of the seventeenth century. As a
result, the colonies of Virginia and Maryland remained tottering frontiers of
settlement. Not until the cultivation of tobacco emerged as a cash crop during
the late seventeenth century, an economic development that led to pressure for ever
larger tracts of land in the Chesapeake occupied by Native Americans, and worked
increasingly by African slaves, did the colonial economy develop and significant
territorial expansion begin.12 The demographic heterogeneity that emerged in
seventeenth-century Virginia, as in other English settler colonies, meant that the
Old World social, economic, and political frameworks that defined epistemological
categories such as master and servant, or landowner and lessee, were rethought to
meet the historical and geographic demands of settler colonialism in the North
American colonies.13
Demographic heterogeneity, a defining characteristic of English settler colonial-
ism in North America, necessitated an elaborate system of colonial regulation of
human behaviour. To the English in North America, racial and ethnic cosmopoli-
tanism was an impediment to the growth of politically stable settler communities
10 Daniel Gookin, Historical Account of the Doings and Sufferings of the Christian Indians, in the
Years 1675, 1676, 1677 (1836; New York: Arno Press, 1972), 434–40, passim; Kristina Bross, Dry Bones and
Indian Sermons: Praying Indians in Colonial America (Ithaca, NY: Cornell University Press, 2004), 21;
John Docker, ‘Are Settler-Colonies Inherently Genocidal? Re-reading Lemkin’, in A. Dirk Moses (ed.),
Empire, Colony, Genocide: Conquest, Occupation, and Subaltern Resistance in World History (New York:
Berghahn Books, 2008), 86. 89, 95.
11 Alfred W. Crosby, Germs, Seeds and Animals: Studies in Ecological History (Armonk, NY:
M. E. Sharpe, 1994), 29; David Scott, Conscripts of Modernity: The Tragedy of Colonial Enlightenment
(Durham: Duke University Press, 2004), 44.
12 Lorena S. Walsh, ‘Slave Life, Slave Society, and Tobacco Production in the Tidewater
Chesapeake, 1620–1820’, in Ira Berlin and Philip D. Morgan (eds), Labor and the Shaping of Slave
Life in the Americas (Charlottesville: University Press of Virginia, 1993), 170.
13 Reinhart Koselleck and Keith Tribe, Futures Past: On the Semantics of Historical Time (New York:
Columbia University Press, 2004), 256; Ian Tyrell, ‘Beyond the View from Euro-America:
Environment, Settler Societies, and the Internationalization of American History’, in Thomas
Bender (ed.), Rethinking American History in a Global Age (Berkeley: University of California Press,
2002), 170–1.
and a source of social anxiety.14 As a result, colonialists, particularly English

colonial authorities, articulated a binary logic in which settler colonial civilization
was imagined and created in opposition to the chaos of the American wilderness.
This binary logic became increasingly elaborate as English settler colonies devel-
oped and expanded territorially during the eighteenth century, ultimately helping
settlers to structure and organize wilderness spaces with a combination of town
planning, trade relations with other Europeans as well as Native Americans, and
the delineation of a social hierarchy that by the American Revolution was overtly
racial and white supremacist in nature.15 As Eyal Ben-Ari and Yoram Bilu observe,
‘space and place are central organizing principles in all complex societies.’16 To
Ben-Ari’s and Bilu’s observation we must add that English colonists in North
America adapted Old World economic, political, and social structures to order
their New World existence. Recognition of the commitment of English settlers to
these ideals, and the colonial structures that they helped to establish, moves us
towards an understanding of how English colonists rationalized anti-Indian vio-
lence, or even genocidal actions.
The historian Patrick Wolfe has presented one of the most sustained frameworks
for understanding settler colonialism. Wolfe argues that ‘Settler colonies were (are)
premised on the elimination of native societies.’17 In an influential 2001 essay in the
American Historical Review, Wolfe presents a ‘typology of strategic phases’ for
understanding settler colonialism. These phases include ‘confrontation’, which
for indigenous peoples involves the loss of life through disease and violent conflict
with settlers; ‘carceration’, or the removal and segregation of indigenous peoples
away from settler society; and ‘assimilation’, which can involve biological and/or
cultural assimilation of indigenous peoples with settler populations.18 Wolfe argues
that ‘the recalcitrant persistence of extraneously constituted indigenous socie-
ties’—be they Australian Aborigines or North American Indians—posed a serious
problem to settler colonialism.19 Settlers, coveting the land that indigenous peoples
occupied, turned settler colonialism into a ‘zero-sum contest over land on which
14 A. G. Roeber, ‘The Origins of Whatever Is Not English among Us: The Dutch-Speaking and the
German Speaking Peoples of Colonial British America’, in Bernard Bailyn and Philip D. Morgan (eds),
Strangers within the Realm: Cultural Margins of the First British Empire (Chapel Hill: University of
North Carolina Press, 1991), 226; Michael Hechter, Debra Friedman, and Satoshi Kanazawa, ‘The
Attainment of Social Order in Heterogeneous Societies’, in Michael Hechter (ed.), Theories of Social
Order: A Reader (Palo Alto, CA: Stanford University Press, 2003), 331.
15 George M. Fredrickson, White Supremacy: A Comparative Study in American and South African
History (New York: Oxford University Press, 1981), 5.
16 Eyal Ben-Ari and Yoram Bilu, ‘Introduction’, in eidem (eds), Grasping Land: Space and Place in
Contemporary Israeli Discourse and Experience (Albany: SUNY Press, 1997), 6–7.
17 Patrick Wolfe, Settler Colonialism and the Transformation of Anthropology: The Politics and Poetics
of an Ethnographic Event (London: Cassell, 1999), 2.
18 Patrick Wolfe, ‘Land, Labor, and Difference: Elementary Structures of Race’, American Historical
Review 106:3 (2001), 871.
19 Ibid. 874.
conflicting modes of production could not ultimately coexist’.20 In the United

States, however, historians of colonial North America and the republic of the
United States have generally overlooked Wolfe’s framework.
Part of the reason that American historians have neglected Wolfe’s analysis lies in
the well-established and nuanced historiography of English–Indian relations and
colonial violence in colonial North America.21 American historians have analysed in
great detail the complexity of colonialism during the seventeenth and eighteenth
centuries, observing that European settlers, traders, militiamen, and explorers ar-
rived in colonial North America with very different motives. Some, such as French
and Spanish traders, were motivated by a desire to seek their fortune in the
Americas.22 It was therefore in the best interest of traders, and the trickle of settlers
to regions as diverse as Quebec, New England, the Chesapeake, and the Mississippi
and Arkansas, to cultivate trade and diplomatic relations with indigenous Amer-
icans. In this vast colonial context, the ‘elimination’ of the ‘native’ was not a viable
option if trade, let alone settler colonialism, was to take root in North America.
Kathleen Duval has demonstrated that indigenous Americans in the Arkansas Valley
believed that ‘Incorporating alien [European] peoples proved to be an effective
means of maintaining and increasing power’, while ‘Colonial administrators gener-
ally delineated the boundaries of their empires by referring to the native peoples with
whom they had forged alliances’.23 This historiography suggests that there existed no
predetermined ‘structural’ rationale for violence or even genocide in colonial North
America; rather, early settler colonialism in North America was a messy business, as
Old World cultural ideals and socioeconomic structures were forced to adapt to new
landscapes and were creolized because Native Americans held the upper hand, at least
20 Ibid. 868. See also Patrick Wolfe, ‘Settler Colonialism and the Elimination of the Native’, Journal
of Genocide Research 8:4 (2006), 398.
21 Of an enormous historiography, see Richard White, The Middle Ground: Indians, Empires, and
Republics in the Great Lakes Region, 1650–1815 (Cambridge: Cambridge University Press, 1991);
Kathleen Brown, ‘Native Americans and Early Modern Concepts of Race’, in Martin Daunton and
Rick Halpern (eds), Empire and Others: British Encounters with Indigenous Peoples, 1600–1850
(Philadelphia: University of Pennsylvania Press, 1999), 79–100; Karen Ordahl Kupperman, Indians
and English: Facing off in Early America (Ithaca, NY: Cornell University Press, 2000); Ann Marie Plane,
Colonial Intimacies: Indian Marriage in Early New England (Ithaca, NY: Cornell University Press,
2000); James H. Merrell, ‘The Indians’ New World: The Catawba Experience’, in Stanley N. Katz, John
M. Murrin, and Douglas Greenberg (eds), Colonial America: Essays in Politics and Social Development
(Boston: McGraw Hill, 2001), 301–28; Jenny Hale Pulsipher, Subjects unto the Same King: Indians,
English, and the Contest for Authority in Colonial New England (Philadelphia: University of
Pennsylvania Press, 2005); Alan Taylor, The Divided Ground: Indians, Settlers and the Northern
Borderland of the American Revolution (New York: Alfred A. Knopf, 2006).
22 J. H. Elliott, Empires of the Atlantic World: Britain and Spain in America, 1492–1830 (New Haven,
CT: Yale University Press, 2006); Christopher Scmidt-Noware, The Conquest of History: Spanish
Colonialism and National Histories in the Nineteenth Century (Pittsburgh: University of Pittsburgh
Press, 2008).
23 Kathleen Duval, The Native Ground: Indians and Colonists in the Heart of the Continent
(Philadelphia: University of Pennsylvania Press, 2006), 8, 24, 26, 188–90.
initially, in terms of military strength, access to economic resources, and political

power. As Daniel Richter presciently observes, ‘British-American Indian relations
during the long eighteenth century defy a single narrative.’24
How, then, might we attempt to understand anti-Indian violence and episodes of
genocide in English colonial North America? While an ever growing historical
literature informs us that English colonists, like Spanish and French colonists, forged
socioeconomic networks and cultural identities that were adapted to the New World,
the English colonists maintained a dogged determination to see themselves, and the
settler communities they were creating, as English in political, social, and economic
form. Confronted with the racial and ethnic heterogeneity of settler colonialism in
North America, English colonists quickly developed a deep sense of anxiety about
their collective identity, political independence—from both England and local in-
digenous peoples—and their economic prosperity.25 In this respect, Michael Zucker-
man argues that the ‘early colonists were at once more free and more controlled,
more concerned about themselves and more sensitive to the opinions of others, than
their European forebears had been; and those who came after them . . . sought
simultaneously a new purity of personal identity and a new consummation of
conscious community.’26 In the binaries of civilization and wilderness that colonists
hoped would structure settler colonial life, and in the settler’s often-frustrated hopes
for the future acquisition of land, economic security, and political independence, the
seeds of anti-Indian violence and genocide germinated.
N AT I V E A M E R I C A N S A N D T H E Q U E S T I O N
O F C O LO N I A L G E N O C I D E
................................................................................................................
American historians routinely begin their assessments of settler colonialism by
pointing to the devastating consequences of disease among indigenous American
peoples. Alfred Crosby’s now famous description of ‘virgin soil epidemics’, or
24 Daniel K. Richter, ‘Native Peoples of North America and the Eighteenth-Century British
Empire’, in P. J. Marshall (ed.), The Oxford History of the British Empire: The Eighteenth Century
(Oxford: Oxford University Press, 2001), 347.
25 See for example Jean M. O’Brien, Dispossession by Degrees: Indian Land and Identity in Natick,
Massachusetts, 1650–1790 (New York: Cambridge University Press, 1997), 10; James Horn and Philip D.
Morgan, ‘Settlers and Slaves: European and African Migrations to Early Modern British America’, in
Elizabeth Mancke and Carole Shammas (eds), The Creation of the British Atlantic World (Baltimore:
Johns Hopkins University Press, 2005), 39–40.
26 Michael Zuckerman, ‘Identity in British America: Unease in Eden’, in Nicholas Canny and
Anthony Pagden (eds), Colonial Identity in the Atlantic World, 1500–1800 (Princeton: Princeton
University Press, 1987), 130.
outbreaks of diseases to which Native Americans had no previous exposure and

were ‘therefore immunologically almost defenseless’, have led historians to estimate
that Native American populations declined by as much as ninety per cent within a
century of first contact.27 Russell Thornton argues that after suffering the debilitat-
ing consequences of disease in the colonial era, the Native American population
continued to decline during the nineteenth century, falling from 600,000 in 1800,
to 228,000 by 1890.28
Observing the death of indigenous people from disease in New England, John
Winthrop observed in 1629 that ‘God hath consumed the Natives wth a great plague
in those parts soe as there be few inhabitants left.’29 Such comments have led some
historians to speculate that disease transfer was an example of the genocidal
consequences of settler colonialism in North America. According to Steven Katz,
Native American depopulation from disease was an unintentional consequence of
settler colonialism. He argues that ‘the greatest demographic disaster in history, the
depopulation of the New World, for all its death and terror, was largely an
unintended tragedy, a tragedy that occurred despite the sincere and indisputable
desire of Europeans to keep the Indian population alive’.30 According to Katz, the
spread of European diseases and the death of Native American populations was not
27 Alfred W. Crosby, ‘Virgin Soil Epidemics as a Factor in the Aboriginal Depopulation in America’,
William and Mary Quarterly 3rd ser. 33:2 (1976), 289; Alfred Cave, ‘Genocide in the Amercias’, in Dan
Stone (ed.), The Historiography of Genocide (Hampshire: Palgrave Macmillan, 2008), 273.
28 Russell Thornton, ‘Aboriginal Population Size of North America’, in Michael R. Haines and
Richard H. Steckel (eds), A Population History of North America (New York: Cambridge University
Press, 2000), 24. See also Henry F. Dobyns, ‘Disease Transfer at Contact’, American Review of
Anthropology 22 (1993), 275–6; James Mooney, ‘The Aboriginal Population of America North of
Mexico’, in John R. Swanton (ed.), Smithsonian Miscellaneous Collections, vol 80 (Washington, DC:
US GPO, 1928), 1–40; William Cronin, Changes in the Land: Indians, Colonists, and the Ecology of New
England (New York: Hill and Wang, 1983), 88; Henry F. Dobyns, Their Number Become Thinned: Native
American Population Dynamics in Eastern North America (Knoxville: University of Tennessee Press,
1983); Russell Thornton, ‘Aboriginal North American Population and Rates of Decline, ca. a. d. 1500–
1900’, Current Anthropology 38:2 (1997), 310–15; Melissa L. Meyer and Russell Thornton, ‘Indians and
the Numbers Game: Quantitative Methods in Native American History’, in Colin G. Calloway (ed.),
New Directions in American Indian History (Norman: University of Oklahoma Press, 1992), 5–30.
29 Winthrop quoted in Peter C. Mancall (ed.), Envisioning America: English Plans for Colonization
of North America, 1580–1640 (Boston: Bedford/St. Martin’s, 1995), 137.
30 Steven T. Katz, The Holocaust in Historical Context, vol i: The Holocaust and Mass Death before
the Modern Age (New York: Oxford University Press, 1992), 20. Native American historians have taken
exception to this view, seeing a seamless connection between what they interpret as the impact of
disease and the ensuing cultural and physical genocide of Native American peoples. See George E.
Tinker, Missionary Conquest: The Gospel and Native American Cultural Genocide (Minneapolis:
Fortress Press, 1993), 4–5; Ward Churchill, A Little Matter of Genocide: Holocaust and Denial in the
Americas, 1492 to the Present (San Francisco: City Lights Books, 1997), 137–43. See also Edward D.
Castillo’s preface in Clifford E. Trafzer and Joel R. Hyer (eds), Exterminate Them! Written Accounts of
Murder, Rape, and Enslavement of Native Americans during the California Gold Rush (East Lansing:
Michigan State University Press, 1999), x.
an intentional effort to perpetuate genocide.31 In contrast, Ward Churchill refers to

settler colonialism in North America as ‘the American holocaust’, and David
Stannard similarly portrays the European colonization of the Americas as an
example of ‘human incineration and carnage’.32
Over the past forty years, American historians have endeavoured to move beyond
the polemical debate about disease transfer in an effort to understand the nuances of
settler rationalizations for the often-violent treatment of Native Americans.33
A closer recognition for the importance of chronology has been critical to such
analysis. For example, the historian Karen Kupperman has observed that the initial
encounters between northeastern Native Americans and English settlers were not
characterized by violence, but mutual curiosity and a desire for trade, something
particularly important to the survival of English settlements.34 To the west, Richard
White’s analysis of Algonquin–French encounters in the Great Lakes region notes
that ‘campaign[s] of genocide’ against indigenous peoples did not occur until long
after the early years of contact, years characterized by the mutual search for trade
allies.35 Kupperman and White remind us that settler colonialism in North Ameri-
ca occurred in uneven chronological and geographical phases. What is important is
that only after traders and settlers established the foundations of permanent settler
communities, as Wolfe’s framework suggests, did competition for land and natural
resources intensify, and some of the most extreme examples of settler colonial
violence emerged.
Chesapeake Bay settlers provide us with an early example of frontier violence.
Between 1607 and 1644, scholars point out that a series of ‘wars’ were fought
between the English and the Powhatan Indians. These battles were sparked by
the aggressive aspirations of English traders, the desire of settlers for land and
natural resources, and the sense among the Powhatan that the English were not
conforming to local trade practices and land-use agreements.36 By 1622 the English
in Virginia had become particularly nervous about the intentions of local Indians
31 M. Annette Jaimes offers a sustained analytical rethinking of disease transfer and genocide in her
The State of Native America: Genocide, Colonization, and Resistance (Boston: South End Press, 1992),
31–3.
32 David E. Stannard, American Holocaust: The Conquest of the New World (New York: Oxford
University Press, 1992), xi, 62, 69.
33 R. David Edmunds, ‘The Indian in the Mainstream: Indian Historiography for Teachers of
American History Surveys’, History Teacher 8:2 (1975), 242–62; Reginald Horsman, ‘Well-Trodden
Paths and Fresh Byways: Recent Writing on Native American History’, Reviews in American History
10:4 (1982), 234–44.
34 Kupperman, Indians and English, 173, passim. See also Norbert Finzsch, ‘“The Aborigines . . . were
never annihilated, and still they are becoming Extinct”: Settler Imperialism and Genocide in Nineteenth-
Century America and Australia’, in Moses (ed.), Empire, Colony, Genocide, 254.
35 Richard White, Middle Ground: Indians, Empires, and Republics in the Great Lakes Region, 1650–1815
(New York: Cambridge University Press, 1991), 25, 168.
36 April Lee Hatfield, Atlantic Virginia: Intercolonial Relations in the Seventeenth Century
(Philadelphia: University of Pennsylvania Press, 2007), 24.
after Powhatan warriors murdered several Englishmen. To the English, these

murders reflected the increasingly aggressive stance that the Powhatan were assum-
ing towards them. In this uneasy colonial context, settlers needed little prodding to
take up arms after they received orders from the Virginia Company ‘to root out
[the Powhatan Indians] from being any longer a people . . . Wherefore, as they have
merited, let them have a perpetual war without peace or truce, and, although they
have desired it, without mercy, too.’37
A similar pattern of mutual suspicion and violence emerged in other North
American colonies. In 1644, Dutch authorities in New Amsterdam (later renamed
New York) hired John Underhill to burn and slaughter some five hundred Native
Americans after settlers became nervous about ‘unprovoked’ attacks on colonial
communities.38 Violent clashes with Native Americans produced in settlers a
growing suspicion of the ‘wild’ frontier Indians, a psychology that in moments
of extreme stress or anxiety resulted in a determination to exterminate the local
indigenous populations. In 1763, for example, Pennsylvania settlers, led by the
infamous Paxton boys, surrounded and killed twenty peaceful Susquehannock
Indians after a period of heightened settler anxiety.39
One of the most controversial historical case studies of English–Indian violence
occurred in New England between 1636 and 1637. The Pequot War had its origins in
the Pequot Indians determination to prevent rival Indian tribes—particularly the
Narragansett—from attaining more favourable trade relations with Dutch and
English traders in the Connecticut River Valley. Pequot leaders felt that the
emergence of the Massachusetts Bay Colony, and its increasing trade with rival
Indian tribes, squeezed the Pequot out of a position of dominance with European
traders. This loss of influence proved irksome to Pequot leaders and worried Dutch
merchants. The Dutch attempted to shore up trade relationships with Indian tribes
by drafting formal agreements. For example, in 1633 the Dutch–Pequot Agreement
was designed to give the Dutch access to trade with Indian tribes other than the
Pequot. This trade arrangement rankled Pequot warriors, and prompted them to
assault and kill traders from the rival Narragansett tribe. A series of Dutch–Pequot
reprisals and counter-reprisals ensued, culminating in Pequot warriors mistaking
the captain of an English trading vessel, John Stone, and six of his crew, as
37 James Mooney, ‘The Powhatan Confederacy, Past and Present’, American Anthropologist 9 (1907),
138; John Esten Cooke, Virginia: A History of the People, 6th edn (Boston: Houghton, Mifflin, 1885),
124–5; Hatfield, Atlantic Virginia, 234; Ben Kiernan, Blood and Soil: A World History of Genocide and
Extermination from Sparta to Darfur (New Haven, CT: Yale University Press, 2007), 219–25.
38 John J. Anderson and Alexander Clarence Flick, A Short History of the State of New York (New York:
Maynard, Merrill, 1902), 31.
39 James H. Merrell, Into the American Woods: Negotiators on the Pennsylvania Frontier (New York:
Norton, 1999), 16, 37–9.
Dutchmen.40 The raid and subsequent murder of Stone and his crew prompted
Massachusetts Bay authorities to send a military party under the leadership of John
Endecott to ensure the peace of the colonists ‘by killing the barbarians’.41
It is at this point that historians express a wide range of views about the historical
significance of the Pequot War. Specifically, historians disagree over the nature of the
Puritan attack on a Pequot village at West Mystic, Connecticut, on 26 May 1637. At
West Mystic, English soldiers and their Narragansett allies killed approximately four
hundred Pequot men, women, and children, and also burned the Pequot village. The
English hoped that this attack would cripple the Pequot. The following year the
English attempted to seal the fate of the Pequot. The Treaty of Hartford extinguished
the legal right of surviving Pequot Indians from using their tribal name or attaining
access to ancestral lands. Through a combination of violence and legal manoeuvring,
the Puritans helped to establish English settler colonial patterns of economic, politi-
cal, and social practices designed to separate the Indians from the English, and
ultimately exterminate indigenous peoples from the land that settlers coveted.42
Steven Katz argues that the Pequot War was neither genocidal nor racially
motivated. He insists that because Indian warriors and English soldiers fought
together against the Pequot, the Pequot War was therefore non-racial in charac-
ter.43 Moreover, the regrouping of the Pequot by 1650 proved that genocidal intent
did not motivate the English.44 In a forceful response, Michael Freeman argues that
the violence of the Pequot War and the subsequent Treaty of Hartford did indeed
reflect Puritan intentions to commit both biological and cultural genocide.45
40 Jennings, Invasion of America, ch. 13; Neal Salisbury, Manitou and Providence: Indians,
Europeans, and the Making of New England, 1500–1643 (New York: Oxford University Press, 1984),
214–25; Adam J. Hirsch, ‘The Collision of Military Cultures in Seventeenth-Century New England’,
Journal of American History 74:4 (1988), 1194–6; Alden T. Vaughan, New England Frontier: Puritans and
Indians, 1620–1675, 3rd edn (Norman: University of Oklahoma Press, 1995), ch. 5; Alfred A. Cave, The
Pequot War (Amherst: University of Massachusetts Press, 1996), 56–8, 72–3.
41 Cave, The Pequot War, 109.
42 Jennings, Invasion of America, 220–6; Vaughan, New England Frontier, 144–7; Salisbury, Manitou
and Providence, 221–3; Hirsch, ‘The Collision of Military Cultures’, 1202–3; Cave, Pequot War, 146–51;
Wilbur R. Jacobs, ‘British-Colonial Attitudes and Policies toward the Indian in the American
Colonies’, in Howard Peckham and Charles Gibson (eds), Attitudes of Colonial Powers toward the
American Indian (Salt Lake City: University of Utah Press, 1969), 95.
43 Steven T. Katz, ‘The Pequot War Reconsidered’, New England Quarterly 64:2 (1991), 206–24. Katz
takes aim at what he labels Frances Jennings’s ‘genocidal thesis’, and Richard Drinnon’s ‘genocidal
intentions’ argument in his Facing West. Francis Jennings, The Invasion within: Indians, Colonialism,
and the Cant of Conquest (Chapel Hill: University of North Carolina Press, 1975).
44 Katz, ‘Pequot War Reconsidered’, 222–3.
45 Katz rejects the bulk of Freeman’s argument by insisting that acts of genocide must be defined by
the criteria of ‘actualized intentions’. Steven T. Katz, ‘Pequots and the Question of Genocide: A Reply
to Michael Freeman’, New England Quarterly 68:4 (1995), 641–9, esp. 648; Michael Freeman, ‘Puritans
and Pequots: The Question of Genocide’, New England Quarterly 68:2 (1995), 290.
Borrowing the words of one Puritan captain, Freeman insists that the English
intent was ‘to cut off the remembrance of them [the Pequot] from the earth’.46
The Pequot War, as was the case with the Powhatan massacre in Virginia, highlighted
one of the most important patterns in English–Indian relations: as settlers established
communities and sociopolitical hierarchies developed, the presence of Native American
peoples elicited a nagging sense of vulnerability and a growing racialized distrust of
American Indians, and sparked outbursts of anti-Indian violence. In the cases of the
Powhatan massacre and Pequot War, the perception of imminent Indian attacks
prompted English aggressions with the objective of protecting settler society by elim-
inating the threat posed by the Native American warriors. Therefore, the actions of
Virginian and Puritan settlers can be interpreted as offensive actions to defend an
emerging settler colonial order that settlers imagined themselves creating. This was why
John Winthrop claimed that ‘we went not to make war upon them [the Pequots], but to
do justice, etc.’47 To Puritan leaders, as with colonial officials in the mid-Atlantic and
Chesapeake, doing ‘justice’ meant eliminating the Indian threat to settler society.48
Thus, imbedded in the fluid economic structures, developing political hierarchies, and
colonial culture of eighteenth-century North America was an increasingly explicit belief
that exterminating Native American peoples could indeed be seen as a ‘just’ means for
settler colonialism to expand from the Atlantic to the Pacific Ocean.
THE AMERICAN REPUBLIC AND THE

QUESTION OF GENOCIDE
................................................................................................................
By the mid-eighteenth century, settler colonial patterns of settlement, colonial poli-
tics, and economic exchange informed Anglo-American expansion into the southern
and western frontiers of North America. In the south and southwest in particular,
settler colonial expansion involved the dispossession of Native Americans and the
forced labour of African-American slaves, factors that united all white Americans in a
socioeconomic community of common interest. The historian Norbert Finzsch refers
to American territorial expansion after the Revolutionary War as ‘settler imperialism’,
or ‘a process of constant deterritorialization and reterritorialization’.49 American
settler imperialism was in many ways influenced by the social and legal precedents
46 Freeman, ‘Puritans and Pequots’, 280, 285, 289.

47 Ronald Dale Karr, ‘ “Why Should You Be So Furious?”: The Violence of the Pequot War’, Journal
of American History 85:3 (1998), 903; Kiernan, Blood and Soil, 325–6.
48 Karr, ‘ “Why Should You Be So Furious?” ’, 904.
49 Finzsch, ‘ “The Aborigines . . . were never annihilated” ’, 261.
set in eighteenth-century North America. For example, as early as 1732, the Preamble
to the Georgia Charter explicitly stated that ‘Our provinces in North America have
been frequently ravaged by Indian enemies.’50 Adding to the racial anxiety that
accompanied Anglo-American settler expansion was the ever present fear of slave
revolts. For instance, in the wake of the Stono rebellion in South Carolina in 1739,
colonial officials observed that the uprising of African-American slaves reminded
settlers that ‘every one that had a life to lose were in the most sensible Manner shocked
at such Danger daily hanging over their Heads.’51 So important was white supremacy
to the social order in settler colonial North America that the Revolutionary War
between the colonists and Britain unleashed a new wave of anti-Indian and anti-
Negro sentiment. Indeed, rather than sooth racial anxieties, political independence
from Britain magnified them. Thus, in 1790 the US Congress passed legislation that
defined citizenship on the basis of race. The Naturalization Act stated that
all free white persons who, have, or shall migrate into the United States, and shall give
satisfactory proof, before a magistrate, by oath, that they intend to reside therein, and shall
take an oath of allegiance, and shall have resided in the United States for one whole year,
shall be entitled to the rights of citizenship.52
The emergence of a racially defined American republic in North America added

a level of ideological intensity to colonial patterns of frontier violence and the
Anglo-American quest for territorial and economic aggrandizement. Standing in
the path of a republican settler colonial empire that stretched from the Atlantic to
the Pacific were the Native American communities of the south, southwest, the
Great Plains, and the west. American explorers, ranchers, settlers, and gold-seekers
took to the United States’ frontiers of settlement their hopes and dreams for a
better life, and a moral repugnance for indigenous peoples that expressed itself, in
the words of one recent scholar, as the ‘beastilization’ of Native Americans.53 It is in
the nineteenth- and twentieth-century context of Anglo-American expansion that
Patrick Wolfe’s ‘eliminationist’ framework best helps us understand the often-
genocidal violence and government policies associated with the Cherokee ‘Trail
of Tears’ in the late 1830s, and military expeditions against American Indians from
the southeast, Great Plains, and the Pacific coast that resulted in such events as the
Mountain Meadow Massacre in Utah (1857), the Bannock River Massacre in Idaho
50 Matthew Frye Jacobson, Whiteness of a Different Color: European Immigrants and the Alchemy of
Race (Cambridge, MA: Harvard University Press, 1998), 24.
51 Peter H. Wood, ‘Anatomy of a Revolt’, in Mark M. Smith (ed.), Stono: Documenting and
Interpreting a Southern Slave Revolt (Columbia: University of South Carolina Press, 2005), 59.
52 Jacobson, Whiteness of a Different Color, 22.
53 Karl Jacoby, ‘ “The Broad Platform of Extermination”: Nature and Violence in the Nineteenth-
Century North American Borderlands’, Journal of Genocide Research 10:2 (2008), 249–67; See also Ned
Blackhawk, Violence over the Land: Indians and Empires in the Early American West (Cambridge, MA:
Harvard University Press, 2006), 172–3.
(1860), and the many postbellum ‘expeditions’ against the Sioux in Wyoming and
Nebraska, and the Ute in Colorado, to name just a few examples.54
One of the clearest examples of how indigenous communities were adversely
impacted by the disease and settler violence that accompanied Anglo-American expan-
sion was in California. Albert Hurtado observes that Spanish colonialism in California
led to the dramatic decline of the Indian population from approximately 300,000 to
200,000 between 1769 and 1821. By the end of the 1850s, and in the wake of the California
Gold Rush, the Native American population teetered around 30,000. The rapid decline
in California’s Native American population was the result, Hurtado argues, of ‘disease,
starvation, homicide, and a declining birthrate’.55 Thus, California’s indigenous com-
munities struggled to endure European diseases and violence, first from Spanish
colonists and missionaries moving north from south and central America, and by the
mid-nineteenth century, Anglo-American colonists migrating from the east to the west
coast of North America.56 Hurtado’s statistics portray a settler colonial context in which
the mere survival of California’s indigenous communities hung in the balance.
The citizens and political leaders of the United States had coveted California and
other lands west of the Mississippi for much of the nineteenth century.57 The
discovery of gold in the 1840s added an additional material motivation in the drive
to colonize the west and eliminate Native Americans from land rich in natural
resources.58 As the number of US migrants increased in California in the 1840s
and 1850s, a written record of anti-Indian racism and violence emerged.59 This
54 See for example William L. Anderson, Cherokee Removal: Before and After (Athens: University of
Georgia Press, 1991); Dee Brown, Bury My Heart at Wounded Knee: An Indian History of the American West
(New York: Henry Holt, 1970); Jerry Keenan, Encyclopedia of American Indian Wars, 1492–1890 (New York:
W. W. Norton, 1999); Peter Cozzens, Conquering the Southern Plains: Eyewitnesses to the Indian Wars,
1865–1890 (Mechanicsburg, PA: Stackpole Books, 2003); Jeffrey Ostler, The Plains Sioux and U. S.
Colonialism from Lewis and Clark to Wounded Knee (Cambridge/New York: Cambridge University Press,
2004); Benjamin Madley, ‘Patterns of Frontier Genocide 1803–1910: The Aboriginal Tasmanians, the Yuki
of California, and the Heroero of Namibia’, Journal of Genocide Research 6:2 (2004), 167.
55 Albert L. Hurtado, Indian Survival on the California Frontier (New Haven: Yale University Press,
1988), 1, 66–8, passim; Kiernan, Blood and Soil, 350–1.
56 Douglas Monroy, ‘Guilty Pleasures: The Satisfaction of Racial Thinking in Early-Nineteenth-
Century California’, in Paul Spickard (ed.), Race and Nation: Ethnic Systems in the Modern World (New
York: Routledge, 2005), 33–52.
57 Estwick Evans, A Pedestrious Tour, of Four Thousand Miles, through the Western States and
Territories, during the Winter and Spring of 1818. Interspersed with Brief Reflections upon a Great
Variety of Topics: Religious, Moral, Political, Sentimental, &c. &c. (Concord, NH: Joseph C. Spear,
1819), 198; James M. McCaffrey, Army of Manifest Destiny: The American Soldier in the Mexican War,
1846–1848 (New York: New York University Press, 1994), 67–8; Hurtado, Indian Survival, 86–7; John S.
D. Eisenhower, So Far From God: The U. S. War with Mexico, 1846–1848 (Norman: University of
Oklahoma Press, 2000), 19, 172, 269, 286;
58 Hurtado, Indian Survival, 97, ch. 6; Alan G. Bogue, ‘An Agricultural Empire’, in Clyde A. Milner
II, Carol A. O’Connor, and Martha A. Sandweiss (eds), The Oxford History of the American West (New
York: Oxford University Press, 1994), 275–313; H. W. Brands, The Age of Gold: The California Gold Rush
and the New American Dream (New York: Anchor Books, 2002).
59 Madley, ‘Patterns of Frontier Genocide’, 169.
literature was framed by a belief in the ‘White man’ being both racially superior to
the California Indians and the embodiment of civilization. In contrast, California’s
Indian population embodied a ‘savage’ impediment for settlers to overcome in
their desire to exploit the natural resources of California.60 Writing in 1854, the
historian Elisha Smith Capron argued that Spanish missionary efforts to bring
‘civilization’ to California’s Indians had failed. Capron wrote:
The general characteristics of the Indians of Upper California may be inferred, with much
correctness, from what has been said respecting the missions. It is very certain that the
reverend fathers would not have been able, with the means they had, and the system they
pursued in California, to tame the spirit and enslave the bodies of the tall, athletic, haughty-
souled savages of the more northern and eastern country.61
Missionary efforts to Christianize California’s Indians falls under what Lemkin

called a ‘subtle kind of cultural genocide’.62 ‘Cultural genocide’ included such
practices as taking Indian children from their parents and placing them in mission
schools, imposing lessons in Christianity and Western forms of economic practices,
and most significantly, teaching indigenous children a European language so that
linguistic ties with indigenous cultural traditions would be forever severed.63 With
the lessons of Spanish colonialism in California and Anglo settler expansion in the
eastern United States to draw upon, Anglo-Americans understood implicitly that
the success of settler imperialism, or what I have labelled elsewhere ‘republican
settler colonialism’, hinged on their ability to control, transform, and if necessary,
eliminate both the untamed wilderness and the Native American peoples.64
Anglo-Americans in California articulated a progressive, forward-looking concep-
tion of their envisioned settler colonial civilization. They did this, in part, by ascribing
negative racial stereotypes to California Indians. For example, the Indian Agent E. A.
Stevenson argued in 1853: ‘Many of them [California Indians] being without any settled
place of habitation; and many of them have already imbibed the very worst vices of
civilization, and are becoming vitiated and degraded, a pest and nuisance to the
60 Madley, ‘Patterns of Frontier Genocide’, 177.

61 Elisha Smith Capron, History of California: From Its Discovery to the Present Time: Comprising a
Full Description of its Climate, Soil, Rivers, Towns, Beasts, Birds, Fishes, State of Its Society, Agriculture,
Commerce, Mines. &c. (Boston: John P. Jewett, 1854), 19.
62 Docker, ‘Are Settler-Colonies Inherently Genocidal?’, in Moses (ed.), Empire, Colony, Genocide, 90.
63 Bernard W. Sheehan, Seeds of Extinction: Jeffersonian Philanthropy and the American Indian (New
York: W. W. Norton, 1974), 139–41; Blanca Tovias, ‘Navigating the Cultural Encounter: Blackfoot
Religious Resistance in Canada (c. 1870–1930)’, in Moses (ed.), Empire, Colony, Genocide, 271; Gregory
D. Smithers, ‘The “Pursuits of the Civilized Man”: Race and the Meaning of Civilization in the United
States and Australia, 1790s–1850s’, Journal of World History 20:2 (2009), 245–72.
64 Smithers, ‘The “Pursuits of the Civilized Man” ’, 247.
localities where they resort.’65 White Americans insisted that the California Indians were
immoral savages because they lacked the intelligence to moderate their behaviour.66
Quite simply, the Indians represented an uncivilizable impediment to white settlement
and a threat to the safety of white settlers. As a result, Anglo-American settlers expressed
a determination ‘to exterminate these merciless foes, or drive them from us’.67
California settlers believed they were under siege from ‘savage’ Indians. According
to one nineteenth-century historian, the Anglo-American settlers knew from their
own experience that acts of Indian aggression were a product of the ‘character of the
Indian—a mischievous disposition and desire for plunder’ that convinced settlers
that they needed to ‘either unite and exterminate the Indians in their neighborhood
or withdraw from it altogether’.68 However, by the mid nineteenth-century, with gold
to be discovered, land to be farmed, and racial ‘science’ reinforcing the Anglo-Ameri-
can belief in white supremacy, settlers came to the conclusion that California’s Indians
presented an innate biological and sociocultural threat from within the expansionist
republic. Republican settlers, therefore, wrote often about the anxiety that California’s
indigenous inhabitants were causing them. Most Anglo-Americans agreed. California’s
natives were ‘wild Indians’, most being ‘in the habit of killing great numbers of hogs
and cattle and stealing provisions and other articles from the houses and camps of the
whites’.69 The alleged harassment of white settlers by California Indians continued
unabated for three decades after gold fever ended. One settler, Walter Van Dyke,
informed Governor Stanford in 1862 that the white residents in his county ‘have
been so harassed and frightened by the Indians that many had been forced out of
fear to leave their homes’. Van Dyke echoed the racialized anxieties harboured by most
white Californians, and opposed any government scheme to protect Native Americans
on reserves. According to Van Dyke, ‘It is impossible, utterly so, to keep these Indians
on any reservation’; therefore, it was not only ‘a waste of time but a swindle on the
Government and an insult and injury to our people to continue this plan’.70
California governors heard this message from their citizens on a regular basis.
Because white Californians felt utterly unprotected from what they saw as
65 Robert F. Heizer (ed.), The Destruction of California Indians (Santa Barbara: Peregrine Smith,
1974), 14. The quotation ‘White man’ was Stevenson’s, and is typical of the racialized language of
American settlers and officials to California.
66 These sentiments, as Benjamin Madley observes, were articulated in the face of white settlers
kidnapping Indian women and children. See Madley, ‘Patterns of Frontier Genocide’, 178.
67 James J. Rawls, Indians of California: The Changing Image (Norman: University of Oklahoma
Press, 1984), 178.
68 Theodore Henry Hittell, History of California, 4 vols (San Francisco: N. J. Stone, 1898), iii.904.
69 Citizens of Tehema County to Governor Weller, May 29, 1859, Folder F 3753: 348, Military
Department, Indian War Papers, 1850–1880, MF 3:6 (22), roll 3, California State Archives,
Sacramento, California [hereafter Indian War Papers, CSA]; Heizer, The Destruction of California
Indians, 130.
70 Walter Van Dyke to Governor Stanford, November 25, 1862, F 3753: 623, MF 3:6 (24), roll 3, Indian
War Papers, CSA. See also Traftzer and Hyer (eds.), Exterminate Them!, 40, 44.
unprovoked Indian attacks, many settlers decided to take matters into their own
hands and dispensed a brand of frontier justice designed to force the Indians off
their homelands and into the mountainous regions of California. The Sierra, for
example, presented a bleak prospect for many tribes, as hunger and the desperate
quest for survival forced Indians to act in ways that whites deemed threatening and
uncivilized.71 The Native American struggle for survival did not register with most
settlers, nor did it engender sympathy from the governors of California, for most
supported the genocidal inclinations of the white population. In 1851, Governor
Burnett gave official voice to the genocidal intent of settler violence against the
California Indians, stating: ‘[T]he white man, to whom time is money, and who
labors hard all day to create the comforts of life, cannot sit up all night to watch his
property . . . after being robbed a few times he becomes desperate, and resolves
upon a war of extermination.’ According to Burnett, ‘A war of extermination will
continue to be waged between the races until the Indian race becomes extinct.’72
John McDougal, Burnett’s successor in the governor’s office, made the genocidal
intent of the State of California even clearer. McDougal warned that the State
would ‘make war upon the [Indians] which must of necessity be one of extermi-
nation to many of the tribes’.73
Not all California governors—or settlers for that matter—supported these
genocidal intentions.74 Governor Weller, for instance, insisted that the State’s
responsibility was ‘not to wage a war of extermination against the Indians’, but to
protect lives and property.75 Unfortunately for Weller, the official sanction that his
predecessor gave to genocidal operations had become entrenched in settler culture
and behaviour. For example, in November 1850, William Rogers, who had been
directed by the California governor to form a militia of two hundred men, reported
the discovery of fifteen dead Mowok. Rogers reported that there existed ‘numerous
Trails Marked with Blood’, suggesting the death toll could have been much higher.
Even more brutal, the Daily Alta California reported in 1852 that in Trinity County
a ‘rancheria of 148 Indians, including women and children, was attacked, and
nearly the whole number destroyed’.76 These accounts highlight what one contem-
porary reported was ‘a deliberate design to exterminate the Indian race’.77
71 Heizer (ed.), The Destruction of the California Indians, 282. See also Frank H. Baumgardner,
Killing for Land in Early California: Indian Blood at Round Valley: Founding the Nome Cult Indian
Farm (New York: Algora, 2005), 2.
72 Hurtado, Indian Survival, 134–6; Edward D. Castillo, ‘Preface’, in Traftzer and Hyer (eds),
Exterminate Them!, x; Kiernan, Blood and Soil, 351–2.
73 Hurtado, Indian Survival, 134, 136.
74 Madley, ‘Patterns of Frontier Genocide’, 179. Baumgardner, Killing the Land in Early California, 11.
75 Governor Weller to Capt. W. Jarboe, October 23, 1859, F 3753: 399, Indian War Papers, MF 3:6
(22), roll 1, CSA; Baumgardner, Killing for Land in Early California, 160.
76 Heizer (ed.), The Destruction of the California Indians, 249.
77 Ibid. 254.
The genocidal intent of California settlers and government officials was acted out
in numerous battles and massacres (and aided by technological advances in weapon-
ry, especially after the Civil War), in the abduction and sexual abuse of Indian women,
and in the economic exploitation of Indian child labourers.78 While the California
government proved reluctant to pass laws that would have protected California
Indians from white attacks, the federal government responded with a combination
of indifference and incompetence. The response—or lack there of—from the Cali-
fornia and federal officials sent white settlers the message that no official obstacle
would impede their efforts to culturally and physically destroy the California Indians.
The intent behind settler attacks on California’s Native American population was
therefore unequivocal: assert white supremacy and exterminate the Indians. Accord-
ing to an anonymous correspondent to the Daily Alta California in 1851, ‘God has
given us in California a goodly heritage’ and the ‘well-known energy and enterprise of
the Anglo-Saxon race’ was claiming its Providential inheritance from the California
Indians.79 By the early twentieth century, the American educator and journalist Henry
Kittredge Norton articulated the Anglo-American belief that the West had been
cleared of indigenous communities and successfully colonized by republican settler
societies. Kittredge claimed that ‘nowhere are there any Indian neophytes to be seen,
for with a few exceptions the descendents of the California Indians are in their graves,
literally exterminated by the onward march of a stronger race’.80
C O N C LU S I O N
................................................................................................................
Colonial North America in the seventeenth and eighteenth centuries constituted an
ever growing number of racially and ethnically heterogeneous sites of trade,
exploration, and settlement. As Europeans ventured westward into the North
American wilderness, territorial expansion, changing land-use patterns, new eco-
nomic networks, and different systems of coerced labour all motivated settlers to
think and act with different colonial motives that contributed to a sense of
instability and flux in settler communities. What bound Europeans together, and
provided the ideological and political basis for ordering settler societies, was an
increasingly explicit racialized anxiety and disgust for Native Americans. The
78 Ibid. Chs. 6–9 provide primary source accounts of such activities. See also Lt. Edward Dillen to
Maj. W. W. Mackall, January 27, 1860, F 3735: 423, Indian War Papers, MF 3:6 (23), roll 2, CSA.
79 Traftzer and Hyer (eds), Exterminate Them!, 42.
80 Henry Kittredge Norton, The Story of California from the Earliest Days to the Present, 2nd edn
(Chicago: A. C. McClurg, 1913), 142.
settlers’ sense of disgust was important to the genocidal intentions behind different
forms of colonial violence. Certainly, by the Revolutionary era white Americans
saw Indian attacks on their communities as examples of a ‘savage’ and ‘treacherous’
race engaging in unprovoked pillaging.81 As David Hollenbach explains, ‘repeated
exposure to images of violence . . . can give rise to public moral disgust, leading to a
haughty sense of superiority, [and] a deepened perception of divisions between the
civilized “us” and the savage “them”.’82 This binary logic lay at the centre of a
developing colonial order of civilized settlers and uncivilized Indians (and Afri-
cans, for that matter), and led to phases of officially sanctioned efforts to extermi-
nate the ‘natives’, as was the case at Fort Mystique, or inspired frontier settlers
anxious to defend their land and homes from Indian attacks to violently strike out
against indigenous people with the intent of exterminating the threat posed to
settler colonial society.83
By the time of the early republic and antebellum era, the cultural, legal, eco-
nomic, and social precedents established by colonial relations between settlers and
Indians led frontier settlers and government officials to adopt a loose mixture of
official military incursions against indigenous warriors, or inspired the type of
frontier exterminatory violence that state and federal officials sanctioned either in
word or through inaction in California.84 In numerous other sites across the
American republic’s vast western frontier, a combination of military and settler
violence placed many indigenous communities on the edge of extinction by the
early twentieth century. Much research remains to be done on the genocidal
violence visited upon the Native American peoples of the Great Plains and south-
west.85 By placing this analysis in a settler colonial framework, as suggested by
scholars such as Patrick Wolfe and Ian Tyrell, the death and destruction experi-
enced by Native Americans in the half decade after the American Civil War (1861–5)
emerges less as an anomalous moment in US history, and comes into focus as an
important example of the transnational drive of Western settler colonial powers to
dispossess the world’s indigenous peoples, reorder nature by imposing Western
political and economic structures on the landscape, and if necessary, eliminating
81 Gregory D. Smithers, Science, Sexuality, and Race in the United States and Australia, 1780s–1890s
(New York: Routledge, 2008), ch. 1.
82 David Hollenbach, The Common Good and Christian Ethics (Cambridge: Cambridge University
Press, 2002), 237.
83 Kiernan, Blood and Soil, 234.
84 Benjamin A. Valentino, Final Solutions: Mass Killings and Genocide in the Twentieth Century
(Ithaca, NY: Cornell University Press, 2004), 50–1.
85 Excellent new research is being done, for example, by Ostler, The Plains Sioux, 15, 39, passim; Rob
Harper, ‘Looking the Other Way: The Gnadenhutten Massacre and the Contextual Interpretation of
Violence’, William and Mary Quarterly 3rd ser. 64:3 (2007), 621–44; Kiernan, Blood and Soil, ch. 8;
Madley, ‘Patterns of Frontier Genocide’; Ned Blackhawk, Violence over the Land: Indians and Empires
in the Early American West (Cambridge, MA: Harvard University Press, 2006); Jacoby, ‘ “The Broad
Platform of Extermination” ’.
indigenous people from the land.86 As Richard Cole Harris observes, the ‘geo-
graphical core’ of settler colonialism—to which I include the United States—was
‘about the displacement of people from their land and its repossession by others’.87
This basic analytical premise inspired what Ben Kiernan refers to as the ‘selective
threat of genocide’ against Native Americans, a threat that resulted in various
forms of organized and spontaneous acts of violence since the seventeenth century
was driven by the intent to exterminate the American ‘natives’.88
F U RT H E R R E A D I N G
Anderson, William L., Cherokee Removal: Before and After (Athens: University of Georgia
Press, 1991).
Baumgardner, Frank H., Killing for Land in Early California: Indian Blood at Round Valley:
Founding the Nome Cult Indian Farm (New York: Algora, 2005).
Blackhawk, Ned, Violence over the Land: Indians and Empires in the Early American West
Brown, Dee, Bury My Heart at Wounded Knee: An Indian History of the American West
(New York: Henry Holt, 1970).
Brown, Kathleen, ‘Native Americans and Early Modern Concepts of Race’, in Martin
Daunton and Rick Halpern (eds), Empire and Others: British Encounters with Indigenous
Peoples, 1600–1850 (Philadelphia: University of Pennsylvania Press, 1999), 79–100.
Cave, Alfred, The Pequot War (Amherst: University of Massachusetts Press, 1996).
Jacoby, Karl, ‘ “The Broad Platform of Extermination”: Nature and Violence in the Nine-
teenth-Century North American Borderlands’, Journal of Genocide Research 10:2 (2008):
249–67.
Madley, Banjamin, ‘Patterns of Frontier Genocide 1803–1910: The Aboriginal Tasmanians,
the Yuki of California, and the Heroero of Namibia’, Journal of Genocide Research 6:2
(2004), 167–92.
Merritt, Jane T., At the Crossroads: Indians and Empires on a Mid-Atlantic Frontier, 1700–1763
(Chapel Hill: University of North Carolina Press, 2003).
Smithers, Gregory D., Science, Sexuality, and Race in the United States and Australia, 1780s–1890s
(New York/London: Routledge, 2009).
86 Wolfe, ‘Land, Labor, and Difference’; Tyrell, ‘Beyond the View of Euro-America’, 170; Smithers,
‘The “Pursuits of the Civilized Man” ’.
87 Harris, Making Native Space, xxiv.
88 Kiernan, Blood and Soil, 322.
part iv
.............................................................................................
GENOCIDE IN THE
LATE MODERN
WORLD
.............................................................................................
chapter 17
.............................................................................................
GENOCIDE AND
M A SS VIOLENCE I N
T H E ‘ H E A RT OF
DA R K N E S S ’
AFRICA IN THE COLONIAL
PERIOD
.............................................................................................
dominik j. schaller
................................................................................................................
In autumn 2001, representatives of the Namibian Herero filed lawsuits against the
government of the Federal Republic of Germany, the Deutsche Bank, and two other
German companies at the US Federal Court. The so-called Herero People’s Repa-
ration Corporation claimed reparations in the amount of US$4 billion because of
these companies’ alleged involvement in a military campaign in the former Ger-
man colony of South West Africa between 1904 and 1908 that had resulted in the
deaths of almost 80,000 Africans and the destruction of the traditional Herero
society. The Herero were the first African people to demand reparations for
346 dominik j. schaller
colonial policies in terms of genocide.1 The lawsuit and the associated discussions
about German colonial policies in Africa have contributed to a heightened schol-
arly awareness of the close relationship between the phenomena of colonialism and
genocide.
This insight is not as new as first apparent. From the 1950s, the struggle of anti-
colonial resistance movements in Africa and Asia as well as the corresponding
process of decolonization mobilized intellectuals and led to political debates about
the exploitative nature of colonialism. Many activists, philosophers, and scholars
like Frantz Fanon, Hannah Arendt, Jean-Paul Sartre, and Georges Balandier re-
cognized and denounced the genocidal potential of colonialism, although most of
them did not resort to the term ‘genocide’ coined by the Polish-Jewish interna-
tional lawyer Raphael Lemkin in 1944.2 Lemkin himself was well aware that
genocide cannot be separated analytically from colonization when he stated:
Genocide has two phases: one, destruction of the national pattern of the oppressed group;
the other, the imposition of the national pattern of the oppressor. This imposition, in turn,
may be made upon the oppressed population which is allowed to remain or upon the
territory alone, after removal of the population and the colonization of the area by the
oppressor’s own nationals.3
For all that, Lemkin was not an anti-imperial or anti-colonial thinker like
Cesaire and Fanon. Although he regarded the German and Belgian excesses in
Central and Southern Africa as genocidal, he considered colonialism and the
European ‘mission civilisatrice’ an important step in overcoming the assumed
backwardness of the ‘dark continent’.4
Lemkin’s ambivalence about colonial rule in Africa persists in the field of
genocide studies. In much of the scholarly literature it is suggested that genocide
is an exclusively modern phenomenon. The murder of the Ottoman Armenians by
the Young Turks during World War I and the Holocaust are widely perceived as
ideal types of genocide because of the perpetrators’ alleged ‘irrational’ motives:
nationalism and racism. The perpetrators of colonial mass violence, by contrast,
are believed to be driven by ‘rational’ motives like greed and revenge. Such an
1 Allan D. Cooper, ‘Reparations for the Herero Genocide: Defining the Limits of International
Litigation’, African Affairs 106:422 (2007), 113–26 (113).
2 A. Dirk Moses, ‘Empire, Colony, Genocide: Keywords and the Philosophy of History’, in idem
(ed.), Empire, Colony, Genocide: Conquest, Occupation, and Subaltern Resistance in World History
(New York: Berghahn, 2008), 3–54.
3 Raphael Lemkin, Axis Rule in Occupied Europe (Washington DC: Carnegie Endowment for
International Peace, 1944), 79.
4 On this mission in Africa, see Alice L. Conklin, A Mission to Civilize: The Republican Idea of
Empire in France and West Africa, 1895–1930 (Stanford: Stanford University Press, 1997); Dominik J.
Schaller, ‘Colonialism and Genocide: Raphael Lemkin’s Concept of Genocide and Its Application to
European Rule in Africa’, Development Dialogue 50 (December 2008), 75–93; Moses, ‘Empire, Colony,
Genocide’.
genocide and mass violence in the ‘heart of darkness’ 347
artificially constructed distinction ignores the fact that génocidaires never rely on a
single motive, and it leads to the distorted but widespread conclusion that the
category ‘colonial genocide’ is analytically useless, and that no genocides occurred
in colonial Africa. In traditional socio-scientific genocide studies, the history of
European colonialism in Africa is often portrayed as a success story and a ‘triumph
of humanitarianism’, as the following excerpt from a popular ‘world history’ of
genocide demonstrates: ‘The effect of the European presence was manifestly
ameliorative, as were the long-term effects of the introduction of Western medi-
cine, education and legal and customary practices, and the inclusion of these places
in the modern global economic system.’5
Lemkin’s own multifaceted conception of genocide shows that the spread of
European civilization by colonial and imperial rule cannot be so easily enlisted in
an optimistic narrative of progress. Brutal suppression of indigenous resistance,
forced labour, and the expulsion of Africans from arable land were the norm rather
than the exception. Violence was the essential element in the management of
colonial empires.6 Could the inclusion of Africans in a modern European-style
capitalist economy be seen as a blessing when it was designed to lead to the
deliberate destruction of indigenous social and political institutions as well as
African customs and cultural heritage?7 Lemkin had these policies (amongst
others) in mind when he developed the concept of genocide. For these reason,
the study of genocide in colonial Africa requires systematic methodological reflec-
tion. Five main issues need to be addressed before various aspects of colonial mass
violence in Africa can be further explored.
Firstly, the legacy of colonial rule in Africa is highly controversial. Whereas
Western apologists praise the supposed technological and cultural benefits of
European colonialism, African voices claim the ‘underdevelopment’ and almost
all current political and socio-economic problems from which their continent
suffers can directly be traced back to both the transatlantic slave trade and imperial
conquest. The most prominent adherent of this position was Walter Rodney whose
‘How Europe Underdeveloped Africa’ influenced a whole generation of Africa-
nists.8 Although the critical literature rightly highlighted how the roots of many
African conflicts lie in the colonial period, it tended to ignore African agency. Too
5 William D. Rubinstein, Genocide: A History (London: Longman, 2004), 102–3.

6 John McCulloch, ‘Empire and Violence, 1900–1939’, in Philippa Levine (ed.), Gender and Empire
(Oxford: Oxford University Press, 2004), 220–39.
7 The Africanist Mahmood Mamdani opposes this view. He argues that the British abandoned the
idea of ‘civilizing’ the Africans by the time of the ‘scramble of Africa’ and that they preferred to
conserve the ‘traditional’ African societies in order to facilitate their rule over the continent.
See Mahmood Mamdani, Citizen and Subject: Contemporary Africa and the Legacy of Late
Colonialism (Kampala: Fountain, 1996).
8 Walter Rodney, How Europe Underdeveloped Africa (London: Bougle-L’ouverture, 1972). Equally
important were the early works of Immanuel Wallerstein on Africa and European-style capitalism.
See Immanuel Wallerstein, ‘Africa in a Capitalist World’, Issue: A Journal of Opinion 3:3 (1973), 1–11.
often, Africans were portrayed as helpless victims of Western power and violence.
Nor did historiography pay much attention to indigenous resistance and how the
Africans’ manifold reactions to European imperialism fuelled the dynamics of
colonial violence. Colonial masters were far from omnipotent. There were limits
to their power claims and many Africans succeeded in defying the colonial system.
These insights are important for our understanding of the excessive violence
inherent to colonial wars in Africa: European settlers and colonial soldiers resorted
to genocidal methods of warfare when they were afraid that they could lose the
control over the situation. Even so, although African societies were deeply affected
by genocidal violence and economic exploitation, they usually managed to recover
and to remain living and innovative entities. African survival strategies and pro-
cesses of societal reconstruction after foreign rule and genocide have not attracted
the interest of historians until recently. Not only was disinterest responsible for the
neglect of these important aspects, but also most Western historians’ inability to
deal with original African sources, i.e., oral testimonies.
Thirdly, the genocide concept can oversimplify the genocidal conjuncture. As an
originally legal framework, it suggests a clear-cut distinction between perpetrators
and victims that is difficult to reconcile with empirical evidence.9 Africans were not
only victims of colonialism. The Europeans could not have conquered the conti-
nent without indigenous collaboration. Colonial armies and administrations con-
sisted mainly of African mercenaries and support staff, and some African groups
benefited enormously from the advent of the colonizers and relied on the Eur-
opeans’ help to climb to a position of power.10 Consider the Tutsi of Rwanda: it was
their rulers’ readiness to collaborate with the German invaders that contributed
decisively to the extension of their supremacy over the Hutu. The cases of the Zulu
in Southern Africa and of the Ethiopian Empire also show intra-African conquest
and even genocide. If both the Zulu and the Ethiopians fell victim to excessive
violence unleashed by British and Italian forces in 1879 and 1936, respectively, the
Zulu and Ethiopian Amharas had themselves used genocidal violence in order to
build and extend their empires decades earlier.11
9 For a fundamental and much debated criticism of the genocide concept, see Christian Gerlach,
‘Extremely Violent Societies: An Alternative to the Concept of Genocide’, Journal of Genocide Research
8:4 (2006), 455–71.
10 Ronald Robinson, ‘Non European Foundations of European Imperialism: Sketch for a Theory of
Collaboration’, in Roger Owen and Bob Sutcliffe (eds), Studies in the Theory of Imperialism (London:
Longman, 1972), 117–41.
11 Michael R. Mahoney, ‘The Zulu Kingdom as a Genocidal and Post-Genocidal Society, c. 1810 to
the Present’, Journal of Genocide Research 5:2 (2003), 251–68. On the violent Amharization of the
Oromo, see P. T. W. Baxter, ‘Ethiopia’s Unacknowledged Problem: The Oromo’, African Affairs 77:308
(1978), 283–96. On the systematic suppression of Oromo culture and language in the Ethiopian
Empire: Mekuria Bulcha, ‘The Politics of Linguistic Homogenization in Ethiopia and the Conflict
over the Status of “Afaan Oromoo” ’, African Affairs 96:384 (1997), 325–52.
Fourthly, the study of colonial mass violence in Africa must resist generalizing
conclusions. Seven European nations participated in the ‘scramble for Africa’ in the
1880s. Their motives, financial as well as infrastructural requirements, and colonial
cultures differed considerably, as did their modes of governance and administra-
tion. Almost all of these empires lacked a coherent structure and were chaotic
clusters of settler, trade, plantation, and military colonies that all entailed different
forms of rule. The spectrum ranged from direct and almost totalitarian control to
indirect rule where African elites enjoyed a great deal of autonomy. Some colonies
were marked by a mosaic of diverse types of rule. In German East Africa (present-
day Tanzania), for example, some territories were governed by civil administra-
tions, others by military authorities. The treatment of the indigenous populations
and the degree of violence used to establish and to maintain the colonial order
depended decisively on the intended use of a colony and the administrative
structures of its government. Colonial administrations dominated by the military
usually resorted to extreme measures like genocidal violence in order to suppress
African resistance. The complexity and polymorphy of imperial rule in Africa
makes it difficult to formulate a grand theory of colonial mass violence of universal
application.
Fifthly, only in recent years have historians realized the importance of colonial
mass violence in the global history of collective violence.12 Colonial genocide has
even become a widely used distinct category. However, it is important to note that
genocidal violence in most of colonial Africa differs in some considerable ways
from genocides committed in North American and Australian settler colonies:
European colonization of Africa did not inevitably lead to the expulsion and/or
annihilation of the indigenous populations. There are two reasons for this differ-
ence: whereas colonization preceded the formation of bureaucratic colonial states
in America and Australia, European settlement followed the establishment of
colonial administration in Africa. As a result, the colonial states in the British
New World territories were almost unlimitedly dominated by settlers’ interests. In
Africa, by contrast, the settlers’ influence and ability to fight and expel the Africans
on their own was more restricted because the colonial states were still weak and
their power relied on the cooperation with indigenous chiefs.
A more important reason why Europeans normally did not envisage physically
exterminating the Africans in large parts was their dependency on indigenous
labour power.13 Most parts of Africa were not attractive for European migrants
12 On how European colonialism provided radical ideologies of the twentieth century with the
idea of population economy and genocidal thinking, see Jürgen Zimmerer, ‘Colonialism and the
Holocaust: Towards an Archaeology of Genocide’, in A. Dirk Moses (ed.), Genocide and Settler Society.
Frontier Violence and Stolen Indigenous Children in Australian History (New York: Berghahn, 2004),
49–76.
13 Patrick Wolfe, ‘Land, Labor, and Difference: Elementary Structures of Race’, American Historical
Review 106:3 (2001), 866–905.
who could have replaced the Africans as indentured servants or contract workers.
Even so, this does not mean that genocide did not occur in Africa. As will be
shown, colonial wars often degenerated into genocidal slaughter. Southern Africa,
which is a special case in the colonial history of Africa because of its appeal for
European settlers, saw many of these genocidal wars.
SOUTHERN AFRICA AS A ZONE OF VIOLENCE

................................................................................................................
The strategic importance of the Cape of Good Hope for the sea route from Europe
to India had a significant impact on the history of Southern Africa. This region
experienced the beginnings of European colonization and settlement much earlier
than other parts of sub-Saharan Africa. The relatively mild climate conditions and
the absence of tropical diseases in the coastal areas made South Africa a favourable
living space for European settlers. Another catalyst for European colonization was
the availability of arable land and the region’s diversity and richness of natural
resources, which include most notably gold, copper, iron, and platinum. For these
reasons, Southern Africa’s colonial history is in many ways comparable to the
European conquest of North America: the establishment of trading posts at
the coast line, a lasting influx of settlers and constant extensions of the frontier,
the permanence of colonial wars and the expulsion of the autochthonous popula-
tions from their ancestral lands, the import of slaves, and the formation of systems
of strict racial segregation are key characteristics of the historical processes in these
two regions.
This is not the only significant development which shaped the history of
Southern Africa. The violent empire-building of the Zulu in the early nineteenth
century in combination with environmental degradation and drought led to
massive streams of migration and an ethnic reorganization in many areas. The
nine Cape Frontier Wars in the nineteenth century and the constant movement of
white settlers further east also contributed to the population migration and politi-
cal disruption in the region. Thus, the Mfecane, the demographic and ethnic
transformation of the 1820s and 1830s, cannot be traced back to the nation-building
of the Zulu alone because of its interplay with European colonization.14 What also
sets South Africa apart from many other African regions under colonial control was
the grave conflict between two emerging European settler communities. It should
be noted, though, that hostilities between the Boers and the British 1899–1901 were
14 Carolyn Hamilton (ed.), The Mfecane Aftermath. Reconstructive Debates in Southern African
History (Johannesburg: Wits University Press, 1995).
not ‘white man’s war’, because both parties relied to a great extent on African
support troops.
Several interconnected clusters of conflicts thus raged in this region. Their
origins lie in the mid-seventeenth century with the modest beginnings of European
colonization. When the Dutch East India Company (Vereenigde Oostindische
Compagnie, VOC) decided to establish a trading post at the Cape of Good Hope
in 1651, there were no plans for extensive European settlement in the region.
Nevertheless, the presence of the VOC attracted Dutch, German, and Huguenot
immigrants in the following decades. At the end of the eighteenth century, a small
European settler community consisting of about 15,000 members existed at the
Cape. The indigenous peoples of the region were either pastoralists or hunter-
gatherers. In the early years of colonization, the pastoralist Khoikhoi and the
Europeans maintained a more or less cooperative relationship and traded goods.
The Africans provided the Europeans with meat in exchange for iron, copper, and
tobacco. But the relations between these two groups worsened when the VOC
started to foster the colonization of the Cape region. Former employees of the
company, the so-called free burghers, were encouraged to move to the interior of
Southern Africa and to settle there to avoid the intermediate trade of the Khoikhoi.
Regular attacks on these outposts by the Khoikhoi, and both the Boers’ religious
conviction that they were a ‘chosen people’ and that the Africans were ‘Canaanites’
doomed to be destroyed by God contributed to a radicalization of the European
settlers’ attitude and behaviour. For the Boers, the most imminent danger for a
secure existence as independent and self-sustaining farmers was these ‘Bushmen’,
traditional hunter-gatherer groups or impoverished and dispossessed Khoikhoi.
Commandos of the Boers launched frequent attacks on the Bushmen and forced
the hunter-gatherers to retreat to the Kalahari desert or to mountain areas where
most of the refugees were hunted down by Europeans in the second half of the
eighteenth century.15 The primary goal of the assaults by the Boers was the
extinction of the Bushmen as independent and viable groups. In 1774, for example,
a raid by European settlers led to the killing of 500 Bushmen alone. Women and
children were usually captured and pressed into slavery.16
Although Europeans perceived the vanishing of Southern Africa’s indigenous
populations as a natural and inevitable development, humanitarian and philanthropic
15 Clifton C. Crais, White Supremacy and Black Resistance in Pre-Industrial South-Africa: The
Making of the Colonial Order in the Eastern Cape, 1770–1865 (Cambridge: Cambridge University
Press, 1992).
16 On the so-called General Commando of 1774, see Nigel Penn, ‘Land, Labour, and Livestock in
the Western Cape during the Eighteenth Century’, in Wilmot G. James and Mary Simons (eds), The
Angry Divide. Social and Economic History of the Western Cape (Cape Town: D. Philip, 1989), 2–19,
especially 16–18; Jan-Bart Gewald, ‘Untapped Sources: Slave Exports from Southern and Central
Namibia up c.1850’, in Carolyn Hamilton (ed.), The Mfecane Aftermath: Reconstructive Debates in
Southern African History (Pietermaritzburg: Natal University Press, 1995), 412.
circles in Great Britain denounced the Boers’ policy of extermination.17 When the
British annexed the Cape Colony in 1806, they soon aimed at the dismantling of
slavery. Nevertheless, the British presence in Southern Africa did not lead to the
pacification of the region, but to a further escalation of violence as British control
expanded territorially. The British pushed the frontier eastwards and encouraged
European settlement. In 1820 alone, up to 4,000 white colonists intruded into the
eastern borderland of the Cape Colony. There had been several hostile encounters
between the trekboers (Dutch settlers) and the Xhosa around the Great Fish River
region of the Eastern Cape. The Xhosa were a Bantu-speaking people who relied
mainly on cattle breeding. In the 1850s, about 100,000 Xhosa lived in Southern
Africa. The British advance resulted in a significant intensification of violence. Both
British troops and European settlers fought six major frontier wars (often referred to
as ‘Kaffir Wars’) against the Xhosa in the nineteenth century. The Xhosa far out-
numbered the European settlers and resisted the colonial conquest tenaciously.
European farms were regularly attacked and in 1819 the city of Grahamstown was
almost overrun by the Africans.18 The persistence of the Xhosa contributed to a
radicalization of the settlers and led them to ever more extreme measures. The
following appeal, published in the Grahamstown Journal on 10 April 1847 is typical in
that respect:
Let war be made against the Kaffir huts and gardens. Let all these be burned down and
destroyed. Let there be no ploughing, sowing or reaping. Or, if you cannot conveniently, or
without bloodshed prevent the cultivation of the ground, take care to destroy the enemy’s
crops before they are ripe, and shoot all who resist. Shoot their cattle too wherever you see
any. Tell them the time has come for the white man to show his mastery over them.19
Scorched-earth tactics characterized European warfare in Africa and had disastrous

effects on the indigenous societies. When the Xhosa were deprived of much of
their land and pushed further east by the British, their desperation made them
susceptible to millenarian fantasies. The visions of a teenage girl made the Xhosa
slaughter about 400,000 of their cattle from April 1856 to February 1857. Although
historians still debate the Xhosas’ motivation, it is generally believed that the Xhosa
17 Patrick Brantlinger, Dark Vanishings: Discourse on the Extinction of Primitive Races, 1800–1930
(Ithaca: Cornell University Press, 2003), 73–4.
18 On the frontier wars against the Xhosa, see Christopher Saunders, ‘Political Processes in the
Southern African Frontier’, in Howard Lamar and Leonard Thompson (eds), The Frontier in History:
North America and Southern Africa Compared (New Haven: Yale University Press, 1981), 149–71; Ben
Maclennan, A Proper Degree of Terror: John Graham and the Cape’s Eastern Frontier (Johannesburg:
Ravan Press, 1986); Timothy Stapleton, Maqoma: Xhosa Resistance to Colonial Advance, 1798–1873
(Johannesburg: Jonathan Ball, 1994); Christoph Marx, ‘Kolonialkrieg und rassistische Dämonologie:
Das südliche Afrika im 19. Jahrhundert’, in Mihran Dabag, et al., Kolonialismus. Kolonialdiskurs und
Genozid (Paderborn: Fink, 2004), 167–84.
19 Cited in Allister Sparks, The Mind of South Africa. The Story of the Rise and Fall of Apartheid
(London: Ballantine Books, 1990), 65.
thought the destruction of their herds would reawaken their ancestors who could
defeat the foreign conquerors.20 In the aftermath of this seemingly irrational event,
almost 40,000 Xhosa starved to death. The survivors lost their lands to European
settlers and were driven into the service of the Europeans. The cattle killing was a
catalyst in the process of the obliteration of the Xhosas’ traditional lifestyle and
their identity as independent pastoralists.
Another group that suffered massively from European imperialism and the
consequences of colonial warfare were the Zulu. Until 1800, the chiefdom of the
Zulu was only one among many others in KwaZulu-Natal. Under the auspices of
their king Shaka (c.1787–1828), the Zulu revolutionized their method of warfare,
defeated their neighbours, and established hegemonic rule in the region. To
maintain their authority, the Zulu relied on policies of mass violence, such that
one scholar has called their kingdom a genocidal society.21 Although the Zulu have
been portrayed as courageous warriors and evenly matched adversaries in literature
and film, the wars between the British colonial troops and the Africans were far
from symmetrical, and non-combatants were gravely affected. The European
settlers’ fear of the Zulu were heightened by a painful defeat of the British army
in Isandhlwana in 1879. Rumours of the Zulus’ alleged habit of mutilating British
soldiers contributed to a further radicalization of the Europeans and to their desire
to annihilate the Africans. The demand for a radical solution of the Zulu problem
was widespread in the Cape Colony, as a correspondent of the Daily News observed:
A desire for extermination is, I must confess, one of the most painful peculiarities of the
present time. If the ideas at present prevailing in some circles were allowed to have free play,
I do not think there would be many Zulus of any age or of either sex alive this day
twelvemonth.22
Indeed, the British campaign against the Zulu was a war of extermination. The
systematic destruction of the economic foundation of Zululand, the refusal to take
prisoners, and deliberate attacks against villages and civilians brought the Zulu to
their knees. They were only saved from a policy of outright genocide by their king’s
readiness to surrender unconditionally.23
It is interesting to note that the same radicalization and totalization of war
marked the military conflict between the Boers and the British 1899–1902. The
Boers did not accept British rule of the Cape that had begun in 1806 and they
20 For a critical discussion of the event, see Jeffrey Brian Peires, The Dead Will Arise: Nonqqawuse
and the Great Xhosa Cattle-Killing of 1856–7 (Bloomington, IN: Indiana University Press, 1989);
Timothy Stapleton, ‘“They No Longer Care For Their Chiefs”: Another Look at the Xhosa Cattle-
Killing of 1856–57’, International Journal of African Historical Studies 24:2 (1991), 383–92.
21 Mahoney, ‘The Zulu Kingdom as a Genocidal and Post-Genocidal Society.’
22 The Illustrated London News, 12 April 1879. Cited in Michael Lieven, ‘“Butchering the Brutes All
Over the Place”: Total War and Massacre in Zululand, 1879’, History 84:276 (1999), 617.
23 Lieven, ‘“Butchering the Brutes All Over the Place”’.
regarded the abolition of slavery an illegitimate interference by the British autho-

rities. Between 1835 and 1846, about 14,000 Boers left the Cape and migrated
eastward and north-eastward, where they founded independent republics. When
gold and diamonds were found in the interior, the British increased their interest in
the Orange Free State and Transvaal regions and tried to regain control over the
Boers. In the war from 1899 to 1902, both the Boers and the British relied on their
experiences of warfare against the Khoikhoi, Bushmen, Xhosa, and Zulu. Because
the Boers were organized in small commandos, the British resorted to a policy of
scorched earth and violence against non-combatants. To cut the Boer commandos
from their supply sources, British army commanders and authorities ordered the
incarceration of the Boer civilian population in concentration camps. Hygienic
conditions and food supply were completely insufficient in these camps and led to
the deaths of about 28,000 detainees, mainly women and children.24
Although the British were fighting against Europeans, they understood the
conflict as a traditional colonial war and used the same methods with which they
had defeated the Xhosa and the Zulu. In this way, the Anglo-Boer war is an
important precursor of modern total war, as experienced in Europe only a few
years later. But what exactly are the key characteristics of colonial wars? And why
did so many colonial wars in Africa reach genocidal proportions?
T H E G E N O C I DA L P OT E N T I A L O F C O LO N I A L
WA R FA R E I N A F R I C A
................................................................................................................
On 2 October 1904, the supreme commander of the German colonial troops in
South West Africa, Lothar von Trotha (1848–1920), issued his infamous genocide
order:
I the great General of the German troops send this letter to the Herero people. The Herero
are no longer German subjects. . . . The Herero people must leave the country. If the nation
doesn’t do this I will force them with the Groot Rohr [cannon]. Within the German borders,
every Herero, with or without gun, with or without cattle will be shot. I will no longer
accept women and children, I will drive them back to their people or I will let them be
shot at.25
24 For a discussion of the origins and the course of the Anglo-Boer War, see the contributions in
Donald Lowry (ed.), The South African War Reappraised (Manchester: Manchester University Press,
2000).
25 Proclamation by Lothar von Trotha (copy), 2 October 1904, German Federal Archive, Berlin
Lichterfelde (GAF), Bestand Reichskolonialamt (R1001), 2098, 7–8.
In a letter to the chief of the German general staff, Lothar von Trotha reaffirmed his
genocidal intention by noting he would not be ready to accept women and children
as prisoners and that he would drive them into the waterless Omaheke desert where
they would face death from starvation and exhaustion:
To accept women and children who are for the most part sick, poses a great risk to the force,
and to feed them is out of the question. For this reason, I deem it wiser for the entire nation
to perish than to infect our soldiers into the bargain and to make inroads into our water
and food supplies.26
The German general’s measures were—from a military standpoint—not neces-

sary anymore. The rebelling Herero had already been defeated in the battle at
Hamakari on 11 August 1904. Lothar von Trotha’s motives for the annihilation of
the Herero were racist and social-Darwinist, as the following excerpt from an
article written by the general reveals: ‘At the outset, we cannot do without the
natives. But they finally have to melt away. Where the climate allows the white man
to work, philanthropic views cannot banish Darwin’s law “Survival of the Fittest”.’27
Lothar von Trotha’s superiors in Berlin and even the Emperor shared the racist
worldview of their general.
As might be expected, colonial soldiers’ and officials’ behaviour and actions
during wars against Africans were deeply shaped by racist attitudes. Most Euro-
peans did not perceive their African counterparts as human beings who deserved
respect and fairness. According to a widespread and influential British handbook
for colonial soldiers, ‘small wars’ were understood as ‘expeditions against savages
and semi-civilised races’.28 Even so, ideological factors such as racism alone cannot
sufficiently explain the genocidal outcome of so many colonial wars. The complete
physical annihilation of Africans was generally perceived as counterproductive and
irrational because the colonizers were heavily dependent on their labour power for
the work on plantations, farms, and in mines. Von Trotha’s genocidal campaign, for
example, met with strong disapproval in Germany. Paul Rohrbach, the most
influential German colonial propagandist at the time, denounced the general’s
policy and stated: ‘It was a mistake . . . to send a general to Southwest [Africa] . . . who
did not understand that it would not be the important thing to destroy the natives as
26 Letter by Lothar von Trotha to the chief of the German general staff von Schlieffen, 4 October
1904, GFA, R 1001, 2089, 5. Horst Drechsler, Let Us Die Fighting: The Struggle of the Herero and Nama
against German Imperialism (London: Zed Press, 1980), 161.
27 Cited in Gesine Krüger, Kriegsbewältigung und Geschichtsbewußtsein: Realität, Deutung und
Verarbeitung des deutschen Kolonialkriegs in Namibia 1904–1907 (Göttingen: Vandenhoeck und
Ruprecht, 1999), 66.
28 Charles Edward Callwell, Small Wars: Their Principles and Practice, 3rd edn (London: His
Majesty’s Stationery Office, 1906), 21.
enemies. . . . Southwest-Africa with natives was of much more value for us than
without.’29 But why then did most colonial wars degenerate into genocidal slaughter?
The transition from war to genocide is a complex process that undergoes various
phases of cumulative radicalization and depends on a range of situational factors.
A very decisive factor was military setbacks suffered by European colonial armies. It
was the humbling defeat at Isandhlwana in January 1879 and the loss of up to 1,200
soldiers that seemed to confirm British settlers’ claim that the ‘Zulu problem’
ought to be solved once and for all. Another example of such a military setback
is the fall of Khartoum and the death of Gen. Charles Gordon in January 1885 at the
hands of the Mahdists, which was perceived as a national disgrace and led to a
severe political crisis in London. This is the reason why the British reconquest of
the Sudan in 1898 resulted in the carnage of the Mahdists. The French conquest of
Algeria in the 1830s and 1840s reached genocidal dimensions when Arab tribes
under Abd al-Qadir launched successful attacks against the invaders. Even Alexis
de Tocqueville, who later became an ardent critic of French exterminationist
policies in Algeria, had been in favour of radical means to defeat Abd al-Qadir:
We shall never destroy Abd-el-Kader’s power unless we make the position of the tribes who
support him so intolerable that they abandon him. . . . All means of desolating these tribes
must be employed. . . . I believe that the right of war authorizes us to ravage the country and
that we must do it, either by destroying harvests during the harvest season, or year-round
by making those rapid incursions called razzias, whose purpose is to seize men or herds.30
A comparative analysis of colonial wars in Africa shows that colonial powers are
ready to act against their economic interests and to use genocidal violence in order
to avoid or avenge humiliating defeats against ‘semi-civilized’ adversaries. Eur-
opeans perceived indigenous resistance not only as a threat to their ambitions in
the colonies but to their imperial project as such because a possible failure
diminished their prestige in the eyes of imperial rivals.
The radicalization of military campaigns in the African colonies manifested itself
in the use of methods of warfare that were morally and legally outlawed (at least
theoretically) in the European theatres of war. In the following, five of these deadly
methods will be outlined.
Most colonial conflicts were asymmetric wars because of the Africans’ guerrilla
tactics. As many of the examples and quotation above have shown, the burning of
indigenous settlements and crops were integral elements of colonial warfare.
Scorched-earth tactics usually resulted in famines and had disastrous effects on
African societies. In German East Africa (Tanzania), the colonizers deliberately
29 Paul Rohrbach, Um des Teufels Handschrift. Zwei Menschenalter erlebter Weltgeschichte

(Hamburg,: Hans Dulk, 1953), 64.
30 For a discussion of the genocidal dimension of the French conquest of Algeria 1830–75, see Ben
Kiernan, Blood and Soil. A World History of Genocide and Extermination from Sparta to Darfur (New
Haven, CT: Yale University Press, 2007), 364–74. Quote on pages 370–1.
used hunger as a weapon to suppress the Maji-Maji movement 1905–8, as the

following quotation by the German captain Wangenheim reveals: ‘In my view, only
hunger and distress can bring about a final submission. Military actions alone will
remain more or less a drop in the ocean.’31 Fields and granaries were systematically
destroyed by German troops who prevented cultivation and confiscated everything
eatable. Hunger and consequent diseases led to the deaths of about 250,000
Africans. Entire areas in Southern Tanzania were depopulated.32
Although the Geneva Protocol of 17 June 1925 prohibited the use of gas and
poison in war, European powers still resorted to chemical warfare in the colonies.
Before the adoption of the international treaty, the British army used mustard gas
against Arabs in Iraq in 1920 and against Afghans on the northwest frontier.
Winston Churchill, Secretary of State for War at the time, made clear that he
considered gas to be the ideal weapon for colonial wars: ‘I do not understand this
squeamishness about the use of gas. I am strongly in favour of using poison gas
against uncivilised tribes.’33 Between 1921 and 1927, the ‘Spanish Army of Africa’
launched gas attacks to put down the Riffian Berber rebellion led by Abdel Karim
in Northern Morocco. The ‘intensive and continuous bombing with the aid of the
most harmful of all gases’ did not only aim at defeating the Berber rebels. Accord-
ing to the Spanish King Alfonso XIII, the intention of the gas campaign was ‘the
extermination, like that of malicious beasts, of the Beni Urriaguels [a Berber tribe]
and the tribes who are closest to Abdel Karim’.34 The reason for this extremist
position was the military disaster in Annual where 3,000 Berbers managed to kill
more than 10,000 Spanish colonial soldiers in July 1921. This embarrassing defeat
caused a political crisis in Madrid and put both the King and the military
leadership under immense pressure. Italy faced similar problems in Libya and
regularly bombed the oases of the resisting tribes with phosgene and mustard gas
in the 1920s. The quality and quantity of chemical warfare reached a new dimen-
sion in Mussolini’s imperial war against Ethiopa 1935–41. The Ethiopian Emperor
Haile Selassie (1892–1975) made the Italian war crimes and their consequences
public when he called for international support at the headquarters of the League
of Nations on 30 June 1936:
Special sprayers were installed on board aircraft so that they could vaporize, over vast areas
of territory, a fine, death-dealing rain. Groups of nine, fifteen, eighteen aircraft followed one
another so that the fog issuing from them formed a continuous sheet. It was thus that, as
31 Gustav Adolf Graf von Götzen, Deutsch-Ostafrika im Aufstand 1905/06 (Berlin: Dietrich Reimer,
1909), 149.
32 Felicitas Becker, ‘Traders, “Big Men” and Prophets: Political Continuity and Crisis in the Maji
Maji Rebellion in Southeast Tanzania’, Journal of African History 45:1 (2004), 1–22.
33 David E. Omissi, Air Power and Colonial Control: The Royal Air Force 1919–1939 (Manchester:
Manchester University Press, 1990), 160.
34 Sebastian Balfour, Deadly Embrace: Morocco and the Road to the Spanish Civil War (Oxford:
Oxford University Press, 2002), 135.
from the end of January 1936, soldiers, women, children, cattle, rivers, lakes and pastures
were drenched continually with this deadly rain. In order to kill off systematically all living
creatures, in order to more surely to poison waters and pastures, the Italian command made
its aircraft pass over and over again. . . . These fearful tactics succeeded. Men and animals
succumbed. The deadly rain that fell from the aircraft made all those whom it touched fly
shrieking with pain. All those who drank the poisoned water or ate the infected food also
succumbed in dreadful suffering.35
Although the Italian Regia Aeronautica did not drop gas bombs on the major cities, the
Ethiopian population suffered for a long time from chemical pollution. There are no
official figures but the Italian gas attacks likely killed tens of thousands of Africans.36
The persistence of many African guerrilla movements and their backing by the
autochthonous society often allowed military leaders and colonial authorities to
conclude that only forced migration or the deportation of the indigenous population
could lead to sustainable ‘pacification’. In June 1930, the Italian marshal Badoglio
ordered the total evacuation of the Djebel al-Akhdar in Libyan Cyrenaica. Up to
100,000 Bedouins, half of the whole region’s population, were driven away. The
Italian marshal was well aware that this measure would in all likelihood result in
mass death. In a letter to his superior, he stated:
As for overall strategy, it is necessary to create a significant and clear territorial separation
between the controlled population and the rebel formations. I do not hide the significance
and the seriousness of this measure, which might be the ruin of the so-called subdued
population . . . But by now the course has been set and we must carry it out to the end, even
if the entire population of Cyrenaica must perish.37
After the suppression of the Herero and Nama revolts in German South West
Africa in 1908, colonial officials and settlers suffered from a paranoid fear of new
indigenous unrest, and they considered the deportation of all surviving Herero to
New Guinea. German authorities saw the deportation of about 200 Nama to Togo
and Cameroon from 1904 to 1910 as a social engineering experiment about whether
Namibians could survive in tropical regions. Only a handful of the deportees in fact
survived and returned to South West Africa. The outbreak of World War I averted
the further implementation of these ideas.38
35 Speech of H.M. the Negus Haile Selassie (Ethiopia), in: League of Nations Official Journal,
Special Supplement No. 151: Records of the 16th Ordinary Session of the Assembly. Plenary meetings
of 30 June to 4 July 1936. Text of the debates, Part II, Geneva 1936, 25.
36 Aram Mattioli, Experimentierfeld der Gewalt. Der Abessinienkrieg und seine internationale
Bedeutung 1935–1941 (Zürich: Orell Füssli, 2005), 109, 114–16; Giulia Brogini Künzi, Italien und der
Abessinienkrieg 1935/36. Kolonialkrieg oder Totaler Krieg (Paderborn: Schöningh, 2006), 260–6.
37 Alexander de Grand, ‘Mussolini’s Follies: Fascism in its Imperial and Racist Phase, 1935–1940’,
Contemporary European History 13:2 (2004), 131.
38 Dominik J. Schaller, ‘Kolonialkrieg, Völkermord und Zwangsarbeit in “Deutsch-Südwestafrika” ’,
in idem et al. (eds), Enteignet, Vertrieben, Ermordet. Beiträge zur Genozidforschung (Zürich: Chronos,
2004), 147–232. On the deportations in particular, 189–95.
Many colonial wars in Africa saw the confinement of expelled civilians or

prisoners of war in concentration camps. Although these concentration camps
cannot be equated to the Nazis’ extermination camps like Auschwitz or Treblinka,
they turned out to be death camps nonetheless. Malnutrition, insufficient hygienic
conditions, forced labour, and the brutal behaviour of the guards caused the deaths
of tens of thousands of Africans in European camps in the twentieth century alone.
The mortality of African prisoners in German South West Africa during and
immediately after the 1904–8 war was almost fifty per cent according to official
German documentation.39 To defeat the Mau Mau uprising led by the Kikuyu
in Kenya 1952–60, the British colonial government incarcerated hundreds of
thousands of Africans in concentration camps or in settlements ringed with barbed
wire.40 Although the British claimed that their goal was the ‘pacification’ of the
region and the ‘civilizing’ of the alleged barbarian African rebels, it is more likely
that their murderous campaign aimed at the elimination of the Kikuyu as decisive
political actors in late colonial Kenya. The British understood the use of excessive
mass violence as the only way to suppress the Kikuyu rebellion and to save their
African empire.41
No colonial wars occurred without massacres and the systematic killings of
prisoners. The young Churchill, who joined the British campaign against the
Sudanese Mahdists in 1898, denounced the military leadership’s deliberate policy
of not taking prisoners and of killing wounded enemies: ‘I shall merely say that the
victory at Omdurman was disgraced by the inhuman slaughter of the wounded and
that Kitchener was responsible for this.’42 Churchill’s statement suggests that
the murder of prisoners was not just a spontaneous act. In the Ethiopian case,
the murder of prisoners can directly be traced back to Benito Mussolini. When the
Duce heard about the Amhara rebellion in 1937, he told the Italian Governor-
General of Ethiopia: ‘Prisoners and their accomplices and the uncertain will have to
be executed.’43
39 Ibid. 177.
40 There is no consensus about the actual number of detainees in British concentration camps in
Kenya. While Caroline Elkins speaks of up to one and a half million Kikuyu prisoners, David
Anderson estimates that ‘only’ up to 70,000 Kikuyu were incarcerated. Caroline Elkins, Britain’s
Gulag. The Brutal End of Empire in Kenya (London: Pimlico, 2005), xii; David Anderson, Histories of
the Hanged: Britain’s Dirty War in Kenya and the End of the Empire (London: Weidenfeld and
Nicolson, 2005).
41 Thus, the British intentions were not genocidal as Caroline Elkins suggests in her work: ‘Mau
Mau became for many whites in Kenya . . . what the Armenians had been for the Turks, the Hutu for
the Tutsi, the Bengalis to the Pakistanis, and the Jews to the Nazis’ (Britain’s Gulag, 49).
42 Cited in Martin Daly, Empire on the Nile: The Anglo-Egyptian Sudan, 1898–1934 (Cambridge:
Cambridge University Press, 2004), 4.
43 Cited in G. Bruce Strang, On the Fiery March: Mussolini Prepares for War (Westport, CT: Praeger,
2003), 22.
Although genocidal warfare may be ordered by the military command or the

colonial authorities, massacres were committed by ordinary soldiers or settler
militias. Therefore, it is important to focus on these ‘men on the spot’ to fully
understand the dynamics of radicalization inherent to so many colonial wars.
When colonial soldiers reported about atrocities in their memoirs or in letters
home they explained the violence they committed against African civilians by
referring to social Darwinist ideas. Thus, they claimed that their actions were
part of an alleged natural process that would inevitably lead to the vanishing of
‘inferior races’. These ideological explanations often merely served as retrospective
justifications, however. As social-psychological literature on the development of
soldiers’ readiness to kill non-combatants has shown, overstraining, stress, and fear
are decisive reasons for genocidal behaviour. This is especially true for the colonial
situation in Africa where settlers on the frontier were living in a constant fear of the
indigenous population and developed paranoid fantasies about African revolts.
European soldiers who had been brought to Africa to suppress indigenous resis-
tance were not at all familiar with the geography and topography of the region and
had no idea about its inhabitants. Furthermore, many Europeans suffered from the
harsh climate and from tropical diseases. All these situational factors shaped the
Europeans’ extremely violent behaviour towards Africans during colonial wars.
As the colonial period in Africa ranged from the sixteenth century up to the end
of the cold war and the collapse of the apartheid regime in South Africa in 1994 it is
difficult to estimate how many colonial wars exactly the Europeans waged and how
many Africans fell victim to them. But as I have shown above, there is no doubt
that most of these military campaigns reached genocidal dimensions because
excessive violence was deliberately targeted against both combatants and women
and children.
C U LT U R A L G E N O C I D E AND F O RC E D L A B O U R
................................................................................................................
Colonial mass violence should not be reduced to the physical killing of Africans
alone. European colonizers deliberately used another form of violence in Africa:
the dissolution of indigenous culture encompassing political and socio-economic
institutions, religious beliefs, and historical heritage. For Lemkin, the intended
annihilation of a group’s culture and identity constitutes an act of genocide.
Anthropologists usually call practices of cultural destruction ‘ethnocide’.44
44 Lemkin, Axis Rule, 82–91; Robert Jaulin, La Paix Blanche: L’introduction à l’Ethnocide (Paris:
Seuil, 1970).
The driving force for the ‘Scramble for Africa’ since the 1880s was the European
powers’ wish to extend their spheres of economic and political influence so that
they were able to compete with their imperialist rivals. Africa was perceived as an
important reservoir for much desired raw materials, and its inhabitants were
expected to plant, extract, and process these materials as cheap labourers. Unsur-
prisingly, Africans were generally not interested in abandoning their subsistence
economies and tried to avoid work on European-owned plantations or mines. The
consequence was a chronic shortage of indigenous labour power in most sub-
Saharan colonies. To overcome the Africans’ reluctance and to integrate them into
a ‘modern’ European economic system, the imperial conquerors embarked on the
strategy of eliminating the indigenous populations’ economic independence. This
objective effectively required the destruction of traditional cultures and ways of life.
Thus the forced settlement of nomadic groups and the transformation of hunter-
gatherer societies into proletarians was an important precondition for the estab-
lishment of a capitalist economy in Africa. Missionaries served as agents of this
kind of cultural genocide by inculcating a colonial working morale among the
Africans. Some extremists even claimed that Africans ought to be deprived of any
‘tribal identity’ and transformed into a helot class of workers. The German
publicist Paul Rohrbach, who had himself served in the colonies, was an ardent
advocate of these utopias of white supremacy in Africa:
Only the necessity of losing their free national barbarianism and of becoming a class of
servants for the whites provides the natives—historically seen—with an internal right of
existence. . . . The idea that the Bantus would have the right to live and die according to their
own fashion is absurd. It is true for peoples as well as for individuals that their existence is
only justified if they contribute to general progressive development.45
Surprisingly, literature on European colonial rule in Africa has not yet paid
sufficient attention to the destruction and theft of indigenous cultural heritage by
Europeans. Probably the best known case is the looting of Benin by a British army
under the command of Harry Rawson in 1897. The invaders systematically de-
stroyed a great amount of artefacts and seized the famous Benin Bronzes, many of
which are still displayed in the British Museum. Another well-known example is
the demolition of the Mahdi’s tomb in Omdurman by Kitchener in 1898. The
general’s act caused a popular outrage in Great Britain. In his report on the
Sudanese campaign, Winston Churchill attacked Kitchener’s rudeness and de-
nounced the destruction of this ‘fine building which might attract the traveler
and interest the historian’ as ‘an act of vandalism and folly’.46
45 Paul Rohrbach, Deutsche Kolonialwirtschaft: Südwest-Afrika (Berlin: Buchverlag der Hilfe, 1907),
285.
46 Winston S. Churchill, The River War: An Historical Account of the Reconquest of the Soudan,
2 vols (London: Longmans, 1899), ii.215.
Europeans established coercive systems all over the continent and exploited
hundreds of thousands of Africans as slave labourers. Two of the most brutal and
inhuman regimes of forced labour were practised in German South West Africa
and in the Belgian Congo. Colonial authorities in Namibia proclaimed that all
‘tribal organization’ had come to an end and issued decrees that restricted the
indigenous populations’ freedom of movement. Furthermore, all Africans had to
carry a tiny identity badge around the neck. In Leopold II’s ‘Congo Free State’ the
Africans paid a high price for the profitable rubber exploitation. The colonizers
severely punished and burned the villages of the local population when it did not
manage to provide the assigned rubber quotas. The Belgians resorted to hostage-
taking, massacres, and flogging to make their African subjects collect the precious
and much desired natural resource. It is estimated that the Belgian reign of terror in
Central Africa resulted in a population decline of fifty per cent or, in absolute
numbers, in the deaths of several million Congolese.47
The cases of German South West Africa and the Belgian Congo have become
symbols for the exploitative character of European colonial rule in Africa. Raphael
Lemkin understood them as genocides and emphasized that the extermination of
the ‘natives’ was the official state policy of both the Germans and the Belgians. In
his unpublished study on German rule in Africa, Lemkin described the German
idea of colonization as such:
The Germans did not colonize Africa with the intention of ruling the country justly, living
in peace with the true owners of the land and developing its resources for the mutual
advantage of both races. Their idea was to settle some of the surplus German population in
Africa and to turn it into a German white empire. Bismarck said, ‘A German who can put off
his Fatherland like an old coat is no longer a German for me,’ and it was undoubtedly this
idea which encouraged the policy of deliberate extermination.48
For Lemkin, forced labour in the Belgian Congo and in German South West Africa
was thus more than mere economic exploitation of the indigenous population. He
perceived it—particularly in the German case—as a form of annihilation through
work. But Lemkin was off the mark when he praised British colonialism as a
blessing and examplar for other European nations. Although British and French
colonial authorities proclaimed the ‘cultivation of the natives’ and the promotion
of the Africans’ general welfare as their official aims, forced labour and excessive
flogging were integral elements of colonial rule all over Africa. The weakness of
many colonial states and the lack of authority of colonial officials over settlers
fostered a culture of violence on farms and plantations. Colonial governments in
47 Adam Hochschild, King Leopold’s Ghost: A Story of Greed and Heroism in Colonial Africa
(Boston: Houghton Mifflin, 1999).
48 Raphael Lemkin, ‘The Germans in Africa’, unpublished typewritten manuscript (c.1950), Jacob
Rader Marcus Center of the American Jewish Archives, Hebrew Union College, Cincinnati, Raphael
Lemkin Papers, Box 6, Folder 9, 49–50, cited in Schaller, ‘Colonialism and Genocide’, 86.
Africa regularly issued decrees to advancing the security of African workers,

although European farmers and entrepreneurs did not want to be subject to state
regulations and successfully managed to evade state control.49
C O N C LU D I N G R E M A R K S : N O S I L E N C E
O F T H E G R AV EYA R D
................................................................................................................
While European colonial rule lasted only several decades, it had a profound impact
on Africa. The history of European colonialism in Africa is above all of unprece-
dented socio-economic, political, and cultural change, mass violence, and exploi-
tation. Until recently, the historiography of colonialism and genocide has
portrayed the Africans as passive and apathetic victims of European power and
violence. The German historian Horst Drechsler’s notion of ‘the silence of the
graveyard’ regarding the situation of the Herero after the German campaign of
genocide in 1904 is typical in that respect. But Africa did not degenerate into a
graveyard because of the Europeans’ attempt to transform the continent and its
inhabitants according to their ideas. European colonialism did not succeed in
completely destroying African cultures and identities. Africans always found ways
to preserve their cultures and to reconstitute their social organizations, however
totalitarian and coercive the colonizers’ policies and fantasies about absolute power
were. At the same time, the legacies of colonial rule, including policies of divide
and rule and unequal treatment of different peoples, and the establishment of
arbitrary interstate and intrastate boundaries that would remain contested after the
departure of the Europeans, combined to influence future patterns of mass inter-
group violence, as illustrated by the chapters by Alex De Waal and Omar McDoom
elsewhere in this volume.
Balfour, Sebastian, Deadly Embrace: Morocco and the Road to the Spanish Civil War (Oxford:
Ben-Ghiat, Ruth and Mia Fuller (eds), Italian Colonialism (New York: Palgrave Macmillan,
2005).
49 McCulloch, ‘Empire and Violence’, 227.

Elkins, Caroline, Britain’s Gulags. The Brutal End of Empire in Kenya (London: Pimlico,
2005).
Hull, Isabel V., Absolute Destruction. Military Culture and the Practices of War in Imperial
Germany (Ithaca, NY: Cornell University Press, 2005).
Klein, Thoralf, and Frank Schumacher (eds), Kolonialkriege. Militärische Gewalt im Zeichen
des Imperialismus (Hamburg: Hamburger Edition, 2006).
Lieven, Michael, ‘“Butchering the Brutes All Over the Place”: Total War and Massacre in
Zululand, 1879’, History 84:276 (1999), 614–33.
Lowry, Donal (ed.), The South African War Reappraised (Manchester: Manchester Univer-
sity Press, 2000).
Marchal, Jules, Lord Leverhulme’s Ghosts: Colonial Exploitation in the Congo (London: Verso,
2008).
Schaller, Dominik J., ‘Colonialism and Genocide: Raphael Lemkin’s Concept of Genocide
and its Application to European Rule in Africa’, Development Dialogue 50 (December
2008), 75–93.
—— ‘From Conquest to Genocide: Colonial Rule in German Southwest Africa and German
East Africa’, in A. Dirk Moses (ed.), Empire, Colony, Genocide. Conquest, Occupation, and
Subaltern Resistance in World History (New York: Berghahn, 2008), 296–324.
Zimmerer, Jürgen, and Joachim Zeller (eds), Genocide in German South-West Africa: The
Colonial War of 1904–1908 and its Aftermath (Monmouth: Merlin Press, 2008).
chapter 18
.............................................................................................
G E N O C I D E AT T H E
TWILIGHT OF THE
OT TOMA N EM PI R E
.............................................................................................
hilmar kaiser
World War I saw the almost complete annihilation of the Ottoman Armenians.
The destruction stands at the centre of an intense debate. Some historians claim
that Armenians organized uprisings in the rear of the Ottoman army and compa-
rably few Armenians were killed during an anti-insurrection campaign.1 Others do
not address the genocide directly, but provide insights into the functioning of the
state military, the ‘Special Organization’ (TM—Teşkilat-ı Mahsusa),2 an organiza-
tion that had been founded by the ruling Committee of Union and Progress (CUP)
for intelligence, counter-insurgency, and other tasks that the government did not
want to be associated with, and Ottoman demographic policies.3 These studies
1 Yusuf Halaçoğlu, Ermeni Tehciri ve Gerçekler (1914–1918) (Ankara: Türk Tarih Kurumu, 2001);
idem, Sürgünden Soykırıma. Ermeni İddiaları (Ankara: Babıali Yayıncılığı, 2006); Hilmar Kaiser,
‘Dall’impero all republicca: le continuità del negazionismo turco’, in Marcello Flores (ed.), Storia,
verità, guistizia. I crimini del XX secolo (Milan: Bruno Mondadori, 2001), 89–113; idem, ‘Le génocide
arménien: négation ‘à l’allemande’, in Comité de Défense de la Cause Arménienne (ed.), L’actualité du
Génocide des Arméniens (Paris: Edipol, 1999), 75–91.
2 Erdal Aydoğan, ‘Teşkilât-ı Mahsusa’, in Hikmet Özdemir (ed.), Türk–Ermeni İhtilafı. Makaleler,
Turkish–Armenian Conflict. Articles (Ankara: Türk Büyük Meclisi Kültür, Sanat ve Yayın Kurulu
Yayınları, 2007), 399–427.
3 Fuat Dündar, İttihat ve Terakki’nin Müslümanları İskân Politikası (1913–1918) (Istanbul: İletişim
Yayınları, 2001); idem, Modern Türkiye’nin Şifresi. İttihat ve Terakki’nin Etnisite Mühendisliği (1913–
1918) (Istanbul: İletişim Yayınları, 2008).
366 hilmar kaiser
discuss Ottoman wartime policies and undermine claims of an Armenian uprising,

as large-scale demographic programmes would have been unthinkable under such
conditions: the deportation of over three million people (Armenians and non-
Armenians) depended on the absence of substantial resistance.
Other authors assume that the genocide was part of a policy that used the
‘opportunity’ afforded by the war to realize a scheme that had been in the making
for years. Such interpretations stress the role of the CUP. The party had supposedly
used the TM to execute the genocide, but the organization was a part of the
Ottoman military.4 Critically, the approach suggests coherent planning ahead of
the crime while fragmented evidence is seen as a proof for a conspiracy.
Recent case studies have avoided generalizations based on limited sources.5
Accordingly, claims of long-term genocidal planning appear to be untenable.
Instead, it has become evident that the development of the war determined the
timing of deportations and massacres.6 Moreover, the genocide cannot be sepa-
rated from demographic policies that targeted other communities, like the Greeks,
Nestorians, Syrian Orthodox Christians, Circassians, and Druzes, and depended on
the extraction of Armenian assets. The CUP’s goal was the establishment of total
control over state and society.7 This chapter will consider the development of the
genocide in the context of wider Ottoman demographic policies and late Ottoman
history.
T H E O T TO M A N C O N T E XT
................................................................................................................
Since the founding of the empire, deportations formed part of Ottoman policies by
which unwanted groups were replaced with controllable populations. A solid tax
base was crucial for the maintenance of bureaucracy and army.8 A major divide in
Ottoman society was that between Muslims and non-Muslims. While Muslim
4 Vahakn N. Dadrian, The History of the Armenian Genocide. Ethnic Conflict from the Balkans to
Anatolia to the Caucasus (Providence, RI/Oxford: Berghahn Books, 1995); Taner Akçam, A Shameful
Act: The Armenian Genocide and the Question of Turkish Responsibility, trans. Paul Bessemer (New
York: Metropolitan Books, 2006).
5 Uğur Ü. Üngör, ‘“A Reign of Terror”: CUP Rule in Diyarbekir Province, 1913–1913’, MA Thesis,
University of Amsterdam, 2005.
the Ottoman Armenians (Oxford, Oxford University Press, 2005).
7 Fikret Adanır and Hilmar Kaiser, ‘Migration, Deportation, and Nation-Building: The Case of the
Ottoman Empire’, in René Leboutte (ed.), Migrations and Migrants in Historical Perspective.
Permanencies and Innovations (Brussels: Peter Lang, 2000), 273–92.
8 Rudi Paul Lindner, Nomads and Ottomans in Medieval Anatolia (Bloomington, IN, Research
Institute for Inner Asian Studies, 1983).
genocide at the twilight of the ottoman empire 367
communities constituted a majority by 1914, Christian communities formed a

sizeable part of the empire’s population. Importantly, Greeks and Armenians
played critical roles in the Ottoman economy. Non-Muslims had some autonomy
in internal community affairs, while being subjected to discriminatory taxation
and an inferior legal status.9 Following a weakening of government control, local
competitors for the tax revenue emerged from urban and rural elites in the
seventeenth and eighteenth centuries.10 Early in the nineteenth century, the central
authorities began eliminating this competition.11 This re-centralization policy,
however, did not check the rise of nationalist movements and the founding of
independent Balkan states. Increasingly, Ottoman governments depended on Eu-
ropean support. In return, they granted concessions and other privileges to Euro-
pean nationals. The governments were further weakened by increasing foreign debt
that led to bankruptcy and tightening European control over the tax basis.12 The
loss of territory to European powers and secessionist groups intensified tensions
even more.
Following a defeat by Russia, in the 1878 Congress of Berlin the Sublime Porte
accepted territorial losses and reforms, including improvements for Ottoman
Armenians. For years, Ottoman authorities had increased taxation while local elites
had not stopped levying their own duties. The government did, however, not stop
this abuse nor provide security, allowing continued double taxation. Competition
for land increased the insecurity of Armenian life and property. The central
government was partly responsible. For years, the Sublime Porte had settled
Muslim immigrants to strengthen control over areas feared to be threatened by
foreign occupation or national movements like the Armenian highlands. Many
newcomers had survived ethnic cleansing in the Balkans, Caucasus, and the
Aegean.
Lacking funds, the government regularly abandoned settlers without supplies,
tacitly supporting their occupation of non-Muslim property.13 In the Armenian
highlands, Kurdish leaders joined in the competition for land, registering
9 Benjamin Braude and Bernard Lewis (eds), Christians and Jews in the Ottoman Empire: The
Functioning of a Plural Society (New York: Holmes and Meyer, 1982).
10 Bruce McGowan, ‘The Age of the Ayans, 1699–1812’, in Halil İnalcik and Donald Quataert (eds),
An Economic and Social History of the Ottoman Empire, 1300–1914 (New York: Cambridge University
Press, 1994), 637–758.
11 Andrew G. Gould, ‘Pashas and Brigands: Ottoman Provincial Reform and Its Impact on the
Nomadic Tribes of Southern Anatolia, 1840–1885’, PhD Dissertation, University of California: Los
Angeles, 1973.
12 J. A. R. Marriott, The Eastern Question: An Historical Inquiry in European Diplomacy, 3rd edn
(Oxford: Oxford University Press, 1924); Donald C. Blaisdell, European Financial Control in the
Ottoman Empire: A Study of the Establishment, Activities, and Significance of the Administration of
the Ottoman Public Debt (New York: Columbia University Press, 1929; rpt. New York: AMS, 1966).
13 Abdullah Saydam, ‘XIX. Yüzyılda Adana Eyaletinin Sosyo-Ekonomik Yapısı Hakkında Bir
Rapor’, Belgelerle Türk Tarih Dergisi 71 (1991), 27–33; Gould, Pashas and Brigands.
368 hilmar kaiser
Armenian land in their names with the support of the authorities. Kurdish tribes
used Armenian fields as pastures, demanded shelter for the winter, often taking
possession by killing the owners and abducting women and children. The erosion
of Armenian landholding was part of a policy to create a Muslim majority in
Armenian districts. In the face of government indifference to Armenian protests,
Armenians appealed to the Berlin Congress; thereby the ‘Armenian Question’
entered the agenda of European diplomacy.14
Reforms did not materialize. The Ottoman government continued its settlement
programme, attracting as many Muslims to its territory as possible.15 Increasing
tensions between Armenians and Muslims favoured the rise of Armenian political
organizations in the 1880s. These parties intended to protect villages. For its part,
the government created the Kurdish ‘Hamidieh’ cavalry in 1890. The units were
used against Armenian organizations and rewarded with tax relief. They enriched
themselves openly at the cost of Armenians or other Kurds.16 When in 1894
Armenian mountain dwellers in Sassun opposed double taxation, Ottoman troops
and Kurds slaughtered them, starting a series of massacres. Throughout 1895 and
1896, Hamidiehs and Muslim villagers and townsmen massacred Armenians,
plundering property, seizing land, and abducting women. The government fur-
thered the weakening of Armenian communities by deporting survivors from
Sassun, making room for more Muslim settlers.17
In the absence of foreign intervention, the Ottoman government organized the
massacre of thousands of Armenians in Constantinople in 1896.18 It survived the
ensuing diplomatic crisis with a promise for reform. The Ottoman authorities
continued to suppress political discontent as opposition remained weak and split
along ethnic and ideological lines. By 1908, however, a severe economic crisis
intensified social discontent.19 An army mutiny forced the weakened government
to accept parliamentary rule. The coup triggered intercommunal fraternizations
14 Arshag O. Sarkissian, History of the Armenian Question to 1885 (Urbana: University of Illinois
Press, 1938).
15 United Kingdom, National Archives, Kew Gardens, Foreign Office (hereafter: FO) 195/2073,
Massy to Chargé d‘Affaires, Adana, Dec. 31, 1900, no. 31; ibid. FO 195/2095, Jan. 16, 1901, no. 7; ibid.
Massy to O‘Conor, Adana, May 5, 1901, no. 24.
16 Martin M. Van Bruinessen, Agha, Scheich und Staat. Politik und Gesellschaft Kurdistans (Berlin:
Edition Parabolis, 1989), 248–9; Jelle Verheij, ‘Die armenischen Massaker von 1894–1896: Anatomie
und Hintergründe einer Krise’, in Hans-Lukas Kieser (ed.), Die armenische Frage und die Schweiz
(1896–1923) (Zürich: Chronos Verlag, 1999), 69–129. Louise Nalbandian, The Armenian Revolutionary
Movement: The Development of Armenian Political Parties through the Nineteenth Century (Berkeley/
Los Angeles: University of California Press, 1963).
17 Avedis Nakashian, A Man Who Found a Country (New York: Crowell, 1940), 135.
18 Hugo Von Köller, Von Pasewalk zum Bosporus. Ein abenteuerliches Junkerleben (Berlin-Leipzig:
Brunnen Verlag, 1927), 245–8.
19 Donald Quataert, ‘The Economic Climate of the “Young Turk Revolution” in 1908’, Journal of
Modern History 51 (1979), D1147–61; idem, ‘Machine Breaking and the Changing Carpet Industry of
Western Anatolia, 1860–1908’, Journal of Social History 19 (1986), 473–89.
and Muslim and Armenian parties cooperated in parliament.20 In April 1909, a

counter-revolution caused a setback. Massacres had been prepared in many places
as well but were prevented by CUP members. In Cilicia, 25,000 to 30,000 Arme-
nians were slaughtered by reactionaries who took advantage of religious fervour,
dire economic conditions, inflation, and the competition for land and jobs. Despite
their defeat, the reactionaries’ strength impressed the CUP, which decided to
integrate the oppositional Muslim provincial elites. Accordingly, Armenian de-
mands for reform and the restitution of property met with new resistance, giving
fresh impetus to Kurdish outrages and occupation of Armenian land. Armenian
communities lost their representatives through the killing of teachers, priests, party
leaders, and village headmen. Disillusioned, the leading Armenian party, the
Armenian Revolutionary Federation, broke with the CUP in 1912, on the eve, as
it transpired, of another series of catastrophic events. British ambassador Gerard
Lowther concluded that the 1890s massacres had been ‘succeeded by fifteen years of
draconian methods spelling even a worse extensive “elimination” of the Armenian
element.’21
ETHNIC CLEANSING OF GREEKS AND B U LG A R I A N S

................................................................................................................
During the Balkan Wars of 1912–13, a coalition of Balkan states defeated the
Ottoman Empire, which lost Macedonia, Epirus, Kosovo, and Western Thrace.
The consequences were felt far beyond the war zone. During the fighting all states
had massacred and displaced civilians. The rationale was to eliminate competing
claims to occupied or re-occupied regions. At least 339,074 Muslims fled to
Ottoman-held territory, while 376,186 still had remained behind by January 1915.
The government was unprepared for the influx of destitute refugees and many of
them died of starvation or disease.22
In response, the Ministry of the Interior reorganized its settlement policies
while the CUP unleashed a clandestine campaign against Ottoman Greeks and
20 For an overview, see Erik Jan Zürcher, Turkey: A Modern History (London/New York: I. B. Tauris,
2004).
21 FO 424/237/570, Lowther to Grey, Constantinople, Dec. 31, 1912, no. 1129.
22 Carnegie Endowment for International Peace, Report of the International Commission to Inquire
into the Causes and Conduct of the Balkan Wars (Washington, DC: Carnegie Endowment for
International Peace, 1914); Murat Bardakçı, Talât Paşa’nın Evrak-ı Metrûkesi. Sadrazam Talât
Paşa’nın Özel Arşivinde Bulunan Ermeni Tehciri Konusundaki Belgeler ve Hususı̂ Yazışmalar
(Istanbul: Everest Yayınları, 2008), 39, 45.
370 hilmar kaiser
Bulgarians.23 Party cadres and government officials coordinated attacks along the
Aegean littoral.24 The authorities’ complicity became evident as criminals were
allowed to continue their outrages and force thousands of Greeks to flee. Then they
recorded displaced Greeks and their property. The state coordinated the displace-
ments and the settlement of immigrants, thereby obstructing the return of the legal
owners.25 The campaign continued into World War I when Greek communities
were officially deported and their property seized. A minimum of 163,975 Greeks
were expelled or fled to Greece, while at least 93,088 were deported to interior
provinces. As many as 58,039 Bulgarians were expelled to Bulgaria.26
THE ARMENIAN REFORM SCHEME

................................................................................................................
Another major challenge for the CUP following the Balkan Wars was the resur-
gence of the Armenian Reform question. Armenian leaders had brought up their
grievances with European powers. The latter forced a new reform plan on the
Ottoman government. The eastern provinces were divided into two regions, each
under a European high commissioner overseeing reforms. The CUP opposed the
scheme and stalled its implementation while threatening the Armenian leadership.
The issue remained unresolved until the beginning of World War I provided the
CUP with an opportunity to abandon the scheme.27
W O R L D WA R I
................................................................................................................
A secret military alliance with Germany gave the CUP new political leverage. In
October 1914, the Ottoman government abrogated the ‘Capitulations’ that had
23 H. Yıldırım Ağanoğlu, Osmanlı’dan Cumhuriyet’e Balkanlar’ın Makûs Talihi Göç (Istanbul: Kum
Saatı, 2001), 346–54.
24 Midhat Şükrü Bleda, İmparatorluğun Çöküşü (Istanbul: Remzi Kitabevi, 1979), 54–6.
25 Bundesrepublik Deutschland, Auswärtiges Amt, Berlin, Politisches Archiv (hereafter: AA-PA),
Türkei 168/11, A 12811, Wangenheim to Bethmann Hollweg, Therapia, June 25, 1914, no. 175; ibid. A
13109, Humbert to Wangenheim, Smyrna, June 25, 1914, no. 2795; copy enclosure in Wangenheim to
Bethmann Hollweg, Therapia, June 6, 1914, no. 179; ibid. A 13846, Schönberg to Mutius, Therapia, July
9, 1914; enclosure in Mutius to Bethmann Hollweg, Therapia, July 10, 1914, no. 187.
26 Bardakçı, Evrak-ı Metrûkesi.39, 45, 79, 80–1.
27 Armen Garo, Bank Ottoman. Memoirs of Armen Garo, trans. Haig T. Partizian, ed. and intro.
Simon Vratzian (Detroit: Armen Topozian, 1990), 182–92.
granted foreign nationals some immunity from Ottoman law. The CUP also
wanted to pre-empt foreign interventions in support of the non-Turkish commu-
nities.28 Now the CUP targeted Zionist settlers in Palestine. The Ministry of
Interior installed Manastirli Bahaeddin Bey, its expert for Zionist affairs and a
member of the TM, as governor of Jafa. The official announced that he would
destroy the Zionist movement; all foreign Jews would have to leave while new
arrivals would not be permitted. Jewish settlers who had become Ottoman citizens
were harassed and others discouraged from applying. On 5 November 1914, autho-
rities searched Jewish houses for arms and documents, demanding the surrender of
all guns legally owned by Zionists. The latter were forced to purchase guns to meet
the number of guns demanded from them. At the same time, Bahaeddin promised
Muslim villagers the Zionist settlers’ land and women so that fear of massacre
spread. On 17 December 1914, police rounded up and deported Jews to Egypt. Due
to German and US opposition, however, the campaign slowed down until it gained
new momentum in 1917.29
N E S TO R I A N S
................................................................................................................
Throughout the summer of 1914, the Ottoman government was concerned about
Russian advances towards the Kurdish and Nestorian population. On 26 October
1914, Talaat ordered the deportation of the Nestorians from Hakkari to western
provinces where they should be dispersed among Muslims. Three days later, the
scheme was, however, postponed for lack of forces and soon abandoned for the
time being.30 The deportation scheme was probably connected with the imminent
Ottoman attack on Russia. Apparently, the CUP suspected that the Nestorians
would join forces with Russians and tried to pre-empt that danger. Seeing that its
suspicions did not materialize, the CUP shelved the project. Avoiding deportation
did not, however, mean escaping massacre during Ottoman cross-border raids into
28 Ulrich Trumpener, Germany and the Ottoman Empire, 1914–1918 (Princeton: Princeton
University Press, 1968), 38–9; AA-PA, Türkei 159, no. 2/14, A 32771, Neurath to Bethmann Hollweg,
Pera, Nov. 5, 1915, no. 654.
29 Israel, Central Zionist Archives, Jerusalem, Z 3/52, Ruppin to Lichtheim, Jaffa, Mar. 24, 1915,
enclosure to Action–Committee to Members, Berlin, Apr. 4, 1915; Isaiah Friedman, Germany, Turkey
and Zionism, 1897–1918 (Oxford: Oxford University Press, 1977); Henry Morgenthau, Ambassador
Morgenthau’s Story (Garden City, NY: Doubleday, Page, 1918), 370.
30 Türkiye Cumhuriyeti, Başbakanlık Osmanlı Arşivi, Istanbul, Dahiliye Nezareti (hereafter: DH)
Şifre Kalemi (hereafter: ŞFR) 46–78, Talaat to Van province, Oct. 26, 1914, Emniyet-i Umumiye
Müdiriyeti (hereafter: EUM) Spec. 104; ibid. 46–102, Minister to Van province, Oct. 29, 1914, EUM
Spec. 107; ibid. 46–195, Talaat to Van province, Nov. 5, 1914, EUM Spec. 113.
372 hilmar kaiser
Iran or following the collapse of the Ottoman front along the Iranian border in
1915. Further inland, Syrian Orthodox Christians were massacred alongside Arme-
nians in Diarbekir province.31 Their killing appears, however, not to have been part
of a central government policy. The provincial authorities were repeatedly in-
structed to treat Syrian Orthodox differently than Armenians.32
ARMENIANS
................................................................................................................
In August 1914, Ottoman TM units started undercover cross-border operations into
Russian territory. The goal was the softening of Russian border defences and to
foment a rebellion among the Muslim population.33 Following an Ottoman naval
attack on Russian Black Sea ports on 29 October 1914, France, Britain, and Russia
declared war on the Ottoman Empire. On 14 November 1914, the Ottoman Empire
declared a ‘Holy War’. In December 1914, the Ottoman Third Army launched an
offensive, hoping to take the Russians by surprise. The attack failed, however, and
the army was almost annihilated. During their retreat, irregular units and fleeing
Muslim civilians began plundering and massacring Armenian villages. The atro-
cities were motivated by the lack of supplies and a desire for revenge.34 By spring
1915, the Ottoman army suffered further defeats on Sinai and the Mesopotamian
and Iranian fronts. On 18 March 1915, an allied fleet nearly forced the Dardanelles.35
In view of this danger, the Ottoman government prepared to evacuate the govern-
ment and the imperial court from Constantinople.
Following defeat in Iranian Azerbaijan, Ottoman troops turned the region
before the advancing enemy into a wasteland. They annihilated local Armenians
although the latter had remained loyal. The campaign’s principal target was the
Armenian community in Van. The city was the Armenian political centre of the
eastern provinces. Here, Armenian leaders organized and coordinated political
31 Precise information about these atrocities is still lacking, but it appears that almost all surviving
Nestorians fled the area. David Gaunt, Massacres, Resistance, Protectors: Muslim-Christian Relations in
Eastern Anatolia During World War 1 (Piscataway, NJ: Gorgias Press, 2006); Üngör, Reign of Terror.
32 See for instance: DH.ŞFR 57–102, Minister to Diarbekir, Bitlis, Harput, Aleppo provinces, Urfa
district, Oct. 25, 1915, EUM Spec. 91, 51, 86, 113, 42; ibid. 70–79, Shukru to Diarbekir province, Nov. 23,
1916, Aşâir ve Muhâcirı̂n Müdiriyeti Umûmiyyesi, Settlement Dept. Gen. 1089.
33 Arif Cemil, I. Dünya Savaşı’nda Teşkilât-ı Mahsusa (Istanbul: Arba Yayınları, 1997), 27–32.
34 Hilmar Kaiser, ‘ “A Scene from the Inferno”: The Armenians of Erzerum and the Genocide, 1915–
1916’, in Hans-Lukas Kieser and Dominik Schaller (eds), Der Völkermord an den Armeniern und die
Shoah / The Armenian Genocide and the Shoah (Zürich: Chronos Verlag, 2002), 129–86.
35 Carl Mühlmann, Das deutsch-türkische Waffenbündnis im Weltkrieg. Mit einem Geleitwort von
W. Foerster (Leipzig: Koehler and Amelang, 1940), 43.
campaigns and, in case of need, the community’s self-defence. Therefore, the

Ottoman civilian and military authorities gave the elimination of this suspected
internal foe priority over combating the Russians. The government forces attacked
on 20 April 1915 and massacred surrounding Armenian villages with few excep-
tions. In the city, however, the government forces were defeated by Armenian
defenders and withdrew on 17 May 1915 before Russian forces reached the city.36
Having received news about the defence at Van and expecting an imminent
attack on the Dardanelles, Talaat ordered the arrest of Armenian community
leaders empire-wide on 24 April 1915. In Constantinople, the police rounded up
Armenian, among others, journalists, clerics, politicians, and teachers. The arrested
were sent to the interior where the majority was killed. Armenian elites in the
provinces suffered the same fate.37
I N I T I A L D E P O RTAT I O N S
................................................................................................................
The April arrests were not the start of general deportations although Armenians
from Zeitun were re-directed to Der Zor, the new main destination for deportees.
On 2 May 1915, the Ministry of War suggested the deportation of all Armenians
from the eastern border regions to the Russian lines or to interior provinces. As in
the case of the Greeks, the Armenians should immediately be replaced by Muslim
settlers.38 Within a week, the governors of Erzerum, Bitlis, and Van provinces were
ordered to deport, in cooperation with the army, Armenians towards the Syrian
Desert and Mosul.39 On 23 May 1915, the deportations were extended to districts in
historic Cilicia and border areas of Mosul province. Armenians could keep their
movable possessions. They were to be distributed in villages and small towns. New
Armenian villages had to be at least twenty-five kilometres away from railway lines
and could not exceed fifty households. Moreover, Armenians must never exceed
36 Onnig Mukhitarian, An Account of the Glorious Struggle of Van—Vasbouragan, trans. Samuel S.

Tarpinian ([Detroit, MI]: General Society of Vasbouragan, 1980).
37 Hikmet Özdemir and Yusuf Sarınay (eds), Türk-Ermeni İhtilafı. Belgeler (Turkish–Armenian
Conflict: Documents) (Ankara: Türk Büyük Meclisi Kültür, Sanat ve Yayın Kurulu Yayınları, 2007),
18–27. Yusuf Sarınay, ‘What Happened on April 24, 1915?: The Circular of April 24, 1915, and the Arrest
of the Armenian Committee Members in Istanbul’, International Journal of Turkish Studies 14:1–2
(2008), 75–101.
38 Özdemir and Sarınay, İhtilâfı, 33–7.
39 DH.ŞFR 52–281, Minister to Tahsin Bey, May 7, 1915 EUM Spec. 4090; ibid. 52–282, Minister to
Tahsin, Mustafa Abdülhalık, Djevdet, May 9, 1915, EUM Spec. Dept. 409[0?]; Türkiye Cumhuriyeti,
Başbakanlık Devlet Arşivleri Genel Müdürlüğü, Osmanli Arşivi Daire Bakanlığı, Osmanlı Belgelerinde
Ermeniler (1915–1920) (Ankara: Başbakanlık Basımevi, 1994), 28–9.
374 hilmar kaiser
ten per cent of the Muslim population. Local authorities were responsible for the
execution of the plan.40
On 24 May 1915, the Entente powers declared that they would hold all Ottoman
citizens and officials personally accountable for their role in the persecution of the
Armenians. The declaration defined the outrages that had occurred so far as crimes
against humanity.41 The Ottoman government responded on 27 May 1915 with a
‘provisional law’ that gave Ottoman military commanders the right to deport
anyone they wished to. Three days later, the Ministry of the Interior issued a
manual for its Department for the Settlement of Tribes and Immigrants, specifying
the deportations’ implementation.42 Armenian immovable property was to be
seized and used for the settlement of Muslims. Evidently, the deportations were
not a temporary emergency measure but should permanently change the demo-
graphic map.43
A R M E N I A N P RO P E RT Y
................................................................................................................
On 10 June 1915, the Ministry of Interior regulated the liquidation of Armenian
property. All property had to be registered while perishable goods and livestock
were to be auctioned off immediately. The Muslim migrants were settled first, then
nomads. The newcomers had to be capable of farming or operating the businesses
they received. Non-distributed real estate was to be auctioned off. In sum, Arme-
nian owners had lost their property rights and could not object.
On 26 September 1915, the government passed a provisional law concerning the
property of Armenian deportees. Nominal attention was given to the rights of
creditors who could present their claims, but obstacles rendered the pursuit of their
interests impracticable. Even if claimants won their case, they could not obtain
already liquidated properties. On 6 January 1916, Talaat stated that the Ottoman
economy had to become an exclusively Muslim one. He decreed that Armenian
property must fall into Muslim hands. On 23 November 1916, the government also
sanctioned the seizure of property of Armenians who had not been deported.
40 Özdemir and Sarınay, İhtilâfı, 46–9, 66–9; DH.ŞFR, 53–91, Minister to Mosul province, Urfa, Zor
districts, May 23, 1915, EUM; ibid. 53–92, Minister to Adana, Aleppo provinces, Marash district, May
23, 1915, EUM; ibid. 53–93, Minister to Erzerum, Van, Bitlis provinces, May 23, 1915, EUM; Osmanlı
Belgelerinde Ermeniler, 33–4; Bloxham, Great Game, 84–5. For an overview based largely on Armenian
sources see Raymond Kévorkian, Le génocide des arméniens (Paris: Odile Jacob, 2006).
41 Arthur Beylerian (ed.), Les grandes puissances, l’Empire ottoman et les Arméniens dans les archives
françaises (1914–1918), Recueil de documents (Paris: Publications de la Sorbonne, 1983), 29.
42 Özdemir and Sarınay, İhtilâfı, 64–5, 72–81; Kaiser, ‘Négation’, 81–3.
43 Özdemir and Sarınay, İhtilâfı, 72–6, 78–81.
The Ottoman government faced strong competition for the loot throughout the
expropriation. Local elites, government officials, common criminals, and Arme-
nians themselves succeeded in obtaining or hiding parts of property that otherwise
would have been transferred to the government.44
T H E S E C O N D WAV E OF D E P O RTAT I O N S
................................................................................................................
As with the expropriations, by the extension of deportations far beyond the war zone
the Ottoman government demonstrated that military considerations were not the
driving force behind the deportation programme. On 19 June 1915, the Third Army
ordered the extension of deportations and the removal of all Armenians from the
Erzerum, Trebizond, Van, Bitlis, Harput, Diarbekir, and Sivas provinces.45 The vast
extension of the deportation programme necessitated adjustments by the authorities.
Given the massive number of deportees, recordkeeping emerged as a major task.
Repeatedly, the Ministry of Interior demanded information on Armenian villages,
their location and agricultural potential, number of deportees and the route along
which they had been deported, Armenian real estate, and suggestions for settlement
of prospective settlers.46 On 22 June 1915, Armenians who had converted to Islam
were allowed to stay behind for the time being and had to undergo special registra-
tion. The order was personally addressed to the provincial governors, who were all
trusted CUP members. They had to keep it secret and destroy the evidence.47
Throughout July 1915, the central authorities requested population data and
information on the progress of the deportations and their impact on the ethnic
make-up of the areas in question. The region around Diarbekir had been ear-
marked for Turkification and the settlement of other Muslim nationalities prohib-
ited.48 As Armenians were not to exceed ten per cent of the population, the
44 Hilmar Kaiser, ‘Armenian Property, Ottoman Law, and Nationality Policies during the Armenian
Genocide, 1915–1916’, in Olaf Farschild, Manfred Kropp, and Stephan Dähne (eds), The First World War
as Remembered in the Countries of the Eastern Mediterranean (Würzburg: Ergon Verlag, 2006), 49–71.
45 Türkiye Cumhuriyeti and Genelkurmay Başkanlığı, Arşiv Belgeleriyle Ermeni Faaliyetleri 1914–
1918, vol. i (Ankara: Genelkurmay Askeri Tarih ve Stratejik Etüt Başkanlığı Yayınları, 2005), 187–8.
46 DH.ŞFR 54–15, Ali Munif to Adana, Aleppo, Erzerum, Bitlis, Van, Diarbekir provinces, Marash
district, June 14, 1915, İskan-ı Aşair ve Muhacirin Müdiriyeti (hereafter: IAMM); ibid. 54–136, Ali
Münif to Harput, Trebizond, Sivas provinces, Canik district, June 24, 1915, IAMM.
47 DH.ŞFR 54–100, Minister to Governors of Van, Trebizond, Erzerum, Bitlis, Harput, Diarbekir,
Sivas, Djanik, June 22, 1915, EUM Spec. 4531.
48 DH.ŞFR 54–272, Minister to Diarbekir province, July 3, 1915, IAMM 54; ibid. 54/A-51, Minister to
all provinces districts, July 20, 1915, EUM Spec. 397; ibid. 54/A-100, Minister to Erzerum, Adana, Bitlis
Aleppo, Diarbekir, Sivas, Trebizond, Harput, Van provinces, Urfa, Djanik, Marash districts, July 24,
1915, EUM Gen. 400.
376 hilmar kaiser
government had to face an increasingly large ‘surplus’ population that survived the
death marches from the eastern provinces. As a countermeasure the settlement
zones were broadened within the Mosul and Aleppo provinces and extended to
today’s Southern Syria and Jordan.49
The unfolding human disaster soon reached such proportions that the authorities
were faced with an unexpected problem. Thousands of Armenian children had been
left behind along supply lines and in cities and villages. In many places, local Muslims
had taken children in, be it for altruistic or other reasons. Now the government tried
to regain control of the situation. The Ministry of Education was put in charge of the
children. Those younger than ten years of age were to be admitted to government
orphanages, but adequate facilities were lacking or non-existent.50
Many families had given up children in face of imminent massacre or because
they could no longer provide for them. Massacres in villages and towns were mostly
avoided to preclude loss of property by plundering and burning like in the 1890s and
in 1909.51 The killing of Armenians during the deportations took place in several
stages. While most Armenian men had been drafted into the army, there were still
many older men, teenagers, deserters, and those who had paid the military exemp-
tion tax among the deportees. The government escorts, mostly gendarmes or local
militia, usually separated most of the men from the caravans during the first days of
the deportation. The men and boys were then taken away and killed. Males who
either succeeded in disguising themselves as women or managed to bribe the guards
gained a temporary reprieve from massacre. The men usually did not resist their
murderers, not wanting to endanger the lives of their families.52
At specific locations along the route the deportees were camped for some days
while several convoys were merged. During this respite, the deportees were
informed that a massacre was imminent and asked to give up their children and
teenage girls or to buy their way out of the predicament by paying a ransom. In this
way, the authorities induced the deportees to surrender large amounts of money
and valuables that otherwise could be obtained only with difficulty and effort as the
deportees often swallowed precious stones and coins. Similar to a modern turnpike
system, the convoys passed successive notorious massacre sites; where those who
could not longer pay were murdered. Following the massacres, the killer squads
searched the corpses for valuable documents, like life insurance policies and
financial documents that could be used by the state to claim the benefits as being
the legal heir of the victims.
49 Özdemir and Sarınay, İhtilâfı, 144–8.

50 DH.ŞFR 54–163, Minister to Harput province, June 26, 1915, EUM Spec. 4573; ibid. 54–411, Ali
Munif to provinces, July 12, 1915, IAMM 378; ibid. 54/A-54, Ali Munif to Sivas province, July 20, 1915,
IAMM 66; Özdemir and Sarınay, İhtilâfı, 118–19.
51 Kaiser, ‘Armenian Property’.
52 AA-PA Türkei 183/40, A 32613, Sarkis Manukian to Rössler, Aleppo, Nov. 25, 1915; enclosure 2 in
Rössler to Bethmann Hollweg, Aleppo, Nov. 30, 1915, K. no. 110, no. 2725.
The killing squads operated under the command of trusted CUP members who
were either administrators or officers, or had been assigned to the TM. The
composition of these units varied from district to district. While TM units were
responsible for massacres close to the combat zone, the gangs in the interior
districts included local militiamen, gendarmes, Kurdish tribal groups, or cut-throats
hired from the local population for the occasion. The convoys’ escorts cooperated
with the killing squads in coordinating the delivery of the victims at the assembly
points so that comparably few men could murder large numbers of victims with
swords and daggers within a short period of time. Major massacre sites, located near
transit camps like the Kemah gorge near Erzindjan, Lake Khazar near Harput, and
the valleys south of Firindjilar near Malatia, shared certain characteristics. The sites
were mountainous but still close to overland roads. Thus, the deportees could be
easily transported to the massacre sites while having relatively few escape routes.
Rivers and deep gorges facilitated the hiding or disposal of corpses.
The killers were rewarded with a part of the stolen property and women and
children, depending on availability or preferences. Armenian boys survived in
Muslim villages and among tribes as slaves working as either shepherds or farm
hands. Their survival was, nevertheless, uncertain as replacements were cheaper
than adequate food rations. Girls were mostly used as house servants or sex slaves
and, if lucky, as brides within the family. This way the family saved payment of a
dowry, a major expense within the life cycle of a Muslim peasant family. Moreover,
converted Armenian children who were raised by Muslims could keep the inheri-
tance from their original families, thereby generating considerable new income.
The absorption of Armenian women into Muslim families presented no problem as
the women were forced to become Muslim and the nationality of the males defined
the identity of the family. Some, however, objected to the CUP’s extermination
policy and wanted to save a person from abuse and almost certain death.53
Following the murder of most men and older boys during the initial phase of the
deportations, the history of the Armenian genocide became by and large that of
women and children. It marked a reversal of gender roles within the Armenian
community where women were subjected to patriarchal structures that limited
their independent decision-making and contacts with outsiders. Moreover, women
from exclusively armenophone communities had to cope with situations where they
could only communicate with other deportees. Mostly women and children became
the target of random and systematic rape. The violence was targeted at destroying the
Armenian individuals’ and community’s self-perception by inflicting lasting psycho-
logical harm. Rape meant an irreparable transgenerational loss of self-esteem, or
‘honor’, for Armenians, even if they lived beyond the reach of the Ottoman govern-
ment. The women’s counter-strategies included disguising their age, forms of physical
53 Kaiser, ‘A Scene’; Özdemir and Sarınay, İhtilâfı, 194–5.

378 hilmar kaiser
individual or joint resistance, but often suicide was the only ‘protection’ from rape
left.
Escape from the caravans was no common option as the deportations had elimi-
nated places of refuge within whole provinces. Nevertheless, Armenian women
developed strategies defying the CUP’s genocidal logic. Elder women gave up their
food ration for the children and girls were given up before the boys. Once only boys
were left, the mother tried to protect at least one male descendant at the cost of her
own life. Thus, women focused not on individual survival but that of the family as
they understood it, and were willing to engage in activities that stood in extreme
contrast to their social values.54
T H E T H I R D WAV E OF D E P O RTAT I O N S
................................................................................................................
By early August 1915, the government extended the deportations to the central and
western provinces. Areas with small Armenian communities were excluded as were
some urban communities in part.55 Now the government sanctioned the ongoing
dispersion of young Armenian children among non-Turkish Muslim villages. The
government’s own assimilation efforts gained new momentum in 1916. It appears
that the authorities relied on private and foreign institutions to keep the children
alive until government institutions were put in place.
Unlike in the eastern provinces, large numbers of Armenians were deported by
railway or along the railway lines. Again, the authorities tried to keep track of the
deportees, their number, place of origin, and current location. Following protests
from the US, German, and Austro-Hungarian embassies, Catholic and Protestant
Armenians, who had so far not been deported, were exempted, unless they con-
stituted a concern. Moreover, the number of remaining Armenians and their
percentage of the total population had to be ascertained. Protests and production
breakdowns secured the permanent or temporary exemption of Armenian
54 Eliz Sanasarian, ‘Gender Distinction in the Genocidal Process: A Preliminary Study of the
Armenian Case’, Holocaust and Genocide Studies 4:4 (1989), 449–61; Ara Sarafian, ‘The Absorption
of Armenian Women and Children into Muslim Households as a Structural Component of the
Armenian Genocide’, in Omer Bartov and Phyllis Mack (eds), In God’s Name. Genocide and Religion
in the Twentieth Century (New York/Oxford: Berghahn Books, 2001), 209–21. Armen Anush, Passage
Through Hell: A Memoir, trans. Ishkhan Jinbashian (Studio City, CA: Hagop and Knar Manjikian,
2005). Donald E. Miller and Lorna Toryan Miller, An Oral History of the Armenian Genocide (Berkeley:
55 DH.ŞFR 54/A-200, Minister to 4th Army command, July 31, 1915, IAMM Statistical Dept. 63.
Osmanlı Belgelerinde Ermeniler, 81.
specialists and workers, while families of Armenian soldiers that still remained
behind were distributed among Muslim villages.56
Already at the end of August 1915, Talaat was certain that the Armenian Question
had been solved.57 In the western provinces, however, the large number of depor-
tees created considerable problems for operating the railways and increased the
chaos in and around Aleppo, the cross-roads of the deportation. Streamlining the
transfer of deportees and recordkeeping remained a major challenge. In response,
Talaat dispatched a special envoy, Shukru Bey, to the railway and northern Syria.
Increasingly, deportee convoys from central and eastern regions were re-directed
towards Mosul.58 The measure came in response to the rapid progression of
deportations in the western and central provinces. By 18 September 1915, many
districts had already been entirely emptied. Exempted Armenians had been dis-
persed among Muslim villages up to five per cent of the population. Two months
later, Talaat decreed that those exempted had to become Muslims.59
On 7 October 1915, Shukru Bey issued new regulations for the deportations to the
Syrian Desert. He had established new administrative offices along the lower
Euphrates for the coordination of the deportations. Henceforth, the optimal size
of a deportation caravan was 1,000 deportees. Rakka, Harran, and Der Zor were
designated as settlement regions for hundreds of thousands who had to be individ-
ually registered. The extension of the deportations increased the government’s
expenses. In response, the authorities used proceeds from Armenian property for
the financing of deportations.60
In spring 1916, the deportations from interior and eastern provinces came to a
halt as relatively few Armenian women and children were left. These were to be
dispersed in exclusively Muslim villages or taken into orphanages. Women of
childbearing age had to marry Muslim men.61
56 DH.ŞFR 55–206, Minister to Sivas province, Aug. 25, 1915, EUM Gen. 5256; ibid. 55–323, Talaat to
Angora province, Aug. 30, 1915, EUM Spec. 67; ibid. 59–111, Hasan Shafi, Dept. IAMM Director, to
provinces, Dec. 27, 1915, IAMM Gen. 790; Hilmar Kaiser, ‘The Baghdad Railway and the Armenian
Genocide, 1915–1916: A Case Study in German Resistance and Complicity’, in Richard G. Hovannisian
(ed.), Remembrance and Denial: The Case of the Armenian Genocide (Detroit: Wayne State University
Press, 1998), 67–112; idem, Cross-roads; Özdemir and Sarınay, İhtilâfı, 174–5, 178–9, 208–9; DH.ŞFR 54/
A-294, Minister to Izmid district, July 28, 1915, EUM Spec. 43; DH.ŞFR 55–211, Minister to Syria,
Adana, Konia, Ankara, Aleppo provinces, Urfa, Zor Marash districts, Aug. 25, 1915, EUM Spec. 5258;
ibid. 55–208, Minister to provinces, Aug. 25, 1915, EUM Spec. 5260; ibid. 55/A-230, Talaat to Adana
province, Sunday Sept. 12, 1915, EUM Spec. 64.
57 AA-PA Türkei 183/38, A 26474, Hohenlohe to Bethmann Hollweg, Pera, Sept. 4, 115, no. 549;
Özdemir and Sarınay, İhtilâfı, 234–7.
58 Özdemir and Sarınay, İhtilâfı, 262–3, 272–3; DH.ŞFR 55/A-145, Talaat to provinces, Sept. 18, 1915,
EUM Gen. 568.
59 Osmanlı Belgelerinde Ermeniler, 93–96; DH.ŞFR 57–281, Talaat to provinces, Nov. 4, 1915, EUM
Spec. 6270.
60 Özdemir and Sarınay, İhtilâfı, 300–15, 380–3.
61 Ibid. 424–6, 438–9, 442–3.
380 hilmar kaiser
T H E L OW E R E U P H R AT E S
................................................................................................................
Those deportees that survived the death march and reached the Syrian Desert
found themselves trapped in a system of camps that stretched along the Euphrates
to Der Zor. In the camps, the Ottoman authorities exposed them to contagious
diseases like typhus which killed hundreds daily. Refusal to provide water and food
increased the high mortality among the emaciated victims. Here as well, Muslims
were eager to obtain Armenian women. Authorities registered such marriages but
did not record the deaths of the former Armenian husbands.62
In January 1916, the route along the Euphrates was littered with human corpses
and bones, bearing testimony to the ongoing killing of deportees and death by
exhaustion, famine, and disease.63 The government monitored the number and
death of deportees but recordkeeping became more complicated when the autho-
rities started emptying outlying camps and forced the Armenians into the desert
camps around Der Zor. In June 1916, deportations to Mosul and Syria were
stopped. Consequently, Armenians had to be distributed within the Der Zor region
and adjoining districts of Aleppo province.64
The local economy could not support the tens of thousands of Armenians in the
small desert settlements while the authorities watched the impact of a killing
famine and ravaging pandemics. Nevertheless, Talaat accelerated the dispatch of
Armenians into the disaster zone. The authorities reacted swiftly when they learned
that bribed gendarmes were allowing deportees to escape from the desert and
return to Aleppo.65
In July 1916, Talaat repeatedly ordered the dissolution of Armenian camps near
the military supply lines along the Euphrates River. He declared that Armenians
were a military threat and must be moved to appropriate locations. Accordingly,
62 Raymond Kévorkian, ‘Le sort des déportés dans les camps de concentration de Syrie et de
Mésopotamie’, in Revue d‘histoire arménienne contemporaine 2 (1998), 7–61; idem, ‘Témoignages sur les
camps de concentration de Syrie et de Mésopotamie’, ibid. 62–215; idem, ‘Autres témoignages sur les
déportations et les camps de concentration de Syrie et de Mésopotamie (1915–1916)’, ibid. 219–44;
Helmut Becker, Äskulap zwischen Reichsadler und Halbmond. Sanitätswesen und Seuchenbekämpfung
im türkischen Reich während des Ersten Weltkriegs (Herzogenrath: Verlag Murken-Altrogge, 1990); DH.
ŞFR 58–114, Talaat to Suad Bey, Oct. 24, 1915, EUM Spec. 47.
63 AA-PA Türkei 183/41, A 5498, Rössler to Bethmann Hollweg, Aleppo, Feb. 9, 1916, no. 366 K, no. 18;
Rahmi Apak, Yetmişlik Bir Subayın Hatıraları (Ankara, Türk Tarih Kurumu, 1988), 149.
64 DH.ŞFR 62–199, Minister to Zor district, Apr. 1, 1916, EUM Spec. 6; ibid. 64–165, Talaat to Zor
district, May 31, 1916, EUM 22; ibid. 64–175, Talaat to Zor district, June 1, 1916, EUM Gen. 47469; ibid.
64–239, Talaat to Aleppo province, June 7, 1916, EUM Gen. 47701 Spec. 79; ibid. 64–248, Talaat to Zor
district, June 7, 1916, EUM Gen. 47702 Spec. 25; ibid. 65–51, Talaat to Aleppo province, June 18, 1916,
IAMM, Settlement Dept. Spec. 85.
65 DH.ŞFR 64–292, Minister to Zor district, June 13, 1916, IAMM 28; ibid. 66–21, Talaat to Diarbekir
province, July 18, 1916, EUM Gen. 49557 Spec. 39; ibid. 66–43, Talaat to Aleppo province, July 20, 1916,
EUM Spec. 72.
Armenians in the employ of military transport had to be replaced by Muslims,

thereby depriving the deportees of their only income. On 8 August 1916, Talaat
ordered a count of all Armenians in the region and inquired into their percentage
of the population. By the time of this order, the governor of Der Zor organized
large-scale massacres. Near Der Zor, the governor burned alive approximately one
thousand Armenian children. At the Khabur River, Chechen killing squads that
had already massacred the inhabitants of the Rasulain camp slaughtered under
government supervision Armenians by the tens of thousands. Again, the concen-
tration of Armenian deportees assured that comparably few perpetrators could kill
huge numbers of victims with axes and knives within a relatively short time while
the authorities secured the victims’ remaining valuables.66
Following these major massacres, the Ottoman government geared down its
extermination programme. Der Zor remained a destination for deportees, partic-
ularly political suspects. The Armenians were continuously registered. In October
1916, the settlement commission at Rakka forwarded a list to Constantinople
identifying 373 deportees from Amasia, Marash, and Hadjin.67 Tracking surviving
Armenians remained a concern for Talaat. He continued to monitor the progress,
assimilation, and annihilation through population surveys while ideas for new
settlement projects continued to be developed.68
The annihilation of the Armenians provided the government with a material
basis for a restructuring of the Ottoman demographic map. In 1916, the authorities
used the Armenian real estate in western and central provinces for the settlement of
Kurdish tribal confederations. Like the last remaining Armenians, the newcomers
were settled in small groups among Turkish-speaking Muslims to assimilate them.
Their leaders were detained in distant locations. The area around Diarbekir, Urfa,
and Adiaman received Muslim migrants, especially Albanians, and was designated
as a ‘Turkification region’. Kurds, Druze, Greeks, Shiites, and a variety of smaller
groups were subjected to the programme. By ‘Turkifying’ entire regions as well as
other non-Turkish and non-Muslim communities, the CUP sought to remove
competing claims to Ottoman territory. The Kurds of Dersim had become alarmed
by the extermination of Armenians as early as July 1915, fearing that they would be
66 DH.ŞFR 66–19, Talaat to Zor district, July 19, 1916, EUM; ibid. 66–94, Talaat to Zor district, July
29, 1916, EUM Gen. 43 Spec. 1676; ibid. 66–159, Talaat to Zor district, Aug. 6, 1915, EUM Spec. 46; ibid.
66–170, Talaat to Zor district, Aug. 8, 1916, EUM Spec. 47; Ara Sarafian, compiler, United States Official
Records on the Armenian Genocide 1915–1917 (Princeton/London: Gomidas Institute, 2004), 555–60;
AA-PA Türkei 183/45, A 31831, Rössler to Radowitz, Aleppo, Nov. 5, 1916, K. no. 104; no. 3045; Anush,
Passage; Kaiser, ‘A Scene’, 184–5.
67 DH.ŞFR 69–261, Talaat to Marash district, Nov. 13, 1916, EUM Gen. 2942 Spec. 34; DH. EUM.
Gen. 7–17/A, Rakka Settlement Commission to Ministry of Interior, Oct. 27, 1916.
68 DH.,ŞFR 69–120, Talaat to provinces, Oct. 29, 1916, EUM KLU 53999 Spec. 559; Türkiye
Cumhuriyeti, Genelkurmay Başkanlığı, Arşiv Belgeleriyle Ermeni Faaliyetleri 1914–1918, vol. vii
(Ankara: Genelkurmay Askeri Tarih ve Stratejik Etüt Başkanlığı Yayınlari, 2007), 320.
382 hilmar kaiser
killed next. In response they sheltered Armenian refugees and formed joint
Kurdish–Armenian resistance groups.69
T H E D E AT H T O L L
................................................................................................................
The number of Armenian victims is unknown. Ottoman data filed in 1914 suggest
an Armenian population of 1,281,173 or, adjusted for undercounting, at least
1,718,132.70 Government data from 1915 shows that in Erzerum and Bitlis provinces
98,178 Armenians had been killed in clashes or had fled. The 67,792 Armenians of
Van province were not deported, reflecting massacres and also Armenian resistance
and flight. Of these 165,970 Armenians about 65,081 survived by escaping to
Russian lines.71 By April 1917, 284,157 Armenians remained, bringing the number
of Armenians accounted for to 349,238. The statistics probably did not include
Armenians in hiding and those who had converted. These formed, however, only a
fraction of the 1,368,894 Armenians who had disappeared from Ottoman records.
This suggests that over 1.1 million Armenians had lost their lives due to government
policies. Possibly more than 150,000 Armenians had been forcibly assimilated.
These low estimates are based on government data and do not cover Armenian
losses in 1917 and 1918.72 It must be emphasized that most survivors suffered from
physical injuries, diseases, and psychological traumata. Many had been maimed.
When these medical emergencies are taken together with the survivors’ age and
gender composition, and the annihilation of the secular and religious elites, the
69 DH.ŞFR 54/A-128, Talaat to Erzerum, Diarbekir Bitlis provinces, July 25, 1915, Spec. Dept.; Adanır
and Kaiser, ‘Migration’. See also Dündar, Türkiye’nin Şifresi.
70 This figure represents a low estimate. An Armenian census arrived in 1913 at 1,914,620 Armenians
while leaving out some communities. Mutlu adjusted 1914 Ottoman data by 34% for undercounting.
Ottoman statistics from the Turkish military archives gives a total of 1,219,323 Armenians, but omits at
least 65,850 Armenians of Diarbekir province. Thus bringing the adjusted Armenian population to
1,718,132. Servet Mutlu, ‘Son Dönem Osmanlı Nüfusu: Müslüman ve Ermeniler’, in Şafak Ural, Feridun
Emecen, and Mustafa Aydın (eds), Türk-Ermeni İlişkilerinde Yeni Yaklaşımlar. Uluslararası Sempozyum
15–17 Mart 2006 (Istanbul: Istanbul Üniversitesi Yayınları, 2008), 57–77; Raymond H. Kévorkian and
Paul B. Paboudjian, Les Arméniens dans l’Empire ottoman a la veille du génocide (Paris: Les Éditions
d’Art et d’Histoire, 1992), 60; Genelkurmay Başkanlığı, Ermeni Faaliyetler, i, 607, 609.
71 Genelkurmay Başkanlığı, Ermeni Faaliyetler, i, 159–70. I have estimated Armenian losses among
those killed at 50% for Van and Erzerum provinces, and at 70% for Bitlis.
72 Bardakçı, Evrak-ı Metrûkesi, 108–39. The document is undated; however, the data on Beirut were
taken from a report compiled in March 1917. DH.EUM 2 Şb 74–55, Muhiddin to Ministry of Interior,
Beirut, March 12, 1917, Chief Secretariat Gen. 374 Spec. 19 Secret.
remaining Armenians were a disintegrating fragment of their former community

rather than resembling just a numerically reduced population.
T H E A F T E R M AT H
................................................................................................................
After the Ottoman defeat, the victorious allies, Armenian survivors, and the
Ottoman opposition tried to bring the CUP criminals to justice. Many of the key
figures, among them Talaat, had, however, fled. Others were protected by post-war
Ottoman administrations. Nevertheless, some perpetrators were condemned by
Ottoman courts. The government had an interest in putting the blame for the
killings on the TM and Kurds, thereby exculpating the state. The proceedings
produced documents that provided insights into the organization of the mas-
sacres.73 Within months, the CUP regrouped and began, under its new name, the
National Movement, a successful campaign against Armenian survivors, Greeks,
and the Western powers. The CUP overcame its internal opponents and competi-
tors for territories within the Ottoman core provinces outside the Arab-speaking
regions. In the aftermath of the 1921–2 Greco-Turkish war with its huge interethnic
violence, the international community sanctioned with the Treaty of Lausanne in
1923 the realities created by the CUP during the Armenian genocide, and furthered
the ethnic homogenization of Anatolia by sanctioning the Greco-Turkish ‘popula-
tion exchange’.74
C O N C LU S I O N
................................................................................................................
The Armenian deportations were not the result of an Armenian rebellion. On the
contrary, Armenians were deported when no danger of outside interference ex-
isted. Thus Armenians near front lines were often slaughtered on the spot and not
deported. The deportations were not a security measure against rebellions but
depended on their absence. The initial deportations resembled earlier measures
against Greeks, Nestorians, and Zionists. The dramatic extension of deportations
in May 1915 showed signs of improvisation. No detailed regulations for the
73 Osman Selim Kocahanoglu, Divân-ı Hârb-i Örfı̂ Muhamekatı Zabıt Ceridesi. Tehcir
Yargılanmaları (1919) (Istanbul: Temel Yayınları, 2007); Kaiser, ‘Dall’impero’.
74 See Ben Lieberman’s chapter in this volume.
384 hilmar kaiser
administration of Armenian property or the assimilation of women and children

existed. The execution of deportations was largely left to provincial administrators
who had to find their own means for the task at hand. Nevertheless, the CUP
understood that the deportations afforded it a unique opportunity to acquire the
Armenian communities’ assets and finance not only the government’s budget but
also an ambitious demographic engineering programme, the Turkification of the
empire’s core provinces. Talaat viewed the negative economic consequences of the
deportations as a price the government had to pay.75
The assimilation of Armenian children and women overwhelmed the state’s
resources and local Muslim initiative became decisive. Thus, many Armenian
children and women entered Kurdish and Arab households while the state
intended the Turkification of Armenian children and young Armenian women.
Local competition for Armenian property was another field where the Turkifica-
tion programme encountered sustained competition from local circles.
The central authorities monitored the deportations and tried to keep precise
counts of deportees. The loss of life caused by death marching, famine, diseases,
and systematic massacres along the routes appears to have been part of the
government’s programme. The government assigned to the deportees settlement
areas in thinly populated districts with very limited arable land. Here they were not
to exceed ten per cent of the Muslim population. Thus, the Ottoman authorities
were faced with an Armenian ‘surplus’ population that could not be accommo-
dated or assimilated. Financing Armenian deportee camps for extended periods of
time was out of the question already at the start of the deportations as the Ministry
of Interior had declared that it was impossible to feed the comparably few Zeitun
deportees. The deaths caused by starvation were therefore not unforeseen but
appear systematic.76 The same holds true for deaths caused by disease. By May
1915, the Ottoman authorities had adequate information about the prevention of
infectious diseases, particularly typhus, but deliberately exposed Armenian depor-
tees to infection. The employment of so-called natural causes for death and
massacres at remote locations in the eastern provinces, hidden from foreign
observers, caused hundreds of thousands of deaths.
Nevertheless, far too many Armenians still survived and reached the lower
Euphrates. Armenian resilience and a series of survival strategies as well as under-
cover relief work made this survival possible. In response, in the summer of 1916,
the central authorities coordinated the concentration of survivors in a few localities
before they were slaughtered under government supervision. However, it seems
that in 1915, the CUP had no coherent single plan for the extermination of the
75 Hilmar Kaiser, ‘Die deutsche Diplomatie und der armenische Völkermord’, in Fikret Adanır and
Bernd Bonwetsch (eds), Osmanismus, Nationalismus und der Kaukasus: Muslime und Christen, Türken
und Armenier im 19. und 20. Jahrhundert (Wiesbaden: Ludwig Reichert Verlag, 2005), 226–7.
76 DH.ŞFR 52–292, Minister to Konia province, May 9, 1915, EUM Spec. 22.
Armenians. It is more likely that Talaat and his associates accepted mass murder as
a viable policy option to be employed in case problems arose. The CUP leadership
habitually opted for genocide in the interest of Turkish nationalism.
Bryce, James, and Arnold Toynbee (eds), The Treatment of Armenians in the Ottoman
Empire, 1915–1916. Documents Presented to Viscount Grey of Fallodon by Viscount Bryce,
uncensored edition, ed. and intro. Ara Sarafian (Princeton: Gomidas Institute, 2000).
Bloxham, Donald, The Great Game of Genocide: Imperialism, Nationalism, and the Destruc-
tion of the Ottoman Armenians (Oxford, Oxford University Press, 2005).
Captanian, Pailadzo, Mémoires d’une déportée (Paris : Flinikowski, 1919).
Dadrian, Vahakn N., The History of the Armenian Genocide: Ethnic Conflict from the Balkans
to Anatolia to the Caucasus (Providence, RI/Oxford: Berghahn Books, 1995).
Kaiser, Hilmar (in collaboration with Luther and Nancy Eskijian), At the Crossroads of Der
Zor: Death, Survival, and Humanitarian Resistance in Aleppo, 1915–1917 (Princeton, NJ:
Gomidas Institute, 2001).
Kévorkian, Raymond, Le génocide des arméniens (Paris: Odile Jacob, 2006).
Kieser, Hans-Lukas, and Dominik Schaller (eds), Der Völkermord an den Armeniern und die
Shoah (The Armenian Genocide and the Shoah) (Zürich: Chronos Verlag, 2002).
Miller, Donald E., and Lorna Toryan Miller, An Oral History of the Armenian Genocide
Odian, Yervant, Accursed Years: My Exile and Return from Der Zor, 1914–1919, trans. Ara
Stepan Melkonian, intro. Krikor Beledian (London: Gomidas Institute, 2009).
Trumpener, Ulrich, Germany and the Ottoman Empire, 1914–1918 (Princeton: Princeton
chapter 19
.............................................................................................
MASS
D E P O RTAT I O N S ,
ETHNIC
CLEANSING, AND
G E N O C I DA L
P OLITICS IN THE
L AT E R RU S S I A N
EMPIRE AND THE
USSR
.............................................................................................
nicolas werth
Between the middle of the nineteenth century and the middle of the twentieth, the
immense areas of the Russian Empire and then the Soviet Union were the scene of
extreme forms of state violence, though of course in very different historical
contexts. This violence included massive deportations, ethnic cleansing, and fam-
ines intentionally inflicted in order to end peasant and nationalist resistance to
genocide in the later russian empire and the ussr 387
Soviet policies. Can we discern continuities in the policies of state violence as they
were conducted by the Czarist regime and by the Bolshevik regime that issued from
the revolution of October 1917?
At first sight, the categories of the population targeted by the two regimes seem
very different: the Czarist state’s violence was directed chiefly against ethnic
minority groups, and took place in areas of ‘internal colonization’ on the margins
of the immense Russian Empire—the Caucasus and Central Asia. In contrast,
Soviet state violence was aimed first at ‘class enemies’ (‘kulaks’, ‘bourgeois’, and
‘people of the past’) in the framework of a revolutionary struggle whose objective
was the ‘construction of socialism’. However, if we look more closely, this difference
becomes less sharp in the mid-1930s. At that time we see a remarkable change in the
criteria of discrimination: class criteria are replaced by ethnic discriminations that
culminate in the great wave of deportations of vast numbers of ‘punished peoples’
from 1941 to 1944. This change must surely be seen as paralleling the change in the
Soviet political system: once ‘socialism’ and the ‘classless society’ were constructed
(which had been officially proclaimed ‘completed’ at the Party’s Seventeenth
Congress in early 1934), emphasis was henceforth placed on strengthening the
cohesion of the enormous ‘state of nations’ that the USSR had become, a new
type of empire. In a context of growing international tensions, and with another
world war on the horizon, the question of minorities in diaspora (Soviet citizens of
Polish, German, Baltic, Korean, Chinese, Iranian, Afghan, Turkish, etc. origin)
suspected of being a ‘hotbed of spies and saboteurs in the pay of hostile enemy
powers’ became increasingly crucial.
At the same time, the problem of national minorities that were still not very
‘Sovietized’ twenty years after the Revolution arose with a vengeance and those
involved were for the most part the same mountain peoples of the Caucasus who
had given the Czarist regime so much trouble. During the ‘Great Patriotic War’, the
Soviet regime used alleged collaboration with the German occupying forces as a
pretext for dealing ‘definitively’ with these peoples who were resisting Sovietization
by ‘punishing’ them, that is, by deporting all of them ‘in perpetuity’. The persis-
tence of the imperial dimension, certain unintegrated groups’ resistance to Russi-
fication or Sovietization, the context of armed conflicts (colonial conquest, the
First and Second World Wars)—these three parameters of ‘long duration’ justify
our comparison of different forms of state violence over the period of a century.
Nonetheless, the extraordinary magnitude of the practices of mass deportation and
ethnic cleansing as a way of managing populations and territories (not to mention
genocidal policies making use of famines) are particularly distinctive of the Soviet
experience between 1917 and 1953. Between these two dates, more than six million
people were deported, of whom a million and a half (most of them children and
elderly persons) died premature deaths; six million more died as a result of famines
from 1931 to 1933—famines caused not by climate but by the predatory policies of
the Stalinist state. These figures must be compared with the three million people
388 nicolas werth
who were forced to leave their homelands or displaced as a result of policies related
to ethnic cleansing during the last fifty years of the Czarist regime.
This chapter addresses the main specificities of the great episodes of deportation
and ethnic cleansing in the later Russian Empire and in the Soviet Union, as well as
an immense event that remained completely hidden for more than half a century,
the ‘man-made famines’ of the early 1930s.1 It also incorporates discussion of the
applicability or otherwise of the word ‘genocide’ to the Ukrainian famine of 1932–3
and the deportation of the ‘punished peoples’ from 1941–4.
D E P O RTAT I O N S A N D E T H N I C C L E A N S I N G
I N T H E R U S S I A N E M P I R E (1860–1914)
................................................................................................................
It was during the final phase of the wars of conquest in the Caucasus (from the late
1850s to the early 1860s) that the Czarist regime resorted for the first time to a policy
of ethnic cleansing. The latter took place in the more general framework of a
‘populations policy’ that sought to catalogue and classify, according to their degree
of ‘reliability’ and ‘loyalty’ to the regime, the diverse peoples of the Russian
Empire.2 In 1858, Alexander II approved a plan proposed by Prince Bariatinskii,
Viceroy of the Caucasus, for ‘cleansing’ the western Caucasus of its ‘autochthonous
tribes’ and settling Russian colonists there. During an important military council
presided over by Prince Bariatinskii in late August 1860, it was clearly indicated that
‘only a policy of terror could force the tribes to leave their auls [mountain villages]
to make room for colonists.’3 Over five years, half a million people were forced to
leave under threat: their villages were burned and razed by the army, and civilians
massacred in large numbers. It is estimated that nearly two million Caucasians
were driven from their homes and lands between the end of the 1850s and the end
of the 1870s.4 At the same time, several hundred thousand colonizers (Russians,
Ukrainians, Cossacks) received various subsidies from the army or the state in
1 The expression ‘man-made famine’ is James Mace’s. See his pioneering article, ‘The Man-Made
Famine of 1932–1933: What Happened and Why’, in The Great Man-Made Famine in Ukraine, ed.
Ukrainian Weekly (Jersey City: Svoboda Press, 1983).
2 Peter Holquist, ‘To Count, to Extract, and to Exterminate: Population Statistics and Population
Politics in Late Imperial and Soviet Russia’, in Ronald Grigor Suny and Terry Martin (eds), A State of
Nations: Empire and Nation-Making in the Age of Lenin and Stalin (Oxford: Oxford University Press,
2001), 111–44. This view is challenged by Bob Geraci in A. Dirk Moses (ed.), Empire Colony Genocide
(New York: Berghahn, 2008).
3 Holquist, ‘To Count’, 117.
4 Ibid. 119.
exchange for settling on part of the territories that had been emptied of their
inhabitants. After the conquest of the Caucasus, Central Asia became a second area
for experimenting with this new ‘populations policy’.5 In the event of a major
European conflict involving Russia, a certain number of preventative measures
were planned (hostage-taking, massive arrests, and deportations) to dissuade any
revolt on the part of the native populations considered ‘unreliable’. ‘Muslim
natives’ were not the only targets. Among the ‘suspect populations’ that might
help the enemy and betray Russia were not only the ‘Caucasian peoples’ but also,
on the western fringes of the Empire, Jews and Russian subjects of German origin.
The First World War was to give military men an opportunity to put these theories
largely into practice.
F O RC E D D I S P L AC E M E N T S A N D ‘P R EV E N TAT I V E
D E P O RTAT I O N S ’ O F J EWS A N D G E R M A N S
D U R I N G WA R (1914–16)
................................................................................................................
During the First World War, the military authorities, which were all-powerful in
areas near the front, displaced or deported about a million persons solely because
of their nationality or ethnic origin. They were for the most part German or
Austro-Hungarian nationals,6 but they also included Russian subjects of German
or Jewish origin who were considered ‘unreliable elements’. Nonetheless, it is not
always easy to reconstitute the precise context of these forced displacements. They
were part of a vast exodus of civilian populations from areas near the front, a
movement that was partly spontaneous and partly brought about by the ‘scorched
earth’ policy carried out by the Russian army as it retreated.7 The expulsions and
displacements were only occasionally transformed into organized, systematic de-
portations. Moreover, analysis of the forced displacements and deportations brings
out the crucial role played in these operations by the military, whereas the
government remained divided and powerless to control the actions of army
commanders, and especially those of the head of the General Staff, General
Yanushkevich, who was the main instigator of this policy.
5 Holquist, ‘To Count’, 120.

6 This point will not be developed here. Cf. S. G. Nelipovic, ‘Nemeskuju pakost’ uvolit’bez
neznostej . . . Deportacii v Rossii 1914–1917’ (‘Getting Rid of the German Filth without
Hesitations . . . Deportations in Russia, 1914–1917’), Voenno-istoriceskii Zhurnal 1 (1997), 35–49.
7 A. N. Kurcev, ‘Bezency pervoi mirovoi voiny v Rossii, 1914–1917’ (‘Refugees in the First World War
in Russia, 1914–1917’), Voprosy Istorii 8 (1999), 98–112.
390 nicolas werth
The Russian army’s first expulsions of Jewish communities took place in Poland,
during the first weeks of the war. On 5 September 1914, the whole Jewish population
of the little town of Pulawy (pop. 3,000) was forced by troops to leave within
twenty-four hours and to find its own means of transportation. During the
following weeks, many similar events occurred: on 14 October, in Grodzin (Warsaw
province), the military authorities gave the town’s 4,000 Jews three hours to clear
out.8 The commander justified these ‘preventative’ expulsions, which were often
accompanied by violence perpetrated by the troops, by pointing out that the Jews,
who spoke Yiddish, a language ‘close to German’, were just so many potential spies
and traitors.9 Starting in late 1914, expulsions of Jews, aggravated by the practice of
taking hostages chosen among the local notables, took on a massive character in
Austrian Galicia, which was temporarily occupied by Russian troops after a suc-
cessful military campaign.10 With the setbacks and then the rout of the Russian
army during the summer of 1915, the expulsions of Jewish communities in areas
near the front increased and were transformed in some places into genuine
organized deportations using special rail convoys. Thus in a few weeks (May–
June 1915), ninety-eight per cent of the 30,000 Jews living in Curland were deported
to the Ukrainian province of Ekaterinoslav.11 The expulsions and deportations of
Jews culminated in September and October 1915: during these two months, the
military forced the departure of several hundred thousand Jews from the province
of Minsk, as well as the whole of the Jewish community of Pskov.12 These massive
deportations were often accompanied by violence indeed, veritable pogroms,
which aroused the reprobation of the Prime Minister, Goremykin. The discussions
among high civilian and military officials concerning the ‘Jewish question’ re-
vealed, in fact, two different forms of discrimination with regard to the Jewish
community: among conservative bureaucrats, there was a traditional anti-Judaism
that sought to confine Jews to precisely defined regions; among the military men,
there was an anti-Semitism that had more to do with the effort to identify
populations considered ‘suspect’ and to move them away from zones of military
operations.
Among the other ‘suspect populations’ displaced by force were Russian subjects
of German origin, notably German ‘colonists’, farmers who had long before settled
in the Polish provinces of the Empire and in western Ukraine (the provinces of
Volhynia, Podolia, Kherson, and Ekaterinoslav). On 5 January 1915, the Army Chief
8 Peter Gatrell, A Whole Empire Walking. Refugees in Russia during World War I (Bloomington:
Indiana University Press, 1999), 18–22.
9 Quoted by Eric Lohr, Enemy Alien Politics within the Russian Empire during World War I, PhD
dissertation, Harvard University, 1999, 123.
10 The new military governor of Lvov, Count Bobrinski, was a notorious anti-Semite and played a
major role in this systematic policy of expulsions, which were accompanied by numerous pogroms.
11 Lohr, Enemy, 135.
12 Gatrell, A Whole Empire Walking, 23–8.
of Staff, General Yanushkevich, ordered the expulsion of not just the colonists, but
all Russian subjects of German origin living in the Polish provinces. Nonetheless,
the implementation of this plan was hampered by the rapid advance of the German
army in Poland, starting in April and May 1915. On the other hand, in the regions
remaining under the control of the Russian army, about 400,000 German colonists
in the provinces of Volhynia, Podolia, and Kiev were expelled and expropriated
between the summer of 1915 and the beginning of 1917.13
In order to assess these repressive actions, we have to recall that they took place
in a context of total war, in regions near the front where the boundaries between
the civilian and military spheres were largely erased, and were connected with a
massive flight of civilian populations who had been driven out of their homes by
the scorched earth policy followed by the retreating Russian army starting in 1915.
About five million civilians joined the exodus.14 Moreover, within the confines of
the Russian Empire, the Central Powers conducted an identical policy of expelling
and deporting populations considered ‘suspect’ or ‘hostile’: thus the German Ober
Ost administration ordered the departure of hundreds of thousands of ‘Slavs’ from
areas conquered in Poland and the Baltic countries; similarly, Austro-Hungarian
authorities drove out of reconquered Galicia large parts of the Ruthenian popula-
tion, which was thought favourable to the Russians.15
P U N I T I V E D E P O RTAT I O N S D U R I N G T H E C I V I L
WA R S : T H E E M B L E M AT I C E P I S O D E O F
‘D E -C O S S AC K I Z AT I O N ’ (1919–20)
................................................................................................................
The end of the ‘imperialist war’, made official by the Treaty of Brest-Litovsk in
March 1918, went almost unnoticed by Russian society in the midst of revolution,
the armed conflict being continued along shifting fronts via a civil war that had
already begun. At that time, the boundaries between the front and the rear areas,
between the civilian and the military spheres, had disappeared.16 The civil war
that began in the spring of 1918 was far more than a military conflict between
two structured camps, that of the Revolution (the ‘Reds’) and that of the
13 Lohr, Enemy, 140.

14 Gatrell, A Whole Empire Walking, 145.
15 Holquist, ‘To Count’, 124–6.
16 On these processes, cf. Nicolas Werth, ‘Les déserteurs en Russie: violence de guerre, violence
révolutionnaire et violence paysanne (1916–1921)’, in Stéphane Audoin-Rouzeau et al. (eds), La
violence de guerre, 1914–1945 (Bruxelles-Paris: Complexe/IHTP, 2002), 99–117.
392 nicolas werth
Counter-Revolution (the ‘Whites’). It was in reality an extraordinarily complex

overlapping of various forms of conflict. If none of the camps involved had a
monopoly on violence, only the Bolshevik revolutionary project, which was based
on a theory of civil war, indeed, a ‘scientific’ vision of society and the cult of the
strong state, provided the coherence necessary for conceiving certain planned opera-
tions of social engineering such as ‘de-Cossackization’—an unprecedented attempt
to eliminate, by means of ‘mass terror’ and deportation,17 a social group that had a
specific organization and its own status. This is not the place to recount the history of
this project,18 but rather to highlight a few striking characteristics that make de-
Cossackization an important stage in the history of the practices of deportation.
Some of the Cossacks (particularly the Cossacks of Kuban) who had been
stripped in December 1917 of the advantageous status they had enjoyed under the
old regime,19 and whom the Bolsheviks had classed as ‘kulaks’ and ‘class enemies’,
joined the White forces that had been assembled in southern Russia during the
spring of 1918. During the Red Army’s advance towards Ukraine and southern
Russia in early 1919, punitive detachments of special Cheka troops exterminated
tens of thousands of Cossacks in the Don region. These operations were not
measures of military retaliation taken in the heat of combat; they were a response
to a political directive issued at the highest level of the Party (the Orgburo of the
Central Committee) on 24 January 1919. This secret directive gave the order to
‘implement mass terror against the rich Cossacks who will have to be exterminated
and physically liquidated down to the last one, and in general, against all Cossacks
who have participated directly or indirectly in the struggle against Soviet power’.20
Confronted by this orgy of violence, the Cossacks along the Don rebelled.
Operating behind the Red Army, which was fighting General Denikin’s troops
farther south, the insurgent Cossacks made a major contribution to the striking
successes of the White armies during the summer of 1919. To weaken the Cossacks,
Iosif Reingold, the chairman of the Revolutionary Committee of the Don,21
proposed playing on the antagonism between the Cossacks and their non-Cossack
neighbours and implementing, along with ‘Red Terror’, a policy of ‘rearranging’
Cossack territories and deportation. This policy was experimented with towards
the end of 1920, after the final defeat of the White armed forces. In October and
November 1920, five Cossack stanitsy (large towns) in the Terek region (North
17 The expression ‘mass terror’ is used in the secret directive issued by the Orgburo of the Central
Committee of the Bolshevik Party, 24 January 1919. Cf: Izvestia TsK KPSS 6 (1989), 177–8.
18 On de-Cossackization, cf: Peter Holquist, ‘Conduct merciless mass terror: decossackization on the
Don, 1919’, Cahiers du monde russe 28:1–2 (1997), 127–62; V. L. Genis, ‘Raskazacivanie v Sovetskoi Rossii’
(‘The de-Cossackization in Soviet Russia’), Voprosy Istorii 1 (1994), 42–56; N. F. Bugaı̈ and A. M. Gonov,
Kavkaz: narody v eselonax (The Caucasus: Deported Peoples) (Moscow: Insan, 1998), 81–96.
19 Having the privilege to bear arms, to self-organize, and receiving a considerable quantity of land.
20 The text of this circular can be found in Izvestia TsK KPSS 6 (1989), 177–8.
21 RGASPI, 5/2/106/7.
Caucasus) were entirely emptied of their inhabitants (9,000 families, or about

45,000 persons). All the men between the ages of eighteen and fifty were deported
to the Donets Basin, where they were forced to work in the mines. As for the
women, the elderly, and the children, they were expelled ‘with authorization to
settle in other towns further north’.22 All the deportees’ livestock was confiscated.
Their houses and lands were redistributed to non-Cossacks, and especially to the
neighbouring Chechens, ‘who have always shown their deep attachment to
the Soviet government’, the responsible military officer noted.23 In order to make
the irreversible character of this deportation clear, the names of the Cossack stanitsy
were changed. Finally, to make an example, the stanitsa Kalinovskaia, considered the
most ‘hostile’ to Soviet power, was completely razed.
It will be noted, however, that given the size of the country and the extreme
violence of the civil war, these punitive deportations remained relatively limited.
Despite its successes, in 1920 the Bolshevik regime hesitated to engage in another
test of strength with the Cossack community which had been weakened but was
still ready to defend itself by force of arms. Moreover, as is shown by the field
reports submitted by the head of the deportation operations in October and
November 1920, the logistics were notoriously inadequate for carrying out large-
scale deportations.24 De-Cossackization remained a large project that was never
fully carried out.
T H E ‘L I Q U I DAT I O N O F T H E K U L A K S
A S A C L A S S ’ (1930–2)
................................................................................................................
‘De-kulakization’ or the ‘liquidation of the kulaks as a class’ was part of the
Bolshevik government’s offensive against part of the peasantry, which began during
the civil war and was momentarily interrupted by the NEP ‘truce’ declared by the
regime in 1921 when it was confronted by the magnitude of peasant resistance.
However, the context was different: whereas in 1920–1 large segments of the
peasantry were actively resisting Bolshevik policy, in 1929–30 it was a peasant
society that had been pacified that was the target of the Stalinist Revolution
‘from above’. To justify his attack, Stalin referred to the ‘threat’ that ‘kulaks’
posed to the very survival of the Soviet regime, which was being strangled by the
22 The rest of the group was authorized to ‘settle’ in villages at least fifty kilometres away.
23 Order issued on 23 October 1920 by General Medvedev, the officer responsible for the operation
(RGASPI, 85/ 11/131/ 11).
24 The slowness of the operations of deportations is explained by the lack of railway cars.
394 nicolas werth
kulaks’ deliberate refusal to sell their wheat to the state. Stalin and many Party
leaders were still traumatized by the threat of starvation experienced by many town
dwellers during the civil war and wanted to ensure that never again would such a
possibility recur.
The de-kulakization campaign begun towards the end of 1929 had in reality a
twofold objective: to ‘extract’ (the term used in official directives) the elements
likely to resist the collectivization of the countryside being launched at the same
time, and to ‘colonize’ the vast, inhospitable areas of Siberia, the Great North, the
Urals, and Kazakhstan. The first objective corresponded to the view, clearly ex-
pressed by the Bolsheviks when they took power, that peasant society contained
‘exploitative elements’ that were irremediably hostile to the regime and that had to
be ‘liquidated’. The second objective was, in a more contextual manner and in
connection with the launching of the first Five Year Plan, part of a vast project of
using labour provided by prisoners or deportees to develop the country’s inhospi-
table areas, which were rich in natural resources.25 In this context, expropriation
and deportation were chosen as the principal modalities for liquidating the kulaks
as a class. In three years (1930–2), more than five million kulaks were expropriated,
more than two and a half million men, women, and children deported, and several
hundred thousand arrested and sent to camps. At the end of January 1930, each
region received, from the Party’s highest political authority, the Politburo, ‘quotas
for de-kulakization in the first and second categories’. The ‘first-category kulaks’,
defined as ‘activists engaged in counter-revolutionary actions’, were to be arrested
and transferred to camps after a ‘rapid appearance before a troika’ (a special court
under the jurisdiction of the political police), or even ‘shot, in the case of the most
hardened among them, who are likely to resist’. The ‘second-category kulaks’,
defined as ‘rich peasants, but less actively engaged in counter-revolutionary activ-
ities’, were to be arrested, expropriated, and deported, along with their families, to
remote regions of the country.
To carry out these arrests and deportations, a veritable military logistics, un-
precedented in peacetime and mobilizing hundreds of rail convoys and tens of
thousands of special troops provided by the OGPU, was set up. However, the vast
scale of this project led to immense problems in coordinating the militarized
operations of deportation carried out by the OGPU and the settlement of the
deportees, which was left to the initiative of local authorities who were over-
whelmed or indifferent. Thus ‘deportation-colonization’ often became ‘deporta-
tion-abandonment’: hundreds of thousands of deportees were left to their fate on
the steppes or in the taiga, along railway tracks, without regular food supplies or
25 On de-kulakization, cf. in particular the collections of documents published under the direction
of V. P. Danilov, Tragedia sovetskoı̈ derevni (The Tragedy of the Soviet Countryside), vol iii: 1930–1933
(Moscow: Rosspen, 2002); idem, Sovetskaia derevnia glazami VCK-OGPU-NKVD (The Soviet
Countryside seen by the Cheka-OGPU-NKVD), vol. iii: 1930–1931 (Moscow: Rosspen, 2003).
work. Epidemics and shortages (and even, in some cases, famine) decimated the
deportees, first of all the children and the elderly. About fifteen per cent of the
deportees died during the first year. In the following years, the deportees’ condi-
tions improved very little. They were given a ‘special status’ that made them
veritable social pariahs, ‘special displaced persons’ (their official name); were
stripped of their civil rights; were forbidden to leave ‘special villages’ that were
generally isolated from everything in the most remote regions of Siberia, the Great
North, or Kazakhstan; and were subjected to a veritable forced labour in agricul-
tural or industrial structures controlled by the OGPU. The archives of the Depart-
ment of Special Populations attached to the Gulag mention very heavy losses
among the deportees: in 1930–1, more than 500,000 had disappeared, died, or
escaped out of the two million peasants deported; in 1932, more than 200,000
escapees, nearly 100,000 deaths (or an annual mortality rate of eight per cent); in
1933, more than 200,000 escapees and more than 150,000 deaths (or a mortality rate
of fourteen per cent)! Let us add that children, and especially infants, constituted
the largest portion of the deaths. In 1933, for example, the annual rate of infant
mortality in the special villages for deportees in Western Siberia reached sixty per
cent!26 These few figures give some idea of this great operation of social engineering
that was supposed to eliminate a social class, and beyond that, to subject the
peasantry as a whole to an unprecedented expropriation.
T H E G R E AT F A M I N E S OF 1931–3
................................................................................................................
Between 1931 and 1933, more than six million people, the great majority of them
peasants and livestock-raisers, died of hunger in the Soviet Union as a result of
terrible famines. In 1931, the first famines struck Kazakhstan, the most important
livestock-raising area in the USSR, where the collectivization of the herds resulted
in a decline of eighty-five per cent in the number of animals in only three years.27
Between 1930 and 1933, the population of Kazakhstan, estimated at six and half
million at the end of the 1920s, dropped by two million (600,000 fled; 1,400,000
died of hunger or in epidemics). That is, the population decreased by thirty per
cent—a proportion unequalled elsewhere, in any of the regions of the USSR struck
by the shortages and famines following the forced collectivization of the country-
side. Beginning in spring 1932, famine resulting from excessive State procurements
26 Juri Poliakov (ed.), Naselenie Rossii v XX veke (The Population of Russia in the Twentieth
Century), vol i (Moscow: Rosspen, 2000), 278–80.
27 On the famine in Kazakhstan, cf. Isabelle Ohayon, Du nomadisme au socialisme. Sédentarisation,
collectivisation et acculturation des Kazakhs en URSS, 1928–1945 (Paris: Maisonneuve, 2006).
396 nicolas werth
reached not only Western Siberia28 and the regions of the lower and middle Volga
(300,000 to 400,000 deaths in one year), but also the rich plains of Kuban, in the
North Caucasus, which were administratively connected to Russia, but populated
mainly by Ukrainians and Cossacks (400,000 to 500,000 deaths), and the Ukraine
(3.5 million to 3.8 million deaths).
In 2006, the Parliament of the Republic of Ukraine officially recognized that the
famine of 1932–3 was a genocide perpetrated by Stalin’s regime against the Ukrai-
nian people. The term used today in Ukraine to designate this famine, Holodomor,
is explicit: it combines the words holod (hunger) and moryty (to kill by privations,
to starve, to exhaust). Thus it clearly emphasizes the intentional aspect of the
phenomenon. The description of the 1932–3 famine as a genocide is not, however,
universally accepted among historians who have studied the question, whether
they be Russians, Ukrainians, or Westerners. We can distinguish two main inter-
pretive trends. In one camp are historians who see the famine as having been
artificially organized by Stalin’s regime in 1930 to break the back of the Ukrainian
peasants’ resistance, which was particularly strong, to the kolkhoz system and in
addition, to destroy the Ukrainian nation in its ‘peasant-national’ specificity, which
constituted a serious obstacle to the transformation of the Soviet Union into an
imperial state of a new kind, dominated by Russia. These historians adhere to the
genocide thesis. In the other camp are historians who, while recognizing the
criminal nature of Stalin’s policies, consider it necessary to study all the famines
of the years 1930–3 (in Kazakhstan, Ukraine, part of Western Siberia, and the Volga
regions) as a complex phenomenon in which several factors, ranging from the
geopolitical situation to the imperatives of industrialization, played an important
role alongside Stalin’s ‘imperial intentions’.29 For these historians, the term ‘geno-
cide’ is not appropriate to describe the famine of 1932–3 in Ukraine and Kuban.
A third position is emerging. As Andrea Graziosi rightly suggests, in light of the
most recent research, it is possible to propose a way of moving beyond the two
main existing lines of interpretation. According to Graziosi, who draws particularly
on the work of the American historian Terry Martin30 as well as on that of the
Ukrainian historians Juri Shapoval and Vassili Vassiliev,31 from the late summer of
1932 on, the Ukrainian famine is characterized by a high degree of specificity with
regard to Soviet famines as a whole during the period from 1931 to 1933. The latter
(including the terrible Kazakh famine) appear to be direct but unforeseen,
28 On the famine in Western Siberia, cf. M. K. Malyseva and V. S. Poznansii, Kazaki-bezentsy ot

goloda v Zapadnoi Sibiri, 1931–1934 (Almaty: Iz. Terra, 1999).
29 On the development of historical writing on the Ukrainian famine, Andrea Graziosi, ‘Les
famines soviétiques de 1931–1933 et le Holodomor ukrainien. Une nouvelle interprétation est-elle
possible et quelles en seraient les conséquences?’, Cahiers du monde russe 46:3 (2005), 453–72.
30 Terry Martin, The Affirmative Action Empire. Nations and Nationalism in the Soviet Union, 1923–
1939 (Ithaca, NY/London: Cornell University Press, 2001).
31 Yuri Shapoval and Valeri Vasiliev, Komandiri velikogo golodu (Kiiv: Geneza, 2001).
unprogrammed, unintentional consequences of policies inspired by ideology

that had been implemented since the end of 1929: forced collectivization,
de-kulakization, sedentarization of nomads and semi-nomads (in Kazakhstan),
and the imposition of the kolkhoz system, which the peasants saw as a ‘second
form of serfdom’, arbitrary or totally disproportionate levies on their harvests of
cereals and on their herds.
Until the summer of 1932, the Ukrainian famine, which was already underway,
was related to other famines that began earlier. But starting in August and Septem-
ber 1932, the Ukrainian famine changed in nature when Stalin developed his
‘national interpretation of the famine’ (Terry Martin). As is shown by the recently
published correspondence between Stalin and his main aides during the summer of
1932,32 Stalin then persuaded himself that a vast resistance front ranging from
simple kolkhozians to Ukrainian communist leaders had been constituted in
Ukraine in order to refuse to deliver to the state the agricultural products necessary
to supply food for the cities and for export. In a letter sent to Lazar Kaganovitch on
11 August 1932, Stalin went so far as to write: ‘We are in danger of losing the
Ukraine—“the party, the state, and even the organs of the political police and the
republic” are infested with nationalist agents and Polish spies.’33 Henceforth, Stalin
had made up his mind to use the weapon of hunger, to aggravate the famine that
was beginning, to instrumentalize it, to intentionally amplify it in order to punish
the peasants who, according to him, were ‘carrying on a war of sabotage against the
Soviet government . . . , a war to the death’.34 His two closest collaborators, Vya-
cheslav Molotov and Lazar Kaganovitch, were sent to Ukraine and the North
Caucasus as ‘plenipotentiaries’ charged with ‘purging’ and bringing into line the
local communist organizations that were ‘dragging their feet’—and sometimes
even refusing to implement excessive plans for cereals deliveries. Reinforcements
composed of ‘activists’ were sent from Russia, and armed detachments of the
political police engaged in veritable punitive actions in the Ukrainian kolkhozes
in order to requisition all the available wheat, including seed grain for the next
harvest.35 If the peasants were hardest hit—by hunger, leading to the death, under
the most atrocious conditions, of millions of persons—repression also affected
32 R. W. Davies, O. Khlevniuk, et al. (eds), Stalin i Kaganovich. Perepiska, 1931–1936 (Moscow:

Rosspen, 2001).
33 Martin, The Affirmative Action Empire.
34 Ibid. 273–4.
35 In this respect, a document recently exhumed from the archives seems to be fundamental. This is
a secret directive issued by Stalin on 22 January 1933, ordering ‘an immediate end to the massive
exodus of peasants fleeing Ukraine and Kuban on the pretext that they are looking for bread. The
Central Committee and the Council of the People’s Commissars’, Stalin wrote, ‘have proof that this
exodus from Ukraine is being organized by enemies of the Soviet government, socialist-
revolutionaries and Polish agents, for the purposes of propaganda intended to discredit, by means
of the peasants fleeing toward regions of the USSR north of Ukraine, the kolkhoz system in particular
and the Soviet system in general’ (RGASPI, 558/11/45/109).
398 nicolas werth
local officials and Ukrainian intellectuals who were arrested and imprisoned. In
December 1932, two secret decrees issued by the Poliburo put an end—in Ukraine
and in Ukraine only—to the policy of ‘indigenizing’ managers that had been
pursued since 1923 in all the federal republics; Ukrainian ‘nationalism’ was firmly
condemned.
Recent research undoubtedly shows the strong specificity of the Ukrainian case,
at least from the second half of 1932 on. On the basis of these new elements, how
should we describe the whole of the measures Stalin’s regime took to punish by
hunger and terror the Ukrainian peasantry as an ethnically defined group, mea-
sures whose result was the death of more than four million people in Ukraine and
the North Caucasus? The crucial point—so far as the use of the term genocide is
concerned—on which historians, including Ukrainian historians, differ is this: Did
Stalin target the peasants of Ukraine and Kuban qua peasants or qua Ukrainians?
For some historians, the main objective of the famine was to break down peasant
resistance, not national resistance. Other historians insist, on the contrary, that the
peasants of Ukraine and Kuban were targeted first of all as Ukrainians; for Stalin
the question of the Ukrainian peasantry was in fact ‘essentially a national question,
the peasantry constituting the main force behind the nationalist movement.’36
Breaking the Ukrainian peasantry amounted to breaking the most powerful na-
tionalist movement capable of opposing the process of constructing the Soviet
Union.37 This specifically anti-Ukrainian orientation, these historians conclude,
allows us to describe as genocide the whole of the political actions Stalin’s regime
undertook intentionally, starting at the end of the summer of 1932, to punish by
hunger and terror the Ukrainian peasantry, actions whose consequence was the
death of more than four million persons in Ukraine and the North Caucasus.
D E P O RTAT I O N S A S ‘C L E A N S I N G ’ A N D ‘M A K I N G
S E C U R E ’ T H E USSR’ S B O R D E R A R E A S (1935–9)
................................................................................................................
Beginning in 1935, the Soviet regime launched a number of deportations aimed at
‘cleansing’ the Soviet Union’s border areas (which were increasingly seen as verita-
ble front lines) of their ‘ethnically and socially suspect elements’. Since 1923,
36 On these divergent interpretations, cf. Roman Serbyn, ‘The Ukrainian Famine of 1932–1933 as
Genocide in the Light of the UN Convention of 1948 ’, Ukrainian Quarterly 62:2 (2006), 87–106.
37 According to the letter sent by Stalin to the writer Cholokhov on 6 May 1933, quoted in N. Werth,
‘Un Etat contre son peuple’, in S. Courtois and N. Werth et al. (eds), Le Livre Noir du Communisme
(Paris: R. Laffont, 1997), 237–8.
the border areas—twenty-two kilometres deep—had been put under a special

system of surveillance carried out by border guards connected with the political
police. Starting in 1934, this system was further strengthened. In February and
March 1935, several operations were begun to cleanse the borders considered most
strategic (the region around Leningrad, the Soviet–Polish border). In the Lenin-
grad region, an initial operation resulted in the deportation of some 3,500 families
of Finnish, Latvian, or Estonian origin, who were sent to Kazakhstan, Siberia, and
Tajikistan.38 At the same time, 8,300 families (41,650 individuals) were deported
from the border districts around Kiev and Vinnitsa to eastern Ukraine; citizens of
Polish and German origin represented more than half the deportees, the others
being classified as ‘socially alien elements’.39 In these first operations, which were
still limited and selective, the ethnic criterion was ‘blended’ with class considera-
tions, which were more accepted in communist political culture. Thus among the
Soviet citizens of Finnish origin deported from the Leningrad region were first of
all the ‘individual’ peasants who had not joined the kolkhozes, small craftsmen,
individuals deprived of the vote, and other representatives of socially stigmatized
groups. The cleansing operations in border areas were continued and expanded in
1936, encompassing 20,000 families from the Leningrad region and Western Uk-
raine.40 Officially, ‘administratively transferred’ contingents—to use the NKVD’s
bureaucratic terminology—were not assimilated stricto sensu to the ‘special dis-
placed persons’, expropriated and stripped of the civil rights. A few concessions
were made to ease their resettlement. Nevertheless, the sole fact that they were
under the administrative control of the office for ‘special populations of the Gulag’
and that they were required to stay put for an indefinite period, made their
condition very similar to that of the special displaced persons.
The obsession with cleansing the border areas to prevent ‘enemy agents’ from
infiltrating the country within the various diasporas (Polish, German, Finnish,
Baltic, Romanian, Korean) grew still stronger in 1937–8, the years of the Great
Terror. Several hundred thousand Soviet citizens of Polish, German, Baltic, Finn-
ish, or Romanian origin, along with many other citizens who had (or had earlier
had) any professional, familial, or simply geographic (inhabitants of border regions
were particularly vulnerable) connection with countries identified as hostile, were
arrested and executed.41 In this repressive context of unprecedented violence (let us
recall that in 1937–8, more than a million and a half individuals were arrested by the
NKVD, about half of whom were executed), the collective deportation of some
200,000 persons expelled from border areas seems almost anodyne. In September
38 Martin, Affirmative Action Empire, 333.

39 Ibid. 330.
40 GARF, 9479/1/36/12–16.
41 Cf. Nicolas Werth, ‘Repenser la Grande Terreur’, Le Débat 122 (November–December 2002),
118–39.
400 nicolas werth
and October 1937, the great programme of deportation struck the Korean commu-
nity that had settled in the border districts on the Soviet Far East (the regions of
Vladivostok, Khabarovsk, and Birobidzhan). In a secret resolution passed by the
Central Committee on 21 August 1937, this massive deportation was justified by
the fact that the Koreans constituted ‘a hotbed of spies and diversionists’ for the
Japanese secret services, which were said to be ‘particularly active since the occu-
pation of Manchuria by Japanese troops’. For the first time, the totality of a
national minority—no less than 172,000 individuals—was deported to Uzbekistan
or Kazakhstan.42 The Koreans were settled, under precarious conditions, in settle-
ment villages and kolkhozes specially reserved for them.
D E P O RTAT I O N S O F P O P U L AT I O N S TO
T E R R I TO R I E S A N N E X E D F O L LOW I N G
T H E G E R M A N –S OV I E T N O N -A G G R E S S I O N
PAC T (1940–1)
................................................................................................................
In the instructions he gave in September 1939 to NKVD officers charged with
Sovietizing the western Ukraine and Belarus, Vsevolod Merkulov, the head of State
Security, emphasized the historical importance of the operations about to be
undertaken: in a few months, the NKVD was going to complete a cleansing of
the country’s enemies that the USSR’s ‘organs’ had worked on for twenty years. All
the repressive measures that had been tried out since 1917 would have to be used.
However, the political, operational, and logistic expertise that had been acquired
over two decades made it possible to predict victory over the enemy.43
These instructions help us understand the exemplary shortcut represented by
the deportations carried out in the territories annexed by the USSR after the
German–Soviet Non-Aggression Pact. For the first time, the repressive practices
that had been widely tested on Soviet society were exported. Their brutality
profoundly traumatized Polish, Latvian, Lithuanian, and Estonian societies that
were subjected to Sovietization; but for the occupation authorities, the policies
implemented were only duplicating, in an almost routine way, measures already
being applied to Soviet citizens, without crossing any additional threshold of
violence.
42 Martin, Affirmative Action Empire, 333.

43 RGANI, 89/18/1/ 3–6.
In the wake of the mass killings committed by the NKVD two years earlier in the
context of the ‘secret repressive operations’ of the Great Terror (within sixteen
months, between 700,000 and 800,000 persons were executed in the USSR), about
25,000 Polish citizens belonging to Poland’s military, political, and economic elites
who had been incarcerated in three ‘special camps’ in Kozielsk, Ostachkov and
Starobielsk, were executed in April 1940 as ‘sworn enemies of the Soviet system’.44
In addition, between September 1939 and June 1941, about 110,000 persons were
arrested and sentenced by the NKVD’s special courts. A detailed analysis of these
sentences allows us to draw two main conclusions. If the Polish social and political
elites were particularly affected in the first months of the Sovietization of Poland,
during which the ‘revolutionary criterion’ (‘crushing the dominant classes of the
bourgeois state’) was pushed all the way, the Jews, who were suspect as ‘speculators’,
‘Bundists’, or ‘Zionists’, were also widely repressed (contrary to a notion long put
about, according to which a Jewish minority was supposed to have ‘profited’ from
the Soviet occupation), as were Ukrainians and Belorussians, when they were
classified as kulaks or ‘nationalists’.45 Moreover, a study of the distribution of the
punishments inflicted by the NKVD’s special courts and by military tribunals from
1939 to 1941 shows that sentences handed down were not more severe for foreign
nationals in an occupied country than for Soviet citizens arrested at the same
time.46
A third form of repression used in occupied Poland, the deportation of some
320,000 Poles (and 80,000 Balts and Moldavians in other territories annexed by the
USSR in 1940) was also carried out in accord with a schema that had already been
thoroughly tested over the preceding years. From February 1940 to June 1941, four
major deportation operations were organized, each targeting specific categories of
the population that had become suspect in the eyes of the authorities. The first
operation, implemented on 10 February 1940, was supposed to concern primarily a
particular group of ‘colonists and military foresters’ to whom the Polish govern-
ment had granted lands in border districts as a reward for their services during the
Soviet–Polish war of 1920. In reality, the 27,000 families (about 140,000 indivi-
duals) deported in February 1940 represented a much larger contingent of socially
alien elements (landowners, industrialists, government officials, policemen). The
sources used by A. Gurianov47 enable us to follow with great precision the
preparations for these deportations, the planning for the assignment of contingents
44 Among the most complete studies on this period, see R. G. Pikhoia, V. P. Kozlov et al. (eds),
Katyn (Moscow: Fond Demokratia, 1997).
45 O. A. Gorlanov and A. B. Roginskii, ‘Ob arestax v zapadnyx zonax Belorussii I Ukrainy v
1939–1941 g’, in A. E. Gurianov (ed.), Repressii protiv Poliakov I pol’skix grazdan (Moscow:
Zvenia, 1997), 77–113.
46 Ibid. 105.
47 A. Gurianov, ‘Polskie specpereselentsy v SSSR v 1940–1941’ (‘Polish Special Displaced Persons in
the USSR in 1940–1941’), in Repressii. 124.
402 nicolas werth
to hundreds of ‘colonization villages’ distributed over twenty-four regions from

Arkhangelsk to Irkutsk, and the setting up of logistics (convoys, police forces, lists
of persons to be deported) that were intended to avoid a new ‘deportation-
abandonment’ and to ‘rationalize the economic management of the contingents’.48
This rationalization, much vaunted in bureaucratic reports, should not lead us to
forget the suffering endured by the deportees, both during the journey and when
they arrived. From 1 March 1940 to 1 July 1941, the office of ‘special settlements’
registered, among the contingent deported in February 1940, 10,864 deaths, or 7.7
per cent of the total number of persons deported.49
A second operation provided for the deportation before 15 April of three new
categories that were very different but each bore a stigma on the scale of exclusion:
the stigma of class, for the families of Polish officers taken prisoner, the stigma of
being high officials in the state apparatus, the stigma of being large landowners or
industrialists who had been arrested (and on the point of being executed in the
three camps where they had been regrouped);50 for prostitutes, the stigma of being
‘social parasites’; for refugees from western Poland who had moved into the zone of
Soviet occupation and who had tried to return to the German zone but had been
turned back, the stigma of being migrants.51 This operation was implemented on 13
April; about 61,000 persons were deported. In contrast to the first contingent,
which had been assigned to colonization villages with the status of special displaced
persons, this second contingent, composed chiefly of women, the elderly, and
children, was scattered over hundreds of kolkhozes and other small factory
towns in Kazakhstan, which further increased the total isolation of these families
of outcasts, impoverished and knowing nothing about what had happened to their
loved ones. A third operation, implemented on 28 and 29 June 1940, targeted
mainly Polish refugees, most of them of Jewish origin, who had sought refuge in
the Soviet zone to escape Nazi persecution. These refugees were suspect in the eyes
of the Soviet occupation authorities for several reasons: because of their social
position, their vocations (many of them were merchants or professionals), their
political past (Bundists and Zionists were rather numerous among the Jewish
intelligentsia), and their refusal—which was frequent—to adopt Soviet nationality
and to conform to ‘Soviet legislation regarding passports’ (which forbade them,
among other things, residence in large cities). In the framework of this third great
operation of deportation, about 75,000 refugees (eighty-four per cent of which
48 Resolution of the Council of the People’s Commissars no. 2122–617 ss, 29 December 1939 (GARF,
5446/1/510/ 163–165).
49 Gurianov, ‘Polskie specpereselentsy’, 118–19; N. Bugaı̈, L.Beria-I.Stalinu, Soglasno vasemu
ukazaniju . . . (L. Beria to J. Stalin, ‘In accord with your instructions . . . ’) (Moscow: AIRO-XX, 1995),
12–13.
50 Werth, ‘Un Etat contre son peuple’, 234–5.
51 Gurianov, ‘Polskie specpereselency’, 116–17.
were Jews) were deported to various regions of northern Russian and Siberia, and
confined to colonization villages.52
Finally, between 22 May and 20 June 1941, the NKVD implemented a fourth great
operation of deportation that was particularly ambitious because it concerned not
only the territories of Ukraine and Belarus that had been attached to Poland
between 1920 and 1939, but also Moldavia (annexed from Romania in August
1940) and the three Baltic countries incorporated into the USSR in 1940. The
deportation plan ratified by the Politburo on 16 May 1941 defines ten categories
of individuals to be arrested, sent to a camp, or deported. Most of these categories
had already been used in the earlier operations; the places where the deportees were
to be assigned or interned varied depending on the categories defined. In all,
between 22 May and 20 June 1941, between 105,000 and 110,000 persons—including
about 45,000 Poles—were arrested. Of this number, about 20,000 were sent to a
camp, and 85,000 to 90,000 deported to Siberia (the regions of Novossibirsk, the
Altai, Omsk, and Krasnoiarsk) and Kazakhstan.53 When the last convoys of depor-
tees from western Belarus, which had been Polish territory, passed through Minsk
on 22 June 1941, the German invasion of the USSR had already begun.
T OTA L E T H N I C D E P O RTAT I O N S O F
‘P U N I S H E D P E O P L E S ’ D U R I N G
T H E ‘G R E AT P AT R I OT I C WA R ’
................................................................................................................
In the course of the Great Patriotic War, more than two million Soviet citizens
belonging to ethnic minorities accused of being a hotbed of potential agents for the
Nazi invader (Soviet citizens of German origin),54 or of having ‘collaborated with
the occupier’ (Chechens, Ingushes, Kalmyks, Balkars, Karachays, Crimean Tatars)
were deported and confined to special villages in one of the inhospitable regions of
the country, thus joining the vast cohort of special displaced persons. In compari-
son with the deportations analysed up to this point, those carried out between 1941
and 1944 have a number of specific characteristics, the most remarkable of which is
no doubt that they were aimed at destroying whole national minorities that had
been declared to be ‘enemies of the Soviet regime’. The goal was not to physically
eliminate all the individuals belonging to these national groups, but rather to make
52 See note 49.

53 A. Gurianov, ‘Masctaby deportatsii naselenija v glub’ SSSR, mai-ijun’ 1941’ (‘The Scope of the
Deportation of Populations to Remote Regions of the USSR in May–June 1941’), in Repressii, 149–60.
54 Werth, ‘Un Etat contre son peuple’, 241.
404 nicolas werth
the ‘punished’ nation disappear as such, to exclude it from the ‘great family of
Soviet socialist nationalities’. The three waves of deportation took place between
the summer of 1941 and the end of 1944. The first of these (September 1941 to March
1942) struck all Soviet citizens of German origin (more than a million persons), who
were deported to Kazakhstan or Siberia. Since Soviets of German origin were relative-
ly widely dispersed over a large area (about a third of them lived in the German
autonomous region of the Volga, but the other two-thirds were distributed over a
dozen regions), the NKVD, with the approval of the highest state authorities, began a
whole series of targeted deportations, region by region, prepared several weeks in
advance, and implemented on the basis of lists provided by various offices. In order to
ensure that the cleansing would be as complete as possible, the NKVD went so far as to
hunt up and arrest several tens of thousands of soldiers and officers of German origin
serving in the Soviet army. Some of the deportees, those most capable of working
were, in addition, put into the battalions of the ‘labor army’ in which living and
working conditions were similar to those in the camps of the Gulag.55
From November 1943 to June 1944, a second wave of deportations (about
900,000 persons) struck six non-Slavic peoples (Kalmyks, Karachays, Balkars,
Chechens, Ingush, Crimean Tatars) who were accused of ‘collective collaboration’
with the enemy during the brief period when their regions had been under partial
German occupation. This was the pretext under which the government sought to
‘definitively settle’56 the question of these areas that were little Sovietized and
poorly controlled, despite frequent police raids.
The third wave of deportations (July–November 1944) was more limited (it
affected about 200,000 persons) and was part of the policy, already largely im-
plemented during the second half of the 1930s, of ‘securitizing border zones’. The
deportees included notably the Bulgar, Greek, and Armenian communities along
the coast of the Black Sea, as well as Turkish–Mezkheti, Khemshin, and Kurd
minorities that had settled along the borders with Turkey and Iran.
In November 1948, a decree issued by the Presidium of the Supreme Soviet of the
USSR stipulated that the ‘peoples punished during the Great Patriotic War’ would
keep their defamatory status ‘in perpetuity’. This decision implied that every
member of the punished community transmitted, from one generation to the
next, the ‘collective crime’ committed by his ancestors. But can we deduce from
this that ‘elements of racial politics had surreptitiously slipped into Stalinist ethnic
cleansing’?57 An analysis of Stalinist policy regarding nationalities indicates that the
regime did not persecute this or that people because of alleged ‘biological defects’.
55 P. Polian, Ne po svoiei vole. Istoria i geografija prinuditel’nyx migratsii v SSSR (Moscow: OGI
Memorial, 2001), 114–15.
56 In the terms used by Bogdan Kobulov, Vice-Minister of the Interior, who was responsible for the
deportation of the Chechens and the Ingushes in February 1944, among other things.
57 Eric Weitz, ‘Racial Politics without the Concept of Race’, Slavic Review 61:1 (2002), 18.
Its goal was not to eliminate this or that race or ethnic group, but rather to
eradicate any form or manifestation of ethnic or national particularism that
might hamper the project of constructing a community of Soviet socialist nation-
alities and slow the realization of the communist utopia. Based on the conviction
that nationalities, like social classes, were socio-historical formations and not racial
or biological entities, the treatment inflicted on ‘punished peoples’ or ‘enemy
nations’ was more closely related to a form of ‘ethno-historical excision’.58 The
regime sought more to eradicate the national, cultural, and historical identities of a
community than to physically eliminate every one of its members. This probably
explains why a regime perfectly capable of implementing vast genocidal operations
did not set up extermination camps on the Nazi model.
Nevertheless, a few historians think that the deportations and ethnic cleansing of
the Soviet punished peoples, whose intentional nature is indubitably documented,
along with all the planning stages and the implementation, constitute genocides in
the sense defined by the UN Convention.59 They argue that in these operations the
Soviet regime was ‘deliberately inflicting on the group conditions of life calculated
to bring about its physical destruction in whole or in part’, recalling, for example,
that of the 600,000 persons deported from the Caucasus from late 1943 to early
1944, 147,000 (almost a quarter) were dead four years later (the highest death rate
being that among children); whereas only 28,000 births were registered!60
What these differing interpretations among historians show is both the weak-
nesses of the definition of genocide adopted in 1948—a definition at the same time
too restrictive and too general—and the porous nature of the borderline between
ethnic cleansing and genocide. As Norman Naimark has rightly noted, ‘ethnic
cleansing opens the way to genocide, mass crimes necessarily being involved in
emptying a land of its people.’61
A rapidly drawn up inventory of the deportations, ethnic cleansings, and
genocidal policies conducted by the Soviet regime is devastating. One observation
is inescapable: the distinction between peacetime and wartime no longer appears
pertinent when one is studying the specific form of mass violence constituted by
massive deportations of groups considered to be socially or ethnically ‘alien’ to the
national community. Over the years, deportations became a more or less perma-
nent way of managing not only these groups, but also territories. In this respect,
they differ radically from the forced displacements of populations implemented by
the Czarist regime during its wars of colonial conquest (in the Caucasus, and to a
lesser extent, in Central Asia) and from the preventative deportations organized by
58 See Francine Hirsch, ‘Race without the Practice of Racial Politics’, ibid. 40.
59 See, for example, J. Otto Pohl, Ethnic Cleansing in the USSR, 1937–1939 (Westport CT:
Greenwood, 1999).
60 On these figures, cf. Werth, ‘Un Etat contre son peuple’, 244.
61 Norman Naimark, Fires of Hatred: Ethnic Cleansing in Twentieth-Century Europe (Cambridge,
406 nicolas werth
Russian military authorities during the First World War. Every repressive project
whose genesis, degree of intentionality, preparation, execution, and consequences
we can now reconstruct on the basis of archival documents that have recently been
made available must naturally be situated in its specific historical context and
analysed in its continuum and its duration. The Soviet example has in any event one
great advantage: it shows the juridical frameworks and the definitions laboriously
worked out by political scientists and historians in order to ‘classify’ crimes
perpetrated by governments remain generally unsatisfactory when confronted by
the extraordinary inventiveness of regimes seeking to persecute people.
Bugai, Nikolai, The Deportation of Peoples in the Soviet Union (Commack, NY: Nova
Science, 1996).
Courtois, Stéphane, Nicolas Werth, et al., Le Livre noir du communisme (Paris: R. Laffont,
1997).
Gatrell, Peter, A Whole Empire Walking: Refugees in Russia during World War I (Blooming-
ton: Indiana University Press, 1999).
Graziosi, Andrea, ‘Les famines soviétiques de 1931–1933 et le Holodomor ukrainien. Une
nouvelle interprétation est-elle possible et quelles en seraient les conséquences?’, Cahiers
du monde russe 46:3 (2005), 453–72.
Holquist, Peter, ‘Conduct merciless mass terror: decossackization on the Don, 1919’, Cahiers
du monde russe 28:1–2 (1997), 127–62.
Mace, James, ‘The Man-Made Famine of 1932–1933: What Happened and Why’, in The Great
Man-Made Famine in Ukraine, ed. Ukrainian Weekly (Jersey City: Svoboda Press, 1983).
Martin, Terry, The Affirmative Action Empire. Nations and Nationalism in the Soviet Union,
1923–1939 (Ithaca, NY/London: Cornell University Press, 2001).
Polian, Pavel, Against Their Will: The History and Geography of Forced Migrations in the
USSR (Budapest: Central European University Press, 2004).
Serbyn, Roman, ‘The Ukrainian Famine of 1932–1933 as Genocide in the Light of the UN
Convention of 1948 ’, Ukrainian Quarterly 62:2 (2006), 87–106.
Suny, Ronald Grigor, and Terry Martin (eds), A State of Nations: Empire and Nation-Making
in the Age of Lenin and Stalin (Oxford: Oxford University Press, 2001).
chapter 20
.............................................................................................
T H E NA Z I E M P I R E
.............................................................................................
christopher r. browning
Genocide in the Nazi Empire issued from a confluence of traditions: anti-Semitism,

racism, imperialism, and eugenics. None of these was unique to Germany, but they
came together in a lethal combination in Germany under the Nazi dictatorship to
provide the ideological underpinnings for three clusters of genocidal projects. The first
was the ‘purification’ of the German race through the mass murder of the mentally and
physically handicapped within the Third Reich and the expulsion and mass murder of
‘Gypsies’ from the Third Reich. The second was a demographic revolution or ethnic
restructuring within the lands deemed to be Germany’s future Lebensraum through the
decimation, denationalization, and expulsion of the predominately Slavic populations
living there. The third was the systematic and total mass murder of every Jew—man,
woman, and child—caught within the German Machtbereich or ‘sphere of power’. It
was this genocidal project that the Nazi regime called the ‘Final Solution to the Jewish
Question’, and it is the total historical experience of the Nazi persecution of the Jews,
culminating in the ‘Final Solution’, that I refer to as the ‘Holocaust’.
I D E O LO G I C A L U N D E R P I N N I N G S
OF NA Z I G ENOCIDE
................................................................................................................
Anti-Semitism, with its origins in the Christian–Jewish adversarial relationship
dating back two millennia, was the oldest of these fateful traditions. What had
408 christopher r. browning
begun as one of many dissident or reform movements within Judaism during the
ministry of Jesus became a rival monotheistic sect with incompatible theological
claims concerning his identity following his death. This theologically driven con-
flict became immensely asymmetrical in the fourth century, when Christianity
became the official religion of the Roman Empire. Judaism—unlike various pagan
sects—was not outlawed and destroyed, but Jews were left in the precarious status
of a permitted albeit despised religious minority.
At the beginning of the second millennium, the position of the Jewish minority
in Western Europe became more precarious, as theological anti-Judaism trans-
formed into anti-Semitism. After centuries of contraction and decline, Western
Europe experienced an explosive growth and change in the eleventh–thirteenth
centuries. The underside of this first great European modernization was disorien-
tation and embitterment on the part of many. The Jews were not only the most
obvious symbol of religious anxiety and doubt in a new era of rational challenge to
religious orthodoxy but also became the symbol of usury and debt in a new
monetary, commercial economy, of clannishness amidst unsettling social change.
In short, the Jews became the subject of a broad, multifaceted negative stereotype.
Normal human individuality was denied them while negative traits were both
generalized to all Jews and predicated upon an essential inferiority, of which
stubborn religious unbelief was now just one among many indicators.1
This broad negative stereotype remained deeply embedded in European culture,
even though growing secularization diminished the centrality of religious differ-
ence. The ‘Dual Revolution’ of the nineteenth century—democracy and national-
ism on the one hand and urbanization and industrialization on the other—
brought legal equality, economic opportunity, and social mobility to the newly
emancipated Jews, but Europe’s second great modernization also inflamed its
‘social losers’ against its most visible beneficiaries. European anti-Semitism expe-
rienced a major revival, with Jews now portrayed in an updated negative stereotype
as exploitive capitalists, subversive revolutionaries, avant garde destroyers of tradi-
tional values and cultural standards, and aliens to the national community.
If the Dual Revolution had complex repercussions in European society, so did
the Scientific Revolution. Confidence in the capacity of human rationality to
master nature bred confidence that human behaviour could be understood with
similar scientific certainty, and hence explaining the past and prescribing solutions
to current political problems was possible through knowing the ‘laws of history’.
Marxism, based on class struggle as the law of history, was the most coherent
ideological product of this intellectual development. More diffuse but even more
widespread was the belief that race, not class, was the fundamental explanatory
category of human behaviour, and Social Darwinism—applying selective
1 Gavin Langmuir, History, Religion, and Antisemitism (Berkeley: University of California Press,
1990), 275–305.
the nazi empire 409
catchphrases and metaphors such as ‘struggle for survival’ and ‘survival of the
fittest’ to human relations—provided the appearance of scientific legitimacy to
racial theories of history.
European expansion and conquest (with especially lethal consequences for the
native populations of the western hemisphere) had been proceeding for centuries,
but reached its apogee in the nineteenth, due to Europe’s exponential growth in
power owing to the mobilizing capacities of the modern nation-state and the
economic and technological capacities of the Industrial Revolution. Europe’s easy
domination of other peoples and territories was accepted as the natural and
inevitable historical culmination of European superiority and native inferiority,
explained by race, rather than the result of a temporary and historically contingent
imbalance of power. A competitive nation-state system and a competitive capitalist
economy drove Europeans to maximize the exploitation of their imperial domains,
in which coercion and repression of the native populations were only partially
curtailed by notions of the ‘white man’s burden’ to bring the blessings of higher
civilization to those deemed inferior and primitive. Europeans became accustomed
to treating dominated populations abroad in ways no longer thinkable in Europe
itself, with perhaps German South West Africa and the Belgian Congo the most
extreme cases but still symptomatic of a much broader pattern of attitudes and
behaviour.
The temptation to enlist racial theorizing, seemingly legitimized by science and
validated by European expansion, in support of anti-Semitism was too great to
resist, and Wilhelm Marr articulated the conceptual marriage in 1873. Jews, he
argued, were not a group defined by religious belief or ethnic custom but rather by
immutable racial characteristics impervious to conversion or assimilation. The
accumulated traits of the centuries-old but now updated negative anti-Semitic
stereotype—formerly perceived as a mysterious affliction—were now explained as
the inescapable Jewish racial inheritance.
If racial anti-Semitism was one unfortunate fruit of enlisting seeming scientific
legitimacy in the service of politics, eugenics was another. Underlying the eugen-
ics movement was the belief in fundamental human inequality based on heredity
and thus the conviction that limiting the procreation of inferior people and
maximizing that of superior people would improve mankind. By the early
twentieth century advocates of eugenics were pushing for laws authorizing the
sterilization of individuals deemed to be carriers of hereditary defects, but several
potentially lethal ‘slippages’ in the rhetoric of eugenics were apparent. Some of
the allegedly hereditary defects were behaviours that frequently were conditioned
by both heredity and environment. And not just individuals but entire groups—
often categorized by class or race—were ranked as hereditarily inferior (on the
grounds of such subjective qualities as feeble-mindedness, shiftlessness, and lack
of sobriety) and were perceived collectively as threatening to cause society-wide
degeneration.
All of these traditions were present throughout Europe and North America by
the early twentieth century. But did they have a particular prominence or reso-
nance in Germany? The clearest case seems to be eugenics. Though the USA was
home to eugenics-driven laws authorizing the sterilization of individuals that
unequally impacted poor people, non-whites, and recent emigrants, in Germany
eugenics enjoyed widespread respectability in the universities and medical profes-
sion. Called ‘race hygiene’ by some, eugenics was seen not just as a defence of
society from the perceived threat of individual deficiencies but a defence of the
superior German, Nordic, or Aryan race. And it was in Germany that Binding and
Hoche went beyond the typical advocacy of sterilization and openly argued for laws
permitting the killing of those deemed ‘unworthy of life’ in order to relieve both
families and society of the ‘burden’ of caring for incurable and degenerate people
whose lives had no purpose or value.
The fate of the Herero and Nama in German South West Africa certainly likewise
suggests a particular genocidal continuity between Germany’s colonial practices in
its Wilhelmine empire in Africa and Nazi empire in Europe. Isabel Hull has made a
strong case that one element of continuity from genocide in South West Africa to
Germany’s fateful defeat in World War I was the institutional culture of the German
military obsessed with a doctrine of ‘absolute destruction’.2 But Hull does not argue
for a simple, straight line from South West Africa to Auschwitz. Indeed, the
example of South West Africa must be qualified in several ways. In quantitative
terms of loss of life, for instance, the Belgians in the Congo outdid the Germans in
South West Africa many fold. And South West Africa was not the sole template of
German colonial practices. In East Africa they waged brutal ‘wars of destruction’ to
repress native resistance, but the goal was unconditional surrender and total
submission, not genocide. Tribes that submitted were immediately enlisted and
armed as auxiliaries to combat those still resisting. In short, German colonial
practices in Africa were not distinct from those of other imperial powers.
Vejas Liulevicius has suggested an important link between German colonial
practices in Eastern Europe during World War I and subsequent Nazi practices
there. The German military established a model colony called Ober Ost (in terms of
current boundaries: northeast Poland, Lithuania, southern Latvia, and western
Belarus). They aimed to impose total control but also to carry out Kulturpolitik,
lifting up the native populations from filth, disease, and ignorance to become
disciplined, productive, and grateful subjects. Instead, hyperexploitation and de-
grading controls caused growing resentment, and the presence of a foreign con-
queror awakened national consciousness. Having hoped to manipulate and
cultivate ‘lands and peoples’, the Germans increasingly surrendered to a mindset
of managing ‘space and race’. Deprived of what they perceived as their rightfully
2 Isabel Hull, Absolute Destruction: Military Culture and the Practice of War in Imperial Germany
the nazi empire 411
earned conquest of the East, they longed to return to these hated lands and
despised populations but next time without any illusion of a cultural mission.
Indeed, the post-war depredations of the Freikorps foreshadowed the destructive
potential of rage over defeat, unlimited imperial appetite, and fanatical anti-
Bolshevism.3
Ironically, continuity between nineteenth-century anti-Semitism in Germany
and the Holocaust is more tenuous. Less than one per cent of the population,
highly assimilated German Jews had achieved a degree of prominence in economic,
professional, and cultural life, accompanied by a deep attachment to their country,
unequalled elsewhere in Europe. Unlike German Catholics and Socialists, they were
not subjected to outright discriminatory legislation, even if more subtle forms of
exclusion limited their role in the officer corps and civil service. Noisy, single-issue
anti-Semitic parties failed conspicuously. Anti-Semitism as a political platform was
co-opted by the Conservative Party in the 1890s, and—as Shulamit Volkov has
argued—became a ‘cultural code’ or political buzzword for all that conservatives
opposed: liberal democracy, socialism, unfettered capitalism, internationalism, and
cultural experimentation.4 But anti-Semitism among the traditional conservatives
did not obtain the intensity and priority that it would in the burgeoning German
Right of the 1920s. And the behaviour of German troops in Eastern Europe during
the First World War confirmed among the local Jewish populations Germany’s
image as the land of decency and opportunity for Europe’s Jews.
HITLER AND N AT I O NA L S O C I A L I S M
................................................................................................................
Between 1914 and 1933, Germans experienced a staggering accumulation of disas-
ters: a prolonged war of attrition and starvation blockade, unexpected military
defeat, revolution, a humiliating treaty settlement, hyperinflation, and finally the
unprecedented unemployment of the Great Depression. The tenuous democratic
majority that had emerged in the 1912 elections and established the Weimar
Republic in 1919 rapidly withered away thereafter. Neither an authoritarian tradi-
tional Right nor a revolutionary Left could fill the void in an increasingly frag-
mented and polarized German society. Longing for economic recovery and
prosperity, the restoration of great power status and national pride, the end of
political gridlock, and a return to the mythical social unity of the spirit of 1914,
3 Vejas Liulevicius, War Land on the Eastern Front: Culture, National Identity, and German
Occupation in World War I (New York: Cambridge University Press).
4 Shulamit Volkov, ‘Antisemitism as a Cultural Code’, Leo Baeck Institute Yearbook 23 (1978), 25–46.
Germans provided the National Socialists with a degree of broad cross-class

support unattainable by any other party and hence the decisive electoral plurality
that—in alliance with the traditional conservatives and key elites—brought Hitler
to power.
The Nazi dictatorship was certainly sensitive and catered to the longings and
illusions of German society, but how Hitler perceived reality now also mattered. As
German historians like Eberhard Jäckel established in the 1960s, Hitler was not
simply a power-hungry and opportunistic demagogue.5 He had a coherent ideo-
logical vision or Weltanschauung. Moreover, his was an ideological vision shared by
an ‘uninhibited generation’ of committed young men whose formative years were
the two decades of German disasters and who would provide the fanatical and
energetic leadership of key components of the Nazi regime.6 Also important, many
aspects of this ideological vision at least partially ‘overlapped’ with broadly held
sentiments in German society, providing a common vocabulary, the comfort of
seeming familiarity, and a shared emotional resonance between regime and society
on key issues.
Hitler’s ideological vision has been aptly summed up under the terms ‘race and
space’. For Hitler, history was the outcome of the natural and unfettered struggle
between races for the ‘living space’ or Lebensraum that underpinned each race’s
relative strength to compete and capacity to reproduce. In this struggle, the various
European races had emerged as the superior ‘culture-creating’ races of the world,
and the hardy German/Aryan race—shaped and winnowed by the harsh environ-
ment of northern Europe—had the potential to triumph over its rivals. However,
Germany’s deserved victory in World War I had been thwarted and now its very
viability and continued existence were threatened by the suffocating terms of the
Versailles Treaty and the debilitating effects of Weimar democracy.
German defeat and democratic revolution had a common source in an alleged
betrayal or ‘stab in the back’ by Jewish revolutionaries who subverted the German
war effort by undermining unity and continued willingness to sacrifice on the
home front. For Hitler, the Jews were not just another race in the natural competi-
tion for space, but an unnatural and subhuman element that could not hold its
own land and establish its own state and therefore could exist only as a parasite
living off other races. Jews destroyed their host races in two ways. First, they
destroyed the ‘purity’ of the host race by race-mixing. As ‘pure blood’ was
axiomatically assumed to be vital to the strength of a race, such race-mixing was
deemed fatal to a race’s capacity to wage the unceasing struggle for Lebensraum
5 Eberhard Jäckel, Hitler’s Weltanschuung: A Blueprint for Power (Middleton, CT: Wesleyan
6 Michael Wildt, Generation des Unbedingten. Das Führerkorps des Reichssicherheitshauptamtes
(Hamburg: Hamburger Edition, 2002); and Ulrich Herbert, Best. Biographische Studien über
Radikalismus, Weltanschauung und Vernuft, 1903–1989 (Bonn: Dietz, 1996).
the nazi empire 413
(witness the decline of Rome when its hardy peasant-soldier stock mixed with its
conquered subjects). Second, Jews were the carriers and disseminators of insidious
ideas that weakened a people’s will to wage a no-holds-barred struggle for survival.
In this regard, the subversive effects of three supposed Jewish conspiracies were
held to provide ample historical evidence: Christianity, with its pacifistic doctrine
of turn the other cheek and love thy neighbour; liberalism, with its belief in
fundamental human equality; and Marxism, with its proclamation of proletarian
solidarity above national loyalty.
If Hitler’s obsession with and hostility towards Jews was an implacable element
of his ideological outlook, and if the destruction of Weimar democracy and the
Versailles Treaty settlement were inherent preliminary goals, the territorial site of
Germany’s Lebensraum was theoretically open. Here the attraction of the East—as
the site of successful German expansion in World War I (framed as the continua-
tion and culmination of Germany’s historic Drang nach Osten), inhabited by
primitive Slavs undeserving of and incapable of developing such vast spaces,
the demographic centre of Jewish reproduction, and now the headquarters of
the ‘Jewish Bolshevik’ conspiracy for world revolution—proved irresistible. That
the Hitler regime would culminate in an attack on the Soviet Union—character-
ized as a military campaign for Lebensraum at the expense of Slavs, an ideological
crusade against Bolshevism, and a race war against the Jews—was the logical
implication of a shared ideological vision. The march towards this destiny in turn
spawned a series of genocidal projects that drew upon past legacies and were
consistent with Nazi ideology but which were nonetheless historically contingent.
They fell into three categories: those implemented within the Third Reich, those
implemented within German Lebensraum, and those implemented throughout
the German sphere of power.
G E N O C I DA L P RO J E C T S W I T H I N
THE THIRD REICH
................................................................................................................
Upon coming to power in January 1933, Hitler and the National Socialists set about
establishing their dictatorship and ‘coordinating’ German society with frightening
speed. In addition to crushing their political enemies and demolishing pluralism,
they also set out to purify the German people of those they deemed to be biologi-
cally defective or ‘degenerate’ members. Their reproduction would ‘dilute’ and
‘weaken’ the vitality, strength, and purity of the German racial community and
thereby undermine its capacity to wage relentless and unending struggle success-
fully, and their continued existence also would be an economic ‘burden’ that
drained resources from rather than contributed to that struggle. Hence, the regime
pursued quite literally the surgical exclusion of those deemed hereditarily defective
through compulsory sterilization, as decreed in the ‘Law for the Prevention of
Hereditarily Diseased Offspring’ in July 1933.7
However, those Germans considered biologically or hereditarily defective could not
be neatly and easily categorized in mass by legal definition. The categories of affliction
justifying compulsory sterilization included hereditary physical defects and illnesses
as well as nebulous mental and behavioural categories considered hereditary,
such as feeblemindedness, manic-depression, schizophrenia, and severe alcoholism.
‘Applications’ for compulsory sterilization could be made by doctors, institution
directors, and public health officials, and were adjudicated on an individual basis
by ‘hereditary health courts’, whose verdicts (routinely around ninety per cent in
favour of sterilization) were enforced by the police. Congenital feeblemindedness was
both the most imprecise diagnosis and most frequently invoked justification (roughly
fifty per cent followed by schizophrenia at twenty-five per cent). In the pre-war years,
over three hundred thousand Germans were sterilized.8
The transformation of the Volksgemeinschaft from the mythic inclusive national
community of August 1914 to the Nazi ideal of an exclusive biological-racial
community was intensified by a number of additional measures. A whole battery
of anti-natal and pro-natal measures supplemented the 1933 law. Moreover, com-
pulsory sterilization was fatefully expanded in 1937 to include strictly racial char-
acteristics. The offspring of German mothers and African fathers, who were among
the French army troops that took part in the postwar occupation in Germany, were
pejoratively referred to as the ‘Rhineland bastards’. As the oldest of these African-
German children approached maturity, the Nazi regime took action. In the sum-
mer of 1937, hundreds of these children were summarily sterilized.9
Between political opponents to the regime, who could be recovered for the racial
community through altering behaviour by punishment, coercion, and re-education
on the one hand, and the biologically defective who were barred from reproducing
through sterilization on the other, was a murky borderland inhabited by people
identified and stigmatized as ‘asocials’. These people were seen not only as aesthetic
blemishes on the Nazi image of the racial community, but also as stubborn non-
conformists who constituted a dissident threat to the Nazis’ capacity to impose
both uniform and productive behaviour.10 The tendency was always to subsume
7 Gisela Bock, Zwangssterilisation im Naizionalsozialismus: Studien zur Rassenpolitik und

Frauenpolitik (Opladen: Westdeutscher Verlag, 1986).
8 Henry Friedlander, The Origins of Nazi Genocide: From Euthansia to the Final Solution (Chapel
Hill: University of North Carolina Press, 1995).
9 Rainer Pommerin, ‘Sterilisierung der Rheinlandbastarde’: Das Schicksal einer farbigen deustchen
Minderheit 1918–1927 (Düsseldorf: Droste, 1979).
10 Klaus Scherer, ‘Asozial’ im Dritten Reich: Die vergessenen Verfolgen (Munich: VOTUM Verlag,
1990).
the nazi empire 415
asocial behaviour, when judged irremediable, within the biologically defective. Two
groups in particular within the Third Reich increasingly fell victim to this tendency
to treat asocial behaviour as a racial-biological threat: homosexuals and ‘Gypsies’
(Sinti and Roma). Homosexuals were perceived as offensive to public morality,
symbolic of the sexual licence of Weimar, subversive of Nazi notions of manly
camaraderie, and treasonously withholding their procreative powers from the
community. Over time, those deemed incorrigible homosexuals (the ‘seducers’
who were allegedly the real source of the problem) were subjected not only to
incarceration but also to sterilization, castration, or execution.11
Nowhere can the tendency of the Nazi regime to merge its categories of habitual
criminal, feebleminded, asocial, and racial alien be seen more clearly than in its
treatment of Sinti and Roma, pejoratively referred to as Zigeuner.12 The victims of
widespread prejudice and discrimination before 1933, the Sinti and Roma were
stereotypically characterized as parasitical, criminal, lazy, and rootless. After 1933 in
Germany they were disproportionately subjected to the Nazi regime’s measures
against asocial habitual criminals, vagrants, and beggars. In fact, one category of
asocial behaviour subject to ‘preventive detention’ was simply exhibiting a ‘Gypsy-
like’ lifestyle, even when those involved were not Gypsies. After 1935, Sinti and
Roma were increasingly confined to Gypsy camps. The Sinti and Roma were
likewise disproportionately subjected to compulsory sterilization on the grounds
of feeblemindedness. The Nürnberg Laws did not mention Zigeuner, but
subsequent commentaries declared them to be of ‘alien blood’ and subject to the
same prohibitions that affected Jews. Himmler in turn set up a Central Office for
the Fight against the Gypsy Nuisance and declared the ‘Gypsy problem’ to be a
‘matter of race’.
In the war years persecution, incarceration, and sterilization gave way to wide-
spread killing in the case of two groups whose very existence within the boundaries
of the Third Reich was now to come to an end: the mentally and physically
handicapped on the one hand and Sinti and Roma on the other. In the summer
of 1939, Hitler set in motion planning for the ‘euthanasia’ of both severely de-
formed newborns and institutionalized mentally ill adults deemed both incurable
and incapable of productive work. With the outbreak of war, institutionalized
Polish mental patients in the territories of the Polish corridor quickly annexed to
11 Geoffrey Giles, ‘The Institutionalization of Homosexual Panic in the Third Reich’, in Robert
Gallately and Nathan Stoltzfuss (eds), Social Outsiders in Nazi Germany (Princeton, NJ: Princeton
University Press, 2001), 233–55; and idem, ‘Männerbund mit Homo-Panik: Die Angst der Nazis vor der
Rolle der Erotik’, in Burkhard Jellonnek and Rüdiger Lautmann (eds), Nationalsozialistischer Terror
gegen Homosexuelle: Verdrängt und Ungesünht (Paderborn: Ferdinand Schönigh, 2002), 105–18.
12 Michael Zimmerman, Rassenutopie und Genozid: Die nationalsozialistische ‘Lösung der
Zigeunerfrage’ (Hamburg: Christians, 1996). Guenter Lewy, The Nazi Persecution of the Gypsies (New
York: Oxford University Press, 2000) contains much valuable information but argues for a very
restricted notion of genocide that excludes the Sinti and Roma as genocidal victims.
the Third Reich were killed by firing squad. This particular killing action inaugu-
rated in the fall of 1939 expanded to include Polish patients in the Warthegau as
well as German patients in Pomerania and East Prussia. It also undertook the initial
experiments in using carbon monoxide to poison rather than shoot its victims,
both in prototype gas chambers in Posen and in a prototype itinerant gas van.13
The nationwide programmes then got underway. Deformed newborns were sent to
special wards in select hospitals and killed by drug overdose. The centralized killing
programme of adults, designated as T 4, coordinated the inventory of patients in
institutions throughout Germany, after which the selected victims were sent to one
of six killing centres equipped with gas chambers utilizing bottled carbon mon-
oxide. This centralized programme was halted by Hitler in August 1941, when its
operations became too public and evoked too much unease and protest. Nonethe-
less, decentralized killing of targeted adults continued in selected hospitals, usually
through lethal injection or medication, to the end of the war. All Jewish patients
were killed without selection, either in the euthanasia centres in Germany or
following deportation to Poland. And finally, in a programme designated Opera-
tion 14f13, touring teams of T 4 doctors conducted selections in the concentration
camps, after which the victims were sent to one of the euthanasia killing centres.
Altogether, by the end of the Nazi regime, some 150,000 people had been subjected
to ‘medicalized’ killing in these various operations aimed at ridding Germany of
those deemed ‘unworthy of life’ due to mental or physical handicaps as well as
some 20,000 concentration camp prisoners who were deemed medically ‘unfit’.14
In contrast to Hitler’s direct instigation of and involvement in the killing of the
German handicapped, he exhibited virtually no interest in the fate of the Sinti and
Roma. The fact that the Nazi regime nonetheless carried out a genocidal assault
against them demonstrates how lethal can be the combination of pervasive popular
prejudice, institutionalized racism, empowered local authorities, and a bureaucrat-
ic police state that develops the habit of solving problems through repression and
mass murder. The Sinti and Roma were subjected to vacillating, contested, and
sometimes contradictory policies, which provided some loopholes for survival. But
ultimately more than two-thirds of the Sinti and Roma in the Third Reich and
Protectorate perished in Nazi camps, and many of those spared death were
sterilized and thus subjected to a form of ‘delayed genocide’. In 1938–9, some
3,000 Austrian Gypsies from the Burgenland were sent to German concentration
camps as part of several actions against the ‘work shy’ and asocials, and few
survived the high mortality rates of prolonged incarceration. Plans in the fall of
1939 to deport all Sinti and Roma, along with all Jews, into the General Govern-
ment failed, but in May 1940 2,800 were deported from western Germany (at the
13 Volker Riess, Die Anfänge der Vernichtung ‘lebensunwerten Lebens’ in dem Reichsgauen
Danzig-Westpreussen und Wartheland 1939/40 (Frankfurt: Peter Lang, 1995).
14 Friedlander, The Origins of Nazi Genocide.
the nazi empire 417
instigation of the military, who pronounced them a security risk) into the General
Government, from where few ever returned.15 In November 1941, 5,000 Gypsies
were deported from Austria to the Lodz ghetto and then killed in the gas vans of the
Chelmno death camp the following January. In February 1942 some 2,000 Gypsies
were expelled from East Prussia to Bialystok, some of whom were then expelled yet
further east to Brest the following fall. In both places they suffered high mortality
under terrible conditions. In December 1942, Himmler ordered the deportation of
more than 20,000 Sinti and Roma from the Third Reich and the Protectorate to
Auschwitz-Birkenau. Most perished from the terrible living conditions in the
‘Gypsy family camp’ there or after re-assignment to other camps. Three contin-
gents—1,700 from Bialystok in March 1943, 1,000 sick prisoners in May 1943, and
the final remnant of some 3,000 with the closing of the Gypsy family camp in
August 1944—were sent to the Birkenau gas chambers.16
The killing of the Sinti and Roma from the Third Reich was not total, however,
for at least 6,000 were exempted from the deportations Himmler ordered in
December 1942. One exempted category was ‘racially pure Gypsies’, reflecting
Himmler’s idiosyncratic views (contested by Martin Bormann among others)
that the Gypsies—in distinct contrast to Jews—were ‘Aryans’ whose worst traits
resulted from all-too-frequent race-mixing. Also exempted were Zigeuner-
mischlinge, who had been accepted by ‘pure’ groups, ‘socially adjusted Gypsies’
with regular jobs and residences, Gypsies married to Germans, and Gypsies either
in military service or working in war-important jobs. Eligibility for exemption was
decided by the local criminal police, and many Zigeunermischlinge were threatened
with loss of their exemption from deportation if they did not submit to steriliza-
tion. Such exceptions combined with the lack of a clear programme and premedi-
tated ‘intent’ to kill all Gypsies has led at least one scholar, Guenther Lewy, to argue
that Nazi mass killing in this case did not constitute genocide—a view not widely
shared by others.17
While Germans deemed ‘unworthy of life’ as well as the vast majority of Sinti
and Roma were no longer to exist within the boundaries of the Third Reich, similar
genocidal projects targeting these specific groups did not extend beyond those
boundaries. Institutionalized patients elsewhere in occupied Europe suffered griev-
ously and died in large numbers from lack of care and starvation. In places where
the German occupiers wanted the buildings they occupied or to subject them to
experiments (as in Belarus), they were massacred. But there was no plan or
programme to kill all institutionalized, handicapped people throughout German-
occupied Europe. Likewise, Gypsies beyond the boundaries of the Third Reich were
15 Zimmermann, Rassenutopie und Genozid, 172.

16 Ibid. Lewy, The Nazi Persecution of the Gypsies.
17 Lewy, The Nazi Persecution of the Gypsies, 221–4. In contrast, see: Zimmermann, Rassenutopie
und Genozid, 369–81.
often killed on local initiative but not as part of a continent-wide programme

mandated from Berlin. On Soviet territory itinerant or ‘wandering Gyspies’ were
killed in large numbers as alleged spies, partisans, and disease carriers, but ‘seden-
tary Gyspies’ were generally spared. However, in certain areas—such as Estonia18
and the Crimea—local German authorities killed all Gypsies. In Serbia, a distinc-
tion was made by gender and age. The German military was short of victims to fill
its self-imposed reprisal quotas and shot adult male Gypsies alongside adult male
Jews. But Gypsy women and children were released from the Semlin camp outside
Belgrade before the Jewish women and children interned there were gassed. The
same distinction of different treatment between victims within the Third Reich and
beyond can be made for the persecution of homosexuals. Himmler, who was
phobic about German homosexuals, had no desire to discourage or repress homo-
sexuality in the occupied territories, for he viewed non-procreative sex among
non-Germans as beneficial to Germany.
G E N O C I DA L P RO J E C T S W I T H I N
G E R M A N L E B E N S R AU M
................................................................................................................
As the German path to seizing Lebensraum in the Soviet Union led through other
East European countries of mixed populations, the Nazi regime had to negotiate an
unpredictable diplomatic and military course as well as experiment in the demo-
graphic reconstruction of the populations that came under its control. Hitler had
hoped to ally with Poland against the Soviet Union, but when no agreement was
reached with the Poles, he instead invaded and partitioned that country in cooper-
ation with the Soviet Union. The German-occupied half of partitioned Poland then
became the first real ‘laboratory’ for transforming non-German lands into German
Lebensraum through engineering a demographic revolution. The outlines were
sketched in late September and early October 1939. The German share of Poland
was to be split yet again in two, with the western regions ‘incorporated’ directly
into the Third Reich while the remainder was reduced to the status of a German
colony designated as the General Government. Heinrich Himmler was adamant
that ‘one only possesses a land when even the last inhabitant of this territory
belongs to his own people.’19 Thus making the ‘incorporated territories’ truly
part of the Third Reich was to be achieved by expelling all eight million
18 Anton Weiss-Wendt, ‘Extermination of the Gypsies in Estonia during World War II: Popular
Images and Official Policies’, Holocaust and Genocide Studies 17/1 (spring 2003), 31–61.
19 National Archives microfilm, T175/122/266598ff: Himmler Memorandum, 25 June 1940.
the nazi empire 419
non-German inhabitants into the General Government—a policy later dubbed

‘ethnic cleansing’ by Slobodan Milošević, but one that Hitler then referred to as
Flurbereinigung or ‘basic cleansing’. These lands, emptied of all but the ‘ethnic
Germans’ already living there, were to be repopulated by ‘calling back home to the
Reich’ all the ethnic Germans living in the lands of Eastern Europe ceded to Stalin
in the Nazi-Soviet Non-Aggression Pact (namely the Baltic states, eastern Poland,
and Bessarabia).20 The initial uprooting and dissolution of ethnic German com-
munities in much of Eastern Europe was not, therefore, the result of German defeat
and Red Army revenge at the end of the war, but rather the calculated policy of
Heinrich Himmler and his SS demographic engineers in 1939–41.
The genocidal implications for Poles under German occupation were twofold.
Not only were millions of Poles in the incorporated territories to be expelled into
the General Government, with total loss of property and regardless of loss of life,
but within the General Government the Poles were to be reduced to a denationa-
lized mass of slave labourers. Polish leadership classes and carriers of Polish
national identity and culture were targeted for elimination. The rest of the Polish
population was to be deprived of meaningful education, held to minimum food
rations, and subjected to forced labour either at home or, increasingly, in Germany.
The sheer logistical magnitude of the Nazi expulsion plans and the reality of
wartime economic needs for continued production prevented complete ethnic
cleansing from the incorporated territories, but the decimation of the Polish
intelligentsia and exploitation of Polish forced labour continued unfettered.21
Until the unforeseeably quick victory over France in the west, the time frame for
continued expansion of German Lebensraum in the east had to remain uncertain.
Thus the German occupiers were constructing a defensive wall along the Nazi-
Soviet demarcation line into the summer of 1940, after which German planning
shifted from the defensive to the offensive. When Hitler decided to abandon the
Non-Aggression Pact and invade the Soviet Union, he also exhorted and pro-
claimed that this campaign would take the form not of a conventional war bur
rather of a ‘war of destruction’. The once distant vision but now immediate
prospect of territorial conquest of vast Lebensraum in the east, an ideological
crusade against communism, and the ultimate confrontation with the demograph-
ic centre of world Jewry found resonance and produced feverish planning through-
out the Nazi regime.22 The military removed Soviet civilians from the protection of
martial law, mandated the execution of captured communist ‘commissars’ and the
20 Götz Aly, ‘Final Solution’: Nazi Population Policy and the Murder of the European Jews
(London: Arnold, 1999); and Phillip Rutherford, Prelude to the Final Solution: The Nazi Program for
Deporting Ethnic Poles, 1939–1941 (Lawrence: Kansas University Press, 2007).
21 Czeslaw Madajczyk, Die Okkupationspolitik Deutschlands in Polen 1939–1945 (Berlin: Akademie
Verlag, 1987).
22 Alex Kay, Exploitation, Resettlement, Mass Murder. Political and Economic Planning for German
Occupation Policy in the Soviet Union, 1940–1941 (New York: Berghahn Books, 2006).
infliction of collective reprisal against any manifestation of resistance behind the

lines, and arranged for military cooperation with and logistical support for mobile
SS killing squads or Einsatzgruppen. The economic planners envisaged a systematic
exploitation and looting of Soviet territory and accepted as self-evident that
‘umpteen million people will doubtless starve to death, if we extract everything
necessary for us from the country.’23 Meeting with his SS leaders on 12–15 June 1941,
Himmler confided that the coming conflict ‘will be a racial struggle of pitiless
severity, in the course of which 20 to 30 million Slavs and Jews will perish through
military action and crisis of food supply’.24 When victory seemed imminent in
mid-July 1941, Hitler proclaimed that the newly conquered territories would be
transformed into a German ‘Garden of Eden’. Over the next year Himmler’s
planners worked on various versions of a Generalplan Ost, which envisaged both
the settlement of Germans in the conquered Lebensraum and the reduction of the
native population from 45 to 14 million through death or expulsion into Siberia.25
The death—mostly from starvation and exposure—of more than two million
Soviet soldiers in German POW camps in the first nine months of the war,26 the
calculated starvation blockade of Leningrad, and the escalating death of Soviet
civilians from starvation and massacre behind German lines indicate clearly that
the even vaster decimation of the Soviet population envisaged by Nazi planners was
no idle speculation. Only German defeat precluded full realization of the genocidal
intentions of Generalplan Ost and even vaster loss of life.
THE GENOCIDE OF THE J EWS

................................................................................................................
Nazi persecution of the Jews was initially one component of the other two evolving
genocidal projects: first, the biological-racial purification within the Third Reich,
and second the massive population decimation and ethnic cleansing within Ger-
man Lebensraum. In 1941–2, however, the Final Solution gained an autonomy,
priority, and singularity apart from all other persecutory and genocidal policies
of the Nazi regime. Its goal was the total and systematic elimination of every last
23 Nürnberg Document 2718-PS: note on conference of state secretaries, 2 May 1941.

24 The testimony of Erich von dem Bach-Zelewski in Trials of the Major War Criminals before
the International Military Tribunal, iv.482–8, and in Justiz und NS-Verbrechen, xx.413.
25 Helmut Heiber (ed.), ‘Der Generalplan Ost’, Vierteljahrshefte für Zeitgeschichte 6 (1958), 281–325;
and Mechtild Rössler and Sabine Schleiermacher (eds), Der ‘Generalplan Ost’: Hauptlinien der
nationalsozialistischen Planungs- und Vernichtungspolitik (Berlin: Akademie Verlag, 1993).
26 Christian Streit, Keine Kameraden. DieWehrmacht und die sowietischen Kriegsgefangenen
1941–1945 (Stuttgart: DVA, 1978).
the nazi empire 421
Jew—man, woman, and child—within the Nazi sphere of power, and was therefore
a genocidal project that ultimately had no geographical limit or boundary other
than the reach of the German military.
During the pre-war years of the Nazi regime, the persecution of the Jews aimed
at excluding them from both the German Volksgemeinschaft and German soil. The
anti-Jewish legislation of 1933 constituted the ‘civic death’ of German Jews. It ended
Jewish emancipation by denying them equality before the law and expelling them
from the civil service as well as cultural and associational life and sharply curtailing
their activities in the professions. The Nürnberg Laws of 1935 completed the ‘social
death’ of German Jews, outlawing marriage and sexual relations between Jews and
‘Aryans’ and rendering any continuing social ties vulnerable to denunciation and
criminal prosecution. Numerous laws in 1938 resulted in the ‘economic death’ of
Germans Jews through completing the transfer (euphemistically dubbed ‘aryaniza-
tion’) and expropriation of Jewish property and eliminated the last free participa-
tion of Jews in the German economy—a process that had been steadily underway
since 1933. By the mid-1930s, SS specialists on Jewish policy had clearly articulated
the goal of this escalating persecution—to create a Third Reich ‘pure of Jews’
(judenrein). Life was to be made so intolerable that all German Jews would
voluntarily emigrate despite the near total loss of property entailed in leaving
Germany and the ever rising barriers to immigration elsewhere. When Germany
annexed Austria in March 1938, SS experts (Adolf Eichmann in particular) could
move from theory to practice, devising measures to compel the departure of Jews
from Austria at a rate far exceeding Jewish emigration from the Old Reich.
Despite the effort to create a German Third Reich pure of Jews through ever
intensifying persecution and expulsion, nearly half the German Jews remained
trapped in the Old Reich (as defined by 1937 boundaries) when war broke out in
1939. And the addition of Austrian and Czech Jews in 1938 and 1939 underlined the
paradoxical dilemma facing the Nazis, namely that every diplomatic and military
success that added territory was simultaneously a step backwards in achieving a
Third Reich pure of Jews. With the outbreak of war and the prospect of even greater
territorial gain (as well as ever greater numbers of Jews) imminent, Hitler sought to
cut the Gordian knot by signalling his followers through his January 1939 Reichstag
speech about his expectations. If the Jews caused another world war, he prophesied,
it would result in the destruction of the Jewish race in Europe.
Hitler’s followers, especially Heinrich Himmler, understood the signal and acted
accordingly. They included all Jews within the German sphere in three successive
plans for ethnic cleansing and population decimation. Thus, in the fall of 1939, all
Jews—beginning with the Jews of the incorporated territories but then encom-
passing all Jews of pre-war Germany—were to be expelled to a Lublin Reservation
in the furthest corner of the German empire. This territory was deemed especially
suitable since its marshy character would ensure a decimation of the Jewish
population. With victory over France and its overseas empire at German disposal
in June 1940, SS and Foreign Office experts devised the Madagascar Plan, in which
all Jews of the German sphere were to be sent to that French colony in the Indian
Ocean as soon as Britain dropped out of the war and its merchant marine could be
seized for that purpose. Finally, with the decision for Barbarossa, the Nazis planned
that following victory all European Jews would be expelled to ‘a territory yet to be
determined’—code language for the Siberian and Arctic wastelands of a defeated
Soviet Union. With each successive plan the scope and murderous implications
intensified, as did the frustrations of the planners when the first two proved
impractical in turn.
Preceding Barbarossa, Himmler had noted that 20 to 30 million ‘Jews and Slavs’
would perish. Military guidelines for troop behaviour called for ruthless measures
against ‘bolshevist agitators, guerrillas, saboteurs, Jews’, and military propaganda
identified the commissars and party functionaries targeted for summary execution
as ‘mostly filthy Jews’.27 Heydrich likewise targeted ‘Jews in party and state posi-
tions’ for summary execution. And the planners of mass starvation surely knew
from the experience of the Polish ghettos that when food was scarce, Jews were
always the first to starve. In short, genocide of the Soviet Jews, through some
unspecified combination of execution, expulsion, and starvation and according to
some unspecified timetable, was implied in Nazi plans for the war of destruction.
The implicit became explicit over the summer of 1941. Following Hitler’s proc-
lamation in mid-July that the conquered Soviet territories should be transformed
into a German Garden of Eden, Himmler reinforced the 3,000 men of the Einsatz-
gruppen with nearly 20,000 Order Police and Waffen-SS for behind-the-lines killing
operations. And following Himmler’s various tours of the eastern front to visit
such units, targeting shifted from the selective mass murder of Jewish leadership
and more broadly Jewish men of military age to the mass murder of all Jews but
especially women and children. By late August 1941 the systematic ‘liquidation’ of
Soviet Jewry was underway.
Concerning the fate of the European Jews west of the Soviet Union, the funda-
mental change in vision—from ‘solving the Jewish question’ through a combina-
tion of expulsion and decimation to solving it through systematic and total mass
murder—occurred in the fall of 1941. Between mid-September and mid-October,
three key decisions were taken. Himmler’s role in all three can be documented,
though a surviving paper trail leads to Hitler only in the first. These three decisions
were as follows: to begin deporting Jews from the Third Reich to intermediate stops
in the eastern ghettos of Lodz, Minsk, and Riga before sending them ‘yet further to
the east’ the following spring; to construct camps equipped with gassing facilities in
Belzec, Chelmno, and elsewhere; and to ban all further Jewish emigration since,
27 Helmut Krausnick and Hans-Heinrich Wilhelm, Die Truppe des Weltanschauungskrieges: Die
Einsatzgruppen des Sicherheitspolizei und des SD, 1938–1942 (Stuttgart: DVA, 1981), 125, 136.
the nazi empire 423
according to Heydrich, such Jews ‘would be too much out of the direct reach of the
measures for a basic solution to the Jewish question to be enacted after the war’.28
The Final Solution was to be implemented throughout the German sphere of
power, not just within German Lebensraum. Eichmann’s preparatory document for
Heydrich at the Wannsee Conference (20 January 1942) listed all 11 million Euro-
pean Jews, including those of Portugal, England, Ireland, Finland, and the Europe-
an areas of Turkey. But Europe was not the limit. As Hitler informed the Grand
Mufti of Jerusalem in a meeting on 28 November 1941: ‘Germany has resolved, step
by step, to ask one European nation after the other to solve its Jewish problem, and
at the proper time, direct a similar appeal to non-European nations as well.’ When
German troops broke through the Caucasus into the Middle East, he had no
imperial goals, he assured the Grand Mufti. ‘Germany’s objective would then be
solely the destruction of the Jewish element residing in the Arab sphere.’29 While
Hitler’s sincerity in renouncing territorial gains in the Middle East is doubtful, his
commitment to killing Jews there is not. When Rommel’s troops invaded Egypt
and threatened Palestine in the summer of 1942, plans were made to dispatch an
Einsatzgruppe.30
Though the Third Reich lasted for twelve years and the Second World War for
six, most of the killing of the Final Solution was in fact compressed into a very short
period. As of March 1942, some twenty to twenty-five per cent of all victims of the
Holocaust had already perished. Just eleven months later, by February 1943, some
seventy-five to eighty per cent of all victims had been killed. This massive genocidal
assault was accomplished through three distinct but occasionally overlapping sets
of killing operations. East of the old Nazi–Soviet demarcation line, numerous
mobile firing squads—usually organized by the Higher SS and Police Leaders or
Security Police successors to the Einsatzgruppen but comprising a bewildering array
of available manpower, including local militias—conducted regional sweeps and
liquidated the Jewish communities in one city, town, and village after another. In
Poland, repeated Aktionen emptied one ghetto after another, with the bulk of the
Jews put on trains and sent to their immediate death in the gas chambers of
Chelmno, Belzec, Sobibor, and Treblinka, and a small minority granted a brief
stay of execution in various slave labour camps.
In the other countries of Europe, Germany approached various puppet and
satellite governments on the one hand or its own occupation regimes on the other
28 Political Archives of the German Foreign Office, Pol. Abt. III 246, Luther memorandum, 17
October 1941.
29 Akten zur deutschen auswärtigen Politik, Series D, vol. 13/2, 718–21 (Schmidt memorandum, 30
November 1941).
30 Klaus-Michael Mallmann and Martin Cüppers. ‘ “Beseitigung der jüdisch-nationalen Heimstätte
in Palästina”: Das Einsatzkommando bie der Panzerarmee Afrika 1942’, in Jürgen Mathäus and Klaus-
Michael Mallmann (eds), Deutsche, Juden, Völkermord: Der Holocaust als Geschichte und Gegenwart
(Darmstadt: WBG, 2006). 153–76.
to round up and turn over their Jews for deportation. This was usually done with
the help of local collaborators, and the most common destination of these depor-
tation trains was Auschwitz-Birkenau. Major deportation programmes began from
Slovakia, France, Belgium, and the Netherlands in 1942, and from Bulgaria and
Greece in the spring of 1943. In the wake of defeat in Stalingrad in February 1943,
however, German leverage on collaborating regimes declined, and the pace of
deportations slackened from those countries in which Germany was dependent
upon the cooperation of others. The key exception in this regard was the tragic
deportation of nearly one-half million Hungarian Jews in May/June 1944, immedi-
ately following the German installation of a more compliant satellite government
there. The last phase in the genocide of the European Jews then resulted from the
‘death marches’, as the Germans continually forced surviving Jewish camp inmates
to withdraw before the Allied advance under the most lethal conditions in the last
months of the war.
Ultimately, the Nazi Revolution was to be a racial revolution that reconfigured
the demographic make-up of the Nazi empire. The destruction of the Jews was the
most comprehensive and far-reaching component of this racial revolution and has
become the paradigmatic historical example of total genocide. The further the
genocide of the Jews progressed while the chances of German victory simulta-
neously diminished, the more obsessed and paranoid Hitler and other leading
Nazis became about their imagined Jewish threat as a ‘world Jewish conspiracy’.31
While no other group targeted by the Nazis took on such mythical and threatening
proportions, numerous other groups perceived as either constituting a danger to
German racial purity and strength within the Third Reich or occupying ‘space’
claimed as vital German Lebensraum were also targeted for decimation or elimina-
tion through various programmes of expulsion, starvation, and mass murder.
However imprecise the notion of destruction ‘in part’ of the genocide convention
may be, the actual fate of the handicapped and Sinti and Roma within the Third
Reich and the intended fate of the Slavs of Eastern Europe pass this threshold.
Aly, Götz, ‘Final Solution.’ Nazi Population Policy and the Murder of the European Jews
(London: Arnold, 1999).
Browning, Christopher, and Jürgen Matthäus, The Origins of the Final Solution: The
Evolution of Nazi Jewish Policy, September 1939-March 1942 (Lincoln: University of
Nebraska Press, 2004).
31 Saul Friedländer, Nazi Germany and the Jews, vol ii: The Years of Extermination (New York:
HarperCollins, 2007). Jeffrey Herf, The Jewish Enemy: Nazi Propaganda during World War II and the
Holocaust (Cambridge, MA: Harvard University Press, 2006).
the nazi empire 425
Burleigh, Michael, and Wolfgang Wippermann, The Racial State: Germany 1933–1945 (New
York: Cambridge University Press, 1991).
Friedlander, Henry, The Origins of Nazi Genocide: From Euthanasia to the Final Solution
(Chapel Hill: University of North Carolina Press, 1995).
Friedländer, Saul, Nazi German and the Jews, vol i: The Years of Persecution, and vol ii: The
Years of Extermination (New York: HarperCollins, 1996, 2007).
Gerlach, Christian, Krieg, Ernährung, Völkermord. Forschungen zur deutschen Vernichtung-
spolitik im Zweiten Weltkrieg (Hamburg: Hamburger Edition, 1998).
Hilberg, Raul, The Destruction of the European Jews, 3rd edn (New Haven, CT: Yale
Kay, Alex, Exploitation, Resettlement, Mass Murder. Political and Economic Planning for
German Occupation Policy in the Soviet Union, 1940–1941 (New York: Berghahn Books,
2007).
Lewy, Guenther, The Nazi Persecution of the Gypsies (New York: Oxford University Press,
2000).
Longerich, Peter, Politik der Vernichtung. Eine Gesamtdarstellung der nationalsozialistischen
Judenverfolgung (Munich: Piper, 1998).
Madajczyk, Czeslaw, Die Okkupationspolitik Deutschlands in Polen 1939–1945 (Berlin: Aka-
demie Verlag, 1987).
Rutherford, Phillip, Prelude to the Final Solution: The Nazi Program for Deporting Ethnic
Poles, 1939–1941 (Lawrence: Kansas University Press, 2007).
Schleunes, Karl, The Twisted Road to Auschwitz: Nazi Policy toward German Jews 1933–1939
(Urbana: University of Illinois Press, 1970).
Wendling, Paul, Health, Race and German Politics between National Unification and Nazism,
1870–1945 (Cambridge: Cambridge University Press, 1989).
Zimmerman, Michael, Rassenutopie und Genozid: Die nationalsozialistische ‘Lösung der
Zigeunerfrage’ (Hamburg: Christians, 1996).
chapter 21
.............................................................................................
TWENTIETH-
C E N T U RY C H I NA
E T H N I C A S S I M I L AT I O N A N D
INTERGROUP VIOLENCE
.............................................................................................
uradyn e. bulag
GENOCIDE AND ITS DISCONTENTS

................................................................................................................
Dissident nationalists of the Mongol, Tibetan, and Uyghur minorities in the
People’s Republic of China (PRC) now contend that China has committed physical
and cultural genocide against their nationalities. The Inner Mongolian People’s
Party, an exiled Mongolian dissident group, claims that genocide was carried out
by the Chinese against the Mongols during the Cultural Revolution, and some
insist that ‘it is proper to state that the so-called policy of “improving Mongolian
people’s traditional way of production” is a type of cultural genocide.’ Uyghur
nationalists also maintain that they have been suffering from the twin genocides.
Indeed, for Tibetan nationalists, Inner Mongolians and Uyghurs have already been
culturally genocided as they have been outnumbered by Chinese settlers in their
own homelands, so their fates now serve as handy mirrors for what will become of
Tibet. A Tibetan in exile opined on the eve of the railway connection to Lhasa in
2006 thus: ‘When the railway will be fully operational, Chinese migration to Tibet
twentieth-century china 427
will likely accelerate and crush the last hopes of Tibetan cultural survival as similar
projects have done in Eastern Turkestan [Xinjiang] and Inner Mongolia.’1
Cultural genocide has now become a staple of the Dalai Lama’s criticism of
Chinese treatment of Tibetans. As early as October 1996, in an interview in Le
Monde, the Dalai Lama remarked: ‘A kind of cultural genocide is in progress in
Tibet. And even if losing her independence is acceptable, then still the destruction
of our spirituality, of Tibetan Buddhism, is unthinkable. Protecting the Tibetan
heritage has become my primary occupation.’2 On 16 March 2008, two days after
the Chinese suppression of riots in Lhasa which triggered a month-long pro-
Tibetan protest against the worldwide Beijing Olympic torch relay, the Dalai
Lama again accused the Chinese government of carrying out ‘cultural genocide’
by promoting the influx of Chinese migrants into Tibet and by means of restric-
tions on Buddhist practices: ‘Whether intentionally or unintentionally, some kind
of cultural genocide is taking place.’3
Genocide, a neologism introduced in the 1940s characterizing a state’s or a
dominant group’s mass killings of a culturally, ethnically, or religiously different
minority group, carries legal, moral, and political consequences for the accused.
The term has empowered victimized groups to redress wrongs, often becoming a
basis for demanding increased autonomy or independence. Fearing secession and
state disintegration, the Chinese government, like all national governments, has
rejected categorically all such allegations, calling them ‘nothing but lies’, insisting
that the Tibetans, and all ethnic minorities in the PRC for that matter, enjoy full
political rights of autonomy and have prospered under the Chinese rule; their
population has multiplied, and the Chinese state has been investing heavily to
maintain and promote their cultural heritage.4
Some Western scholars seem to concur. Donald Lopez Jr. has criticized the Dalai
Lama and the Tibetans in exile for desiring national independence of Tibet based
on a claim to protect Tibetan ‘national culture and nature’ even while Tibetan
Buddhism has become phenomenally successful only after it has been removed
1 Wu Yunna, ‘Another High Tactic of the Cultural Genocide by the CCP’, [online] (2002), available at
http://web.radicalparty.org/pressreview/print_right.php?func=detail&par=3685; ‘China’s Occupation of
East Turkistan, Genocide of Uyghurs’, [online] (21 July 2008), available at http://www.ireport.com/docs/
DOC-48003; Tenzin Dargyal, ‘Bombardier and the Tibetan Cultural Genocide’, Op-Ed, International
Campaign for Tibet, [online] (30 May 2006), available at http://www.savetibet.org/media-center/tibet-
news/bombardier-and-tibetan-cultural-genocide (accessed 29 July 2008).
2 Quoted in Donald S. Lopez, Jr., Prisoners of Shangri-La: Tibetan Buddhism and the West (Chicago/
London: University of Chicago Press, 1998), 199.
3 ‘Dalai Lama accuses China of “cultural genocide” ’, USA Today, 16 March 2008, available at http://
www.usatoday.com/news/world/2008–03–16-china-tibet_N.htm (accessed 29 July 2008).
4 Information Office of the State Council of the People’s Republic of China, ‘Regional Ethnic
Autonomy in Tibet’ (May 2004), available at: http://english.peopledaily.com.cn/200405/23/
eng20040523_144150.html
428 uradyn e. bulag
from its native soil.5 Barry Sautman has launched a vehement attack on the Tibetan
accusations of cultural genocide, insisting that Tibetans, instead of being extin-
guished either as a culture or as a population, have thrived during the Chinese
communist rule.6 He denies that Tibet is an internal colony of China; rather
Tibetans enjoy equal rights as citizens, and he buttresses his claim by quoting the
Dalai Lama, who has expressed willingness to accept the Chinese rule, and has
acknowledged that Tibetan culture is part of Chinese culture. Tom Grunfeld went
further by discrediting the foundational text of the genocide case—the findings of
the International Commission of Jurists in the wake of the 1959 Tibetan uprising
which made the original allegation that the Chinese had committed cultural
genocide—as a product of cold war anti-communism.7
These open rebuttals of the Tibetan genocide charge are significant, not just for
their high-profile defence of the Chinese record in Tibet, but for the fact that theirs
are not isolated voices. It is notable that the Western academic community has been
by and large reluctant to use the term genocide for China’s treatment of ethnic
minorities, except perhaps for describing the Manchu destruction of the Dzungar
Mongols in the early period of the Qing dynasty (1644–1911),8 even if they are
sympathetic with the minorities and concerned with the human rights situation in
Inner Mongolia, Tibet, and Xinjiang. To be sure, before the 1990s, Western studies
of China’s nationalities were much more inclined to talk of China’s assimilation of
nationalities, though they never denounced China as a ‘prison of nationalities’ as
they did the Soviet Union, Yugoslavia, and many other communist multinational
countries. Since the late 1980s, scholars have rejected the theories of assimilation,
moving instead to emphasize the Chinese state’s ‘creation’ of fifty-six minzu or
nationalities and allotment of autonomies to all but the Han, the majority.9 In the
post-structural theorization about majority and minority relationship as being
intersubjective, minority differences are said to have been accentuated or reified
for the very purpose of constructing and sustaining the Han Chinese majority.10
The paradigm shift in conceptualizing the ‘nationality question’ in China as ethnic
5 Lopez, Prisoners of Shangri-La.

6 Barry Sautman, ‘Colonialism, Genocide, and Tibet’, Asian Ethnicity 7:3. (2006), 243–65. Idem,
‘Tibet and the (Mis-)Representation of Cultural Genocide’, in idem (ed.), Cultural Genocide and Asian
State Peripheries (New York: Palgrave Macmillan, 2006), 165–279.
7 Tom Grunfeld, The Making of Modern Tibet (Armonk, NY: M. E. Sharpe, 1996), 146–9.
8 Peter Perdue, China Marches West: The Qing Conquest of Central Eurasia (Cambridge, MA:
Harvard University Press, 2005). Mark Levene, Genocide in Age of Nation State, vol ii: Rise of the West
and the Coming of Genocide (London: I. B. Tauris, 2005).
9 Dru C. Gladney, Muslim Chinese: Ethnic Nationalism in the People’s Republic (Cambridge, MA:
Harvard University Press, 1991); Katherine Palmer Kaup, Creating the Zhuang: Ethnic Politics in China
(Boulder, CO: Lynne Rienner, 2000).
10 Dru C. Gladney (ed.), Making Majorities: Constituting the Nation in Japan, Korea, China,
Malaysia, Fiji, Turkey, and the United States (Stanford: Stanford University Press, 1998).
relationship poses a significant challenge to minority nationalists’ perception of

their fates.
How do we reconcile these two diametrically opposed views of the Chinese state
behaviour to ethnic minorities? Has it ‘created’ minorities or ‘destroyed’ them? We
have now come to an impasse largely caused by disagreement not only over the
meaning of ‘genocide’ but more importantly over how even to conceptualize
ethnicity in China.
In this chapter, I invoke a Chinese political concept of ‘sinicization’ (hanhua,
literally becoming Han Chinese), with a hope to capture the nature of ethnic
relations in China historically, and the political fate of ethnic groups in contempo-
rary China. As will be clear, sinicization has powerful genealogical and governmen-
tal dimensions; it is not primarily an ‘acculturation’ process as it is understood
generally.11 Sinicization may not kill people directly, but it murders the non-
Chinese sense of genealogical differences and their polities.
S I N I C I Z AT I O N AND POLITICIDE
................................................................................................................
To focus on sinicization is to take up a theme indigenous to China and key to
understanding China’s treatment of non-Chinese people historically. It is not an
uncontroversial concept in Chinese studies circles, however. In the Western acade-
mia, sinicization is at the centre of assessing what are called ‘conquest dynasties’,
such as the Mongol Yuan (1271–1368) and the Manchu Qing (1644–1911), especially
the latter. Promoted by Mary Wright and Ping-ti Ho, sinicization was a major
paradigm from the 1950s to the 1980s to explain the Manchu success in conquering
and ruling China for more than two and half centuries.12 The paradigm held that
the Manchu adopted a policy of ‘systematic sinicization’ by sponsoring neo-
Confucian norms of government, so much so that in the second half of the dynasty,
the Manchu became largely sinicized, losing their own language and ethnic identity
and becoming simply a ruling class.
The sinicization thesis has recently received powerful head-on challenges from
historians of the Qing period. Evelyn Rawski, in a major polemic against Ping-ti
Ho in 1996, argued that the key to Qing success in empire building lay in its ability
11 Cf. Lemkin’s objection to using terms like ‘Germanization’, ‘Magyarization’, ‘Italianization’.

Raphael Lemkin, Axis Rule in Occupied Europe: Laws of Occupation, Analysis of Government,
Proposals for Redress (Washington DC: Carnegie Endowment for International Peace, 1944), 80.
12 Mary Wright, The Last Stand of Chinese Conservatism: The T’ung-Chih Restoration, 1862–1874
(Stanford: Stanford University Press, 1957); Ping-Ti Ho, ‘The Significance of the Ch’ing Period in
Chinese History’, Journal of Asian Studies 26:2 (1967), 189–95.
430 uradyn e. bulag
to use its cultural links with the non-Han peoples of Inner Asia and to differentiate
the administration of the non-Han region from the administration of the former
Ming realm. She further argued that sinicization was nothing but twentieth-
century Chinese nationalist ideology. More recently, Pamela Crossley and Mark
Elliot countered the sinicization thesis by insisting that the Manchu rulers
were deeply concerned with maintaining what Elliot called Manchu ‘ethnic
sovereignty’.13
Should we then abandon sinicization as an analytical concept as we have
genocide? Surely sinicization also refers to the fate of the Yueh and of Nanzhao—
originally non-Chinese areas in the south that were indeed submerged into Chinese
culture. In fact, the Manchu embarked on systematic sinicization of themselves and
the non-Chinese polities towards the end of the Qing even though trying to
maintain their own sovereign domination. Moreover, the theoretical recognition
of our common intersubjective ground hasn’t abrogated the genealogical and
political problems of sinicization in China.
The genealogical underpinning of sinicization is imbedded in the Confucian
order of xia-yi zhibian (Chinese-barbarian distinction), and its avowed goal of
yixia bianyi (using Chinese to transform barbarians) as preordained law of the
world. It is most eloquently embodied in the famous ancient adage fei wo zu lei, qi
xin bi yi (if he is not of our lineage, he is sure to have a different heart), suggesting
that even if, as a non-Chinese, one speaks Chinese and behaves like a Chinese in
every way, one is bound to be perfidious. Thus, for those who have been culturally
sinicized to pass as Chinese must invent new genealogies to prove their Chineseness
through patrilineal descent.14 It follows that sinicization is inherently ‘geneacidal’,
not so much in the sense of killing physically a non-Chinese people (though killing
was often rampant and mutual), nor necessarily forcing them to speak Chinese, but
in the sense of altering or obliterating their collective genealogical descent. Once
the sense of foreign descent is eliminated, a non-Chinese group is deemed Chinese
and is allowed to retain some of their ethnic customs. This explains the regional
differences of the Chinese communities.
Sinicization subscribes to the Confucian political ideology whereby the civilized
or Chinese be administered separately from the barbarians, and it has a mission to
13 Evelyn S. Rawski, ‘Presidential Address: Reenvisioning the Qing: The Significance of the Qing
Period in Chinese History’, Journal of Asian Studies 55:4 (1996), 831; Pamela K. Crossley, A Translucent
Mirror: History and Identity in Qing Imperial Ideology (Berkeley: University of California Press, 1999);
Mark C. Elliot, The Manchu Way: The Eight Banners and Ethnic Identity in Late Imperial China
(Stanford: Stanford University Press, 2001).
14 For the dialectics of Chinese culturalism and racism, see Patricia B. Ebrey, ‘Surnames and Han
Chinese Identity’, in Melissa J. Brown (ed.), Negotiating Ethnicities in China and Taiwan (Berkeley:
Institute of East Asian Studies, University of California, 1996), 19–36; Frank Dikötter, The Discourse of
Race in Modern China (Stanford: Stanford University Press, 1992); Prasenjit Duara, Rescuing History
from the Nation: Questioning Narratives of Modern China (Chicago: University of Chicago Press, 1995).
bring the barbarians under the same political administration for the Chinese. It is
in carrying out this mission that we can find the negation and intended destruction
of the non-Chinese peoples’ organizational structure and their ethnic sovereignty.
Thus, I suggest that essentially sinicization constitutes ‘politicide’ whose purpose is
the destruction of the capacity to produce or reproduce a polity with the eventual
aim of eliminating the ‘political identity’ of an ethnic group.15
Sinicization as a Confucian political programme should be located in the
unification of China under the Qin dynasty (221–206 bce), which established a
junxian or centralized bureaucracy to govern Chinese local regions under an
autocratic emperorship, and banished the former fengjian or feudal system. The
fengjian system did not die out, but was allowed to persist amongst ‘barbarians’ at
the margins of Chinese dynasties, granting them a certain amount of autonomy.
Barbarians were seen by the Chinese as genealogically different as they were
removed from the Chinese civilization. They were imagined to have bestial origins,
hence dangerous to the Chinese, a conviction borne out by repeated wars. This
difference warranted separate administrations under the fengjian system, but the
difference did not have an intrinsic value as it had to be destroyed through
changing the benxing (nature) of the barbarians by means of both altering their
genealogy and bringing them under the centralized junxian administrative system.
This was best manifest in a policy called gaitu guiliu, meaning replacing the tusi
(native) officials with the court-appointed civilian officials (see below). One might
argue that the fengjian was a concession to peoples beyond the reach of the Chinese
power, and junxian system was the political arm of sinicization, predicated as it was
on transformation of so-called barbarians, serving as a mechanism to expand
Chinese polity, limited only by its military power.
The fengjian system had a different fate in ‘conquest dynasties’ established by
Inner Asians, whom the Chinese call barbarians. In a conquest dynasty, the Chinese
Confucian civilization would often be deprived of its hegemonic status, made equal
to, if not lower than, other value systems brought into the dynasty. Buddhism, for
instance, was powerfully promoted by the non-Chinese rulers in Northern Wei,
Tang, Liao, Jin, Yuan, and Qing dynasties. Of course, not all conquest dynasties
were the same. The Mongol Yuan and Manchu Qing statecrafts, closely reflecting
their original confederate structure, were predicated on divide and rule, rather than
centralization. The Mongol Yuan recognized four major ethnic groups—Mongol,
Semu (coloured-eyed people, referring to the Turks, Arabs, and Europeans),
15 Politicide as I use here does not refer to mass killings for political reasons as originally proposed
by Barbara Harff. It is used in the sense defined by Baruch Kimmerling as ‘a process that covers a wide
range of social, political, and military activities whose goal is to destroy the political and national
existence of a whole community of people and thus deny it the possibility of self-determination’.
Barbara Harff, ‘Recognizing Genocides and Politicides’, in Helen Fein (ed.), Genocide Watch (New
Haven, CT: Yale University Press, 1992), 28; Baruch Kimmerling, Politicide: Sharon’s War against the
Palestinians (London: Verso, 2003), 4.
432 uradyn e. bulag
Hanren (Han people, including Khitan, Jurchen, and Han subjects of the northern
Song) and Nanren (southern people, referring to the subjects of the southern
Song), and the Manchu Qing five ethnic groups—Manchu, Mongol, Tibetan,
Hui (Muslims), and Chinese. Neither the Mongols nor the Manchu tried to
Mongolize or Manchuize the populations. The primary concern of these two
conquest dynasties was to maintain the ethnic sovereignty of the ruling groups
by means of largely separate administrations for the different components of their
empires, with ‘China’ ruled in a Chinese way and the other parts ruled according to
their own local political cultures. Sinicization would have defeated the very pur-
pose of their conquest.
However, non-Chinese conquest dynasties were not always immune from adopt-
ing the sinicizational model. When a conquest dynasty adopted the sinicizational
model, it would often have more devastating effect on their non-Chinese allies in
the empire than a Chinese dynasty would. This was because the policy came from
‘the inside’ rather than from the outside. The Northern Wei dynasty (386–534 ce) is
a case in point. Of Särbi (Chinese: Xianbei) or Inner Asian nomadic origin, after
moving the capital to Luoyang in 494, Emperor Xiaowen (467–499 ce) adopted a
policy of systematic sinicization, issuing edicts to ban Särbi clothing, language, and
surnames, all to be replaced by those of the Chinese. Sinicization marginalized the
Särbi’s own martial tradition, ultimately exposing them to both internal rebellion
and external invasions.16 Similarly, the Manchu attempt to destroy non-Chinese
polities towards the end of the empire was symptomatic of sinicization, and it led
to the loss of the support of their allies.
The tusi office was an indirect rule institution established by the Mongols during
the Yuan dynasty—in the place of the Han, Tang, and Song institution of jimi fu
(haltered-and-bridled prefecture), a concessionary administration for frontier
peoples—to rule non-Chinese groups in today’s Yunnan, Guizhou, and Sichuan.
The institution was taken over by the Chinese Ming dynasty, which, however,
repeatedly tried to eliminate it by establishing prefectures, departments, and
counties, a policy called gaitu guiliu. The Qing initially maintained the tusi system,
but it embarked on gaitu guiliu around 1700. The rationale was not so much ‘the
logic of the expanding empire demanded it,’ as John Herman suggested,17 but
rather because the Manchu rulers had never been able to identify with the non-
Chinese peoples such as the Lolo, Miao, and Turen, as they had with the Mongols,
16 For a dissenting view, see Dorothy C. Wong, ‘Ethnicity and Identity: Northern Nomads as
Buddhist Art Patrons during the Period of Northern and Southern Dynasties’, in Nicola Di Cosmo
and Don J. Wyatt (eds), Political Frontiers, Ethnic Boundaries and Human Geographies in Chinese
History (London: RoutledgeCurzon, 2003), 80–118.
17 John E. Herman, ‘The Cant of Conquest: Tusi Offices and China’s Incorporation of the
Southwest Frontier’, in Pamela Kyle Crossley, Helen F. Siu, and Donald S. Sutton (eds), Empire at
the Margins: Culture, Ethnicity, and Frontier in Early Modern China (Berkeley: University of California
Press, 2006), 161.
Turkic Muslims, and Tibetans. In this region, the Chinese were deemed more
useful to the Manchu conquerors, so the direct rule took on the Chinese institu-
tional structure, establishing prefectures and counties.
The Manchu controlled the Inner Asians through a variety of means such as
military conquest, marital alliance, religious patronization, and conferring titles.18
The Manchu administration of various Inner Asian groups was extremely
variegated, at times brutal, even genocidal to some less compliant groups, but
most groups enjoyed some degrees of autonomy vis-à-vis the Qing court, and they
were not subject to cultural assimilation to the Manchu or the Chinese. In fact, for
the greater part of the Qing dynasty, the Mongols, Muslims, and Tibetans enjoyed
higher status than the Chinese, and their lands were protected from Chinese
migration.
The Manchu alliance with the Inner Asians was awkward from early on, due
largely to the perennial Russian threat. After the first Opium War in 1840, as
European powers and Japan encroached on Qing territories, native rules, originally
designed to rally the Inner Asian peoples or to pacify them, now became a menace,
transforming Inner Asia from the rear base to the perilous frontier, constituting a
‘problem’. ‘China proper’ (zhongguo benbu/bentu), on the other hand, changed
from a target of conquest to heartland.
The Qing suppression of the Muslim rebellions in the mid-nineteenth century
heralded a sinicizational transition in the Qing empire, turning against its erstwhile
Inner Asian allies. For the first time in Qing history, a Chinese army under the
command of a Chinese general was deployed in Inner Asia. In 1884, on crushing the
rebellion in Huijiang (Muslim regions), the native rules were abrogated, and
the lands were renamed Xinjiang sheng (New Dominion province) with officials
directly appointed by the Qing court.19 The key to understanding the provincial
(xingsheng) system in the frontier was that it was not just an establishment of a
direct rule from the centre following the Confucian administrative system of
junxian. Rather it was settler-colonialist in character, intended to make the Chinese
population the bulwark of the new frontier regime.
Settler-colonization became an official Qing policy in 1902 when xinzheng (New
Policy) was introduced in the wake of the Boxer Rebellion. It was a typical case of
‘official nationalism’, viz., the Manchus almost totally identified with the Chinese
interest, and tried to extend the Chinese skin over the gigantic imperial body.
18 See Chia Ning, ‘The Li-fan Yuan of the Early Qing Dynasty’, PhD dissertation, Johns Hopkins
University, 1992. Joseph Fletcher, ‘Ch’ing Inner Asia c. 1800’, and ‘The Heyday of the Ch’ing Order in
Mongolia, Sinkiang and Tibet’, in John K. Fairbank (ed.), The Cambridge History of China, vol x: Late
Ch’ing, 1800–1911, Part I (Cambridge: Cambridge University Press, 1978), 35–106; 351–408. Nicola Di
Cosmo, ‘Qing Colonial Administration in Inner Asia’, International History Review 20:2 (1998),
287–309.
19 James A. Millward, ‘ “Coming onto the Map”: “Western Regions” Geography and Cartographic
Nomenclature in the Making of Chinese Empire in Xinjiang’, Late Imperial China 20:2 (1999), 61–98.
434 uradyn e. bulag
Mongolia was now turned into a frontier, subjected to a new policy of ‘open up’—
fangkeng mengdi—in which Mongolian land was slated for cultivation by bringing
in Chinese farmers. Of course, there had already been significant Chinese settle-
ment in the Mongol regions, especially in Josotu league, and in 1891 the Chinese
farmers massacred tens of thousands of Mongols in a rebellion called Jindandao.
Under this new official programme, large areas of fertile Mongol land, instead of
being rented from greedy or needy princes as had happened previously, were
forcefully confiscated and sold to Chinese settlers, a measure bringing tens of
thousands of Chinese peasants within a short period of time, and provoking
open resistance from Mongol princes. The same open-up policy had been carried
out unofficially much earlier in Manchuria, the Manchu’s own homeland, so that
in 1907 Manchuria was seamlessly turned into three provinces—Heilongjiang, Jilin,
and Fengtian (changed to Liaoning in 1954), making it unmistakably Chinese. Less
than half a century later, Manchuria lost its name, and began to be known as
Dongbei (the Northeast) or Dongsansheng (the Eastern Three Provinces).20
If the provincialization of Manchuria went unopposed thanks to the Manchu’s
own initiative and perhaps more importantly because Manchuria was largely
emptied of the Manchu who had been scattered in garrison towns throughout
the empire, their Inner Asian allies were less compliant. The threat of the New
Policy to the autonomy of the Mongol princes and the wide-scale violence between
Mongols and Chinese settlers in Inner Mongolia prompted Outer Mongolia to
declare independence in December 1911 with overwhelming response from Inner
Mongolia. Tibet soon followed suit. It is ironic that the Manchu sinicization and
their official nationalist programme in the Inner Asian frontier did not save them,
for Chinese nationalists, subscribing to the orthodox Confucian xia-yi distinction,
treated Manchu as alien barbarians unfit to rule China, holding them accountable
not only for humiliating the Chinese by conquering and ruling them, but for failing
to defend China against European and Japanese encroachments.21 The Qing
20 Burensain Borjigin, ‘The Complex Structure of Ethnic Conflict in the Frontier: Through the
Debates around the “Jindandao Incident” in 1891’, Inner Asia 6:1 (2004), 41–60. Tsai Sheng Luen,
‘Chinese Settlement of Mongolian Lands: Manchu policy in Inner Mongolia/A Case Study of Chinese
Migration in Jerim League’, PhD dissertation, Brigham Young University, 1983; Thomas. R.
Gottschang and Diana Lary, Swallows and Settlers: The Great Migration from North China to
Manchuria (Ann Arbor, MI: Center for Chinese Studies, University of Michigan, 2000); Mark C.
Elliot, ‘The Limits of Tartary: Manchuria in Imperial and National Geographies’, Journal of Asian
Studies 59:3 (2000), 603–646; Li Narangoa, ‘The Power of Imagination: Whose Northeast and Whose
Manchuria?’, Inner Asia 4:1 (2002), 3–25.
21 Peter C. Perdue, ‘Erasing the Empire, Re-racing the Nation: Racialism and Culturalism in
Imperial China’, in Ann Laura Stoler, Carole McGranaham, and Peter C. Perdue (eds), Imperial
Formations (Santa Fe: School for Advanced Research Press, 2007), 141–69; Peter Zarrow, ‘Historical
Trauma: Anti-Manchuism and Memories of Atrocity in Late Qing China’, History and Memory 16:2
(2004), 67–107.
empire crumbled in 1911, and with it the integrity of the Manchu as an organized
political entity.
The new Republic of China founded in 1912 was thoroughly sino-centric even
though it proclaimed a ‘union of five races’ (wuzu gonghe)—Chinese, Manchu,
Mongol, Tibetan, and Muslims.22 The multinational veneer was soon replaced by a
new concept—zhonghua minzu, the Chinese nation, in which the four non-Chinese
races were not recognized as having separate ethnic identities; instead, they were to
be treated as lineages (zongzu) of the Chinese nation stemming from the same
ancestors of the Chinese—Yan and Huang emperors.23 This genealogical national-
ism was unprecedented in Chinese history, not only in imagination, but also in
scope. The denial of different genealogical descent lines of the non-Chinese peoples
was accompanied by efforts to destroy their native polities implemented during the
Qing and bring them under the provincial system, a sinicizational policy carried
out in full force in Mongolia first.
Outer Mongolia’s independence was lost when it was forced to participate in a
tripartite conference with Russia and China, resulting in an autonomous state in
1915, recognizing China’s suzerainty. The autonomy was annulled by a Chinese
invasion in 1919 only to provoke another Mongol rebellion, eventually resulting in a
Mongolian People’s Republic (MPR) proclaimed in 1924 under the Soviet aegis.
The ‘loss’ of Outer Mongolia further strengthened the Chinese determination to
eliminate native administrations.
Inner Mongolia bore the brunt of the Chinese attacks. As early as 1915, the
Republic of China established three tebie xinzheng qu (special administrative
regions)—Suiyuan, Chaha’er, and Rehe—in the territory of Inner Mongolia over
the Mongolian leagues and banners. In 1928, as the Chinese Nationalist Party (the
Nationalists) won the civil war and built a Nationalist government, further pushing
for an exclusively Chinese nationalist agenda, these special administrative zones
were formally turned into provinces. ‘Inner Mongolia’ thence disappeared from the
map of the Republic of China. Several more provinces were built in China’s Inner
Asian frontiers in the same year: Ningxia in the Muslim-inhabited areas to the
north of the Gansu corridor, Qinghai in the Mongol–Tibetan–Muslim area of
Kokonuur, and Xikang in the Khamba region of eastern Tibet.
Sinicization, as we have seen, became first and foremost a nationalist project in
the Republican China. The vehemence the project took points to the extreme
anxiety of the Chinese Nationalists to build both a strong unitary state and a
homogeneous nation to stand up to challenges by other nations in the age of
22 Joseph W. Esherick, ‘How the Qing Became China’, in Joseph W. Esherick, Hasan Kayali, Eric
Van Young (eds), Empire to Nation: Historical Perspectives on the Making of the Modern World
(Lanham: Rowman AND Littlefield, 2006), 229–59.
23 James Patrick Leibold, ‘Constructing the Zhonghua Minzu: The Frontier and National
Questions in Early 20th Century China’. PhD dissertation, University of Southern California, 2002.
436 uradyn e. bulag
imperial nationalism. In this anxiety, the Chinese nationalists tolerated not even
Chinese local autonomy, still less frontier non-Chinese autonomy. In declaring the
Manchu, Mongols, Muslims, and Tibetans as genealogically Chinese, ethnicity
took the form of bianzheng (frontier governance). Since there was no ethnic
question as there were no officially recognized ethnic groups, the nationalist
project was declared accomplished upon destroying frontier administrations and
establishing provinces and counties settled by Han Chinese ‘lineage’ of the Chinese
nation. After 1928, only the MPR and the ‘Tibet proper’ ruled by the Dalai Lama
were beyond the reach of the Chinese power, but their demolition and provinciali-
zation remained the undying dream of the Chinese Nationalists until their own
demise in 1949.
S I N I C I Z AT I O N T H RO U G H N AT I O NA L I T Y
R E G I O NA L A U TO N O M Y
................................................................................................................
The Republican China’s politicidal programme of sinicization provoked intense
resistance from the non-Chinese peoples who now began to fight for autonomy
and some even for independence. Some turned to external powers as the only hope
for their survival as a collective group, appealing to the principles of national self-
determination and colonial liberation. In the 1930s, an Inner Mongolian autono-
mous government emerged with Japanese help. Like the Japanese, the Chinese
Communist Party (CCP), as a militant opposition party aspiring to take over the
government from the ruling Nationalists, found in the non-Chinese nationalism a
useful energy to tap, hoping to use minorities as useful allies to fight a common
enemy. It was in the CCP’s effort to win the confidence of minorities that the
party’s harshest and most effective denunciation of the Nationalists’ sinicization
programme was heard.
In 1935, when the beleaguered CCP settled in Yan’an of north China, facing
pressure from the Nationalists and a looming war with Japan, Mao Zedong made a
solemn declaration to the Mongols of Inner Mongolia, in which he denounced the
Nationalists’ non-recognition of ‘minority nationalities’ and their ‘sinicization’
policy for the suffering it caused to the Mongolian people. He vowed that his
Soviet government would restore the original Mongol league and banner system,
pledging, ‘Under no circumstances should other nationalities be allowed to occupy
the land of the Inner Mongolian nation or expropriate it under various excuses.’ In
1945, poised to challenge the Nationalists in the post-WWII settlement, Mao
condemned the Nationalists for denying the existence of nationalities in China
and particularly for the crime of ‘the massacre of Mongolians of the Ikhchao
[Yekejuu] League in 1943, the armed suppression of the minority nationalities in

Sinkiang [Xinjiang] since 1944 and the massacres of the Hui people in Kansu
Province in recent years’.24
‘Promise’, ‘recognition’, ‘sympathy’ were, as we have seen, the strategies of the
CCP used to intervene on behalf of the beleaguered minorities, but these concepts
are as illocutionary as governmental. Inner Mongolian communists-cum-nation-
alists were the first to respond to the CCP and established an Inner Mongolia
Autonomous Government in May 1947 with the CCP support. Since the founding
of the People’s Republic of China (PRC) in 1949, an autonomous system called
Minzu quyu zizhi (Nationality Regional Autonomy) has been implemented for
newly classified minority nationalities—55 as of today—each organized under
a territorial administrative unit. By the end of 2002, in addition to the five
province-level autonomous regions, there were 76 autonomous prefectures, 28
prefecture-level autonomous cities, 699 autonomous counties, and 68 county-
level autonomous counties. These autonomous areas cover 63.72 per cent of China’s
territories, and officially recognized minorities constitute 8.41 per cent of China’s
total population with a substantial number living in non-autonomous areas of
China.25
The post-revolutionary official storyline of the CCP nationality policy is that it has
‘saved’ the minorities from ‘extinction’ under the imperialists and the Nationalists,
and the minorities have been living happily and gaily in a great family of nationalities
since 1949. But why do minority nationalists by and large decry CCP’s nationality
policy as genocidal, and nothing short of genuine autonomy would guarantee their
continued survival as mentioned at the beginning of this chapter? The answer requires
a closer look at the kind of autonomy the CCP instituted to minorities.
Autonomy, for the CCP, is a tool of political expediency, and is granted accord-
ing to the power relations in a given situation. It is above all a mechanism for
integrating a politically organized group into the Chinese sovereignty, rather than
an institution in recognition of their intrinsic difference. Autonomous power is
reduced and even rescinded according to the degree of integration. This is a
systemic issue having its foundation in the ‘unitary’ national structure of the
PRC rejecting the federal system.26 Below are some features of the Nationality
Regional Autonomy (NRA) as implemented for minority nationalities.
First, a non-Chinese group is not ‘granted’ an NRA, but ‘brought into’ NRA only
when a new native leadership is ideologically and organizationally incorporated or
24 Stuart R. Schram (ed.), Mao’s Road to Power: Revolutionary Writings 1912–1949 (Armonk, NY:
M. E. Sharpe, 1999), v.71. Mao Zedong, ‘On Coalition Government’, in Selected Works of Mao Tse-tung
(Peking: Foreign Languages Press, 1967), 256.
25 National Bureau of Statistics of China, China Statistical Yearbook (Beijing: China Statistics Press,
2003), 43.
26 Baogang He, ‘Minority Rights with Chinese Characteristics’, in Will Kymlicka and Baogang He
(eds), Multiculturalism in Asia (Oxford: Oxford University Press, 2005), 56–79.
438 uradyn e. bulag
subordinated to the CCP which practises ‘democratic centralism’. The founding of

the Inner Mongolia Autonomous Government in 1947 was initially mired in a bitter
struggle about whether it should be led by the Inner Mongolia People’s Revolu-
tionary Party or the CCP. With the CCP leadership firmly established, the autono-
mous government was seamlessly changed to the Inner Mongolia Autonomous
Region, an integral part of the PRC in October 1949. Similarly, it took the
destruction of the semi-independent Tibetan Government in 1951, flight of the
Dalai Lama to India in 1959, and the removal of the Panchen Lama from Tibet in
1964, to prepare a condition to establish a Tibet Autonomous Region in 1965.
Bringing into NRA is therefore not recognition of native polity, but its reorganiza-
tion into a system little different from the Chinese provincial and county adminis-
trative system, with the leadership directly appointed by the CCP. In fact, although
the executive leaders of the NRA areas have been from titular nationalities (except
during the Cultural Revolution), the main power—the posts of Party secretary and
chairmanship of the People’s Congress—has been in the hands of the Han, with
very few exceptions.
Second, the NRA is a combination of nationality autonomy and regional
autonomy; it is institutionally designed to use the regional principle to trump
the nationality principle. It is characteristic of all autonomous places to include
substantial number of Han Chinese, and in many cases, the Chinese outnumber the
titular minorities in their own autonomous locality. Indeed, the NRA was justified
on an ‘objective’ reality of da sanju, xiao juju—implying that minorities lived in
small compact communities, and scattered amidst other groups, especially the Han
Chinese. The insistence on the Han incorporation was justified also on the ground
of evolutionism and Confucian ethics. In this logic, since socialist modernity
required economic and political advancement, Han, as an advanced people by
default, had to be there, with the ostensible altruistic purpose to ‘help’ minorities to
‘progress’. A more practical concern was national security, seeing minority nation-
alities as secessionist and subversive. Thus, with the Han as the indispensable
component of NRA, this autonomy has attained key features of settler-colonialism.
The imbrication of nationality autonomy and regional autonomy, buttressed by
ideological and national security imperatives, pit the minorities and the Han
against each other. The PRC established a large Chinese colony called Xinjiang
Production and Construction Corp created in 1954, a year before the founding of
the Xinjiang Uyghur Autonomous Region, and the Corp has 2.5 million Han
Chinese working there now. In addition, Han settlers gradually built up their
presence throughout Xinjiang, increasing from six per cent in 1949 to over forty
per cent of the region’s entire population at present. In Tibet, the number of Han
settlers is the most serious contention between the Tibetans and the Chinese,
constituting the core basis for the Tibetan independence movement and the Dalai
Lama’s demand for ‘genuine’ autonomy. Tibetans contend that they have already
become a minority in their own homeland; the Chinese government and some
academics insist, however, either that this phenomenon is confined to Lhasa or that
Chinese residents there are largely seasonal migrants, but not settlers.27
In Inner Mongolia, the issue of Chinese settlement has equally exasperated
ethnic tension, forcing even Mongol cadres to rebel. The issue was not so much
about the Chinese becoming the majority—by 1949, the Chinese had already
outnumbered the Mongols by seven to one. Rather, it was about the Chinese
settlers challenging the nominal Mongol autonomous rights, thus prompting
Ulanhu, the most prominent Mongol CCP cadre, to reissue Mao’s 1935 declaration
to the Inner Mongols in 1965, using Mao’s promise as the source of Mongolian
autonomy. In the subsequent struggle, the entire Mongol cadre corps became the
target of a witch-hunt as they were suspected of being underground nationalists
conspiring with the Soviet Union and the Mongolian People’s Republic to tear
away Inner Mongolia from China. Between 1968 and 1969, an Inner Mongolia-wide
campaign against Mongol nationalists was carried out, in which 16,222 people were
killed by official Chinese acknowledgement. Between 1969 and 1979, the Inner
Mongolia Autonomous Region was territorially dissected, with the western part
divided up by Gansu province and Ningxia Hui Autonomous Region, and the
eastern part by Heilongjiang, Jilin, and Liaoning provinces, leaving only the central
part under the direct Chinese military regime. This destruction is the very source of
a new wave of Mongolian nationalism. The high death toll has been characterized
as genocidal.28
Third, the NRA regions follow the same socialist modernization path as in China
proper, but development is characterized by explicit ethnic division of labour. The
developmentalist state prioritized developing heavy industry and natural resource
extraction industry in Inner Mongolia from the first Five-Year Plan in the early
1950s. In 1964, fearing a war with the Soviet Union and the United States, the
Chinese government launched a large-scale programme called the Third Front,
transferring many military and heavy machinery plants in the coastal cities and
Manchuria to the western and south-western mountainous regions inhabited by
27 Yan Hao, ‘Tibetan Population in China: Myths and Facts Re-examined’, Asian Ethnicity 1:1
(March 2000), 11–36. Barry Sautman, ‘ “Demographic Annihilation” and Tibet’, in Barry Sautman
and June Teufel Dreyer (eds), Contemporary Tibet: Politics, Development, and Society in a Disputed
Region (Armonk, NY: M. E. Sharpe, 2005), 230–57.
28 Uradyn E. Bulag, ‘Inner Mongolia: The Dialectics of Colonization and Ethnicity Building’, in
Morris Rossabi (ed.), Governing China’s Multiethnic Frontiers (Washington, DC: University of
Washington Press, 2004), 84–116; Tumen and Zhu Dongli, Kang Sheng yu Neirendang Yuan’an
(Kang Sheng and the Wrongful Case of the Inner Mongolia People’s Revolutionary Party) (Beijing:
Zhonggong Zhongyang Dangxiao Chubanshe, 1995); Wu Di, ‘ “Neirendang” Da Xue’an Shimo’ (‘The
Beginning and the End of the “Neirendang” Massacre Case’), in Song Yongyi (ed.), Massacres during
the Cultural Revolution (Hong Kong: Kaifang Zazhi She, 2002), 59–109; Altandelekei, Mongolian
Genocide during the Cultural Revolution in Inner Mongolia, trans. Yang Haiying (in (Shizuoka
University Departmental Bulletin Paper, 2008). [in Japanese] Available at: http://ir.lib.shizuoka.ac.
jp/bitstream/10297/2552/1/080716001.pdf
440 uradyn e. bulag
minorities such as the Yi and Tibetans. More than thirty new industrial cities
have developed over the decades with their population being largely Chinese
settlers. In 2000, as China’s coastal industries began to shift from assembling to
manufacturing, thereby requiring increasing amount of natural resources which
the western region has in abundance, China launched a new national programme
called Develop the West, xibu dakaifa, in imitation of the American westward
expansion model. In all these programmes, minority participation is minimum.
Fourth, the NRA purports to enjoy one of the world’s most favourable affirma-
tive action policies, but it proves to have cynical outcome for minorities. Minorities
are allowed to speak their languages, if they still can.29 They are also exempted from
the one-child policy. Although individually minorities have benefited from such
favourable policies, so have the majority individuals, many of whom have adopted
minority identity to accrue personal benefits. These individuals do not need to
satisfy any criteria for being a minority, nor do they need to demonstrate any traits
associated with a minority, except wearing minority dresses on ceremonial occa-
sions. They do not enrich minority cultures and languages, but ‘water down’ or
‘hollow out’ them, in Juha Janhunen’s terms, resulting in what he calls an ‘ethnic
inflation’, whereby populations of many minority groups have multiplied dramati-
cally. ‘Altogether, ethnic inflation is best seen as a government tool for the rapid
assimilation of the minorities under a minimal danger of ethnic conflict.’30 One
may add that such inflation can happen only when an ethnic group is deprived of
any internal mechanism to control its internal boundary.
As is clear, although the CCP recognized the existence of nationalities, nonethe-
less solving the ‘nationality question’ has been an integral part of the CCP’s overall
plan to ‘reconstruct’ China. This has been carried out by means of the NRA. Rogers
Brubaker’s characterization of the early Soviet nationality policies was equally
applicable to China:
first, to harness, contain, channel, and control the potentially disruptive political expression
of nationality by creating national-territorial administrative structures and by cultivating,
co-opting, and (when they threatened to get out of line) repressing national elites; and
second, to drain nationality of its content even while legitimizing it as a form, and thereby
to promote the long-term withering away of nationality as a vital component of social life.31
29 A major exception appears to be Xinjiang, where the government has intervened to discourage
Uyghur language. Many Uyghur language schools have been merged with Chinese language schools
beginning in the mid-1990s, which has been characterized as ‘linguicide—the forced extinction of
minority languages’. Arienne M. Dwyer, The Xinjiang Conflict: Uyghur Identity, Language Policy, and
Political Discourse (Washington, DC: East-West Center Washington, 2005), 39.
30 Juha Janhunen, ‘Tungusic: An Endangered Language Family in Northeast Asia’, International
Journal of the Sociology of Language 173 (May 2005), 43.
31 Rogers Brubaker, Nationalism Reframed: Nationhood and the National Question in the New
Europe (Cambridge: Cambridge University Press, 1996), 25.
We thus have in NRA a unique paradox of nationality-building for the purpose of

nationality-destruction; politicide was a built-in feature of China’s nationality building.
R E C T I F I C AT I O N OF N A M E S : T H E F I NA L S O LU T I O N ?
................................................................................................................
The rapidly changing ethnoscape and administrative landscape in China’s ethnic
minority frontier are unmistakably the success story of sinicization, aided by a
number of techniques, including but not limited to corporatism, developmental-
ism, nationalism, and the NRA system. This new ‘reality’ has to be reconceptua-
lized, renamed.
Recall that the Dalai Lama’s demand for ‘genuine autonomy’ is based on his
conviction that the NRA does not serve as an institution to protect Tibetan cultural
identity, quite the reverse. Ironically, the Chinese government and the Chinese
settlers are not happy with this ‘form’ of autonomy either, even though it is
defended as the best possible measure for fully guaranteeing minority identity. In
the remainder of this chapter, I focus on zheng ming (rectification of names), the
quintessential Confucian technique of governance—‘to govern is to rectify’—now
used by the Chinese government and some academics to ‘name away’ the political
identities of minorities and to fuse them into a revamped Chinese nation in the
recent two and half decades.
We can now identify a distinct governmental measure to rectify names in many
areas of ethnicity in China. There have been increasing calls to change the name
‘autonomous regions’ to ‘autonomous provinces’. In Inner Mongolia, the rectifica-
tion of place names is explicitly linked to a new mode of governance, for many
prefecture level leagues (Mongolian: aimag; Chinese: meng) have been changed to
zhixiashi (municipality) under a programme called che meng she shi—revoking the
league to establish municipality, some even losing their Mongol names to Chinese
ones. In 1981, barely two years after Jo’uda League was returned to the Inner
Mongolia Autonomous Region, it was renamed Chifeng Municipality. This was
followed by Jerim League, which was changed to Tongliao Municipality in 1999. As
of 2008, four more leagues have been turned to municipalities, though their
Mongol names have survived. The rationale for this rectification of names was
given by the party secretary and the mayor of Tongliao Municipality, both ethnic
Chinese, who wrote in a glowing piece published in People’s Daily on 6 October
1999: ‘This is the result of deepening reform, expanding opening-up, and accel-
erating development undertaken by the people of Tongliao under the leadership of
the Party . . . By replacing league with municipality (che meng she shi), history once
again gives Tongliao people a development opportunity that comes only once in a
442 uradyn e. bulag
thousand years.’32 What this suggests is that development opportunities for ethnic
Chinese require changing an NRA area into a cosmopolitan, and better an explic-
itly Chinese, space marked by a Chinese name.
The rectification of names are already taking place in many NRA places where
some native toponyms, even ethnonyms, have been distorted or rendered only in
Chinese, sometimes with an explicit aim of domesticating a transnational people.
For instance, the Kirghiz inside China is rendered Ke’erkezi, whereas outside China
it is Ji’erjisi; the Evenki inside China is called Ewenke, but Aiwenji for those in
Russia.33 Place names are not just names reflecting the topographical features of the
NRA regions; they reflect the cultural, physical, and social environments of the
name-givers. In changing the names of the places, these places are expropriated
from their native residents, making them unable to culturally associate with their
own land. A new ethno-archaeology is now on the rise whereby minority scholars
collect toponyms as a way of keeping memory of their homeland alive.
The most fundamental rectification pertains to renaming ‘the Chinese People’
(zhongguo renmin) as ‘the Chinese Nation/Nationality’ (zhonghua minzu) since the
late 1980s. This is a move away from the communist statehood to embrace the
principle of national state whose subject is not class-based ‘people of various
nationalities’ but nation. In this new move, the term minzu (nation/nationality)
has come to refer to the entire citizens of China, thereby creating a question of how
to call minority nationalities (shaoshu minzu). Since the mid-1990s, the English
translation of minzu has been officially changed from ‘nationality’ to ‘ethnic group’
or simply ‘ethnic’. Although the Chinese term minzu for minorities is retained in
official usage, the academic community has opted for a neologism, zuqun, literally
meaning ethnic group, in an effort both to follow the Euroamerican academic
norm and to ‘depoliticize minzu’. In this rendering, nationality is understood as a
political and territorial concept, and the designation of minorities as ‘nationalities’
is deemed detrimental to ‘national unity’, conducive to secessionism.
In the cacophony of depoliticizing minzu, some scholars have even suggested
changing the very term of zizhi (autonomy; literally self-rule) to gongzhi (literally
joint rule), translating the latter as ‘jointnomy’. The key purpose of this putative
innovation is to render the Chinese settlers as legitimate residents of autonomous
areas, making them indigenous to the land, thereby breaking the zhuti (titular)
32 Uradyn E. Bulag, ‘From Yeke-juu League to Ordos Municipality: Settler Colonialism and Alter/
Native Urbanization in Inner Mongolia’, Provincial China 7:2 (2002), 207.
33 Naran Bilik, ‘Names Have Memories: History, Semantic Identity and Conflict in Mongolian and
Chinese Language Use’. Inner Asia 9:1 (2007), 23–39. See also Magnus Fiskesjö, ‘The Autonomy of
Naming: Kinship, Power and Ethnonymy in the Wa Lands of the Southeast Asia-China Frontier’, in
Yangwen Zheng and Charles MacDonald (eds), Personal Names in Asia: History, Culture and Identity
(Singapore: University of Singapore Press, 2008).
principle in the current system that endorses minorities as the sole proprietors of
their respective autonomous areas.34
Most of the new grand conceptual narratives have made inroads among Chinese
officials, the population at large, including so-called liberal intellectuals and Chinese
students or new immigrants living in Western countries. They now see the Tibetan
and Uyghur ‘questions’ partly as a result of a semiotic concession to minorities. For
instance, the most common claim today is that the Tibetan demand for genuine
autonomy harms ‘the people of all ethnics in Tibet’ (Xizang gezu renmin, or for short
Xizang ren). We may say that many new conglomerate ‘jointnomous’ peoples have
been created: Xizang Ren for the Tibet Autonomous Region, Xinjiang Ren for the
Xinjiang Uyghur Autonomous Region, and of course Neimeng Ren for the Inner
Mongolia Autonomous Region, and so on. These new peoples are all ‘indigenous’!
Depoliticized, ethnic minorities can now be safely embraced and appropriated as
constituting the property of the reconstituted Chinese nation and culture. Today,
minorities are once again genealogically modified and secured as Yanhuang zisun
(children of Yan and Huang emperors) sharing the same ancestors with the Han
Chinese. It is not surprising that despite the demise of their native polities or
autonomies, or perhaps precisely because of it, minority cultures, especially their
ceremonial parts, have been ‘revitalized’. Minorities sing and dance, and ethnic
tourism is booming. The kind of ethnic cultures that exist in China today can only
be described as ‘spectacular’, forming a dazzling new ethnoscape.
It may now be concluded that sinicization has made a remarkable success in the
PRC more than any other time in Chinese history. Chinese policies have been
directed at destroying the possibility that non-Chinese national identity might
have any political meaning, at destroying the minorities’ capacity to think and do
politics independently as sovereign ethnic groups. In the PRC, the power of sinici-
zation has derived from several major sources: the nationalizing imperative of the
Chinese state that has not only domesticated non-Chinese peoples as Chinese
minorities, but subtly redefining minority identity to be an aspect of Chineseness,
rather than different from Chinese; the hegemonic representation of sinicization as
modernization which renders the Chinese culture as superior a priori, and which
imposes the Chinese culture as a gesture of benevolence that cannot be rejected;
large-scale Han Chinese settlement in minority areas and repression of minority
dissent; and most crucially the penetration into and tearing apart the inner fabric of
the minorities through the CCP’s organizational restructuring. What results is a
unique complex of both politicide and simulacrum of cultural efflorescence.
34 Ma Rong, ‘Lijie Minzu Guanxi de Xin Silu: Shaoshu Zuqun Wenti de “Qu Zhengzhi Hua” ’
(‘New Perspective to Understand Ethnic Relations: De-politicization of Ethnicity’), Bejing Daxue
Xuebao 41:6 (2004), 122–33; Zhu Lun, ‘Minzu Gongzhi Lun: Dui Dangdai Duo Minzu Guojia Zuji
Zhengzhi Shishi de Renshi’ (‘On Ethno-national Jointnomy: A Study of the Political Realities in
Ethno-national Communities in Contemporary States with Multiple Ethnic Groups’), Zhongguo
Shehui Kexue 4 (2001), 95–105.
444 uradyn e. bulag
Blondeau, Anne-Marie and Katia Buffetrille (eds), Authenticating Tibet: Answers to China’s
100 Questions (Berkeley: University of California Press, 2008).
Bovingdon, Gardner, Autonomy in Xinjiang: Han Nationalist Imperatives and Uyghur
Discontent (Washington, DC: East-West Center Washington, 2005).
Brown, Kerry, The Purge of the Inner Mongolian People’s Party in the Chinese Cultural
Revolution, 1967–69: A Function of Language, Power and Violence (Folkestone: Global
Oriental, 2006).
Bulag, Uradyn E., The Mongols at China’s Edge: History and the Politics of National Unity
(Lanham: Rowman and Littlefield, 2002).
Gladney, Dru C., Dislocating China: Reflections on Muslims, Minorities, and Other Subaltern
Subjects (London: C. Hurst, 2004).
Herman, John E., Amid the Clouds and Mist: China’s Colonization of Guizhou, 1200–1700
Liu, Xiaoyuan, Reins of Liberation: An Entangled History of Mongolian Independence,
Chinese Territoriality, and Great Power Hegemony, 1911–1950 (Stanford: Stanford Univer-
sity Press, 2006).
Margolin, Jean-Louis, ‘Mao’s China: The Worst Non-Genocidal Regime?’, in Dan Stone
(ed.), The Historiography of Genocide (Houndmills: Palgrave Macmillan, 2008), 438–67.
Millward, James, Eurasian Crossroads: A History of Xinjiang (New York: Columbia Univer-
sity Press, 2007).
Shakya, Tsering, The Dragon in the Land of Snow: A History of Modern Tibet since 1947
(London: Pimlico, 1999).
Smith, Warren W., Jr., China’s Tibet?: Autonomy or Assimilation (Lanham: Rowman and
Littlefield, 2008).
Starr, S. Frederick (ed.), Xinjiang: China’s Muslim Borderland (Armonk: M. E. Sharpe,
2004).
chapter 22
.............................................................................................
POLITICAL
GENOCIDES IN
P O S TC O LO N I A L
A S IA
.............................................................................................
robert cribb
ETHNIC AND P O L I T I C A L G RO U P I D E N T I T Y
................................................................................................................
Mass political killing presents a special analytical problem in genocide studies. The
slaughter of human beings because of the political beliefs and attitudes they held,
or were presumed to hold, cost millions of lives during the twentieth century. In
terms of the number of victims, the most extensive such killings—in the Soviet
Union, Indonesia, China, and Cambodia—match events unambiguously regarded
as genocide. Yet mass political killing is not prima facie covered by the 1948 UN
Convention that makes genocide a crime in international law. In identifying
genocide as acts intended to destroy a ‘national, ethnical, racial or religious’
group, the definition seems almost pointed in its exclusion of political killing.
Nor does mass political killing match the broader perception that genocide is
primarily a phenomenon of interethnic relations. Raphael Lemkin, who coined
the term, succinctly characterized genocide in 1945 as ‘deliberately wiping out
446 robert cribb
whole peoples’.1 His broader work, too, emphasized the character of genocide as an
act of cultural and social destruction, which compounded the crime of mass
murder. Generations of genocide scholars since Lemkin have invested great energy
into refining this characterization, but this refinement has worked mainly to
identify more complex processes of ethnic extermination. There remains a persis-
tent reluctance in the field of genocide studies to allow mass political murder to
qualify as genocide. Schabas put this position forcefully: ‘Confusing mass killing
of the members of the perpetrators’ own group with genocide is inconsistent with
the purposes of the Convention, which was to protect national minorities from
crimes based on ethnic hatred.’2 Even though the Holocaust against the Jews was
accompanied by a savage persecution of German socialists and other political
enemies of the Nazis, there is a powerful inclination to see the ethnic dimension
of the persecution as raising the Holocaust to a different level of turpitude
from that of the political murders. This inclination crystallized in the 1988 coining
of a new word, ‘politicide’, by Harff and Gurr to designate mass killings in which the
victims were targeted for their political affiliation, rather than their ethnicity.3 The
term, now two decades old, has had a mixed academic career. Although it has
been used by those who, like Harff and Gurr, are reluctant to regard political
killings as genocide, its form suggests that it applies to any kind of political killing,
including perhaps assassinations. Other scholars, Schabas amongst them, prefer to
use the term ‘crimes against humanity’ when discussing mass killing with political
motives.
Every definition is suspended in a web of explicit and implicit understandings of
the world. The definition of genocide that separates it from mass political killings is
located within a deep-seated perception in Western intellectual life that ethnic
identity is primordial, or at least historical, whereas political identity is a matter of
choice. Only a minority of scholars today see national identity as arising from the
deep cultural identity of a distinct people, but the mainstream of scholarly writing
on nationalism nonetheless emphasizes that the national identity of each individ-
ual is a product of powerful, even inescapable, forces of socialization. One is not
Indonesian, or Iranian, or Italian, or Icelander by choice, but rather by virtue of the
circumstances in which one grows up. By contrast, political identity is understood
to be malleable. One is left or right or centrist as a consequence of reflection on the
problems of the world. Even if the circumstances in which one grows up are
influential, they are not decisive—a conservative upbringing can and often does
produce a radical; a political career can shift from right to left and back again.
1 Raphael Lemkin, ‘Genocide: a Modern Crime’, Free World 9:4 (1945), 39.
2 William Schabas, Genocide in International Law: The Crimes of Crimes (Cambridge: Cambridge
3 Barbara Harff and T. R. Gurr, ‘Toward Empirical Theory of Genocides and Politicides:
Identification and Measurement of Cases since 1945’, International Studies Quarterly 37:3 (1988),
359–71.
political genocides in postcolonial asia 447
This distinction between primordial ethnic identity and acquired political iden-
tity leads to a significant difference in moral judgement. Aside from our condem-
nation of any atrocity or injustice, we feel especial indignation when people are
punished for what they are, rather than what they have done. This indignation
appears when we suspect, for instance, that members of minority communities
receive harsher treatment in the legal system than do members of the majority.
Indignation is most acute, however, in that quintessential genocidal image of the
massacre of women and children, the aged and infirm, for no other reason than the
ethnic group that they belong to. In the case of political killings, by contrast,
indignation is often tempered by the rationalization that the victims had chosen
to engage in politics and that their doom, even if unjust, was a consequence of their
own fatal decisions. This moral contrast between the innocent and the defiant—the
moral distinction between those who are made victims for what they are and those
who are made victims for what they have done, or failed to do—in fact reaches into
the heart of genocide studies itself. Although the UN Convention definition
identifies four straightforward categories of victim (‘national, ethnical, racial, or
religious’), genocide scholars have often not been inclined to explore religious
victimization as a distinct form of genocide unless it is linked with ethnicity, as in
the persecution of Jews.
The broader human consequences of ethnic destruction and political killing also
appear to be different. Ethnic identity is often perceived as part of the cultural
heritage of humankind, an irreplaceable human resource which, if lost, can no more
be recovered than can an animal or plant species that has been driven to extinction.
Genocide, Lemkin stressed, is a crime against humanity not just because the mode
of killing repudiates basic human values but also because the destruction of human
cultural diversity is a crime against all of us, not just the immediate victims. By
contrast, again, the dominant image of the victims of political and religious
persecution is that they go to their deaths confident that their beliefs will live on
after them and that their deaths may even hasten progress towards eventual victory.
Tertullian’s epigram, ‘The blood of martyrs is the seed of the church,’ has been
echoed in countless variants across the centuries. A host of political rationalizations
exists moreover for the destruction of political enemies. Wading through a sea of
blood pouring from slain political enemies in order to achieve a glittering future for
humankind may be a repellent image, but in the modern world it has been far less
repellent than the genocidal slaughter of children.
In the decades since the term genocide was coined and its official definition was
set, however, important reasons have emerged for regarding the distinctions between
ethnic and political mass killing as rather less significant than they appeared to be
in the 1940s. First, trends in scholarship have challenged the presumption that
ethnic identity is simply primordial or historically determined. In 1983, Eric
Hobsbawm and Terence Ranger coined the term ‘invention of tradition’, which
drew attention to the fact that many of those elements that constitute the most
448 robert cribb
visible and most appealing features of cultural identity were in fact created con-
sciously or half-consciously and rather recently as cultural symbols, rather than
being deeply embedded in a cultural-historical past.4 This insight implies a mallea-
bility of culture that sits uneasily with primordialist and historical assumptions of
the deep roots of ethnic identity. It implies that cultural identity can be acquired in
similar way to political identity. The cultural studies turn in scholarship has
reinforced this insight. Cultural studies scholars have deconstructed the notion
of cultural authenticity, showing that the privileging of some essentialized cultural
form, on the basis of its antiquity and or some other marker of validity, is itself a
cultural construction.5 Cultural studies scholars have also celebrated cultural
innovation, especially hybridity.6 This new understanding of culture as something
constantly in flux, constantly reinvented, constantly created, has eroded the notion
that cultural heritage is a fragile treasure to be preserved by interventionist protec-
tive measures. None of these insights devalued culture in moral terms; on the
contrary, cultural studies celebrated cultural diversity and heterogeneity in a way
that the older canonical approach did not. But they cast profound doubt over the
presumption that any culture could define a ‘whole group’ in the way specified by
the UN Convention. By asserting that culture, as a living, changing phenomenon,
could not be preserved without destroying it, cultural studies eroded the broad
philosophical underpinning of ethnic identity as a special category in social affairs.
With this erosion, the argument for restricting the term genocide to ethnically
directed killing was also diminished.
Second, other analytical trends in social science began to highlight similarities
between political and ethnic identity. Some of these similarities are especially
marked in the so-called pillarization (verzuiling, pilarization) of Dutch and Belgian
society. For much of the nineteenth and twentieth centuries—the period following
the emergence of mass politics—Dutch and Belgian societies were vertically seg-
mented into pillars (zuilen) according to religion and political ideology. Dutch
society was marked by four pillars, Protestant, Catholic, socialist, and liberal, each
of them with its own political parties, trade unions, newspapers, schools, univer-
sities, hospitals, sporting and cultural associations, and building societies. Even
large private firms were often strongly associated with a single zuil and would hire
staff only from that segment of the population. The zuilen were defined by belief,
including political belief, and they constituted whole social worlds within which
people were born, raised, trained, matched, employed, entertained, cared for, and
eventually buried. Political differences gave rise to cultural differences, including
4 Eric Hobsbawm and Terence Ranger (eds), The Invention of Tradition (Cambridge: Cambridge
5 Chris Barker and Paul Willis, Cultural Studies: Theory and Practice, 2nd edn (London: Sage,
2003), 392.
6 Ibid. 23–4.
differences in language use, which gave them a quasi-ethnic character.7 In Belgium,

in fact, ethnicity was part of the pillarization structure, with zuilen defined also by
the ethnic distinction between Dutch-speaking Flemings and French-speaking
Walloons. The Dutch–Belgian case in other words shows that ideological identity
can be expressed in public life in ways very close to the public expression of
ethnicity.
During the nineteenth and twentieth centuries, moreover, we also saw from time
to time a close association between national and political identities. This associa-
tion is encapsulated in the term ‘un-American’, used at the highest levels of the
United States in the 1950s in an attempt to align national identity with a conserva-
tive ideological approach to the world. A similar strong overlap between ideology
and nationality has been apparent at times in the identification of Nazism with
Germany, Communism with the Soviet Union and China, liberal internationalism
with Sweden, and Islam with Iran.
Of course, none of the considerations outlined above means that national and
ethnic identities are the same as ideological identities. Nonetheless, they blur the
boundary between these forms of identity in a way that removes the obvious
distinction between ‘acts committed with intent to destroy, in whole or in part, a
national, ethnical, racial or religious group’ and acts committed with similar intent
to destroy a political group. The ethnic genocides that dominate the discussion in
this volume had a variety of overlapping motives. In some cases, the aim was
dispossession—the displacement of original occupants from land so that others
could use it, or the removal of a category of people from powerful positions in
society so that others could take those positions. In some cases, the aim was to
prevent secession or alliance with an external power. In other cases, the genocide
was driven by a doctrinaire belief that members of the targeted group contaminated
the national polity, polluting society in a moral sense and blocking policies that
would have been in the interests of those to whom the nation properly belonged.
The four great mass political killings that took place in the twentieth century
shared this complex of motives, but in all cases that third motive was the most
salient. These mass killings were attempts to rid the body politic of a vast group of
people whose very existence seemed inimical to the nation as the perpetrators
conceived it. The mass killings arose from fundamental conflicts over the national
character of the Soviet Union, Indonesia, China, and Cambodia. The protagonists
in each case agreed on the physical framework provided by national borders, but
they were engaged in a struggle to reshape the fundamentals of national character
within those borders. The destruction of political enemies in these cases was not
just a grab for factional power or economic interest but the extermination of an
alternative nation within the state. In this respect, the victims indeed constituted a
7 Arend Lijphart, Verzuiling, Pacificatie en Kentering in de Nederlandse Politiek (Amsterdam:

J. H. de Bussy, 1968).
450 robert cribb
‘national’ group in the terms of the UN Convention. The mass political killings,
moreover, did not simply target political protagonists but rather victimized whole
families. Unlike the punishment of the families of political enemies in earlier times,
when it was intended to enhance the savagery of punishment and to diminish the
possibility that vengeful survivors might form a renewed threat to the regime, this
destruction was rationalized at the time by claims that the individual victims were
themselves culpable. Sometimes on the general basis of class, sometimes on the
more specific basis of family connections, the perpetrators of these killings turned
their enemies into quasi-ethnic groups and committed genocide upon them.
The political genocide in the Soviet Union is analysed in this volume by Nicolas
Werth, and therefore will be discussed here only in passing. Of the other three, the
Indonesian killings targeted communists whereas the Chinese and Cambodian
killings were carried out by communists. All three took place in the course of little
more than a decade (1965–1979) at the height of the cold war. Intense global
political antagonisms were thus acted out on three national stages, as was also
the case in the genocides of secessionist movements discussed in this volume by
Geoffrey Robinson. Although we know that the main protagonists in the cold war
were capable of immense direct brutality against their enemies—the United States
in Indochina, the Soviet Union in Afghanistan—and although we know that the
United States had some complicity in the Indonesian killings, the roots of all three
political genocides lay primarily in the national histories of the countries
concerned and in the vastly different futures which communism and its alternatives
appeared to map out for modernizing societies.
INDONESIA
................................................................................................................
Between October 1965 and March 1966, the Indonesian army and its political
allies carried out the murder of approximately half a million members and associ-
ates of the Indonesian Communist Party (Partai Komunis Indonesia, PKI). The
killings were part of a seizure of power by anti-communist sections of the army led
by General Suharto, who went on to become Indonesia’s longest-serving president,
removing from office his predecessor President Sukarno, who had declared the
country’s independence from the Dutch in 1945. Suharto’s long regime was marked
by the brutal repression of other opponents, including secessionists in Aceh, East
Timor, and Papua, and by deep-seated corruption, but it also delivered impressive
economic growth and a dramatic reduction in poverty.
The killing of the communists was presented as a response to an ambiguous
coup launched in Jakarta early on the morning of 1 October 1965 by a group calling
itself the 30th September Movement. Much controversy has existed over this coup.
Mid-ranking army officers were involved in the kidnapping and killing of six senior
anti-communist generals (a seventh narrowly escaped). The plotters claimed to
have acted to forestall a coup that they claimed had been planned by the anti-
communist generals for 5 October. After their bold initial strike, however, the
plotters appeared confused and were easily suppressed by Suharto as head of the
army’s Strategic Reserve (KOSTRAD). The controversy over the coup relates to the
question of who may have been involved aside from the immediate organizers.
Responsibility for devising the coup has been attributed variously (and in various
combinations) to the PKI, to Sukarno, to the air force (which was generally
sympathetic to the Left), and to Chinese intelligence operatives. It has also been
identified as a ‘black’ operation carried out by variously Suharto, the CIA, and MI6,
designed to provide a pretext for the suppression of the Left in Indonesia. Recent
research has indicated that the coup was most likely a joint endeavour by the army
officers directly involved and a small group around the communist party chairman,
D. N. Aidit, and that it was intended both to forestall an anti-communist coup and
to push Indonesia decisively to the Left.8
The PKI was at that time the third largest communist party in the world, the
largest in the non-communist world. It had risen from the ashes of an unsuccessful
rebellion in 1948 during Indonesia’s War of Independence to run fourth in the 1955
general elections. It had publicly eschewed armed revolution and was pursuing the
strategy, later identified with Euro-communism, of aiming to win power by
electoral means.9 It was a strong advocate for social justice and was considered
unusual amongst Indonesian political parties for its efficiency and freedom from
corruption. The party, however, had enemies, including those who remembered
the brutality of its 1948 uprising, religious groups apprehensive about the party’s
atheism, landowners fearful of the party’s base amongst landless peasants, and
establishment members of the bureaucratic, political, and military elites who
recognized in the party a hungry new elite that would displace them if it had the
chance. Fear that the PKI would win enough seats in the planned 1959 elections to
guarantee it a place in government was an important element in anti-communist
support for President Sukarno when he dismantled the democratic system between
1957 and 1959, replacing it with what he called ‘Guided Democracy’, a semi-
authoritarian system under his own direction.10
Although anti-communists initially saw Guided Democracy as a freezing of
existing political strengths, the PKI appeared to thrive. It continued to win support
8 John Roosa, Pretext for Mass Murder: The September 30th Movement and Suharto’s Coup d’Etat in
Indonesia (Madison: University of Wisconsin Press, 2006).
9 Robert Cribb, ‘The Indonesian Marxist Tradition’, in Colin Mackerras and Nick Knight (eds),
Marxism in Asia (London: Croom Helm, 1985), 251–72.
10 Herbert Feith, ‘Dynamics of Guided Democracy’, in Ruth McVey (ed.), Indonesia (New Haven:
Human Relations Area Files, 1963), 309–409.
452 robert cribb
in the countryside, the bureaucracy, and even sections of the armed forces. It also
developed a community of affiliated organizations, covering peasants, women,
trade unions, artists, and so on with an even larger membership.11 President
Sukarno, moreover, increasingly saw the PKI as a political counterweight to the
power of the army. He wove the PKI into a new national ideology, NASAKOM
(nationalism, religion, communism), giving it a formal place at the centre of the
Indonesian national identity, and his public rhetoric increasingly adopted the
terminology of the party.12 The party gave vehement support to Sukarno’s cam-
paign against the creation of the new state of Malaysia out of British colonies in
maritime Southeast Asia, and urged that the armed forces pursue an armed
struggle against the new state, a struggle that would take them far away from the
centre of power in Java. The party began to argue (though unsuccessfully) for the
arming of workers and peasants, and began to launch direct actions in the
countryside to implement a 1960 law on redistribution of land.13 As President
Sukarno aged and showed signs of illness, it appeared increasingly possible that the
party might slip effortlessly into political power, despite the fact that it held no
senior cabinet posts and despite its lack of access to weapons. These circumstances
created a perception in Indonesia that the country was approaching a decisive
moment in its history, in which either the PKI would come to power or the army
would move decisively to eliminate the party as a political force.
Research on the killings that followed the 1965 coup is still meagre, but at least
three patterns of killing are evident.14 In some regions, army units, especially the
para-commando RPKAD, moved systematically from village to village, using lists
obtained from ransacked PKI offices or from local informers to identify party
members and leftist activists, who were then summarily shot. This was the pattern
in places where the PKI had been a prominent and not necessarily antagonistic part
of the local landscape, so that it had relatively few enemies and its support base
included many people who were attracted by its general social activism and
reputation for incorruptibility. In other regions, the killings were carried out
11 Donald Hindley, The Communist Party of Indonesia 1951–1963 (Berkeley: University of California
Press, 1964).
12 Donald E. Weatherbee, Ideology in Indonesia: Sukarno’s Indonesian Revolution (New Haven, CT:
Yale University Press, 1966).
13 Rex Mortimer, The Indonesian Communist Party and Land Reform, 1959–1965 (Clayton, Vic.:
Centre of Southeast Asian Studies, Monash University, 1972).
14 Robert Cribb (ed.), The Indonesian Killings of 1965: Studies from Java and Bali (Clayton, Vic.:
Monash University Centre of Southeast Asian Studies, 1990); Geoffrey Robinson, The Dark Side of
Paradise: Political Violence in Bali (Ithaca, NY: Cornell University Press, 1995); John Roosa, Ayu Ratih,
and Hilmar Farid (eds), Tahun Yang Tak Pernah Berakhir: Memahami Pengalaman Korban 65: Esai-
Esai Sejarah Lisan (Jakarta: Lembaga Studi dan Advokasi Masyarakat, 2004); Hermawan Sulistyo,
‘The Forgotten Years: The Missing History of Indonesia’s Mass Slaughters Jombang-Kediri 1965–1966’,
PhD thesis, Arizona State University, 1997; R. A. F. Paul Webb, ‘The Sickle and the Cross: Christian and
Communist in Bali, Flores, Sumba and Timor, 1965–1967,’ Journal of Southeast Asian Studies 17:1
(1986), 94–112.
primarily by local militias drawn from amongst the PKI’s enemies. These killers
had often been engaged in earlier conflicts with the party over issues such as land
reform and they responded to clear signals from the army that violent action
against communists would not be punished. In still other regions, the army actively
sought to engage other sections of society in the killing. Party members were
ordered to report to the authorities. When they did so, or if they were caught on
the run, they were detained in makeshift jails, before being handed over in batches
to execution teams assembled from local communities. The enthusiasm of these
teams seems to have varied widely. Some participants were recruited unwittingly
and suddenly found themselves in a forest clearing, expected to cut the throats of a
handful of hooded, anonymous victims. Others took part to establish their own
anti-communist credentials. Others had been moved by lurid propaganda stories
about communist excesses put about to stoke popular hostility to the party. The
rightwing generals who died in the October coup were said to have been sexually
mutilated by communist women before they were killed; communists all over the
country were accused of preparing pits for the bodies of the anti-communist
victims they allegedly planned to slaughter on party orders.
There is no evidence of systematic records being kept of the killings. Most were
carried out clandestinely, often at night. Graves were unmarked and recorded only
in the memories of perpetrators and occasional bystanders. Memoirs by witnesses
to the killings are exceedingly sparse and an unknown number of victims, includ-
ing a few relatively senior party members, survived by fleeing and changing their
identities. Nonetheless, circumstantial evidence suggests that around half a million
people were killed, one-sixth of the party’s claimed membership.15 Members of the
women’s organization Gerwani, affiliated with the PKI and accused of involvement
in torture of the generals, were especially targeted, but a large majority of the
victims were men. Although it has often been suggested that Indonesians used the
opportunity presented by the purge to settle private scores, there is very little
evidence of the killing of people who were not believed to be communist. Nor is
there evidence that Indonesia’s Chinese minority was targeted for killing, though
there was some lesser violence in the form of burning of shops and expulsion from
small towns.
In the aftermath of the killing, more than a million people passed at least briefly
through military detention camps.16 Some prisoners spent ten years in detention
on the remote island of Buru in eastern Indonesia. Only a few dozen senior figures
from the party and the left wing of the army were brought to trial; the rest were
15 Robert Cribb, ‘How Many Deaths? Problems in the Statistics of Massacre in Indonesia
(1965–1966) and East Timor (1975–1980)’, in Ingrid Wessel and Georgia Wimhöfer (eds), Violence
in Indonesia (Hamburg: Abera, 2001), 82–98.
16 Greg Fealy, The Release of Indonesia’s Political Prisoners: Domestic Versus Foreign Policy,
1975–1979 (Melbourne: Monash Asia Institute, 1995).
454 robert cribb
detained on the basis of military fiat. Even after they were released, their identity
cards were marked with the letters ET (for eks-tahanan, ex-prisoner) and they were
forbidden to vote or to work in a wide range of supposedly sensitive jobs, including
education and the oil industry. The alleged sins of the party members, moreover,
were visited upon their families for decades thereafter. Thirty years after the 1965
coup, Indonesians wishing to work in sensitive areas or even, in some circum-
stances, to complete their education, were required to show that they were ‘en-
vironmentally clean’, meaning that they had no close family connections with
anyone who had been ‘involved’, according to the government’s flexible sense of
the term, with the PKI.17
Since soon after the emergence of the idea of Indonesia in the early twentieth
century, communism had presented a distinctive set of ideas about the fundamen-
tal character of the nation. The extermination of the party in 1965–6 thus repre-
sented not just the exclusion of a political ideology from national legitimacy but
also the genocidal destruction of the people who bore a certain idea of the nature of
the Indonesian nation.
C H I NA
................................................................................................................
The Chinese Communist Party (CCP) came to power in 1949, defeating the anti-
communist Nationalist government of Chiang Kaishek. Once in power, the Party
embarked on a thoroughgoing restructuring of Chinese society, collectivizing
agriculture, placing businesses under Party supervision, and promoting rapid
industrial development. Within the Party itself, however, there was a wide range
of opinion about the extent of change that was needed and about the pace that
should be maintained. The Party Chairman, Mao Zedong, favoured more radical
and more rapid change, and on his initiative the Party pursued what proved to be a
disastrous programme of rapid industrialization in 1958–61, known as the Great
Leap Forward. Under the programme, peasant labour was redirected into indus-
trial production. Along with unfavourable weather, these policies led to a collapse
in harvests and a famine in which millions died of starvation. The death toll is
officially stated to have been 14 million, but some observers have suggested
substantially higher numbers.18 In contrast with the famine genocide in Ukraine
17 Ariel Heryanto, State Terrorism and Political Identity in Indonesia: Fatally Belonging (London:
Routledge, 2006).
18 Peng Xizhe, ‘Demographic Consequences of the Great Leap Forward in China’s Provinces’,
Population and Development Review 13:4 (1987), 639–70; Judith Banister, ‘Population Policy and
Trends in China, 1978–83’, China Quarterly 100 (1984), 717–41.
in 1932–3, there is no reason to believe that the Great Leap Forward Famine was
deliberately engineered to destroy or weaken any social group. Rather, it was the
consequence of the blind confidence of Party leaders in the capacity of human will
to achieve official goals and the unwillingness of officials to report anything except
abundant fulfilment of those goals.
From the start, the CCP was fearful of forces in Chinese society that might block
or hamper its policies and which might even seek to remove it from power.
Although the Party had come to power with a formidable reputation for patriotism
and efficiency, it was well aware that vast numbers of Chinese were unconvinced of
the need for its radical programmes. Remnants of the Nationalist government,
moreover, remained in power on the island of Taiwan, and publicly affirmed their
intention to reconquer the mainland. From October 1950, too, Communist Chinese
troops were in direct conflict with US forces in the Korean War. The CCP
responded to this sense of threat with a series of savage purges intended to remove
various kinds of ‘dangerous’ element from Chinese society. The targets of these
purges in the 1950s and early 1960s included those defined as class enemies (land-
lords and business owners), people politically associated with the previous regime,
alleged spies for outside powers, dissident intellectuals, followers of organized
religion, and party members suspected of the abuse of power. The fate of those
purged varied from public executions to beating, to jailing or exile in the country-
side, to public humiliation, confiscation of possessions, and loss of career pro-
spects. Estimates of the number of executions in this period range from 400,000 to
2 million,19 but very little empirical work has been done to test these estimates.
These purges were marked by the Party’s recruitment of a new generation of young
revolutionaries who had not participated directly in the earlier struggle against the
Nationalists but who now became engaged in internal revolution.
The fears of Party chairman Mao Zedong that the revolution might be reversed
grew more acute after the death of Stalin and the rise of more moderate leaders in
the Soviet Union. Because the Soviet Union had been the CCP’s most important
ally and had provided its basic development model, Mao was particularly alert to
the risk that moderate forces would arise, as they had in the Soviet Union, within
the Communist Party itself. Whereas the purges of the 1950s had mainly targeted
categories of people who were demonstrably not of the Party, in the 1960s
Mao began to search for a way of purifying the Party itself, along with society as
a whole. The mass killing of communists in Indonesia in 1965–6 added to Mao’s
19 Maurice Meisner, Mao’s China and After: a History of the People’s Republic (New York: Free Press,
1986), 81; Benedict Stavis, The Politics of Agricultural Mechanization in China (Ithaca, NY: Cornell
University Press, 1978), 29. Jean-Louis Margolin, ‘China: A Long March into Night’, in Stéphane
Courtois (ed.), The Black Book of Communism: Crimes, Terror, Repression (Cambridge MA: Harvard
456 robert cribb
apprehension. Here was a communist party that had allowed itself to be defeated
by resurgent rightists. He was determined that no such fate would befall the CCP.
To counter these supposed threats, Mao formally launched the so-called Great
Proletarian Cultural Revolution in May 1966.20 The action was characterized as
‘cultural’ because it demanded a fundamental change in attitudes, rather than what
was portrayed as lip-service to the ideals of the revolution. In August 1966, the
Party’s Central Committee summed up the issues as follows:
Although the bourgeoisie has been overthrown, it is still trying to use the old ideas, culture,
customs, and habits of the exploiting classes to corrupt the masses, capture their minds, and
endeavor to stage a comeback. The proletariat must do just the opposite: It must meet
head-on every challenge of the bourgeoisie in the ideological field and use the new ideas,
culture, customs, and habits of the proletariat to change the mental outlook of the whole of
society. At present, our objective is to struggle against and crush those persons in authority
who are taking the capitalist road, to criticize and repudiate the reactionary bourgeois
academic ‘authorities’ and the ideology of the bourgeoisie and all other exploiting classes
and to transform education, literature and art, and all other parts of the superstructure that
do not correspond to the socialist economic base, so as to facilitate the consolidation and
development of the socialist system.21
A small group of ultra-radical party leaders, later derogatorily known as the ‘Gang
of Four’, systematically purged the upper echelons of the Party of those considered
to be potentially revisionist.
Because the heart of the problem, in Mao’s view, lay within the Party itself, the
principal agents of the Cultural Revolution were young, radicalized Chinese, male
and female, who were described collectively as Red Guards. Their movement
emphasized intense loyalty to Mao and his teachings, summarized in the celebrated
Little Red Book, and an implacable hostility to individuals, institutions, and forms
of behaviour that were not unambiguously embedded in the Maoist view of the
world. The Red Guards surged through Chinese society to root out all traces of
what were called the ‘Four Olds’—old customs, old habits, old culture, and old
thinking. Included in these vices were any sign of affection for or interest in China’s
traditional culture, the culture of any of the country’s ethnic minorities, or Western
thought and culture. Anyone who possessed suspect books or works of art, who
listened to Western or traditional music, who had Western education or connec-
tions with Western businesses, or who seemed hesitant to embrace the full exuber-
ance of the Maoist revolution was a likely target. Victims were routinely removed
from their jobs and their unacceptable possessions were burnt or smashed. They
20 On the Cultural Revolution in Chinese politics, see Harry Harding, ‘The Chinese State in Crisis,
1966–9’, in Roderick MacFarquhar (ed.), The Politics of China: The Eras of Mao and Deng, 2nd edn
(Cambridge: Cambridge University Press, 1997), 148–247; and Roderick MacFarquhar and Michael
Schoenhals, Mao’s Last Revolution (Cambridge, MA: Harvard University Press, 2006).
21 Central Committee of the CCP, ‘Decision Concerning the Great Proletarian Cultural
Revolution’, 8 August 1966.
themselves would be dragged into the street, self-incriminating placards placed

around their necks or dunce’s caps placed on their heads. They were forced to
confess publicly to their shortcomings while crowds jeered at them. The beating of
victims was widespread and as the campaign grew in intensity, increasing numbers
were beaten to death or were permanently crippled. A remarkable number of Party
leaders committed suicide.
This early phase of the Cultural Revolution attracted much attention because of
the character of the Red Guards. Energy and enthusiasm with which the Red
Guards celebrated violence and sought to smash every vestige of the old order
seemed to be the world’s most extreme manifestation of a general revolt of the
young that later found expression in the anti-establishment movements of 1968 in
the West. The Cultural Revolution was also remarkable for the destruction it
caused to China’s physical heritage, the distrust it created between family members,
the disruption it caused to the lives and careers of members of the middle levels of
Chinese society, and the chaos it brought within the Party. Mao as chairman of the
Party appeared to be orchestrating the destruction of the central political institu-
tion that he had helped to create.
By early 1967, not only had the education system been laid waste and the system
of local government largely dismantled by the Red Guards, but workers’ communes
had begun to seize control of factories and businesses. In many parts of the country
rival factions—Red Guard, worker, military—competed for practical control in
what was close to civil war. Under these conditions, Mao began to engineer a
reversal of the extreme devolution of power, calling for the establishment of
‘revolutionary committees’ that included both Party cadres with demonstrated
‘revolutionary’ qualities and, crucially, representatives of the People’s Liberation
Army. The creation of these committees intensified the conflict in China’s pro-
vinces, as rival factions tested each other’s power and struggled for inclusion in the
emerging elites.22 At the same time, the Gang of Four widened the purge of the
Party under the slogan ‘Campaign to Purify Class Ranks’. In the purge, a secret
committee investigated the backgrounds of all party cadres and removed anyone
with previous links to Westerners or the Nationalists or whose family background
included bourgeois elements. As part of the attempt to purge counter-revolution-
ary attitudes, millions of students, bureaucrats, and intellectuals were sent to the
countryside to learn peasant values through intense indoctrination and the disci-
pline of hard physical labour and difficult conditions.23 About 12 million ‘educated
22 Wang Shaoguang, Failure of Charisma: The Cultural Revolution in Wuhan (Hong Kong: Oxford
23 In a provocative essay, Jin Qiu discusses the preference of former Red Guards for remembering their
own suffering in this period while ignoring the suffering they inflicted on others; see ‘Victim or
Victimizer: The Reconstruction of the Cultural Revolution through Personal Stories’, in Kenneth
Christie and Robert Cribb (eds), Historical Injustice and Democratic Transition in Eastern Asia and
Northern Europe: Ghosts at the Table of Democracy (London: Routledge, 2002), 13–23. A particularly
powerful memoir of this period is Rae Yang, Spider Eaters (Berkeley: University of California Press, 1997).
458 robert cribb
youth’ were caught up in this process. Although the transfer to the countryside was
couched as a measure against counter-revolutionaries, in fact a large number of
Red Guards were removed in this way from the cities and from all opportunities to
continue agitating. Something close to normal rule began to be restored only
in 1971.
Early accounts of the Cultural Revolution emphasized the violence of the Red
Guards and implied that the revolutionary committees had been Mao’s way of
putting back into the bottle the genie of youth radicalism that he had released in
1966 once it had served the purpose of removing his enemies within the Party.24
Recent research, however, has shown that this period of revolutionary committees
was a time of intense political killings, in addition to the broader phenomenon of
civil war between rival armed groups. The evidence lies in official government
publications that record massacres for the years 1966–71.25 The deaths are recorded
in county almanacs, sometimes simply referring to ‘many killings’ or reporting that
they took place ‘frequently’, sometimes giving specific numbers, as in the following
report from Quanzhou County in Guangxi:
October 3, [1967]. In Sanjiang Brigade, Dongshan Commune, the militia commander
Huang Tianhui led [the brigade militia] to geng-sha [push into a cliff to kill] 76 individuals
of the brigade—former landlords, rich peasants and their children—in the snake-shaped
Huanguaneng canyon. . . . From July to October, [other] 850 individuals [in the county]—
the four-type elements (Landlords, Rich Peasants, Counterrevolutionaries, and Bad
Elements) and their children—were executed with firearms.26
The killing of children was particularly marked in the Cultural Revolution

massacres, probably because they were believed to share their parents’ inappro-
priate class background.27 The pattern of killings was different from county to
county and province to province, with the Inner Mongolian Autonomous Region
apparently suffering most heavily.28 Walder and Su, who have investigated these
records most closely, conclude that between 750,000 and 1.5 million people were
24 Juliana Pennington Heaslet, ‘The Red Guards: Instruments of Destruction in the Cultural
Revolution’, Asian Survey 12:12 (1972), 1032–47; Harding, ‘The Chinese State in Crisis’, 243–44
(Harding cites an official death toll of 34,800 and a ‘not . . . unreasonable’ estimate of half a million).
25 For a discussion of the circumstances in which these reports were compiled, see Andrew
G. Walder and Yang Su, ‘The Cultural Revolution in the Countryside: Scope, Timing and Human
Impact’, China Quarterly 173 (2003), 79.
26 Quoted in Yang Su, State Sponsorship or State Failure? Mass Killings in Rural China, 1967–68,
Paper 03–06 (Irvine, CA: Center for the Study of Democracy, University of California, 2003), 5.
27 See Yang Su, ‘Mass Killings in the Cultural Revolution: a Study of Three Provinces’, in Joseph
Esherick, Paul Pickowicz, and Andrew George Walder (eds), The Chinese Cultural Revolution as
History (Stanford: Stanford University Press, 2006), 107–8.
28 See Uradyn Bulag’s chapter in this volume, and Wu Di, ‘The Aftermath of the Cultural
Revolution in Inner Mongolia’, in Kenneth Christie and Robert Cribb (eds), Historical Injustice and
Democratic Transition in Eastern Asia and Northern Europe (London: Routledge, 2002), 24–37.
killed in rural areas during these years, with a similar number being maimed by
beatings.29
Much is still uncertain about the circumstances of most of the killings and of the
identities of the victims. As in Indonesia, however, the mass political killings of the
Cultural Revolution were marked by a sense that the future of the country was at
stake, meaning that extreme measures were justified to suppress the enemies of the
newly consolidating order. In both cases, the characterization of these enemies was
both casual and precise: in China, the victims were defined as ‘class enemies’, in
Indonesia as ‘communists’, both categories representing a kind of person, rather
than a culpable act. Whereas the perpetrators of the Indonesian killings have been
almost entirely silent on their actions, detailed records exist of the revolutionary
committee meetings at which executions were planned and the names of the
victims were decided. Although the committees recruited militias to assist on the
task of extermination, the actions took place under clear official mandate.30 And
although the Chinese government has since rehabilitated most of the surviving
victims of the Cultural Revolution (unlike Indonesia’s government, which continues
a range of discriminatory practices against those linked with the Communist Party),
there has been no attempt to rehabilitate or to memorialize the dead. Those who
took part in the killings were often promoted within the Party in subsequent years.31
The Chinese killings, like those in Indonesia, were aimed at the elimination of a
whole group of people believed to embody a different idea of Chinese nationhood.
Whereas the victims in Indonesia were defined principally by their formal affilia-
tion with the Communist Party or one of its affiliates, the victims in China were
defined more broadly by class, so that the slaughter often encompassed whole
families.
CAMBODIA
................................................................................................................
The Communist Party of Kampuchea (CPK), better known as the Khmer Rouge
(Red Khmers, or Cambodians), came to power in Cambodia in the final phase of
the Second Indochina War (1954–75). Little was known about the party in the rest
of the world and most observers were surprised, sometimes disbelieving, when
news emerged that the party had launched a massive re-engineering of society to
29 Walder and Yang Su, ‘The Cultural Revolution in the Countryside’, 74–99.
30 Yang Su, ‘Mass Killings in the Cultural Revolution’, 108.
31 Ibid.
460 robert cribb
create a pristine communist state and had killed its political and class enemies on a
massive scale. The regime was removed in 1979 by a Vietnamese invasion, following
Khmer Rouge attacks on Vietnam that aimed to recover former Cambodian
territory in the Mekong Delta. The Khmer Rouge resumed guerrilla resistance in
the countryside and, with the support of China and the United States (which
regarded Vietnam as an ally of the Soviet Union), retained Cambodia’s United
Nations seat until 1982. From 1982 until 1991, it remained the strongest element
within the Coalition Government of Democratic Kampuchea which, though inter-
nationally recognized, controlled only a small part of the country.
In early historical times, what is now Cambodia had been the centre of a series
of powerful kingdoms, whose splendour is still reflected in the vast monuments of
Angkor. In later centuries, however, the Khmer had been unable to prevent the
inroads of neighbouring Siamese and Vietnamese states. By the end of the
seventeenth century, Cambodia had lost to Vietnam control of its sea access
through the Mekong Delta, and it ceased to be a major power. During the
nineteenth century, France incorporated Cambodia, along with neighbouring
Vietnam and Laos, into French Indochina but Cambodia experienced few of the
direct consequences of colonialism that were visited on Vietnam. The Cambodian
monarchy remained in place under the French protectorate, bolstered by French
power and in turn conferring legitimacy on French domination. The modern
sector of the economy and the bureaucracy was dominated by French, Vietnam-
ese, and other outsiders.
Cambodia became independent in 1953 as a constitutional monarchy in the
context of the closing phases of the First Indochina War, whose most important
outcome was the partition of Vietnam between rival northern (communist) and
southern (anti-communist) states in 1954. This partition set the framework for
the Second Indochina War, in which the United States and its southern allies
fought the north, which had backing from the communist bloc. Although the
Cambodian leader, Prince Norodom Sihanouk, asserted Cambodia’s neutrality
and was one of the pioneers of the Non-Aligned Movement, the country was
increasingly drawn into the conflict by the fact that Vietnamese communist
forces used it as a sanctuary and by the heavy-handed US response, which
included massive bombing of communist-dominated regions. In 1970, Sihanouk
was overthrown in a pro-Western military coup by Gen. Lon Nol, who created
the Khmer Republic.
The bombing and the consequent hardship in the countryside caused, for
instance, by the death of an estimated seventy-five per cent of the draught animals
needed to cultivate the fields appear to have generated increasing support for the
CPK. Although the party had depended on Vietnamese military support and
protection in its early days, it increasingly pursued its own line and defended the
separate priorities of the Cambodian revolution against the ‘Indochina’ impera-
tives of the Vietnamese. Khmer Rouge forces captured the capital, Phnom Penh, on
17 April 1975, two weeks before the fall of Saigon to Vietnamese communist forces,
and the CPK renamed the country ‘Democratic Kampuchea’.32
CPK doctrine demanded the rapid creation of a fully communist society. The
party believed that communism could be established within a single country and
that industrialization and urbanization, conventionally seen as essential in the
Marxist historical sequence, could and should be bypassed. ‘We will be the first
nation to create a completely communist society without wasting time on inter-
mediate steps,’ one of their leaders is reported to have said.33 Party ideologues
argued that a just, prosperous, and entirely self-reliant society could be created by
‘unleashing the productive power of the masses’ in a programme of vast irrigation
works designed to allow rice to be grown all year round. This programme required
that the energies of the Khmer people be redirected from commerce (money
and banking was abolished and replaced with barter), from religion (religious
observance was punished and places of worship were destroyed),34 and from
the corruption and distractions of urban life and foreign values in general. Accord-
ingly, nearly all inhabitants of the cities were shifted to rural areas, where they were
set to work on infrastructure projects. Private property was abolished and work was
collectivized to varying degrees from region to region. In some places, collectiviza-
tion included communal dining. The party had previously implemented many
aspects of this programme in ‘liberated’ areas in the countryside before the fall
of Phnom Penh.
Whereas many communist parties allowed a respected but subordinate role to
‘progressive’ elements from enemy classes, the CPK expressed a doctrinaire hostili-
ty to these classes from the start. In theory, the party permitted people who had
held official positions or possessed some wealth under the old order to accept re-
education and to immerse themselves in the peasant values of the new order. In
practice, however, the party was vigilant for any sign amongst these old elements of
failure to conform to the new order. The strongest social division in Cambodia
under the Khmer Rouge was thus between the ‘old people’—those who had been
under effective communist government in the ‘liberated’ areas before 1975—and
the ‘new people’ (perhaps numbering two million or more), most of them dis-
placed from the cities. It was made clear to these people that their survival
depended on utter compliance with Khmer Rouge norms. Many survivors reported
that they were told words to the effect, ‘To spare you brings no benefit; to destroy
32 Ben Kiernan, How Pol Pot Came to Power: A History of Communism in Kampuchea, 1930–1975
(London: Verso, 1985).
33 Karl D. Jackson, ‘The Ideology of Total Revolution’, in idem (ed.), Cambodia, 1975–1978:
Rendezvous with Death (Princeton: Princeton University Press, 1992), 63.
34 Some 40,000–60,000 Buddhist monks were defrocked and forced into labour brigades;
Christians and Muslims, including the indigenous Muslim Cham minority, were targeted as
followers of foreign beliefs. Temples, mosques, and churches were destroyed, along with sacred
objects such as Buddha images and copies of the Bible and the Qur’an.
462 robert cribb
you brings no loss.’35 New people were subject to intense re-education, and they
were set to hard physical labour under difficult, often unhealthy conditions,
generally with meagre food rations. Families were often split up and sent to
different parts of the country. They were closely monitored by Khmer Rouge
cadres and trivial infringements of rules or signs of persisting bourgeois or foreign
values, such as expressions of nostalgia for pre-revolutionary times, could be
punished by summary execution. It was reported that people were executed for
wearing glasses (a sign of pre-revolutionary wealth and intellectualism) or for
having soft hands that betrayed a lack of experience in hard work. Survivors
reported that Khmer Rouge children were particularly vigilant in identifying and
punishing signs of un-revolutionary attitudes amongst the new people.36 Members
of Cambodia’s ethnic minorities—indigenous Muslim Chams and immigrant
Chinese and Vietnamese, as well as CPK cadres in the east of the country who
had been in contact with Vietnamese communists—were especially suspect be-
cause of their presumed foreign sympathies and suffered an even higher death rate
than the general population.37
In the period 1975–9, some 17,000 people passed through the Tuol Sleng (S-21)
interrogation centre in Phnom Penh. Victims of the regime’s growing paranoia,
these people included both alleged foreign spies and growing numbers of party
cadres who were interrogated and tortured. They were forced both to confess to
complex conspiracies against the regime and to reveal the names of their supposed
co-conspirators who were then called in for interrogation in a widening circle of
victimization, which also included small children.38
There are significant difficulties in estimating both the total number of victims
of the Khmer Rouge and the precise causes of death. Researchers have used
demographic data, sample surveys amongst survivors, and forensic evidence
from grave sites, as well as comparative evidence concerning the practicalities of
mass killings from other contexts. Although much uncertainty remains, there are
strong indications that the death toll was a little over two million out of a 1975
35 See, for instance, Ben Kiernan, The Pol Pot Regime: Race, Power, and Genocide in Cambodia under
the Khmer Rouge, 1975–1979, 2nd edn (New Haven, CT: Yale University Press, 2002), 4.
36 François Ponchaud, Cambodia Year Zero (London: Allen Lane, 1978); Michael Vickery,
Cambodia 1975–1982 (Sydney: Allen and Unwin, 1984); Elizabeth Becker, When the War Was Over:
The Voices of Cambodia’s Revolution and Its People (New York: Simon and Schuster, 1986); Jackson
(ed.), Cambodia, 1975–1978; David P. Chandler, The Tragedy of Cambodian History: Politics, War, and
Revolution since 1945 (New Haven, CT: Yale University Press, 1991); Alexander Laban Hinton, Why Did
They Kill: Cambodia in the Shadow of Genocide (Berkeley: University of California Press, 2005)
discusses the motivation for the killings using insights from psychology. There is now a substantial
literature of survivor accounts of the Khmer Rouge period in Cambodia. Two powerful memoirs are
Loung Ung, First They Killed My Father: A Daughter of Cambodia Remembers (Pymble, NSW:
HarperCollins, 2000), and Pin Yathay, Stay Alive, My Son (Chiang Mai: Silkworm Books, 2000).
37 Kiernan, The Pol Pot Regime, 251–309.
38 David P. Chandler, Voices from S-21: Terror and History in Pol Pot’s Secret Prison (Berkeley:
population of around 8 million. Of these victims, it seems plausible that forty to

fifty per cent were executed, with the remainder dying of disease and starvation
under conditions created deliberately or carelessly by the Khmer Rouge authori-
ties.39 The deaths were unevenly distributed across the population of Cambodia.
Vietnamese, Cham, and other ethnic minorities suffered the most; new people
suffered much more than old people.
C O N C LU S I O N
................................................................................................................
The Indonesian killings of 1965–6 cost half a million lives. The killings during the
Cultural Revolution in China claimed a million. Although these are massive death
tolls, both episodes involved only a relatively small proportion of the total
populations of those two countries. The political killings in the Soviet Union
took place on a much larger scale, but they proceeded as a multitude of separate
episodes, some involving ethnically defined victims, some in which hardship
played a greater role in the death toll than did execution. As a result, the Soviet
killings are difficult to comprehend as a single event. The Cambodian killings, by
contrast, claimed twenty-five per cent of a population of eight million in little
more than three years. Even though the perpetrators shared the ethnicity of most
of their victims, it seemed immediately unreasonable to many genocide scholars
to treat the Cambodian killings as anything other than genocide. A UN rappor-
teur is said to have coined the term ‘auto-genocide’ as an uncomfortable ac-
knowledgement of the fact that the events in Cambodia seemed to match the
enormity of ethnic genocide, yet did not fit the conventional interpretation of the
UN definition.40 Since the 1980s, however, the term auto-genocide has largely
disappeared and the experience of Cambodia continues to sit uncomfortably on
the sharp frontier of genocide studies, bringing other mass political killings such
as Indonesia and China in its wake.
Whether these cases can properly be considered as genocide depends above all
on the changing scholarly understanding of the nation, as it influences the meaning
of the term ‘national’ in the UN Convention. The political killings in Indonesia,
China, and Cambodia here and there crossed ethnic boundaries—the Khmer
Rouge’s treatment of the Cham was genocidal under any definition of the term—
but at their core they were political conflicts within a single ethnic group. They
39 Bruce Sharp, ‘Counting Hell: the Death Toll of the Khmer Rouge Regime in Cambodia’, [online].
Available at http:// www.mekong.net/cambodia/deaths.htm
40 Martin Shaw, What is Genocide? (London: Polity Press, 2007), 76.
464 robert cribb
were not, however, mere expressions of a culture of political violence in which

disputes were routinely resolved with the gun, the knife, or the crowbar. Rather
they were the outcome of fundamental struggles over national character. In each
case, the perpetrators were not simply eliminating a political enemy. They saw
themselves as shaping the character of their nation by removing a category of
people who could never be a legitimate part of it. This category was defined by
membership of the communist party in Indonesia and by imputed class member-
ship in China and Cambodia, but the rationale for purging was similar in all three
cases. The nation as it was envisaged by those in charge could not survive the
presence of masses of people with different national conceptions. The mode of
purging, too, involved the persecution of whole communities of people, not just
leaders or cadres.
The fact, however, that these three political genocides took place within little
more than a decade also demands explanation. They occurred in the context of
the cold war, a time of acute polarization between secular ideologies that is
probably unprecedented in world history. The fate, not just of individual nations
but also of the world as a whole, seemed to be at stake. This polarization almost
certainly strengthened the inclination of both sides in these conflicts to demonize
their enemies. It also gave an international dimension to all three political geno-
cides: the perpetrators believed themselves to be eliminating enemies who were all
the more dangerous for the powerful allies that they were believed to have on the
other side of the cold war divide. In the case of Indonesia, the awareness of
the Indonesian military that they had the backing of the United States played a
role in the confidence with which they acted.41 In the early twenty-first century,
the ideological polarization that marked the twentieth century appears distinctly
muted, replaced by a religious polarization that few in the middle of the twentieth
century would have considered possible. Political genocides may turn out to
be a phenomenon of a rather specific phase in human history, but it would be
distressing if the twenty-first century turns out to be one in which the religious
dimension of genocide needs urgent attention.
Becker, Elizabeth, When the War was Over: The Voices of Cambodia’s Revolution and its
People (New York: Simon and Schuster, 1986).
Chandler, David P., The Tragedy of Cambodian History: Politics, War, and Revolution since
1945 (New Haven, CT: Yale University Press, 1991).
41 Bradley R. Simpson, Economists with Guns: Authoritarian Development and U.S.-Indonesian

Relations, 1960–1968 (Stanford: Stanford University Press, 2008).
Cribb, Robert (ed.), The Indonesian killings of 1965–1966: Studies from Java and Bali
(Clayton, Vic.: Monash University Centre of Southeast Asian Studies, Monash Papers
on Southeast Asia 21, 1990).
—— ‘Genocide in Indonesia, 1965–1966’, Journal of Genocide Research 3:2 (2001), 219–39.
Harding, Harry, ‘The Chinese State in Crisis, 1966–9’, in Roderick MacFarquhar (ed.), The
Politics of China: The Eras of Mao and Deng, 2nd edn (Cambridge: Cambridge University
Press, 1997), 148–247.
Kiernan, Ben, The Pol Pot Regime: Race, Power, and Genocide in Cambodia under the Khmer
Rouge, 1975–1979, 2nd edn (New Haven, CT: Yale University Press, 2002).
MacFarquhar, Roderick, and Michael Schoenhals, Mao’s Last Revolution (Cambridge, MA:
Robinson, Geoffrey, The Dark Side of Paradise: Political Violence in Bali (Ithaca, NY: Cornell
Roosa, John, Pretext for Mass Murder: The September 30th Movement and Suharto’s Coup
d’Etat in Indonesia (Madison: University of Wisconsin Press, 2006).
Walder, Andrew G., and Yang Su, ‘The Cultural Revolution in the Countryside: Scope,
Timing and Human Impact’, China Quarterly 173 (2003), 75–99.
Yang Su, ‘Mass Killings in the Cultural Revolution: A Study of Three Provinces’, in Joseph
Esherick, Paul Pickowicz, and Andrew George Walder (eds), The Chinese Cultural
Revolution as History (Stanford: Stanford University Press, 2006), 96–123.
chapter 23
.............................................................................................
S TAT E-S PON SORED

VIOLENCE AND
SECESSIONIST
REBELLIONS
IN ASIA
.............................................................................................
geoffrey robinson
................................................................................................................
The second half of the twentieth century was marked by the phenomenon of
state-sponsored violence against secessionist rebellions. That was certainly true in
Asia where newly independent states, including India, Sri Lanka, Pakistan, Burma
(Myanmar), Thailand, Indonesia, and the Philippines, all sought to quell one or
more armed movements for autonomy or independence by resort to violence.
In most of those places, moreover, state-sponsored violence against armed
secessionist movements has continued with only slight interruption into the
twenty-first century.
This chapter examines comparatively four instances of such violence. Focusing
on East Pakistan, the Karen areas of Burma, West Papua in Indonesia, and East
Timor, it begins with an empirical account of each case, asking: what were the
origins and dynamics of the violence, who were the perpetrators, and who were the
state violence and secessionist rebellions in asia 467
victims? In addressing these questions, it also considers how these events have been
analysed. What terms and concepts have been used to describe the violence, how
consistently have they been applied, and how useful are they in understanding what
happened? Building upon these accounts, the chapter considers whether there are
discernible patterns in these four cases and, if so, what they might signify. What can
they tell us, for example, about broader theories of genocide and violence, about
the role of states, of ideology, and of ‘underlying’ cultural, social, and economic
tensions in genocide?
In its simplest form, the argument here is that the violence against secessionist
movements in all of these places was shaped by the confluence of powerful historical
trends in the post-war period, including the reification of ethnicity as a political
category during and immediately after the colonial period; the legitimacy accorded
the use of violence both as a tool of liberation and as a means to maintain political
order; the rise of military-dominated authoritarian states in the decades after the
Second World War; the support of those regimes, and their use of violence, by powerful
Western states; the generally permissive normative legal and political environment
particularly in the context of the cold war; and the hegemonic discourse and practice of
‘development’ and ‘modernization’ that have characterized the entire post-war period.
At the same time, I argue that state violence in these cases did not flow inevitably
from such broader historical trends, but was invariably shaped by the specific
attitudes, policies, and practices of both secessionist movements and state autho-
rities, and by the historical dynamic of the conflict between them. After all, there
were places where these broad structural conditions did not give rise to secessionist
movements, or where states responded to them in different ways. The violence that
might reasonably be called ‘genocidal’, moreover, generally did not begin with
genocidal intentions. Rather, it was the result of a process of escalation, and of the
development over time of distinctive repertoires and institutional cultures among
states and secessionist movements that together made extreme violence and genocide
more likely. Finally, I argue that, with some exceptions, accounts of the violence in
these cases—and in particular their designation as genocide or not—have been
shaped less by an objective accounting or analysis of the facts of each case, than by
considerations of their political significance to major world powers, and by their fit
with prevailing, but always shifting, political and scholarly agendas.
E A S T PA K I S TA N
................................................................................................................
Over several months in 1971, East Pakistan became the site of widespread,
targeted violence that some scholars and other observers have described as
468 geoffrey robinson
genocide.1 The violence left an estimated one million people dead, and resulted in
the mass exodus of between eight and ten million others to neighbouring India.2 In
addition, tens of thousands of women are said to have been raped in the course of
the conflict, though exact numbers have been impossible to establish, in part
because of the systematic silencing of women’s voices since 1971.3 The violence
ended abruptly in December 1971, with the surrender of Pakistani forces in the
face of Indian military intervention. Shortly thereafter, East Pakistan became the
independent state of Bangladesh.
The vast majority of the victims of this violence were Bengalis, the majority
ethnic group in East Pakistan, with real or alleged supporters of independence
among the principal targets. While many of those killed and displaced were Bengali
Muslims, substantial numbers were Hindus.4 The perpetrators were overwhelm-
ingly members of the Pakistani Army and local militia groups mobilized for the
purpose. That neat picture was complicated, however, by the fact that Bengalis,
including members of the pro-independence Awami League and its armed sup-
porters, the Mukti Bahini, also engaged in acts of violence, including murder and
rape, against those they perceived to be their enemies. The latter group included
the Biharis, an Urdu-speaking ethnic minority group in East Pakistan, and Bengalis
who had sided with Pakistan during the conflict.
As this brief portrait suggests, the violence of 1971 was related to long-standing
cultural and linguistic differences between the peoples of East and West Pakistan.
Geographically separated by some 1,200 miles, the majority of Punjabis in the West
and Bengalis in the East spoke different languages and embraced different cultural
practices, and these differences had helped to generate tensions between East and
West within years of Pakistan’s formation in 1947. But these differences did not
automatically give rise to—and cannot on their own explain—the extreme violence
of 1971. Rather, the violence stemmed from the wider political context of the post-
independence period.
1 See, for example, Rounaq Jahan, ‘Genocide in Bangladesh’, in Samuel Totten et al. (eds), Century
of Genocide: Eyewitness Accounts and Critical Views (New York: Routledge, 2004), Chap. 9. It was also
called genocide at the time. See, for example, Sydney H. Schanberg, ‘Kennedy, in India, Terms
Pakistani Drive Genocide’, New York Times, 18 August 1971.
2 Estimates of the number killed vary widely, from 300,000 to 3 million, but there is now general
agreement that the total was about one million, roughly half of whom died as refugees fleeing the
Pakistani Army. See Mathew White, ‘Death Tolls for the Major Wars and Atrocities of the Twentieth
Century’, [online]. Available at http://users.erols.com/mwhite28/warstat2.htm#Bangladesh.
3 See Nayanika Mookherjee, ‘ “Remembering to Forget”: Public Secrecy and Memory of Sexual
Violence in the Bangladesh War of 1971 ’, Journal of the Royal Anthropological Institute 12 (2006),
433–50.
4 See International Commission of Jurists, The Events in East Pakistan, 1971, [online]. Available at
http://www.globalwebpost.com/genocide1971/docs/jurists/1_preface.htm
A vitally important element of that context was the rising political power of the
Pakistani Army. Although Pakistan remained a parliamentary democracy in form,
by the late 1950s the Army had come to play a central role in national politics. In
1970–1, when the crisis came to a head, the country was led by a former Army Chief
of Staff, Gen. Yahya Khan, the Army was the pre-eminent political power in the
country, and its extreme nationalist ethos of Punjabi supremacy permeated the
state. The dominance of the military at that juncture was critically important in
generating the violence of 1971, primarily because it facilitated and made more
likely the use of excessive force to solve a complex political problem.
The violence of 1971 also stemmed from the conscious articulation of ethnic and
regional differences for political ends after independence in 1947. Political mobili-
zation on the basis of ethnic identity served to entrench differences, and set in
motion the cycle of hostility that led to violence. Very soon after independence,
cultural and linguistic differences became the focus of intense political contesta-
tion, with Bengalis complaining that, despite their numerical superiority in the
country, political and economic power were concentrated in the West, and that
Bengalis were being treated as second-class citizens. These perceptions had a strong
basis in fact, and were the foundation for growing Bengali demands for autonomy
or independence, articulated by the Awami League under the leadership of Sheikh
Mujibur Rahman. The party’s demands, summarized in the latter’s ‘Six Points’,
amounted to a form of confederalism which would have ended West Pakistan’s
dominance of the country. Pressure from the Awami League coincided with
mounting unrest in West Pakistan and insistent demands from other opposition
parties for an end to military rule. In the face of this opposition, Gen. Yahya Khan
promised direct elections. The elections, held in December 1970, resulted in what
would have been an absolute majority for the Awami League in parliament.
Unwilling to countenance such an outcome, in late February 1971 political leaders
in Islamabad indefinitely postponed the convening of parliament.
It was against this backdrop that the violence of 1971 unfolded. The postpone-
ment led quickly to angry demonstrations in East Pakistan, and to some acts of
violence. Then, on the night of 25 March 1971 Pakistani Army forces launched an
unusually heavy-handed crackdown in the East, known as Operation Searchlight.
Using artillery and thousands of ground troops, the initial assault targeted univer-
sity student dormitories and Hindu villages in and around the capital, Dhaka,
leaving hundreds dead. The extreme brutality of that operation prompted Mujibur
Rahman of the Awami League to declare the independence of ‘Bangladesh’ the
following day. Over the next several months, the behaviour of Pakistani forces
aided the rebels in garnering support from the local population, in mobilizing their
own armed force, the Mukti Bahini, and in winning the backing of neighbouring
India.
A central element in the strategy of the Pakistani forces, and one that
led inevitably to the killing of a great many non-combatants, was the classic
counter-insurgency tactic of separating rebel forces from the civilian population. In

practice, that approach led to sweeping assaults on villages suspected of supporting
the rebels. In the course of those operations, entire villages were burned to the
ground, civilians were summarily executed, and women were raped. The violence
was also fuelled by problems of command and control. Pakistan Army soldiers
quickly came to view all Bengalis as the enemy, and acted in anger at reports that
their comrades had been killed or captured by enemy forces. Official Pakistani
inquiries and victims alike concurred that resulting atrocities were partly the result
of a failure by officers to constrain or limit the unlawful actions of soldiers—and
indeed of their encouragement of such actions.5 Beyond these general patterns, it is
clear that particular groups were deliberately targeted by the Army, on the grounds
that they constituted enemies of the state. These included, most notably, members
of the Awami League, students, and Hindus, a great many of whom were killed as
they fled the violence to India.6
Finally, the violence was shaped by the prevailing international context and by
the actions of key states in 1971.7 That environment was broadly conducive to the
exercise of power by military regimes, and to the use of extreme violence against
secessionist movements. This was, after all, the height of the cold war and a time
when President Nixon and Henry Kissinger were engaged in the most cynical use of
military force and secret diplomacy to achieve US political and strategic objectives.
As a US ally, and a crucial interlocutor in the US search for rapprochement with
China, Pakistan had good reason to believe that its harsh repression of the rebels in
East Pakistan would be met with either silence or support by the United States. And
so it was. Despite the internal assessment of US officials in Dhaka (East Pakistan)
that the events constituted ‘genocide’, the Nixon administration publicly denied
any knowledge of atrocities, and provided Pakistan with assurances of its diplo-
matic and military support.8 Pakistan’s leaders were less astute in their assessment
of the likely posture of neighbouring India, which they deemed unlikely to
intervene. In fact, after the Pakistani Army crackdown of March 1971, India
provided substantial political, economic, and military support to the rebels and a
place of refuge for millions of Bengalis, and eventually launched a major military
assault both in the East and the West, forcing Pakistan to surrender in December
1971 and bringing a swift end to the violence.
5 For an official Pakistani account of the atrocities, see The Report of the Hamoodur Rehman
Commission of Inquiry into the 1971 War (Vanguard: Lahore, 2000).
6 See International Commission of Jurists, The Events in East Pakistan, esp. Part II.
7 On the international dimensions of the conflict, see Robert Victor Jackson, South Asian Crisis:
India, Pakistan, and Bangla Desh; A Political and Historical Analysis of the 1971 War (New York: Praeger,
1975); and O. Marwah, ‘India’s Military-Intervention in East Pakistan, 1971–1972’, Modern Asian
Studies 13 (1979), 549–80.
8 See National Security Archive, ‘The Tilt: the U.S. and the South Asian Crisis of 1971 ’, [online].
Available at http://www.gwu.edu/~nsarchiv/NSAEBB/NSAEBB79/.
Given the extent and severity of the violence in East Pakistan, and the evidence
that it was part of a planned military operation by government forces, it is perhaps
not surprising that most Bangladeshis, as well as contemporary observers and some
foreign scholars, have used the term genocide to describe it. And yet, when
compared to some other instances of state-sponsored killing—such as those by
the Khmer Rouge in 1975–9—it is striking how little serious scholarly work has
been devoted to this case since the 1970s. Beachler has argued that this curious
silence stems from two important conditions that have little to do with the violence
itself.9 First, in contrast to the Khmer Rouge killings of 1975–9 which were initially
condemned as genocide by major Western powers,10 the United States in particular
had no interest in drawing critical attention to the violence committed by its ally
Pakistan, and certainly no interest in describing it publicly as genocide. As a
consequence, the idea that this was a genocide simply did not gain traction
among key states or within intergovernmental bodies like the UN. Second, in
contrast to other instances of state-sponsored killing, in which major powers
were known to have been directly complicit, the evidence of direct Western
involvement in the violence in East Pakistan was neither compelling nor well
known.11 As such, after a brief flurry of criticism, the case fell off the intellectual
and political agendas of scholars and activists on the Left. In contrast to some other
cases, moreover, the case was not embraced by the community of genocide scholars
until quite recently. One possible explanation for that lapse is that the field of
genocide studies has been slow to describe as genocide instances of violence that
differ significantly from the Holocaust.12 A final reason for the lack of serious
attention paid to the events of 1971 may be that the secessionists themselves were
armed, and were responsible for acts of violence against minority groups. At least
in public perception, and probably for some scholars, that rendered the attribution
of blame, and the use of the term genocide, more problematic. In short, because the
violence did not resonate in obvious way with the political agendas of Western
states, or with the interests and propensities of scholars, and because it lacked a
simple narrative of good versus evil, it has received far less attention than might
seem warranted.
9 Donald Beachler, ‘The Politics of Genocide Scholarship: The Case of Bangladesh’, Patterns of
Prejudice 41 (2007), 467–92.
10 US characterizations of the killings shifted significantly after the Vietnamese invasion of
Cambodia in 1979. For many years thereafter, the United States worked with China and Thailand to
minimize the Khmer Rouge brutality in order to blacken Vietnam’s reputation.
11 Some evidence of US complicity was revealed by Seymour Hersh in the early 1980s, but his work
was an exception to the rule. Seymour M. Hersh, The Price of Power: Kissinger in the Nixon White
House (New York: Summit Books, 1983).
12 I am grateful to Robert Cribb for this insight.
THE KAREN IN BURMA

................................................................................................................
Whereas the state violence in Pakistan in 1971 targeted a single opponent, and the
conflict was resolved quickly in favour of the secessionists, in neighbouring Burma
(Myanmar) the state has battled dozens of different armed secessionist movements
since independence in 1948, and none of those conflicts have been resolved in favour
of a rebel group. Among the many ethno-linguistic groups that have taken up arms
against the government, the largest and most tenacious has been the Karen, who
currently number some five to seven million (among Burma’s total population of
47 million), most of whom live in the hilly eastern areas of the country bordering
Thailand.13 Under the leadership of the Karen National Union (KNU), and its
armed wing, now called the Karen National Liberation Army (KNLA), they have
been fighting for autonomy or independence in some form since 1948.14 Although
exact figures have not been established, human rights organizations and scholars
have estimated that tens of thousands of Karen civilians have died as a consequence
of government military campaigns since then, and more than one million others
have been forcibly displaced, or subjected to torture, forced labour, and sexual
violence.15
In a superficial sense, the conflict and violence have their roots in the ethnic and
cultural differences that exist between the Karen minority and the Burman major-
ity that has controlled the state since 1948.16 As in the case of Pakistan, however,
Karen ethnic identity has been at least as much the product of a history of political
contestation—and of violence itself—as a ‘natural’ or primordial fact of life. As in
Pakistan, for example, the violence in Burma has stemmed in part from the
imposition of colonial and post-colonial boundaries that sought to incorporate
territories and peoples never previously governed as one citizenry.17 Before the
British established full colonial rule in Burma in 1886, the Karen had seldom been
13 Human Rights Watch, ‘They Came and Destroyed Our Village Again’: The Plight of Internally
Displaced Persons in Karen State (New York, 2005), 16.
14 See Martin J. Smith, Burma: Insurgency and the Politics of Ethnicity (London: Zed Books, 1999),
and Hazel J. Lang, Fear and Sanctuary: Burmese Refugees in Thailand (Ithaca, NY: Cornell Southeast
Asia Program Publications, 2002).
15 For eyewitness accounts and analysis of the killings and other abuses, and for casualty estimates,
see Human Rights Watch, ‘They Came and Destroyed Our Village Again’; Amnesty International,
Burma: Extrajudicial Execution and Torture of Members of Ethnic Minorities (London, 2001); and
Martin Smith and Annie Allsebrook, Ethnic Groups in Burma: Development, Democracy and Human
Rights (London: Anti-Slavery International, 1994).
16 On Karen culture, see Charles F. Keyes, Ethnic Adaptation and Identity: The Karen on the Thai
Frontier with Burma (Philadelphia: Institute for the Study of Human Issues, 1979).
17 Ananda Rajah, ‘A “Nation of Intent” in Burma: Karen Ethno-nationalism, Nationalism and
Narrations of Nation’, Pacific Review 15:4 (2002), 517–37.
part of a political entity called ‘Burma’.18 Quite the contrary, like many ethnic
groups now located inside Burma’s modern border—such as the Shan, Chin,
Kachin, Mon, Kayah, and Karenni—the Karen had for many years belonged to
relatively autonomous political power centres, as often as not hostile to the rulers
of various Burman kingdoms based in the lowlands.19 Indeed, their political
distance from those kingdoms led many Karen to welcome the British conquest
of Burma, to embrace the message of Protestant missionaries, and to become loyal
supporters of British rule. The British repaid Karen loyalty by recruiting them into
the colonial army, and granting them a significant measure of self-rule. Their close
ties to British authority rendered the Karen objects of suspicion and hatred by pre-
war Burman nationalists. That enmity was exacerbated during the Second World
War as the nationalists fought on the side of the invading Japanese armies,
ruthlessly attacking Karen who had remained loyal to the British.20 This longer
history of conflict helped to cement a sense of Karen identity while giving rise to
anxiety about the position of Karens in an independent Burma. It was also one of
the key reasons, when autonomy was not granted by the new regime, the Karen
took up arms against the state.
Nevertheless, on their own, the existence of a distinct identity and a history of
enmity cannot explain the violence that has characterized the state response to
Karen demands since independence. As in the case of Pakistan, the explanation lies
at least in part in the character of the post-colonial Burmese state, and the central
role of the Army in the process of state formation.21 Like Pakistan, Burma gained
its independence in 1948 through negotiation with the British, and for the next
decade or so, it was governed by a civilian political elite that had embraced
parliamentary democracy. By the late 1950s, however, growing dissatisfaction
with civilian rule—and particularly its inability to handle various rebel move-
ments—encouraged the Burmese Army to intervene in political life, and in 1962 to
seize power outright in the name of national unity and order. Notwithstanding
some changes in the name of the regime, the military has remained in power ever
since. Certain general features of Burma’s military state—including its extreme
intolerance of dissent, its insistence on national unity, and its reliance on force to
answer any challenge to its authority—have made violence against the Karen (and
other rebel groups) virtually inevitable.
18 On the colonial history of Burma, see Hugh Tinker, The Union of Burma (London: Oxford
University Press, 1967), 1–33; and Lang, Fear and Sanctuary, 26–31.
19 On this point, see Keyes, Ethnic Adaptation and Identity; and Lang, Fear and Sanctuary, 26–9.
20 Tinker, Union of Burma, 9–10.
21 Callahan has argued, for example, that Burma’s state has acted as it has, in part, because it has
been dominated by ‘war fighters’ who have adopted coercion-intensive institutions in the process of
state formation. Mary P. Callahan, Making Enemies: War and State Building in Burma (Ithaca, NY:
Cornell University Press, 2003).
The particular character of state violence against the Karen has also been shaped
by specific aspects of Burmese military doctrine, policy, and practice. Against the
Karen, and other armed insurgents, the Burmese Army (the Tatmadaw) has
employed a counter-insurgency strategy known as the ‘four-cuts’, formally adopted
in 1968. Aimed at separating guerrillas from their civilian support base by ‘cutting’
their access to food, intelligence, recruits, and finances, in practice this strategy has
resulted in the subjection of civilians to systematic violence, terror, and forcible
displacement.22 The violence has also been fuelled by the government’s heavy-
handed approach to economic ‘development’ in Karen and other minority areas. In
the name of development projects aimed primarily at extracting oil and other
natural resources—and with the acquiescence of large transnational corporations
like Unocal and Total—the authorities have forcibly displaced the populations of
entire Karen villages. They have also conscripted tens of thousands of Karen
villagers to work without compensation and under appalling conditions on infra-
structure projects, such as road and pipeline construction, or as porters for military
units.23 These tendencies have been further compounded by a culture of impunity
that has developed within the Tatmadaw over more than half a century. Recogniz-
ing that they are completely beyond the reach of the law, and seeing that officers
condone and even encourage brutal behaviour, ordinary soldiers have little incen-
tive to act in a humane way.
As in East Pakistan, it is widely accepted that the extreme brutality of the
Tatmadaw has served to heighten the antagonism of Karen and other minorities
towards Burmans and the Burmese state.24 That hostility has helped to reinforce
demands for independence, and to sustain recruitment efforts, despite what appear
to be very dim prospects for victory. The resilience of the resistance has also been
sustained by the economic resources controlled by the Karen and some other
minority groups. Most important, by virtue of their geographical position along
the border between Burma and Thailand, for long periods they have been able to
control a black market cross-border trade in valuable products including timber,
gems, arms, and drugs.25
Finally, the contours and trajectory of violence against the Karen have been shaped
significantly by the wider regional and international political context within which
they have occurred. In contrast to their resort to partition to address competing
22 On the four-cuts strategy and its consequences for civilian populations, see Lang, Fear and
Sanctuary, 37–43; and Smith and Allsebrook, Ethnic Groups in Burma, 78–84.
23 On the problems of development and forced labour, see Human Rights Watch, ‘They Came and
Destroyed our Village Again’; and Karen Human Rights Group, Forced Labour Orders since the Ban:
A Compendium of SPDC Order Documents Demanding Forced Labour since November 2000 (Bangkok,
2002).
24 Lang, Fear and Sanctuary, 42–3.
25 Mary P. Callahan, Political Authority in Burma’s Ethnic Minority States: Devolution, Occupation
and Coexistence (Washington, DC: East-West Center, 1997); Lang, Fear and Sanctuary, 41.
political demands in the subcontinent, in Burma the British managed to deflect calls
for the creation of separate ethnic or religious states. Like virtually all post-colonial
political elites, moreover, Burma’s new leaders accepted the colonial boundaries as the
basis for the new nation-state, and sought to maintain those boundaries at all costs.
That task was rendered more difficult not only by the emergence of armed ethnic rebel
groups throughout the country, but also by the arrival on Burmese soil of hundreds of
thousands of Kuomintang (KMT) troops from China in 1949. Supported by the
United States and its allies, those troops soon became involved in the illicit trade of
weapons and drugs, not only helping to destabilize the civilian regime in Rangoon,
but also militarizing the largely minority areas bordering China and Thailand.
More broadly, from the 1940s to the early years of the twenty-first century, the
actions of key powers encouraged the resort to arms both by rebels and by the state,
and facilitated a pattern of extreme brutality against civilian populations on the part
of Burmese forces. During the cold war, for example, key Western states either
encouraged or acquiesced in the establishment of military regimes, and turned a
blind eye to the abuses they committed in the name of security and order. Beginning
in the late 1980s, Western governments began to criticize Burma’s rulers for their
systematic violations of human rights, but their acquiescence in those abuses over
almost three decades had already led to the deep militarization of ethnic minority
areas, both through the presence of Burmese troops there and through the routine
resort to military means by the ethnic groups themselves. Meanwhile, major trans-
national corporations continued to do business with the regime, thereby facilitating
the continuation of state violence. The violence against the Karen and other minor-
ities has also continued because some key Asian states, most conspicuously China,
but also the regional grouping known as ASEAN, have preferred to maintain cordial
ties with the regime, and have consequently remained largely silent on the question.26
Discussion of state violence against the Karen has been largely excluded from
more general studies of genocide. Apart from Karen partisans, moreover, those
most familiar with the case have not generally described the violence as genocidal.
More commonly, it has been portrayed as part of a general problem of militariza-
tion and systematic human rights abuse in Burma. The reluctance to use the term
genocide in this instance is in some respects understandable. For one thing, the
numbers of Karen killed, even if large in proportion to the group’s population,
appear relatively small in comparison with some of the better known genocides of
the twentieth century. Moreover, international human rights organizations, like
Human Rights Watch and Amnesty International, which have been at the forefront
of efforts to document the violence against Burma’s ethnic minorities, have gener-
ally adopted a very strict, legalistic, definition of the term, and consequently
seldom apply it even in cases of widespread and systematic violence.
26 Lang, Fear and Sanctuary, 45.

At the same time, it is difficult to avoid the conclusion that other factors have
played a role. One problem is that for much of the past fifty years, powerful
Western states have had no compelling political interest in the Karen that would
warrant calling the attacks upon them a genocide. Their near silence has undoubt-
edly served to limit public interest in the matter, and may also have kept scholarly
attention to a minimum. The few scholars who have examined the case, moreover,
have quickly found that it is a very complicated story. The Karen, after all, are only
one of many ethnic groups that have been targeted by the state of Burma, and like
the secessionists in East Pakistan, they are not entirely blameless in the cycle of
violence. Further complicating matters, there have been serious conflicts among
Karen over matters of strategy and leadership. In short, the evidence does not lend
itself to an easy designation of good and bad actors.
W E S T PA P UA
................................................................................................................
State violence against secessionist groups has also been a recurring theme in
Indonesia, an archipelagic state comprising scores of ethno-linguistic groups, and
considerable religious diversity. That pattern has been especially clear in West
Papua where, in a population of just over two million, between 10,000 and
30,000 civilians have been killed and countless others displaced by Indonesian
military campaigns and ‘development’ efforts over the past fifty years.27 Since the
1960s, West Papua (also known as West New Guinea, West Irian, and Irian Jaya) has
also been the site of heavy-handed state initiatives of cultural assimilation and
‘Indonesianization’.
As in Pakistan and Burma, Papuan secessionists have typically justified their
demands for independence on the basis of claimed ethnic and cultural distinc-
tiveness from the rest of Indonesia, and especially its dominant ethnic group, the
Javanese. As elsewhere, however, contemporary Papuan identity is, at least in part,
the product of European colonial rule, and of Indonesian state policy and prac-
tice.28 In the final negotiations over Indonesian independence in 1949, the Dutch
insisted that Papua should remain under their control, arguing that the Papuans
27 On the violence in West Papua, see Robin Osborne, Indonesia’s Secret War: The Guerilla Struggle
in Irian Jaya (Sydney/Boston: Allen and Unwin, 1985). For estimates of numbers killed, see Mathew
White, ‘The Lesser Unpleasantries of the Twentieth Century’, [online]. Available at: http://users.erols.
com/mwhite28/warstat5.htm#Irian62.
28 On the historical development of Papuan identity, and the Organisasi Papua Merdeka, see David
Webster, ‘ “Already Sovereign as a People”: A Foundational Moment in West Papuan Nationalism’,
Pacific Affairs 74:4 (2001–2), 507–28.
were entitled to develop towards a separate independence by virtue of their very

different appearance from most Indonesians. Over the next decade or so, Dutch
authorities encouraged a sense of Papuan cultural distinctiveness and contributed
to the emergence of a fledgling Papuan political identity.29 Nevertheless, by the
early 1960s that shared identity was still limited to a rather small group of educated
Papuans. Dispersed through some of the least accessible terrain in the world, with
little in the way of communications or common historical experience to bind them,
most of the population did not yet share that vision.
That pattern began to change significantly through the 1960s as the Dutch, facing
considerable international pressure, relinquished control of Papua to Indonesia. The
de facto transfer was effected under UN auspices in 1962–3, and then formalized
through a deeply flawed, but UN-approved, plebiscite in 1969.30 The 1960s saw the
start of the first serious demands for independence and the emergence of an armed
rebel movement, officially named the Organisasi Papua Merdeka (OPM) in 1969. The
OPM emerged largely in response to the Indonesian takeover, and with the recogni-
tion that peaceful strategies had failed. Significantly, the 1960s also saw the consoli-
dation of the power of the Indonesian Army, and with it the formation of a state,
known as the New Order, whose leadership was notoriously intolerant of dissent,
and obsessed with national unity, stability, and security. It adopted the mantra of
‘development’ as its central rationale, and routinely used force to deal with its real or
perceived enemies.31 Needless to say, the OPM—officially described as a ‘security
disruptors movement’—was among those deemed to represent an unacceptable
threat to national stability and unity, and on those grounds subjected to concerted
campaigns of violence by state forces over the next three decades.
The persistence and the appeal of the OPM since then has stemmed, in large
part, from the harshness of the measures taken by the central state to repress the
rebels. Beginning in the late 1960s, Indonesian military forces deployed in Papua
adopted a repertoire of counter-insurgency tactics that had been developed in
other operations—most notably against the Darul Islam movement in the 1950s,
and the Indonesian Communist Party (PKI) in 1965–6. These included the mobili-
zation of local militia forces as Army proxies, the systematic use of torture and
harsh interrogation against suspected OPM supporters, and attacks on settlements
thought to be supporting or protecting rebel fighters.32 Over time, the negative
29 On the colonial period, see C. L. M. Penders, The West New Guinea Debacle: Dutch
Decolonisation and Indonesia, 1945–1962 (Honolulu: University of Hawai’i Press, 2002).
30 On the Act of Free Choice and the role of the UN, see John Saltford, The United Nations and the
Indonesian Takeover of West Papua, 1962–1969 (London/New York: Routledge, 2003).
31 On the history of the Indonesian Army, see Harold Crouch, The Army and Politics in Indonesia
32 For accounts of these military campaigns and their consequences, see Osborne, Indonesia’s Secret
War; and Amnesty International, Indonesia: Continuing Human Rights Violations in Irian Jaya
(London, 1991).
effects of these tactics have been exacerbated by what I have called the ‘institutional
cultures’ of impunity and terror that have become deeply embedded in the
Indonesian Army and its proxies.33 Like the Tatmadaw in Burma, then, through
its systematic brutality the Indonesian military has actually helped to create in
Papua the enemy it seeks to crush.
Indonesia’s presence has been rendered even more odious to Papuans by the
routine cultural arrogance of government officials, as evidenced in state efforts to
‘Indonesianize’ the population. Describing Papuans as backward or stone-age
peoples in need of civilization, Indonesian authorities have adopted both the
language and the policies of a colonial authority.34 In the early 1960s, for example,
Indonesia’s foreign minister explained that the government’s policy towards the
people of West Papua was ‘to get them down out of the trees, even if we have to pull
them down’.35 To that end, campaigns have been launched since that time to
compel local people to wear Western-style clothing, to teach them the Indonesian
language, and to settle nomadic populations in villages and towns.36 In the name of
development, moreover, Indonesian authorities have energetically pursued the
colonial-era project of ‘transmigration’ under which people from the more densely
populated islands of Java and Bali have been transferred to Papua, ostensibly to
make it more productive, but also to instruct local people in wet-rice agricultural
techniques of those regions.37 Finally, in search of tax and other revenue to fill state
coffers, the state has parcelled out to transnational corporations the rights to
Papua’s vast mineral resources, which include some of the richest gold and copper
deposits in the world. The huge mining operations undertaken by Freeport have
become the focus of intense criticism by Papuans and others, who have complained
that they have sullied the natural habitat, and provided a pretext for local officials
and soldiers to run protection rackets and other illicit schemes, while offering little
in the way of jobs or other economic benefit to the local community.38
As in the case of East Pakistan and of the Karen in Burma, the dynamic of
violence in Papua has also been influenced by the wider international context. As
noted earlier, the emergence of a Papuan identity and demands for independence
were to some extent the result of Dutch colonial policy. UN acquiescence in the
plebiscite of 1969, for example, arguably made the OPM’s resort to violence much
more likely, while depriving the movement of a strong international legal basis for
33 See Geoffrey Robinson, ‘If You Leave Us Here, We Will Die’: How Genocide Was Stopped in East
Timor (Princeton: Princeton University Press, 2010), ch. 3.
34 J. Pouwer, ‘The Colonisation, Decolonisation and Recolonisation of West New Guinea’,
Journal of Pacific History 34 (1999), 157–79.
35 Cited in Osborne, Indonesia’s Secret War, 136.
36 See D. Gietzelt, ‘The Indonesianization of West Papua’, Oceania 59 (1989), 201–21.
37 On transmigration and its critics, see Osborne, Indonesia’s Secret War, 125–36.
38 See, for example, D. Leith, ‘Freeport and the Suharto regime, 1965–1998’, Contemporary Pacific 14
(2002), 69–100.
its independence struggle. Although some small regional states like Vanuatu have
shown a measure of sympathy for rebel demands, their influence has been over-
shadowed by key Western states whose silence in the face of Indonesian violence
has virtually guaranteed its continuation. Viewing the New Order regime as an
important bulwark against communism at the height of the cold war, as a vital
economic prize and, after 2001, as a valuable ally in the so-called war on terror,
those states have been reluctant to do or say anything that might upset good
relations with Indonesian authorities. In this regard, Australia has played an
especially dubious role, denying virtually all claims for political asylum by Papuans,
and doing little else to constrain Indonesian violence.39
Given its long duration and its dire consequences for the population, it is
striking that the violence in West Papua has seldom been described as genocidal
or included in general studies of genocide. States, scholars, and international
human rights organizations have tended instead to portray the violence as a
problem of the violation of human rights, as a denial of the right to self-determi-
nation, or as forced cultural assimilation. As in the case of the Karen, the reluctance
to use the term genocide may simply reflect the fact that the absolute numbers of
people killed in West Papua has been small in comparison to other cases. But, if
East Pakistan and Burma are any guide, it is probably also because major Western
powers have had no compelling political interest in highlighting the plight of
Papuans. On the contrary, as noted above, it has been in their interest to maintain
cordial relations with Indonesia, and so to remain silent on the subject of Papua.
That silence has gone largely unchallenged by genocide scholars who appear to be
only dimly aware of the existence of West Papua, or of the state-sponsored violence
that has been committed there for more than 40 years. International church and
NGO networks have campaigned energetically to draw attention to the problem of
human rights abuse and forced cultural assimilation in the territory, but they have
been unable to generate the kind of political pressure that alone might bring a real
change in the posture of major powers.
EAST TIMOR
................................................................................................................
Among the cases reviewed here, East Timor stands out as perhaps the most
notorious instance of state-sponsored violence against an independence
39 R. R. Premdas, ‘The Organisasi Papua Merdeka in Irian Jaya—Continuity and Change in Papua
New Guinea’s Relations with Indonesia’, Asian Survey 25 (1985), 1055–74. Also see F. K. Kalidjernih,
‘Australian Indonesia-Specialists and Debates on West Papua: Implications for Australia-Indonesia
Relations’, Australian Journal of International Affairs 62 (2008), 72–93.
movement. A Portuguese colony for more than 300 years, East Timor was invaded
by neighbouring Indonesia in December 1975, and occupied until 1999 when some
eighty per cent of the population voted in favour of independence in a UN-
supervised referendum. In the course of the twenty-four-year occupation, at least
100,000 people died, most of them as a result of hunger and disease brought about
by Indonesian policies.40 Indonesian authorities have insisted that the figures have
been inflated and that any casualties were the unavoidable consequence of war. By
contrast, most independent observers have concluded that the deaths were the
direct and predictable consequence of Indonesian policy and practice, and some
have described the outcome as genocide.41
As in the other cases examined here, the conflict and violence in East Timor were
in important respects a legacy of European colonial rule. Although the population
spoke, and still speaks, at least a dozen different languages and had some history of
internal political division, by the time the Portuguese announced their plans to
leave in 1974, a small but influential elite had come to view East Timor as having a
distinct political identity which justified its designation as an independent nation-
state.42 Yet that sense of national identity was not by itself a reason for conflict with
Indonesia, or for the appalling violence that followed. Nor did the conflict stem
from the fact that East Timor’s population was predominantly Catholic or animist
while Indonesia’s was mainly Muslim. In fact, before 1975 there was no indication
of any serious religious or cultural hostility between the peoples of East Timor and
Indonesia. It was only after the December 1975 invasion, and the direct experience
of Indonesian rule, that a consciousness of such differences spread widely among
the population. In other words, the idea of an East Timorese national identity was
reinforced, and the will to resist was deepened, not by the prior existence of
‘primordial’ differences, but by the experience of Indonesian rule itself.43
The dynamic of violence in East Timor from 1975 to 1999 was in many respects
very similar to the pattern observed in other areas under Indonesian rule. As in
West Papua, for example, the Indonesian military described the independence
movement as ‘security disruptors’ or terrorists, and sought to cut them off from
the civilian population. Similarly, the tactics employed to achieve that end entailed
the forced displacement and relocation of the civilian population, the deliberate
40 See Amnesty International, East Timor–Violations of Human Rights. (London, 1985); and East
Timor Commission for Reception, Truth, and Reconciliation (CAVR), Chega! (Dili, 2006), [online].
Available at: http://www.cavr-timorleste.org/en/chegaReport.htm.
41 See, for example, James Dunn, Timor: A People Betrayed (Sydney: ABC Books, 1996); and Ben
Kiernan, Genocide and Resistance in Southeast Asia: Documentation, Denial and Justice in Cambodia
and East Timor (New Brunswick: Transaction Publishers, 2008).
42 On the colonial period and the origins of the nationalist movement in East Timor, see John
Taylor, Indonesia’s Forgotten War: The Hidden History of East Timor (London: Zed Books, 1991); and
José Ramos-Horta, Funu: The Unfinished Saga of East Timor (Trenton, NJ: Red Sea Press, 1986).
43 On this point, see Robinson, ‘If You Leave Us Here, We Will Die’.
destruction of villages, crops, and livestock, and the torture, rape, arbitrary im-
prisonment, and summary execution of suspected supporters of independence.
The victims of these campaigns were overwhelmingly civilians living in rural areas.
Most died of starvation and disease in the first five years of the occupation, but a
considerable number were killed, or died following torture or arbitrary detention,
by Indonesian forces and their proxies.44
As in the other cases discussed here, the violence was fuelled by the distinctive
doctrines, practices, and attitudes of military authorities. The Army’s doctrine of
‘total peoples’ defense’, for example, was the basis for the mobilization of local
militia forces, which spread violence more widely among the population.45 Mean-
while, the deeply rooted pattern of impunity, and the ‘institutional culture of
terror’ within the Indonesian Army and its proxies, ensured the continued use of
torture, rape, and other forms of terror against presumed supporters of indepen-
dence. The clearest manifestation of that pattern, but hardly the only one, was the
Santa Cruz massacre of 12 November 1991, in which some 270 young people were
killed by Indonesian Army troops who opened fire on a peaceful pro-independence
demonstration at a cemetery in Dili. The cycle of violence was further accelerated
by the arrogant attitudes and policies of Indonesian government authorities, who
tended to look upon the East Timorese as backward and in need of civilization. To
that end, they imposed a variety of schemes aimed at Indonesianizing the popula-
tion—including birth control, Indonesian language education, and transmigra-
tion—which East Timorese regarded with deep suspicion and hostility, and which
had the effect of deepening their sense of identity and grievance.
Finally, as in the other cases discussed here, the violence in East Timor was
shaped in significant ways by the international context in which it occurred. Most
obviously, it was facilitated, even encouraged, by the support New Order autho-
rities received from key states, notably the United States, the UK, and Australia.46
The 1975 invasion, for example, took place less than 24 hours after Indonesian
President Suharto met US President Ford and Secretary of State Kissinger, and
received from them an unequivocal green light to proceed.47 After the invasion, the
United States and other states provided vital diplomatic, economic, and military
44 The most complete account of these tactics and their consequences is in CAVR, Chega!
45 On the history of this doctrine, and of the militias, see Geoffrey Robinson, ‘People’s War: Militias
in East Timor and Indonesia’, South East Asia Research 9:3 (2001), 271–318.
46 The best analysis of the role of the international community in aiding and abetting the violence
in East Timor is Joseph Nevins, A Not-So-Distant Horror: Mass Violence in East Timor (Ithaca, NY:
47 The memorandum of that conversation and other US government documents related to the
Indonesian invasion can be found at National Security Archive, ‘East Timor Revisited: Ford, Kissinger
and the Indonesian Invasion, 1975–76’, [online]. Available at: http://www.gwu.edu/~nsarchiv/
NSAEBB/NSAEBB62/. On Australia’s role, see Wendy Way (ed.), Documents on Australian Foreign
Policy: Australia and the Indonesian Incorporation of Portuguese Timor, 1974–1976 (Canberra:
Department of Foreign Affairs and Trade, 2000).
assistance to Indonesia, and played an active role in silencing the evidence of

violence that made its way out of the territory. Some of that assistance was
provided at precisely the time that Indonesian forces were committing their most
egregious acts of violence against the civilian population. The devastating military
campaigns of the late 1970s, for instance, entailed the use of sophisticated weapons
systems, including counter-insurgency aircraft (OV-10 Broncos), napalm, and
high-powered weapons, provided by the United States and other Western allies.
It is safe to say that without such support neither the invasion nor the genocide
would have happened.
If in these ways East Timor’s experience was similar to the other cases considered
here, there were some respects in which it was very different. For example, whereas
secessionist movements elsewhere stuck steadfastly to the use of violent means in
pursuit of their objectives, from the mid-1980s East Timor’s resistance gradually
shifted towards a strategy that emphasized diplomacy, clandestine organization,
and non-violent demonstrations. And where others continued to use the language
and symbolism of ‘national liberation’, East Timor’s leaders began increasingly to
speak the language of universal human rights.48 Together, these shifts helped to win
the movement significant international support, and to ensure that East Timor’s
cause received a sympathetic hearing, not only among human rights NGOs and
solidarity groups, but also within the Catholic Church, at the UN, and among
legislators and policy-makers in major states. The cause received a further boost in
1996 when two East Timorese were awarded the Nobel Prize for Peace. Despite its
small size, by the late 1990s these developments had made the problem of East
Timor an issue powerful Western states could no longer ignore.
East Timor’s contested international legal status also helped to disrupt the cycle
of violence, and eventually to end it. In contrast to West Papua, where the UN
ratified the 1969 plebiscite transferring formal sovereignty to Indonesia, East
Timor’s ‘incorporation’ as an Indonesian province was never recognized by the
UN. That important legal issue ensured that East Timor remained a matter of
international dispute throughout the period of the occupation, providing a strong
basis for those demanding independence, and highlighting the violence there. The
singular importance of this wider context became evident in 1998–9, when mount-
ing international pressure at a time of domestic economic and political crisis in
Indonesia prompted the country’s new President to propose a UN-supervised
referendum on East Timor’s political status. It became even clearer in September
1999, when escalating violence in the aftermath of the vote led some to predict a
second genocide. At that critical juncture, major powers that had for years turned a
blind eye to Indonesian violence in East Timor chose instead to authorize an armed
48 On these shifts see Robinson, ‘If You Leave Us Here, We Will Die’, esp. ch. 4.
intervention to end it. That move, in turn, paved the way for the withdrawal of
Indonesian forces, and for East Timor’s formal independence in 2002.49
Finally, East Timor differs from the other cases discussed here in the way it has
been portrayed by outside observers and scholars. Like the Pakistani Army’s
campaign in East Pakistan, Indonesia’s war in East Timor in the late 1970s was
quickly described by nationalists and a handful of sympathetic scholars and human
rights activists as genocidal. These outside observers, for the most part on the Left,
saw in East Timor a perfect example of what they considered to be the general
political and moral failings of Western governments. Yet, whereas scholars soon
lost interest in the case of East Pakistan (or began to question the applicability of
the term genocide in that case), interest in East Timor grew steadily, and it began to
be included in general studies of genocide.50 The shift was due, in part, to a
growing public awareness of conditions in the territory, particularly after the
Santa Cruz massacre of 1991, and the wealth of credible information that became
available for scholarly examination. But it also stemmed from the fact that by the
1990s—in marked contrast to all of the other cases discussed here—the East Timor
issue had become a subject of serious political concern among major powers and
within the UN. In large part, that was because of the unusual success of the
resistance and of international solidarity groups in highlighting the direct com-
plicity of those states in the violence in East Timor; nothing even remotely similar
had been achieved in East Pakistan or the other two cases. Finally, it seems likely
that Indonesia’s campaign of violence in East Timor has come to be described as
genocide because, over the years, the resistance movement and its sympathizers
abroad managed to develop and disseminate a compelling narrative of the conflict,
in which East Timorese were portrayed entirely as courageous victims, and the
Indonesian state as wholly evil. It is now clear that that portrait was an oversimpli-
fication, but the narrative is well established, and will not easily be rewritten.
DISCUSSION
................................................................................................................
At first glance, the experiences of state-sponsored violence in East Pakistan, Burma,
West Papua, and East Timor seem to display a bewildering diversity. There were
marked differences, for example, in the ethnic and cultural character of the seces-
sionist movements, in the duration and extent of the violence, in the numbers killed
49 For a detailed account of the violence surrounding the 1999 referendum, and the reasons for
international intervention, see ibid.
50 See, for example, Totten et al. (eds), Century of Genocide ; and Kiernan, Genocide and Resistance
in Southeast Asia.
and otherwise affected, and in the ways in which the violence has been portrayed.
A closer examination, however, reveals what appear to be significant similarities
among these cases, most notably in the historical conditions that shaped the
secessionist movements and the states that sought to crush them, in the character
of the states committing the violence, in the broader international environment
within which the violence occurred, and in the processes through which political
conflict turned to violence, and violence itself escalated. It also reveals some
intriguing patterns in the ways that scholars and others have portrayed the violence.
While recognizing that these cases are in many respects unique, my sense is that,
viewed together, they may offer insights into the study of state-sponsored violence
and genocide more generally.
First, they highlight important caveats about the relevance of cultural and ethnic
identity in the dynamic of secessionist and state violence. Notwithstanding the
claims of state authorities, secessionists, and many outside observers, it is clear that
the violence in each of these places was not primarily an expression of pre-existing
or primordial cultural or ethnic difference. Indeed, the political objectives of those
who challenged state authority—the secessionist movements—were rooted in, or
justified by reference to, ethnic or racial identities either created by or reified under
colonial rule. Likewise, in all of these cases, the ‘nation’ whose unity and security
state authorities sought to maintain was, to a very significant degree, the legacy of
colonial rule. Political conflict and violence, moreover, was clearly not the product
of pre-existing or primordial ethnic identities but, in contrast, actually helped to
create and to solidify those identities.
The wider implication of these observations is that history itself has a defining
importance in the dynamic of genocide and other forms of political violence. Most
obviously, past violence significantly increases the likelihood of future violence. In
part, that is because the experience or memory of violence can help to create or
deepen a sense of group identity and enmity. In part too it is because history,
including memories of past violence, provides the essential raw material for
political leaders seeking to mobilize populations to take part or at least acquiesce
in mass violence.51 Importantly, historical experience and memory also provide the
organizational and behavioural models, as well as the rhetorical tool kit, that are
the foundation of future violence, and that shape its character. These observations
certainly appear to make sense for the cases considered here, which all involve long
histories of violence, where political leaders on all sides have appealed to that
history in mobilizing their followers, and where both identities and enmities
appear to have stiffened through the experience of violence.
51 For variations on this argument, see Paul R. Brass, Theft of an Idol: Text and Context in the
Representation of Collective Violence (Princeton: Princeton University Press, 1997); and Stanley
Tambiah, Levelling Crowds: Ethnonationalist Conflicts and Collective Violence in South Asia
Second, these cases suggest that state-sponsored violence is shaped in significant

ways by the character of state institutions, and more specifically by the attitudes,
policies, and practices of state authorities. It is especially noteworthy that all of
these cases involved violence committed by states dominated by the military, and
obsessed by national unity, security, and order. All invoked the absolute right to
maintain their territorial integrity, and to meet any attempt to breach national
unity with unlimited force. More specifically, in their effort to maintain national
unity, security, and order each of these states developed and employed repertoires
of violence that entailed the massive use of force against civilians. The counter-
insurgency strategies employed in each case, moreover, typically strengthened the
resolve of the secessionists, primarily because of their brutality and because they
targeted civilians thought to be supporting resistance movements.
These observations are pertinent to a common theme in much genocide litera-
ture, which points to the importance of state ideology in fuelling genocide and mass
violence. Weitz, for example, has highlighted the centrality of a utopian or revolu-
tionary vision in some of the twentieth century’s best known genocides.52 The
evidence from the four cases considered here lends some support to that view, but
also suggests the need for its refinement.53 For while the state ideologies of Pakistan,
Burma, and Indonesia evinced a strident nationalism, and contained a powerful
undercurrent of racism, they could hardly be characterized as utopian or revolu-
tionary. Indeed, if any ideology can be said to have driven the state-sponsored
violence in these cases, it was the ideology of an arrogant, bellicose militarism,
wrapped in the guise of legitimate nationalism and benign ‘developmentalism’.54
These four cases also highlight the need to look beyond the conventional state
ideologies envisioned by Weitz and others, to consider the conditioning effects on
genocide of broader ‘systemic’ factors including geopolitical competition and
‘development’.55
Third, in all four cases discussed here the international environment was critical
in shaping both the strategies and language of secessionist movements and the
state violence against them. Most obviously, violence against secessionists was
52 Eric D. Weitz, A Century of Genocide: Utopias of Race and Nation (Princeton: Princeton
53 A number of authors have taken issue with this argument. Straus, for example, has noted that the
leaders who instigated the genocide in Rwanda ‘did so primarily to win a civil war, not to radically
restructure society’. Scott Straus, The Order of Genocide: Race, Power, and War in Rwanda (Ithaca, NY:
Cornell University Press, 2006), 11. Also see Dirk Moses, ‘Toward a Theory of Critical Genocide
Studies’, On-line Encyclopedia of Mass Violence [online]. Available at: http://www.massviolence.org/
Toward-a-Theory-of-Critical-Genocide-Studies.
54 That conclusion is broadly consistent with a wider scholarship suggesting a close link between
nationalism and mass killing. See, for example, Michael Mann, The Dark Side of Democracy:
Explaining Ethnic Cleansing (New York: Cambridge University Press, 2005).
55 Notable proponents of this approach to genocide studies include Donald Bloxham, Dirk Moses,
Mark Levene, and Christian Gerlach.
conditioned by the provision of military, diplomatic, and economic assistance by

key states, or simply through their silence and acquiescence. These observations are
broadly consistent with a substantial body of literature that has highlighted the role
of powerful states in conditioning mass violence and genocide.56 And yet, this
simple portrait of national self-interest tells only a part of the story. The case of East
Timor, in particular, suggests the potentially critical importance of the wider
geopolitical context, and within it the role of non-state actors of international
norms and legal regimes, and of individual acts of courage, in shaping—even
stopping—state violence.
That observation is not entirely new. A number of scholars have highlighted the
ways in which geopolitical context can shape state violence. Others have pointed to
the possibility that acts of conscience on the part of non-state actors—including
the media, religious groups, and NGOs—might prevent, stop, or at least slow the
dynamic of mass violence.57 In addition, some scholars have drawn attention to
the ways in which shifts in international norms and legal regimes might affect the
prospects for intervention to stop mass killing and genocide.58 What is needed is a
framework that draws these elements together. Apart from adding a new dimen-
sion to the analysis of state violence, it would serve to highlight the essentially
contingent quality of genocide and mass violence, and thereby reject any notion
that they are inevitable or unstoppable.
Finally, given the broad similarities among these cases, it is striking that some
have very commonly been described as genocide while others have not. In this
respect, East Timor and, to a lesser extent, East Pakistan stand out. The violence in
those places has been described as genocide, and has been included in general
comparative studies of the subject while, with some rare exceptions, the egregious
violence in Burma and West Papua has not. It is fair to ask why this is the case. The
answer does not appear to lie in any obvious sense in the nature of the violence
itself, in the intentions of the perpetrators, or in the religious or ethnic composition
of the population affected. On all of these dimensions, the differences among the
four cases are not especially significant. That leads inevitably to the conclusion that
other factors were at work.
The discussion of the four cases above suggests at least two possibilities. First, it
would appear that the term genocide has been selectively applied to those cases of
state-sponsored violence that have entailed substantial political costs for Western
powers, while the rest have typically been described as instances of civil war, ethnic
cleansing, cultural assimilation, or human rights abuse. Here the case of East
56 See, for example, Samantha Power, ‘A Problem from Hell’: America and the Age of Genocide (New
York: Basic Books, 2002).
57 See, for example, David Webster, ‘Non-State Diplomacy: East Timor 1975–99’, Portuguese Studies
Review 11:1 (2003), 1–28.
58 See, for example, Geoffrey Robertson, Crimes against Humanity: The Struggle for Global Justice
(New York: New Press, 2000).
Timor is especially instructive. In the first fifteen years of the Indonesian occupa-
tion, Western states were able to deny or obscure the extent and nature of the
violence. With the rise of powerful religious and secular solidarity groups after 1991,
however, and a growing public awareness of the violence, the political costs of
acquiescence and complicity for those states grew exponentially, and transformed
the conflict into an international issue of considerable importance for those states.
Only then did the case of East Timor come to be widely regarded as an instance of
genocide.
A second possibility is that scholars, political actors, and the general public are
less likely to describe as genocide a case in which the narrative is too complicated—
and particularly where both sides have engaged in violence. Secessionists used
violence in all of the cases discussed here, but East Timor was again something
of an exception. As long as the resistance was engaged primarily in armed resis-
tance, only a handful of solidarity groups and scholars were prepared to describe
the Indonesian state violence as genocide. However, as the resistance began to shift
away from armed struggle in the mid-1980s, and to focus instead on a diplomatic
approach that adopted the language of universal human rights rather than national
liberation, international sympathy grew, and the term genocide came to be more
widely accepted.
To sum up, these cases suggest the need for a reconsideration of some of the
more widely accepted approaches to the study of state violence and genocide, and
further reflection on the use and utility of the term itself. More specifically, greater
attention needs to be paid to the dynamic historical interaction between the broad
structural forces that condition state violence and the specific intentions, policies,
and practices of both states and secessionist movements. In that regard, approaches
attentive to the effects of historical experience and memory, to processes of state
formation, to the impact of international norms and legal regimes, to geopolitical
context, and to the role of a wide range of non-state actors would be especially
helpful. Finally, there is a need for continued introspection on the part of scholars
and others with respect to the use of the term genocide. If the term is to have any
analytical value at all—and if it is to be useful as something more than a moral and
political epithet—we must adopt a far more rigorous understanding of its mean-
ing, and examine our own motivations in choosing, or declining, to invoke it.
Amnesty International, Indonesia: Continuing Human Rights Violations in Irian Jaya
(London, 1991).
Callahan, Mary P., Making Enemies: War and State Building in Burma (Ithaca, NY: Cornell
East Timor Commission for Reception, Truth, and Reconciliation (CAVR). Chega! (Dili,
2006); available at http://www.cavr-timorleste.org/en/chegaReport.htm
Human Rights Watch, ‘They Came and Destroyed Our Village Again’: The Plight of Internally
Displaced Persons in Karen State (New York, 2005).
Imama, Jahanara, Of Blood and Fire: The Untold Story of Bangladesh’s War of Independence
(New Delhi/New York: Sterling, 1989).
Osborne, Robin, Indonesia’s Secret War: The Guerilla Struggle in Irian Jaya (Sydney/Boston:
Allen and Unwin, 1985).
The Report of the Hamoodur Rehman Commission of Inquiry into the 1971 War (Lahore:
Vanguard, 2000).
Robinson, Geoffrey, ‘If You Leave Us Here, We Will Die’: How Genocide Was Stopped in East
Timor (Princeton: Princeton University Press, 2010).
Smith, Martin J., Burma: Insurgency and the Politics of Ethnicity (London: Zed Books, 1999).
Taylor, John, East Timor: The Price of Freedom (London: Zed Books, 1999).
chapter 24
.............................................................................................
NAT IONA L
SECURITY
DOCTRINE IN
L AT I N A M E R I C A
THE GENOCIDE QUESTION
.............................................................................................
daniel feierstein
During the second half of the twentieth century, large sections of the population
were exterminated in various parts of Latin America. Most of these events followed
a similar pattern and were the result of what became known as the National
Security Doctrine. Developed primarily by the United States, this policy widened
the sphere of international conflict to Latin America in the belief that the region
could play a strategic role in the fight against communism, an ideological struggle
that had no territorial boundaries.
The National Security Doctrine was inspired by the cold war but also by the
methods developed by Western powers in various counter-insurgency struggles.
In particular, the methods applied by the ‘French school’ in Indochina and Algeria
and adopted by the Americans during the Vietnam War were later taught at
numerous military and ideological training centres in Latin America. The most
important of these was the School of the Americas, first established in the Panama
Canal Zone in 1946 to train Central American forces. Following the success of the
490 daniel feierstein
Cuban Revolution in 1959, another branch of the School was opened at Fort
Benning, Georgia, in 1963 to teach ‘French’ counter-insurgency tactics. Thus, the
practice of systematic annihilation of political enemies in Latin America, which
began as early as 1954 with the military coup in Guatemala, continued almost until
the beginning of the twenty-first century, spreading throughout practically all of
Latin America.
This chapter analyses the general characteristics of these developments, their
similarities and differences, and the possible connections between civil wars in the
region and processes of mass extermination, taking into account that there were no
real wars in many of the territories in which these practices were applied. It also
examines the controversial question of whether some of these events—particularly
the cases of Guatemala and Argentina, as they have been analysed in court
sentences and academic research—could be understood through the genocide
concept, bearing in mind especially what the concept of genocide can contribute
to a better understanding of the differences among these events and the strategic
purpose of the political repression in some of the Latin American societies.
E A R LY A N N I H I L AT I O N P RO C E S S E S
I N L AT I N A M E R I C A
................................................................................................................
During the nineteenth century, most of the newly independent Latin American
states annihilated various indigenous populations that inhabited their territories.
This was done to secure borders, establish a state monopoly on violence, and
change the juridical status of Indian communities that had formed separate
‘republics’ during the colonial period. Argentina and Chile are good examples of
nation-states founded on the annihilation of indigenous groups. Elsewhere, like
Bolivia and Peru, however, these goals were achieved by forcing indigenous popu-
lations into submission rather than annihilating them completely.1
Indigenous populations were also attacked and sometimes deliberately annihi-
lated in wars between these newly independent states. For example, Argentina,
Brazil, and Uruguay all but destroyed isolationist Paraguay and annexed large tracts
of its land in the War of the Triple Alliance (1864–70). However, instances can be
1 In Argentina more than 13,000 Mapuches and thousands of Wichi, Pilagá, and Guarani were
annihilated during 1850–90. In Chile, thousands of Mapuches were also annihilated at the end of the
nineteenth century.
national security doctrine in latin america 491
found well into the twentieth century, such as the massacres of Haitians in the
Dominican Republic in 1937.2
In fact, during the twentieth century, the indigenous populations of Latin
America were often slaughtered by private companies. In order to open up new
territories, the national governments of the region gave entrepreneurs a free hand
in areas historically occupied by indigenous groups. Territories where this policy
was applied during the twentieth century include Southern and Western Mexico,
where indigenous groups were repeatedly attacked by paramilitary forces; various
parts of the Amazon forest in Brazil; Eastern Paraguay, where the Ache Indians and
other groups were enslaved or starved to death from the nineteenth to the end of
the twentieth century; Southern Paraguay and Northern Argentina; where the
Wichı́s, Tufas, and other groups were annihilated; Southern Chile and Argentina;
where the Mapuches were permanently harassed; and numerous parts of Central
America.
After this necessarily brief account of the annihilation of indigenous peoples
during the nineteenth and early twentieth centuries, we shall now consider the role
played by the so-called Doctrine of National Security in unifying existing processes
of political repression and extermination throughout Latin America. This doctrine
also marked a shift from the targeting of indigenous groups as such to a situation in
which ethnicity played only a secondary role, if any, in the choice of victims.
POLITICAL REPRESSION UNDER THE

N AT I O NA L S E C U R I T Y D O C T R I N E
................................................................................................................
Although the National Security Doctrine was promoted throughout the region by
the United States during the cold war, its first true antecedent was the Argentine
Conintes Plan (Plan for Civil Insurrection against the State), which appeared soon
after the visit of a French military mission to Argentina in 1957. The Conintes Plan
spoke of a ‘national menace’ from an ‘internal political enemy’, defined clearly as a
‘political identity’ and not from an ‘ethnic viewpoint’. This doctrine, which was
already being used by a democratic government to repress political opponents in
Argentina at the end of the 1950s, was to be applied even more harshly by military
governments in the following decades in Latin America.
From the late 1950s onwards, the so-called Meetings of American Armies became
a regular institution and an ideal channel for spreading the National Security
2 Richard Lee Turtis, ‘A World Destroyed, A Nation Imposed: The 1937 Haitian Massacre in the
Dominican Republic’, Hispanic American Historical Review 82:3 (2002), 589–635.
Doctrine among the military all over Latin America. US involvement in the
overthrow of Jacobo Arbenz in Guatemala in 1954, Stroessner’s dictatorship in
Paraguay from 1954 onwards and Duvalier’s dictatorship in Haiti from 1957, togeth-
er with a unanimous concern among right-wing Americans about the ‘communist
threat’ after the Cuban Revolution in 1959, all helped to spread the new American
vision of conflict in Latin America, a vision focused on the ‘war against the
Communism’ in which Latin America was seen as a key battleground in a cold
war—and sometimes even a Third World War—of global dimensions. The ‘enemy’
was not only revolutionary movements but any populist, religious, or indigenous
movements with progressive ideas aimed at bringing about social change.3
The concept of ‘dirty war’, originating in French counter-insurgency doctrines,
was soon applied to all situations of conflict in Latin America whether these involved
guerrilla movements waging civil war against dictatorial regimes (as in Colombia,
El Salvador, or Nicaragua), or military-political opposition groups too small to
spark a civil war or even to control sizeable portions of territory (as in Guatemala or
Peru). In most countries, including Argentina, Bolivia, Brazil, Chile, Paraguay, and
Uruguay, armed opposition groups were not even professional insurgents and so to
describe these conflicts as ‘wars’ was exaggerated, to say the least.
Nevertheless, the concept of ‘dirty war’, ‘counter-insurgent war’, or ‘anti-subver-
sive war’ became the ideological justification for turning Latin America’s armed
forces into armies of occupation on their own territories. Generally, a military
regime took control of government and/or paramilitary forces operating in con-
junction with the armed forces. The regime then proceeded to transform society
through the institutionalization of terror at every level of daily life. To do so, it
resorted to a whole arsenal of terror, including concentration camps, systematic
murder of entire groups (families, villages, ethnic and religious communities), and
extensive use of torture and rape as weapons of physical and psychological
destruction.
A BRIEF OUTLINE OF CASES

................................................................................................................
This section describes briefly the main developments in the region country by
country before going on to examine their shared characteristics and consequences.
3 During the 1970s many of the Argentinean military believed that they were fighting a Third World
War, as can be seen in conference papers and other writings from this period. The inspiration for this
idea came especially from André Beaufre’s Introduction to Strategy and the works of other French
counter-insurgents. For an analysis of this way of perceiving conflict, see Horacio Verbitsky, La última
batalla de la Tercera Guerra Mundial (Buenos Aires: Sudamericana, 2002).
In 1954, a US-sponsored military coup in Guatemala toppled Jacobo Arbenz’s

democratic government after it began to introduce land reforms. In the twelve
months following the coup there were around 3,000 political murders. After a
revolt by junior military officers in 1960, which led to armed insurrection in the
countryside, repression intensified and it was not until 1996 that the government
signed peace accords, ending the internal conflict.
During this thirty-six-year period, the military and paramilitary Guatemalan
forces assassinated and/or disappeared more than 200,000 people. This was a
deliberate attempt at political and social engineering involving indigenous com-
munities, villages, groups, and whole regions, with terror reaching its peak roughly
between 1978 and 1990. The methods used in the Guatemalan case were copied, in a
limited range, all over Latin America.
Most experts agree about the appropriateness of the term ‘genocide’ to describe
the atrocities committed in Guatemala, not only because of the sheer number of
victims (almost ten per cent of the total population was killed) but also because of
the targeting of ethnic groups and nations, including different Mayan popula-
tions.4 The term genocide is justified in this case—but not solely for the reasons
just mentioned. Rather, it was the systematic nature of the annihilation that made
it genocide, as the groups were defined by the perpetrators. Groups were killed in
Guatemala under the National Security Doctrine for exactly the same political
reasons that they were killed in some of the other Latin American countries, the only
difference being that in Argentina, Bolivia, Chile, and Haitı́ the number of victims
was much smaller. However, in countries like Brazil, Uruguay, and Honduras, the
killing was neither systematic nor clearly defined as groups. Maybe the genocide
concept could clarify the differences and similarities among different processes.
In 1954, the year of the military coup in Guatemala, Gen. Alfredo Stroessner
seized power in Paraguay from another military dictator, Federico Chávez.
Stroessner’s regime, which lasted until 1989, immediately aligned itself with
American policies and anti-communism in the region, and later with the National
Security Doctrine. From the mid-1970s, Paraguay participated actively in Operation
Condor, a clandestine programme for political repression drawn up by South
America’s right-wing dictatorships under which Argentina, Brazil, Bolivia, Chile,
Ecuador, Paraguay, Peru, and Uruguay exchanged intelligence and prisoners, and
cooperated in carrying out repressive measures, including joint operations.
The Association of Relatives of the Detained and Disappeared in Paraguay estimates
that under Stroessner there were between 3,000 and 4,000 political murders, with
4 Among them, for example, we could include Frank Chalk and Barbara Harff. Concerning this
kind of analysis see, particularly, the work of Marc Drouin, ‘Atrocity Crimes and the Genocide
Continuum in Guatemala, 1978–1984’, Paper presented at the Second International Meeting on
Genocidal Social Practices: From Europe to Latin America and Beyond: The Continuity of the
Genocidal Social Practices, Buenos Aires, November 20–22.
many thousands more imprisoned and tortured. In fact, a large number of docu-
ments relating to Operation Condor were found in Paraguay in 1992 and although
Stroessner was overthrown by his own party in a military coup in 1989, Paraguay
remains one of the few countries in the region where democracy still has not been
restored.
Haiti, on the Caribbean island of Hispaniola, had remained within the American
sphere of influence ever since it was invaded and occupied by the United States in
1915. Thus, when François Duvalier (Papa Doc) became president in 1957, he was
quick to implement the National Security Doctrine, a policy continued by his son,
Jean Claude Duvalier (Baby Doc), who succeeded him in 1971. The paramilitary
groups known as the ‘tonton macoutes’ that Papa Doc set up to exert repression
continued to operate after the overthrow of Baby Doc in 1986 and are still
active. Haitian paramilitary forces are reckoned to have claimed around 30,000
victims, with up to 500,000 people driven into exile during the Duvalier family’s
dictatorship.5
The Dominican Republic, which shares the island of Hispaniola with Haiti, was
ruled by the pro-American dictator Trujillo from 1930 until his assassination in
1961. Thus, when Juan Bosch became president in 1963 after the first free elections
in over thirty years, his government lasted less than a year. An expeditionary force
of US Marines was sent ‘to prevent Communism taking hold of the Republic’, and
Bosch was replaced by Joaquin Balaguer, who had acted as interim president after
Trujillo died. The exact number of people killed during the invasion and the
repression that followed is unknown but it is estimated at several thousands.6
In March 1964, another military-led coup ousted the populist President of Brazil,
João Goulart. Again, the coup was allegedly ‘to frustrate the communist plan to
seize power’.7 The dictatorship that followed lasted until 1989, and although there
was a gradual transition to democracy after 1985, the first ten years of military
government were marked by harsh repression which went under the name of
‘Operação Limpeza’ or ‘Operation Clean-Up’. Although the Brazilian State passed
a law in 1995 admitting responsibility for the murder of 136 people between 1961
and 1979, new cases came to light the following year. In a book called ‘The Right to
Memory and Truth’ published by the Special Commission on the Political Dead
5 Marı́a Paz Fiumara; Exitos y fracasos: las fuerzas armadas y de seguridad en El Salvador y Haitı́
(Bibliotecas CLACSO (http://www.clacso.org.ar), 2004); Elizabeth Abbott, Haiti: The Duvaliers and
Their Legacy, (New York: McGraw-Hill, 1988).
6 Bosch was one of the first to emphasize the paradoxically systematic role that paramilitary forces
would play in annihilation processes in Latin America. See Juan Bosch, El pentagonismo, sustituto del
imperialismo, cited in Verbitsky, La última batalla de la Tercera Guerra Mundial, 19. See also G. Pope
Atkins, Arms and Politics in the Dominican Republic (Boulder, CO: Westview Press, 1981).
7 ‘Ato Institucional I, 9, Government of Brazil, April 1964’, cited in Marcelo Raffin, La experiencia
del horror (Buenos Aires: Editores del Puerto, 2006), 132.
and Disappeared of Brazil at the end of 2007, the government officially admitted
responsibility for the murder of 479 people.
In November 1964, a military coup also put an end to the government of the
Nationalist Revolutionary Movement (NRM) in Bolivia. It is true that the NRM
itself had resorted to political repression during the 1950s, setting up concentration
camps after it came to power in 1952. However, the military junta that replaced it
was quick to align its policies with US interests and extend the repression to
include political assassination and frequent periods of martial law. It is estimated
that from 1966 to 1986 between 3,000 and 8,000 people were executed by ‘death
squads’ under the orders of various military governments. During this period,
Ernesto ‘Che’ Guevara, who had gone to Bolivia to spread his own brand of
revolution, was captured with the help of US Army Special Forces and executed.
For a brief period in 1971, Bolivia was ruled by an anti-imperialist revolutionary
government, with Gen. Juan Jose Torres as president. But Torres was overthrown in
the same year by another military coup led by Gen. Hugo Banzer. Banzer’s regime
became increasingly repressive, especially after the military coup in Chile in 1974
and the subsequent alignment of Bolivia and Chile with the National Security
Doctrine. Torres himself was murdered in Buenos Aires in 1976 under ‘Operation
Condor’.
Finally, Banzer was replaced by a new military junta in 1978, the first of many
between 1978 and 1982. In 1980, democracy triumphed briefly after a popular
uprising forced general elections. But the president-elect, Socialist Hernán Siles
Suazo, was so unacceptable to both the military dictatorships of the region and the
United States that he was deposed before he could take office. In the same year,
1980, Luis Garcı́a Meza Tejada seized power in the so-called ‘cocaine coup’, coordi-
nated from Buenos Aires by an Argentinean army intelligence unit and financed
largely by drug trafficking. Garcı́a Meza was one of the most brutal dictators of the
period and during his one year in office he had more than 500 political opponents
murdered or ‘disappeared’. However, international isolation eventually forced his
resignation on 3 August 1981 and the Bolivian military remained in power for only
another year. Siles Suazo was finally sworn in as president in 1982.
To implement repression, a succession of military governments hired many
international perpetrators of genocide such as former SS officer Mario Busch and
his team of pro-Nazi Croatian Ustaschas/Ustaše, and former Gestapo officer Klaus
Barbie, known as the Butcher of Lyon, who was finally sentenced to life imprison-
ment in 1987 for the deaths of 4,000 people during World War II.8
In 1972, the president of Uruguay’s Colorado Party, Juan Maria Bordaberry,
came to power during an institutional crisis and placed the Uruguayan Armed
8 See Jorge Gallardo Lozada; De Torres a Banzer : diez años de emergencia en Bolivia (Buenos Aires:
Ediciones Perisferia, 1972) and, particularly, Gregorio Selser, Bolivia : el cuartelazo de los cocadólares
(México: Mexsur, 1982).
Forces in charge of the campaign against the urban guerrilla of the Tupamaros
movement.9 The following year, he dissolved Congress, banned trade unions, and
outlawed the Communist Party and other left-wing organizations. Although the
governments of Bordaberry (1972–6) and later Aparicio Méndez (1976–81) were
headed by civilians, effective power remained in the hands of the Armed Forces.
However, in 1980, a change in the constitution proposed by the Uruguayan military
was rejected in a referendum. In 1984, the so-called ‘Naval Club Agreement’
between the military and the main Uruguayan political parties led to general
elections, which were won by the Colorado Party. The newly elected democratic
government finally took power in 1985.
It is reckoned that between 1972 and 1985 the Uruguayan military forces mur-
dered about 100 political prisoners and a further 160 people disappeared.
In addition, thousands more were arrested and tortured and tens or even hundreds
of thousands were forced into exile. Indeed, around 300,000 Uruguayans emigrated
during this period—just over ten per cent of the country’s population—many of
whom were surely forced to leave for political reasons.10
On 11 September 1973, Salvador Allende’s democratic socialist government in
Chile was overthrown in a bloody military coup strongly backed by the US
Department of State. Allende himself was killed or committed suicide during the
bombing of the presidential palace. General Augusto Pinochet took control of the
country and, although his regime became more permissive in the late 1980s, he did
not step down until March 1990 after losing a plebiscite in 1988.
In February 1991, after examining some 3,000 cases of alleged human rights
abuses, the National Commission for Truth and Reconciliation determined (in the
so-called Rettig Report) that during the Pinochet era more than 1,000 people had
been murdered and another 1,000 political detainees had disappeared, also pre-
sumably murdered.11 The Commission concluded that the real number of victims
was probably greater and that many cases will never be reported.
In addition, an unspecified number of people were interned in concentration
camps. Different researchers have calculated the number of detainees at between
several thousand and several tens of thousands but—as with the numbers of
murdered and/or missing persons—it is likely that the true figures exceed these
9 For declarations by the Uruguayan governments of the period and a clear analysis of the
process, see Luis Roniger and Mario Sznajder, El legado de las violaciones a los derechos humanos en
el Cono Sur: Argentina, Chile y Uruguay (La Plata: Ediciones al Margen, 2005). The document cited
above can be found on p. 32.
10 See Marcelo Raffin, La experiencia del horror (Buenos Aires: Editores del Puerto, 2006) and Luis
Roniger and Mario Sznajder, El legado de las violaciones a los derechos humanos en el Cono Sur:
Argentina, Chile y Uruguay (La Plata: Ediciones al Margen, 2005).
11 See Elı́as Padilla Ballesteros, La memoria y el olvido. Detenidos Desaparecidos en Chile (Santiago
de Chile: Ediciones Orı́genes, 1995). Ongoing judicial inquiries in Spain have since brought to light
more victims and have helped to clarify the choice of victims.
estimates, given that most of the perpetrators remain at large and that there is no
protection for victims or witnesses in Chile.
As mentioned earlier, a forerunner of the National Security Doctrine was im-
plemented in Argentina in the late 1950s. In 1966, Gen. Juan Carlos Organia imposed
a hard-line military regime to exclude the Peronists from power. Numerous popular
uprisings followed until Perón (in exile since 1955) finally returned to Argentina and
formed a democratic government in 1973. However, following Perón’s death in July
1974, a systematic campaign of terror known as ‘Operation Independence’ began in
the province of Tucumán. This model was repeated throughout the country after
Gen. Jorge Rafael Videla seized power in 1976 and until the military dictatorship
collapsed in 1983.
Officially, over 13,000 people disappeared or were murdered between 1974 and
1983 although Argentinean human rights organizations put the total at between
20,000 and 30,000. The systematic nature of the repression was unique in Latin
America. A complex network of over 500 concentration camps was set up across
the country and numerous political organizations, trades unions, student unions,
and neighbourhood associations were ruthlessly dismantled and suppressed. That
is why some researchers and Argentine courts have applied the genocide concept to
understand the Argentinean experience.
Moreover, the Argentine ‘success’ was exported to Central America in the late
1970s and early 1980s, when the Argentinean military trained ‘special’ forces in
Honduras, El Salvador, and Nicaragua, in torture and mass murder, as well as how
to dispose of the bodies. As mentioned earlier, the Argentinean army also played a
key role in the military coup in Bolivia in 1980.
In 1980, after more than twenty years of uninterrupted rule, the military govern-
ment in Honduras created counter-insurgency units, specialized military intelli-
gence groups and ‘death squads’, based on the French counter-insurgency model
used in the Latin American Southern Cone, concerning the decision to annihilate
the ‘political enemies’. The Argentinean and US governments both played a key role
in training these repressive forces. Despite the election of a civilian president in
Honduras in 1981, political kidnappings and murders actually increased, persisting
until the end of the decade.
The Committee of Relatives of Disappeared Prisoners in Honduras places the
number of disappeared and murdered at more than 200. The victims were mostly
political leaders, lawyers, and members of trades unions, student movements,
groups expressing solidarity with the victims of other conflicts in Central America,
or members of religious groups. During the 1980s, this pattern was repeated in
Ecuador, Mexico, Venezuela, and Panama.
All the instances of repression considered so far occurred in peacetime. However,
civil wars in Colombia, Nicaragua, El Salvador, and Peru have also claimed tens of
thousands of victims, many of them civilians. Again, the pattern is similar to that
found in other Latin American countries: devastation of towns and villages,
executions of rural indigenous communities, disappearances, and political mur-

ders. In Colombia and Peru (and more recently in Mexico), political repression
became intertwined with the fight against the drug trafficking, generating a con-
fused layering of conflicts. Even more recently, attempts to brand organizations like
the Revolutionary Armed Forces of Colombia (FARC) and the National Liberation
Army (ELN) as ‘terrorist organizations’ and to include them in the new global
strategies of the ‘war on terror’ have led to a further endorsement of repressive
policies.
C H A R AC T E R I S T I C S O F T H E R E P R E S S I O N
I N L AT I N A M E R I C A
................................................................................................................
As this brief summary of events has shown, terror was used systematically in
different Latin American countries in the second half of the twentieth century in
order to transform whole societies and so destroy any possibility of political
opposition to pro-US policies. Intelligence was gathered and state and paramilitary
forces were established to intimidate, kidnap, torture, and murder political
opponents. The widespread practice of disappearing the victims served to eliminate
evidence, as well as instilling terror in the communities to which the victims
belonged and denying relatives and friends the chance to grieve for their loved ones.
Although similar methods were used throughout the region, the scale and the
systematic character of the repression varied from country to country. In some
cases, violence was ferocious but limited to small, clearly defined people, as in
Brazil, Ecuador, Uruguay, and Honduras, where the number of political murders is
estimated in hundreds. By contrast, thousands or tens of thousands of people from
all walks of life were murdered in Chile, Argentina, Paraguay, and Bolivia. Here, the
sheer extent of the repression shows a deliberate policy of social transformation
involving the systematic destruction of trades unions, political parties, student
groups, and so on. In short, the aim was not the political persecution of specific
individuals but the devastation of social groups as such.
Guatemala has become the archetype of this modus operandi because of the
scale and duration of the repression and its impact on the population, with whole
communities being wiped out, as happened during the civil war in El Salvador.
However, in the late 1970s and early 1980s it was Argentina that had the dubious
distinction of organizing repression and political intervention throughout the
rest of Latin America. The Argentinean army not only trained Latin America’s
military—particularly in Honduras and El Salvador in counter-insurgency tactics,
torture and the art of disappearing people, but also designed and participated in
the military coup in Bolivia in 1980.
THE CONCEPT OF WA R
................................................................................................................
Many analysts and international observers, as well as the perpetrators themselves,
have used the term ‘war’ (counter-insurgency war, civil war, anti-subversive war,
dirty war, etc.) to describe the terror implemented by some of Latin America’s
dictators. The concept of war is inappropriate because it suggests that the victims
were mostly armed combatants, rather than the unarmed civilians that the vast
majority really was. It also tends to gloss over the true nature and purpose of the
atrocities committed against them.
Although modern warfare has blurred the distinction between combatants and
non-combatants, producing ever growing numbers of civilian victims, even many
armed conflicts of Latin America lacked the basic conditions that military theorists
consider necessary to speak of war. Minimally, these consist of a confrontation
between professional armies; control of territory; and, in the case of civil war, the
alignment of most of the population with the different parties in conflict.12 If we
accept these three prerequisites for war, only Colombia, Peru, Nicaragua, and
El Salvador were ever actually at war. In Argentina, the Ejército Revolucionario
del Pueblo (People’s Revolutionary Army)—which had little to do with most of the
victims in Argentina—never managed to control even the mountainous region in
the province of Tucumán where it had set up its headquarters, while guerrilla forces
in Guatemala were unable to liberate any area of the country, their only support
coming from indigenous and peasant farming villages. In other countries, the
insurgents did not have the backing or the resources to control any territory.
Moreover, none of the insurgent forces in these countries—with the exception of
the Unión Revolucionaria Nacional Guatemalteca (Guatemalan Revolutionary
National Union) during the 1980s—commanded professional troops. Instead,
military–political groups divided their time between political and military
activities, without ever becoming professional armies.
The concept of war, then, has proved more useful for legitimizing the actions of
the perpetrators than for constructing a theoretically principled understanding of
the specific nature of these conflicts. Attempts to describe such conflicts only in
terms of ‘repression’ are also problematic, because they tend to look at individual
cases in isolation, treating them as ‘excesses’ committed by Latin American dicta-
tors, rather than as a set of common policies based on a shared ideology of
12 For an analysis of two different theories of war and their possible application to events in Latin
America, see Juan Carlos Marı́n, Los hechos armados. Argentina, 1973–1976. La acumulación primitiva
del genocidio (Buenos Aires: PICASO/La Rosa Blindada, 1996), which describes the events in Argentina
as a ‘civil war’; and Eduardo Luis Duhalde, El Estado Terrorista Argentino. Quince años después, una
mirada crı́tica (Buenos Aires: EUDEBA, 1999) or Carlos Flaskamp, Organizaciones polı́tico-militares.
Testimonio de la lucha armada en Argentina (1968–1976) (Buenos Aires: Ediciones Nuevos Tiempos,
2002), which take a different view.
‘national identity’ and ‘security’ and aimed at transforming and ‘purifying’ the
societies of Latin America, in their own words.
A RG E N T I NA : A D E BATA B L E E X A M P L E O F
THE USE OF THE GENOCIDE CONCEPT
................................................................................................................
The term genocide has been widely used to describe the extermination of indige-
nous communities in Guatemala. It has generally been avoided when referring to
similar developments in other Latin American countries, because the destruction
of political groups was excluded—at the request of Great Britain and the Soviet
Union (among others)—from the definition of genocide adopted by the UN
General Assembly in the Convention on the Prevention and Punishment of the
Crime of Genocide in 1948.
This exclusion focuses narrowly on ethnicity rather than a wider vision of the
political objectives of genocide. This restriction occurred not only at the drafting
stage of the Convention with regard to the strategy of annihilation pursued by the
Third Reich, but also in analyses of subsequent conflicts in Indonesia, Cambodia,
the Balkans, and Africa.
The military repression in Argentina is a good testing ground for the notion that
in modern times—except in wartime and during colonial occupation—genocide
has been perpetrated ‘from within’. Firstly, there are some legal sentences concerning
that genocide happened in Argentina.13 Secondly, unlike what happened in Guate-
mala, ethnicity and religion played an almost negligible role in the choice of victims
with the exception of the ‘non-systematic’ anti-Semitism of the perpetrators in
Argentina, a fact that has forced many genocide scholars to go beyond the analysis
about which specific groups were annihilated and reconsider the strategic purpose
of genocide in Argentina as a whole, as a practice directed against the whole
Argentinian group.
Let us look first at the legal position on the Argentine genocide. In 1999,
following extensive enquiries, Judge Baltazar Garzón of the Fifth Central Court
of Instruction in Madrid indicted ninety-eight members of the Argentinean mili-
tary for crimes of genocide and terrorism. Although Spain’s Audiencia Nacional
later altered the charges to crimes against humanity (Adolfo Scilingo was finally
13 See, particularly, the sentences against Miguel Osvaldo Etchecolatz, 2006, and Christian Von
Wernich, 2007, by the Federal Criminal Oral Court No. 1 of La Plata, Buenos Aires Province,
Argentina, available in Spanish at http://www.ladhlaplata.org.ar/juicios.htm. More than likely other
sentences will follow, as the trials in Argentina continue.
convicted on this count in 2005), the Federal Criminal Oral Court No. 1 of La Plata,
Argentina, has since sentenced Miguel Osvaldo Etchecolatz (former Director
General of Investigations for the Buenos Aires Police) and Christian Von Wernich
(a police chaplain), for ‘crimes against humanity within the framework of the
genocide that occurred in Argentina between 1976 and 1983’.
In the case of Etchecolatz, the Court recognized that the charge of genocide had
been rejected on a technicality known as the ‘principle of congruity’ (the charge
had not been included in the original investigations) but went on to give a lengthy
justification of why the genocide label should be applied to the Argentinean
experience.
The grounds of the sentence included the following points:
1) Political groups were included in all drafts of the Convention on Genocide
following UN Resolution 96/1 of the General Assembly, December 11, 1946.
Political groups were not excluded for legal reasons from the Convention but
because of ‘the prevailing political circumstances at the time’, a reason not
supported by the philosophy of law.
2) Even though the Convention excludes political groups, the Court considered
that ‘there is no impediment to the use of the term genocide’ in describing what
happened in Argentina, namely, the ‘partial annihilation of a national group’,
since Argentina’s Supreme Court ruled (in Causa 13/84 against the Argentine
Military juntas) that ‘we consider as proven the practices of mass des-
truction implemented by those calling themselves the Process of National
Reorganization . . . , a process that was practically identical throughout the
country and prolonged in time’.
3) The Court also deemed that ‘the plural and pluripersonal acts alleged were acts
against a group of Argentineans or residents of Argentina that could be differ-
entiated, and which no doubt were differentiated by those who organized
the persecution and harassment’ and these actions ‘consisted of deaths, pro-
longed illegal detentions . . . , tortures, confinements in clandestine detention
centers . . . , removing detained children and giving them to other families—
forcibly transferring children of the group to another group—so that the idea is
clearly present of the extermination of a group of the Argentinean population.
This was not done in a random or indiscriminate fashion, but with the intention
of destroying a section of the population . . . composed of those citizens who did
not fit the type pre-established by the promoters of the repression as necessary
for the new order to be installed in the country.’
4) Developing this argument, the Court further considered that those targeted for
extermination were not specific individuals; rather ‘thousands were disappeared
or killed for no political or ideological reason other than the fact that they
belonged to certain communities, sectors or groups of the Argentinean nation
(national group) which [the perpetrators] considered, in their inconceivable

criminal logic, to be incompatible with the Process.’
5) Finally, the Court considered that ‘the term “national group” is absolutely valid for
analyzing what happened in Argentina since the perpetrators set out to destroy
part of the social fabric in order to produce a sufficiently substantial change so as to
affect the State in its entirety. Given the inclusion of the term “total or partial” in
the definition of the 1948 Convention, it is evident that the Argentinean national
group has been annihilated “partially” and to a sufficiently substantial extent as to
alter the social relations within the nation . . . the annihilation in Argentina was not
spontaneous, was not fortuitous, was not irrational: this was the systematic
destruction of a “substantial part” of the Argentine national group, with the
intention of transforming it as such, redefining its way of life, its social relations,
its destiny, its future.’
For this reason, the sentence concludes that ‘from all that has gone before it is
indisputable that we are not dealing as we previously expected with a mere
succession of crimes, but rather with something significantly greater than deserves
the name of genocide.’14
At the time of writing, preliminary investigations have begun into the repression
in the Argentinean province of Tucumán in 1974–5, which is now being treated as a
crime of genocide from the very beginning.
The specific charge presented in all these cases has been the partial destruction of
the Argentinean national group, in accordance with article II of the Convention on
the Prevention and Punishment of the Crime of Genocide.
However, labelling the Argentine experience as genocide has been controversial,
particularly outside Argentina. One of the main arguments of the lawyers and
international organizations that rejected Garzón’s and Argentina’s sentences has
been that
[T]he victims were individually chosen for their political beliefs and not because they
belonged to a group, which would imply that those responsible for their disappearance and/
or murder did not have the necessary criminal intent as they did not intend to destroy a
group but only to eliminate political dissidents.15
That is the reason human rights organization Nizkor has presented an amicus
curiae under Spain’s courts to modify the sentence from genocide to crimes against
14 All the quotes were extracted to the sentence against Echetcolatz by the Federal Oral Criminal
Court No. 1 of La Plata, September 2006, available in Spanish at http://www.ladhlaplata.org.ar/juicios.
htm
15 The amicus curiae presented by the human rights organization Nizkor in the trial against Adolfo
Scilingo requested he be sentenced for ‘crimes against humanity’, but not under the concept of
‘genocide’. For the complete amicus curiae and the sentence of Spain’s Audiencia Nacional, see
http://www.nizkor.org
humanity. At the moment, national and international lawyers and human rights
organizations are discussing these two ways of understanding the facts: genocide or
crimes against humanity. The question continues to be controversial.
The researchers and lawyers that supported the idea that genocide occurred in
Argentina says that there is ample evidence that the vast majority of the victims
were social activists belonging to trades unions, student organizations, or neigh-
bourhood associations, and so the notion of ‘individual political dissidence’ has
proved difficult to sustain, because there was a group dimension. Also, there are
more than 500 cases in which the children of activists were kidnapped and, in some
cases, tortured, murdered, or disappeared—children that by no stretch of the
imagination could be classed as ‘political dissidents’.
Nevertheless, it is insufficient to prove that political activists and their children
were systematically persecuted and murdered or even that a network of over 500
concentration camps was set up all over Argentina for this purpose. Indeed, the
defenders of the genocide position, says that focusing too narrowly on the fact that
the victims belonged to certain groups distracts attention away from the real
intentions of the perpetrators—the transformation of the rest of Argentinean
society by eliminating these groups and what they represented.
For a better understanding of the debate, we turn now to public declarations
made by some of the perpetrators themselves and which are widely quoted in the
academic literature and, particularly, as a ‘genocide proof ’ in the Argentine Court
sentences.16
T H E VO I C E OF THE P E R P E T R ATO R S
................................................................................................................
Nine months after the military coup, Gen. Jorge Rafael Videla, who headed the
military government between 1976 and 1980, the worst years of repression, declared
in an interview in the magazine Gente that
Argentina is a Western and Christian country, not because a notice at Ezeiza Airport says so,
but because of its history. It was born Christian under Spanish rule; it inherited its Western
culture from Spain and it has never renounced this condition; on the contrary, it has
defended it. It is to defend this Western and Christian condition as a way of life that this
16 This is just a small sample of the declarations that have been collected. A much larger sample can
be found in the grounds of the accusations and sentences against Miguel Osvaldo Etchecolatz and
Christian Von Wernich at http://www.ladhlaplata.org.ar/juicios.htm. For an analysis of these
declarations see Daniel Feierstein, El genocidio como práctica social (Buenos Aires: Fondo de Cultura
Económica, 2007).
struggle has begun against those who have not accepted this way of life and have tried to
impose a different one.17
A year later, in a newspaper interview published in La Prensa, Videla added that

it was not the fact that some individuals might ‘think differently’ that was being
persecuted by the military dictatorship, but the effects this might have on the way
other members of society behaved. In Videla’s own words:
Within our way of life, nobody is deprived of freedom just because they think differently;
but we consider it a serious crime to attack the Western and Christian way of life and try to
change it for one that is completely alien to us. The aggressor in this type of struggle is not
just the bomber, the gunman or the kidnapper. At the intellectual level, it is anyone that
tries to change our way of life by promoting subversive ideas; in other words, who tries to
subvert, change or disrupt [our] values . . . A terrorist is not just someone who kills with a
gun or a bomb, but anyone who spreads ideas that are contrary to Western and Christian
civilization.18
It was not just Videla who thought in this way. One of the main promoters of
French counter-insurgency methods, Gen. Alcides Lopez Aufranc, who had at-
tended courses at the École de Guerre in Paris in 1957, argued that ‘The Argentinean
political spectrum must be modified. Otherwise, we will have to call elections
sooner or later and the alternatives are Peronists, radicals and Marxists.’19
Many similar declarations by the repressors are on record. But perhaps the
clearest indication of the military dictatorship’s intention to systematically reorga-
nize Argentinean society is to be found in the name the regime gave itself—‘Proceso
de Reorganización Nacional’ (‘Process of National Reorganization’). The process
was worked out in detail by Gen. Diaz Bessone, the Minister of Planning, in his
‘National Project’, which argued that ‘the real objective is to organize a new and
viable political system and to make the achievements of [our] armed intervention
irreversible.’
Going on to analyze what he calls ‘the foundational stage’ of the Process of
National Reorganization, Diaz Bessone emphasized the following:
Founding a new republic is no easy matter . . . The armed forces must be sufficiently alert,
determined and resourceful to act simultaneously as an efficient fighting force against
guerrillas and terrorists; an efficient surgeon that will remove the evil from all social classes
and walks of life; an efficient government that will steer the ship of the state skillfully and
prudently; and last but not least, parents of the new republic, strong, united, just, free,
supportive of others, clean, exemplary . . . But it is only fair to point out that since no
national project was outlined beforehand, little has been achieved so far to accomplish the
remaining objectives, which are to defeat not only the guerrillas but subversion ‘in totum,’
17 Jorge Rafael Videla, in Gente, 22 December 1976.

18 Jorge Rafael Videla, in La Prensa, 18 December 1977.
19 Declarations cited in Enrique Vázquez, La última. Origen, apogeo y caı́da de la dictadura militar
(Buenos Aires: EUDEBA, 1985), 115.
so laying strong foundations for the birth of the new republic . . . This national project, the
political project, the creative project of life in common, will have no meaning, nor will it
illuminate Argentina’s path ahead, unless it is applied now. Otherwise, we run the risk of
going astray or falling behind those nations that actively determine the course of history.
Moreover, our failure so far to solve the basic problems may give our opponents the chance
to regroup as long as those who create and sustain subversion remain alive.20
In short, Diaz Bessone is arguing that this regime, which defines itself as a
‘Process of National Reorganization’, must supplement military action against
guerrilla forces with ‘surgery’ to ‘remove the evil’ from every part of society and
so make way for a ‘New Republic’. In other words, a series of individuals and
groups must be annihilated to achieve the transformation—or to guarantee secu-
rity and purity—of society. The enemy is not just armed guerrilla forces, nor is this
an exercise in indiscriminate terror against the population as a whole. Instead, it is
a clearly defined ‘surgical operation’ on previously defined sections of the popula-
tion whose disappearance is meant to have an ‘irreversible’ effect on Argentinean
society.
Although state repression in Argentina is better documented than that in Chile,
Bolivia, or Guatemala, the logic applied by the perpetrators in these countries was
similar, although the Argentine systematic nature of the concentration-camps
structure was not present in almost any other experience.
On the other hand, in the Guatemalan case, the other Latin American experience
analysed as genocide by many researchers into the genocide studies field, the
annihilation of indigenous communities had less to do with ethnicity or religion
per se than with the perpetrators’ perception of these communities as natural allies
of the guerrillas. Thus, the Guatemalan genocide was motivated primarily by a
decision to politically transform Guatemalan society rather than by racism. The
same is true of the many so-called ethnic killings in Bolivia and El Salvador, where
racism among the ‘rank and file’ was manipulated instrumentally for strategic
political ends.
By contrast, Brazil is a clear example of state repression that targeted specific
individuals, but not necessarily whole groups. Rather, terror was imposed through
imprisonment, mistreatment, and sporadic, selective killings. While these crimes
clearly fit the category of crimes against humanity, it is difficult to classify them as
genocide in the absence of any clearly identifiable cases of systematic annihilation
against any group.
Most researchers have refused to consider the possibility of genocide unless it
involves the persecution of ethnic groups. It is one of the controversial points in the
analysis about Latin American experiences.
20 A fundamental part of the ‘National Project’—including the passage cited above—is reproduced
as an annex in Vázquez, La última, 299–327.
However, one of the risks of this kind of approach could be the distortion of the
strategic purpose of the annihilation processes, that in any case was directed to
ethnic groups ‘as such’ but was directed against different groups (many of them,
ethnic groups) by ‘political reasons’. This viewpoint also blurs the distinction
between cases like Argentina, where society ‘transformation’ or ‘purification’ was
brought about by the systematic annihilation of whole groups, and other cases like
Brazil, where the state pursued the same goal by different means.
On the contrary, some researchers are convinced that to extend the genocide
concept to those crimes committed under ‘political reasons’ could trivialize the
genocide concept and turn it into a non-precise legal tool.
The debate continues openly, not only in courts, but also in academic research.
T H E C O N S T RU C T I O N OF C O L L E C T I V E M E M O RY
................................................................................................................
This chapter has argued that state repression carried out in various Latin American
countries under National Security Doctrine constituted attempts to transform
whole societies. Although repression in some countries was limited in scope, in
others—particularly in Guatemala, Bolivia, Chile, Argentina, Paraguay, and El
Salvador—the policy chosen was the systematic annihilation of some groups of
the population.
It has also argued that the notion of genocide as the partial destruction of a
national group by members of the same group highlights the strategic nature of acts
committed by agents of the state in some Latin American countries. This view of
genocide may, in turn, provide us with a broader and deeper understanding of
other instances of mass destruction in the twentieth century, from Nazi Germany
to Indonesia, Cambodia, and the former Yugoslavia. These cases are often mis-
understood because of a refusal by scholars to examine the deeper political nature
of almost all the modern genocides.
Many Nazis are on record as saying that their objective was to transform German
society (and then the whole of Central Europe) and that the eradication of the
Jewish, Sinti, Roma, and Slav populations from the Reich was aimed at achieving a
Reich free from ‘ideological infections’. This is not to deny the racist philosophy of
National Socialism—simply to emphasize its predominantly political nature, al-
though it could be debatable.
Similarly, the policy of the Khmer Rouge between 1975 and 1979 of forcibly
depopulating cities and relocating people on collective farms, killing any that
resisted, was aimed at a global transformation of Cambodian society. Again, the
first victims were political groups and intellectuals. More recently, the policies of
‘ethnic cleansing’ carried out in Bosnia during the late 1980s and early 1990s were
intended to break up the Yugoslav federation and transform Bosnia into a homo-
geneous Serbian society or, if this were not possible, to transform the Yugoslav
federation into territorially independent nations that no longer considered them-
selves to be members of a Yugoslav society. In fact, the first victims were people of
mixed race in Bosnia-Herzegovina, especially in the cosmopolitan city of Sarajevo,
together with Serbian, Croatian, and Bosnian politicians and intellectuals who
argued for the need to preserve a Yugoslavian identity.
C O N C LU S I O N
................................................................................................................
The history of Latin America in the second half of the twentieth century illustrates
with unusual forcefulness what lies at the heart of modern genocide processes.
It demonstrates clearly how, in some cases, the systematic annihilation of certain
groups serves as a tool for the partial destruction and transformation of society
from within. The fact that in Latin America these groups were clearly chosen for
political rather than ethnic or religious reasons facilitates the understanding of
similar processes where the ethnic or religious elements are more intertwined.
Although discussion of the appropriateness of the genocide concept to analyse
some of these facts continues to be controversial, the mere debate could produce a
better understanding of the mass annihilation processes all over the world, through
the Latin American peculiarities.
On the other hand, maybe if the collective memory of genocide is one of
definitively differentiated groups annihilating one other, the strategic aims of
genocide will be fulfilled. Jews and Sinti and Roma will never be seen again as
Germans; Serbs, Croatians, and Bosnians will never again be Yugoslavs; ‘Commu-
nist Indians’ will never again be Guatemalans; and ‘subversive delinquents’ will
never be Argentineans.21
And so these ‘alienated’ ways of constructing memory gradually force those who
survive to align themselves on one side or another. The survivors are either Ger-
mans or Jews, either Serbs or Bosnians—until they can no longer see how the
processes of annihilation have transformed their societies and the ways they
construct their own identities. The trials currently taking place in different
countries into what happened in Latin America in the second half of the twentieth
21 One of the best and most readable analyses of the modes of construction and possible
deconstruction of the processes of identity is Benedict Anderson, Imagined Communities: Reflections
on the Origin and Spread of Nationalism (London/New York: Verso, 1991), which questions the
so-called ‘eternal’ or ‘essential’ nature of ‘national identities’.
century may be an opportunity for us all to reframe these ways of constructing

memory.
Alternatively, we can forget that these processes of annihilation ever occurred
and accept the fragmentation of our societies, as well as the dissolution of our
identities that has resulted from them. We can also forget that the concept of
‘national identity’—the recurrent justification for genocide in our modern world—
is no more than a social construction dating back some two hundred years. During
this time, our ability to kill one another has increased almost daily because we have
failed to understand that ‘national identity’ is no more than a pretext to annihilate
‘others’ as well as the ‘other’ hidden within ourselves.22
Armony, Ariel, Argentina, the United States and the Anti-Communist Crusade in Central
America, 1977–1984 (Athens: Ohio University Press, 1997).
Corradi Juan, et al., Fear at the Edge: State Terror and Resistance in Latin America (Berkeley:
Feierstein, Daniel, ‘Political Violence in Argentina and Its Genocidal Characteristics’,
Journal of Genocide Research 8:2 (2006), 149–68.
—— El genocidio como práctica social. Entre el nazismo y la experiencia argentina (Buenos
Aires: FCE, 2007).
Feitlowitz, Marguerite, A Lexicon of Terror (Oxford: Oxford University Press, 1998).
Ibarra, Carlos Figueroa, ‘The Culture of Terror and Cold War in Guatemala’, Journal of
Grandin, Greg, The Last Colonial Massacre: Latin America in the Cold War (Chicago:
University of Chicago Press, 2004).
McSherry, Patrice J., Predatory States: Operation Condor and Covert War in Latin America
(Lanham: Rowman and Littlefield, 2005).
Menjivar, Cecilia, and Néstor Rodrı́guez, When States Kill: Latin America, the U.S. and
Technologies of Terror (Austin: University of Texas Press, 2005).
Roniger, Luis, and Mario Sznajder, The Legacy of Human Rights Violations in the Southern
Cone (Oxford: Oxford University Press, 1999).
22 I thank Dirk A. Moses, Donald Bloxham, and Douglas Town, whose detailed suggestions greatly
improved the final manuscript.
chapter 25
.............................................................................................
GENOCIDE AND
P O P U L AT I O N
D I S P L AC E M E N T I N
P O S T- C O M M U N I S T
E A S T E R N E U RO P E
.............................................................................................
cathie carmichael
................................................................................................................
Between 1990 and 2010, the political map of the Balkans and Caucasus changed
as Communist regimes collapsed and border disputes escalated. Some of the most
bitter conflicts were in areas where there were mixed populations with large
‘minority’ populations that did not recognize potential changes in borders (Rus-
sians in Chechnya or Serbs in Croatia, Bosnia, and Kosovo). Not all ethnically
mixed regions did experience conflict: the Baltic and other mixed regions such as
Transylvania avoided significant bloodshed. For conflict to take place there must be
sufficient ideological mobilization to allow the politics of violence to flourish.
Religious identities, heavily repressed during the Communist era, became a defin-
ing part of the rejection of that system. Orthodoxy and Catholicism were revita-
lized as a political alternative after years of official atheism. A revival of militant
510 cathie carmichael
Islamism after the success of Ayatollah Khomeini in Iran in 1979 radicalized politics
in both the Balkans and the Caucasus. This religious revival exacerbated tensions
between ethnic communities in contested regions, giving politicians an apparently
‘traditional’ pretext for augmenting their power within narrow ethnically based
constituencies.
The decline of Soviet power in the Transcaucasian region precipitated fighting
in the Azerbaijani autonomous oblast of Nagorno-Karabakh between the Azeri
minority and Armenians, with the exodus of the former in the 1990s. Conflict
between the former oblast of South Ossetia and the Georgian government erupted
in both 1991 and 2008, with military intervention by the Russians against the Tbilisi
government. The dissolution of the Chechen-Ingush Autonomous Soviet Socialist
Republic in 1991 into the republics of Chechnya and Ingushetia sparked two brutal
wars between the Russian Federation and those Chechens who wanted full sover-
eign independence. In the former Communist state of Yugoslavia, over 120,000
people died in a series of local wars (mali ratovi) triggered by organized Serb
resistance to the independence of Bosnia and Croatia where many ethnic Serbs
were living. Both conflicts were settled in 1995, with significant territorial revisions
in Bosnia. In 1999, conflict over the status of the Serbian region of Kosovo, largely
populated by pro-independence Muslim Albanians, led to NATO intervention,
ethnic cleansing spearheaded by the government troops of Slobodan Milošević,
and the subsequent rout of the Serbs.
The ‘unmixing’ of populations before the 1990s also followed historical trends.
The long collapse of the Ottoman Empire in the Balkans had led to a dispersal of
populations, particularly Muslims and Orthodox Serbs. When the Serbs were
unified in the first Yugoslavia after 1918, the other nationalities resented their
dominance. In the Caucasus, the rejection of Russian hegemony and the embrace
of a wider Islamic identity had its origins in the colonial expansion of the Tsarist
state in the region over the previous two hundred years. Recent violence dramati-
cally simplified the ethnic composition of many areas and millions became
permanent refugees. During the 1990s, about sixty per cent of Bosnia’s population
was either temporarily or permanently displaced. At the time of the declaration of
Bosnian independence in 1992, approximately seventeen per cent of the republic’s
population were Croats; now the number is about fourteen per cent. This pattern
was repeated elsewhere. The Serb minority in Krajina, Slavonija, and more
recently Kosovo shrank considerably, but in Bosnia, their percentage of the overall
population actually rose from thirty-one per cent in 1991 to thirty-seven per cent
in 2006. Bosnian Muslims now live in confined areas of the Western Federation
territory rather than in the towns of the eastern part of the republic, where once
they formed a majority, and they have also moved to other former Yugoslavian
republics.
genocide and population displacement 511
T H E ‘N AT I O NA L Q U E S T I O N ’ AND COMMUNISM
................................................................................................................
Marxist views on national self-determination evolved in the years immediately
before the First World War.1 Somewhat reacting against classic dialectical materi-
alism, which tended to view national identity as part of the ‘superstructure’, Josef
Stalin argued that national identity was unlikely to wither away and was therefore
infrastructural: ‘a historically evolved, stable community of language, territory,
economic life and psychological make-up manifested in a community of culture.’2
When the Soviet Union was founded in 1922, it was supposed to be a union of
equal nation-states, but some nations were more equal than others. It initially
contained four republics, subsequently dividing into fifteen discrete republics,
albeit under the leaden influence of Moscow. More problematically, the Russian
Republic included sixteen ‘autonomous republics’ (including the Chechen-Ingush
republic) and numerous administrative units or oblasti. The Communist states
created borders that in some cases could only be maintained by their firm political
control.
For the most part, the internal ‘republican’ borders that Communist authorities
fashioned in Europe and Eurasia were to prove durable. The USSR’s 1922 bound-
aries largely survived the collapse of the country in 1991, creating new independent
states ‘from within’. The estimated 35 million Communist-era ethnic Russian or
Russian-speaking migrants in the ‘near abroad’ did not generally aspire to join the
new Russian Federation, preferring to gain citizenship in Lithuania, Latvia, Esto-
nia, Belarus, Ukraine, or Kazakhstan wherever possible.3 In Europe, the status of
Transdniester (a small unrecognized territory between Moldova and Ukraine with
a mixed Moldovan, Russian, and Ukrianian population) has sparked tensions, but
as the peaceful situation in the Baltic indicates, ethnic nationalism is not always an
overwhelming political force.4 The Czech and Slovak Socialist Republics, created in
1969 (from the one reform to survive the crackdown after the Prague Spring),
separated fairly amicably in 1992. The controversial establishment of the People’s
Republic of Poland in 1952 had involved a dramatic push to the West and incor-
poration of former German territory and the expulsion of German speakers as well
as bitter disputes between Ukrainians and Poles in the late 1940s. Nevertheless, with
no will to contest these borders or reverse the ethnic cleansing of the previous
1 Erik van Ree, ‘Stalin and the National Question’, Revolutionary Russia 7:2 (1994), 214–38.
2 Eric Hobsbawm, Nations and Nationalism since 1780: Programme, Myth, Reality (Cambridge:
Cambridge University Press, 1992), 5.
3 Vera Tolz, ‘Conflicting “Homeland Myths” and Nation-State Building in Postcommunist Russia’,
Slavic Review 57:2 (1998), 267–94.
4 Henry E. Hale, Foundations of Ethnic Politics: Separatism of States and Nations in Eurasia and the
World (Cambridge : Cambridge University Press, 2008).
generations, Poland evolved peacefully into a modern democratic state from 1989
onwards. The German Democratic Republic founded in 1949 was fairly easily
absorbed by the former West Germany in 1990 without bloodshed, territorial
revisions, or major protest. The contrast between violent and peaceful transition
to post-Communism is instructive, suggesting that without individuals willing to
spark protest and lead territorial revisionist movements, borders can attain legiti-
macy over time, whatever their actual ethnic or religious composition. Despite the
relative lack of violent political unrest associated with the break-up of these
Communist states, there proved to be major unresolved problems linked primarily
to unresolved border questions. It was these areas in which the most serious and
genocidal conflicts developed after the 1990s. It is also likely that (often unac-
knowledged) prejudice against Islam hampered the creation of stable Muslim
nations in both the Soviet Union and Yugoslavia.
T H E S TAT U S OF N AG O R N O -K A R A BA K H
................................................................................................................
The war in Nagorno-Karabakh from 1988 until 1994 was a direct consequence of
the collapse of Communist authority and unresolved border questions from the
pre-Communist period. In 1922 a ‘Transcaucasian Federal Republic’ had been
created, consisting of Georgia, Azerbaijan, and Armenia, which was to last until
1936. This region had seen intense fighting during the First World War between
Imperial Russia and the Ottomans and the early 1920s and bitter divisions
already existed between the region’s Christian Armenian and Muslim Azeri
populations.5 After 1936, the individual republics enjoyed full national status
within the USSR, which did not itself solve the status of minorities. Although
Nagorno-Karabakh was in Azerbaijan, over ninety per cent of the oblast popula-
tion was Armenian. In February 1988, the Armenian political authorities in
Nagorno-Karabakh, emboldened by Mikhail Gorbachev’s promises of restructur-
ing (perestroika), voted to unify the oblast with the Armenian Republic; the event
is generally seen as precipitating the crisis as it is likely that they remained deeply
inimical to Muslim neighbours and were no longer forced to share power in the
absence of Soviet pressure. In March 1988, the Armenian authorities began to
expel Azeris from the mountainous region after an Azeri pogrom in the city
of Sumqayit/Sumgait on the Caspian Sea resulted in the deaths of 26 local
the Ottoman Armenians (Oxford: Oxford University Press, 2005), 232–3.
Armenians and destructive riots which led to thousands more being injured,
including Azeris. Although the Soviet media tried to repress news of the tragedy
for several weeks, the pogrom was a signal that Soviet authority had dwindled
and that restructuring would bring problems. Both the Soviet Union and Yugo-
slavia had low homicide rates before the breakdown of authority.6 Fear of state
reprisal against manifestations of ‘divisive’ nationalism had helped to minimize
public disorder in Communist societies. So great a shock was the news of the
Sumgait pogrom that the Dagestani poet Razul Gamzatov called it a ‘Chernobyl
of the spirit’.7
In the following six years, Armenia and Azerbaijan fought over the issue, with
peace finally brokered by the Russians in 1994. Historical prejudices had enflamed
relations between the two states. The Azerbaijani capital, Baku, had been the site of
bitter pogroms between Muslims and Armenians in 1905.8 The latter were often
more prosperous and some anti-Armenian actions took on the character of ‘class’
warfare. The Armenians, still collectively traumatized by the genocide of 1915, saw
the (generally Shi’ite) Azeris as a Muslim ‘menace’ to a Christian identity under
threat in the entire region. The Soviet invasion of Afghanistan in December 1979
under the terms of the Brezhnev Doctrine, which promised to prop up troubled
Communist regimes, led directly to a revival in Islamic fundamentalist politics,
which spilled over the Soviet border. Anti-Soviet sentiment had already been
inspired by the dramatic surge of radical religious politics during the Iranian
revolution earlier that year. Muslim volunteers or Mujahideen joined the pro-
Azerbaijan forces during the Afghan war, just as they were to help the Bosnian
government in the early 1990s. The long-term result of the war was the exodus of
non-Armenians from Nagorno-Karabakh and the deaths of up to 30,000 people
and displacement of thousands. Armenians now constitute over ninety-five per
cent of the population. Although now the region is now de facto independent of
Azerbaijan and enjoys the status of a ‘quasi-state’, it has been heavily dependent on
Armenia for trade and most other forms of support, in the way that Republika
Srpska has been reliant on Serbia.9 Like Bosnia it remains one of the most heavily
land-mined parts of the world with several hundred deaths and injuries reported
since the conflict began.10
6 John B. Allcock, Explaining Yugoslavia (London: Hurst, 2000), 383.

7 Roy Medvedev and Giulietto Chiesa, Time of Change: An Insider’s View of Russia’s Transformation
(London: I. B. Tauris, 1991), 224.
8 The events of that year were dramatically described in Luigi Villari, Fire and Sword in the
Caucasus (London: Unwin, 1906).
9 Pål Kolst, ‘The Sustainability and Future of Unrecognized Q uasi-States’, Journal of Peace
Research 43:6 (2006), 733.
10 Physicians for Human Rights, Landmines: A Deadly Legacy (New York: Human Rights Watch,
1993), 144.
THE CHECHEN CRISIS AND

T H E R U S S I A N F E D E R AT I O N
................................................................................................................
Although largely Sunni Muslim Chechens were a clear majority in Chechnya, the
capital, Grozny, had a large number of Russians living there before 1991, perhaps
fifty per cent of the population. Relations between the two groups, difficult since
Russian expansion in the late eighteenth century, were badly enflamed by the
deportation of Chechens to Kazakhstan and Siberia in 1944 on the grounds that
the whole nation was deemed to have collaborated with the Nazis. Many were also
executed before deportation or died en route, a tragedy that played an important
part in the repudiation of the Soviet legacy.11 Effectively barred from political
participation until the mid-1980s, the All-National Congress of the Chechen People
(NCCHP) sensed the weakness of the authorities in turmoil in Moscow in Sep-
tember 1991 and killed the local Soviet representative, moving towards de facto
independence. The Russian President Boris Yeltsin opposed these moves, sending
in troops in November 1991. The Chechens resisted, declaring independence in
1993. Unlike the other new states that sprung from the ruins of the Soviet Union,
defined as ‘union republics’ in the Soviet Constitution, Chechnya did not have the
legal ‘right’ to succeed, a status not recognized by the NCCHP, but crucial to our
understanding of why Moscow opposed this independence so bitterly, when it had
accepted the departure of Ukraine and the other former republics.12 Many Muslim
states recognized Chechnya, creating further regional polarization. The situation
between ethnic groups began to deteriorate. Russians and other non-Chechens
living there were subjected to intimidation and began to leave. The government of
Dzhokhar Dudayev, who like Tudjman and Mladić had been a Communist general,
faced internal opposition, which Yeltsin exploited. In December 1994, the Russians
invaded Chechnya and came under heavy international criticism for their intense
bombing of the capital, Grozny, and use of special troops from the interior ministry
who hunted down and slaughtered Chechen fighters in a so-called zachistke or
‘cleansings’. As the Chechen troops retreated from their lost capital, they killed as
many Russian soldiers as they could and were joined by Mujahideen from other
parts of the Muslim world. Dudayev, who had grown up as the child of exiles in
Kazakhstan and imbibed their resentment,13 was killed by the Russian military in
11 Brian Glyn Williams ‘Commemorating “The Deportation” in Post-Soviet Chechnya: The Role of
Memorialization and Collective Memory in the 1994–96 and 1999–2000 Russo-Chechen Wars’, History
and Memory, 12:1 (2000), 101–34.
12 Brian Stormo, ‘The Unsuccessful Secession of Chechnya’, International Relations Journal 1
(2004), 22.
13 Mike Bowker, ‘Conflict in Chechnya’, in Cameron Ross (ed.), Russian Politics under Putin:
Normality, Normalcy or Normalisation (Manchester: Manchester University Press, 2004), 256.
1996. His successor, Zelimkhan Yandarbijev, signed a peace agreement later that
year, sensing the Russian public’s lack of support for the war and Yeltsin’s declining
authority. Over half a million civilians were displaced between 1994 and 1996 and
many went to neighbouring Ingushetia.
In the late summer of 1999, the Russians started bombing Chechnya again in
retaliation for terrorist activities in Moscow and Chechen support for the ongoing
war in Dagestan, which had radicalized Russian public opinion. By the winter they
retook Grozny, committing atrocities in the village of Alkhan-Yurt and the suburbs
of the capital.14 Russian President Vladimir Putin introduced a new constitution in
May 2000, but still effectively faced guerrilla insurgency in the mountains and a
sharp rise in Islamic militancy. Action against the Chechens may have raised the
profile and popularity of Putin, as he deployed increasingly tough language and
military tactics, but also led to terrible ‘retribution’ in the form of bombings in
Moscow and Beslan in 2004. Over 200,000 Russians and 20,000 Armenians left
Chechnya in the 1990s, and non-Muslims now number less than five per cent of the
total population. At least 4,000 Russian soldiers died during the two wars and the
number of Chechen fighters killed may well be higher. The number of Russian and
Chechen non-combatants who died probably was at least twice this number.15
WA R AND ETHNIC CLEANSING IN SOUTH

OSSETIA AND ABKHAZIA
................................................................................................................
Like Nagorno-Karabakh, South Ossetia had also been granted oblast status within
the Georgian SSR. Prior to Soviet disintegration, just under thirty per cent of the
South Ossetian population was Georgian and Christian, the rest being largely
Muslim Ossetians (who also lived across the border in the Russian oblast of
North Ossetia). The new Georgian government wanted to dissolve the oblast status
and integrate the region into their state, although the Ossetian population had
already voted for pro-independence candidates in elections in October 1990. In
January 1991, the Georgian Army entered the capital Tskhinvali and were reported
to have killed over 1,000 people before a ceasefire was announced the following
year, backed by Moscow.16 About 100,000 also fled to North Ossetia during the
14 James Hughes, Chechnya: From Nationalism to Jihad (Philadelphia: Pennsylvania University

Press, 2007), 120.
15 Matthew Evangelista, The Chechen Wars: Will Russia Go the Way of the Soviet Union?
(Washington, DC: Brookings Institution, 2002), 84.
16 Mike Bowker, Russia, America and the Islamic World (Dartmouth: Ashgate, 2007), 143.
fighting. Distance from the Georgian government increased and a referendum on

independence was passed in 2006, supported by ninety-nine per cent of those who
voted. This was boycotted by ethnic Georgians, just as Serbs boycotted referenda in
Kosovo and Bosnia in the 1990s. On 8 August 2008, Georgian troops re-entered
Tskhinvali, provoking an armed response from the Russian Federation. ‘Volun-
teers’, many of whom were from outside South Ossetia, had already begun to attack
Georgians living there and the government of Mikheil Saakashvili reacted to
documented reports of ethnic cleansing. After Russian troops invaded Georgian
territory proper the following week, bombing the town of Gori, Saakashvili agreed
to a ceasefire. According to the terms drawn up by French President Nicolas
Sarkozy, the International Community would open ‘international discussions’ on
the regions stability and security. The rhetoric of genocide, self-defence, and
humanitarian intervention was used by both sides.
The Black Sea region of Abkhazia, which had been an autonomous republic
within Soviet Georgia, resisted incorporation into Georgia in 1992. After Georgia
lost a war against the Abkhaz and their Russian allies in 1993, at least half of the
Georgian population and most of the other ethnic groups numbering about
250,000 fled or were forced to leave (in some cases by Muslim Chechens who
had joined the Abkhaz). As political divisions have hardened, this is likely to be a
permanent displacement. In late August 2008, the Russian Federation recognized
the independence of both Abkhazia and South Ossetia. Earlier that month they also
had backed the Abkhaz militarily and moved into Georgian territory around the
town of Poti before the Saakashvili government capitulated. Arguably the recogni-
tion of Kosovo in 2008 gave considerable impetus in both cases of secession and has
‘unfrozen’ conflicts which had been previously thought to have stagnated.17 For-
getting Chechnya, Putin had already declared in 2007 that ‘if we decide . . . that the
principle of self-determination is more important . . . than territorial integrity, then
we must apply this principle to all parts of the world and not only to regions where
it suits our partners.’18
CRISIS AND GENOCIDE IN Y U G O S L AV I A

................................................................................................................
When the Communist partisans began to establish a new government during the
War in Yugoslavia (1941–5), they realized their ultimate success lay in balancing the
interests of the different nationalities that had been the subjects of the Karadjordjević
17 Rick Fawn, ‘The Kosovo—and Montenegro—effect’, International Affairs 84:2 (2008), 269.
18 Ibid. 286.
monarchy before the war. Fanatically loyal to their charismatic leader Josip Broz Tito,
Yugoslavs remained united by the policy of ‘brotherhood and unity’ until the 1970s or
even later in some republics, notably Bosnia. The internal boundaries drawn up in
1946 by the Stalin-influenced Edvard Kardelj, and revised again by him in 1974, were
dramatically unstable after the demise of Communism. Although the new Yugosla-
vian state had been ‘largely manned by the Serbs of Bosnia and Croatia’,19 it had also
been specifically designed to keep Serbia weak in a confederation in which they
represented forty-two per cent of the overall population. By the death of Tito in
1980, cracks in the system began to emerge, especially as the 1974 Constitution
designated the Serbian regions of Kosovo and Vojvodina as ‘autonomous’, a status
they had been denied in 1946.20
The legacies of previous wars and genocide remained part of the lived experience
of the subsequent generations. More than one million Yugoslavs had died in the
1940s, most the victims of other Yugoslavs rather than the occupying Germans and
Italians. Many were ‘possessed’, as the historian Milorad Ekmečić put it, by their
memories of the 1940s.21 Ivan Čolović argued that the ‘discourse of warlike ethnic
nationalism places contemporary events . . . outside the co-ordinates of historical
time . . . (offering) a mythic, anti-historical perception of time.’22 Certainly during
the 1990s, past grievances frequently seemed more important to extremists than the
human rights of the living.
Some Serbs also feared that increasing population growth amongst the largely
Muslim Kosovo Albanians (Kosovars) threatened the status of Serbs living there.
Ethnic Serbs had begun to drift out of the region attracted by jobs in industrial
centres and their numerical share of the population dwindled to just over ten per
cent (it had been more than twenty-five per cent when Serbia annexed Kosovo
from the Ottomans in 1912). Although few were inclined to actually live there, Serbs
felt attached to their medieval legacy, particularly splendid Orthodox monasteries,
and regarded it as a fault line between Christian and Islamic civilization. The
Communists had dithered over the status of Kosovo in 1945, ultimately preferring
to keep it rather than hand it over to Enver Hoxha in Albania. This decision was
made to avoid ethnic violence and a Serb backlash, but perhaps merely prolonged
the controversial question of the region’s status.
After the death of Tito, nationalist Serbs became less restrained about criticizing
the new status of Kosovo. They claimed, without much empirical foundation, that
19 Stevan K. Pavlovitch, ‘Serbia and Yugoslavia—The Relationship’, Southeast European and Black
Sea Studies 4:1 (2004), 102.
20 Daniele Conversi, ‘Central Secession: Towards a New Analytical Concept? The Case of Former
Yugoslavia’, Journal of Ethnic and Migration Studies 26:2 (2000), 340–1.
21 Tim Judah, The Serbs: History, Myth and Destruction of Yugoslavia (New Haven, CT: Yale
22 Ivan Čolović, The Politics of Symbol in Serbia: Essays in Political Anthropology (London: Hurst,
2002), 13.
they were subject to ‘genocide’ in the region,23 that women were frequently raped,
and that the whole population felt intimidated by Kosovars. Popular fears were
stirred up by the media, the church, and the Serbian Academy’s memorandum of
1986, which addressed the ‘tortured’ question of Kosovo.24 In this atmosphere of
frenzy and a sharp rise in publically expressed nationalism across the country, an
apparatchik Slobodan Milošević was able to manoeuvre and intimidate his way
into power. He brought a new rhetoric to the masses by promising to defend the
rights of Serbs everywhere and by surrounding himself with a retinue of Kosovo
Serb squadristi. Serb nationalists simultaneously revived plans set out by the
extremist Četnici,25 who had murdered Bosnian Muslims and Kosovo Albanians
during the 1940s.26 At the same time, these nationalists accused the Croatians, led
by dissident historian Franjo Tudjman, of being descended from the Ustaša, who
had run a Nazi-puppet state from 1941 to 1945 and committed genocide against the
Serbs.
Shortly after the outbreak of fighting in Bosnia, the British newspaper The Daily
Mirror ran a headline that reverberated around the world: ‘Belsen 92: The picture
that shames the world’. The picture was of emaciated prisoners in an ad hoc
detention centre that had been set up by the Serbs in their breakaway republic.
At the time, comparisons between the terrible predicament of the Bosnian Muslims
(and the Croatians in the previous year) and Third Reich victims seemed apt as
propaganda, particularly as international intervention to stop the fighting during
the first post-Communist Balkan crisis was deemed to be so ineffective. The
Croatian sociologist Stjepan Meštrović insisted that the war in Croatia was ‘geno-
cide’, as did his colleague Norman Cigar in relation to Bosnia.27 In 2007, Serbia as a
state was cleared of genocide at The Hague Tribunal (ICTY). The international
court did rule, however, that the slaughter of an estimated 8,000 men and boys as
‘combatants’ at Srebrenica in July 1995 was genocide. In 1999, Milošević became the
first serving head of state to be indicted for war crimes. Many regarded him as the
primary instigator of the crisis because of the support he gave to Serb nationalists
23 The use of the word genocide when referring to the declining position of Serbs in Kosovo was
popularized by Dimitrije Bogdanović, Knjiga o Kosovu (Belgrade: SANU, 1985), 7.
24 Bože Čović (ed.), Izvori velikosrpske agresije (Zagreb: August Cesarec and Skolska knjiga, 1991),
296–300.
25 Safet Bandžović, ‘Koncepcije Srpskog kulturnog kluba o preured-enju Jugoslavije 1937–1941’,
Prilozi 30 (2001), 163–93.
26 On Bosnia see, Tomislav Dulić, Utopias of Nation: Local Mass Killing in Bosnia and Herzegovina,
1941–42 (Uppsala: Acta Universitatis Upsaliensis, Studia Historica Upsaliensia, 2005); and Vladimir
Dedijer, Genocid nad muslimanima 1941–45, Zbornik documenta i svjedočenja (Sarajevo: Svjetlost,
1990). On Kosovo, see Jozo Tomasevich, War and Revolution in Yugoslavia: Occupation and
Collaboration (Stanford: Stanford University Press, 2001).
27 Stjepan G. Meštrović, Genocide after Emotion: The Post-Emotional Balkan War (London:
Routledge, 1996); Norman Cigar, Genocide in Bosnia: The Policy of ‘Ethnic Cleansing’ (College
Station: Texas A&M University Press, 1995).
between 1987 and 1999. The judgement against him would have been an important
landmark in legal history, but his death while on trial in 2006 prevented his case
from being concluded.28
WA R IN C ROAT I A 1990–5
................................................................................................................
Milošević and his supporters precipitated the break-up of the League of Commu-
nists in January 1990. The Yugoslavian republics, now with increasing autonomy
and acting like nations-states in waiting, moved towards free elections, choosing
from a range of non-Communists, many of whom were nationalists. In Croatia, a
government was formed by Tudjman, notorious for his 1990 revisions of the
estimated numbers killed in fascist Croatia, especially the Jasenovac death
camp.29 His government made further provocative gestures towards the minority
Serbs, banning Cyrillic road signs and rehabilitating Ustaša fascists such as Mile
Budak.30 In his role as a propagandist, Budak had revived the Habsburg anti-Serb
1914 slogan ‘Srbe na vrbe’ (‘Hang Serbs on the willows’),31 and the significance of
the attempt to restore his reputation in 1990 was not lost on local Serbs. During the
Second World War, the Ustaša government had perpetrated what Tomislav Dulić
has called an ‘attempted genocide’, which led to the deaths of sixteen to seventeen
per cent of the total Serb population.32 The element of trauma amongst Serbs
should not be underestimated. They had also suffered huge losses during the First
World War due to aggressive military actions by the Habsburg Army and a terrible
typhus epidemic.33 With their fear of Islamic encroachment, which stemmed from
the years of Ottoman oppression and the legacy of earlier genocides,34 ordinary
28 For a discussion of the trial see William A. Schabas, ‘International Justice for International
Crimes: An Idea whose Time Has Come’, European Review 14 (2006), 421–39.
29 Franjo Tudjman, Bespuća povijesne zbiljnosti: rasprava o povijesti i filozofiji zlosilja (Zagreb:
Nakladni zavod Matice hrvatske), 1990.
30 Dubravka Ugrešić, The Culture of Lies: Antipolitical Essays (University Park, PA: Penn State Press,
1998), 228.
31 Milan Bašta, Rat je završen sedam dana kasnije (Belgrade: OOUR, 1986), 151.
32 Tomislav Dulić, ‘Mass Killing in the Independent State of Croatia, 1941–1945: A Case for
Comparative Research’, Journal of Genocide Research 8:3 (2006), 274.
33 Mark Levene, Genocide in the Age of the Nation State, vol ii: The Rise of the West and the Coming of
Genocide (London: I. B. Tauris, 2005), 323; Andrej Mitrović, Serbia’s Great War 1914–1918 (London:
Hurst, 2007).
34 Cathie Carmichael, ‘“Neither Water, nor Wine, Neither Turks nor Slavs, But Odious Renegades”:
Anti-Islam and Ideologies of Ethnic Cleansing in the Balkans’, in Steven Vardy and Hunt Tooley (eds),
Ethnic Cleansing in Twentieth Century Europe (New York: Columbia University Press, 2003), 113–32.
Serbs were highly vulnerable to cynical nationalist politicians particularly in areas

such as Krajina, where bodies had been hastily disposed of in the limestone ravines
in the 1940s and were only then being uncovered and reburied.35
Croatian independence triggered a Serb reaction at a time when many across
Yugoslavia had been radicalized by the Kosovo issue. By the time of full indepen-
dence, Tudjman faced an open insurrection led by the Serbian Democratic Party
(SDS), which attempted to form a breakaway ‘autonomous region’ within Croatia.
Terrible fratricidal violence took place in the Krajina region, Slavonija, and the
Adriatic hinterland in late 1991. ‘Ethnic cleansing’ (a term first introduced by the
Serbs themselves to describe their putative fate in Kosovo) was widespread and
horrific. The Danubian town of Vukovar was reduced to a pile of rubble by Serb
paramilitaries backed by remnants of the Yugoslavian National Army (JNA), who
were better armed and trained, although individual soldiers were often demora-
lized by the crimes they witnessed.36 A paramilitary group led by Arkan (the
gangster Željko Ražnatović) killed about 250 Croats in a nearby abattoir.37 Land-
mines were planted in the Dinaric coastal region, accompanied by the gratuitous
destruction of hotels, buildings, orchards, and farms to deter Croats from return-
ing or to ‘punish’ them. At the same time, the Croatian government tolerated
abuses towards Serbs remaining within their territory, most of whom left the
capital, Zagreb, between 1991 and 1995.38 More than 10,000 civilians were killed
in Croatia and by 1995 as many as 300,000 people had been displaced. The United
Nations moved in during January 1992 to supervise the borders between the
autonomous regions and government-held territory, which restored a semblance
of peace, but at the price of dividing and separating the two communities. In
August 1995, after substantial US-backed rearmament the Croatian government
took Krajina in ‘Operation Storm’ and most of the Serb population left, fearing for
their safety. The Croatian Ambassador to France described the actions of the Serbs
as ‘auto-nettoyage ethnique’ (self ethnic-cleansing),39 but Gen. Ante Gotovina who
led the campaign has been accused of crimes against humanity with respect to the
remaining Serbs and is currently on trial in the Hague. As in Nagorno-Karabakh
from whence most Azeris fled, Croatia effectively lost most of its historic Serb
minority in 1995 and since then very few have returned with those remaining in
ghettos. For a triumphant President Tudjman, the exodus was the solution to
35 Bette Denich, ‘Dismembering Yugoslavia: Nationalist Ideologies and the Symbolic Revival of
Genocide’, American Ethnologist 21:1 (May 1991), 367–90.
36 Vukovar. Poslednji Rez (Belgrade: B92 2006), documentary written and produced by Janko Baljak
and Drago Hedl.
37 Marcus Tanner, Croatia: A Nation Forged in War (New Haven, CT: Yale University Press, 1997),
266–7.
38 Brad K. Blitz, ‘Refugee Returns, Civic Differentiation, and Minority Rights in Croatia 1991–2004’,
Journal of Refugee Studies 18:3 (2005), 362–86.
39 Smiljan Simac, ‘Croatie, Serbie: les fausses symétries’, Le Monde (25 May 1999), 13–14.
Croatia’s ‘centuries-old problem’.40 Denying any responsibility for the act, he

stated: ‘We didn’t ask the Serbs to leave. And it’s their problem if 90%of them
packed their bags.’41
WA R AND GENOCIDE IN B O S N I A -H E RC E G OV I NA
................................................................................................................
In many respects, events in Bosnia resembled the deteriorating circumstances in
Croatia in 1991. In parliament, the SDS led by Radovan Karadžić threatened the
government of President Alija Izetbegović with ‘hell’ if Bosnia-Hercegovina left
Yugoslavia.42 Despite this threat, the government called a referendum on indepen-
dence in April 1992, which was boycotted by almost one-third of the population of
the republic, largely on the instruction of the SDS (much as the December 1991
referendum in Nagorno-Karabakh had been boycotted by the Azeri minority of
more than twenty per cent of the population). Of those who did vote, over ninety-
nine per cent opted for full independence from what remained of Yugoslavia.
Karadžić argued that independence was illegal without the support of all three of
the ‘constituent’ Bosnian nationalities and that they would be in the majority had
they not been subjected to genocide in the past. The referendum revealed just how
deep the chasm between those who wanted to leave Yugoslavia and those who
wanted to be in a Serb-dominated state had become.
From this inauspicious foundation, the government declared independence and
promptly faced widespread and highly organized Serb resistance. The armed Serbs
rapidly took over the northern and eastern parts of the country. Just as Zagreb had
remained with the government during the Croatian war, the capital, Sarajevo,
remained a stronghold of the Bosnian government. Muslims fled into the towns
such as Goražde, Srebrenica, and Žepa, which became virtual ghettos, often
without transport links, medical supplies, or basic foodstuffs. In the first weeks
Serb paramilitaries, effectively protected by the JNA, murdered civilians, especially
the vulnerable elderly and left their mutilated bodies on display. The British
journalist Michael Nicholson described their tactics as ‘elitocide’ when it became
clear that the Serbs had also attacked Muslim and Croat political activists and
prominent community leaders.43 Paramilitaries also killed dozens of foreign
40 James Gow, Triumph of the Lack of Will: International Diplomacy and the Yugoslav War (London:
Hurst, 1997), 43.
41 Florence Hartmann, Miloševič: La Diagonale du Fou (Paris: Denoël, 1999), 244.
42 Ibid. 247.
43 David Rieff, Slaughterhouse: Bosnia and the Failure of the West (New York: Simon and Schuster,
1995), 113.
journalists that summer. Backed by an armed wing of the SDS, the Bosnian Serb
Army (VRS) was quickly formed in 1992 and commanded by former Communist
general Radko Mladić, making Bosnia a country under occupation. The Serb forces
set up prison camps at Manjača, Prijedor, and Omarska, inviting ‘weekend war-
riors’ or war tourists to torture their former neighbours,44 thus spreading guilt and
complicity beyond the SDS, the VRS, and the ranks of paramilitaries. The alarming
incidence of crimes against humanity led to the establishment of the International
Criminal Tribunal for the former Yugoslavia (ICTY) to punish breaches of inter-
national law, which was convoked in The Hague in 1993 and has subsequently
sentenced over 100 of those indicted.45
President Izetbegović, a Muslim dissident imprisoned for his views by the
Communists, appealed to the International Community to help him defend the
newly recognized state. During the early stages of the war, a UN embargo on
importing armaments had crippled his government’s ability to defend itself, as
the remnants of JNA armaments were in the hands of the Serb rebels. International
ineptitude made the fate of Bosnia worse, delaying government military successes.46
Peace treaties such as the one set forward by Cyrus Vance and David Owen in 1993,
which proposed dividing Bosnia into ethnic ‘cantons’, may have speeded up the
ethnic cleansing of some regions to make Serb and Croat claims stronger. Embol-
dened by plans to divide the Bosnian state, Croat Defence Council paramilitaries
(HVO) began targeting Bosnian Muslims in Hercegovina in 1993, particularly in the
town of Ahmići, in which they killed over 100 civilians. International human rights
lawyer Payam Akhavan arrived there soon afterwards, describing the ‘all-embracing
form of the destruction of this village and its inhabitants’.47 The Croatian officer
Tihomir Blaškić was deemed by the ICTY in 1997 to have had some ‘command
responsibility’ for the massacre and has since served a nine-year sentence.
After UN Protection Force intervention in June 1992, the fighting stagnated
somewhat, except for an ongoing siege of the capital, Sarajevo, designed to lower
government morale, and the dramatic battle for the crucial city of Bihać, which
would have linked Serb gains in Northern and Eastern Bosnia with Krajina. Many
Bosnian Muslims were confined to so-called ‘safe havens’ in the Eastern towns
supervised by the UN, but lacked supplies and were in effect trapped. Under the
terms imposed by the UN they were also disarmed, which left them completely
dependent on the international community for defence. By 1995, the Bosnian
government had begun to reverse the military stagnation of the previous three
44 One of the best accounts of Bosnian camp atrocities is Roy Gutman’s, A Witness to Genocide: The
First Inside Account of the Horrors of Ethnic Cleansing in Bosnia (Shaftesbury: Element, 1993).
45 http://www.un.org/icty/
46 There are several astute criticisms of international ineptitude, including Brendan Simms,
Unfinest Hour: How Britain Helped to Destroy Bosnia (Harmondsworth: Penguin Press, 2001).
47 http://www.un.org/icty/kupreskic/trialc2/judgement/kup-tj000114e-3.htm, accessed on 27
January 2008.
years. As the rebel Serbs were pushed out of Krajina by Croatian rearmament and
military action and they had failed to take the UN ‘safe haven’ Bihać despite huge
losses, their military tactics became more desperate and the VRS more demora-
lized.48 Inadequate protection from Dutch UN troops in July 1995 led to the
surrender of the ‘safe haven’ of Srebrenica to Mladić and the slaughter of almost
the entire Muslim male population of about 8,000. Over 100,000 people including
soldiers and civilians died as a result of the war, most either in the summer of 1992
by paramilitaries or during the sieges of Sarajevo and Bihać. By far the highest
number of individuals killed were an estimated 65,000 Muslims, although 25,000
Serbs and 8,000 Croats also died. Bosnia’s Muslims and Serbs were also heavily
displaced and an estimated two million left either temporarily or to rebuild their
lives elsewhere.
In late 1995, Bosnian politicians eventually signed the Dayton Peace Treaty,
brokered primarily by the USA. The treaty created an odd administrative arrange-
ment, with Serb-controlled territory (‘Republika Srpska’) to the east and the
Bosnian Muslims and Croats in control of the south-west and capital. The treaty
may have created the precondition for future separate states. It did allow for the
return of some refugees; most towns have not recovered their previous status.49
The communities remain effectively divided with the past being kept alive by
successive administrations. Bomb craters in the capital have since been painted
red and are known as ‘Sarajevo roses’.50 In 2005, a poster commemorating the tenth
anniversary of the genocide at Srebrenica was defaced with ‘it will happen again’.51
Extremism remained in evidence in neighbouring republics. In the second round
of the 2008 presidential elections in Serbia, the candidate Tomislav Nikolić for the
Radical Party, heavily linked to the paramilitaries in the 1990s, won 47.97 per cent
of the popular vote.
T H E K O S OVO C R I S I S
................................................................................................................
The issue of Kosovo, which remained untouched at Dayton, entered Serbian
politics in the mid-1970s and has remained there ever since, fuelling a growth in
48 Phillip Corwin, Dubious Mandate: A Memoir of the UN in Bosnia, Summer 1995 (Durham, NC:
Duke University Press, 1999), 202.
49 Francine Friedman, Bosnia and Herzegovina: A Polity on the Brink (New York, Routledge, 2004), 80.
50 Fran Markowitz, ‘Census and Sensibilities in Sarajevo’, Comparative Studies in Society and
History 49:1 (2007), 49.
51 Paul B. Miller, ‘Contested Memories: The Bosnian Genocide in Serb and Muslim Minds’, Journal
of Genocide Research 8:3 (2006), 313.
extremist political groups, especially the Serbian Radical Party (SRS). The 1983
funeral of Aleksandar Ranković, former Minister of the Interior who had favoured
Serb domination in the region, was an ominous sign of popular discontent as
people lined the streets to pay their respects.52 More than any other Communist
politician, he had been associated with Belgrade centralization and therefore seen
as a foil to Kardelj, architect of the loathed 1974 Constitution. The 1974 rights of
Kosovo were effectively revoked by Milošević in 1989. Thereafter until 1999, the
region was effectively run from Belgrade without significant cooperation with the
vast majority of Kosovars, a situation as ‘separate worlds’53 or even ‘Apartheid’.54
Led by Gandhi-inspired Ibrahim Rugova, eighty per cent of the region’s population
voted for independence in 1992 by a margin of ninety-eight per cent. This vote,
which had been boycotted by the region’s Serbs, was also ignored internationally.
Rugova kept the door open to dialogue with Milošević, even during the NATO
bombing of the Federal Republic of Yugoslavia (FRY) in 1999. While his strategy
kept a semblance of peace, a whole generation of Kosovo’s young people were
deprived of any participation in the state. A large male diaspora working abroad
from 1981 onwards contributed to the formation of a guerrilla Kosovo Liberation
Army (KLA). Ironically its strategy of armed resistance had certain similarities with
the uprising in the ‘Serbian Autonomous Regions’ (SARs) in 1990–2, but this time
with the cautious support of the many UN countries, who saw the Serbs as the
primary villains of the piece. The similarity may not be entirely coincidental.
Kosovars had also been JNA conscripts, imbibing Tito’s philosophy of preparedness
for defence. The KLA, which also drew on local Kaçak bandit traditions in areas like
Drenica,55 deliberately provoked clashes with the Serb authorities and increasingly
found support within their community. In 1998, one of their founders, Adem
Jashari, and his entire family were killed by the police, a massacre that led to an
international backlash. The Belgrade government received a number of ultimatums
from an international community led by the NATO countries in Rambouillet in
1999, which were primarily designed to avoid another Bosnia.
Milošević found himself caught between internal nationalism and international
disapproval at Rambouillet. Unable to abandon Kosovo and agree to the stipulated
referendum due to the wrath of his own nationalists, and definitely unable to
withstand NATO airpower in the long run, he allowed the Yugoslavian Army (VJ)
52 Mitja Velikonja, Religious Separation and Political Intolerance in Bosnia and Herzegovina (College
Station, Texas: Texas A&M Press, 2003), 223.
53 Shkëlzen Maliqi, Kosova: Separate Worlds: Reflections and Analyses 1989–1998 (Priština:
Dukagjini, 1998).
54 Sevdie Ahmeti, ‘Forms of Apartheid in Kosovo’, in Dušan Janjić and Shkëlzen Maliqi (eds), in
Conflict or Dialogue: Serbian-Albanian Relations and the Integration of the Balkans (Subotica: Open
University, 1994), 205–25.
55 Michel Roux, Les Albanais en Yougoslavie: Minorité nationale, territoire et développement (Paris:
Foundation de la maison des sciences de l’homme, 1992), 214–15.
to pursue a desperate policy of ethnic cleansing, possibly believing that terrible acts
of violence would depopulate Kosovo as it had done in Bosnia and parts of Croatia,
and refugees would have to remain in neighbouring Macedonia and Albania where
they had fled. The VJ killed over 10,000 Kosovars during a campaign of air strikes
upon Serbia and Montenegro in 1999 that did not see a single NATO fatality. Up to
500 non-Kosovar citizens of the FRY were also killed by NATO as it started to drop
cluster bombs in an attempt to break morale, especially in some of the larger cities.
Individuals associated with the opposition, such as the journalist Slavko Čuruvija,
were assassinated and an atmosphere of panic and recrimination against the
Milošević regime developed, which resulted in his dramatic downfall the following
year. The Kosovars, emboldened by NATO support, returned to their shattered
homes in the summer of 1999 after the surrender of the FRY. They then began a
process of revenge against local Serbs and Roma (whom they felt had supported the
regime) and as many as 700 were killed that year.
After the war, the region was administered by the UN. More Serbs left over the
following years, leading to a further decline in their numbers. In 2005, the UN
appointed a special envoy, Martti Ahtisaari, to negotiate a situation, which led to
full independence for Kosovo in 2008. The declaration precipitated a Serb backlash
in areas in the north of the country, particularly in the town of Mitrovica, which
has effectively resisted the authority of the new state just as the Serbs had done in
Krajina from 1990 to 1995.
VIOLENCE, CYNICISM, AND

P O S T-C O M M U N I S T I D E N T I T I E S
................................................................................................................
The break up of Yugoslavia cost well over 120,000 lives and led to the displacement
of millions. The root cause of violence was Serb discontent with the borders of the
successor states, which they rejected in order to establish their own quasi-states
before the new administrations had a chance to oppress them as a minority. Extreme
elements revived fascist plans for ethnic cleansing from the Second World War,
during which time Kosovars and Bosnian Muslims had been attacked by Serb
nationalists and Serbs by the Croatian Ustaša and Kosovars. Serb extremists aroused
international opposition primarily because they took up violence as their first resort.
Undoubtedly distrust of Muslim Kosovars and Bosnians as well as a hatred of
Croatian nationalism existed at the level of Serb popular culture, but these emotions
were not the primary cause of the fighting. The war was one of intricate strategy to
gain as much territory and people for any future Serbian state(s) as possible and
frequently coordinated by nationalists from within Serbia itself (or ‘joint criminal
enterprise’ around Milošević as it was called at The Hague). This strategy failed in
Kosovo, Krajina, and Croatia but worked in Bosnia, where violence was ‘rewarded’
by the Dayton Treaty. But this was a pyrrhic victory indeed, as in 2006 Serbia lost its
long-term ally Montenegro when the latter voted for complete independence,
severing it from the Adriatic coast. A considerable number of Montenegrins who
voted for independence were displaced Albanians and Bosnians who had taken up
residence there.
There was very little evidence of interethnic ‘hatred’, although the international
media frequently discussed ‘violent’ Balkan mentalities. Stjepan Meštrović and his
colleagues, for instance, stated that ‘the Yugoslav civil war exhibited barbaric acts of
cruelty, massacres and the mutilation of the living as well as corpses that they beg
for an explanation. Our explanation is that when one examines the history of the
Balkans, such savagery appears to be fairly typical.’56 This view was common and
pronouncements by academics appear to have encouraged policy makers. Speaking
in February 1993, primarily to prevent the implementation of the UN Genocide
Convention, US Secretary of State Warren Christopher thought that ‘the end of
communist domination of the former Yugoslavia raised the lid on the cauldron of
ancient ethnic hatreds . . . It has long been the cradle of European conflict and
remains so today.’57 But there were certainly more plausible and less essentialist
explanations. For the sociologist Bojan Baskar the ‘cynical’ nature of post-Com-
munist politicians and their readiness to use liminal elements within society such
as convicted prisoners, hooligans, and gangsters to break the Hague Conventions is
a more plausible explanation for violence, which was hardly universal through the
former Yugoslavia.58 Muslims living in Serbia were generally left alone during the
entire Bosnian war, as were over 100,000 Kosovars living in Belgrade in 1999. Those
atrocities that did take place were carried out to remove populations from land
during a period of the breakdown of authority, not primarily because of ‘ancient
hatreds’.
To what extent can it be stated that the wars in Yugoslavia in the 1990s constitute a
clear case of genocide? The nationalist Serbs wanted to remove non-Serbs from
regions that they claimed and they were prepared to kill them en masse in the
process, a strategy revealed most starkly in Eastern Bosnia in the summer of 1995 and
Kosovo in 1999. Ethnic cleansing ‘bled’, as Norman Naimark suggested, into geno-
cide.59 Unlike the apocalyptic Nazis, they did not consider it their ‘fundamental
56 Stjepan Meštrović, Slaven Letica, and Miroslav Goreta, Habits of the Balkan Heart: Social
Character and the Fall of Communism (College Station: Texas A&M University Press, 1993), 61.
57 Tim Allen and Jean Seaton, The Media of Conflict: War Reporting and Representations of Ethnic
Violence (London: Zed Books, 1999), 1.
58 Bojan Baskar, Dvoumni Mediteran. Študije o regionalnem prekrivanju na vzhodnojadranskem
območju (Koper, Knjižnica Annales, 2000), 156.
59 Norman M. Naimark, Fires of Hatred: Ethnic Cleansing in Twentieth Century Europe (Cambridge,
duty’ to kill all of their perceived enemies, but as many as necessary to gain what they
wanted.60 In doing so they destroyed ancient populations and civilizations as well as
much of their material culture. In their attempts to control regions they claimed,
armed Serb extremists committed acts designed not only to kill individuals, but also
to destroy morale and break community cohesion such as gang rape and torture.61
In 1993, the European Commission estimated conservatively that 20,000 Bosnian
women had been raped, which was ‘among the largest documented cases of this
kind in recent history’.62
One of the major problems that these wars raised was the integrity of state
borders, the autonomy of individual states, and consistency of political principles.
The international community chose to recognize the Izetbegović government
despite the fact that a third of the citizens of Bosnia did not. It chose to recognize
Tudjman even with open rebellion in Krajina. By ignoring Kosovo for years,
regardless of the wish of eighty per cent of its population for independence as
reflected in the 1992 referendum, Serb nationalists received a clear message that the
international community did not care what happened within the borders of the
FRY so long as they did not have to deal with it. This pattern appears to have
continued in international relations. In 2007, US Secretary of State Condoleeza
Rice declared that South Ossetia was ‘part of Georgia . . . we believe in Georgia’s
territorial integrity,’ thus ignoring the referendum of the previous year in which
ninety-nine per cent of those who took part opted for full independence.63 Only
after direct Russian military intervention did her government recognize the need
for ‘discussions’ on the region’s status.
The conflicts in the Caucasus and Balkans were primarily caused by the collapse
of Communist authority, unresolved border questions in contested areas often
with large minorities, and the frequent willingness of post-Communist leaders to
break international laws with respect to the treatment of both captured soldiers and
non-combatants. With the exception of Croatia, the majority of violence in this
period was concentrated in regions that lay on the border between Islamic and
non-Islamic cultures. The Serb insurrection in Croatia created a violence exemplar
that ricocheted across the Balkans. The collapse of Soviet power coincided with or
even triggered a rise in Islamic militancy in the Caucasus and Middle East and
inspired guerrilla groups to form. When the Afghan Taliban entered the capital,
Kabul, in 1996, they mutilated the body of the former president Mohammad
Najibullah, whose government had previously committed atrocities against them,
60 Aleksa Djilas, The Contested Country: Yugoslav Unity and Communist Revolution 1919–1953
(Cambridge, MA: Harvard University Press, 1991), 123.
61 Caroline Kennedy-Pipe and Penny Stanley, ‘Rape in War: Lessons of the Balkan Conflicts in the
1990s’, in Ken Booth (ed), The Kosovo Tragedy: The Human Rights Dimension (London: Frank Cass,
2001), 67–84.
62 Stefan Wolff, Ethnic Conflict: A Global Perspective (Oxford: Oxford University Press, 2007), 103.
63 Fawn, ‘The Kosovo—and Montenegro—Effect’, 285.
and hanged him from a traffic light to symbolize the complete defeat of pro-Soviet
civilization. Chechen and Azeri militants have often viewed their fighting in terms
of restoring an Islamic moral order.64 In the former Yugoslavia, Serb nationalists
claimed that they were fighting against Islamic encroachment, which became
effectively a self-fulfilling prophecy after 11 September 2001. Although a rise in
religious consciousness may not wholly explain the occurrence of violence in these
regions, successionists and politicians eager to redraw boundaries cynically
exploited the idea of an ideological purpose and traditional identities under threat
to motivate and legitimize armed resistance.
Bracewell, Wendy, ‘Rape in Kosovo: Masculinity and Serbian Nationalism’, Nations and
Nationalism 6:4 (2000), 563–90.
Broz, Svetlana, and Laurie Kain Hart (eds), Good People in an Evil Time: Portraits of
Complicity and Resistance in the Bosnian War (New York: Other Press, 2004).
Chufrin, Gennadi, The Security of the Caspian Sea Region (Oxford: Oxford University Press,
2001).
Greene, Thomas, ‘Internal Displacement in the North Caucasus, Azerbaijan, Armenia and
Georgia’, in Roberta Cohen and Francis Mading Deng (eds), The Forsaken People: Case
Studies of the Internally Displaced (Washington DC: Brookings Institution, 1998), 233–312.
Gow, James, The Serbian Project and its Adversaries: A Strategy of War Crimes (London:
Hurst, 2003).
Hoare, Marko Attila, How Bosnia Armed: The Birth and Rise of the Bosnian Army (London:
Saqi Books, 2004).
Mertus, Julie, Kosovo: How Myths and Truths Started a War (Berkeley: University of
California Press, 1999).
Politkovskaya, Anna, A Small Corner of Hell: Dispatches from Chechnya (Chicago: University
of Chicago Press, 2003).
Ramet, Sabrina P., Nationalism and Federalism in Yugoslavia 1962–1991, 2nd edn (Bloo-
mington: Indiana University Press, 1994).
Sells, Michael, The Bridge Betrayed: Religion and Genocide in Bosnia, 2nd edn (Berkeley:
Thompson, Mark, Forging War: The Media in Serbia, Croatia, Bosnia and Hercegovina
(Luton: University of Luton Press/Article 19, International Centre against Censorship,
1999).
Williams, Brian Glyn, ‘Caucasus Belli: New Perspectives on Russia’s Q uagmire’, Russian
Review 4:4 (2005), 680–8.
64 Brian Glyn Williams, ‘The Russo-Chechen War: A Threat to Stability in the Middle East and
Eurasia?’, Middle East Policy 8, no. 1 (March 2001), 128–48.
chapter 26
.............................................................................................
G E N O C I DA L
WA R FA R E I N
N O RT H - E A S T
AFRICA
.............................................................................................
alex de waal
................................................................................................................
The modern history of the Horn of Africa is marked by protracted violence. The
two powerful states of the region, Ethiopia and Sudan, are hybrid imperial crea-
tions, each one an amalgam of African and European colonialisms. For centuries,
the dominant states of the Ethiopian highlands and the Nile Valley have been
predators on the peoples of their peripheries, inflicting slavery, subjugation, and
massacre upon them. The other states of the Horn, Eritrea and Somalia (and the
entities that replaced the latter), were forged out of resistance to the two dominant
centres of state power, and each exists insofar as it can dispense violence.1 Revolu-
tionary elites in both Ethiopia and Sudan have sought radical transformation of
their states, but ended up replicating and intensifying the patterns of violence they
inherited.
1 Djibouti, as a quasi-imperial military-commercial outpost, will not be discussed in this chapter.

530 alex de waal
The categories ‘war’ and ‘genocide’ do not adequately describe the nature of
violence in the Horn. Mass killing in the region has not been dealt with adequately
by genocide scholars and has yet to develop its own comparative and analytical
literature. Armed conflict is so protracted that peace is the exception. Normal is a
low level of conflict and insecurity, most of it perpetrated by states and state-
associated elites, but also involving armed formations contending for state power,
resisting state power, or pursuing other agendas. These persistent conflicts inter-
mittently erupt into episodes of extreme brutality in which thousands or tens of
thousands of non-combatants are killed. Massacres, subjugation of identity, forced
removals, and violent destruction of autonomous governance have all occurred.
The term ‘genocide’ has been indiscriminately used by local dissidents and foreign
critics to the extent that its political currency has been devalued. The US govern-
ment designated the events in Darfur, Sudan, during 2003–4 as genocide. Sadly, the
scale and nature of the killing, displacement, and famine in Darfur during that
period was not exceptional in the modern history of the Horn of Africa. In
contrast, good arguments could be made that half a dozen or so other episodes
in the previous fifteen years have as strong a claim to that label. The central
argument advanced in this chapter is that isolating individual episodes of killing
and displacement and labelling them as genocide is not a useful exercise. The tasks
before us are to explain the ubiquity of violence and the recurrence of massacre and
displacement of genocidal dimensions.
This chapter consists of four substantive sections. The first outlines the key
themes. A second part briefly surveys the position of the Horn of Africa within
scholarly and legal approaches to genocide. The major part outlines twenty-two
episodes of extreme violence, including mass killing and group-targeted repression,
over the past half century. The final section extracts some key themes and draws
some general conclusions.
V I O L E N C E , I D E N T I T Y, AND S TAT E H O O D
................................................................................................................
Violence is inscribed in the political formations of the Horn. Ethiopia and Sudan
are both conquest states, their modern boundaries carved out of myriad pre-
existing socio-political formations almost always through military suppression,
commonly followed by the imposition of garrison settlements or mercenary-
commercial entrepôts. They have common features and important differences.
Independent African states in the Ethiopian highlands, the Nile Valley, and the
Sudanic belt all consisted of an agrarian core that was settled and taxed and a
periphery from which tribute, slaves, and other resources were extracted.
genocidal warfare in north-east africa 531
Ethiopia was the classic African empire, ruled by a sovereign claiming descent
from the biblical King Solomon. Observers described the king as a despot whose
writ scarcely extended beyond the gates of his capital city2—a nice encapsulation of
the paradox of absolute rulers commanding weak structures of governance. Ethio-
pia retained its independence during the ‘scramble for Africa’ from the 1880s, its
Emperor Menelik II finding it possible to participate as a (junior) partner in that
imperial carve-up, extending his frontiers into areas where his predecessors had
only sent their most audacious raiding parties. By such means, the subjugation of
the Oromo and other peoples of the empire was accomplished.3 In the newly
conquered territories of the empire, the characteristic face of government was the
military governor and the armed settler, the latter known as neftegna, ‘rifleman’. A
racial-religious hierarchy existed among the Emperor’s subjects. A small minority
ruled. The aristocracy, clergy, and settlers applied the pejorative ‘Galla’ to the
Oromo serfs and the even more derogatory ‘Shankila’ to the enslaveable peoples
of the southern and western marchlands. Territorial conquest, the reduction of
subjugated peoples to commodities, and the infliction of legitimized violence
without recourse by the rulers characterized this imperial system.
Sudan was a variant. The states of the eighteenth- and nineteenth-century Nile
Valley and eastern Sahel, including the Funj Kingdom and the Fur Sultanate,
differed only in degree. They were smaller in scale and their rulers sought legitimacy
through Islam. In the nineteenth century they were overthrown by the ambitious
rulers of Ottoman Egypt and replaced by a hybrid empire on the Nile. Rather than
the government having its own military servants and financial structures, the state
was itself a tool of merchants and mercenaries. Outside the ‘metropolitan provinces’
along the river itself north of Khartoum, the characteristic face of governance was
the raiding party—the ghazwa or state-licensed freebooting expedition to acquire
ivory and slaves. The combination of Islam and colour differences created a deep
imprint of racism.4 Sudanese resistance to the conquest took the form of millenari-
an Mahdism, a revolution that expelled the Egyptians and their European merce-
nary generals and ushered in a messianic and also profoundly violent state that
lasted until its warriors were machine-gunned by the British army in 1898.
Colonial invasion was extremely violent in all parts of the Horn of Africa. The
British march up the Nile Valley from 1896 to the 1920s was marked by blood; the
Italian occupation of Eritrea and Somalia in the 1880s and finally Ethiopia itself in
1935 was accompanied by massacre, chemical weapons, and the murder of much of
2 Dame Margery Perham, The Government of Ethiopia (London: Faber and Faber, 1969).
3 Mohamed Hassan, The Oromo of Ethiopia: A History 1570–1860 (Cambridge: Cambridge
University Press, 1860); Bonnie K. Holcombe and Sisai Ibssa, The Invention of Ethiopia: The Making
of a Dependent Colonial State in Northeast Africa (Trenton, NJ: Red Sea Press, 1990).
4 Stephanie Beswick, Sudan’s Blood Memory: The Legacy of War, Ethnicity and Slavery in South
Sudan (Rochester, NY: University of Rochester Press, 2006).
532 alex de waal
Ethiopia’s educated class in a prison that gained the name ‘Alem Bekagn’, ‘farewell
to the world’.
Decolonization in the 1950s and 1960s was marked by more violence and the
beginnings of a series of insurrections that have continued, in one form or another,
for half a century. The creation and sustaining of the post-colonial entities of
Eritrea and Somalia involved both protracted violence against Ethiopia and inter-
nal violence. The intermittent emergence of South Sudan as an autonomous entity,
in opposition to Khartoum, displays similar patterns.
Minority groups—Ethiopia’s Amhara aristocracy and Sudan’s riverain Arabized
elite—kept power in both states through control of administration, commerce, and
the military. They co-opted and repressed provincial elites. A succession of revolu-
tionary counter-elites in Ethiopia and Sudan—mostly from the same social strata as
those they overthrew—sought to transform their states under slogans of socialism,
nationalism, and Islamism, but usually succeeded only in adding new layers of
violence to those that already existed. With its long tradition of statecraft, Ethiopia
has been better able to mobilize big armies and maintain effective administrative
structures. But at moments of crisis, peasant revolts spring up across the country and
the fundamental weakness of the state is revealed.5 The Sudanese state has been
chronically weak and unstable, reproducing the historic pattern whereby it was a
tool in the hands of mercantile–military partnerships.6 The result has been long-term
reliance on purchasing loyalty from provincial elites who command militia, creating
cycles of peripheral violence that both generate endless war and prevent insurgents
from coalescing into forces that can overrun the centre. These long-running conflicts
and struggles for state power form the backdrop to the intermittent but predictable
instances of mass killing and displacement that constitute the third section of this
chapter. The ruling groups sustain themselves despite disorder on their peripheries.
Despite generating instability, these states persist. The secondary states of the Horn—
Eritrea, Somalia and its successors, and the putative South Sudan—reproduce the
same violent political pathologies, but without the remarkable ability of those deeper
rooted states to bounce back from the brink of autodestruction.
There are three main types of peaks of violence that recur within this political
matrix. The first and commonest is ethnically targeted rural massacre. This typically
occurs within counter-insurgency or frontier policing operations. Noting that
the region is characterized by ethnic mobilization for civil politics, armed conflict,
or peripheral control, it is unsurprising that insurgents, counter-insurgents, and
frontier warlords alike tend to target ethnic groups. Given the toolkit for low-cost
5 Gebru Tareke, Ethiopia: Power and Protest: Peasant Revolts in the Twentieth Century (Cambridge:
6 Peter Woodward, Sudan 1989–1989: The Unstable State (London: Croom Helm, 1990); Abdullahi
Gallab, The First Islamic Republic: Development and Disintegration of Islamism in the Sudan (London:
Ashgate, 2008).
counter-insurgency that involves intimidating a civilian population, forcible relo-

cation, control of food and livelihoods, and using local proxies which are paid in
loot, it is unsurprising that most casualties in these operations are civilian. Nor is it
surprising that famine is a typical outcome that causes more deaths than violence
itself.
The second is the repression of political opposition through mass detention,
torture, and execution in the urban centre of state power. Ethiopia’s Red Terror
is the supreme exemplar of this though there are smaller scale instances in every
state in the region.
The third paradigm of extreme violence we may call the ‘blow-around’ of violent
conflict. This occurs when the perpetrator of violence is not acting at the direct
behest of a state, though the group may initially have been armed or even created
by a state. The massacres in Somalia after the collapse of the Siad Barre government
are examples of this. One may fairly point the finger of blame at Siad Barre for
having deliberately fostered militarized tribalism as a stratagem for either remain-
ing in power or creating mayhem should he be removed. Much of the violence in
Darfur since the 1980s is similarly attributable to the high price of loyalty extracted
from patrons in Khartoum, Libya, and Chad by military-tribal entrepreneurs on
the frontier, who then pursue their local ambitions through violence.
THE HORN OF AFRICA AND

T H E S T U DY O F G E N O C I D E
................................................................................................................
Before 2004, an essay on genocide in the Horn of Africa would have been, by
definition, an exercise in contentious labelling. Despite the numerous cases of mass
killing in the Horn over the past two centuries, unabated in recent decades, most
scholars of genocide dealt with the region in passing or not at all. Cases of cursory
treatment include Totten,7 Harff and Gurr,8 Fein,9 Kiernan,10 and Jones.11
7 Samuel Totten, Teaching about Genocide: Issues, Approaches and Resources (Greenwich CT:
Information Age, 2004).
8 Barbara Harff and Ted Robert Gurr, ‘Towards Empirical Theory of Genocides and Politicides:
Identification and Measurement of Cases since 1945’, International Studies Quarterly 32:3, 1988, 359–71,
at 364–5; Barbara Harff, ‘No Lessons Learned from the Holocaust? Assessing Risks of Genocide and
Political Mass Murder since 1955’, American Political Science Review 97:1 (2003), 57–73, at 60.
9 Helen Fein, ‘Accounting for Genocide after 1945: Theories and Some Findings’, International
Journal on Group Rights 1 (1993), 79–106, at 87.
10 Ben Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to
Darfur (New Haven, CT: Yale University Press, 2007), 594–7.
11 Adam Jones, Genocide: A Comprehensive Introduction (London: Routledge, 2006).
534 alex de waal
However, the Horn of Africa is also the locus of two important innovations in
the use of the label genocide to apply to instances of mass killing. One case is the
Ethiopian designation of the 1977–8 Red Terror as genocide, in conformity with the
country’s penal code.12 In this case, the guerrilla movement that took power in 1991
prosecuted officials of its predecessor, including the former head of state, Mengistu
Haile Mariam, who was convicted of genocide in absentia on 13 December 2006.
The second case in which the label genocide has been used by a state with the
power to act on its words was on 9 September 2004 when the US Secretary of State
determined that the conflict in Darfur, Sudan, constituted genocide. Following an
investigation sponsored by the State Department and carried out by the Coalition
for International Justice,13 Colin Powell told the US Senate Committee on Foreign
Relations that ‘genocide has been committed in Darfur and that the government of
Sudan and the Jingaweit[sic] bear responsibility—and genocide may still be occur-
ring.’14 In his next sentence, he said that no change in policy followed from this
determination. If we were to apply Powell’s criteria more widely, then we would
identify genocides very widely in the region. Equally, we could attempt to apply more
restrictive definitions and find no cases at all.
Comparable though less remarked difficulties arise with identifying wars. The
beginnings and ends of wars are typically defined by the belligerents to suit their
own political purposes. Insurgents create heroic myths of origin around their first
military actions, not mentioning their rivals who may have started fighting earlier,
or glossing over the fact that their first actions were criminal raids. The parties to a
peace treaty—both belligerents and mediators—concur that the signing of an
agreement marks the end of a war, even though the immediate post-war period
may actually be just as violent, as ‘other armed groups’ (to use recent Sudanese
terminology) are removed from the scene. Thus, Sudan’s first civil war is common-
ly dated 1955–72—from the Torit Mutiny to the Addis Ababa Agreement—though
in fact there was little organized violence during 1957–60. About 1,000 people were
violently killed in Darfur in the year after January 2005 (when major hostilities
ended), while larger numbers died in ‘peaceful’ South Sudan over the same period,
many of them in disarmament exercises. Most wars are defined by an act of
political collusion between the adversaries. An interesting exception is the thirty-
year ‘shifta war’ in the Somali regions of Kenya—shifta means ‘bandit’ and it suited
both colonial and post-colonial authorities to dismiss the Somali insurgents as
12 In an important anomaly, Ethiopian law (Penal Code of the Empire of Ethiopia of 1957,
Proclamation 158 of 1967, Articles 281–6) contains a definition of genocide that includes ‘political
groups’ as a category, in line with Lemkin’s earlier formulation, allowing for charges of genocide to be
brought against those who ordered mass killings of political opponents.
13 Samuel Totten and Eric Markusen, Genocide in Darfur: Investigating the Atrocities in the Sudan
(New York: Routledge, 2006).
14 Colin L. Powell, ‘The Crisis in Darfur’, written remarks before the Senate Foreign Relations
Committee, Washington DC, 9 September 2004.
bandits while also maintaining military rule in the district. This exception illus-
trates how coercive rule in frontier regions can be as violent in ‘peacetime’ as in
‘war’—the distinction may have little relevance to the experience of life and death
of the local inhabitants.
What drives the recurrent massacres and forced removals in the Horn is the
attempt of state leaders to sustain their dominance in a turbulent system. How
does a centre of state power with limited legitimacy and severe resource con-
straints control its far-flung peripheries? And how does a minority elite sustain
itself in power in the metropolis? It is by exemplary violence and by purchasing
the loyalty of local leaders with supplies of guns and a licence to pillage their
neighbours.
E P I S O D E S O F M A S S AC R E A N D F O RC E D
R E M OVA L I N T H E H O R N
................................................................................................................
As the discussion above will have indicated, studying lethal violence against
civilians in the Horn of Africa can present major problems of definition. One
concept that does travel well is ‘massacre’, as utilized by Semelin: ‘a generally
collective form of action, involving the destruction of non-combatants.’15 This
section provides twenty-two illustrative cases of either individual or recurrent
massacre and forced removal, which form the sharp peaks of violence and violation
amid a broken landscape of conflict, frontier governance, and struggles for power.
Ethiopia and Eritrea

Seven episodes of mass killing and one of mass expulsion demand consideration in
Ethiopia and Eritrea.
Massacres in Eritrea and the Ogaden, 1960s

The extension of imperial rule into the newly acquired territories of Eritrea and the
Ogaden (the latter inhabited by ethnic Somalis) was marked by military campaigns
that involved the widespread burning of villages and killing of civilians. These were
akin to colonial pacification campaigns, intended to demonstrate the power of the
15 Jacques Semelin, Purify and Destroy: The Political Uses of Massacre and Genocide (London: Hurst,
2007), 323.
536 alex de waal
state and to punish those who dared resist. The biggest of these campaigns was
mounted in lowland Eritrea in 1966–7.16
The Red Terror, Addis Ababa, 1977–8

The Ethiopian revolution was a bloody affair almost from the outset. While the
deposition of Emperor Haile Selassie in September 1974 was bloodless—the army
controlled the city with just half a dozen tanks and the ageing monarch offered no
resistance when he was taken away into detention and ultimately to his death—it was
rapidly followed by a peasant jacquerie against aristocrats and landlords, the sum-
mary execution of sixty ministers from the previous government, a crackdown in the
Eritrea city of Asmara, and shootouts among the revolutionaries themselves.17
Beginning in February 1977, the military junta that had taken power repressed an
urban insurrection with exceptional brutality. The challenge was mounted by the
Ethiopian People’s Revolutionary Party (EPRP), a leftist militant group with a
strong following among students, that had helped initiate the revolution but felt
that it had been hijacked by the soldiers. The EPRP began a campaign of assassi-
nating government officials. At the same moment, an internal power struggle
within the ruling Provisional Military Administrative Committee (known as the
Dergue) was resolved in favour of Col. Mengistu Haile Mariam. Mengistu set about
the systematic destruction of the urban opposition, which entailed giving power
and impunity to neighbourhood leaders to arrest and execute individuals on the
merest suspicion of opposition sympathies. Estimates of the number killed in
Addis Ababa alone are in excess of 10,000,18 plus many others elsewhere in the
country. An entire generation of educated Ethiopians was either killed or driven
from the country. In a second wave of killing, the Dergue turned on its own civilian
supporters, eradicating the All-Ethiopia Socialist Movement (known by its Amhara
acronym, MEISON).
The two rival leftist parties, EPRP and MEISON, had developed ethnic profiles
during the early 1970s. The EPRP was identified as mostly Amhara and Tigrayan
while MEISON was chiefly Oromo. However, ethnicity was not the dominant
characteristic of either party and neither was it the reason for the repression. The
struggle was political and military. The aim was to cow the opposition into
complete acquiescence. The Dergue called its campaign the ‘Red Terror’, conscious-
ly evoking precedents in Revolutionary France and Bolshevik Russia. The term
‘genocide’ only became current later, especially after 1991 when the incoming
government set up the Special Prosecutor’s Office to bring charges against the
individuals responsible for many instances of killing and torture.
16 Africa Watch, Evil Days: Thirty Years of War and Famine in Ethiopia (London: Africa Watch,
1991), ch. 2.
17 Rene LeFort, Ethiopia: An Heretical Revolution? (London: Zed Books, 1983).
18 Babile Tola, To Kill a Generation: The Red Terror in Ethiopia (Free Ethiopia Press, 1989).
War in the North, 1977–91

The Emperor Haile Selassie’s failure to resolve peacefully Eritreans’ grievances
about the loss of their status as partners in a federation was a major reason for
his downfall in 1974. His army’s military campaigns in rural Eritrea involved
burning villages, confiscating livestock, and killing civilians. His successors did
no better and by 1977 were resolved on a military solution to the Eritrean national-
ist rebellion. Over the subsequent fourteen years, a succession of immense offen-
sives were mounted against the positions of the Eritrean People’s Liberation Front
(EPLF), which quickly developed from a classic guerrilla movement into a conven-
tional army with mechanized brigades, lacking only an air force. Meanwhile,
insurrection also erupted immediately to the south in Tigray, where the Tigray
People’s Liberation Front (TPLF) mounted a mobile guerrilla war. Most of the time
the EPLF and TPLF were in alliance and their combined forces ultimately defeated
the Mengistu government in 1991.
During the 1980s, Ethiopia fielded the largest conventional army in sub-Saharan
Africa, with more than 300,000 regular soldiers in addition to militia. It was a
mechanized force equipped by the Soviet Union, with thousands of tanks and
scores of aircraft. By the final campaign, the EPLF and TPLF (the latter now the
leading member of the Ethiopian People’s Revolutionary Democratic Front) each
fielded armies of more than 80,000. Estimates of battlefield fatalities ran into
hundreds of thousands for the course of the war.
The Ethiopian government counter-insurgency sought to destroy the social and
economic base of the rebels. The aim was not extermination but victory and
control, but the partial destruction or removal of ethnic groups was employed as
an instrument in pursuit of those goals. Killing of civilians was routine during
military operations and patrols and at checkpoints. The air force was used against
civilian targets. In the worst single incident, an estimated 1,800 market-goers were
killed during a day’s aerial bombardment in Hausien in Tigray on 22 June 1988.19
Hausien was not a military target but the government was trying to destroy the
morale and economic base of the insurrection. The counter-insurgency also creat-
ed famine, partly deliberately and partly as the inevitable outcome of a military
strategy that included offensives aimed at the agriculturally productive areas of
northern Ethiopia and Eritrea, restrictions on trade and movement in government-
controlled areas, bombing of markets in rebel-controlled areas, and efforts to stop
relief aid going to rebel areas while assistance in government areas was used to
further military aims, including feeding militia (reserves and auxiliaries to the
army).20 A final element in the counter-insurgency was forcible relocation of
19 Africa Watch, Evil Days: Thirty Years of War and Famine in Ethiopia (London, 1991), 258–64.
20 Alex de Waal, Famine Crimes: Politics and the Disaster Relief Industry in Africa (London: James
Currey, 1997), ch. 6.
538 alex de waal
populations, targeting especially those thought to be sympathetic to the rebellions,

to southern Ethiopia.21
The Ethiopian government of the time was a Marxist-Leninist military dictator-
ship imposed upon a centuries-old tradition of statehood.22 The army and state
institutions, rather than local proxies, were the main instrument of violence.
War in the South, 1977–84

The conduct of the war in Eritrea and Tigray was foreshadowed by the Ethiopian
government’s counter-insurgency in the south, targeting Somalis and Oromos.
This overlapped with the war to expel the invading Somali army in 1977–8. This
war was almost entirely unreported.23 Among the methods used were coordinated
ground and air offensives against rebel strongholds, scorched earth operations
through areas in which civilians were supporting the rebels, closing of markets
and withholding of relief, forced relocation to planned villages (‘villagization’), and
(towards the end of the period) the resettlement of people from the north in areas
that remained rebellious. Journalists estimated 25,000 civilian fatalities in 1978–9
alone.24 A smaller rebellion in the adjoining district of Sidamo in 1981 saw some of
the largest single massacres, possibly because the area is thickly populated and it
is difficult for people to flee. For example, more than 1,000 were reported killed
when the air force used a fuel-air explosion in the valley of Gata Warrancha
in March that year.
The Lower Omo Valley

By the late 1980s, the south-western corner of Ethiopia, bordering Kenya and
Sudan, had become a lawless and militarized area beyond the control of the three
governments and indeed rebel forces too. This area has a history as a violent
frontier plundered by states and freebooters.25 Local cattle herders’ social structures
also came to function as military formations, both as resisters and as collaborators
in this depredation, and also for local raiding. Emblematic of violent transformation
is the fact that one group in the far south of Ethiopia, the Nyam-Atom tribe (whose
name means ‘Elephant Eaters’) renamed themselves Nyang-Atom (‘Carriers of New
Guns’).26 Warfare in this remote region became far more deadly when automatic
21 Jason Clay, Sandra Steingraber, and Peter Niggli, The Spoils of Famine: Ethiopian Famine Policy
and Peasant Agriculture (Cambridge, MA: Survival International, 1985).
22 Christopher Clapham, Continuity and Change in Revolutionary Ethiopia (Cambridge:
23 Africa Watch, Evil Days, ch. 5.
24 Ibid. 86.
25 Donald Donham and Wendy James, The Southern Marches of Imperial Ethiopia: Essays in History
and Social Anthropology (Athens: Ohio University Press, 2002).
´
26 Jan-Åke Alversson, Starvation and Peace or Food and War: Aspects of Armed Conflict in the Lower Omo
Valley, Ethiopia (Uppsala: Scandinavian Institute for African Studies, 1989), 87 (quoting Serge Tornay).
weapons became readily available.27 Feuding and raiding, including reciprocal cases
of homicide, were common between a host of groups, among them the Nyangatom
and their smaller and less well-armed neighbours, the Mursi.28 One cycle of killing
occurred in 1987, culminating in the slaughter of 600–800 Mursi—over ten per cent
of the entire Mursi population. The next year, in reprisal for the killing of 15
policemen, Kenyan forces killed 200 Nyangatom raiders and 500 civilians and burned
five villages inside Ethiopia as well as others in Kenya. Similar reprisals against the
Toposa, who live inside Sudan, were reported.
The Expulsions of 1998

In 1991, the EPLF-PRDF overthrew the Mengistu government. A new government was
formed in Ethiopia and a new independent state in Eritrea. Rarely had partition gone
so smoothly. It did not last. In 1998, a border incident in an otherwise insignificant
village called Badme quickly escalated into a two-year conventional war. This was
fought by well-disciplined regular armies along a well-defined frontline. There were an
estimated 70,000 fatalities among the troops on both sides. Civilian casualties were few.
The principal exception to civilian immunity was the expulsion of 75,000 Ethiopians of
Eritrean origin beginning a month after the war began. There was a smaller reciprocal
expulsion of Ethiopians from Eritrea. Many of those initially arrested were members of
Eritrea’s ruling party, former Eritrean soldiers, or prominent in Eritrean organizations
in Ethiopia (all of them entirely legal at the time). Round-ups and expulsions
continued for well over a year, broadening to include anyone of Eritrean origin.
Gambella 2003–4
In December 2003, an estimated 400 civilians belonging to the Anuak ethnic group
were killed by mobs that included soldiers in the town of Gambella, state capital of
the western Ethiopian region of the same name.29 Further reports of killings and
the destruction of villages continued. The violence was rooted in political tensions
consequent on political decentralization, in which different ethnic groups, which
took on political personalities during the resistance to Mengistu and the subsequent
ethnic federalism of the EPRDF, contested for control of regions. In Gambella,
tensions over voting and land rights among the indigenous Anuak and Nuer and
highland settlers led to violence. The central government was at first reluctant to
intervene and failed to contain the crisis, and then after the 2003 massacre sealed off
the area and dispatched the army to crush the incipient rebellion.
27 The Kenyan government did not distribute automatic weapons and instead tried to rely on its
police and army to control its borderlands. Regularly, livestock raiders win military encounters with
the state forces.
28 David Turton, ‘Warfare, Vulnerability and Survival: A Case from Southwestern Ethiopia’,
Cambridge Anthropology 13:2 (1988–9).
29 Human Rights Watch, ‘Targeting the Anuak: Human Rights Violations and Crimes against
Humanity in Ethiopia’s Gambella Region’ (New York, March 2005).
540 alex de waal
Coda: Eritrea
After 2001, the Eritrean government led by President Isseyas Afewerki has regressed
into a peculiarly nasty form of dictatorship. Challenged by many of his colleagues
in government over his failures to deliver on promises of democratization and the
blunders that had led to a lost war, Isseyas responded by arresting eleven leading
party members. None has been seen since. Subsequently all civil society and
independent media were repressed. Eritrea is a small country with an efficient
party and the President has exercised total personal control of political life even as
the country has spiralled into economic collapse. It has been efficient politicide.
Eritrea has not witnessed genocidal massacre but few believe that Isseyas will hand
over power willingly or liberalize. The dangers of mass murder are considerable.
Somalia
Four episodes of mass killing and forced displacement in the Somali wars warrant
our attention.
War in the North 1988–9

Following its defeat by Ethiopia in 1978, the Somali government of Siad Barre
became discredited. The Somali National Movement (SNM) headed armed oppo-
sition, with core support among the Isaaq clan family of north-west Somalia. In
May 1988 the SNM nearly captured Hargaisa, the main city of the north-west, and
another town, Burao. Siad Barre responded by reportedly declaring that the Isaaq
should be wiped out. His son-in-law and commander of the operation, Gen.
Mohamed Said Hersi Morgan, reportedly answered that the order couldn’t be
fulfilled because there were too many of them to kill. This is the closest case of
attempted extermination of a group in north-east Africa, thwarted by the intrinsic
difficulty of carrying out such a task when faced with fierce armed resistance and
the ability of the population to flee across a nearby border.
The city of Hargaisa was destroyed in the government’s counter-attack. (No
other city in contemporary Africa has suffered comparable destruction.) Tens of
thousands of people were killed. Virtually the entire populations of Hargaisa and
other towns fled the country. The livelihoods of the people of north-west Somalia
were all but destroyed by looting, the collapse of markets, the destruction of
infrastructure, and the dissemination of landmines which meant that camel
herds were unable to move safely to many areas of pasture. Testimonies from the
war are extraordinarily harrowing, comparable in the intensity of fear and violence
to the depths of the Rwanda genocide.30
30 Africa Watch, Somalia: A Government at War with Its Own People: Testimonies about the Killings
and the Conflict in the North (New York, 1990).
Wars in the South 1990–3

President Siad Barre reportedly promised that he would take Somalia to hell with
him if he were driven from power. He made good on that promise. In the last year
of Siad Barre’s rule and the two years immediately following his overthrow, a
number of overlapping wars of extreme brutality unfolded with much of the
violence directed at the uprooting of ethnic groups. Among them was the dispos-
session, massacre, and expulsion of the farming minorities of southern Somalia.
The two river valleys of Jubba and Shebelle and the rainlands in between were
inhabited by two minority Somali clans, the Digil and Rahanweyn, clusters of
Cushitic peoples distantly related to the Somalis, including the Shebelle and the
Gabwing (also known as Gabaweyn), and Somali Bantus. These people occupied
the most valuable farmland in the country and as the economy shifted to irrigated
schemes, merchants and government officials acquired land leases and dispossessed
the minority farmers. The land grab reached its arbitrary and violent peak as the
state imploded.31 The Somali Bantus were the worst hit. Almost all the surviving
members of this community were refugees in Kenya by the mid-1990s and sub-
stantial numbers have been resettled in the United States.32 This is the most
complete case of ethnic cleansing in the Horn.
A second component was the ethnic cleansing of Mogadishu as the rebel militias
took control of the city in 1991, culminating in urban warfare that cost an estimated
10,000 lives between November 1991 and February 1992. A third component was the
war and famine that unfolded across a swathe of territory in 1992 as Siad Barre’s loyalist
militias attempted a comeback and were repulsed. A campaign of terror, looting, and
revenge killing caused hunger and disease which claimed an estimated 250,000 lives.33
The War of 2006–9

Somalia gained a modicum of stability in the years between 1994 and 2006. The
north-west (target of Siad Barre’s war of 1988) seceded to create the unrecognized
but stable state of Somaliland. In the south, there were fourteen failed attempts to
put together a government. These didn’t succeed, but by 2006 Mogadishu itself
achieved recognizable municipal governance through a network of Islamic courts.
The Islamists’ overreached, sparking a fightback from their adversaries and an
Ethiopian invasion. Fighting intensified in October 2007, after which 300,000
people fled the city and thousands were killed. The killings and expulsions were
31 Catherine Besteman and Lee Cassanelli (eds), The Struggle for Land in Southern Somalia: The
War behind the War (Brighton: Haan, 2003).
32 Catherine Besteman, ‘Genocide in Somalia’s Jubba Valley and Somali Bantu Refugees in the U.S.’,
SSRC Webforum, How Genocides End, 9 April 2007, available at http://howgenocidesend.ssrc.org/
Besteman/
33 de Waal, Famine Crimes, ch. 8.
542 alex de waal
targeted politically, with both sides assassinating clan elders and civil society
leaders deemed unsympathetic.
This war has also involved the Somali region of Ethiopia in another conflict, with
the collapse of local government and the conduct of wide-ranging and violent
counter-insurgency and frontier policing.
The Wajir Massacre

North-east Kenya is inhabited by ethnic Somalis. In 1963, as Kenya’s independence
approached, it became clear to the British that the majority wanted to join Somalia.
Rather than accede to this demand, the British placed the Northern Frontier District
under military rule. Irredentist insurrection was defeated but emergency powers were
exercised for thirty years. Somalis were treated as second-class citizens.34 The largest
single massacre was perpetrated against the Degodia clan at the Wagalla airstrip near
the town of Wajir, in February 1984.35 Over a four-day period the Kenyan army
systematically killed many hundred, probably well over one thousand, men.
Sudan
Sudan has been in a state of armed conflict for the majority of the fifty-plus
years since it achieved independence in 1956. Since the outbreak of the second
north-south war in 1983, armed conflict has been continuous in the country, until
2004 in the south and intermittently in several regions of the north.
There have been few attempts to quantify the death toll in southern Sudan,
but one analytical compilation estimated the death toll at 1.9 million up to 1998.36
The data are better for Darfur though the figures remain controversial, with best
estimates at 200,000 excess deaths or more.37 In both cases, hunger and disease
killed many more than violence. Nine episodes of mass killing are examined here.
The First Civil War

Sudan’s first civil war (1955–72) opened with a massacre of northern traders and
government officials in the town of Torit and witnessed a number of massacres of
southerners by government soldiers and policemen over the subsequent years,
notably in the town of Wau in 1965.
34 Africa Watch, Kenya, Taking Liberties (London, 1991).

35 S. Abdi Sheikh, Blood on the Runway: The Wagalla Massacre of 1984 (Nairobi: Northern
Publishing House, 2007), available at http://www.scribd.com/doc/2551019/BLOOD-ON-THE-
RUNWAY-The-Wagalla-Massacre-of-1984
36 Millard Burr and Robert Collins, ‘Quantifying Genocide in Southern Sudan and the Nuba
Mountains, 1983–1998’ (Washington, DC: US Committee on Refugees, 1998).
37 Centre for Research on the Epidemiology of Disasters, ‘Darfur: Counting the Deaths’ (Brussels,
May 2005); John Hagan and Alberto Polloni, ‘Death in Darfur’ Science 313 (2006), 1578–9.
The Massacre of the Mahdists and the Elimination of the Communists

A year after seizing power in 1969, the leftist government of Jaafar Nimeiri faced
rebellion from the armed followers of the Mahdi’s grandson in Omdurman, the
twin city of the capital, Khartoum, and at their headquarters at Abba Island on the
Nile. Resistance was crushed with overwhelming force including sorties flown by
the air force, resulting in more than one thousand fatalities, including civilians.
The following year, Nimeiri’s allies in government, the Sudan Communist Party,
tried to seize power in a coup. After three days, the coup failed and Nimeiri turned
on his erstwhile comrades with a level of savagery rarely seen among the Sudanese
elite. Dozens were hanged and hundreds imprisoned in a wide-ranging crackdown.
Slaughter in Bahr al Ghazal 1985–9

The second civil war of 1983–2005 was larger in scale than the first and marked by the use
of tribal militia from the earliest days. In July 1985, the transitional government of Gen.
Abdel Rahman Suwar al Dahab took the decision to arm Baggara Arab tribes in south
Kordofan and south Darfur to fight against the insurrection of the SPLA. This was a
fateful decision, taken partly because of lack of money and partly because the regime
feared that a stronger army might overthrow the government itself. In the early days, the
militiamen were not paid, simply encouraged to reward themselves with looted cattle. It
set a pattern for plunder and impunity and local partnerships between traders, military
intelligence officers, and militia commanders to profit from the violence.
In 1987, two Sudanese academics documented ethnically targeted mass killings in
the war zone of Bahr el Ghazal and the town of ed Da’ien, the latter outside the war
zone.38 The following year, an anonymous report, ‘Sudan’s Secret Slaughter’, docu-
mented mass killings in Wau, carried out by the army and a proxy militia, targeting
Dinka. In 1989, a report by Amnesty International on militia killings in Bahr el Ghazal
detailed a similar pattern.39 Relief aid to the displaced population was obstructed.40
Exceptionally high death rates were recorded in several displaced camps during 1988.41
In the 1990s, the militia raids abated as the leaders of the Baggara tribes recognized
that participating in the war on these terms was no longer in their interests.
The Juba Massacres 1992

In June and July 1992, the SPLA carried out military attacks on the southern capital
of Juba and nearly captured the city. One of the reasons for the near success was the
38 Ushari Ahmad Mahmud and Suleyman Ali Baldo, ‘Al Diein Massacre: Slavery in the Sudan’
(Khartoum: University of Khartoum, 1987).
39 Amnesty International, ‘Sudan: Human Rights Violations in the Context of Civil War’ (London,
1989).
40 David Keen, The Benefits of Famine: A Political Economy of Famine and Relief in Southwestern
Sudan (Princeton: Princeton University Press, 1994); African Rights, Food and Power in Sudan:
A Critique of Humanitarianism (London, 1997).
41 de Waal, Famine Crimes.
544 alex de waal
simultaneous mutiny of the southern army units inside the city. In response to the
attacks and the mutiny, Sudanese security implemented a thorough and brutal
crackdown. This was done having first ensured that there was no international
scrutiny, by expelling foreign diplomats and aid workers. The details of the arrests,
detentions, and executions have never come to light. However, they are reliably
reported to include thousands of extrajudicial executions.
The Nuba Mountains Jihad 1988–93

The Sudan government’s assault on the Nuba Mountains culminated in 1992–3
with the most ambitious campaign of the entire war. The SPLA began operations in
the area in 1987, sparking a vicious counter-insurgency that included the execution
of hundreds of community leaders and educated people, and the use of militia to
burn villages and punish groups suspected of supporting the rebels. The campaign
of 1992 was officially declared a jihad.42 The aim of the campaign extended beyond
crushing the SPLA rebellion to include forcibly relocating the entire Nuba popula-
tion out of their homeland into ‘peace camps’ where they would take on a new
identity.43 The systematic use of rape was documented in the Nuba, as a means of
destroying communities and creating a generation with a new identity. Unlike most
other cases in which the use of rape as weapon is inferred rather than directly
documented, Sudanese security officers went on the record saying that it was
policy.44 This campaign, mounted by a revolutionary government at the height
of its ideological hubris, represented a more far-reaching attempt at violent social
re-engineering than anything attempted before or since. At its head were civilian
Islamist ideologues who did not conceal their contempt for the Nuba way of life
with its traditions of music, wrestling, and body-painting, and the Nuba’s tolerant
mix of Islam, Christianity, and traditional religions. Had the campaign succeeded,
the Nuba would have ceased to exist as an identifiable society. As it was, the plan
was defeated by a combination of Nuba resistance and divisions within the Sudan
government, with the army preferring to restrict the aims of the campaign to
military victory rather than Islamist transformation as well.45
SPLA Factional Infighting 1991–4

In August 1991, three senior SPLA commanders launched a coup against the
leadership of John Garang. They succeeded only in splitting the movement and
setting in motion a bloody internecine war. The ‘Nasir Faction’ consisted of the
would-be putschists, while the ‘Torit Faction’ was the mainstream SPLA led by
42 Alex de Waal and A. H. Abdelsalam, ‘Islamism, State Power and Jihad in Sudan’, in Alex de Waal
(ed.), Islamism and Its Enemies in the Horn of Africa (London: Hurst, 2004).
43 African Rights, Facing Genocide: The Nuba of Sudan (London, 1995).
44 Sudan’s Secret War, BBC Television, July 1995.
45 Alex de Waal, ‘Averting Genocide in the Nuba Mountains of Sudan’, SSRC Webforum, How
Genocides End, December 2006, available at: http://howgenocidesend.ssrc.org/de_Waal 2/
Garang, which ultimately prevailed. Khartoum security became actively involved

in deepening the split and furthering the fragmentation of the SPLA.
The split became an ethnic war. Its nadir was in September–November 1991
when Nuer forces of the Nasir Faction attacked the Dinka heartland of Bor and
massacred many hundreds, probably thousands, of civilians in a series of raids.
Many women were raped or abducted. The countryside was stripped of cattle and
other property, and more people died of hunger and disease as a direct result.46
Witnesses to the immediate aftermath describe a scene of complete devastation
with corpses scattered throughout villages and along roads. The following two
years saw reprisal raids and counter-attacks as the SPLA split degenerated into a
Nuer–Dinka tribal war which created what became known as the ‘hunger triangle’
in the heart of southern Sudan.
Clearing the Oilfields, 1997–2000

Subsequent counter-insurgency campaigns in the Southern Sudanese oilfields47 and in
Bahr el Ghazal48 were also mounted with a combination of extreme violence and
scorched earth tactics. In these cases, the Sudan army operated alongside southern
militia leaders and rebel defectors who opposed the SPLA. The objective of the oilfields
campaigns was straightforward: the government wanted to exploit oilfields that lay in
the middle of the war zone. It was a bold plan that succeeded through the support of
China and a handful of foreign companies, the disarray of the SPLA, and because the
government was prepared to remove virtually the entire civilian population from the
key areas. As the transport, security, and oil infrastructure went in, the population was
cleared out. It was the largest scale of successfully forced relocation of the entire war.
Islamist slogans often accompanied government mobilization for the oilfields
campaigns. However, the country’s leading Islamists disparaged these as ‘jihad for
oil’. By the end of the 1990s, Sudan’s Islamist experiment had clearly failed and the
Islamist movement itself was split, with calls for jihad sounding increasingly hollow.
Darfur 2003–4
Significant conflicts occurred in Darfur in 1987–9, 1991, 1995–7, 1999, and 2001–3
prior to the recognized outbreak of insurrection in February 2003.49 These conflicts
included different configurations of the following factors: local ethnic disputes,
spillover of the Chadian civil war, SPLA incursion, and government overreaction to
local insurgency, especially by mobilizing proxy militia forces. The 2003 rebellion
was more significant than its predecessors because of the large-scale involvement
of Chadian army officers, the extensive military supplies provided by the SPLA
46 African Rights, Food and Power in Sudan, 289–90.

47 Human Rights Watch, Sudan: Oil and Human Rights (New York, 2003).
48 Human Rights Watch, Famine in Sudan: The Human Rights Causes (New York, 1999).
49 Julie Flint and Alex de Waal, Darfur: A New History of a Long War (London: Zed, 2008).
546 alex de waal
(which was at the time negotiating a peace agreement with Khartoum) to the
Sudan Liberation Army (SLA), the emergence of the Justice and Equality Move-
ment (JEM) drawn from former Darfurian Islamists suspected of having ties to
would-be putschists in Khartoum, and the extent to which local state officials had
been backing and arming Arab militia. After the rebels overran Darfur’s main
airbase in April 2003, the government decided on a military response at scale. True
to form, it used proxy militia in the frontline of its offensives, in this case drawn
principally from camel-herding Arab tribes, and popularly known as janjawiid.
From June 2003 to March 2004, the Sudan army, militia, and air force mounted
a series of vast scorched earth offensives in northern and western Darfur. The main
method of the forces was targeting civilian communities, especially Fur, Masalit, and
Zaghawa, the groups that formed the mainstay of the SLA and JEM. Estimates for the
numbers of civilians killed are in the scores of thousands. The scorched earth tactics
combined with obstruction of relief created a humanitarian disaster that caused a
further 150,000 deaths. More than two million people were forcibly displaced.
Between January and March 2004 the government inflicted crushing military
defeats on the SLA and JEM, and in April they signed a ceasefire. This was violated
by both sides, with the rebels responding to their calamitous setbacks by taking the
offensive in eastern and southern Darfur, where the government deployed the same
tactics, albeit on a smaller scale, over the succeeding nine months. Data on attacks
and fatalities indicate that the number of civilian casualties dropped away sharply
after the ceasefire and dropped further with the end of major hostilities in January
2005. The situation thereafter was one of a low-intensity conflict, with the rebel
groups fragmenting and occasionally fighting one another, and the government’s
erstwhile proxies losing their enthusiasm for the war and becoming freelance,
turning on one another, or turning their guns on the government. The conflict
during 2005–9 consisted of a mixture of banditry, interethnic conflict, harassment
of civilians including displaced persons, and occasional offensives and counter-
offensives by government and rebels. The pattern was a familiar one from the
protracted wars in the south and elsewhere, of a background level of killing (in this
case, perhaps 100 people per month) with occasional sharp spikes with several
hundred being killed. By 2006, Darfur’s war was intermingled with the conflict in
Chad, with many of the belligerent forces fighting on both sides of the border.
Islamist ideology played no role in the Darfur war, with former Islamists on both
sides. However, a number of the Arab leaders involved were associated with an Arab
supremacist organization known as the Arab Gathering. Originally created in Libya,
with adherents in Chad as well as Sudan, this took many forms over the years, and
consists more of a flexible and ad hoc network than a disciplined organization.
Nonetheless, Arab extremist and supremacist slogans were commonly heard during
the campaigns of 2003–4. Some of those militants subsequently reconsidered their
positions, reaching out to the rebels or even switching sides, and deploying similar
rhetoric against the government.50 By 2007, Arab supremacist calls were hollow, and
the government’s overriding interest was simply holding onto power.
C O N C LU S I O N
................................................................................................................
Across the Horn of Africa over recent decades, massacres, forced displacement,
and famine are the common occurrences during organized warfare and in the
frontierlands of states. Lethal violence against rival elites is also the ultimate
resort of those in power when under threat. Cases lie on a spectrum between
modestly effective and centralized state authority and the complete absence of
state authority. Ethiopia’s military campaigns in the 1980s are an instance of a
centrally planned counter-insurgency that failed only because the EPLF and TPLF
were comparably efficient as military–administrative entities. The fragmentation
that marked the demise of Siad Barre’s government in 1991 stands at the other
extreme. The Sudanese cases and the destruction of north-west Somalia lie in-
between, where governments have sufficient power to unleash the forces of
destruction, but not enough to control those forces once unleashed. Gambella
is an interesting anomaly, where a relatively capable central government initially
remained distant from a growing crisis, but then intervened to stop it, albeit
brutally. Possibly, what we can draw from this is that the principal difference
between a stronger state (Ethiopia) and a weaker one (Somalia, Sudan) is the
differential capacity to stop violence when its immediate goal has been achieved.
Urban intra-elite killing was prominent in the 1970s, notably in Ethiopia, but has
not disappeared.
Ideologies have been pressed into service for state-building and war-making in
the Horn, including communism, nationalism, and Islamism. Militant rhetoric
accompanies every military mobilization but should rarely be taken at face value.
At particular instances in Ethiopia and Sudan, extremist ideology has gone
beyond cynical deployment as a tool of popular agitation, to determine the actual
objectives and methods of a military–political campaign. This was the case at
the height of the Ethiopian revolution, during and after the Red Terror, and at
the zenith of the Islamist hubris in Sudan in the early 1990s. The failures of the
Ethiopian programmes for coercive socialist transformation and the Sudanese
jihad against the Nuba have much to do with the internal contradictions of
ambitious extremist plans in complex societies, where governments possess
limited financial and political resources.
50 ‘Meet the Janjaweed’, Channel 4 News, UK, 13 March 2008.

548 alex de waal
If we look further back in history, we see important continuities with the

comparable episodes of violence in the nineteenth and early twentieth centuries.
These historic parallels are often alive in the minds of contemporary actors, for
example the southern Sudanese leaders who dwell on their people’s experience of
slavery. Over a century’s time span, political ideologies including Islamism and
Arabism appear ancillary to these ongoing struggles for power at the centre and
control at the periphery. The continuities emerge stronger than the transforma-
tions, and the regimes in each country tend to revert to type.
Sudan demands a special case in the study of massacre and displacement. Its
wars and massacres have many similarities with those of its neighbours but it also
possesses its own distinctive way of war. Sudan is a country of paradoxes. Its
government is too weak to build truly effective mechanisms for repression and
military control, but despite international ostracism, the ruling elite is strong
enough to remain secure.51 The persistence of conflicts in the peripheries reflects
the facts that the central government can survive regardless, but also that it uses
local proxies that it cannot control. Khartoum can start wars but very rarely can it
stop them. Racism persists in Sudan and is thoughtfully analysed by Sudanese from
all quarters,52 but exhortations to racial violence are usually opportunistic. This is
true of both central ruling elites which control states and the provincial elites which
use violence to assert their claims on the centre. For example, the same Darfurian
Arab militia leaders who called for the ‘eradication’ of African tribes have at other
moments made political deals with those tribes and called for war against Khar-
toum. As noted by the historians of the triangular Sudan–Libya–Chad wars, with
reference to the political about-turn of the most ideologically Arab supremacist of
Chad’s factional leaders, ‘Ideology, principle and even honor were no substitute for
self-preservation by the chieftains of the Sahara.’53
One important conclusion to be drawn from this survey of massacre and related
events in the Horn of Africa is that selecting the Red Terror and the Darfur war for
the label ‘genocide’ is arbitrary. They are two exemplars of patterns of violence that
have recurred in the region for decades. If scholars of genocide, lawyers, and
policymakers are resolved that one or either of these cases is indeed genocide,
then numerous other instances warrant the same consideration. Such a reclassifi-
cation of the Horn of Africa as the location of recurrent genocide would have
important implications for the field of genocide studies and international law
and policy.
51 Alex de Waal, ‘Sudan: The Turbulent State’, in idem (ed.), War in Darfur and the Search for Peace
(Cambridge MA: Harvard University Press, 2007).
52 Francis M. Deng, War of Visions: Conflict of Identities in Sudan (Washington, DC: Brookings
Institute, 1995); A. H. Abdel Salam, ‘Race Relations, Ethnicity and Human Rights’, in A. H. Abdel
Salam and Alex de Waal (eds), The Phoenix State: Civil Society and the Future of Sudan (Trenton, NJ:
Red Sea Press, 2001); Alex de Waal, ‘Who are the Darfurians?’, African Affairs 104 (2005), 181–205.
53 Millard Burr and Robert Collins, Darfur: The Long Road to Disaster (Markus Weiner, 2006), 236.
African Rights, Facing Genocide: The Nuba of Sudan (London, 1995).
Africa Watch, Somalia: A Government at War with Its Own People: Testimonies about the
Killings and the Conflict in the North (New York, 1990).
Besteman, Catherine, and Lee Cassanelli (eds), The Struggle for Land in Southern Somalia:
The War Behind the War (Brighton: Haan, 2003).
Clay, Jason, Sandra Steingraber, and Peter Niggli, The Spoils of Famine: Ethiopian Famine
Policy and Peasant Agriculture (Cambridge, MA: Survival International, 1985).
Deng, Francis M., War of Visions: Conflict of Identities in Sudan (Washington DC: Brookings
Institute, 1995).
Flint, Julie, and Alex de Waal, Darfur: A New History of a Long War (London: Zed Books,
2008).
Keen, David, The Benefits of Famine: A Political Economy of Famine and Relief in Southwestern
Sudan (Princeton: Princeton University Press, 1994).
LeFort, Rene, Ethiopia: An Heretical Revolution? (London: Zed Books, 1983).
Tareke, Gebru, The Ethiopian Revolution: War in the Horn of Africa (Princeton: Yale
Turton, David, ‘Warfare, Vulnerability and Survival: A Case from Southwestern Ethiopia’,
Cambridge Anthropology 13:2 (1988–9).
chapter 27
.............................................................................................
WA R A ND
GENOCIDE IN
A FRICA’S GREAT
LAKES SINCE
INDEPENDENCE
.............................................................................................
omar mcdoom
It is a sad fact that every generation born since independence in Uganda, the
Democratic Republic of Congo (DRC), Rwanda, and Burundi has lived through
either a war or genocide.1 No single figure exists for the overall death toll in this
troubled subset of countries in the Great Lakes. A low-end estimate would stand at
1.4 million while 2.6 million would be at the high end.2 Moreover, these numbers
reflect violent deaths only. They would increase considerably if the indirect effects
of war and genocide, notably disease and hunger, were also counted. These
1 A generation is assumed to be twenty-five years.

2 This range of estimates was calculated by aggregating scholarly estimates for all individual
episodes of violence across all four countries. The low and high estimates respectively are Uganda
(270,000, 821,500), DRC (204,000, 354,000), Rwanda (509,500, 871,000), and Burundi (450,000,
620,000). See the Appendix for a list of these individual episodes.
Low intensity Mid intensity High intensity
100-999 deaths on average p.a. 1000-9999 deaths on average p.a. 10,000+ deaths on average p.a.
Uganda UG1 UG2 UG3
D.R.C. C1 C2 C3 C4 C5 C6 C7 C8
Burundi B1 B2 B3 B4 B5
Rwanda R1 R2 R3 R4 R5 R6
62
64
72
74
82
84
94
04
60
66
70
76
80
96
98
8
92
02
58
0
19
19
19
19
19
19
19
19
19
19
19
20
20
19
19
19
19
19
19
19
19
19
20
20
20
19
Fig. 1. Timeline comparing the duration and estimated intensity of violence against civilians in the Great Lakes since Indepedence
(author’s compilation)
552 omar mcdoom
numbers have also reinforced clichéd perceptions of the region as Africa’s heart of
darkness. But what accounts for the Great Lakes’ terrible civilian death toll? The
web of violence that has entangled the region is complex and individual episodes
are often interconnected. This chapter has been written with the first-time observer
of the region in mind. It has two main goals: first, it aims to summarize the post-
colonial record. It describes all the principal episodes of violence against civilians in
each country since independence. Figure 1 illustrates the chronology of this vio-
lence in a comparative timeline. Second, the chapter provides a critical overview of
the violence. It identifies the characteristics that define the regional context in
which these numerous episodes of violence occurred. Readers will find it helpful to
refer to the individual cases of violence described in the Appendix before they read
on. The analysis that follows frequently draws on them to illustrate its main points.
H I S TO R I C A L S Y N O P S I S
................................................................................................................
As Figure 1 illustrates, the Great Lakes do not constitute a single zone of continuous
violence.3 While the period 1990 to 2007 marks a major intensification in violence
across all four countries, until that time each country followed a distinct trajectory
in which the violence varied in its origins, frequency, and intensity.
Uganda has experienced one of the longest periods of prolonged violence against
civilians. It begins in 1971 with Gen. Idi Amin Dada whose brutal police state is
responsible for killing an estimated 50–300,000 Ugandans until he was himself
forcibly deposed in 1979 (event UG1). His eventual successor, Milton Obote (1980–
6), seized power in a coup and then engaged in a brutal counter-insurgency to
retain it, claiming another 200–500,000 lives (event UG2). Yet the 1986 victory of
Yoweri Museveni’s rebel group over Obote still did not bring peace to the country.
Insurgencies flared again, concentrated in the north of the country. The rebel
Lord’s Resistance Army emerged in 1987 to destabilize northern Uganda for
much of the next two decades until peace talks began in 2006 (event UG3).
The Congolese post-colonial experience with violence began with independence
in 1960. Secession attempts by two renegade provinces were bloodily suppressed by
government forces (event C1). At the same time, the country’s first Prime Minister,
Patrice Lumumba, was deposed after only 82 days in office and assassinated shortly
thereafter. Discontent with the new regime inspired a pro-Lumumbist rebellion
in 1964 (event C2). While the rebellion was quashed in 1965, the country’s
3 Figure 1 counts only deaths attributable to acts of violence. The figures would be considerably
higher if the deaths caused indirectly as a result of disease and hunger for example were also counted.
war and genocide in africa’s great lakes 553
commander-in-chief, Joseph-Désiré Mobutu, exploited the political opportunity

to seize power. For the next thirty-two years, he used repression and patronage to
rule the country until his grip on power began to weaken from a combination of
gross economic mismanagement and pressure for political liberalization in the
early 1990s (events C4 and C5). In 1996, as the Mobutist regime faltered, Rwanda
and Uganda initiated the first Congo War (event C6). They led a military alliance
that definitively toppled Mobutu. However, the man installed as the country’s new
President, Laurent Kabila, turned against his former sponsors. This was in part the
motivation for the Second Congo War in 1998 (event C7). The war was complex,
involving numerous domestic rebel groups and external actors with varying agen-
das. Although many of the key actors finally reached a peace settlement in 2002,
several spoiler groups chose not to participate. Their continued activities in the east
of the country leave it a lawless and insecure region for ordinary Congolese at the
end of 2007 (event C8).
Rwanda and Burundi, each tiny, land-locked countries composed of ethnic Hutu
majorities and Tutsi minorities, followed different paths to violence after their
independence in 1962. Burundi quickly became a Tutsi ethnocracy. Hutu uprisings
in 1965, 1972, 1988, and 1993 drew brutal responses from its Tutsi-controlled
government and army (events B1–B4). The Hutu civilian population was the
primary victim. The 1972 insurrection infamously resulted in a selective genocide
of Burundi’s Hutu elite. The Tutsi army, militia, and elements of the civilian
population killed an estimated 100–200,000 Hutu. The 1993 uprising was triggered
by the election and subsequent assassination of the country’s first Hutu President,
Melchior Ndadaye. The Tutsi army’s harsh reaction to the uprising marked Bur-
undi’s descent into a civil war that, at the end of 2007, had still not been definitively
resolved (event B5).
Rwanda’s post-colonial experience is almost the converse of its neighbouring
cousin. Even before independence, in 1959 a revolution began (event R1) that
ultimately overthrew the country’s Tutsi monarchy and instead ushered in two
successive Hutu-dominated republics (1962–73 and 1973–94). The revolution and
ensuing marginalization of the Tutsi population caused tens of thousands to go
into exile. On several occasions, this Tutsi refugee community attempted to return
to Rwanda by force of arms (events R2 and R4). Its most determined effort in 1990
provoked a four-year civil war (event R4) that culminated in the Rwandan Pre-
sident’s assassination and the 1994 genocide. A newly-installed extremist govern-
ment unleashed the army and militia groups and, with the participation of
elements of the Hutu civilian population, they killed an estimated 500–850,000
individuals, overwhelmingly Tutsi (event R5). It is only through the victory of the
largely Tutsi rebel group that the slaughter ended. The Hutu génocidaires were
pushed out of Rwanda, mostly into the DRC, where elements continued a low-
grade insurgency against Rwanda’s new government (event R6).
554 omar mcdoom
REGIME TRANSITIONS
................................................................................................................
What, if any, then is the underlying feature behind many of the individual episodes
of violence summarized above? The region’s numerous wars and occasional geno-
cides have most often occurred in the context of extra-constitutional regime
changes. In fact, there have been very few legal and peaceful transfers of power
from one individual or group to another in the post-colonial history of the Great
Lakes.4 Instead, regime transitions have tended to occur in one of three ways:
1) military coup;
2) victory in civil wars; and
3) assassination of the head of state.
Often some combination of these events has occurred to install a new regime. Most
have been accompanied by violence.
Military coups have been the most common method. On occasion, these have been
‘palace revolutions’: a bloodless change in leadership, often originating from within
the ranks of the ruling elite. Burundi exemplifies this pattern with coups in 1966, 1976,
and 1987, all led by ethnic Tutsi men, who were all from the same ethnic subgroup
(Hima), and all from the same region of the country (Bururi). However, more often
than not, military takeovers of moribund civilian administrations have been accom-
panied by violence. These have usually occurred for one of two reasons. First, military
men have either fomented or taken advantage of violence to seize power. In Rwanda
in 1973, a young Gen. Juvénal Habyarimana in collaboration with other military
officers astutely exploited the instability created by popular anti-Tutsi violence to
assume power (event R3). Habyarimana became Rwanda’s second head of state since
independence and presided over its Second Republic until his assassination triggered
Rwanda’s genocide in 1994. Second, in order to consolidate newly established posi-
tions, military autocrats have then engaged in purges to eliminate their opponents. In
Uganda, Gen. Idi Amin Dada ruthlessly eliminated real and imagined enemies after
ousting Milton Obote from power in 1971 (event UG1). By one estimate his paranoia
resulted in the deaths of 300,000 individuals over the course of his eight-year rule.5
Armed rebellion has been the second method for toppling governments in the
Great Lakes. These have, unsurprisingly, proved to be the most violent form of regime
transition. Massive violence and genocide has been most likely to occur when a
rebellion has ended through outright military victory instead of negotiated settlement.
4 When peaceful transitions have occurred, they have almost always been the result of flawed
elections in which incumbents who originally came to power through force have been successful in
legitimizing their position ex post.
5 See A. B. K. Kasozi, N. Musisi, et al., The Social Origins of Violence in Uganda, 1964–1985
(Montreal/Buffalo: McGill-Queen’s University Press, 1994), 104.
As the state has usually had a superior military capability, the asymmetry has com-
pelled challengers to engage in guerrilla warfare. In response the government has often
resorted to counter-insurgency strategies designed to ‘drain the seas’.
A first common tactic in such strategies has been to target civilians systematically
to punish and deter them from supporting the insurgents. Civilians have found
themselves caught between the insurgents and the government troops. In Uganda’s
Bush War the main theatre for war was the infamous Luwero triangle in which the
rebel National Resistance Army established itself (event UG2). Ethnic Bagandan
civilians who inhabited the region became the victims of government orders to
destroy villages as well as atrocities committed by rank-and-file soldiers to which
their superiors turned a blind eye. Similarly, Burundi’s Tutsi-controlled army
consistently punished civilians for Hutu uprisings. Rebellions in 1965, 1972, 1988,
and 1993 each resulted in massacres of Hutu civilians (events B1–B4). The Rwandan
government has also punished civilians for rebel actions. It sponsored massacres of
Tutsi civilians during the country’s civil war in reprisal for rebel attacks (event R4).
In the four years leading up to the genocide in 1994, between two and three
thousand Tutsi civilians were killed.
A second popular tactic used in the region’s counter-insurgency campaigns has
been the quasi-internment of civilians. In these cases, the rural population has been
forced to leave their homes and instead to live in so-called displacement camps. The
tactic is designed to deprive rebels of the opportunity to recruit fighters as well as to
deny them food and intelligence. Individuals found outside of the camps could be
assumed to be rebels. Thus in northern Uganda, locked in a protracted civil war
since 1986, the government interned civilians on a massive scale in its counter-
insurgency campaign against the rebel Lord’s Resistance Army (event UG3). At the
peak, over 1.6 million northern Ugandans lived in these camps, ostensibly for their
own protection, emptying the countryside of people. However, a survey in 2005
found that appalling camp conditions had created an alarming mortality rate.
Disease, hunger, and violence had resulted in an estimated one thousand excess
deaths per week, mostly among children under the age of five.6 The Burundian army
has also employed this tactic, more forcefully, in the multifronted civil war that has
beset the country since 1993 (event B5). Government soldiers looted, raped, and
burned homes in order to drive civilians into ‘re-groupment camps’ where human-
itarian conditions were dire. An unknown number of Burundians died in this way.
Assassinations of heads of state have been the third way in which regime change
has often occurred. Aspiring to the highest public office has been a dangerous
ambition in the Great Lakes. Burundi has experienced the highest number of
assassinations: Louis Rwagasore in 1961; Pierre Ngendandumwe and Léopold
Biha (attempted assassination) in 1965; Charles Ndizeye in 1972; and Melchior
6 See Ministry of Health, Government of Uganda, Health and Mortality Survey among Internally
Displaced Persons in Gulu, Kitgum, and Pader Districts in Northern Uganda. Kampala, Uganda (2005).
556 omar mcdoom
Ndadaye in 1993. The last three each preceded massive violence against civilians
(events B2, B3, and B4). However, the most infamous case has been Rwanda’s
President in 1994. Unknown assassins shot down the plane carrying Juvénal
Habyarimana as it was coming in to land at Kigali airport on 6 April 1994. The
sudden gap created by Habyarimana’s death created an opportunity for Hutu
extremists to seize power. They went on to implement the world’s swiftest genocide
that claimed, according to one estimate, up to 850,000 lives in little over one
hundred days (event R5).7 Similarly in Zaire, a young Joseph Mobutu, the country’s
military commander, led a coup in 1960 that culminated in the execution of Patrice
Lumumba, the country’s first post-independence leader. His assassination inspired
his supporters to launch a violent rebellion in 1964 to challenge those complicit in
his assassination (event C1). Finally, it is worth noting that these assassinations and
coups that have become part of the historical record represent probably only a
fraction of the true number. Unsuccessful attempts to overthrow governments have
usually gone unreported, especially those that originated from within the ruling
elite itself. Incumbents have preferred to avoid the exposure of such serious
internal divisions.
However, it is not only during extra-constitutional regime transitions that violence
has occurred. Conflict has also accompanied sanctioned transitions, in particular the
shift from authoritarian to democratic rule. Incumbent elites, reluctant to surrender
power, have resorted to violence in the run-up to elections to intimidate the opposi-
tion and to assure their own political survival. The democratization wave across
Africa in the 1990s was no exception. In Zaire Mobutu deployed the state’s security
apparatus to suppress student protests in the early 1990s (event C4). In Rwanda, the
sudden switch to a multiparty system in 1991 under international pressure created a
new political space. It quickly led to the creation of opposition parties with youth
wings which violently clashed among themselves and with the ruling party and that
also allowed for the organization of an extremist minority (event RS).
The post-election period has also been susceptible to conflict. Electoral losers
have resorted to violence to contest democratic outcomes. In Burundi, the Tutsi
military elite would not accept the result of the 1993 election in which a Hutu
President, Melchior Ndadaye, was victorious. His assassination a mere four
months into office precipitated the onset of a bloody and chronic civil war
(event B4). Yet violence is not an inevitable by-product of democratization. For
example, contrary to many fears, the DRC held relatively peaceful elections in 2007.
The 2002 peace settlement that had brought many former warlords into a transi-
tional government held. Jean-Pierre Bemba, the former rebel leader who lost in the
second round of the 2007 Presidential elections, did not resort to arms in large part
because of strong external pressure to keep the peace.
7 See G. Prunier, The Rwanda Crisis: History of a Genocide (Kampala, Uganda: Fountain, 1999), 265.
Thus regime transitions, both sanctioned and unsanctioned, in the Great Lakes
have often been accompanied by violence. In the remainder of the chapter, I suggest
seven important characteristics that define the regional context in which these
transitions occurred.
C H A R AC T E R I S T I C O N E :
A U T H O R I TA R I A N R E G I M E S
................................................................................................................
Poor governance is the first contextual characteristic behind violent regime transi-
tions in the region. Authoritarian states run by autocrats have been the norm in the
Great Lakes and democratic states led by elected leaders the exception.
The region’s authoritarian regimes have typically relied on two instruments to
govern: repression and patronage. When autocrats have possessed the means to reward
their security forces richly and to bribe their potential opponents adequately, authori-
tarian regimes have in fact been supportive of peace and order. Indeed, as Figure 1
indicates, the period of least violence was the 1970s when single-party, authoritarian
rule was solidly in place across all four countries. However, when autocrats have lost
these means, their ability to govern has been compromised, which has encouraged
challengers willing to use violence to topple them. It has been the transition between
regimes, a period of disorder, in which wars and genocides have most often occurred.
Violence arises when control of the centre is contested and power is in flux.
In Uganda Idi Amin exemplified the first method—coercive rule—in his bid to
assure a position for which he enjoyed little natural legitimacy (event UG1). The
numerous institutions of the state security apparatus that flourished under Amin held
the power of life and death over ordinary Ugandans. Like their patron, they remained
largely unaccountable for their actions. The second method—neo-patrimonial rule—
has also generated grievances. Privileging a few often meant excluding the many.
Autocrats established clientele networks based on ethnic, clan, and/or regional alle-
giances to assure their influence in the absence of popular legitimacy. Within these
networks, a small minority monopolized power and allocated the state’s wealth
among their own. This form of patronage, while effective in buying loyalty, was
inherently exclusionary and accentuated social cleavages. Thus, for example, small,
elite groups of Hutu and Tutsi each dominated post-independence power at the
expense of the other ethnic group in Rwanda and Burundi respectively.
The inner circles of power within neo-patrimonial regimes in fact have often
gone beyond simple ethnic or tribal markers. In Burundi, two Tutsi subgroups
initially vied for power after independence: the high-status Banyaruguru and the
low-status Hima. Ultimately, Tutsi-Hima came to rule the country from 1965 to
558 omar mcdoom
2003. In Rwanda, the intra-ethnic contest was regional: a north–south split. A

southern Hutu elite under Gregoire Kayibanda controlled the state during Rwan-
da’s First Republic (1962–73) while a group of northern Hutu under Juvénal
Habyarimana dominated during the Second Republic (1973–1994).
In short, the authoritarian nature of governance in the Great Lakes is an
important characteristic of violent regime transition. The reluctance of ruling elites
to pass on the baton of power has compelled opponents to use force to wrest it
from them when the opportunity has presented itself. The vast clientele networks
which/underpinned these regimes have strongly resisted efforts to take away the
personal, material benefits of power that they enjoyed.
C H A R AC T E R I S T I C T WO : N AT U R A L R E S O U RC E
E N D OW M E N T D I F F E R E N C E S
................................................................................................................
The asymmetry in resource endowments between resource-rich DRC and compara-
tively resource-poor Rwanda, Burundi, and Uganda is the second important charac-
teristic of the context for violence in the region. The immense natural wealth of the
DRC has long attracted the attention of outsiders willing to use force to obtain them
and has been likened to a curse.8 The actors—and the resources they extracted—have
varied over the country’s history. Belgium’s King Léopold II, when the Congo was his
private territory, began by squeezing rubber from its population. The Belgian govern-
ment, after seizing the territory, diversified into mining and palm oil. The post-
colonial era saw further expansion of this extraction: coltan, gold, diamonds, copper,
cobalt, oil, gas, and timber have all been extracted and sold to benefit a small minority.
With the weakening of Mobutu’s grip on the country in the 1990s, the DRC’s resource-
poor neighbours, Rwanda, Burundi, and Uganda, seized the opportunity to share in
the wealth. Local warlords and foreign corporations have also been complicit in the
exploitation. In all these cases, there has been very little distribution of the country’s
natural wealth to its people. Ordinary Congolese have, to the contrary, been the
victims of violent conflict motivated and sustained by the DRC’s massive resource
endowment as exemplified by the second Congo War (event C7). Rwanda and
Uganda invaded the DRC in 1998 in part to preserve the economic interests they
had each secured during the first Congo War two years earlier (event C6). Even with
the formal conclusion of the war in 2002 Rwandan forces continued covert incursions
into the east to protect Rwandan material interests there. At the end of 2007, warlords
8 See for example Human Rights Watch, Democratic Republic of Congo: The Curse of Gold (New
York, 2005).
and rebel militia continued to operate in the east financed in part by the sale of these
plundered resources (event C8).
C H A R AC T E R I S T I C T H R E E : D I F F E R E N C E S
I N P H YS I C A L G E O G R A P H Y
................................................................................................................
Differences in both borders and terrain have also helped violence to persist in the
region. Looted resources, arms, rebels, and refugees have all moved easily across the
region’s porous and multiple borders. With nine countries surrounding it, the DRC
has suffered the most from illicit cross-border movements, especially from
Rwanda, Burundi, and Uganda. These flows have contributed to the tangled web
of violence in the region and conflicts have rarely remained purely internal affairs.
Policing the DRC’s borders is a massive task. It has exceeded the capacities of its
capital, Kinshasa, and also taxed those of the international community. MONUC, a
17,000-strong UN peacekeeping mission deployed following the second Congo war
(event C7), has been unable to enforce an arms embargo effectively.9 The flow of
small arms has been an important factor in keeping the flames of conflict in the
region alive.
The contrast in physical terrain between countries in the Great Lakes has also
aided insurgencies. The tiny, densely-populated, and highly-cultivated territories
of Rwanda and Burundi have made it difficult for rebel groups to operate un-
detected. In contrast, the expansive territory of the DRC has provided superb con-
ditions for insurgents. Its dense rainforests, volcanic mountain range, and limited
road network have made it an ideal environment in which to wage a guerrilla war.
With the capital, Kinshasa, far away both physically and mentally, numerous rebel
groups have indeed emerged in the distant east of the country. Rwanda, Burundi,
and Uganda in contrast have been comparatively close to the mineral-rich east.
Their influence in the east has often rivalled or surpassed that of the Congolese
state. In addition to home-grown groups, rebels from other countries have also
been drawn to the region. Rwanda’s former military and militia responsible for its
genocide, Burundi’s Hutu rebel group, the Palipehutu-FNL, and Uganda’s Lord’s
Resistance Army have all operated from the DRC’s lawless east, for example. Its
terrain has provided an attractive cover for many of the region’s insurgents.
9 MONUC: ‘Mission Des Nations Unies en République Démocratique du Congo’.

560 omar mcdoom
C H A R AC T E R I S T I C F O U R : P R E DATO RY A R M I E S
................................................................................................................
Soldiers—both government and rebel—have been responsible for much of the
violence committed against civilians in the region. A shortage of modern, profes-
sional armies trained in international humanitarian norms of warfare is another
defining characteristic of the region. Indeed, civilian deaths outnumber those of
armed combatants. In wartime, rape and pillage continue to be seen as legitimate
rewards. Even in peacetime soldiers ‘tax’ civilians with impunity, a problem
exacerbated when they are poorly paid. The Zairean Armed Forces under Mobutu,
particularly towards the end of his regime (event C4), and rebel forces in the post-
Mobutu eastern Congo have both exemplified this predatory military behaviour.
Military establishments have enjoyed unusual influence within each country,
mainly because many of the region’s authoritarian regimes emerged from coups led
by military elites. The lack of legitimacy of military-headed regimes led their
leaders to rely on and to privilege their armed forces in order to protect their
positions, and to recruit on the basis of ethnic and regional loyalties. In Uganda,
Milton Obote made the northern ethnic groups, the Acholi and Langi, the back-
bone of his army between 1980 and 1986. Yoweri Museveni subsequently purged
them from the national army, replacing them with groups that had been loyal to
him during the Bush War (event UG2). Ethnic affiliation has been important in
rebel recruitment as well, as exemplified by the numerous ethnically based militia
that emerged in the second Congo war (event C7): Lendu, Hema, Hutu, and Tutsi
have all formed the backbone of rebel groups in eastern Congo.
C H A R AC T E R I S T I C F I V E : A N U N S TA B L E
R E G I O NA L B A L A N C E - O F -P OW E R
................................................................................................................
Many of the internal conflicts in the region have been connected to broader
geopolitical rivalries. Proxy wars have been an important means through which
the regional balance-of-power has been altered. The potential profit from natural
resource exploitation in the DRC—especially since the fall of Mobutu—has been
an added incentive for neighbouring countries to sponsor proxy rebel groups.
Often, this has happened covertly. Rwanda and Uganda, for example, have denied
their alliances with various rebel groups in eastern Congo despite reports that they
have advised and armed them since the late 1990s. Alternatively, neighbours have
allowed rebels to use their territories as rear bases from which to launch attacks. In
the 1960s, Burundi allowed Tutsi exiles from Rwanda to make repeated incursions
over its border in their bid to return home (event R2). Similarly, since the early
1990s Hutu rebel groups have initiated military campaigns from Tanzania and the
DRC into Burundi in their bid to topple the Tutsi-dominant regime (event B5).
Rwanda’s genocide in 1994 dramatically shifted the regional balance-of-power.
Rebel victory ended the genocide, overthrew Rwanda’s extremist Hutu govern-
ment, and put a Tutsi-dominant government into power. The genocidal Hutu army
and militia went into exile in eastern Congo. The ensuing instability, and the
insecurity that accompanied it, has become an important rationale and driver of
military action in the region. The former Hutu army and militia continued their
attacks on Rwanda and also targeted Congolese Tutsi in eastern Congo. In
response, Rwanda has made numerous incursions into eastern Congo in the
name of self-defence. It points to the regrouping of the Rwandan Hutu army and
militia first as the ALIR (Armée pour la Libération du Rwanda) and then as part of
the FDLR (Forces Démocratiques pour la Libération du Rwanda) as proof of the
threat at its borders (event C8). It also claims it is acting to prevent another
genocide, this time of Congolese Tutsi. This unstable balance-of-power persisted
at the end of 2007 and has been at the heart of much of the region’s most recent
period of violence.
C H A R AC T E R I S T I C S I X : R E F U G E E C O M M U N I T I E S
A N D T H E C YC L E O F V I O L E N C E
................................................................................................................
The scale of the refugee problem in the Great Lakes is the sixth defining contextual
characteristic for violence in the region. The official number of recognized refugees
in the DRC, Burundi, Uganda, and Rwanda stood at 542,746 at the end of 2006.10
There were an additional 2,675,321 internally displaced persons (IDPs) receiving
assistance from UNHCR. Refugees have been an important mechanism in the cycle
of violence. They have been both the product of conflict and the source of further
conflict in the region. The role of refugee camps in perpetuating violence in the
region has played out in two main ways. First, they have acted as a sanctuary for
rebel fighters. As war has approached civilians invariably have fled en masse,
usually across the nearest border. Rebel soldiers have often fled with them and
distinguishing combatants from civilians has been problematic. Located just over
the border, out of reach of government forces, rebel combatants regroup and
10 See UNHCR, Statistical Yearbook, Trends in Displacement, Protection, and Solutions 2006
(Geneva). In addition there are just over 393,000 Burundian refugees in Tanzania, mainly the
product of Burundi’s 1972 genocide.
562 omar mcdoom
recover. In addition refugee and IDP camps have attracted the humanitarian
industry. The food and medical supplies relief workers have brought have made
easy pickings for both government and rebel combatants. Rwanda’s genocide in
1994 resulted in an estimated two million Hutu civilians fleeing the advance of the
rebel group, the Rwandan Patriotic Front (RPF) (event R5). The majority sought
sanctuary in eastern Congo along with government soldiers and militia implicated
in the genocide. The international community, in contrast to its intransigence
during the genocide, rapidly deployed massive humanitarian resources to the
region to assist the refugees. As a result, it stood accused of also assisting many
of Rwanda’s génocidaires.
The second refugee–conflict linkage has been through armed return. Caught
between unwelcoming hosts and repressive home countries, refugee communities
have often mobilized to fight their way back to the old country. Tutsi refugees from
Rwanda’s 1959 Hutu revolution (event R1) established the Rwandan Alliance
for National Unity in 1979, the antecedent to the Rwandan Patriotic Front.
The RPF went on to launch Rwanda’s civil war in 1990 that culminated in its
genocide. Similarly, Burundi’s 1972 Hutu genocide resulted in Hutu refugee camps
established just over the border in Tanzania (event B2). These camps gave birth
to the radical Hutu rebel group, the Palipehutu, in 1972 and subsequently to
the more moderate FRODEBU in 1980. Each was determined to break the Tutsi
grip on power in Burundi and started a chronic and bloody rebellion to do so
(event B5). Yet it is not just through their support of rebel activity that refugee
camps have made casualties of civilians. The camps established have often been
death traps in themselves. Disease, usually malaria, has been the chief killer. Bodies
weakened from hunger and fatigue have easily succumbed and led to mortality
rates far in excess of those under peacetime conditions. Internal displacement
camps created during Uganda’s and Burundi’s civil wars have been major sources
of death. In both cases, civilians were forced into these camps as part of a
government counter-insurgency strategy. Displacement camps strongly resembled
internment camps.
C H A R AC T E R I S T I C S EV E N : E T H N I C I T Y A N D
E XC LU S I O NA RY I D E O LO G I E S
................................................................................................................
Although tribalism is too simple a lens through which to view violence in the
region, the importance of ethnic identity should not be dismissed altogether. Most
importantly, ethnic differences have served as useful narratives or become the
basis of exclusionary ideologies. As such, they have aided in the mobilization of
public support, bolstered recruitment during wartime, and provided cohesion

among rebel and government soldiers. At times, these narratives have also
provided the rationale for violence. One of the most well-established ethnic
narratives has been that of minority Tutsi domination and manipulation in the
region. In Rwanda, both before and during the genocide, extremists played on
collective Hutu memory of historical subjugation by Tutsi to mobilize popular
anti-Tutsi sentiment. A Tutsi ruling elite had indeed governed Rwanda’s socio-
political order prior to Rwanda’s Hutu revolution in 1959. In Congo, the Tutsi
minority in its eastern territories has historically also been the target of popular
resentment and state exclusion. In 1981, Zaire passed a law that called into
question the citizenship of its Kinyarwandaphone Tutsi community. In South
Kivu province, a subset of Congolese Tutsi, the Banyamulenge, has been the target
of deep-seated animosity from other ethnic groups. This animosity has provided
Rwanda, controlled by the Tutsi-dominant RPF, with the rationale for repeated
incursions into eastern Congo following the end of the second Congo war in 2002
(event C7). The Tutsi-controlled RPF claimed that its ethnic kin were likely to
become victims of another genocide. This same rationale has also provided
eastern Congo’s infamous renegade general, Laurent Nkunda, with the justifica-
tion for not integrating his forces into Congo’s new national army following the
end of the second Congo war. He cited the threat to his Tutsi brethren posed by
the FDLR, the armed rebel group composed in part of Rwandan Hutu génoci-
daires that continues to operate in the East.
Yet the violence cannot be reduced to the mere existence or multiplicity of ethnic
or tribal differences. First of all, the region’s ethnic conflicts have usually involved
only a small number of the myriad ethnic groups within it. Rwanda and Burundi
are highly unusual demographically in sub-Saharan Africa in that they are essen-
tially bi-ethnic societies, each composed of a Hutu majority and a Tutsi minority.
Violence has indeed followed ethnic lines in these countries. Yet Uganda and the
DRC are multiethnic societies and only a small handful of their ethnic groups have
been drawn into violent conflict. The vast majority have coexisted without blood-
shed. Secondly, violence has been as much the product of intra-ethnic power
struggles as it has interethnic rivalries. In the run-up to Rwanda’s genocide, the
principal tension was between Hutu moderates and extremists. The newly-created
political parties split along these lines as hardliners rejected a peaceful settlement of
Rwanda’s ongoing civil war (event R4). In Burundi, the numerous Hutu rebel
groups in its civil war factionalized along similar lines as rival leaders emerged
(event B5). The net effect has been to prolong the fighting.
In short, it is not the mere existence or diversity of ethnic groups that is a
defining characteristic of the region. Rather, it is the degree to which myths and
narratives about certain ethnic groups have become entrenched in societies and are
utilized in wars and genocides that is important.
564 omar mcdoom
C O N C LU S I O N
................................................................................................................
The focus of this chapter has been violence against civilians. The chapter has
suggested some of the important characteristics which have defined the context in
which the wars and genocides responsible for the region’s very high civilian death toll
have occurred. Yet it should be acknowledged that the estimates are problematic. The
estimate in this chapter was aggregated from secondary sources. However, the
component figures are sometimes given without an explanation of the methodology,
if any, used to establish them. We need more reliable estimates. Reports based on
nationally representative surveys have started to appear. Moreover, these have re-
vealed that violence has not been the most important cause of death in wars and
genocide. Disease and malnutrition have been the chief culprits. Damage to the state’s
health and sanitation systems has killed indirectly. In the DRC alone an estimated 5.4
million people died between August 1998 and April 2007 in excess of the norm.11 As
many as 2.1 million deaths occurred after the formal conclusion of the Second Congo
War in 2002. Moreover, the overwhelming majority were not victims of violence but
instead individuals who had succumbed to illness and undernourishment, usually
children under the age of five. Thus, while the region’s wars and genocides evoke
images of AK-47s and machetes, we should not forget the less visible but much more
deadly and chronic effects of these forms of violence on civilian populations.
Appendix
SUMMARIES OF INDIVIDUAL EPISODES

OF VIOLENCE IN REGION
................................................................................................................
I. U G A N DA
................................................................................................................
Event UG1: 1971–9: State terror under Idi Amin
Idi Amin presided over a highly authoritarian regime that used violence to maintain
control of the state and to repress its civilian population. Following his military coup
bringing him to power in 1971, Amin purged the various security institutions of
individuals seen as favourable to his ousted predecessor, Milton Obote. His new state
11 See International Rescue Committee, Mortality in the Democratic Republic of Congo: An Ongoing
Crisis (New York: International Rescue Committee, 2007).
security apparatus institutionalized violence as the basis of authority. Influential elites

and ordinary civilians alike were imprisoned, tortured, and murdered, sometimes for
their perceived opposition to Amin but sometimes arbitrarily. In addition, between
1977 and 1978, Amin authorized death squads to target ethnic Acholi and Langi elites
in response to sabotage activities by Obote agents. Between 1978 and 1979, conspira-
cies and mutinies to topple Amin resulted in further purges. Amin’s policies resulted
in the estimated killing of between 50,000 and 300,000 civilians.12
Event UG2: 1981–5: Civil War

A multipronged rebellion begins following contested national elections. After Idi Amin
was forcibly removed in 1979 by an alliance of forces led by Tanzania, weak transitional
government motivated Milton Obote to seize power a second time in 1980. National
elections to legitimize Obote’s coup were rigged in his favour and this precipitated an
armed challenge to his authority. The most significant rebel group, the National
Resistance Movement under Yoweri Museveni, came to control an area known as
the Luwero triangle, leading to a brutal counter-insurgency campaign by Obote’s
Ugandan National Liberation Army (UNLA). The UNLA committed many atrocities
against mainly ethnic Bagandan civilians in this area. In addition, the UNLA also
perpetrated atrocities in the West Nile region where the rebel group, the Ugandan
National Rescue Front, operated and also preyed on the civilian population. The West
Nile region was the homeland of Idi Amin and the UNLA was composed of mainly
ethnic Acholi and Langi who had been previously targeted by Amin. Altogether it is
estimated between 200,000 and 500,000 civilians were killed in this period.13
Event UG3: 1986–2006: Northern Ugandan Insurgencies

Following the 1986 victory of the National Resistance Movement (NRM) in the civil war,
several insurgent groups emerged in northern Uganda to challenge Yoweri Museveni’s
new government. The most violent and persistent of these, the Lord’s Resistance
Army (LRA), began in 1987 led by Joseph Kony. Claiming a spiritual motivation
based on the Ten Commandments, the LRA began a twenty-year guerrilla campaign
in northern Uganda. The violence against civilians waxed and waned in this period.
The main victims of the violence were the ethnic Acholi population, from which
the LRA also abducted children to serve in the rebel organization. However,
the insurgency spread to affect ethnically Teso and Langi regions. As part of its
counter-insurgency campaign the government forced the civilian population into
12 Kasozi, Musisi, et al., The Social Origins of Violence in Uganda, 104.

13 Ibid. 145.
566 omar mcdoom
displacement camps. At its peak in 2002, over 1.6 million lived in these camps.
According to a 2005 World Health Organization survey, mortality rates reached
1,000 excess deaths per week, mostly from illness. The LRA raided these camps and
conducted ambushes on the roads, killing civilians until peace talks began in
southern Sudan in 2006. As these continued, the security situation remained fragile.
II. D E M O C R AT I C R E P U B L I C OF CONGO
................................................................................................................
Event C1: 1959–61: Transition to Independence
and Secessionist Wars
Almost immediately following independence in 1960 two provinces, Katanga and
S. Kasai, seceded. The national army was ordered to suppress the two rebellions.
In South Kasai, where conflict between ethnic Luba and Lulua had ignited in 1959,
the army killed mostly Luba civilians as it headed to end the Katangan secession. At
the request of Congo’s new Prime Minister, Patrice Lumumba, the UN fielded a
20,000-strong mission to the region. The secessions were brought to an end in 1962.
No figure exists for civilians killed in South Kasai though the UN Secretary-
General, Dag Hammarskjold, characterized it as genocide.
Event C2: 1964–5: Lumumbist Rebellion

Supporters of the assassinated former Prime Minister Patrice Lumumba launched a
rebellion on four fronts in 1964 to topple the government of his rival, President Joseph
Kasa-Vubu. The commander-in-chief of the army, Joseph-Désiré Mobutu, contracted
foreign mercenaries to assist the national army in quashing the rebellion. In Septem-
ber 1964 one rebel group succeeded in capturing Kisangani, in North Congo, where it
declared a separate government. In a combined offensive Belgian paratroopers
dropped by US planes, foreign mercenaries, and elite units of the national army
destroyed the rebel government. Seventy Europeans and nearly 1,000 Congolese were
killed during this operation, code-named Red Dragon. The involvement of foreign
troops made Kasa-Vubu unpopular with other African Heads of State. Mobutu
exploited his unpopularity to seize power in a military coup in 1965.
Event C3: 1977–8: Shaba Secessionist Wars

In 1977 Katangan secessionists, the Front for the National Liberation of Congo (FNLC),
attacked the province of Shaba (aka Katanga) from across the border in Angola. As
their advance captured towns, Mobutu sent in the national army assisted by Moroccan
troops who successfully repelled the invaders. With the rebels routed, the national
army launched a ‘pacification’ campaign against the Katangan civilian population. The
casualty count is unknown but the campaign generated an estimated 200,000 refugees.
The FNLC launched a second attack in 1978 and this time captured Kolwezi town in
South Katanga. This time French and Belgian troops were sent in to expel the FNLC.
The International Red Cross found the bodies of 95 Europeans and 111 Zairean
civilians. Reports conflicted as to whether they were deliberately massacred by rebels,
caught in crossfire, or killed by Congolese government soldiers.
Event C4: 1991: Decline of Mobutu’s Regime

Civilians were the victims of violence committed by and against a weakening Mobutu
regime. Under internal and international pressure to democratize, Mobutu permit-
ted the creation of multiple political parties and a Sovereign National Congress.
Aid donors also dramatically cut their support to Zaire. As a result Mobutu began
to lose control of the state. Student protests in Lubumbashi in 1991 were sup-
pressed, resulting in a massacre of fifty to a hundred. Unpaid soldiers rioted in
Kinshasa in 1991 and again in 1993 that left altogether 145 civilians dead. Further
protests in Kinshasa in 1992 were quashed by the army, resulting in 30 dead.
Event C5: 1993: Decline of Mobutu’s Regime continued

In North Kivu, eastern Congo, violent conflict over land following ethnic lines broke
out. Mobutu sent the Presidential Guard and Civil Guard to quell the fighting.
Without means to sustain themselves, the soldiers instead preyed on the civilian
population. The arrival of the Rwandan Hutu military and militia following
Rwanda’s genocide in 1994 turned the focus of the violence against the Congolese
Tutsi community. No body count was done but newspapers reported thousands of
civilians were killed and 200,000 displaced.
Event C6: 1996–7: First Congo War

As Mobutu’s regime weakened, in October 1996 neighbouring Rwanda and Uganda co-
sponsored an alliance of four rebel groups in eastern Congo, the Alliance of Democratic
Forces for the Liberation of Congo (AFDLC) led by Laurent Kabila, to capture
Kinshasa. The war began when Rwanda, claiming its security was threatened by
Hutu soldiers and militia responsible for Rwanda’s genocide in 1994, crossed into
eastern Congo to dismantle forcibly the refugee camps within which the fugitive
génocidaires were operating. Rwandan soldiers massacred Hutu refugees in the
568 omar mcdoom
camps. The death toll is unknown as the UN was forced to stop before completing its
investigation. AFDLC troops then moved west and captured Kasai and Katanga
provinces, and with the aid of Angolan troops, finally took Kinshasa in May 1997.
Mobutu fled the country and Kabila was installed as the new President.
Event C7: 1998–2002: Second Congo War

Civilians were the principal victims of a second proxy war between the DRC and its
neighbours. In order to improve his domestic popularity, DRC’s new President,
Laurent Kabila, distanced himself from his sponsors by expelling Rwandan and
Ugandan advisors from the country. He instead attacked the unpopular Congo-
lese Tutsi minority in the east with the aid of the fugitive Hutu army and militia
responsible for Rwanda’s genocide. As a result his former allies, along with
Burundi, invaded again in 1998, with their own troops and through two new
proxy rebel groups. Rwanda sponsored the Congolese Rally for Democracy (RCD)
and Uganda supported the Movement for the Liberation of Congo (MLC). On the
other side Kabila received support from Angola, Zimbabwe, and Namibia initially
and Sudan, Chad, and Libya later. As the war progressed, it became more
complex. Rwanda and Uganda turned from allies into enemies and the rebel
groups splintered into factions. Kinshasa did not fall but the country instead
divided into three de facto regions. Broadly, Uganda and aligned rebel groups
controlled the north, Rwanda and its aligned rebel groups controlled the east, and
Kabila and his allies controlled a southern strip running from Kinshasa to
Katanga. Each plundered their territory, taxing civilians and extracting its natural
wealth. Hundreds of thousands were displaced by the violence. The state appara-
tus, including vital healthcare, water, and sanitation services, collapsed. An Inter-
national Rescue Committee survey estimates that by 2002 3.3 million people had
died, mostly of disease and hunger. As much as 1.6 per cent had died of violence
in the survey period. In 1999 the Lusaka Agreement led to the deployment of an
eventually 17,000-strong UN mission (MONUC) but the ceasefire was ignored. In
2001 Laurent Kabila was assassinated by one of his bodyguards and his son, Joseph
Kabila, was installed in his stead. Finally in December 2002 a ‘Global All Inclusive
Agreement’ was signed under which a transitional power-sharing government was
to be installed and a new constitution and national elections were to be prepared.
Although the violence diminishes, it does not stop, least of all in the east.
Event C8: 2002 onwards: Warlordism in Eastern Congo

Civilians continued to be caught between various rebel groups and government forces
in eastern Congo. Following the signing of the Global All Inclusive Agreement in
2002 to end the war, the principal armed rebel groups agreed either to demobilize
or to integrate into the new national army. However, rebel factions that refused to
join emerged in the east and chose to remain warlords. The most important of
these was the North Kivu group led by Laurent Nkunda and aided by Rwanda. It
stated it would not disarm while the fugitive and mainly Hutu rebel group, the
Democratic Forces for the Liberation of Rwanda (FDLR), remained at large and
continued to threaten Congolese Tutsi, in particular an ethnic Tutsi subset, the
Banyamulenge of South Kivu. The national army, FARDC, continued to battle
Nkunda and was widely believed to collaborate with the FDLR against him. At the
same time the integrated FARDC continued to suffer both from defections such as
those of Mai Mai militia units and from ill-discipline with reports of its involve-
ment in looting and rape. All of these groups continue to clash and Nkunda
remains at large at the end of 2007. In another region, Ituri in north-eastern
DRC, conflict ignited in 1999 between ethnic Lendu and Hema rebel groups over
land. Uganda, whose troops controlled the territory played the role of both
‘arsonist’ and ‘fireman’. The fighting quickly devolved into criminal predation on
the civilian population. It is believed several thousand civilians from both groups
were killed during this period. At the end of 2007, the Ituri-based rebel groups
agreed to integrate into the national army. Amid these various conflicts, the UN
mission, MONUC, attempted to remain neutral but its peacekeepers became
targets themselves. Elections in 2006, which passed with little pre- or post-electoral
violence, resulted in victory for Joseph Kabila over former MLC rebel leader Jean-
Pierre Bemba. However, this has had only a marginal impact on violence in the east.
Rebel groups, and their foreign sponsors, continue to prey on the population and
extract the natural wealth of the territories they control. At the end of 2007, Ituri,
North and South Kivu, and Katanga provinces remain insecure and massive dis-
placement, in the order of hundreds of thousands of people, continues. An Interna-
tional Rescue Committee survey in 2004 estimated 3.9 million have perished since
1998. Two per cent of all deaths in the survey period were the result of violence.
III. B U RU N D I
................................................................................................................
Event B1: 1965: Transition to Independence: Political
Assassinations
A failed Hutu coup and Hutu attacks on Tutsi civilians led to reprisal attacks against
Hutu civilians by Tutsi elements of the army and a youth militia. In the run-up to
independence in 1962, Burundi held national elections in September 1961 contested
between two rival groups of traditional princes (ganwa): the Bezi, represented by
570 omar mcdoom
the UPRONA party, and the Batare, represented by the Christian Democrat Party
(PDC). Louis Rwagasore, leader of the UPRONA party, won only to be murdered a
few weeks later by a PDC-hired assassin. In the ensuing leadership crisis the main
political cleavage evolved from a Bezi–Batare ganwa rivalry into a Hutu–Tutsi
ethnic antipathy within the UPRONA party. Burundi’s traditional monarch,
Mwami Mwambutsa—historically popular among both Hutu and Tutsi—resumed
a governing role and called for legislative elections in May 1965 after the Hutu
Prime Minister he appointed, Pierre Ngendandumwe, was also assassinated three
days into his office. Although Hutu won a majority, the Mwami instead appointed
a ganwa, Léopold Biha, as Prime Minister. As a result a small group of Hutu army
officers and gendarmes shot Biha (albeit not fatally) in the Mwami’s Royal com-
pound only to be stopped by Tutsi army officers led by Captain Micombero. At the
same time Hutu mobs in the northern province of Muramvya mistakenly believed
the Tutsi had turned against the Mwami and attacked Tutsi civilians. This resulted
in reprisals by the army and radical UPRONA youth wing, the Jeunesse Révolu-
tionnaire Rwagasore (JRR), in which an estimated 5,000 Hutu civilians were
killed.14 The Biha attack also resulted in a purge of Hutu soldiers, gendarmes,
and politicians, leaving the security forces exclusively within Tutsi control. The
Mwami left the country and in November 1966 the young Tutsi Hima military
officer Michel Micombero seized power from the Mwami’s appointed regent,
Ndizeye, to become the President of Burundi’s First Republic.
Event B2: 1972: Selective Genocide of Hutu Elite

A Hutu insurrection resulted in the army and youth militia committing reprisal
genocidal violence against Hutu elites across the country. Following his coup, President
Micombero moved to make Burundi’s First Republic a Tutsi ethnocracy. Claiming a
conspiracy to overthrow the government, Micombero purged senior Hutu politi-
cians and army officers in 1969. Then in 1971 a rival Tutsi group, the high-status
Banyaruguru from Muramvya province, was politically neutralized in a series of
show trials. It was in this context that on 29 April 1972 an organized Hutu insurrec-
tion arose to challenge Micombero’s authoritarian and ethnocentric regime. The
insurgents orchestrated a simultaneous attack in the capital Bujumbura and in
Rumonge, southern Burundi. The Tutsi-dominant army quickly suppressed the
Bujumbura attack—which involved only 40–60 insurgents—but the Rumonge
attack, comprising 300–500 rebels, killed 1000–2000 Tutsi civilians before it was
stopped on 5 May. The army massacred tens of thousands of Hutu in its counter-
insurgency in the rebellious region. In the first week in the capital senior Hutu
14 R. Lemarchand, Burundi: Ethnic Conflict and Genocide (Washington, DC/Cambridge: Woodrow

Wilson Center Press, Cambridge University Press, 1996), 72.
politicians, civil servants, and army officers were arrested on suspicion of collabora-
tion and then executed. Then after 5 May, the army with the radical UPRONA youth
wing, the JRR, escalated the violence and began to massacre systematically Hutu
elites across the country. State employees, teachers, businessmen, professionals,
university students, and high school students all became victims. The Tutsi civilian
population was mobilized to kill as well, though the extent of its participation and
the degree to which it was willingly complicit are unclear. The total number of Hutu
killed remains uncertain but one estimate places the casualty count at between
100,000 and 200,000 in the space of three months.15 Having eliminated the Hutu
counter-elite, Tutsi hegemony was assured for some time. This ‘selective’ genocide
also led to a massive exodus of several hundred thousand Hutu refugees, most of
whom sought safety in neighbouring Tanzania. The international community, while
aware of the violence, did nothing to intervene and stop it.
Event B3: 1988: Hutu Insurrection

A spontaneous Hutu uprising led once again to reprisal violence from the Tutsi army
against Hutu civilians. Although there had been two palace coups since the 1972
genocide, Tutsi control of the State continued. As with his two ousted predecessors,
the President of the Third Republic, Pierre Buyoya, belonged to the Tutsi Hima
subgroup and originated from Bururi in southern Burundi. After he seized power
in 1987, however, Buyoya stated he wished to improve ‘national unity’. In practice
little altered to redress the Hutu–Tutsi power imbalance. In 1988 two communes in
northern Burundi became the flashpoint for renewed ethnic violence. In response
to rising ethnic tensions the gendarmerie was deployed to the region. However, this
escalated tensions as Hutu feared a recurrence of 1972 and organized self-defence
groups, manned barriers, and destroyed bridges. Hutu then attacked a local Tutsi
businessman known for his role in 1972, which triggered a wave of violence leaving
several hundred Tutsi dead. The army responded using helicopters and armoured
cars to crush the uprising and killed an estimated 15,000 Hutu, causing about
50,000 Hutu to flee into Rwanda.16
Event B4: 1993: Assassination of Hutu President Ndadaye

The assassination of Burundi’s first democratically elected Hutu Head of State
resulted in localized Hutu attacks on Tutsi civilians that were in turn brutally
15 R. Lemarchand, Burundi: Ethnic Conflict and genocide (Washington, DC/Cambridge: Woodrow

Wilson Center Press, Cambridge University Press, 1996), 100.
16 See R. Lemarchand, ‘Burundi at a Crossroads’, in G. M. Khadiagala (ed.), Security Dynamics in
Africa’s Great Lakes Region (Boulder, CO: Lynne Rienner, 2006), 126.
572 omar mcdoom
suppressed by the Tutsi army. In the aftermath of the 1988 violence, President
Buyoya, a Tutsi, continued his quest for ‘national unity’. In March 1993 a new
constitution was adopted that permitted multipartyism. National elections held in
June 1993 resulted in victory for the mainly Hutu FRODEBU party. In a historic
moment Melchior Ndadaye, a Hutu, became Burundi’s new President. However,
on 21 October 1993 low-ranking Tutsi soldiers assassinated Ndadaye (though a UN
Commission suggested the army’s senior command was complicit). In response
FRODEBU Hutu activists took hostage Tutsi men, executed them, and then went
on to massacre Tutsi women and children. On the same day the army moved out to
rescue the Tutsi survivors and killed Hutu civilians wherever it encountered them.
An estimated 30–100,000 Hutu and Tutsi civilians were killed in this way. Several
hundred thousand Hutu fled into Rwanda.
Event B5: 1993–2006: Burundian Civil War

Ndadaye’s assassination led to a multifronted civil war between various Hutu rebel
groups and the Tutsi army in which both Hutu and Tutsi civilians were targeted.
Following the short-lived coup in 1993, Hutu rebel attacks gradually increased and
Burundi slid into civil war. The sudden influx of Hutu refugees in 1994 from Rwanda’s
genocide aggravated this. In 1996 Pierre Buyoya seized power a second time in order
to restore order but the violence continued to escalate. The army, acting increasingly
independently of the civilian government, implemented brutal counter-insurgency
measures. Notably it forcibly moved civilians into displacement camps where harsh
conditions caused a humanitarian crisis. Hutu rebel groups engaged in a guerrilla war
with the Tutsi army. They rarely confronted each other directly and instead civilians
were the main victims of the violence. The Tutsi army commited multiple atrocities
against Hutu civilians and Hutu rebels targeted civilians, both Tutsi and Hutu
collaborators. The second Congo war in 1998 interweaves itself with the violence in
Burundi. It caused rebel groups to factionalize and encouraged cross-border alliances.
A peace process for Burundi eventually led to the Arusha Accord in 2000. It stipulated
a ceasefire and a three-year transition. Although many parties signed on, the two main
rebel groups, the CNDD-FDD and the Palipehutu-FNL, refused to do so and the war
continued. Nonetheless in a historic moment at the end of the transition in 2003,
Buyoya, a Tutsi, stepped down as President in favour of Ndayizeye, a Hutu. Continued
peace talks with the two outstanding rebel groups eventually brought the CNDD-
FDD into the government in 2005. The Palipehutu-FNL agreed to lay down its
weapons in 2006. To date a very rough estimate is that 300,000 Hutu and Tutsi have
been killed as a result of violence in the civil war.17
17 Ibid. 41.
IV. R WA N DA
................................................................................................................
Event R1: 1959–62: Hutu Revolution
Before independence, a short burst of mainly anti-Tutsi violence set in motion a series
of events that led to a revolution overthrowing the Tutsi monarchy. Colonial Belgium
had ruled Rwanda indirectly by reinforcing the rule of a Tutsi minority elite over
the country’s Hutu majority. As independence approached, however, Belgium co-
sponsored with the Church a Hutu counter-elite. Ethnic tensions arose for control
of the post-independence state. The trigger for violence was an attack on a Hutu
political leader that provoked a rapid backlash against Tutsi across the country,
leaving 200 dead. It also started a Tutsi exodus into Tanzania, Burundi, the DRC,
and Uganda that reached 600,000 by 1973. In 1962 Rwanda officially became a
Republic and Gregoire Kayibanda its first Hutu head of state.
Event R2: 1962–4: Tutsi Insurgent Attacks and Massacres

Tutsi exiled in Burundi by the Hutu revolution launched two attacks on Rwanda in
1962 and 1963–4 that resulted in government-sponsored massacres of Tutsi civilians.
These were part of a wider series of insurgent attacks made between 1962 and 1967 to
overthrow the fledgling Hutu Republic, and they left an estimated 6–13,000 dead.18
The 1963–4 attack nearly reached the capital, Kigali, before it was repulsed by the
Rwandan army. The government then organized local self-defence committees to
defend against the enemy but in practice they targeted Tutsi civilians in reprisal.
Event R3: 1973: Ethnic Purges and Transition to Second Hutu

Republic
The over-representation of Tutsi in educational establishments became the rationale
for popular ethnic purges. Ethnic tensions were high in the wake of the Hutu
genocide in Burundi the year earlier. They crystallized in January 1973 when
Hutu students marched and demanded the expulsion of Tutsi students from second-
ary schools and the university to redress the ethnic imbalance. The purges spread
to the public and private sectors and several dozen Tutsi were killed. Hutu army
officers from northern Rwanda exploited the instability to overthrow the regime
18 R. Lemarchand, Rwanda and Burundi (London: Pall Mall Press, 1970), 216, 219. See also S. Straus,
The Order of Genocide: Race, Power, and War in Rwanda (Ithaca/London: Cornell University Press,
2006), 186.
574 omar mcdoom
dominated by Hutu from the centre and south of the country. General Juvénal
Habyarimana became president of the Second Republic.
Event R4: 1990–4: Rwandan Civil War

Tutsi civilians were massacred in response to attacks by the mainly Tutsi rebel group,
the RPF. In October 1990 the RPF, composed mostly of descendants of Tutsi exiles
from the 1959 Hutu revolution, invaded Rwanda from across the Ugandan border.
A guerrilla war ensued. The Rwandan government targeted its own Tutsi civilians
in reprisal, accusing them of collaboration with the rebels. It arrested 13,000 Tutsi
immediately after the initial attack in 1990 and tortured and killed several dozen.
An estimated 2,000 more were killed in several massacres in northern and central
Rwanda in which local authorities incited Hutu mobs to attack Tutsi civilians in
response to alleged enemy threats.
Event R5: 1994: Rwandan Genocide

The assassination of Rwanda’s Hutu president, Juvénal Habyarimana, triggered a
nationwide extermination campaign that principally targeted the Tutsi ethnic group
in its entirety. Since 1973 Habyarimana and a small Hutu elite from the north of the
country had monopolized power to the exclusion of Tutsi and Hutu from other
regions. The confluence of the start of a civil war in 1990 and the advent of
multipartyism in 1991 together seriously threatened the power of the regime. The
rebel RPF was able to capture and hold ground and the newly-created political
parties were able to mobilize popular support against the Habyarimana govern-
ment. In 1993 peace talks in Arusha yielded a power-sharing agreement between
President Habyarimana, the mainly Tutsi rebel RPF, and the mainly south-central
Hutu opposition parties. However, this arrangement divided the ruling Hutu elite
and a hardliner group opposed to sharing power emerged. When Habyarimana’s
plane was shot down on 6 April 1994, it was widely assumed within Rwanda that
the rebel RPF was responsible. The hardliner minority immediately took the
opportunity to seize control of the state. It unleashed the Hutu-dominant army
and youth militia against Tutsi civilians and moderate Hutu. In rural areas local
authorities followed the lead of the new extremist central government and mobi-
lized the Hutu civilian population to ‘defend itself ’ against the Tutsi enemy. Those
local authorities that resisted either were replaced or eventually gave in to the
genocidal project. Many ordinary civilians faced the same choice. Over the next 101
days attack groups composed in the main of Hutu civilians hunted down Tutsi
civilians—men, women, and children often known to them personally. The slaugh-
ter continued until the rebel RPF finally won the war and pushed the extremist
government and armed forces over the border into the DRC. About two million
ordinary Hutu followed them, creating a massive humanitarian crisis in the refugee
camps established in Eastern Congo.
Event R6: 1997–8: North-west Insurgency

In the course of this insurgency rebels killed both Tutsi civilians and Hutu who did not
cooperate while government forces in turn targeted Hutu civilians who collaborated
with the insurgents. Following their defeat in the civil war in 1994 the Hutu army
and militia regrouped in Eastern Congo and continued low-intensity attacks
against RPF-controlled Rwanda. In 1997 they escalated their insurgency in the
north-west and directly engaged RPF forces. They continued attacks on Tutsi
civilians and initially enjoyed the support of the Hutu population—especially as
certain RPF commanders punished Hutu civilians in reprisal for insurgent attacks.
However, as RPF counter-propaganda took effect, the insurgents began to target
uncooperative Hutu as well. The start of the Second Congo War in 1998 dampened
the insurgency as Rwanda pushed its troops west in the DRC towards Kinshasa.
Amnesty International, Human rights in Uganda: Report (London: Amnesty International,
1978).
Hesselbein, G., The Rise and Decline of the Congolese State: An Analytic Narrative on State-
Making, Crisis States Working Papers, DESTIN (London: London School of Economics,
2007).
International Crisis Group, Congo at War: A Briefing on the Internal and External Players in the
Central African Conflict (Nairobi/Brussels: International Crisis Group, 1998).
Kasozi, A. B. K., N. Musisi, et al., The Social Origins of Violence in Uganda, 1964–1985
(Montreal/Buffalo: McGill-Queen’s University Press, 1994).
Lemarchand, R., Burundi: Ethnic Conflict and Genocide (Washington, DC/Cambridge:
Woodrow Wilson Center Press/Cambridge University Press, 1996).
Mamdani, M., When Victims Become Killers: Colonialism, Nativism, and the Genocide in
Rwanda (Princeton, NJ: Princeton University Press, 2001).
Nzongola-Ntalaja, G., The Congo from Leopold to Kabila : A People’s History (London: Zed
Books, 2002).
Prunier, G., The Rwanda Crisis: History of a Genocide (Kampala, Uganda: Fountain, 1999).
Straus, S., The Order of Genocide: Race, Power, and War in Rwanda (Ithaca, NY/London:
Turner, T., The Congo Wars: Conflict, Myth, and Reality (London: Zed, 2007).
Young, C., and T. Turner, The Rise and Decline of the Zairian State (Madison/London:
University of Wisconsin Press, 1985).
part v
.............................................................................................
THE
CONTEMPORARY
WORLD: RULES
AND RESPONSES
.............................................................................................
chapter 28
.............................................................................................
THE UNITED
NAT I O N S , T H E
C O L D WA R , A N D
ITS LEGACY
.............................................................................................
gerd hankel
According to the Preamble of the Charter of the United Nations, the member
states of this organization resolved ‘to save succeeding generations from the
scourge of war’ as well as to act in a way that demonstrates ‘faith in fundamental
human rights, in the dignity and worth of the human person’. This was the spirit in
which the UN General Assembly passed the Convention on the Prevention and
Punishment of the Crime of Genocide on 9 December 1948 and adopted and
proclaimed the Universal Declaration of Human Rights on 10 December 1948.
As demonstrated by the emergence and consolidation of the cold war, the
reality of the situation was very different. The two superpowers pursued their
own agendas based on their respective power politics (Machtpolitik) and the
other states became trapped in block thinking. War and mass crime were only
perceived and condemned as such within the relevant camps if their members
could hold their ideological opponents responsible. For the most part, the United
Nations watched helplessly from the sidelines. The states were meticulous in their
efforts to ensure that the United Nations was not allocated any powers that could
have led to any appreciable infringement of their sovereignty. The significant
extent to which this broad understanding of sovereignty had taken hold is
580 gerd hankel
demonstrated by the difficulties faced by the United Nations since the end of
the cold war when it has tried to adopt an active peace and security policy and
enforce basic human rights.
I N S T I T U T I O NA L H I S TO RY A N D A R R A N G E M E N T S
................................................................................................................
The Second World War had a far greater influence on the world order established in
its aftermath than its predecessor, i.e. the First World War. The avoidance of mass
crimes, genocide, and violent imperial expansionism necessitated the adoption of
measures that, in the interests of humanity, would eliminate threats even remotely
comparable to those that Germany and its allies posed at the time. As in the case of
the First World War, the initiative was once again taken by the USA. Initially, it only
had plans for an alliance between the two leading ‘English-speaking’ democracies,
i.e., the USA and Great Britain, which would be solely responsible for an inter-
national police force for world peace to be deployed in the case of an emergency.
Following the German attack on the Soviet Union on 22 June 1941 and the Japanese
attack on Pearl Harbor on 7 December 1941, however, they both aspired to establish
a wider coalition of states. The Soviet Union and China were included in the
alliance and they, in turn, brought their own proposals to the table. By the Yalta
Conference in spring 1945, the preparations for the establishment of the new
alliance, for which the USA still held overall responsibility, had already reached a
stage where they could be presented as a concrete plan: ‘We are resolved upon the
earliest possible establishment with our allies of a general international organiza-
tion to maintain peace and security. We believe that this is essential both to prevent
aggression and to remove the political, economic, and social causes of war through
the close and continuing collaboration of all peace-loving peoples.’1
The founding conference of the United Nations took place in San Francisco from
25 April to 26 June 1945. The Charter of the United Nations came into force on 24
October 1945. With its 111 articles, it is around four times longer than the Covenant
of the League of Nations, the influence of which is clearly evident in a number of
points. Nation-states were at the centre of the new international system, as had
been the case at the time of the League of Nations. The notion of cooperation was
still relatively novel and was repeatedly trumped by power politics; in the event of
an external threat, the states—either individually or as an alliance—were clearly the
main actors in the international system. This had already been demonstrated in the
1 Quoted in Wilhelm G. Grewe and Daniel Erasmus Khan, ‘History’, in Bruno Simma (ed.), The
Charter of the United Nations: A Commentary, 2nd edn, vol. i (Munich: C. H. Beck, 2002), 7.
the united nations, the cold war, and its legacy 581
aftermath of the First World War and would be confirmed again after the Second,
but with two additional developments. First, the victors’ alliance would outlast the
war. It was to become the guarantor of a future peace, hence the name of the new
organization, the United Nations, which refers to a declaration made by nations in
early 19422 and commemorates the united front adopted in the fight against
military aggression. Second, as already announced in Yalta, the new organization
would help to eliminate the causes of the threat to peace and war, i.e., poverty,
injustice, and oppression.
According to the wording of the Charter of the United Nations, the responsibili-
ty for world peace lies with the Security Council. Article 24, subsection 1 of the
Charter states that ‘In order to ensure prompt and effective action by the United
Nations, its Members confer on the Security Council primary responsibility for
the maintenance of international peace and security, and agree that in carrying out
its duties under this responsibility the Security Council acts on their behalf.’
Compared with the complicated procedure that prevailed between the Council
and Assembly during the League of Nations era, this stipulation amounts to a clear
allocation of powers. However, it does not yet explain how this ‘international peace
and security’ should actually function. The only certainty was that intervention in
the area of state sovereignty had to be allowed for this purpose. Without the
limitation of state freedom of action, which was understood as an expression of
sovereignty, there could be no international peace plan based on the principle of
mutual accommodation. For this reason, Article 25 of the United Nations Charter
states that the decisions of the Security Council are binding on all UN Member
States. For the same reason, two articles later there is another provision that
initially appears to contradict the above statement, but ultimately constitutes
the precondition for its implementation. For, as demonstrated by the League of
Nations, without the involvement of the great powers an international peace
system is doomed to failure. Thus Article 27, subsection 3 of the Charter states
that decisions of the Security Council on matters that are not merely procedural,
i.e., questions concerning peace and international security in general, require the
agreement of all five permanent members of the Security Council—France was
included in this group at the San Francisco conference. In other words, it should
not be possible for any decision to be taken that is incapable of gaining the support
of the permanent members of the Security Council or, based on the less rigid
variant now in force, not acceptable enough to enable them to abstain from voting,
i.e., to waive their right of veto.
Thus the great powers established a privileged position for themselves. If this
may be seen as the expression of a general balancing of power, it should not be
forgotten that the main burden in the fight against the Axis powers was borne by
2 The ‘Declaration by United Nations’ of 1.1.1942, ibid. 7.

582 gerd hankel
Great Britain, her Dominions, the Soviet Union, and the USA, and that the political
realities of the time were such that they only made long-term stability achievable
through the establishment of a militarily strong centre of power.
What Article 2 of the UN Charter has to say in relation to the traditional rights
of states and the idea of the new international order can be explained on the same
predictable basis. It begins with a commitment to the ‘sovereign equality’ of the states
(subsection 1), then refers to the obligation to engage in the peaceful settlement of
disputes (subsection 3) and the prohibition on the use of interstate force (subsection
4), and ends with the prohibition on intervention in the essentially internal affairs of
another state (subsection 7). It goes on to state that, based on its superior position,
the Security Council alone is authorized to breach this principle of non-intervention.
In accordance with Chapter VII of the Charter, it alone has the right to impose
enforcement measures against a state in the event of a threat to or breach of the
peace, or in the context of negotiations in association with an attack.
Given that they are not included in the list provided in Article 2, the elimination
of poverty, injustice, and repression as the possible causes of new conflicts are not
specified as principles of the UN. According to Article 1, subsection 3, the espousal
of these causes is merely among the purposes of the organization; the use of the
qualification ‘merely’ here may be explained by the lack of legal quality of these
purposes. Unlike the principles listed in Article 2, they do not stipulate any rights
and obligations but an outcome that should—but does not have to—be achieved.
According to Article 1, subsection 3, the UN aims ‘to achieve international cooper-
ation in solving international problems of an economic, social, or humanitarian
character and in promoting and encouraging respect for human rights and for
fundamental freedoms for all without distinction as to race, sex, language, or
religion’. This is referred to again and formulated in partly identical terms in
Article 55 where it is also expressly stated that such conditions should give rise to
‘peaceful and friendly relations among nations based on respect for the principle of
equal rights and self-determination of peoples’.
Thus, it may be observed that human rights—if we adopt this term as a general
concept denoting a minimum level of social and economic participation—were
not allocated a particularly prominent position in the UN Charter. The Soviet
Union saw no link between human rights and international security, Great Britain
feared detrimental impacts on the stability of the Commonwealth because rhetoric
about human rights conflicted with its racist colonialism, and the attitudes of many
other states were also characterized by cautious reticence in this regard. Values
relating to states, such as equality and non-intervention, took priority, and consid-
eration of the basic rights that affect the lives of human beings was relegated to a
secondary position; moreover, because these rights were not based on any legal
obligation, compliance was dependent on the voluntary action of the individual
states. The peace, for which the UN stood, was, therefore, primarily a negative
peace, characterized by the absence of war. The additional task of establishing
a positive peace, understood as the establishment of the preconditions that would

make the outbreak of wars less likely due to their ethical unacceptability or
unreasonableness, came second and by some distance.
T H E C O L L A P S E O F T H E WA R C OA L I T I O N , T H E
B I P O L A R D I V I S I O N O F T H E W O R L D, A N D T H E
R E C E N T A C C E P TA N C E O F M I L I TA RY F O RC E
................................................................................................................
The gap between the initial aspirations of the UN and subsequent reality was a
sizeable one. The signs indicating the collapse of the war coalition were already
apparent at the foundation of the United Nations. They hinged on the USA’s
nuclear weapons project, which initially stood for the West’s distrust of Soviet
occupation policy in Central and Eastern Europe. The Soviet Union responded
with a ‘two-camp theory’, i.e., involving the ‘imperialistic-antidemocratic’ Western
camp and the ‘anti-imperialistic-democratic’ Eastern camp. The two camps stood
in irreconcilable opposition. Predicting subsequent developments, George Kennan,
the ‘inventor’ of containment policy, remarked in 1946 that, for Moscow, the UN
was not a mechanism for establishing a stable world order but merely an arena in
which states could pursue their own objectives for the maximum possible success.
The fact that the USA adopted a very similar policy can be seen in its voting
behaviour in the UN Security Council. While the Soviet Union used its veto on 115
occasions up to 1989, the USA did so in 69 cases (Great Britain in 30 cases, France in
18, and China in 3). Security policy was mainly practised outside of the UN. Just
three years after the foundation of the UN, the Western powers formed an alliance
in NATO that curtailed the security thinking of the original collective to its own
sphere of influence. This development was mirrored a short time later on the East-
bloc side with the establishment of the Warsaw Pact. Instead of one new global
security system, there were now two, a situation that prevailed until the end of the
military-political bipolarity. The Security Council imposed military sanctions on a
state on just one occasion—i.e., Korea in 1950, at the beginning of the war there,
and even this was not a real exception; the legality of this move was constantly
contested by the Soviet Union and may be seen as a component of the American
campaign against communism rather than an expression of a collective feeling of
responsibility.
What subsequently happened in the relevant spheres of influence was far more
fatal than the collapse of the collective security system. War after war was fought:
wars of independence, wars of aggression and defence, proxy wars, civil wars, in
584 gerd hankel
which both civilians and combatants were the target of the most heinous of war
crimes and crimes against humanity. Dictatorships were overthrown and replaced by
new ones. The principles of the sovereign equality of states and non-intervention in
the internal affairs of a state proved extremely useful instruments, particularly in
cases involving the arbitrary extension of the borders of the internal domaine réservé
of a state.
The number of war victims from the cold war period is estimated at over
20 million, with just one per cent, i.e., 200,000, originating from the northern
hemisphere. All of the others died in the numerous armed conflicts that shook
the southern hemisphere. Beyond the direct triggers of unresolved border issues,
regional power ambitions, and ethnic and religious tension, in almost all of these
conflicts the interests of the superpowers had a catalytic effect by creating or
restricting action space for local actors, by intensifying, extending, and inter-
nationalizing existing conflicts, and even by curtailing them and establishing
diplomatic solutions. The conflicts resembled those in Korea, Hungary, Vietnam,
and Afghanistan, in which the Soviet Union or the USA intervened, those in the
Congo and Angola where they aimed to consolidate or develop their positions
through local warring parties—in other words, allowed the latter to fight on their
behalf—or, finally in particular, those in the Latin American states, in which, in
accordance with the Monroe Doctrine, suspicious governments were destabilized
and replaced with regimes which complied with the wishes of the USA.3
A typical but less well known case in this regard is that of Guatemala where the
democratically elected president, Jacobo Arbenz Guzmán, was overthrown in 1954.
Guzmán had come to the attention of the USA through reforms that proposed
inter alia the expropriation of the American United Fruit Company and through
the purchase of weapons in the East bloc. The USA reacted with a typical CIA
counter-insurgency operation called ‘Operation Success’ and replaced the Guzmán
government with a politically palatable military dictatorship. This coup marked the
beginning of a development that culminated in the rule of an increasingly oppres-
sive regime which suspected the country’s Indian majority of general support for a
guerrilla movement that had emerged in response to the political repression and
economic exploitation in the country. The war, which involved the repeated and
targeted massacre of the Maya people, lasted a good three decades (1954–86). The
dictatorial regime only managed to stay in power with the help of intensive US
support. Military aid to the tune of hundreds of millions of dollars was paid and
‘experts’ sent to the country to train and arm the Guatemalan army and police.
By the time the violence had subsided in the late 1980s and a peace agreement
between the army commanders and guerrilla leadership was signed in 1996, over
200,000 people had lost their lives. According to the report compiled by the truth
3 Noam Chomsky, World Orders, Old and New (London: Pluto Press, 1997), 37–44.
commission (Comisión para el Esclarecimiento Histórico) established in 1999, the

military was responsible for ninety-three per cent of the approximately 42,000
deaths investigated. Moreover, genocidal acts were perpetrated against the Maya in
at least four of the country’s regions. The USA bore particular responsibility for
this development. It and a series of private American companies exerted pressure
to ensure the perpetuation of the country’s archaic and unjust socio-economic
order. It is known to have been informed about the massacres and atrocities but did
nothing to prevent them.4
From this perspective, and considering the situation in other Latin American
countries at the time of the cold war, the fact that ‘violence [was] exacerbated by
external factors’, and that action motivated by a simplistic friend or foe philosophy
on the part of the self-appointed democratic force for order in the north led to
something ‘known in Latin America as the National Security Doctrine’, which
actually means ‘not defense against an external enemy, but a way to make the
military establishment the masters of the game’.5 Because the policy adopted by the
Soviet Union at the time in its sphere of influence can be characterized in very
similar terms—all that must be done is substitute ‘communist’ for ‘capitalist’ or
‘imperialist’, ‘social worker’ and ‘trade unionist’ for ‘human rights activist’ and
‘dissident’—it is all the more astonishing that the blocs and their protagonists were
able to reach something approaching a joint policy in the context of the UN.
In 1948, the Security Council decided for the first time to send an observer
mission to monitor the ceasefire between Israel and its neighbours. Even if all that
lay behind this action was a minimum common denominator, an emergency
solution, so to speak (for there was and is no express legal basis for this action in
the Charter), an attempt could be made to stabilize the situation and thus contrib-
ute to the establishment of a political solution through the demonstration of
international presence. Because the conflicting parties had to agree to the mission
and the use of violence on the part of the interveners was only authorized in the
case of self defence,6 this decision was less than a collectively borne—necessarily—
armed attempt at conflict resolution from the outside. However, it was also more
than idle fence-sitting and spectating as a result of the self-blockade in the Security
Council. This was what mattered and was also the reason why other missions, to
become known as ‘blue helmet operations’, were carried out and continue to be
carried out to the present day.
4 William A. Schabas, Genocide in International Law (Cambridge: Cambridge University Press,

2000), 168–9; on the report of the Guatemalan Truth Commission, http://www.juridicas.unam.mx/
publica/librev/rev/boletin/cont/106/art/art6.pdf
5 According to a report of the Colombian Human Rights Committee, quoted by Chomsky, World
Orders, Old and New, 61.
6 Moreover, the intervention troop had to be neutral and its deployment implemented on a
voluntary basis.
586 gerd hankel
Likewise in 1948, on 9 December, the UN General Assembly adopted the

Convention on the Prevention and Punishment of the Crime of Genocide and,
a day later on 10 December, it adopted the Universal Declaration of Human Rights.
The lessons of the Second World War were to be drawn in both the Convention
with its legally binding basis and the non-legally binding Declaration. Based on
their normative requirements, the two instruments were also intended to represent
the starting point and objective of the task set out for the General Assembly in
accordance with Article 13 of the Charter. This task, i.e., ‘promoting international
cooperation’ and the codification of international law, only assumed its true
significance with the enlargement of the UN to include states in the southern
hemisphere. While it had only 51 Member States in 1945, by 1960 this number had
increased to 100 and in 1975 it reached 145, most of which were Asian and African.
The extent to which the different economic interests of the new members and
their demands for the fair distribution of goods and better development opportu-
nities could be instrumentalized again for the purpose of the cold war emerged
when it came to casting the requirements of the Declaration of Human Rights in
legally binding form. The General Assembly only succeeded in passing the two
corresponding covenants, i.e., the International Covenant on Civil and Political
Rights and the International Covenant on Economic, Social and Cultural Rights,
eighteen years after the adoption of the Declaration. And it would take another ten
years for the two human rights covenants to come into force in 1976. A few years
earlier, in a resolution passed in October 1970,7 the General Assembly supported
individual provisions of the UN Charter and interpreted them in the light of past
experience and desirable future developments. In this declaration, it named seven
basic principles, including the duty of cooperation (for example, in the promotion
of human rights and basic freedoms and in the abolition of all forms of racial
discrimination and religious intolerance) and the principle of equal rights and
self-determination of peoples. Finally, in 1984, the General Assembly adopted a
Convention on the elimination of torture, i.e., the Convention against Torture and
Other Cruel, Inhuman or Degrading Treatment or Punishment.
In order to protect against the violation of basic human rights, such as the right
to life and bodily integrity, in addition to their binding prohibition, the General
Assembly was also involved in making a universally observed obligation of the legal
institution of the jus cogens through one of its ancillary bodies, the International
Law Commission.8 No state is to be allowed to violate ‘compelling law’. It protected
values that, in the view of the international community, were so essential that all of
7 Resolution 2625 (XXV) of 24 October 1970 is called the ‘Declaration on Principles of International
Law Concerning Friendly Relations and Co-operation among States in Accordance with the Charter of
the United Nations’.
8 Through the inclusion of a jus cogens regulation in the Vienna Convention on the Law of Treaties
between States (Article 53) of 1969, which came into force in 1980 and enjoys almost universal
acceptance today.
their members were obliged to observe them. This purpose was also served by a
second obligation, the erga omnes obligation.9 According to this principle, legally
accepted rights must be protected ‘towards all’, the idea behind it being that the
normative aggregation of international law could no longer permit the serious
violation of the latter, as otherwise the entire structure of international law would
be permanently devastated, ultimately to the disadvantage of all states. For any
body of law that can be violated without consequence corrodes and clears the way
for the lethal political voluntarism of the states.
All in all, this sounds very positive and the UN, under whose auspices the task of
codification proceeded, clearly made a considerable contribution in this regard.
However, even an optimistic view of these developments cannot ignore the gap that
existed between aspirations and reality in those years. There is no doubt that the
Soviet Union and the USA attempted to impose their own conflict resolution
rather than rely on the public institutions of the UN and other international
institutions. The law of the UN Charter was all too often an ephemeral and
negligible force in the face of the considerations of power politics.
T H E L E G AC Y O F T H E C O L D WA R I N R E L AT I O N
TO T H E P ROT E C T I O N O F H U M A N R I G H T S
................................................................................................................
The cold war is now history. The Soviet Union collapsed and the political-military
bipolarity disappeared along with it. As the only remaining superpower, the USA
stressed the power of multilateralism and spoke of a new world order to be
achieved with the participation of as many states as possible.
As we know—and not just since the second Iraq War—this is not quite how
things have unfolded. However, the outcome of this development is contradictory.
The fact that human rights has penetrated far onto the international political
agenda cannot be ignored. The discourse surrounding human rights intensified
after the end of the East–West conflict, in particular and, indeed, to the extent that
apart from a few almost grotesque exceptions, such as North Korea and Myanmar
(Burma), no state now adopts an anti-human rights position. The era in which the
universal validity of human rights and basic freedom was disputed and their—
entirely extant—Western form was held up against an Asiatic or Arab value system
appeared to be over. Thus, the heads of state and government joyously declared at
9 Developed in 1970 by the International Court of Justice in the Barcelona Traction case, cf. ICJ
Reports 1970, here at 32, No. 33.
588 gerd hankel
the UN Summit in 2005: ‘We reaffirm that all human rights are universal, indivisi-
ble, interrelated, interdependent and mutually reinforcing’.10
Even if its semantic content were stretched to the limit, nobody who reads this
sentence could conceive that a single leader or senior politician exists in the world
who is unable to recite the list of main human rights. The participants of the World
Summit ‘corroborate’ their validity; there is virtually no UN report in which they
are not mentioned; working groups on the most wide-ranging topics refer to them;
numerous UN projects, backed by non-governmental organizations, have been
supporting them expressly in their local activity reports and brochures for years.
Yet the scale of the alternative reality, which can be accessed very quickly by
perusing the annual reports of Amnesty International or Human Rights Watch,
astounds. Amnesty International regularly reports of human rights violations in
dozens of countries, of the most heinous of crimes ranging from murder, kidnap-
ping, and torture to the restriction of freedom of opinion and the freedom of the
press, which are also frequently accompanied by random imprisonment and the
use of physical force. Human Rights Watch also lists numerous countries in which
the worst cases of massive and systematic repression are recorded every year.
All of this takes place, moreover, in the presence of a series of complaint and
notification procedures and sanctions, with which human rights violations can be
censured and even prosecuted under criminal law. Enforcement procedures speci-
fied in the relevant conventions or their associated optional protocols exist. These
include, first, the European Court of Human Rights, the Inter-American Court of
Human Rights, and the African Court on Human and Peoples’ Rights. In addition,
the International Convention on Civil and Political Rights and its First Optional
Protocol institute a Human Rights Committee and the Convention against Torture
led to the establishment of a Committee against Torture (Articles 28 and 17).
Finally, the mechanisms known as 1503 and 1235 procedures based on resolutions
of the UN Economic and Social Council, for which the UN Human Rights Council
is responsible, can also be mentioned here. None of these institutions can, however,
resort to coercive means, their only option being persuasion or diplomatic pres-
sure, which, however, are not easy to exercise. For although the violations are often
fundamental in nature, they are not always part of a large-scale or systematic
process; thus the international attention they attract is smaller in scope or more
fleeting. Therefore, the efficiency of these procedures is largely dependent on the
discernment capacity and willingness to cooperate of the states; it is not just their
detractors who describe them as the ‘world’s biggest wastepaper basket’.11
10 Cf. 2005 World Summit Outcome, http://www.un-ngls.org/un-summit-FINAL-DOC.pdf,

section 121.
11 On this point, see Norman Paech and Gerhard Stuby, Völkerrecht und Machtpolitik in den
internationalen Beziehungen (Hamburg: VSA-Verlag, 2001), 676.
The so-called principle of universal jurisdiction offers a far more promising

approach. What this ultimately amounts to is that no state that violates funda-
mental human rights can invoke its sovereignty, because it does not protect people
who have acted on the mandate of such a state against being held to account by the
legal system of a different country. Such a situation arose with perpetrators of the
Rwandan genocide in Belgium and members of the Argentinean military in Spain.
Although the history of the principle of universal jurisdiction—it began with the
fight against piracy, whose perpetrators were seen as ‘enemies of humanity’ (hostes
humani generis)—shows that serious crimes based on a plan were also mainly
thought of in this case, national law enforcement is often more flexible and quicker
than international jurisdiction. On the other hand, as the examples show,12 it is
more susceptible to opportunistic considerations.
That leaves the International Criminal Court, which has been at work since 1 July
2002 and is responsible for the prosecution of the most serious crimes ‘of international
concern’.13 According to the list provided by the Rome Statute, these crimes are
genocide, crimes against humanity, war crimes, and the crime of aggression (which,
however, remains to be defined). Although the Court has carried out investigations of
four countries (Democratic Republic of the Congo, Darfur/Sudan, Uganda, Central
African Republic) and issued arrest warrants against twelve suspects, the implemen-
tation of both of these measures is proving very difficult. States fear excessive inter-
vention in their internal affairs and powerful states such as the USA, China, and Russia
actually do almost everything in their power to sabotage or ignore the Court.14
Therefore, it is possible to see that, even in the case of extensive human rights
violations, the existing instruments of complaint and sanction are limited in their
effect. Without the willingness to cooperate on the part of the states and their actual
partial renunciation of sovereignty, they run dry. Against this background in partic-
ular, it sends out a fatal signal when democracies that describe themselves as liberal
fabricate highly sophisticated exceptions from basic human rights protection for the
purpose of boosting their fights against the real or supposed enemies of their country
or value system. Memories are triggered of the times when political reality forced all
protestation to the sidelines in defiance of the UN. Power politics superceded
international law and, as part of this process, nothing was subject to closer scrutiny
as long as the direction was right and the political alliance was not at risk.
12 See http://www.hrw.org/press/2003/08/belgium080103.htm; German authorities provided a

particularly shameful example in recent years when they allowed the Uzbekistan interior minister,
who allegedly bore prime responsibility for a massacre resulting in several hundred fatalities carried
out in the Uzbek city of Andischan, to leave the country unchallenged despite the fact that charges had
been filed against him. It may be assumed that the continued use of a Luftwaffe airbase in Uzbekistan
was more important.
13 According to the Preamble to the Rome Statute of the International Criminal Court. Cf. http://
www.icc-cpi.int.
14 See the chapter of Donald Bloxham and Devin Pendas in this volume.
590 gerd hankel
T H E L E G AC Y O F T H E C O L D WA R I N T E R M S
O F C O L L E C T I V E I N T E R NAT I O NA L S E C U R I T Y
................................................................................................................
The understanding of collective security at international level has undergone
enormous change since the end of the cold war. For a long time, Article 39 of the
UN Charter, which states ‘The Security Council shall determine the existence of
any threat to the peace, breach of the peace, or act of aggression’ and decides the
measures to be taken ‘to maintain or restore international peace and security’, was
understood as though it only concerned relations between states and did not
involve situations within states. As a result, in the resolutions on Rhodesia and
South Africa in the 1960s and 1970s, the Security Council was at pains to avoid the
vocabulary of Article 39 and to avoid mobilizing other aggravating factors. The
internal situation of a state only came into the focus of the Security Council with
the end of the East–West conflict when the number of interstate conflicts decreased
and intrastate conflicts increased. This arose in a way reminiscent of the original
perceived threat of interstate conflict. Thus, for example, the resolution of 1991 on
the creation of an exclusion zone in north Iraq still stressed the risk that the large
numbers of fleeing Kurds could represent for the stability of the neighbouring
countries of Turkey and Iran, thus enabling China, which usually reacts extremely
sensitively to sovereignty-based interventions, to abstain.15
This changed in the subsequent resolutions. The risk of transnational impacts
was no longer seen as a necessary justification for sovereignty-based intervention;
irrespective of whether it was caused by a despotic regime or civil war, it was
sufficient that violence and terror prevailed against the population in a country.
The development went even further. The re-classification of human rights viola-
tions or breaches of humanitarian international law to threats to international
peace in the sense of Chapter VII of the Charter was often accompanied by the
decision to adopt measures that should counteract the emergence of conflicts and
thus actually come under Chapter VI of the Charter, which, however, does not
grant any corresponding power of enforcement. As a result, it may be said that the
Security Council extended its authority considerably in recent years and drove back
state sovereignty in the same measure.
In concrete terms, this extension of authority is evidenced, first, in the increas-
ing imposition of peaceful sanctions against states in accordance with Article 41
of the Charter as, in this way, targeted pressure can be exerted against both
groups and individuals. The catalogue of measures imposed here ranges from
the imposition of travel bans, the seizure of assets deposited abroad, and the
imposition of economic sanctions—the effect of which is, however, highly
15 See the Resolution of the UN Security Council S/RES/0688 (1991) of 5 April 1991.
questionable16—against aggressive regimes that are hostile to human rights to the

establishment of criminal courts. Second, it is also evidenced by the increasing
debate surrounding the threat and use of military force in accordance with Article
42 of the UN Charter, i.e., surrounding the risking of human life and—what often
appears to assume greater importance—the provision of considerable financial
resources. The fact that the UN peace missions have changed is reflected in this
debate. Blue helmets are no longer blue helmets. The deployments of international
contingents of soldiers, who are strictly neutral and armed only for self-defence,
with the agreement of the conflicting parties were followed by a series of additional
missions which differed considerably in terms of their purpose and scope. Today,
these are divided into four generations or categories. The first two represent peace-
keeping measures while third and fourth generation blue helmet deployments also
incorporate peace-making measures, which do not exclude the use of military force
for purposes other than self-defence (so-called ‘robust peace-keeping’) and can also
involve the assumption of executive tasks. The consensus of the conflicting parties
is not essential; crucial alone is the objective of the operation as formulated in the
Security Council resolution. In most cases, e.g., the mission in the Democratic
Republic of the Congo, it is specified with a general reference to Chapter VII of the
UN Charter (restoration of world peace and security) and substantiated with
intermediary objectives to be achieved using ‘all necessary means’.17 The example
of the UN deployment in the Congo also shows, however, that the adoption of a
Security Council resolution is one thing and its implementation another. The UN
does not have its own armed forces, although this was actually planned according to
the Charter (Article 43 refers to special agreements in which the states make armed
forces available to the United Nations). It depends on its member states’ willingness
to make troops available to it. This is not easy as the latter’s interests are often not
driven by humanitarian urgency but by a logic of their own, in which humanitarian
concern is merely one factor among many. The Permanent Members of the Security
Council constitute no exception here—in contrast. Out of the 34,000 blue helmet
soldiers originally requested by the UN Secretary-General, around 7,000 are
authorized by the Security Council, of which only 4,000 are gradually sent to the
war zone by the member states.18
16 For example, Iraq was subject to trade sanctions for over ten years. The regime itself was not
bothered by them; the victims were the many Iraqis, above all Iraqi children, who showed increasing
signs of malnutrition and illness and died in their thousands due to the unavailability of suitable
medicines. Cf. Ramsey Clark, ‘Letter to the Security Council’, in Adam Jones (ed.), Genocide, War
Crimes and The West: History and Complicity (London/New York: ZED Books, 2004), 273–5.
17 Cf. S/Res/1565 (2004), 1.10.2004 or, more recently, S/Res/1807 (2008), 31 March 2008.
18 According to the former German General and Assistant UN Secretary-General Manfred Eisele in
relation to the example of the protection zones in Bosnia-Herzegovina. Cf. idem, ‘Blauhelme als
Krisenmanager’, in Sabine von Schorlemer (ed.), Praxishandbuch UNO. Die Vereinten Nationen im
Lichte globaler Herausforderungen (Berlin/Heidelberg: Springer, 2003), 34.
592 gerd hankel
How far the disinterest of individual states, political-diplomatic cowardice,

or simply excessive need can go was demonstrated in April to June 1994 in Rwanda
and in July 1995 in Srebrenica. Not only was the United Nations Assistance Mission
for Rwanda not reinforced, soldiers were actually withdrawn from the mission in
response to pressure from the sending states. A few hundred soldiers were sup-
posed to be able to contribute to the ‘immediate cessation of hostilities’19 in a state
that was in the process of descending into a delirium of violence. Despite all the
evidence, the UN Security Council was still avoiding any reference to genocide a
good three weeks after the beginning of the massacre as this would have obliged the
states to intervene in accordance with Article I and VIII of the Convention on the
Prevention and Punishment of the Crime of Genocide.20 Operation Turquoise,
which finally took place under French command and by UN mandate,21 came too
late (it began on 23 June). Moreover, it was tainted from the outset by the suspicion
that France’s involvement was not motivated purely by humanitarian concerns but
also by its desire to salvage its African policy in the region.
When the international community failed again a year later in Srebrenica and
did nothing to prevent the genocide of the Bosnian Muslims who had fled to the
UN-designated ‘safe zone’, demands for a general admissibility of humanitarian
intervention increased, i.e., including that implemented by a state or alliance of
states without the authorization of the Security Council, based on the justifica-
tion that speed is of the essence in such emergency situations. The discussion
surrounding this interpretation of humanitarian intervention is not new. It took
place as far back as the nineteenth century when Europe’s great powers inter-
vened in areas of the Ottoman Empire, claiming to want to assist persecuted
Christians there.22 This approach was also successful; at least in part, however, the
interventions carried out using military means were volatile in terms of power
politics and therefore usually regarded with scepticism or openly condemned by
states not involved in the action. For this reason, in the League of Nations era,
such intervention was supposed to be rendered impossible by a system estab-
lished for the protection of minorities and the collective security provisions of the
Covenant, albeit with limited success which, again, was the reason for the stricter
regulation applied in the UN Charter, which, with the exception of self-defence
(Article 51 of the UN Charter), grants the right of intervention exclusively to the
Security Council. This left no scope for the admissibility of humanitarian inter-
vention in the sense referred to above. Thus, in the past, states that took military
19 See, for example, UN Security Council Resolution 912 of 21 April 1994.

20 For the discussion of this obligation, see Schabas, Genocide in International Law, 447–52;
Samatha Power, ‘A Problem from Hell’: America and the Age of Genocide (London: Flamingo, 2003),
358–64.
21 See UN Security Council Resolution 929 of 22 June 1994.
22 See Donald Bloxham, The Great Game of Genocide: Imperialism, Nationalism, and the
Destruction of the Ottoman Armenians (Oxford: Oxford University Press, 2005), 31–8.
action against other states—Vietnam against Pol Pot’s Cambodia in 1978, Tanza-
nia against Idi Amin’s Uganda in 1979— did not refer to the appalling human
rights situation in these countries in justification of their action but gave prece-
dence to territorial conflicts (Vietnam) and the right of reprisal as retribution for
illegal attacks (Tanzania). The same applies to East Pakistan (Bangladesh) where
India justified its intervention in 1971 with the desire to alleviate the extensive
suffering endured by the civil population as a result of the war with West Pakistan
(Pakistan), although strategic reasons (military weakening of West Pakistan) also
played an important role in its decision to intervene. Military action was not
recognized as legal in any of these cases. As the International Court of Justice
noted in its ruling on Nicaragua in 1986: ‘In any event, while the United States
might form its own appraisal of the situation as to the respect for human rights in
Nicaragua, the use of force could not be the appropriate method to monitor or
ensure such respect.’23
This was the legal situation when NATO started to bomb Serbia in the spring of
1999 without a corresponding Security Council resolution. Due to the certain
imposition of the Russian veto, such a resolution could not be expected to be
forthcoming and the intervention was self-mandated on the basis of the humani-
tarian aim of saving the Kosovo Albanians from expulsion and annihilation.
The threat later emerged as having been highly exaggerated and part of a strategy
to make the war along with its one-sided support unavoidable and to achieve
independence for Kosovo through the military and economic weakening of Serbia.
However, the law of the Charter had been contravened and the UN had been
manoeuvred into a position of inability to preserve the peace through carefully
orchestrated pressure and the forceful assistance of interested states. In response, it
did with due professionalism what tends to be done in such situations involving
confrontation with failure: it commissioned an inventory to obtain information on
the status of the UN peace missions, the ways in which they need to be improved,
and, with express reference to the controversy surrounding the war in Kosovo, how
the law of humanitarian intervention should be shaped using the existing instru-
ments of international law.
The first report was published in summer 2000.24 It clearly lists the weaknesses
of previous and ongoing blue-helmet missions, first and foremost their inadequate
personnel and financial resources, their frequently erroneous analysis and convolut-
ed chain of command, and, finally, the unclear question of what would happen
‘afterwards’, i.e., the consolidation of the ‘secured’ or ‘built’ peace. The second report
23 Cf. ICJ Reports 1986, p. 14–150 (p. 134). As stated in Article 59 of its Statute, the rulings of the ICJ
are only binding on the parties involved and in respect of the particular case in question; however, its
decisions and non-binding advisory opinions are viewed as highly significant statements on existing
(or non-existing) law, in particular customary international law.
24 Report of the Panel on United Nations Peace Operations (Brahimi-Report), A/55/305-S/2000/809.
594 gerd hankel
appeared around a year later.25 Entitled The Responsibility to Protect, it stresses, first,
the role of state sovereignty as a factor in the international order and thus as a
cornerstone of the UN Charter. It then lists criteria that should facilitate interna-
tional reaction to severe and systematic human rights violations in a country—the
examples provided include ethnic cleansing, genocide, and other forms of mass
murder.
The first of the four criteria is the ‘right intention’. This demands that the main
objective of intervention must consist in the termination of human suffering.
According to the second criterion, the intervention must be the ‘last resort’:
sufficient reasons must support the assumption that other non-military measures
have no prospect of success. Third, the military action must not exceed the scope
required to re-establish human safety; thus actions must be based on ‘propor-
tional means’. Moreover, they must, fourth, support sufficient expectation that
the consequences of the intervention will not be worse than non-intervention,
i.e., ‘reasonable prospects’.
The report stated that the Security Council should remain responsible for
decisions concerning intervention. To ensure that its decision-making capacity
is not crippled by the right of veto, the authors proposed a code of conduct, on
which the five Permanent Members of the Security Council were to agree. The most
important point in this code of conduct was that none of the permanent members
would be allowed to adopt an anti-humanitarian position in a situation that
directly involved the self-conception of humanity (as expressed in recognition of
a minimum human rights standard). The use of the veto should not be admissible
in this case. An exception would only be made—and this is where the concession to
realpolitik comes into play—for cases in which a Permanent Member would see its
own vital concerns as coming under threat if it refrained from using the veto.26
As already mentioned, the report was compiled in 2001. Two events then
occurred that brought this expertly formulated conception of international collec-
tive responsibility back down to earth with a bang. First, the Iraq war, which was
justified by the USA and the ‘coalition of the willing’ with the supposedly imma-
nent threat to peace posed by the Iraqi regime and other factors, including the
intolerable human rights situation in the country, began in 2003. A Security
Council resolution empowering the USA and its coalition partners to undertake
this action did not exist. Second, in early 2006, the UN Secretary-General’s Special
25 The Responsibility to Protect (Report of the International Commission on Intervention and State
Sovereignty, December 2001). The report is based on a Canadian initiative which, however, saw itself
as a contribution to the inventory desired by the UN.
26 Section 6.21 of the report refers to ‘constructive abstention’. If this cannot be reached, the
Member States would be called on, as already occurred in 1950 in the case of the Uniting for Peace
Resolution, to provide recommendations for the taking of action and thus increase the pressure on the
Security Council. And if this also fails, regional organizations, which, according to Chapter VIII of the
Charter, are responsible for the settlement of ‘local disputes’. Cf. ibid. sections 6.28–6.35.
Envoy for Sudan finally reported that the situation in the province of Darfur
remained critical. The UN’s sanctions, which were far too weak anyway, had not
even been implemented. The powers that be in Khartoum were making fun of the
UN while the expulsions and mass slaughter continued.27
Thus, here it was again, the gap between aspiration and reality. On the one hand,
proposals for a more efficient peace security system are formulated and acknowl-
edged and supported by the UN bodies, and, on the other hand, reasons or, to be
more accurate, pretexts are found which reveal the responsibility of the states
arising from the UN Charter to be nothing more than rhetoric. Intervention is
instigated when it reflects economic and power-state interests; otherwise, it is not.
In the latter case, the crimes perpetrated are merely, and only occasionally, deemed
‘genocidal acts’ and not crimes of genocide which, as everybody knows, make the
call for intervention impossible to ignore.28 However, the states appear to be able to
live with ‘genocidal acts’, as though, depending on the label attached to the murder
of some 300,000 people, the crime assumes a different dimension and the other-
wise much invoked criterion of collective responsibility or security evaporates.
C O N C LU S I O N : T H E D UA L N AT U R E
O F T H E C O L D WA R L E G AC Y
................................................................................................................
It could be said that no progress worth mentioning has been made in international
law since the end of the cold war. It remains contradictory and subordinate to state
interests, in particular where the prevention and punishment of mass crimes is
concerned.
However, this is just one side of the story. A look back at the past can also reveal
the progress made despite all of the surviving structures. Admittedly they are more
theoretical in nature and, as repeatedly established here, they form the basis of
claims not borne out by reality. But they are in the world, they exist, and there is
hope that, even in practice, if promoted by international public opinion and the
impatience of a globalized world, they will emerge from the long shadow of the
past—albeit without, at the same time, in the name of an allegedly good central
principle, unilaterally doing away with the Charter of the United Nations.
27 ‘Sudan: UN envoy says Security Council must enforce sanctions’, available at: http://www.un.
org/apps/news/story.asp?NewsID=17158&Cr=sudan&Crl=. Kofi Annan had published a report as far
back as December 2004 entitled ‘A More Secure World: Our Shared Responsibility’ (A/59/565) and
repeated almost word for word the list of criteria contained in the 2001 report.
28 The Report of the International Commission of Inquiry on Darfur to the United Nations
Secretary-General, 25.1.2005, section 518–20.
596 gerd hankel
Cassese, Antonio, International Law in a Divided World (Oxford: Clarendon Press, 1992).
Dallaire, Roméo, Shake Hands with the Devil: The Failure of Humanity in Rwanda (Toronto:
Random House Canada, 2003).
Gaddis, John Lewis, The Cold War: A New History (London: Penguin, 2006).
Goldsmith, Jack L., and Eric A. Posner, The Limits of International Law (Oxford: Oxford
Jones, Adam (ed.), Genocide, War Crimes and the West: History and Complicity (London/
New York: Zed Books, 2004).
Schabas, William A., Genocide in International Law (Cambridge: Cambridge University
Press, 2000).
Weiss, Thomas C., and Sam Daws (eds), The Oxford Handbook on the United Nations
(Oxford: Oxford University Press, 2008).
chapter 29
.............................................................................................
M I L I TA RY
I N T E RV E N T I O N
.............................................................................................
alex j. bellamy
The commission of genocide and mass atrocities has provoked calls for military
intervention by external actors since the nineteenth century.1 From the 1820s to the
1870s, groups of activists collectively known as ‘atrocitarians’ agitated for armed
intervention to protect Christians in the Greek, Syrian, and Bulgarian lands of the
Ottoman Empire. At century’s end, the USA invaded Cuba partly in response to calls
from notable figures, including former President Theodore Roosevelt, that it should
act to put an end to Spanish atrocities there.2 In the twentieth century, the Armenian
genocide, Holocaust, and more recent genocides in Bosnia and Rwanda all elicited
widespread agitation in favour of armed intervention. The present century is no
different, with activists in various parts of the world maintaining that those with the
power to do so should intervene in Darfur to protect the civilian population there
from its tormentors. In 2005, two centuries of political agitation produced a land-
mark commitment from the world’s governments when they unanimously declared
their responsibility to protect populations from genocide and to take ‘timely and
decisive’ action in cases where a state manifestly fails in its responsibility to protect.3
Historically, once started, genocides tend to end with either the military defeat of
the perpetrators or the suppression (though not necessarily the annihilation) of the
1 See Gary J. Bass, Freedom’s Battle: The Origins of Humanitarian Intervention (London: Knopf, 2008).
2 Ernest R. May, Imperial Democracy: The Emergence of America as a Great Power (New York:
Imprint, 1961), 127.
3 For accounts of the emergence of the Responsibility to Protect principle see Gareth Evans, The
Responsibility to Protect: Ending Mass Atrocity Crimes Once and for All (Washington, DC: Brookings
Institute, 2008), and Alex J. Bellamy, Responsibility to Protect: The Global Effort to End Mass Atrocities
(Cambridge: Polity Press, 2009).
598 alex j. bellamy
victim groups.4 Only military force can directly prevent genocidal killing, stand
between perpetrators and their intended victims, and protect the delivery of life-
saving aid. But its use entails risks for all parties and does not necessarily resolve the
underlying conflict. Its impact is difficult to predict and force might sometimes
inflame rather than improve situations. Properly used, force can offer physical
protection to populations in immediate danger. But it cannot compel the parties
to build sustainable peace; rebuild shattered governments, economies, and societies;
protect populations in the long term; or provide comprehensive security. Moreover,
there is a real danger that a generalized right to intervene for humanitarian purposes
could be abused, such as Russia’s use of the ‘Responsibility to Protect’ as justification
for its 2008 invasion of Georgia, and the use of humanitarian rhetoric to justify the
2003 invasion of Iraq.5 Then there is the additional hazard that promises of interven-
tion might encourage acts of rebellion that provoke genocidal responses.6
What is more, by the time external military forces can be deployed and the agents of
genocide defeated, the death toll amongst the victim group is likely to be staggeringly
high. Although his analysis may be considered overly pessimistic, Alan Kuperman’s
sober assessment of what intervention could have achieved in Rwanda provides a
cautionary tale for would-be humanitarian warriors. Kuperman maintained that, had
the USA speedily and successfully deployed forces to Rwanda in 1994 once it became
known that genocide was under way, the total number of lives saved would have been
around ‘only’ 125,000 of the approximately 500,000–800,000 victims of the genocide.7
In short, military intervention does not address why genocides happen in the
first place and provides only a short-term palliative at best. Finally, it is important
that we not allow a preoccupation with intervention to obscure the manner in
which hegemomic powers, conceptions of statehood, and neo-liberal economics
sustain the preconditions for genocide—a point ably demonstrated by Mark
Levene in this volume.8
4 The question of how genocides end remains relatively understudied. In 2006, the Social Science
Research Council organized an insightful forum on this question. Available at http://
howgenocidesend.ssrc.org/ (accessed 10 December 2008).
5 I have detailed the misuse of humanitarian arguments in the Georgia case in Alex J. Bellamy
‘Humanitarian Intervention’, in Alan Collins (ed.), Security Studies (Oxford: Oxford University Press,
forthcoming) and in the Iraq case in idem, ‘Ethics and Intervention: The “Humanitarian Exception”,
and the Problem of Abuse in the Case of Iraq’, Journal of Peace Research 41:2 (2004), 131–47.
6 A ‘moral hazard’ documented by Alan J. Kuperman, ‘Humanitarian Hazard: Revising the Doctrine
of Intervention’, Harvard International Review 26:1 (2004), 64–8, and idem, ‘The Moral Hazard of
Humanitarian Intervention: Lessons from the Balkans’, International Studies Quarterly 52:1 (2008), 49–80.
7 The precise death toll of the genocide is contested and estimates vary according to the data sources,
time scales, and method of counting used. Although some analysts—Kuperman included—estimate the
death toll to be significantly lower than 800,000, most studies put the toll somewhere around that figure.
See Linda Melvern, Conspiracy to Murder: The Rwandan Genocide (London: Verso, 2006).
8 Also see Mark Levene, ‘A Dissenting Voice: Or, How Current Assumptions of Deterring and
Preventing Genocide May be Looking at the Problem through the Wrong End of the Telescope, Part I’,
military intervention 599
The purpose of this chapter is to examine the role that military intervention can
play in ending genocide and the political, moral, and legal debates that surround it.
It proceeds in three parts. The first section briefly examines how genocides have
ended since the beginning of the twentieth century, and explores the place of
military intervention by external powers. The second section examines whether
there is a moral and/or legal duty to intervene to end genocide. In the third section,
I consider the reasons why states intervene only infrequently to put an end to
genocide despite their rhetorical commitments.
I N T E RV E N T I O N AND THE ENDING OF GENOCIDE

................................................................................................................
To what extent has military intervention brought genocide to an end? Table 1 sets
out—in necessarily rudimentary form—how some of the most commonly ac-
cepted cases of genocide since 1900 have come to an end and reveals three
important insights. First, with only two partial exceptions, once begun, genocidal
killing ends in only one of two ways—by perpetrators deciding that they have
achieved their objectives or with their military defeat. The partial exceptions are
the Nuba Mountains and Bosnia cases. In the Nuba Mountains case, local
resistance slowed the pace of killing and forced relocation, and divisions within
the government brought about its end.9 In the Bosnia case, a political settlement
(the ‘Dayton Accords’) rather than the military defeat of the Bosnian Serbs ended
the violence, although the Bosnian Serb leadership was coerced into accepting the
accords by a combination of NATO air strikes and ground attacks (Operation
‘Deliberate Force’) and, more importantly, military advances by the Bosnian–Croat
alliance forged and armed by the USA. The single largest act of genocide—the
1995 killing of 7,600 men and boys sheltering in Srebrenica—succeeded in its
immediate aim of eradicating the Muslim males of that town.10 To this day,
Srebrenica remains an exclusively Serb town, an ethnic reality forged by genocide.
Journal of Genocide Research 6:2 (2004), 156. I have argued elsewhere that armed intervention needs to be
seen as one part of a pattern of interaction between powerful and less powerful states and that much of
that interaction contributes to the causes of genocide and mass killing in the first place. A first step to
dealing with this would be to identify those patterns of interaction and encourage powerful states to
engage first in the ‘do no harm’ principle. See Alex J. Bellamy, ‘Humanitarian Responsibilities and
Interventionist Claims in International Society’, Review of International Studies 29:3 (2003), 320–41; and
idem, ‘Conflict Prevention and the Responsibility to Protect’, Global Governance 14:2 (2008), 135–57.
9 De Waal and Conley-Zilkic, ‘Reflections’, 5.
10 Jan Willem Honig and Norbert Both, Srebrenica: Record of a War Crime (London: Penguin, 1996).
Table 1 Commonly Accepted Genocides since 1900 and How They Ended11
Date Where Ending
1904–5 German South West Africa (killing of Decimation and ‘pacification’ of

Herero) targeted group
1915–18 Ottoman Empire (forced deportation Largely successful ‘cleansing’ of
and killing of Armenians) Anatolia. Ottoman defeat in the
First World War
1935–9 Ethiopia (Italy annihilation of Decimation and pacification of
Ethiopians as reprisal—envisaged targeted groups
‘Ethiopia without Ethiopians’)
1937–9 USSR (great purge of kulaks and Decimation and pacification of
others) targeted groups
1937–45 East Asia (Japanese destruction of Defeat of Japanese in the Second
Chinese in Manchuria, ‘rape of World War
Nanking’, and genocidal atrocities
in the Philippines and elsewhere)
1941–5 Nazi-occupied Europe (Jewish Defeat of Nazis in the Second World
Holocaust) War
1965–6 Indonesia (massacre of suspected Suppression of ethnic Chinese and
communists) elimination of communists as a
political force
1967–70 Nigeria (Biafra) (genocidal killing in Suppression of Biafran rebels and
support of Federal forces in civil war) government victory in civil war
1971 Bangladesh (‘cleansing’ of Hindus by Intervention by India
West Pakistan government)
1972 Burundi (elimination of Hutus) Suppression of Hutus, especially
educated class.
1975–9 Cambodia (Khmer Rouge) Intervention by Vietnam
1981–3 Guatemala (killing of ‘communists’ Decimation and pacification of
especially targeting five Maya groups) targeted group
1992 Sudan (Nuba mountains) (Jihad, mass Local resistance and disagreement
killing, and forced relocation of within government of Sudan
Nuba people with genocidal intent)
1992–5 Bosnia ((i) general—mass killing of (i) Peace settlement coerced by
Bosnian Muslims mainly by Bosnian combination of NATO and Croat–
Serbs but also by Bosnian Croats; (ii) Muslim forces; (ii) male population
genocide at Srebrenica—killing of largely exterminated
males by Bosnian Serbs)
1994 Rwanda (killing of Tutsi and Hutu Defeat of militia and government
moderates by Hutu militia forces by Rwandan Patriotic Front
supported by government) (RPF)
2003– Sudan—Darfur (killing and forced Decimation and forced removal of
displacement of various African targeted group lead to reduction
groups by government backed of violence
militias)
11 Based on information in de Waal and Conley-Zilkic, ‘Reflections’; Adam Jones, Genocide: A

Comprehensive Introduction (London: Routledge, 2006); Ben Kiernan, Blood and Soil: A World History
of Genocide and Extermination from Sparta to Darfur (New Haven, CT: Yale University Press, 2008).
However appealing, non-military measures such as economic and political

sanctions, arms embargoes, inducements, ‘naming and shaming’, diplomacy, and
threat of legal punishment have not, historically, sufficed to bring genocide to
an end.
Despite the outpouring of academic literature on the subject and much political
angst at the UN and elsewhere, external intervention to end genocide remains the
exception rather than the norm. Of the fifteen cases identified in Table 1, seven ended
more or less successfully for the perpetrators and a further two (Armenia and
Bosnia-Srebrenica) ended badly for the perpetrators but not before they had
achieved core goals (‘cleansing’ of Anatolia and Srebrenica). Of the genocides that
did not end on the perpetrators’ own terms, three endings were related to their
military defeat in campaigns not directly related to the genocides (the defeat of the
Ottoman Empire, Germany, and Japan in the First and Second World Wars), one was
brought to an end by the perpetrators’ defeat by local actors (Rwanda), and another
by a combination of local and external actors (Bosnia). Across all the cases, local
resistance played a significant part in ending three episodes (Nuba, Rwanda, Bosnia).
Only two episodes were ended by external military intervention specifically aimed at
defeating the perpetrators of genocide (Bangladesh and Cambodia). Somewhat
counter-intuitively, both of these interventions were conducted unilaterally by
post-colonial (not Western) states not primarily motivated by humanitarian con-
cerns.12 Although the West is commonly identified as the principal advocate of
humanitarian intervention, there is scant evidence of Western military activism to
end genocide since the start of the twentieth century. In short, there has been much
less actual external military intervention to end genocide than there has been talk
about such intervention.
Finally, although the broader literature on humanitarian intervention often
suggests that interventions are not tantamount to war and involve a variety of
different military tasks—sometimes labelled ‘military expedients’, such as protect-
ing safe areas, humanitarian convoys, no-fly zones, etc.13—it seems clear that, on
the whole, external intervention to end genocide requires the military defeat of the
perpetrators. In other words, whatever may be required by other forms of external
engagement in armed conflict, such as peace operations, military intervention to
end genocide is identical in form to warfare and has the same objective, namely the
military suppression of the enemy.
12 As Wheeler points out, both interventions were primarily conceived and justified in terms of
self-defence. See Nicholas J. Wheeler, Saving Strangers: Humanitarian Intervention in International
Society (Oxford: Oxford University Press, 2000), 55–110.
13 John G. Heidenrich, How to Prevent Genocide: A Guide for Policymakers, Scholars and Concerned
Citizens (Westport, CT: Praeger, 2001), 163. Also see Bass, Freedom’s Battle.
602 alex j. bellamy
A DUTY TO I N T E RV E N E ?
................................................................................................................
Is there a duty to intervene to ‘save strangers’ from genocide?14 I argue that there is
a clear moral duty to intervene in circumstances where intervention is thought
likely to do more good than harm. There is also evidence of an emerging—but
much less well-established—legal duty. Both ideas, however, remain deeply con-
troversial principally because armed intervention by external actors is rarely
disinterested, leading some to fear that the duty to intervene can be a thinly veiled
justification for a coercive form of Western hegemony or neo-imperialism that
supports the very global system that nourishes the preconditions for genocide.
Usually associated with liberalism, cosmopolitanism, and the Christian Just War
tradition, the case for intervention is typically premised on the idea that external
actors have a duty as well as a right to intervene to halt genocide. For advocates of
this position, all humans have certain fundamental rights—chief among them
being the right not to be arbitrarily killed—and sovereign rights are conditional on
the fulfilment of the state’s responsibility to protect populations under its
care. When states fail in their duties towards their citizens, they lose their right
to non-interference.15 There are a variety of ways of arriving at this conclusion.
Some liberal cosmopolitans draw on the work of the German philosopher Imman-
uel Kant to insist that all individuals have certain fundamental rights that deserve
protection.16 Other advocates of the Just War tradition ground their arguments in
Christian theology.17 Still others argue that today’s globalized world is so
integrated that massive human rights violations in one part of the world have an
effect on every other part. This social interconnectedness, they maintain, creates
moral obligations.18 One of the leading contemporary proponents of this view is
former British Prime Minister Tony Blair. Shortly after NATO began its 1999
intervention in Kosovo, he gave a landmark speech setting out his ‘doctrine of
the international community’. Blair maintained that globalization was changing
the world in ways that rendered traditional views of sovereignty anachronistic.19
14 To use the phrase coined by Nicholas J. Wheeler.

15 Fernando R. Tesón, ‘The Liberal Case for Humanitarian Intervention’, in J. L. Holzgrefe and
Robert O. Keohane (eds), Humanitarian Intervention: Ethical, Legal and Political Dilemmas
(Cambridge: Cambridge University Press, 2003), 93. The principle of non-interference is discussed
in more detail below.
16 Simon Caney, ‘Human Rights and the Rights of States: Terry Nardin on Non-Intervention’,
International Political Science Review 18:1 (1997), 34.
17 Paul Ramsey, The Just War: Force and Political Responsibility (Lanham, MD: Rowman and
Littlefield, 2002), 20.
18 The idea that interconnectedness creates moral responsibilities is eloquently set out by Thomas
Pogge. See Thomas Pogge, World Poverty and Human Rights, 2nd edn (Cambridge: Polity Press, 2008).
19 Tony Blair, ‘Doctrine of the International Community’, speech to the Economic Club of Chicago,
Hilton Hotel, Chicago, 22 April 1999.
A further line of argument is to point to the fact that states have already agreed
to certain minimum standards of behaviour and that military intervention to end
genocide is not about imposing the will of a few Western states upon the many,
but about protecting and enforcing the collective will of international society.
Advocates of this position argue that there is a customary right (but not duty) of
intervention to put an end to genocide and mass atrocities.20 They maintain that
there is agreement in international society that genocide constitutes a grave wrong
warranting external intervention.21 From this perspective, state practice since the
end of the cold war suggests the emergence of a customary right of humanitarian
intervention.22 In particular, they point to the justifications offered to defend the
American, French, and British-led intervention in Northern Iraq in 1991 to support
their case. In that instance, the British argued that they were upholding customary
international law, France invoked a customary ‘right’ of intervention, and the USA
noted a ‘re-balancing of the claims of sovereignty and those of extreme humani-
tarian need’.23
According to this perspective, the movement towards acceptance of a customary
right of humanitarian intervention was reinforced by state practice after Northern
Iraq. Thus throughout the UN Security Council’s deliberations about how to
respond to Rwanda in 1994, no state argued that either the ban on the use of
force (Article 2[4] of the UN Charter) or the non-interference rule (Article 2[7] of
the Charter) should prohibit armed intervention to halt the bloodshed, suggesting
that such intervention would have been legitimate in that case. What stood in the
way of intervention in Rwanda was the fact that no government wanted to risk
the lives of its own soldiers to save Africans (see below). Throughout the 1990s, the
Security Council expanded its interpretation of ‘international peace and security’
and authorized interventions to protect civilians in safe areas (Bosnia), maintain
law and order and protect aid supplies (Somalia), and restore an elected govern-
ment toppled by a coup (Haiti). These cases prompted Thomas Weiss to argue that
‘the notion that human beings matter more than sovereignty radiated brightly,
albeit briefly, across the international political horizon of the 1990s.’24 Progress did
not stop at the turn of the century. Since 2000, the Security Council has mandated
peacekeepers to protect civilians under threat in the Democratic Republic of
20 Wheeler, Saving Strangers, 14.

21 See A. C. Arend and R. J. Beck, International Law and the Use of Force: Beyond the UN Charter
Paradigm (London: Routledge, 1993); Jack Donnelly, International Human Rights, 2nd edn (Boulder,
CO: Westview Press, 1988); and Tesón, ‘The Liberal Case’.
22 See Wheeler, Saving Strangers, and Martha Finnemore, The Purpose of Intervention: Changing
Beliefs about the Use of Force (Ithaca, NY: Cornell University Press, 2003).
23 Adam Roberts, ‘Humanitarian War: Military Intervention and Human Rights’, International
Affairs 69:3 (1993), 436–7.
24 Weiss, ‘The Sunset of Humanitarian Intervention’, 135.
604 alex j. bellamy
Congo, Burundi, Côte d’Ivoire, Liberia, and Darfur though it has usually insisted
on receiving the consent of the host government.25
All this suggests that there is a growing international consensus around a moral
duty to intervene to put an end to massive human suffering. However, there are
also grounds for thinking that there is an emerging legal responsibility to do so as
well, derived from a combination of the Responsibility to Protect principle and a
recent ruling of the International Court of Justice (ICJ). Although it might be
premature to speak of a specific legal obligation to intervene militarily, the emer-
gence of such a duty is important because it places legal obligations on states and
creates the potential for redress if those obligations are not satisfied.
In 2000 the Canadian government created the International Commission on
Intervention and State Sovereignty (ICISS) to develop a way of reconciling sover-
eignty and human rights. The Commission’s report, released in late 2001, was
premised on the notion that the principle of non-interference ‘yields to the
responsibility to protect’ when states are unwilling or unable to protect their
citizens from grave harm.26 The ICISS argued that the Responsibility to Protect
entailed responsibilities to prevent and react to massive human suffering and help
rebuild states and societies afterwards. Of the three responsibilities, the Commis-
sion identified the ‘responsibility to prevent’ as the single most important. In
relation to the use of force for humanitarian purposes, the Commission proposed
the adoption of criteria to guide decision-makers. Drawing from the Just War
tradition, the Commission’s proposed criteria included ‘just cause thresholds’
(large-scale loss of life or ethnic cleansing, actual or apprehended) and ‘precau-
tionary principles’ (right intention, last resort, proportional means, and reasonable
prospects).27
In 2005, the Responsibility to Protect was transformed from a concept advanced
by a Commission of high-profile figures to an international principle unanimously
endorsed by world leaders. At the 2005 World Summit summoned to consider a
UN reform package, world leaders adopted a declaration affirming the Responsi-
bility to Protect, which was subsequently reaffirmed by the UN Security Council in
2006. According to the UN Secretary-General Ban Ki-moon, the Responsibility to
Protect principle agreed to by world leaders rests on three pillars:
25 See Alex J. Bellamy and Paul D. Williams, Understanding Peacekeeping, 2nd edn (Cambridge:
Polity Press, 2009), esp. chs. 5 and 14; Victoria K. Holt and Tobias C. Berkman, The Impossible
Mandate? Military Preparedness, the Responsibility to Protect and Modern Peace Operations
(Washington, DC: Henry L. Stimson Centre, 2006); and Lise Morje Howard, UN Peacekeeping in
Civil Wars (Cambridge: Cambridge University Press, 2008).
26 International Commission on Intervention and State Sovereignty (ICISS), The Responsibility to
Protect (Ottawa: IDRC, 2001), p. xi.
27 Ibid.
1) The responsibility of the state to protect its own populations from genocide,
war crimes, ethnic cleansing, and crimes against humanity;
2) The international community’s duty to assist states in meeting these obligations;
and
3) The international community’s responsibility to respond in a timely and
decisive manner when a state is manifestly failing to protect its population,
using Chapters VI (peaceful means), VII (coercive means authorized by the UN
Security Council), and VIII (regional arrangements) of the UN Charter.28
Although much more work needs to be done to translate the Responsibility to
Protect from words to deeds, the 2005 declaration is important because it is a
politically potent principle based on unanimous consensus produced by one of the
largest gatherings of heads of state ever seen.29 It is important to stress the
unanimity of consensus among Member States. Indeed, two traditional sceptics
about this line of thinking, China and Russia, actually reaffirmed their commit-
ment to the Responsibility to Protect principle in Security Council Resolution 1674
(2006).30 Moreover, by referring to Chapter VII of the UN Charter, world leaders
specifically recognized the need to consider the use of force to protect populations
from genocide. This of course does not mean that governments will agree about the
most appropriate and effective form of engagement with specific crises, but the
principle helps define the parameters and at least means that they can no longer
avoid public consideration of engagement.
It is well known that the 1948 Genocide Convention not only prohibits genocide
but establishes a duty to prevent and punish it (Article 1). This duty to prevent
genocide is widely understood to be a principle of customary international law.31
However, it was never clear whether this general duty amounted to a legal obliga-
tion to intervene to halt specific genocides. Although some legal scholars have
argued that Article I imposes a duty on the UN and its member states to act
(including through military intervention) wherever a genocide takes place,32 the
majority view seemed to support US Secretary of State Colin Powell’s assertion that
the Convention’s language did not impose such a wide-ranging obligation.33
28 Ban Ki-moon, ‘Responsible Sovereigns’, address of the Secretary-General, SG/SM/11701, 15 July

2008; and Edward C. Luck, ‘The United Nations and the Responsibility to Protect’; Stanley Foundation
Policy Analysis Brief, August 2008.
29 Luck, ‘The United Nations’, 3.
30 On China’s position on the Responsibility to Protect see Sarah Teitt, ‘China and the
Responsibility to Protect’, Global Responsibility to Protect 1:2 (2009), 208–36.
31 William Schabas, Genocide in International Law: The Crime of Crimes (Cambridge: Cambridge
32 Stephen J. Toope, ‘Does International Law Impose a Duty upon the UN to Prevent Genocide?’,
McGill Law Journal 46:1 (2000), 187–94.
33 Jerry Fowler, for example, argued that ‘the language of the [Genocide] Convention does not
provide any indication that such an extensive obligation was contemplated. Indeed it would be quite
bizarre to think that the drafters intended in 1948 to make intervention in the internal affairs of other
606 alex j. bellamy
However, important new light was shed on this issue by the ICJ’s 2007 ruling in the
Bosnia and Herzegovina v. Serbia and Montenegro case.
The Court found that Serbia was not guilty of genocide but had violated its
Article 1 obligation to prevent and punish the crime of genocide. The ICJ found
that, whilst states are not obliged to succeed in their efforts to prevent genocide,
they ‘must employ all means which are reasonably available to them to do so’.34 Of
course, it remains to be seen what measures may be thought to be ‘reasonably
available’ to states in given situations. It is unlikely that each individual state
now has a responsibility to intervene militarily to end genocide, and it should be
remembered that measures to end genocide must be consistent with international
law, which prohibits the use of force except in self-defence or with the express
authorization of the UN Security Council.35 However, it has been argued by the
former UN High Commissioner for Human Rights, Louise Arbour, that the
ruling—especially when taken in conjunction with the Responsibility to Protect
principle—imposes specific responsibilities on the members of the UN Security
Council, and the permanent members especially.36 The UN Security Council has
the legal authority to authorize armed intervention whenever it identifies a threat
to international peace and security. Moreover, as the world’s leading military
powers, permanent members of the Security Council almost always have the
military capacity to intervene to halt genocide. Thus armed with both the collective
authority to intervene and—between them—the capacity to do so, it could be
argued that armed intervention to halt genocide falls well within the scope of
‘reasonably available’ measures for permanent members of the Security Council.
This is not to say, of course, that armed intervention will be the most appropriate
course of action in every case. And it should be noted that this line of argument
depends on a very liberal interpretation of the law. Furthermore, whilst the Respon-
sibility to Protect principle enjoys the unanimous support of the UN General
Assembly, it does not resolve difficult questions about the most appropriate and
effective response to specific cases of genocide and mass atrocities. Moreover, even
with these caveats the idea that there might be a right or duty to intervene is
controversial.
Opponents—including many Third World states—maintain that international
order and the preservation of core values such as the right to self-determination
requires something approximating an absolute ban on the use of force outside
states obligatory’ (‘A New Chapter of Irony: The Legal Definition of Genocide and the Implications of
Powell’s Determination’, in Samuel Totten and Eric Markusen (eds), Genocide in Darfur: Investigating
the Atrocities in the Sudan (London: Routledge, 2006), 131).
34 In Marko Milanovic, ‘State Responsibility for Genocide: A Follow-Up’, European Journal of
International Law 18:4 (2007), 686.
35 Mark Gibney, ‘Genocide and State Responsibility’, Human Rights Law Review 7:4 (2007), 767.
36 Louise Arbour, ‘The Responsibility to Protect as a Duty of Care in International Law and
Practice’, Review of International Studies 34:3 (2008), 445–58.
the two parameters set out by the UN Charter—Security Council authorization

(Chapter VII) and self-defence (Article 51)—and that the Security Council should
interpret its remit narrowly.37 The starting point for this position is the assumption
that international society comprises a plurality of diverse communities each with
different ideas about the best way to live. According to this view, international society
is based on rules—the UN Charter’s rules on the use of force first among them—that
permit these communities to coexist relatively peacefully.38 In a world characterized
by radical disagreements about how societies should govern themselves, proponents
of this view hold that a right and duty of humanitarian intervention would create
disorder, as states would wage wars to protect and violently export their own cultural
preferences. What is more, a right of intervention would open the door to potential
abuse. Historically, states have shown a distinct predilection towards ‘abusing’
humanitarian justifications to legitimize wars that were anything but humanitarian.
Most notoriously, Hitler insisted that the 1939 invasion of Czechoslovakia was
inspired by a desire to protect Czechoslovak citizens of German ethnicity whose
‘life and liberty’ were threatened by their own government.39 It was precisely because
of the fear that states would exploit any loophole in the ban on the use of force that
the delegates who wrote the UN Charter issued a comprehensive ban with only the
two limited exceptions: force used in self-defence and under the authority of the
Security Council.40 In addition to the problem of abuse, many post-colonial states
continue to oppose humanitarian intervention because they consider it a dangerous
affront to another core principle, self-determination. They worry that a duty to
intervene would grant a licence for the great powers to interfere in their domestic
affairs, undermining their right to self-government.41
Although these are powerful arguments which should temper enthusiasm for and
analysis of armed intervention, it is important to avoid the belief that the duty to
intervene is an agenda exclusively, or even primarily, concerned with imposing
Western values on the rest. On the one hand, those who caution against intervention
as a matter of principle should be mindful of the evidence about how genocides
end. On the other hand, it is important to not assume that the post-colonial world
speaks as one on this issue. There are good grounds for thinking that there is genuine
consensus that intervention, properly authorized, is a legitimate response to
37 For a good exposition of this position, linking it to the views of several Third World
governments see Mohammed Ayoob, ‘Third World Perspectives on Humanitarian Intervention and
International Administration’, Global Governance 10:1 (2004), 99–118.
38 A position set out and defended at length by Richard H. Jackson, The Global Covenant: Human
Conduct in a World of States (Oxford: Oxford University Press, 2002).
39 Ian Brownlie, ‘Humanitarian Intervention’, in John N. Moore (ed.), Law and Civil War in the
Modern World (Baltimore: Johns Hopkins University Press, 1974), 217–21.
40 An argument put forth in detail by Simon Chesterman, Just War or Just Peace? Humanitarian
Intervention in International Law (Oxford: Oxford University Press, 2001).
41 A position set out by Ayoob, ‘Third World Perspectives’.
608 alex j. bellamy
genocide and mass atrocities. Nine years before NATO intervened in Kosovo without
the authority of the Security Council, a group of West African states (ECOWAS) did
likewise in Liberia. Five years before the adoption of the Responsibility to Protect
principle by world leaders, members of the African Union gave the regional institu-
tion a right to intervene in response to serious humanitarian emergencies.42
Another concern is that military interventions are always tainted by power
politics because an intervener’s motives are never wholly humanitarian.43 Wil
Verwey defined humanitarian intervention as ‘the threat or use of force . . . for the
sole purpose of preventing or putting a halt to a serious violation of human rights.’
He argued that there never had been genuine humanitarian interventions because
prior interventions had almost always been motivated by non-humanitarian con-
cerns. Such interventions could not be considered humanitarian or ‘disinterested’
and were therefore illegitimate.44
Although this line of thinking points to some important concerns, there are at
least three problems. First, there are good reasons to suggest that states will not risk
the lives of their own citizens in order to save the lives of others. As Donald
Bloxham has persuasively argued, ‘humanitarian intervention tends to occur
only when the cause overlaps with the material interests of those intervening.’45
By insisting that interveners be guided by purely humanitarian motives, the bar
is placed so high that no military action could realistically pass the test, even
though in some cases only military measures will remedy human suffering. From
this perspective, it was entirely correct for the West not to intervene in Armenia,
Rwanda, Srebrenica, and Darfur because they could not have done so in a disin-
terested fashion.
Second, motives and purposes are subjective and can be easily disguised by clever
political leaders. Making motives the sole criterion of legitimacy is problematic,
therefore, because it is very difficult for them to be properly assessed, especially
without the help of considerable hindsight. Finally, focusing solely on what moti-
vates an action tells us little about its consequences. Contra those that demand
purely humanitarian motives, a self-interested intervention that ends a genocide
(e.g. Vietnam in Cambodia) is surely preferable to a disinterested intervention that
ultimately fails (e.g. USA in Somalia). Indeed, it is usually the presence of a degree
42 See Paul D. Williams, ‘From Non-Interference to Non-Indifference: The Origins and

Development of the African Union’s Security Culture’, African Affairs 106:423 (2007), 253–79.
43 See Bernard Williams, ‘Humanitarianism and the Right to Intervene’, in Geoffrey Hawthorn
(ed.), In the Beginning Was the Deed: Realism and Moralism in Political Argument (Princeton, NJ:
Princeton University Press 2005), 145–53.
44 Wil Verwey, ‘The Legality of Humanitarian Intervention after the Cold War’, in E. Ferris (ed.),
A Challenge to Intervene: A New Role for the United Nations? (Uppsala: Life and Peace Institute,
1992), 12–36; and Bhikhu Parekh, ‘Rethinking Humanitarian Intervention’, International Political
Science Review 18:1 (1997), 55–74.
45 Donald Bloxham, ‘Genocide: Can We Learn from History?’, BBC History Magazine (January
2007), 48.
of self-interest that makes states prepared to sacrifice the lives of their citizens to
save foreigners from peril.
A more sophisticated position is put forward by Nicholas Wheeler. The starting
point is the idea that humans, much less states, are never prompted to act by a single
motive. Wheeler therefore argues that whilst motives are important they should not
be the ‘threshold’ consideration. That is, actions that produce humanitarian good
should not be condemned because they are not inspired by humanitarian motives.
The key test should be that the means chosen by the intervener must not undermine
the positive humanitarian outcome.46 Thus, Vietnam’s self-interested intervention in
Cambodia should be applauded because the non-humanitarian motives did not
undermine the humanitarian goal of removing the Khmer Rouge from power. By
contrast, because the French government was primarily motivated by a concern to
protect Hutu allies and Francophones in Rwanda, rather than a desire to end the
Rwandan genocide, it chose a strategy that did relatively little to protect the geno-
cide’s victims. In that case, non-humanitarian motives undermined potential hu-
manitarian outcomes. According to Wheeler’s schema this should render the French
intervention (Operation Turquoise) illegitimate.47 This more sophisticated approach
allows us to factor the important concerns expressed by Verwey into our assessment
of intervention while acknowledging that states usually have mixed motives.
I N H I B I TO R S TO I N T E RV E N T I O N
................................................................................................................
Why, when there is so much agreement about the necessity of preventing and
ending genocide, are states so reluctant to intervene militarily? This section exam-
ines three of the most often cited reasons for the failure to intervene—international
law, political will, and prudential considerations.
International law
Questions about intervention tend to be framed around an enduring struggle
between sovereignty and human rights. By this account, sovereignty refers to the
rights that states enjoy to territorial integrity, political independence, and non-
intervention. Where sovereign states are either unwilling or unable to protect the
fundamental freedoms of their citizens, sovereignty and human rights come into
46 Wheeler, Saving Strangers, 33–4.

47 See ibid. chs 3 and 7.
610 alex j. bellamy
conflict. This tension is evident in the UN Charter. Whilst calling for cooperation
to reaffirm faith in fundamental human rights, the Charter (Article 2(4)) outlaws
war as an instrument of policy with only two exceptions (each state’s inherent right
to self-defence [Article 51] and collective measures authorized by the UN Security
Council [Chapter VII]) and affirms the principle of non-interference [Article 2
(7)]) by prohibiting the UN from interfering ‘in matters essentially within the
domestic jurisdiction of states’. These legal rights, it is often argued, constitute a
powerful barrier to intervention, and it has proven very difficult to build sufficient
consensus in the UN Security Council to persuade it to authorize intervention
against a full-functioning state guilty of perpetrating genocide or mass atrocities.
For example, in a March 2005 Security Council debate on whether to refer alleged
crimes in Darfur to the International Criminal Court (ICC), the US representative
explained that country’s abstention by arguing that the Court ‘strikes at the essence
of the nature of sovereignty.’48
States that act without the authority of the Security Council can pay a very high
price. For example, in 1979, when Vietnam invaded Cambodia and ousted the
murderous Khmer Rouge regime, responsible for the death of some two million
Cambodians, it was condemned for violating Cambodian sovereignty. China’s
representative at the UN described Vietnam’s act as a ‘great mockery of and insult
to the United Nations and its member states’ and sponsored a resolution con-
demning Vietnam’s ‘aggression’. The USA agreed.49 France argued that ‘the notion
that because a regime is detestable foreign intervention is justified and forcible
overthrow is legitimate is extremely dangerous. That could ultimately jeopardize
the very maintenance of law and order.’50
However, there are grounds for doubting the extent to which international law
actually is a barrier to intervention. Simon Chesterman has demonstrated that
sovereignty has not in fact inhibited unilateral or collective intervention to uphold
human rights in other countries. In response to arguments that intervention could
be promoted by relaxing the prohibition on the use of force, Chesterman argued
that ‘implicit in many of the arguments for a right of humanitarian intervention is
the suggestion that the present normative order is preventing interventions that
should take place. This is simply not true. Interventions do not take place because
states do not want them to take place.’51 Ultimately, it was not primarily concerns
about sovereignty that prevented timely intervention in Rwanda, but rather the
basic political fact that no state wanted to risk its own troops to save strangers from
genocide (see below).
48 S/PV.5158, 31 March 2005, 12.

49 Cited in Wheeler, Saving Strangers, 90–91.
50 Cited in Chesterman, Just War or Just Peace?, 80.
51 Ibid. 231.
What, though, of Vietnam’s invasion of Cambodia? Was it not the case that
Vietnam paid a heavy political and economic price because it was seen as
violating Cambodia’s sovereignty? This position certainly has merit but needs
to be viewed alongside two other considerations. First, Vietnam was not princi-
pally motivated by humanitarian concerns nor did it justify its invasion as a
humanitarian intervention. Second, and perhaps more importantly, we need to
take the arguments levelled against Vietnam with a pinch of cold war salt.
Whilst not denying the fact that many states, particularly some members of the
Non-Aligned Movement, opposed Vietnam on principled grounds, political
considerations played an important part in shaping the way that international
society reacted to the intervention.52 In the same year as Vietnam’s invasion of
Cambodia, Tanzania—a highly regarded state with a well-respected president,
Julius Nyerere—invaded Uganda and deposed Idi Amin with barely a ripple of
condemnation.53
Either way, it is clear that contemporary international law does not enable the
forging of consensus on collective action to end genocide beyond provisions on
Security Council authorization. This is where recent developments such as the
Responsibility to Protect principle and Bosnia v. Serbia ruling, which seek to
change the relationship between sovereignty, the responsibilities of states, and the
place of non-interference, might have a positive effect.
Political Will
Following on from Chesterman’s insight that interventions do not happen primar-
ily because states do not want them to happen, the second major inhibitor of
armed intervention is political will. Political will works in two ways to inhibit the
chances of intervention to end genocide. The first, and least discussed, is the
presence of prevailing interest. In other words, it is not just that powerful states
lack the will to take risks to save strangers, but that their pursuit of their own
interests leads them to support or shield the perpetrators. The link between China’s
(and to a lesser extent, Russia’s) obstinate support for the government of Sudan and
its interest in Sudanese oil and arms sales is well known.54 But the West has also often
put its own interests ahead of the protection of populations from genocide. As part of
its long war against communism, it supplied arms to the Indonesian government
as alleged communists were slaughtered and East Timor brutally repressed and
helped fund, arm, and train the Khmer Rouge after its defeat by Vietnam. For
52 Wheeler, Saving Strangers, 78–110.

53 On this see Grant Evans and Kelvin Rowley, Red Brotherhood at War: Vietnam, Cambodia and
Laos since 1975, 2nd edn (London: Verso, 1990).
54 For instance, ‘China and Darfur: The Genocide Olympics?’, Washington Post, 14 December
2006, A30.
612 alex j. bellamy
similar reasons, in the early 1980s the USA supported, funded, and armed the
genocidal regime in Guatemala. Motivated mainly by its interest in preserving
its influence in former colonial territories, France funded and armed the Hutu
government in Rwanda and supplied a substantial portion of the guns and machetes
that made genocide possible.55 All of the permanent members of the Security Council,
which has a special responsibility to protect populations from genocide, have there-
fore been implicated in genocide through their support for the perpetrators in the
past half-century. Sometimes, therefore, political will prompts great powers to ac-
tively protect or assist perpetrators, presenting a major obstacle to the goal of ending
genocide. The permanent members of the Security Council would therefore do well to
begin their engagement with the ending of genocide by desisting from actively
supporting its perpetrators.
The second, and more commonly discussed, aspect of political will relates
to the idea that states consider themselves to be responsible first and foremost
for the well being of their own citizens and are reluctant to spend tax money and
risk the lives of their soldiers in order to save strangers from genocide in other
countries. The effects of this lack of will were demonstrated in detail by the 1999
Report of the Independent Inquiry into the UN’s failure to prevent and then halt
the Rwanda genocide of 1994.56 The report opened with a damning but general
criticism, insisting that the Rwandan genocide resulted from the failure of the
whole UN system.57 The lack of resources and will was manifested in the UN
mission deployed in Rwanda (UNAMIR) not being adequately ‘planned, dimen-
sioned, deployed or instructed’ in a way that would have ‘provided for a proactive
and assertive role’ in the face of the deteriorating situation in Rwanda.58 The
mission was smaller than recommended by the UN secretariat, slow to deploy
owing to the reluctance of states to contribute troops, debilitated by administra-
tive difficulties, and when troops did arrive they were generally inadequately
trained and equipped.59
55 On USA and Indonesia see Yves Beigbeder, International Justice against Impunity: Progress and
New Challenges (The Hague: Martinus Nijhoff, 2005), 17–18; on US support for Guatemala’s genocidal
regime see the report of the UN-administered Historical Clarification Commission, Memory of
Silence: Report of the Commission for Historical Clarification: Conclusions and Recommendations at
http://shr.aaas.org/guatemala/ceh/report/english/toc.html (accessed 14 December 2008); on British
support for the Khmer Rouge, see Tom Fawthrop and Helen Jarvis, Getting Away with Genocide:
Elusive Justice and the Khmer Rouge Tribunal (Sydney: University of New South Wales Press, 2005),
68–9. On France and Rwanda see Daniela Kroslak, The French Betrayal of Rwanda (London: Hurst, 2007).
56 This discussion draws on Bellamy and Williams, Understanding Peacekeeping, ch. 5.
57 Independent Commission, Report of the Independent Inquiry into the Actions of the United
Nations during the 1994 Genocide in Rwanda, 12 December 1999, 1.
58 Ibid. 2.
59 Ibid.
The Inquiry concluded, therefore, that the UN’s failure in Rwanda was largely
created by a critical disjuncture—endemic in many UN operations at the time60—
between the tasks given to peacekeepers and their conceptual and material tools.
For largely political reasons UNAMIR was conceived as a small, cheap, and consent-
dependent operation despite evidence at the time that this would be inadequate.
In a tragic coincidence of history, UNAMIR’s mandate came onto the Security
Council’s agenda just one week after the killing of American peacekeepers in Somalia
in the infamous ‘Black Hawks down’ incident. The USA was understandably in no
mood to consider supporting the dispatch of more peacekeepers to Africa and
insisted that any force sent to Rwanda be limited in size and dependent on the
consent of the parties.
Prudential Considerations
Even when states have genuine moral concerns about the commission of genocide
in foreign countries, prudential considerations or competing priorities may augur
against armed intervention. The first prudential inhibitor to intervention is the
calculation that intervention might do more harm than good. Prominent human
rights NGO, the International Crisis Group, and individuals such as Gareth Evans
and Francis Deng all argued against military intervention in Darfur on the grounds
that it would be counterproductive, and indeed exacerbate the violence. They were
supported in this view by a leading commentator on African affairs, Alex de Waal,
whose own position on intervention changed between 2004 and 2006. Initially, de
Waal argued that foreign troops could make a ‘formidable difference’ to the lives
of Darfuri civilians.61 Soon after, de Waal was invited to help the African Union in
its efforts to broker a political settlement but the negotiators failed to persuade
all but one of the rebel groups to sign the Darfur Peace Agreement. The experience
of coming ‘agonisingly close’ to a political settlement and the further complication
of the situation on the ground no doubt contributed to de Waal’s change of heart
on the potential for military intervention to make a ‘formidable difference’. In
2006, he wrote:
The knock-down argument against humanitarian invasion is that it won’t work. The idea of
foreign troops fighting their way into Darfur and disarming the Janjaweed militia by force is
sheer fantasy. Practicality dictates that a peacekeeping force in Darfur cannot enforce its will
on any resisting armed groups without entering into a protracted and unwinnable counter-
insurgency in which casualties are inevitable. The only way peacekeeping works is with
60 See Part 3 of Bellamy and Williams, Understanding Peacekeeping.

61 Alex de Waal, ‘Darfur’s Deep Grievances Defy All Hopes for an Easy Solution’, The Observer, 25 July
2004, available at http://www.guardian.co.uk/society/2004/jul/25/internationalaidanddevelopment.
voluntarysector (accessed 15 December 2008).
614 alex j. bellamy
consent: the agreement of the Sudan government and the support of the majority of the
Darfurian populace . . . Without this, UN troops will not only fail but will make the plight
of Darfurians even worse.62
Which version is more accurate—that foreign forces could make a ‘formidable

difference’ to Darfuri civilians or that this position was ‘sheer fantasy’? These are
the sort of difficult questions that well-intentioned policy-makers must find
answers to in the midst of the genocidal storm. Francis Deng, a well-respected
diplomat and the UN’s Special Representative on Internal Displacement for over a
decade, supported de Waal’s 2006 opinion. Non-consensual intervention, he con-
cluded, would ‘complicate and aggravate’ the crisis by increasing the level of
violence and undermining the potential for cooperation with the Sudanese gov-
ernment.63
There are clearly difficult choices that policy-makers confront when weighing
up whether to commit troops to an intervention. When coupled with either
countervailing interests or an absence of national interests and the danger of
sustaining casualties, it is not hard to see why the uncertainty of success in any
complex operation to combat genocide tends to produce scepticism about the
merits of intervention. In addition to these considerations, we need to bear in
mind that policy-makers often must balance competing priorities. For example,
in relation to Darfur, the USA was required to balance its concern for the victims
of genocide with the pursuit of its national security objectives in the War on
Terror. The imperatives of the War on Terror made it difficult for the USA to
seriously contemplate armed intervention in Darfur for three primary reasons.
First, its military commitments in Iraq and Afghanistan meant that there was
little spare capacity to intervene in Darfur. Intervention in Darfur would have
required scaling back of the other major commitments and US policy-makers
prioritized the War on Terror, stability in Iraq, and support for the Afghan
government over stopping the genocide in Darfur. Second, the government of
Sudan made itself an important source of intelligence information on Islamic
extremism. Thus, American intelligence required a working relationship with the
government of Sudan. Third, given the public relations disaster that accompa-
nied the US intervention in Iraq, the last thing the USA needed was to antagonize
the Arab world further by intervening against Sudan’s Arab government on behalf
of Darfur’s African population.64
62 Alex de Waal, ‘The Book Was Closed Too Soon on Peace in Darfur’, The Guardian, 29 September
2006, available at http://www.guardian.co.uk/commentisfree/2006/sep/29/comment.sudan (accessed
15 December 2008).
63 E/CN.4/2005/8, 27 September 2004, paras. 22, 26, and 36.
64 These points are documented in greater length in Paul D. Williams and Alex J. Bellamy, ‘The
Responsibility to Protect and the Crisis in Darfur’, Security Dialogue 36:1 (2005), 27–47.
None of this is meant to justify inaction in the face of genocide but it goes some
way towards explaining why it is that governments usually choose to stand aside
when there is such a clear moral imperative to intervene to put an end to
genocides once begun. It also helps illuminate the difficult choices that policy-
makers confront.
C O N C LU S I O N
................................................................................................................
Typically, genocide ends either with the suppression and/or destruction of the
victim group or with the military defeat of the perpetrators. Only very rarely are
those military defeats affected by the intervention of external powers spurred
primarily by the intention to put an end to genocide. Indeed, only twice in the
past century have states intervened to put an end to genocide, and in both cases the
interveners had decidedly mixed motives. The problem, then, is not that there is
too much humanitarian intervention in times of putative genocide, but that there
is far too little. International society’s default response to genocide is to stand aside
and hope that the blood-letting comes to an end incidentally. This is despite the
emergence of a clear moral, political—and some would say legal—responsibility to
take timely and decisive action to put an end to genocide. It is, of course, correct to
argue that the responsibility to protect populations from genocide does not create a
duty of armed intervention in every case but we know that once genocide has
begun, only the choice of the perpetrators or their military defeat is likely to bring
it to an end. We should acknowledge, however, that decisions to intervene are
fraught with difficulties. Military intervention is legally problematic, especially
when there is no consensus in the Security Council. States, especially democratic
states, are understandably reluctant to sacrifice their citizens in order to save
foreigners in peril. Policy-makers therefore need to make difficult calculations
about the prospective costs and benefits of armed intervention in a context of
radical uncertainty. And, of course, there is no guarantee that intervention will
succeed in saving lives.
No amount of institutional reform and rhetorical finessing can get around
the fact that armed intervention to end genocide requires leaders who are
prepared to pay the political costs of failure. These decisions will always be
taken on an ad hoc and case-by-case basis and will always involve mixed motives,
and require difficult judgments and a degree of risk taking. Because intervention
to end genocide is so necessary and yet so rare, we commentators might help by
worrying less about the damage done to international order by armed interven-
tion against tyrants and more about the damage done to really existing human
616 alex j. bellamy
beings when international society stands aside. Clearly, non-violent prevention is

much better than violent cure but, ultimately, we must face the fact that once
genocide has begun only war on the perpetrators will bring it to a premature end.
Leaders in democratic countries would find it easier to fulfil their moral commit-
ments if their domestic constituents demanded that they do what was necessary
to put an end to genocide. This could start, of course, with a commitment not to
assist the perpetrators, even when material or strategic interests would seem to
demand such support.
................................................................................................................
The author thanks Dirk Moses and Donald Bloxham for their insightful and
comprehensive comments on an earlier draft of this chapter and assistance with
redrafting. Thanks also to Ruben Reike, Paul D. Williams, and Sara E. Davies.
Bass, Gary J., Freedom’s Battle: The Origins of Humanitarian Intervention (London: Knopf,
2008).
Bellamy, Alex J., Responsibility to Protect: The Global Effort to End Mass Atrocities
(Cambridge: Polity Press, 2009).
Chesterman, Simon, Just War or Just Peace? Humanitarian Intervention in International Law
Evans, Gareth, The Responsibility to Protect: Ending Mass Atrocity Crimes Once and for All
(Washington, DC: Brookings Institute, 2008).
Heidenrich, John G., How to Prevent Genocide: A Guide for Policymakers, Scholars and
Concerned Citizens (Westport, CT: Praeger, 2001).
Holzgrefe, J. L., and Robert O. Keohane (eds), Humanitarian Intervention: Ethical, Legal and
Political Dilemmas (Cambridge: Cambridge University Press, 2003).
Jones, Adam, Genocide: A Comprehensive Introduction (London: Routledge, 2006).
Kiernan, Ben, Blood and Soil: A World History of Genocide and Extermination from Sparta to
Darfur (New Haven, CT: Yale University Press, 2008).
Melvern, Linda, Conspiracy to Murder: The Rwandan Genocide (London: Verso, 2006).
Wheeler, Nicholas J., Saving Strangers: Humanitarian Intervention in International Society
chapter 30
.............................................................................................
P U N IS H MENT AS
P R EV E N T I O N ?
THE POLITICS OF PUNISHING
G É N O C I D A I R E S
.............................................................................................
donald bloxham
devin o. pendas
Alongside military and diplomatic intervention, as well as economic sanctions,

criminal trials have become the international community’s instrument of choice for
reckoning with genocide and crimes against humanity. While the other responses
tend in one way or another to be interventionist, seeking to halt genocide as it is
occurring, what we term the legalist approach to of genocide is generally retrospect-
ive in orientation, seeking to punish acts that have already taken place. Although the
initiation of legal proceedings has recently developed problematic interventionist
aspects, the basic fact remains that law punishes past acts, rather than intervening in
ongoing ones. Why has this preference for punishment arisen, what is it intended to
achieve, and does it achieve it?
The preference does not arise, as some of the advocates of international legalism
would maintain, as a way of depoliticizing the response to genocide. Quite the
opposite: international criminal law remains a political instrument and a particularly
inexpensive one at that. In both economic and political terms, international criminal
trials require only limited expenditures of capital. Accordingly, trials tend to be
618 donald bloxham and devin o. pendas
highly disposable—all the more so because they take place after the fact. As for what
international criminal law is intended to achieve, there are four justifications:
retribution, special prevention (preventing the perpetrator from reoffending), gen-
eral prevention (deterring other potential offenders), and moral pedagogy. This
chapter will argue that none of these are achieved by international criminal law.
Law’s crucial contribution is, instead, articulating a normative consensus that
mobilizes political pressure for compliance. As a mechanism for articulating norms
and forging consensus, the new international legalism offers a chance—albeit a
modest one—to shift patterns of behaviour and place limited checks on the free
hand of international power. But this is an ongoing political battle, one that can never
be resolved by the simple establishment of international courts, conventions, and
case-law, nor any of the other trappings of law. Whatever the aspirations of its jurist
cheerleaders, international law will never be analogous to domestic law, and the two
share little save the outward form. International legalism remains a form of politics.
Law’s role as a continuation of politics by other means brings with it a downside.
Precisely because legal responses to genocide are a form of politics, they are both
readily manipulable for strategic ends and highly dependent upon the constellation
of global political forces at any given moment. Often, for instance, allegations of
human rights violations or genocide are instrumentalized to stigmatize political
opponents, as exemplified by Colombia’s recent attempt to have Hugo Chavez
indicted at the International Criminal Court (ICC) for genocide for his support of
the FARC guerilla movement in Colombia. Alternatively, international legalism can
be used to validate military interventions where politically convenient, and in other
cases to substitute for intervention when such action would not be politically
convenient. Equally often, international trials simply provide a venue for the articu-
lation of competing accounts of mass atrocities which cannot themselves be fully
resolved in the courtroom, what Gerry Simpson has referred to as the ‘procedur-
alized clash of competing ideologies’.1
This chapter will be divided into three roughly chronological sections, each
dealing with an important stage in the chequered history of the legalist paradigm.
Despite the real innovations of the nineteenth century, we take the Nuremberg trials
as our starting point because the legal developments of the immediate post-war
period served as the crucible for most subsequent developments in international
legalism. Almost as soon as the Nuremberg model was developed and implemented
in new legal instruments like the 1948 UN Convention on the Prevention and
Punishment of Genocide, its efficacy was undermined by the global cold war,
which thus marks the second phase of our analysis. The third phase deals with the
re-emergence of the legalist paradigm in the first decade after the end of the cold war.
International legal codes saw significant growth in this period, new international
1 Gerry Simpson, Law, War and Crime (Cambridge: Cambridge University Press, 2007), 13.
punishment as prevention? 619
criminal courts, both ad hoc and, from 2002, permanent, came into being and major
trials for genocide and other crimes against humanity were conducted. The New
World Order proclaimed by the first president Bush was implicitly a juridified one, in
which legal norms would be universally applied through international institutions.
The wave of utopian expectations surrounding these developments obscured the
ongoing reality of power politics. The USA’s reaction to 11 September 2001, its
mounting opposition to the ICC, and its unilateral invasion of Iraq demonstrated
that the limits of the legalist paradigm had not by any means been overcome. Indeed,
the problems besetting the paradigm are intrinsic to the nature of an international
political system based on nation-state sovereignty and influenced by ‘great power’
agendas. The cold war was only a particularly pointed illustration of the primacy of
the political order over the legal.
T H E P A R A D I G M E S TA B L I S H E D :
T H E P O S T WA R M O M E N T
................................................................................................................
In January 1942, the governments in exile of the occupied countries of Europe
declared that among the principal aims of the Second World War should be ‘the
punishment, through the channel of organized justice, of those guilty and respon-
sible for these crimes [by Nazi Germany], whether they have ordered them, or in
any way participated in them.’2 In October 1943, the Big Three Allies followed suit,
proclaiming that any German soldiers or Nazi party members guilty of ‘atrocities,
massacres and executions’ would be ‘sent back to the countries in which their
abominable deeds were done in order that they may be judged and punished
according to the laws of these liberated countries.’3 Those perpetrators whose
crimes were without clear ‘geographical localization’ were to be punished by a
subsequent ‘joint decisions of the government of the Allies’.
With this, the Allies declared that justice, as much as peace or renewed interna-
tional stability, would be an essential goal of post-war policy. There had of course
been similar declarations in the First World War, culminating in the disastrous
trials of German war criminals at Leipzig and of Turkish ones at Constantinople.
What distinguished the declarations of the Second World War from their counter-
parts in the First was that this time, the Allies actually implemented their promises.
2 Arieh J. Kochavi, Prelude to Nuremberg: Allied War Crimes Policy and the Question of Punishment
(Chapel Hill: University of North Carolina Press, 1998), 20.
3 A Decade of American Foreign Policy: Basic Documents, 1941–49 (Washington, DC: Government
Printing Office, 1950), 13.
Indeed, the chief lesson of the First World War for the architects of the post-Second
World War international legal regime was that it was unwise to leave the punish-
ment of war criminals to their own states. There were two distinct and not entirely
consonant principles at work in the Allied response to this insight. First, there was
the territoriality principle articulated in the Moscow Declaration, under which
Nazi criminals would be punished by their victims. The second, not yet clear in the
Moscow Declaration, was an internationalist principle, according to which some
wartime criminals, the political leadership in particular, were in effect too big to be
left to national courts. It was the Americans in particular who in the closing stages
of the war came to embrace the idea of an international trial for the major war
criminals.4 The reasons for this were those of both principle, a moral opposition to
the major alternative of summary executions, and political. Roosevelt’s advisors
wanted to ensure broad public backing for their post-war policy.
Consequently, the International Military Tribunal (IMT) at Nuremberg and its
sister Tribunal in Tokyo were intended from the start to be only one element in a
much broader effort at legal prosecution. Indeed, although Nuremberg’s fame, both
at the time and subsequently, has overshadowed the other trials for Nazi atrocities,
the IMT was truly a drop in the ocean, quantitatively speaking. According to the
latest estimates, more than 95,000 Germans and Austrians were convicted for
wartime crimes throughout Europe.5 Of these, 19 were convicted by the IMT.
Nonetheless, it is the IMT and to a less extent, the twelve so-called successor
trials conducted by the American Military Tribunal, likewise at Nuremberg, that
form the model for the subsequent development of the international legalist
approach to genocide. In part, this is for statutory reasons. The London Charter,
authorizing the IMT, criminalized three categories of offences. The least innovative
of these, and hence the least controversial, was war crimes, namely, ‘violations of
the laws or customs of war’.6 ‘Crimes against peace’ criminalized wars of aggression
or in violation of international treaties. Although this was the crucial charge in the
minds of the Americans at the time, it has been the one with the least subsequent
international legal resonance. Finally, and in this context most importantly, the
London Charter criminalized ‘crimes against humanity’. These included the mass
murder and persecution of civilians, as well as ‘persecutions on political, racial, or
religious grounds’.7 War crimes and crimes against humanity both penalized the
killing of civilians, the crucial difference being whether the victims and the
perpetrators were of the same or different nationalities, and the connection to
4 For the full story, see Kochavi, Prelude.

5 Norbert Frei, ‘Nach der Tat: Die Ahndung deutscher Kriegs- und NS-Verbrechen in Europa—
eine Bilanz’, in idem (ed.), Transnationale Vergangenheitspolitik (Göttingen: Wallstein, 2006), 32. The
figures here only include convictions in Germany (East and West) through 1959.
6 International Military Tribunal, Trial of the Major War Criminals before the International Military
Tribunal (Nuremberg: IMT, 1947), i.11.
7 Ibid.
military operations and military occupation. Crimes against humanity, unlike war
crimes, could be perpetrated against a state’s own citizens. Furthermore, persecu-
tions were criminalized, ‘whether or not in violation of domestic law of the country
where perpetrated’. Thus, the London Charter for the first time prioritized inter-
national over domestic law and the sanctity of national sovereignty was, in principle
at least, subordinated to international jurisdiction.
While the term genocide was used in the IMT indictment and in those of several
of the successor trials to describe the criminal acts in question, it was not one of the
criminal charges articulated in the London Charter. No Nazi criminals were ever
convicted of genocide as such. Although war crimes and crimes against humanity
between them cover most of the actions constituting genocide, the specificity of that
crime as it was articulated by Raphael Lemkin and subsequently codified in the
Genocide Convention was not formally part of the Nuremberg model. Lemkin’s
central insight was that certain atrocities targeted individuals as bearers of largely
non-negotiable identities, that it was the groups themselves, rather than the indi-
vidual victims, who were the real targets of the perpetrators. This view was to a
degree implicit in the concept of persecution contained in crimes against humanity
but Lemkin, who disliked the concept of crimes against humanity, wanted to go
further and provide protection, not just to an amorphous humanity, but to the
identifiable groups actually persecuted in the world. This notion, that certain kinds
of human groups—those based on presumably ‘fixed’ identities—required special
legal protection in a dangerous world, led to the first and most dramatic expansion
of the Nuremberg model with the UN Genocide Convention of 1948.
In December 1946, the first session of the UN General Assembly passed a genocide
resolution, GA Resolution 96 (I), at the behest of Cuba, India, and Panama.8 It
affirmed the criminal character of genocide and urged states to criminalize it under
domestic law. As a non-binding resolution, it also called on the UN to study the
feasibility of a more potent genocide convention. The Economic and Social Council
of the UN and the Sixth (Legal) Committee wrote and debated various drafts of a
genocide convention throughout 1947 and 1948, before finally passing the convention
in December 1948. There was a good deal of bureaucratic back-and-forth in the
legislative history of the genocide convention. This had little to do with bureaucratic
inertia. It was easy (and cheap) to morally abhor genocide, as in the genocide
resolution. Doing something about it invariably came with political costs attached.
The most obvious and important of these was that formally criminalizing genocide
in an international convention would inevitably entail at least nominal restrictions
on the behaviour of states and potentially provide a pretext for international
interference in domestic affairs. The potential for such intervention was deliberately
and tightly limited by the UN Charter (the foundation stone of post-war political
8 William A. Schabas, Genocide in International Law: The Crime of Crimes (Cambridge: Cambridge
University Press, 2000), 42–7, 51–81.
order), which was a ‘Westphalian document par excellence’, stressing the central
significance of the sovereign state as the fundamental building block of the
international system.9 Rhetoric aside, the architecture of that system as a whole
was primarily structured towards prohibiting interstate warfare and the trans-
gression of interstate boundaries, much less so with crimes committed within
established state borders. Indeed, as had been the case in the interwar period and
the nineteenth century, it was in certain circumstances simply preferable for
internal population groups like Turkey’s Kurds, or other potential secessionists,
to be violently repressed in the interests of maintaining borders arrived at
through earlier warfare.10 A genocide convention threatened to undermine this
carefully constructed edifice of post-war stability by introducing a countervailing
principle of humanitarian intervention.
There was also the related question of jurisdiction. If genocide was a crime of
state, who could prosecute it? As many would argue in the subsequent debates, only
an international criminal court was likely to have the independence necessary to sit
in judgment on such crimes. Yet a permanent court laying claim to a superordinate
international jurisdiction proved to be an intolerable threat to national sovereignty
for a great many delegates. Nuremberg was one thing, since it was an ad hoc tribunal
with jurisdiction exclusively over Nazi crimes. A standing court was a different
matter entirely. Finally, there was the simple fact that a convention would require a
careful and authoritative definition of genocide. This could either narrow or expand
the meaning of what was already becoming a particularly potent term of moral and
political opprobrium.
In order to get the genocide convention passed at all, several compromises,
which rendered the document virtually inert from the very start, were necessary.11
The first and most important of these concerned jurisdiction. The initial secretariat
draft of the genocide convention contained as an appendix two draft statutes for a
permanent international criminal court, one with restricted jurisdiction over
genocide alone, another with expansive jurisdiction over international crimes to
be defined by the International Law Commission. The guiding assumption, even if
it was not made explicit in the Convention, was that genocide was generally a state
crime and therefore required international jurisdiction if there was to be any hope
of prosecution. The French delegate to the Sixth (Legal) Committee strongly
supported this notion. ‘Genocide was committed only through the criminal inter-
vention of public authorities; that was what distinguished it from murder pure and
9 Kalevi J. Holsti, The State, War, and the State of War (Cambridge: Cambridge University Press,
1996).
10 Donald Bloxham, The Great Game of Genocide (Oxford: Oxford University Press, 2005), ch. 1 and
the second ‘interlude’.
11 Devin O. Pendas, ‘Towards World Law?: The Failure of the Legalist Paradigm of War, 1945–1980’,
in Stefan-Ludwig Hoffmann (ed.), Human Rights in the Twentieth Century: Concepts and Conflicts
(Cambridge: Cambridge University Press, forthcoming).
simple. The purpose of the convention which the Committee was drawing up was
not to punish individual murders, but to ensure the prevention and punishment of
crimes committed by rulers.’12 Hence, it was imperative, he concluded, to establish
an International Criminal Court forthwith. The French support for linking the
genocide convention to the formation of an international court proved to be the
minority position, however. The Venezuelan representative on the Sixth Commit-
tee, for instance, warned that
the institution of international criminal jurisdiction could only lead to unfortunate results,
in view of the existing world situation. Friction might be created which could disturb the
peace among nations. The establishment of international penal jurisdiction should be
reserved for the future when international relations would be more favorable to such an
institution.13
The Polish delegate concurred, saying that he could on no account ‘sacrifice

questions of principle’ concerning the sanctity of national sovereignty. The Amer-
icans and the British, meanwhile, crafted a compromise that prevented the issue of
an international court from torpedoing the negotiations altogether. At US initia-
tive, Article VI of the final convention left open the possibility that in future, an
International Criminal Court might have jurisdiction over genocide, while the
British inserted into Article IX an option to have the existing International Court
of Justice adjudicate interstate disputes over the interpretation of the convention.
This was enough to satisfy supporters of international jurisdiction without alienat-
ing the advocates of national sovereignty.
The price was rather high, though. Article VI of the final convention states that
perpetrators of genocide ‘shall be tried by a competent tribunal of the state in the
territory of which the act was committed’.14 In other words, as long as the genocidal
regime remains in place, there is no possibility whatsoever of an actual prosecution
for genocide. Regime change would be a prerequisite for the prosecution of
genocide. Given regime change, the trial would not threaten the interstate order
because the perpetrators would already have been deposed and the erstwhile
perpetrator state would remain under a new regime; absent regime change, both
state and regime would remain untouched—all of this is the obverse side of the
achievement of individualizing responsibility for state atrocity under international
law (the state itself remains unprosecuted). The genocide convention was therefore
effectively stillborn. It is hardly surprising that the first successful prosecution of
genocide did not occur until the Akayesu case before the International Criminal
Tribunal for Rwanda (ICTR) in 1998, and then not under the Genocide Convention
12 UN Doc. A/C.6/SR 97, p. 373.

13 UN Doc A/C.6/SR 130.
14 Schabas, Genocide, 566.
but under the Statute for the ICTR. The Nuremberg breakthrough, according to
which international law would trump domestic law, proved to be less a general
principle than a short-term expedient.
T H E D I S A P P E A R A N C E O F T H E PA R A D I G M :
T H E C O L D WA R
................................................................................................................
Nothing better illustrated the transitory nature of ‘Nuremberg’ than the collapse of
the Nuremberg edifice in the 1950s. A growing German opposition to the Allied
trial and occupation regimes found a conducive environment with the onset of the
cold war. The need to placate German national sentiment amid the burgeoning
political conflict with the USSR led first to the ending of the war crimes trials
programmes in all western occupation zones in the late 1940s, in the context of a
general easing of occupation policy. Later it resulted in a series of increasingly
politicized ‘sentence reviews’ which ultimately developed the simple aim of releas-
ing all war criminals, most of them prematurely.15 The final four war criminals in
US custody were released by 1958: the number incarcerated at the beginning of 1955
had been forty-one. Jails in the erstwhile British zone of Germany were empty by
1957.16 Among those released after serving only a few years of life sentences and
commuted death sentences were some of the worst Nazi offenders, including
commanders of the Einsatzgruppen (SS killing squads) and senior members of
the concentration camp hierarchy. Rejection of the legal validity of the trials was
subtly built into articles 6 and 7 of the 1952 Bonn Treaty ending the Allied
occupation statute.17
Treatments of the Nuremberg trials by legal scholars have generally been silent
on the collapse of the legal machinery, focusing instead on the achievements of the
courtroom itself, and the legacies created in law. Yet divorcing the German trials
from any broader political context is to undermine one of the most important
rationales for trial in the first place. That rationale, to paraphrase US chief
prosecutor Robert Jackson’s opening address before the IMT, was to impose the
rule of law on naked power relations. In other words, the collapse of Nuremberg in
the 1950s (along with that of the International Military Tribunal for the Far East)
and Memory (Oxford: Oxford University Press, 2001), Chap. 4; Peter Maguire, Law and War: An
American Story (New York: Columbia University Press, 2001).
16 Maguire, Law and War, 256.
17 Ibid. 229.
illustrates that law may influence the exercise of might but the process also works in
reverse. The particular problem with the Nuremberg case is that much of its
importance rested on the fact that it brought a major world power to book, but
the enduring geopolitical significance of Germany (and Japan) effectively placed a
limit on the extent of this reckoning.
Scholarly silence on the releases of the 1950s may be legitimate for a narrow
legal approach that is only interested in legal instruments, institutions, and
precedents, and can leap across time, like a frog jumping from one water lily to
the next, landing selectively on those moments where the law does seem to come
into its own.18 According to that depiction, which mirrors some of the more
teleological scholarship on the emergence of human rights, the cold war can be
seen as a simple hiatus, an aberration in the development of worldwide democ-
racy and international law; with the end of the cold war, the time of genuine
sovereign accountability has now arrived, and we can just pick up where Nurem-
berg left off. As Geoffrey Robertson puts it on the concluding page of what is
otherwise a measured assessment of the progress of humanitarian law: ‘although
the twenty-first century will have its share of despots, they will be fewer and in the
absence of the Cold War, they will not have superpower support. There will no
longer be any need to say, as FDR said of Grandfather Somoza, “he may be a son
of a bitch, but he’s our son of a bitch.”’19
The alternative view, grounded in an appreciation of the political context so
often missing from the work of legal scholars, is to see the cold war as a pointed
illustration of some of the potentialities of ‘our’ (Western) political-ethical
system. The cold war also illustrated how far legal and humanitarian language
could be abused by strategic interest. Atrocities and aggressive wars were perpe-
trated by both sides and their proxies, while the rhetoric of international law and
human rights was frequently used as a weapon to stigmatize the other side for
things that one’s own side was also doing, in a manner that disillusioned both
former Nuremberg lawyers and historians of the trial.20 Invocation of the geno-
cide convention in the third quarter of the twentieth century over cases like
Nigeria (Biafra) and East Pakistan (Bangladesh) only served to underline the
irrelevance of the document as a concrete ground for political action, much like
the rights declarations of the same period.21 The UN Security Council was
divided among its permanent members into capitalists and communists (neither
18 See, e.g., Gary Jonathan Bass, Stay the Hand of Vengeance: The Politics of War Crimes Tribunals
(Princeton, NJ: Princeton University Press, 2000).
19 Geoffrey Robertson, Crimes against Humanity: The Struggle for Global Justice (New York: New
Press, 1999), 387.
20 Telford Taylor, Nuremberg and Vietnam: An American Tragedy (New York: Bantam, 1971); Eugene
Davidson, The Nuremberg Fallacy (Columbia: University of Missouri Press, 1973).
21 On Pakistan, A. Dirk Moses, ‘The United Nations and the Failure to Prosecute: The Case of East
Pakistan, 1971–1974’, in Hoffman (ed.), Human Rights in the Twentieth Century, forthcoming.
‘side’ internally harmonious) while, in the General Assembly and among the
temporary members of the Security Council, the situation was complicated by
the representatives of a growing number of ‘Third World’ states as UN member-
ship nearly tripled from 1945 to 1975. Some of these states provided the battle-
grounds for the indirect warfare of the First and Second Worlds, and provided
most of the death toll of the cold war. Other states were courted by the major
protagonists in the economic-political-cultural contest, and thus gained tempo-
rary influence. Still others even succeeded in limited ways in using the UN
to press their own agendas against Security Council states, as Algeria did with
France. Almost none were prepared to compromise their hard-won post-colonial
independence by supporting potentially intrusive human rights regimes or general
principles of ‘humanitarian intervention’. Indeed, one could with some justification
view this as the major achievement of the Non-Aligned Movement.
The record of the cold war shows how dangerous it is to extrapolate to general
trends from brief moments of ‘legalist’ triumph such as the year 1945. The legal
optimism greeting the end of the Second World War was repeated at the end of the
cold war; with the benefit of historical perspective we should be equally wary of
succumbing to the temptations of that optimism. The end of the cold war left the
most important ‘vanquished’ protagonist—what became Russia—untouchable in
terms of accountability for crimes committed by the Soviet regime. Even in its
weakened state, Russia remained too powerful for foreign powers to contemplate
even retroactive interference in its ‘domestic’ affairs. There was no question in any
circumstances of bringing the cold war victor to book for its earlier crimes—and
insofar as any of the USA’s former Latin American allies addressed the abuses of
their former right-wing dictatorships, this was generally done outside the court-
room through truth and reconciliation processes, and in ways that could not
threaten the socio-economic order that the dictators had put in place.22 When
criminal law was used in post-authoritarian Latin America, as in for instance
Argentina, it was carefully calibrated so as to not disturb the fragile balance of
domestic power and risk renewed violence.23
In the sense of taming power, therefore, the legal developments after the cold war
are actually less impressive than the temporary achievements of Nuremberg. Yet it
was precisely Nuremberg, alongside the Genocide Convention, that was invoked in
the 1990s. With the end of communism in Eastern Europe, it became possible to
talk more realistically—however temporarily—about a single world order with a
single set of governing frameworks. Alongside firstly the vanguard organizations of
free market capitalism and secondly the apostles of parliamentary democracies,
22 Lawrence Weschler, A Miracle, a Universe: Settling Accounts with Torturers (Chicago: University
of Chicago Press, 1998), and more generally, Robert I. Rotberg and Dennis Thompson (eds), Truth v.
Justice: The Morality of Truth Commissions (Princeton, NJ: Princeton University Press, 2000).
23 Carlos Santiago Nino, Radical Evil on Trial (New Haven, CT: Yale University Press 1996).
Western jurists could make their mark on shaping the norms of that order, and
they were hurried into action by the ethnic cleansing and murder attendant on the
dissolution of the former Yugoslavia.
T H E PA R A D I G M R E A S S E RT E D
................................................................................................................
The unanticipated collapse of the Soviet Empire in 1989 and the ensuing end to the
cold war brought with it sweeping, if short-lived, hopes for a peaceful, stable ‘New
World Order’, in which the entire world increasingly came to resemble the United
States and history itself came to an end as the great ideological struggles of the past
gave way to a universal consumer democracy. Of course history did not end in 1989.
Nor did the New World Order prove to be anything like as peaceful and stable as its
proponents anticipated. Indeed, the collapse of the cold war order brought with it
outbreaks of large-scale violence in unanticipated places, including Europe itself
with the dissolution of Yugoslavia in the 1990s. In the absence of great power
rivalries, much of the violence of the post-cold war period operated below the level
of geopolitical concern for the major international actors. Secretary of State James
Baker’s notorious comment that the United States had no dog in the fight between
Croatia and Serbia can be taken as indicative. Baker was only more blunt, not more
callous than his many counterparts in the United States and Europe.
Given the lack of traditional great power ‘interests’ in many of the world’s new
conflict zones, interventions, even of the cynical proxy war variety typical of the
cold war, were hard to justify.24 At the same time, however, the near universal
mediatization of the world meant that these conflicts were often difficult for
politicians to simply ignore. If action was not necessarily called for, pseudo-
action was. It is in this context that one must understand how the return of
history after 1989 brought with it what Norbert Frei has called the ‘return of
law’.25 International criminal tribunals, with their promise of retrospective pun-
ishment for mass atrocity, were ideally suited to providing the semblance of
action while diffusing pressure to undertake more substantive interventionist
measures. The initial impetus for the formation of the International Criminal
Tribunal for the Former Yugoslavia (ICTY) came, after all, not from the members
24 David Rieff, Slaughterhouse: Bosnia and the Failure of the West (New York: Touchstone, 1995);
Philip Gourevitch, We Wish to Inform you That Tomorrow We Will Be Killed with Our Families: Stories
from Rwanda (New York: Picador, 1998).
25 Norbert Frei, ‘Die Rückkehr des Rechts: Justiz und Zeitgeschichte nach dem Holocaust—eine
Zwischenbilanz’, in Arnd Bauerkämper, Martin Sabrow, and Bernd Stöver (eds), Doppelte
Zeitgeschichte: Deutsch-deutsche Beziehungen, 1945–1990 (Bonn: Dietz, 1998).
of the UN Security Council but from Human Rights Watch.26 In the face of such
pressure from below, it was largely what Pierre Hazan called the ‘opportunistic
steeplechase between France and the United States’ that eventually led to the
creation of the ICTY, as each tried to claim moral leadership while avoiding
military action. That the ICTY was intended to be impotent is apparent in
the complete lack of resources placed at its disposal initially. That in the end
the tribunal was not quite so impotent as its creators initially planned was due
largely to the political skills of its first two chief prosecutors, Richard Goldstone
and Louise Arbour.27 Goldstone, though not so effective as a prosecutor as his
successors, was politically skilled and was able to mobilize private resources to get
the tribunal up and running. Arbour forged an effective prosecutorial strategy
that led to actual trials and, by indicting Slobodan Milošević in May 1999 for his
ongoing crimes in Kosovo, interjected the tribunal directly into the political
process in the Balkans and thereby made it more directly relevant than it would
otherwise have been.
Nonetheless, two points are worth stressing about the two ad hoc UN tribunals
of the 1990s, the ICTY and its sister tribunal for Rwanda, the ICTR. The first is
that neither ever stopped anyone from committing a single crime. Arbour’s
indictment of Milošević during the Kosovo war may have made the tribunal an
indispensible actor in the political process; it did not, however, prevent the
atrocities in Kosovo from continuing. The second point concerns the indepen-
dence of the courts. Obviously, as ad hoc tribunals, the ICTY and the ICTR
were created with deliberately restricted jurisdictions. This in itself limited their
remit considerably and kept them to a degree under the political control of their
sponsors on the UN Security Council. If the tribunals, especially the ICTY,
nonetheless achieved greater independence than their sponsors envisaged, this
ought not to blind us to the ultimate limits the tribunals faced. After all, they
remained dependent on state actors, the Americans in particular, for intelligence
resources, as well, obviously, as on UN troops for the enforcement of their
warrants and indictments. The real boundaries of this independence revealed
themselves in the ICTY’s investigation of NATO for war crimes during the air war
in Kosovo. The simple fact that the ICTY investigated these was, in Louise
Arbour’s words, ‘staggering’ for the NATO powers.28 Whatever the merits of
the case, though, it can hardly be surprising that the ICTY found that NATO had
committed no war crimes. NATO spokesperson Jamie Shea had reminded the
prosecutor’s office,
26 Pierre Hazan, Justice in a Time of War: The True Story behind the International Criminal Tribunal
for the Former Yugoslavia (College Station: Texas A&M, 2004), 14.
27 John Hagan, Justice in the Balkans: Prosecuting War Crimes in the Hague Tribunal (Chicago:
28 Hazan, Justice in a Time of War, 130.
Don’t bite the hand that feeds you . . . The people of NATO are the ones who apprehend the
war criminals indicted by the Tribunal . . . We all want to see war criminals judged and I am
certain that, when Prosecutor Arbour returns to Kosovo and sees the facts, she will indict
the Yugoslav nationals, and no other nationality.29
This was exactly what happened under Arbour’s successor, Carla Del Ponte.
The ad hoc tribunals of the 1990s thus must be seen as at best ambiguous
institutions. On the one hand, there can be no doubt that many of the men and
women working for these tribunals were passionate in their pursuit of justice. Nor
can there be much doubt that those convicted by the ICTY and the ICTR richly
deserved their punishment and that it is preferable to see such criminals punished
rather than left free. However, the inflated promises made on behalf of these
tribunals that, henceforth, war criminals and génocidaires would, in the words of
Boutros Boutros-Ghali, ‘know the sanction of international law’ proved to be
hyperbolic. Moreover, the fiasco of the Milošević trial, which dragged on for
years before terminating with the former dictator’s death in prison, has revealed
an unavoidable tension between the hyper-careful concern for due process embod-
ied in the ad hoc tribunals and the moral and political requirements of swift and
efficacious justice.
The changed atmosphere of the 1990s also gave a renewed impulse to an older
idea: the establishment of an International Criminal Court to go alongside the
International Court of Justice established in 1945. The ICC came into effect in 2002,
and its jurisdiction and functions are based on the Rome Statute of 1998. The ICC
operates on the principle of complementarity with domestic courts: it will only
concern itself with the prosecution of cases in which the state concerned is unable
or unwilling to prosecute its citizens for breaches of international law. Like the ad
hoc tribunals, the ICC’s mandate is to consider genocide, crimes against humanity,
and war crimes. It can also consider aggressive warfare, which, in the eyes of the
Nuremberg lawgivers, was their most important legacy to international law.30 But it
cannot do so until a definition of aggressive war is internationally agreed, which is
no insignificant obstacle, given the past difficulties in framing that concept.31
Enforceability is another thing again, given the age-old use of aggressive warfare
for the most powerful states, and the equally venerable tradition of states finding
‘self-defensive’ pretexts for warmongering.
A particularly important aspect of the ICC, like the ad hoc tribunals, is its
genuinely international constitution. It is no longer possible to talk about
29 Ibid. 132.
30 On the institutional background, M. Cherif Bassiouni, ‘The Permanent International Criminal
Court’, in Mark Lattimer and Philippe Sands (eds), Justice for Crimes against Humanity (Oxford: Hart,
2007), 173–211.
31 Jonathan Bush, ‘“The Supreme . . . Crime” and Its Origins: The Lost Legislative History of the
Crime of Aggressive War’, Columbia Law Review 102 (December 2002), 2324–401.
‘victor’s justice’ in prosecution in the way that it was in the Nuremberg era.
Important elements implicit in the accusation of victor’s justice persist,
however. In particular, genuinely neutral enforcement will be difficult to
achieve. The ICC by definition will depend on the cooperation of powerful
states to enforce its decisions, making it difficult to imagine how it could ever
enforce its will upon those states. This problem, already foreshadowed in the
ICTY’s ruling on the NATO bombing of Kosovo, is unlikely to go away any
time soon.
The ICC cannot escape its subordination to the existing global power struc-
ture. The ICC’s remit is constrained to states that have ratified and acceded to the
Rome Statute. The only exceptions are for cases brought to the court’s attention
by the UN Security Council under Article VII of the UN Charter, concerning acts
likely to disturb the peace internationally. Yet the three most politically and
militarily powerful states with crucial permanent membership of the Security
Council—the USA, China, and Russia—are not Rome signatories. (The USA has
also signed bilateral immunity agreements with around one hundred states,
including Rome signatories, to keep American nationals from the court’s juris-
diction.) The three states are exempt from the court’s scrutiny, and are in the best
imaginable position to keep their allies, whether or not they are Rome signatories,
out of the court’s reach. Given that the Security Council’s approval will often be
necessary to enforce arrest warrants against ICC indictees, there is yet further
scope for the Security Council to undermine the ICC. The extent to which the
ICC, more specifically the office of the prosecutor (OTP), can enforce its will in
the pursuit of politically significant suspects will in turn influence the seriousness
with which the court is taken by the many signatory states, states which might
also have to lend their troops to Security Council-mandated operations in
pursuit of suspects.32
The practical constraints on the ICC help explain why the cases it is pursuing
all stem from a part of the world of relatively low geostrategic significance for the
major powers: Africa. It has been suggested, against a backdrop of long-standing
self-interested Western interventionism in Africa, that the cases pursued in the
Central African Republic, the Democratic Republic of Congo (DRC), Uganda,
and Sudan are illustrations of neo-imperial victimization. This interpretation
does not fit the facts. The first three cases were referred to the ICC by the states
themselves, each of which is party to the Rome Statute—indeed, at least two of
the referrals were partly motivated by the desire to delegitimize domestic political
enemies for the sorts of crimes of which the referring state authorities were
themselves not innocent. The Sudanese case was a Security Council referral
32 On the credibility problems for the ICC, see Phil Clark, ‘Ocampo’s Darfur Strategy Depends on
Congo’, Oxford Transitional Justice Research Working Paper Series, 20 August 2008, available at http://
www.csls.ox.ac.uk/otjr.php.
against a non-signatory state.33 The African marginality that accounts for the
heavy concentration of ICC effort there also helps explain why no greater
international political investment was made to halt the crimes now under inves-
tigation while they were in progress.
It is at the nexus of punishment and intervention that the ICC has arguably
exceeded the boundaries of legitimate behaviour for a legal institution. With some of
the indictments in the case of Sudan, we see a potential conflation of the different
preserves of intervention and legal redress. This is one step further along a road taken
over Rwanda and, in the first instance, Yugoslavia, when legal proceedings were
instituted in the midst of ongoing atrocity, not as a complement to interventionist
action during it, but as a substitute for such action.34 In the Sudanese case, the
indictment of, inter alia, President Omar al-Bashir may prove an obstacle to negotia-
tions over Darfur (and a destabilizing factor in southern Sudan) if the OTP pursues
him irrespective of political consequences, or it may result in investigations being
suspended as part of the bargaining process. In the first instance justice will come at
the cost of possibly prolonging conflict (though in any case the OTP would be reliant
in arresting Bashir and other military or political forces over which it itself has no
direct influence). In the second instance justice will itself come to be seen as a
disposable and explicitly political tool, thus undermining the universalistic claims
of the jurists and threatening the legitimacy of the ICC.35
C O N C LU S I O N : A B A L A N C E
................................................................................................................
Criminal trials are intended to punish crime. Such punishment has classically
been justified in one of three ways, as retribution, as a means for preventing the
33 For a defence of the ICC against the neo-imperialism charges, but one that does not address the
concerns raised in this chapter, see Max du Plessis, ‘The International Criminal Court and Its Work in
Africa: Confronting the Myths’, Institute for Strategic Studies paper 173 (November 2008). See p. 11 on
DRC’s and Uganda’s attempt to use the court for political ends. See also Louisa Lombard, ‘Justice for
Whom? The ICC in the Central African Republic’, on the SSRC Blogs: http://www.ssrc.org/blogs/
darfur.
34 Rachel Kerr, ‘The Road from Dayton to Brussels? The International Criminal Tribunal for the
Former Yugoslavia and the Politics of War Crimes in Bosnia’, European Security 14:3 (2005), 319–37,
here 325, and the material cited there in note 41.
35 Chidi Odinkalu, ‘What if Ocampo Indicts Bashir? 2’; Alex de Waal, ‘Africa’s Challenge to the
ICC’; Heather Adams, ‘Putting the Cart before the Horse’, all on SSRC Blogs: http://www.ssrc.org/
blogs/darfur; Stephen Oola, ‘Bashir and the ICC: The Aura or Audition of International Justice in
Africa’, Oxford Transitional Justice Research Working Paper Series, 15 October 2008 at http://www.csls.
ox.ac.uk/otjr.php. On the issue generally, Steven R. Ratner and Jason S. Abrams, Accountability for
Human Rights Atrocities in International Law (Oxford: Oxford University Press, 2001), 224–5.
perpetrator from committing similar crimes again in future, and as a way of

deterring other potential offenders from engaging in similar crimes themselves. In
addition, trials for genocide and crimes against humanity have often been justified as
forms of political and moral pedagogy.36 In the end, though, none of these justifica-
tions make much sense when applied to genocide.
Retribution, as articulated most clearly by Immanuel Kant, argues that one
punishes the criminal in order to restore the moral balance his or her crime has
upset. Punishment is retribution, meted out by a court, according to a ‘principle of
equality’ so that ‘whatever undeserved evil you inflict upon another within the
people, that you inflict upon yourself.’37 Proportionality is the key element in this
analysis. There are two problems with retributive justice as applied to genocide.
First, retribution is an inherently individualizing approach to punishment. Yet
genocide and crimes against humanity are, by their nature, systematic, mass crimes
in which individual perpetrators operate within broad institutional or social
frameworks. Consequently, the individualized moral claim at the heart of retribu-
tive justice tends to miss its mark. Second, it is difficult to conceive of a propor-
tional retribution for genocide. After all, the most one can do is execute an
individual perpetrator, which is hardly proportional to the thousands of murders
for which he may be responsible. Indeed, pushed to its logical conclusion, pro-
portionality would seem to require reciprocal genocide, which would clearly
violate the principle of individual guilt at the heart of retributive justice, rendering
such an approach internally incoherent.
Special prevention assumes that one punishes a criminal in order to prevent
recidivism. Given that genocide and crimes against humanity are invariably politi-
cal crimes, the risk of recidivism by individual perpetrators would depend entirely
on political circumstances. Eliminate the conditions (the perpetrator state, for
example, or the context of civil war) and one eliminates the chance of recidivism.
In such circumstances, there would be no need to punish individual perpetrators at
all. It might be claimed, as a variant of the special prevention argument, that
indicting leaders for crimes still in progress might deter them from continuing to
commit such offences. At best, the jury is still out on this claim, as in some of the
African cases considered here. It is so far unclear whether indictments have had any
impact on the behaviour of the political leadership of perpetrator states and, to the
extent it has had an impact, whether that impact has been at all beneficial. It might,
in fact, prove to be counterproductive, spurring leaders to greater efforts to
complete their genocide, since they no longer have anything to lose, as was
arguably the case with Milošević.
36 Mark Osiel, Mass Atrocity, Collective Memory, and the Law (New Brunswick, NJ: Transaction,
1997).
37 Immanuel Kant, The Metaphysics of Morals (Cambridge: Cambridge University Press, 1996), 105.
General prevention seems in many ways to be the most plausible explanation for
why we should punish génocidaires. The claim is that punishing perpetrators
deters future genocides. To begin with, there is no empirical evidence that this
works and good reason to assume that it does not. The Serbian state, for instance,
committed crimes against humanity in Kosovo long after the establishment of the
ICTY; the existence of the ICTR in Tanzania did nothing to deter mass criminality
in neighbouring DRC. It is not hard to see why this should be the case. Deterrence
works, to the extent that it does, by raising the potential cost to the criminal of his
or her crime, such that the anticipated benefit is no longer worth the risk
involved.38 Yet in some cases, genocide is, in Weberian terms, a value rational
act, not an instrumentally rational one.39 This means that questions of costs and
benefits are explicitly excluded from consideration. What matters is the normative
consistency of the act, not its anticipated success or failure. Consequently, raising
the costs of acting in the given manner can have no impact whatsoever. In other
cases, genocide can be triggered by a heightened sense of crisis and a paranoid
evaluation of the threat posed by the victim group. In this situation, the cost of the
failure to act, i.e., to commit genocide, is perceived by the perpetrating regime to be
infinitely higher than any punishment that might be imposed in the event of failure
and defeat. Besides, the sheer historical inconsistency of punishment for genocide
removes the most important precondition for deterrence: relative certainty of
punishment.40 In any given genocide, perpetrators often number in the tens or
even hundreds of thousands, while the numbers convicted remain at best in the
hundreds or thousands; and, of course there are many genocides none of whose
perpetrators ever reach trial.
The determination of which genocides or cases of large-scale crimes against
humanity are prosecuted and which are ignored is largely a geopolitical one, as the
proliferation of ICC cases against politically relatively marginal states in Africa
shows. Elsewhere, given great power support for many genocidal regimes over the
past 50 years, there can be a direct continuity between the political context which
led to genocide and the subsequent impunity for génocidaires themselves, as for
instance in Cambodia. It seems highly unlikely that there will be any indictments
brought against perpetrators from close strategic allies of the United States as the
38 John J. Donohue and Justin Wolfers, ‘Uses and Abuses of Empirical Evidence in the Death
Penalty Debate’, Stanford Law Review 58 (December 2005), 791–846; and Richard Berk, ‘New Claims
about Execution and General Deterrence: Déjà vu All over Again’, Journal of Empirical Legal Studies
2 (July 2005), 303–30.
39 Max Weber, Economy and Society: An Outline of Interpretive Sociology (Berkeley: University of
California Press, 1968), i.24–6.
40 William C. Bailey, J. David Martin, and Louis N. Gray, ‘Crime and Deterrence: A Correlation
Analysis’, Journal of Research in Crime and Deliquency 11 (1974), 124–43. Erling Eide in cooperation
with Jorgen Aasness and Terje Skjerp, Economics of Crime: Deterrence and the Rational Offender
(Amsterdam: North-Holland, 1994).
world’s sole superpower. And even if such indictments were brought, the chances
of them resulting in trials would be even slimmer. Absurd as it may be, the
American president has congressional authorization to invade the Netherlands to
rescue any Americans charged at The Hague should this become necessary.
Ours is not a purely materialist interpretation of international law. The very
force of the label ‘genocide’ as mobilization slogan and condemnation illustrates,
if only in the vigorous efforts of states like Turkey and Sudan to avoid its
application, that the language of values can have real currency in the global
arena. Underlying structural factors of the international political economy and
geopolitical strategy do obviously come strongly into play in decisions as to who
reaches trial, but the constitutions of internationally mandated courts, and the
procedures in individual court cases, are generally not functions of those struc-
tural factors. Otherwise, as in the case of the American-dominated trial of
Saddam Hussein, assiduously hived-off from the control of the United Nations,
legal legitimacy would be totally lost. That legitimacy has been maintained is a
credit to the committed individuals and organizations involved in such institu-
tions as the ICTY, the ICTR, and the ICC, and a testament to the existence of an
organized international value community of some sort beyond the international
power system constituted by the world’s most powerful states and multinational
corporations. To what extent this legal community can prevail over the system in
cases where their interests are antipathetic remains—to take the most optimistic
assessment—an open question.
Even in conflicts which have been subject to adjudication, the problems of
equitable prosecution have been enormous, in particular as a result of the post-
genocidal power relations, both domestic and international. Cases where atrocities
have been committed by all sides to a conflict, however unevenly, are particularly
difficult in this regard. There may be prosecutions in which not all parties to a
conflict involving multilateral atrocities are prosecuted, as in Rwanda or Uganda,
and even—despite the extensive efforts of the ICC—the DRC.41 Alternatively,
different problems may arise as for instance with the ICTY’s genuine efforts at
proportionality in prosecution. The evidence seems to indicate that the ICTY’s
efforts have succeeded mainly in generating broad resentment against the court,
with all ethnic parties feeling that the court is biased against them and overly
lenient towards their adversaries.42
41 Lombard, ‘Justice for Whom?’; Lisa Clifford, ‘ICC Risks Losing the Plot in Congo’, Institute for
War and Peace Reporting Comment, 21 November 2008, at http://www.iwpr.net
42 Eric Stover and Harvey M. Weinstein (eds), My Neighbor, My Enemy: Justice and Community in
the Aftermath of Mass Atrocity (Cambridge: Cambridge University Press. 2004); A. Uzelac, ‘Hague
Prosecutors Rest Their Case’, Institute for War and Peace Reporting (27 December 2004); relatedly,
Human Rights Watch, ‘Justice at Risk: War Crimes Trials in Croatia, Bosnia and Herzegovina, and
Serbia and Montenegro’, Human Rights Watch 16:7 (2004), 1–31.
Finally, trials have been justified for their purported value as sites of moral and
political pedagogy, particularly in the context of so-called democratic transitions.
Unlike the other three justifications, which amalgamate trials for mass atrocity
to ordinary criminal law, this argument highlights the specificity of such trials as
elements in transitional justice.
The argument is that such trials can be effective means for establishing the
history of past atrocities and thereby delegitimizing the perpetrator regime.
While it is true that criminal prosecution can be highly effective at gathering
evidence, it is less effective at marshalling this evidence to construct a coherent
and accurate narrative. All too often the history lessons taught by criminal trials
are distorted and misleading.43 This is not due to any malice or incompetence on
the part of the court but rather a result of the quite distinct methods and
objectives of judges and historians.44 Trials are not designed so much to ascertain
what happened and why, as to determine who is to blame. By definition, trials
are less interested in historical processes than they are in concrete manifestations of
criminal intent; their goal is to establish individual guilt, not historical causation. As
a consequence of this necessary focus on the individual in the dock, the remainder of
the perpetrator polity can, with some justification, feel themselves exculpated.
Unintentionally, then, such trials frequently end up serving as alibis for that majority
of perpetrators and bystanders not prosecuted. Trials thus manifest an irreconcilable
tension between any general pedagogical impulse and the individuating character of
criminal justice.
However, even if trials unavoidably tend to teach inadequate history lessons, this
in itself does not mean that they could not contribute to transitional justice. The
form may be more important than the content. Mark Osiel contends that criminal
trials are especially effective venues for ‘stimulat[ing] public discussion in ways that
foster the liberal virtues of toleration, moderation, and civil respect’.45 Because such
trials operate under the ground rules of liberal legalism (individual culpability, due
process, fair defence, etc.), they perforce validate a pluralistic debate about the
meaning of past atrocities, what Osiel terms ‘civil dissensus’. According to this
view, the benefit of the politico-legal system facilitating such a pluralism of views
and the testing of one against the other, without imposing either in an authoritarian
or unquestioning fashion, would become increasingly self-evident. As a consequence,
the population will then come to embrace a liberal democratic polity.
43 Devin O. Pendas, The Frankfurt Auschwitz Trial, 1963–1965: Genocide, History and the Limits of
the Law (Cambridge: Cambridge University Press, 2005); Bloxham, Genocide on Trial.
44 Carlo Ginzburg, The Judge and the Historian: Marginal Notes on a Late-Twentieth-Century
Miscarriage of Justice (London: Verso, 1999); and Norbert Frei, Dirk van Laak, and Michael Stolleis
(eds), Geschichte vor Gericht: Historiker, Richter und die Suche nach Gerechtigkeit (Munich: Beck,
2000).
45 Osiel, Mass Atrocity, 2.
While theoretically appealing, the empirical evidence hardly supports this

hypothesis, as the highly partisan responses to the ICTY show. Osiel’s problem is
ultimately one of scale. Such trials are unlikely to have this kind of liberalizing
impact within post-conflict societies themselves, where they will if anything tend
to harden, rather than ameliorate, the boundaries between conflicting parties.
The exculpatory capacity of such trials for the unindicted can reinforce the sense
of victimization, even among perpetrator groups (as, e.g., in Serbia or post-war
Japan).
If trials are unlikely to deter genocide or democratize post-conflict societies, this
does not mean that they can do no good in the world whatsoever. Above all,
prosecuting genocide and crimes against humanity is a statement of principle, an
act of symbolic disapproval on the part of the international community. This may
have limited significance for the parties to specific conflicts, and can even be
counterproductive for democratization, but it is important to the international
community itself. It is an aspirational statement about what we hope the interna-
tional community can become. To remain silent in the face of genocide would be to
tacitly approve it, as is all too often the case at the moment. By extension,
prosecuting genocide can be a useful instrument for forging an emerging consen-
sus regarding international norms. Such trials pose in insistent terms the question
of what is right and they offer answers, however modest, which can be incorporated
into international political conversations. Such an emergent consensus can be
discerned, again in a modest way, in the relatively wide adoption of the Rome
Statute of the ICC, keeping in mind again that the most important international
powers have thus far refused to join the court. Still, that so many countries are now
ready to adopt a standing international court, when a few short decades ago almost
none were, is a sign of at least limited progress. Of course, the legalist paradigm is
all too easily subject to manipulation. This is why it is important for genuinely
independent NGOs, media outlets, and ordinary citizens to monitor the use and
abuse of international legal norms. International criminal law is too important to
be left to lawyers alone, whose claims to embody a universal class of disinterested
humanitarians evinces an all too evident partiality. Likewise, if international
criminal law is to function at all, even in an aspirational vein, it must not be left
to the sole discretion of politicians, who are the very incarnation of sectional and
sectarian interest. The norms of international criminal law must be incorporated
into an emerging global civil society, the inchoate and fragile international value
community that nevertheless offers the most realistic hope we have for a somewhat
more humane future.
If supporting the norms embodied in international law is important, expecting
too much from the institutions of international law is a potentially dangerous self-
deceit, and one encouraged by the bold claims made by prominent observers of and
participants in the ad hoc tribunals and the ICC. When the institutions of law
trespass into the domain of active intervention in ongoing conflicts involving
genocide or crimes against humanity, they can, on one hand, introduce an inflexi-
ble element into a situation in which diplomatic flexibility is of the essence, or they
can, on the other hand, become overtly politicized and thus delegitimized. Law
courts should only serve as part of the means of negotiating the aftermath of crises.
Law talk should not claim more than it can achieve in the very messy and
quintessentially political world of conflict resolution and regime transition.
Bloxham, Donald, Genocide on Trial: War Crimes Trials and the Formation of Holocaust
History and Memory (Oxford: Oxford University Press, 2001).
Cooper, John, Raphael Lemkin and the Struggle for the Genocide Convention (Basingstoke:
Palgrave MacMillan, 2008).
Earl, Hilary, The Nuremberg SS-Einsatzgruppen Trial, 1945–1958: Atrocity, Law, and History
(Cambridge: Cambridge University Press, 2009).
Hagan, John, Justice in the Balkans: Prosecuting War Crimes in the Hague Tribunal (Chicago:
Hazan, Pierre, Justice in a Time of War: The True Story behind the International Criminal
Tribunal for the Former Yugoslavia (College Station: Texas A&M, 2004).
Maguire, Peter, Law and War: An American Story (New York: Columbia University
Press, 2001).
Osiel, Mark, Mass Atrocity, Collective Memory, and the Law (New Brunswick, NJ:
Transaction, 1997).
Pendas, Devin, The Frankfurt Auschwitz Trial, 1963–1965: Genocide, History and the Limits
of the Law (Cambridge: Cambridge University Press, 2005).
Schabas, William A., Genocide in International Law: The Crime of Crimes (Cambridge:
Simpson, Gerry, Law, War and Crime (Cambridge: Cambridge University Press, 2007).
chapter 31
.............................................................................................
F RO M PAS T TO
F U T U R E*
PROSPECTS FOR GENOCIDE
A N D I T S AVO I D A N C E I N T H E
T W E N T Y- F I R S T C E N T U R Y
.............................................................................................
mark levene
................................................................................................................
Severe problems of overpopulation, environmental impact, and climate
change cannot persist indefinitely: sooner or later they are likely to resolve
themselves, whether in the manner of Rwanda or in some other manner not
of our devising, if we don’t succeed in solving them by our own actions.1
How should we understand the wellsprings of genocide? The above statement

could be read as either a list of potential ingredients or a line of explanatory inquiry
at marked variance with nearly all standard treatments of our subject. Indeed, from
* A longer version of this piece entitled ‘Predicting Genocide in an Age of Anthropogenic Climate
Change: An Interim Report,’ can be found on the Crisis Forum website at http://www.crisis-forum.
org.uk/events/workshop1_resources.php.
1 Jared Diamond, Collapse: How Societies Choose to Fail or Survive (London: Penguin, 2005), 328.
future prospects for genocide 639
Lemkin onwards, most genocide scholars have been at pains to distance the
phenomenon, at least in its contemporary guise, from any explanation of a general-
izing kind. To travel down that road would be to diffuse ‘genocide’ into something
wholly more amorphous. Even in so far as it is clearly a matter of violence,
inclusion of any particular case history as genocide rests on the fulfilment of
criteria that mark it as only belonging to that special category of violence. Thus
to speak of an event as genocide is almost ipso facto to repudiate the possibility that
it might have been shaped or determined by factors or circumstances associated
with the politics, economics, or social or cultural behaviour of dominant interna-
tional society. In contrast, genocide is almost always assumed to mark a radical
rupture with, or from its norms. It is aberrant; abnormal; the outcome of sad,
malfunctioning polities, usually led by seriously mad or bad leaders.
Here I propose an alternative approach, briefly stated, as dependent on underlying
but systemic preconditions broadly common to crises of state and out of which
genocide has regularly emanated. Their historical roots are in some respects quite
straightforward. The avant-garde model of the coherent nation-state developed in a
limited number of early modern polities in Western Europe and then North America
in tandem with efforts to achieve the maximization of their resource potential—
human, biotic, and material—as determined by the needs of an almost perpetual
military competition or actual warfare between these polities. It was no accident that
the states that were most successful in this competition were not only the most
technologically innovative but also the most predatory in their efforts to develop
and utilize their respective resource bases for the capital accumulation necessary in
turn to feed that technological advance. Asset-stripping corporate capitalism, state
formation—or reformulation—and military revolution, though coming through
various, often unrelated pathways, thus coalesced in the late-eighteenth-century
West as a potent nexus of all three. The paradigm also necessarily carried its own
dynamic logic, the shorthand for which we might read in social Darwinian terms not
so much as the survival of the fittest but rather the survival of the fastest.
Thus, we have the protean beginnings of what one historian has dubbed the
‘Great Acceleration’ towards the contemporary globalized political economy.2
To make good, or perhaps more soberly put, simply to stay afloat in a world as
determined by the new Western dispensation, required emulation of its practice.
The alternative was to go under, that is, to be colonized. Even with the later shift
after 1945, to the post-colonial framework in which all formerly Western colonized
zones nominally became sovereign and independent entities, the urge to hothouse,
preferably industrial development became the sine qua non of each and every one,
to the point where ‘advocacy of anything short of maximum economic growth
2 J. R. McNeill, ‘Social, Economic and Political Forces in Environmental Change: Decadal Scale
(1900 to 2000)’, in Robert Constanza et al. (eds), Sustainability or Collapse: An Integrated History and
Future of People on Earth (Cambridge, MA: Dahlem University Press, 2007), 301–29.
640 mark levene
came to seem a form of lunacy or treason.’3 This did not fundamentally shift the
balance of geopolitical and economic power away from the metropoli, at least not
until quite recently. In contrast, it simply intensified the urge of more self-
consciously aware and resentful latecomer states within the periphery and semi-periph-
ery to seek their own short cuts to catch-up.
In an earlier piece exploring the likely contours of violence in the near-future of
our contemporary world, I proposed a three-tier schema with some passing
reference to this model as proffered by Immanuel Wallerstein.4 A first tier consisted
of wealthy First World countries (the ‘liberal West’) closely approximating what
under the era of bipolarity was also referred to as the ‘free world’. A second tier was
made up of the vast majority of modern nation-states, not only those in the former
Soviet bloc but polities in all hemispheres who continued to see themselves as bona
fide players in the international system competition for position and power. A third
tier was posited as more notional than real. Nevertheless, it was based on the
argument that some of the very poorest, weakest, and most underdeveloped
countries who had entered into forced-pace, usually state-driven modernization
to meet the institutional demands of the system were already so broken by the
challenge that it was only a matter of time and/or the termination of tier one
(‘international’) aid before they ceased to operate as effective, infrastructurally
cohesive states altogether. Suggested candidates for this unfortunate grouping
included ‘much, if not all, of sub-Saharan Africa, as well as possibly large chunks
of Central Asia’.5
The further implication of this schema was that potential trajectories, patterns,
and ultimately forms of violence were specific to each tier. In tier one, for instance,
it was posited that while these state-societies were directly or indirectly responsible,
or at least complicit for much of the conflict or threat of conflict, including
genocide in the world at large, they were largely insulated themselves from
suffering extreme, mass violence within their own domestic contexts. In tier
three, by contrast, the actuality or likelihood of violence was endemic and rampant,
yet, paradoxically, was insufficient in the way of state authority to scotch or at least
put a brake on its widespread but diffused prevalence and persistence.
3 J. R. McNeill, ‘Social, Economic and Political Forces in Environmental Change: Decadal Scale
(1900 to 2000)’, in Robert Constanza et al. (eds), Sustainability or Collapse: An Integrated History and
Future of People on Earth (Cambridge, MA: Dahlem University Press, 2007), 302.
4 See Immanuel Wallerstein, The Capitalist World-Economy (Cambridge: Cambridge University
Press, 1979), and The Modern World System, 3 vols (San Diego/New York: Academic Press, 1974–88),
for the Wallersteinian system. Also Mark Levene, Genocide in the Age of the Nation State, vol. i: The
Meaning of Genocide (London/New York: I. B. Tauris, 2005), ch. 3, for a fuller rendition of the
argument herein.
5 Mark Levene, ‘Connecting Threads: Rwanda, the Holocaust and the Pattern of Contemporary
Genocide’, in Roger W. Smith (ed.), Genocide: Essays toward Understanding, Early Warning and
Prevention (Williamsburg, VA: Association of Genocide Scholars, 1999), especially 46–9.
It was thus in tier two that the preconditions of modern genocide were at their
greatest: on the grounds that it was precisely in these states that the driving forces
associated with the developmentalist imperatives of the system were at their most
intense and urgent. However, as this tier two grouping ranged from large, relatively
strong states such as China or Russia at one end, to small, relatively weak ones such
as Rwanda and Burundi, at the other, Wallerstein’s distinction between semi-
periphery and periphery proved ultimately less apposite to the argument than
understanding the crises of states out of which the actual conditions for genocide
emanated. If, however, the overriding implication of the 1999 article was that both
the preconditions and conditions of modern genocide remained closely inter-
twined with the very driving forces, not to say building blocks of our contemporary
global system, some ten years on, the purpose of this chapter is to reconsider the
argument in the light of environmental evidence which now puts the long-term
sustainability of the entirety of the international system in serious doubt.
It has been known for some decades that the scope, scale, and relentlessly
accelerating pace of developmentalism is entirely out of synch with the carrying
capacity of the planet.6 Now, with the full effects of that developmentalism self-
evident in terms of the knock-on consequences of greenhouse gas emissions
(GHG) on the biosphere, one might even propose that the appropriate question
is not so much about whether there will be future genocide but whether there will
be future generations of Homo sapiens upon this planet at all.7
If this of its own might be grounds for deciding that the study of our subject is
facing redundancy we have already hinted at why ongoing predictive analysis could
be of value to the greater cause of humanity’s survival. If the growing scope, scale,
and frequency of genocidal events in the most recent centennial sequence is itself an
indicator of the cul-de-sac nature of systemic drives towards the unattainable, we
might expect the acceleration of those drives set against increasing environmental
blockages—not least global warming—to be an equally strong indicator of where
we are more generally heading. By the same token, if the scale of biospheric
breakdown actually begins to unravel the statist project, then we might expect to
see the specific path of genocide radically diffuse or possibly metamorphose into
other forms of violence.
To be sure, making prognostications about the future is to enter onto dangerous
terrain. That said, developing scenarios for future climate change impacts as set
against different levels of GHG emissions has become practically a staple of climate
6 Donatella Meadows et al., The Limits to Growth (New York: Universe, 1972), for the classic study.
7 See WWF report, Climate Change, Faster, Stronger, Sooner (London: WWF, 2008), available at
http://assets.wwf.org.uk/downloads/cc_science_paper_october_2008_1.pdf. Also David Wasdell,
‘Radiative Forcing and Climate Sensitivity’, initially prepared for the Tällberg Consensus Project,
‘The Tipping Points We Cannot Cross: Defining the Boundary Conditions for Planetary
Sustainability’, 25–26 June 2008, revised and expanded 10 December 2008. Available at http://www.
jimhadams.com/eco/RadiativeForcingEd3.doc
642 mark levene
and earth science modelling. What is needed now, however, is a broader contex-
tualization of genocidal potentialities that take into account the genuine environ-
mental, including climatic factors.
We have sought to develop this analysis—albeit only in the most sparse out-
line—by offering two new routes into the future. Both are necessarily grounded in
realities of the present. Both hold fast to the three-tier approach. However, if the
earlier exposition might be characterized as one of ‘business as usual’, that is, in
which genocide continues to be a symptom of systemic dysfunctionality but in
which political and economic factors are assumed to be paramount, what we seek
to do in our additional Scenario 1 is work up the argument by suggesting how
resource scarcity (linked to the ongoing demands of the global economic system)
plus population pressures are creating a matrix of destabilizing forces in their own
right. Our forecast, as previously, implies that genocide could be one of several
possible outcomes in terms of extreme, mass violence. Indeed, our focus on the
Great Lakes region of Central Africa, more especially the eastern Congo, contains
within it the proposition that tier-three conditions are particularly indicative of
what we describe as post-genocidal conflict.
Does this mean paradoxically, that genocide qua genocide could be on the wane?
As a prelude to Scenario 2—in which we introduce the true elephant into the room:
anthropogenic climate change—we offer the briefest of commentaries on the case
of the Chittagong Hill Tracts (CHT). This is a region where environmental
breakdown, while intermeshed with other more standard factors, can already be
seen to be symptomatic of a descent into genocidal conflict. The key point about
the pursuit of this theme in Scenario 2, however, is that the disruptive potential of
climate change, whether writ small in terms of the single state, or writ large in
terms of the international system, is entirely exponential. All the more reason why
it cannot be ignored by genocide scholars, nor anybody else. Whether climate
change will simply be a ‘threat multiplier’ to already existing conflicts—as security
analysts now repeatedly tout8—or the key factor in a civilizational collapse, only
time will tell. In our concluding remarks, we briefly iterate the current direction of
flow towards ever greater violence, as a consequence of the perpetuation and/or
intensification of present conditions. Gazing into this crystal ball, however, will not
clarify whether genocide will be a major facet of this ravaged landscape. It will
simply confirm the urgent necessity for a paradigmatic shift in our relationship not
only to each other but to our precious planet if we are to avoid not simply genocide
but omnicide.
8 CNA, National Security and the Threat of Climate Change (VA: CNA Corporation, 2007), http://
www.securityandclimate.cna.org/
S C E NA R I O 1: B U S I N E S S A S U S UA L A S S E T AG A I N S T
T H E C A R RY I N G C A PAC I T Y O F T H E P L A N E T
................................................................................................................
Back in 1972, a small team of far-sighted, US-based systems analysts produced a
report for the Club of Rome on future prospects for humanity. They did so by
extrapolating available data, particularly on industrialization, food production,
pollution, and demographic patterns, as set against the carrying resource capacity
of the planet. Their conclusion was stark: exponential growth would lead to
ecological overshoot, the consequence of which would be ‘a rather sudden and
uncontrollable decline’ within a time frame of one hundred years. Limits to
Growth was a landmark event and so duly received a barrage of criticism from
mainstream policy-makers and academics.9 More than thirty years on, however,
leading scientific report after report corroborates the fundamental contours of the
team’s findings. The Millennium Ecosystem Assessment in 2005, for instance,
concluded:
Over the past fifty years, human beings have changed ecosystems more rapidly and
extensively than in any comparable period of time in human history, primarily to meet
rapidly growing demands for freshwater, timber, fiber, and fuel. This has resulted in a
substantial and largely irreversible loss in the diversity of life on Earth.10
More recently, in late 2007, a report from the UN Environment Program

represented simply one more authoritative voice iterating that the planet’s water,
land, air, plants, animals, and fish stocks were all in ‘inexorable decline’.11 Mean-
while, a new generation of ‘ecological footprint’ scientists are setting out, with a
degree of mathematical precision, the gap between the current demands of the
human Oikumene and the limits of planetary supply. One leading figure, for
instance, calculates that while in practice the Earth can offer 1.8 hectares of
cropland, pasture, forest, and fishing ground to each of us, what we are on average
consuming amounts to 2.2. hectares. More sobering still, the Earth’s ability to
regenerate its resources is taking some fifteen months against what we are using up
in twelve. Again, the picture is abundantly clear: our current globalized political
economy as it developed out of a particular but relatively recent historical
9 See Dennis L. Meadows, ‘Evaluating Past Forecasts, Reflections on One Critique of The Limits to
Growth’, in Constanza et al., Sustainability, 399–415, for a more recent assessment.
10 Millennium Ecosystem Assessment (MEA) 2005, extract reprinted in Nathan J. Mantua, ‘A
Decadal Chronology of 20th Century Changes in Earth’s Natural Systems’, in Constanza et al.,
Sustainability, 292.
11 See John Vidal, ‘Global Food Crisis Looms as Climate Change and Fuel Shortages Bite’,
Guardian, 3 November 2007.
644 mark levene
trajectory is radically at odds with nature’s bounty with the consequence that
‘overshoot will ultimately liquidate the planet’s ecological assets’.12
The $64,000 question for us is what does all this mean in terms of human, more
exactly social and political consequences? It should not be rocket science to deduce
that as environmental stress on the human condition sets in and, with it, loss of
control over what previously had been assumed to be normal and predictable,
something—or things—will have to give, with likely violent repercussions. But that
still poses the questions where, when, and how? This, however, hardly needs to be a
matter of future forecasting. If the scientific pronouncements are correct, then
there should be enough evidence in the recent or present-day record to confirm the
relationship between environmental pressures and forms of conflict.
However, while the relationship between Third World population increase and
environmental stress is the standard point of access into this subject the greatest
destroyers of planetary resource in overall global terms are not the poor at all, but
the rich. The average Briton burns up more fossil fuels in a day than a Tanzanian
family uses in an entire year. Indeed, if we were to make further striking compar-
isons, if everyone’s ecological footprint were European we would need 2.1 planet
Earths to sustain us, while if we all followed the US lead, we would need nearly
five.13 But if more needs to be said below about the localized causes of extreme
violence, including genocide in the Third World—again both poor tier two and tier
three—countries, let us just for one moment run with the implications of still
hegemonic tier one efforts to continue a maximized control of Third World
mineral and energy supply against the backdrop of an increasingly undisputed
resource scarcity. Here, for instance, is a report extract from defence analysts
working under the British Ministry of Defence on a possible near-future scenario
for Africa:
Climate change and HIV/AIDS, scarcity of food and water and regional conflict could lead
to Africa becoming a failed continent, where even large, currently self-sustaining states
become chaotic. Outside engagement and intervention would effectively be limited to a
small-number of well-defended entry points and corridors, which would provide access to
raw materials essential to the global economy. Nations or corporations wishing to trade
with Africa would increasingly be required to provide security for their nationals and the
necessary support to sustain critical areas of access and security.14
12 Mathis Wackernagel, quoted in Fred Pearce, Confessions of an Eco-Sinner: Travels to Find Where
My Stuff Comes From (London: Eden Project Books, 2008), 315.
13 Johann Hari, ‘Don’t Call It Climate Change—It’s Chaos’, Independent, 15 November 2005; ‘World
Economy Giving Less to Poorest in Spite of Global Poverty Campaign Says New Research’, 23 January
2006; http://www.neweconomics,org/gen/news.growthisntworking.aspx
14 From DCDC (Development, Concepts, and Doctrines Centre, MOD), ‘Strategic Trends, 2030’,
quoted in Nick Mabey, Delivering Climate Security, International Security Responses to a Climate
Changed World, Whitehall Papers, 69 (London: Royal United Services Institute, 2008), 31.
What is particularly valuable about this assessment is its remarkably frank and,
one might add, naked assertion of the primacy of the national interest. Africa
matters because it has mineral as well as fossil fuel resources. The bottom line,
hence, is that under conditions of instability, Britain must exert maximum
political-cum-military leverage to recover these for herself, and by implication,
prevent other ‘unfriendly’ predators from squeezing her out. The language is
redolent of the nineteenth-century scramble for Africa, some of the consequences
of which were genocidal. More to the point, if this can be taken to be the genuine
bottom line of ongoing British foreign policy,15 it casts a disturbing commentary
on African conflicts in which resource issues have played a prominent role.
Take the most obvious and glaring example; the ongoing conflict in eastern
Democratic Republic of Congo (DRC), more exactly centred on Ituri and North
and South Kivu. The immediate trigger to destabilization was the political crisis,
culminating in the 1994 genocide, in neighbouring Rwanda. Massacre-led interven-
tion in DRC’s east, by the new Rwandese Patriotic Front (RPF) government, initially
against the fleeing Hutu militias, quickly catalyzed a much wider set of military
interventions involving half a dozen additional African states. The primary goad to
each was not geostrategic but venal, that is, to use the opportunity of DRC’s internal
breakdown to maximize their own access to, and exploitation of, its mineral and
natural largesse. They did so by seeking concessions from the failing Kinshasa
government (Mobutu, then Kabila) in return for military support; in the case of
Uganda and Rwanda this was done by a degree of direct intervention, though more
especially in the east, where the competition was greatest, through the military
backing of what became a multiplicity of warlord-led militias.16
So far, one might ask, what has any of this to do with Western involvement or
complicity? The answer is that we are speaking here about a range of resources,
including copper, cobalt, cassiterite (tin oxide), gold, and diamonds, whose value
to their African interlopers only existed if they could be traded for foreign currency,
in other words through purchasing intermediaries willing not to ask difficult
questions about the minerals’ sourcing. This, we must remember, against a late-
1990s surge of market price as industrial demand for minerals in leading developed
countries—including now India and China—began rapidly to outstrip supply.
In eastern Congo, the interrelationship between these diverse factors and the
potential for an exponential violence began to hinge on the mineral compound
coltan. The compound includes the precious metal tantalum much in demand as a
conductor in hi-tech communications and aerospace industries (in other words,
primarily for military purposes), but also for making capacitors in a range of
15 See Mark Curtis, Web of Deceit: Britain’s Real Role in the World (London: Vintage, 2003), esp.
ch. 10, for the necessary confirmation.
16 See Thomas Turner, The Congo Wars: Conflict, Myth and Reality (London/New York: Zed Books,
2007), for background.
646 mark levene
electric devices—computers, Play Stations, digital computers, and especially

mobile phones. Eighty per cent of the world’s reserves of coltan are located in
the Kivus. The Rwandan and Ugandan interlopers in DRC, acting through their
local proxies, thus happened to be sitting on a mineral whose market value, in
direct response to a rapid global take-off in mobile phone demand, went through
the roof in a matter of months, from $65 dollars in late 1999, to $530 in
mid-December 2001.17
We can rather too well state what happened to the region as a consequence.
The traditional, actually thriving pastoral-cum-agricultural economy collapsed as
all able-bodied men and boys scrambled to participate in constructing do-it-
yourself, ramshackle, inherently dangerous as well as highly toxic mines, in addi-
tion to those already overrun by the warlords. There was a growing incidence of
congenital deformities and respiratory problems as a result but with no health care
for the population to fall back on—no administration of course existing to pump
mine revenues back into social infrastructure—mortality from these illnesses
rapidly accelerated. But then, disease-related mortality increased across the
board, as coltan dependency linked to military competition for its control made
rapid inroads into the social cohesion and survivability of the region. With male
employment (including forced labour) all coltan-related—mostly in the mines or
the various militias numbering an estimated 200,000 combatants—young women
and children were sucked into this burgeoning alternative economy primarily as
prostitutes. The statement is shocking but only set against NGO estimates that 30
per cent of the region’s children were also succumbing to severe malnutrition; while
a staggering fifty per cent of the region’s population overall had been displaced.18
Here, then, was a society literally spiralling out of control, where not only was HIV/
Aids rife but previously contained diseases including whooping cough, measles,
even bubonic plague, part and parcel of an ever increasing cycle of degradation,
starvation and of course, atrocity. Indeed, this was exactly the sort of ‘in the midst
of Africa’ breakdown that our British defence analysts had warned against, the
critical caveat being that lack of food and environmental stress were hardly a
consequence of (a localized) resource scarcity but rather of the exact opposite.
All of this, of course, was entirely illegitimate to the UN eyes which, having set
up a panel of experts to investigate the illegal exploitation of DRC’s natural
17 See Mikolo Sofia and Dominic Johnson, ‘The Coltan Phenomenon’, Pole Institute/CREDAP
report, January 2002, available at http://www.odi.org.uk/HPG/papers/bkground_drc.pdf/; Pearce,
Confessions, 273–5, for additional background.
18 Amnesty International, ‘Democratic Republic of Congo “Our brother who help kill us”—
economic exploitation and human rights abuses in the east,’ AI INDEX AFR62/010. 2003 1 April
2003; Oxfam, ‘No End in Sight: The human tragedy of the conflict in the Democratic Republic of
Congo’, August 2001. Available at http://www.oxfam.org.uk/resources/policy/conflict_disasters/
downloads/bp12_drc.rtf
resources, demanded a moratorium on their trade.19 How, then, did countries such
as the UK respond? By duly ignoring or circumventing the UN’s panel request to
investigate the 18 British registered businesses held to be ‘deliberately or through
negligence’ among the 85 named Western companies helping to prolong the
conflict through their economic involvement.20 Nor did the UK freeze its substan-
tial aid programmes to Uganda and Rwanda. Why should it do so when Kigali and
Kampala’s foreign accounts were duly audited as clean?
If, then, this is an example of the practice of business of usual it has to be firmly
set against the carrying capacity of the planet, though now repeatedly involving tier
one states in major resource conflicts. The DRC may illustrate an example where
they have done so at second hand, but also underscores how conflict of this type
carries substantial economic gains which, consciously, or unconsciously, are
accepted by policy-makers as overriding the third world human cost.21 More
cynically, one might propose that because most of these conflicts do not fall within
a rubric of genocide, Western governments are all the better positioned to eschew
responsibility for them.
None of this should greatly surprise. A rising but resource-challenged China was
perfectly willing to give its full backing to the Sudanese government in the late
1990s as the latter focused its efforts on recovering control of major oil fields in its
long-standing genocidal war against the secessionists in southern Sudan—this, of
course, before the present climate-related conflict in Darfur. The US equally
provided covert counter-insurgency support to ensure the Nigerian government
maintained firmer control of its oil-rich delta region.22 In democratic countries,
such as Britain, the intermeshing of relationships between Private Military
Companies(PMCs) such as Executive Outcomes UK and the Canadian-owned
Heritage Oil (the latter the concessionary in a huge but highly dubious 3.1 million
hectares stake-out of Ituri), is known to those with a specialist watching brief but
not something anybody is going to contest in a court of law.23 Where tier one states
can leave corporate business to sort out their camouflaged, old-style mercantilist
methods of access to tier three African resource wealth, or that elsewhere, they will.
Of course, not all contemporary resource conflicts can be so easily packaged in
this corporatized way. Across the DRC border in Rwanda, recent studies have
19 UN Panel of Experts, Illegal Exploitation of Natural Resources and Other Forms of Wealth of the
DRC (UN: New York, 2001).
20 Terry Slavin, ‘DTI failing to act on Africa’s dirtiest war’, Guardian, 6 February 2005.
21 See David Keen, The Economic Functions of Violence in Civil Wars, Adelphi Papers 319, (Oxford:
Oxford University Press, and International Institute for Strategic Studies, 1998), for further
development of the argument.
22 See Doug Stokes, ‘Blood for oil? Global capital, counter-insurgency and the dual logic of
American energy security’, Review of International Studies, 33 (2007). 245–264.
23 Duncan Campbell: ‘Making a Killing: Marketing the New Dogs of War’, 11 July 2008 http://www.
craigmurray.org.uk/archives/2008/07/duncan_campbell.html for the PMC: Heritage connect and, by
extension, the wider world of corporate business.
648 mark levene
suggested that behind the overt Tutsi-Hutu ethnic-cum-political conflict was a

neo-Malthusian style crisis founded on the country’s rapidly burgeoning rural
population as set against a rapidly diminishing ecological resource base.24
Thus, the intense competition for land in Rwanda in the decades up to 1994 not
only produced tensions between land-owning ‘haves’ and ‘have nots,’ they also
drove the marginalized latter increasingly up the slopes of Rwanda’s famous hill-
sides. The further up they went the more they cut down the remaining forest, the
greater the erosion they caused. By 1990, an estimated 8,000 hectares per year
‘enough to annually feed about 40,00 people’ was being washed down the country’s
slopes. Arguably even worse, the rate at which the forest was being cut down and
consumed for fuel was outstripping its ability to regenerate itself by a factor of well
beyond two to one. In turn, that meant the peasants fell back on straw and other
crop residues for fuel, depriving the soil of its normal nutrient cycle.25 When it
came to the crunch in 1994, there were communes where there was intense grass-
roots bloodletting, yet few or no Tutsi among the victims.
There is, however, a point of interconnectedness between the deforestation that
occurred over a period of decades in Rwanda and what happened more rapidly to
much vaster stretches of DRC, as a result of foreign government, especially
Ugandan and Zimbabwean military-cum-corporate concessions, at the height of
the Congo conflict.26 Indeed, in overarching terms, these might be seen as two sides
of the same coin, one localized and demotically-driven, the other venal and
corporate, yet both of which, through the asset-stripping of one of the planet’s
basic ecological reserves are contributing to a planetary backlash that could well
herald what some have already dubbed ‘an anthropocene extinction event’.27 After
all, while on the one hand, tropical forests offer a major CO2 bio-sink mitigating
the effect of anthropogenic climate change, on the other, the ongoing and accel-
erating rate of their loss is estimated to be causing between a fifth and a quarter of
current global carbon emissions.28 Somewhere, in all this, are the people, both
indigenous and incomer, who live in and depend on the tropical forest. It is no
24 Catherine André and Jean-Philippe Platteau, ‘Land relations under unbearable stress: Rwanda
caught in the Malthusian trap,’ Journal of Economic Behaviour and Organisation, 34 (1998), 1–47;
Robert M. McNab and Abdul Latif Mohamed, ‘Human Capital, Natural Resources Scarcity and the
Rwandan genocide,’ Small Wars and Insurgencies 17, no. 3 (2006), 311–332. Also Diamond, Collapse, ch.
10, ‘Malthus in Africa: Rwanda’s genocide.’
25 James Gasana, ‘Remember Rwanda?’ People and Population Pressures report, 6 January 2003,’
http://www.peopleandplanet.net/doc.php%3Fid%3D1780/
26 UN Panel, ‘Illegal Exploitation’, 11–13; See also Patrick Alley, ‘Branching Out, Zimbabwe’s
resource colonialism in DRC,’ (London: Global Witness’ August 2001), http://www.globalwitness.org/
27 See David Wasdell, ‘Beyond the Tipping Point: Positive Feedback and the acceleration of climate
change,’ http://www.meridian.org.uk/Resources/ Global%20Dynamics/TippingPoint/index.htm
28 See both World Rainforest Movement (WRM) http://www.fern.org/pages/about/wrm.html/ and
Biofuelwatch, http://www.biofuelwatch.org.uk/background.php for regular updated information and
articles on this theme.
coincidence that some of the most intense conflicts of the here and now are
between those seeking to maximize its dead-end exploitation and commodification
for quick monetary gain and those who depend upon its sustainability for their
livelihoods and well being. Often lethal struggles between state-backed corpora-
tions and diminishing tribal groups over land and water rights have been part and
parcel of conflict in the Amazon basin for decades, particularly in recent years over
clearances to make way for export-orientated soya bean production. There are
similar processes unfolding in India and Borneo.29
But do any of these instances of what are often disparagingly referred to as ‘low-
level’ violence amount to genocide? And do they serve in any sense as indicators for
how climate change per se might impact on much broader elements of the world’s
populations who are not arboreal but agricultural, or urban? There is one case,
however, where recent historic experience combines with latent conditions of the
present to offer a potentially valuable insight into what could be an aspect of all our
futures: mass genocidal displacement.
F RO M PA S T TO F U T U R E : T H E C A S E
O F T H E C H I T TAG O N G H I L L T R AC T S
................................................................................................................
During the late 1970s and 1980s, efforts by the newly formed state of Bangladesh to
comprehensively integrate, consolidate and develop its sylvan eastern hill region
led to the intensification in an already long-sustained campaign of military-led
terror and violence against its then estimated 700,000 indigenous peoples, collec-
tively known as the jumma. Some NGOs, as well as expert researchers, considered
these, and indeed the wider sequence of events in the CHT, as genocide.30 My own
1999 study was slightly more circumspect, pointing less to any given moment of
29 OECD Development Centre Working Paper 233, ‘Land, Violent Conflict and Development’
(Paris: OECD, 2004); Melanie Jarman, Climate Change (London: Pluto Press, 2007), 121;
‘Indian Maoist Violence’, Reuters, 27 August 2008: http://www.alertnet.org/db/crisisprofiles/
IN_MAO.htm%3Fv%3Din_detail. Also Forest Peoples Programme, http://www.forestpeoples.org/
for regular updates.
30 See The Chittagong Hill Tracts, Militarisation, oppression and the hill tribes, (London:
Anti-Slavery Society, 1984); ‘Life is not ours’; Land And Human Rights in the CHT, Bangladesh
(Copenhagen and Amsterdam: Chittagong Hill Tracts Commission, 1991); Genocide in Bangladesh,
Survival International Review 43 (London: Survival International, 1984); Wolfgang Mey, ‘Genocide in
Bangladesh: The CHT Case’, paper for 7th European Conference on Modern South Asian Studies, 7–11
July 1981 (unpublished). Amnesty by contrast is one NGO which was notable for not articulating the
conflict as genocide per se. Bangladesh, Unlawful Killings and Torture in the Chittagong Hill Tracts
(London: Amnesty International, 1986).
650 mark levene
mass annihilation and more to an ongoing campaign of mass human rights

violations, including some thirteen major massacres in the period 1980 to 1993,
described as elements of a ‘creeping genocide’.31 A quantitative survey of the
fatalities from the conflict has never been conducted. Nevertheless, it is also clear
that the violence reached its high-point in the early 1980s when the then dictator-
ship of General Ershad initiated a full-scale military campaign against a growing
native insurgency. There were clearly some similarities here with contemporaneous
events in the Guatemala highlands, though a slightly closer parallel might be drawn
with the Indonesian military campaigns in Irian Jaya and East Timor. As in the
latter cases, Dhaka’s aim was to eliminate by force the native resistance in order to
clear the CHT ‘frontier’ for mass migration and settlement—in its case, of Bengali
peasant plains farmers, into the highlands valleys.32 With the natives duly sub-
jugated and ultimately swamped by the incomers, the state could then get on with
its more focused, primary agenda, the maximization of the region’s perceived
resource potential: its timber, water supply, mineral and most of all, its believed
oil and gas reserves, for state-corporate development.
In all this we may note close parallels with our wider picture of forcing factors
for violence in the contemporary world. As with Rwanda, demographic pressures
in an appallingly poor, ‘underdeveloped’ third world, agricultural economy were
well-noted in the 1970s and 1980s by Western donor communities. With the
Bangladeshi population already at that stage rising fast from around 40 million
in the 1950s to its present 141 million—with some estimates suggesting further
exponential increase to 340 million before stabilization—here was a country whose
size was equivalent to Nicaragua yet whose demographic weight made it the eighth
largest in the world.33 Moreover, with 80 percent of that population living in
conditions of absolute poverty all policy-makers, whether within the state, or
among first world aid providers, were agreed that only radical, remedial action
could lift the people’s prospects and in the process prevent massive social unrest.
Development of an international market-orientated textile industry employing
mass cheap labour, much of it emanating from a degraded countryside, was part
and parcel of Bangladesh’s master-plan to keep afloat in a globalized economy, if
only in order to service the country’s enormous and growing external debt. An
extreme case of social Darwinism in practice—what has been dubbed ‘the race to
the bottom’—such efforts to earn foreign currency and so avoid the country from
falling out of its already weak tier two status altogether have, however, failed to
transcend the underlying limiting factors.
31 Mark Levene, ‘The Chittagong Hill Tracts: a Case Study in the Political Economy of “Creeping”
Genocide’, Third World Quarterly 20, no. 2 (1999), 339–369.
32 Bernard Nietschmann, ‘Indonesia: Bangladesh, Disguised Invasions of Indigenous Nations,
Third World Colonial Expansion’, Fourth World Journal 1, no. 2 (1985), 96–97.
33 Levene, ‘Chittagong Hill Tracts’, 347.
Bangladesh, at heart, is a great riverine delta region seeping into the Bay of
Bengal. Historically, the source of its fertility and with it of its great human
fecundity, both elements now represent a trap for Bangladesh’s inhabitants. It is
the delta’s ecological fragility, as evident in the increasing severity of monsoon-
driven cyclones, on the one hand, the intensity of riverine erosion from up-river
Himalayan deforestation and glacial retreat, on the other, which are the immediate
cause of this encroaching catastrophe.34 Back in the 1970s and 1980s, nobody in
Bangladesh properly understood that global warming was the key amplifier and
accelerator to these processes. Or that year on year, decade on decade, this situation
could only get worse, not least from sea-level rises which would lead inexorably to
deltaic flooding and ultimately complete inundation. Factor in the rather larger
rises in global temperature than that which had been previously adduced by the
Intergovernmental Panel on Climate Change (IPCC) and figures of over 70 million
Bangladeshis permanently displaced from their domicile have become common
currency.35
Yet in one sense, state and donor policy-making was already fixated on the
problem of a displaced population more than a generation ago. And as nobody in
positions of power either inside or outside the country was prepared to grapple
with the fundamental social issue at stake—namely, the tightening zamindari
(landowner) grip over an increasingly indebted peasant class, and, as a conse-
quence, the former’s consolidation and aggrandizement of their own landholdings,
at the expense of the latter—the focus on some sort of partial internal population
transfer founded on the supposedly almost people-free nine per cent of the country
which was the CHT, had a certain logical ring to it. CHT’s existence within
Bangladesh may have been something of a political accident, emanating from the
rushed nature of the 1947 partition of British India into India and Pakistan but it
was also clearly an undisputed part of the latter’s sovereign territory. In interna-
tional eyes, therefore, Bangladesh was free to legitimately ‘develop’ the region as it
chose. The country has already suffered mass trauma and bloodshed in its 1971
secession from Pakistan. No aid donor was going ‘to endanger the survival of
millions of Bangladeshis just for the sake of the hill tribes—who are 0.5 per cent of
our population.’ So stated an official Dhaka spokesperson, in 1994.36 In principle,
he was correct. Dhaka in the late 1970s and early 1980s received foreign assistance
and funding for its migration programme; foreign consultancies were engaged to
offer advice on how to diversify a return from the region’s forest potential, Western
34 See Abdul M. Hafiz and Nahid Islam, ‘Environmental degradation and intra/interstate conflicts
in Bangladesh’, Environment and Conflicts Project, (ENCOP), Zurich and Bern, Occasional Paper
No. 6, May 1993.
35 Mabey, Delivering, 85.
36 Quoted in Tim McGirk, ‘Fear-filled return home for exiles’, Independent, 25 February 1994.
652 mark levene
counter-insurgency experts too, were soon on hand to assist in stamping out the
jumma insurgency.37
The outcome was genocide or, if not that, something very close to it. Ershad’s
settlement programme found itself stymied by ferocious resistance from the armed
wing of the jumma’s chief political movement, the JSS: clearly the notion of CHT as
practically people-less was false. In response, the military ratcheted up not only its
own anti-jumma terror campaign but also organized Muslim radicals among the
settlers into paramilitary units to do the same. As a consequence, the ethno-
religious elements of the conflict as one between majority Muslim Bengalis and
minority tribal Buddhists and animists became much more pronounced. The
region itself descended into chaos. Tens of thousands of jumma who were not
immediate party to the conflict or who had survived being incarcerated into
military-run strategic hamlets fled; at least 40,000 of them across the border into
India. But if the indigenous population of CHT had now become largely a
displaced one, so too, from a different angle, were the some four to six hundred
thousands settlers who found themselves unable to adapt their traditional plains
husbandry to entirely different conditions. In the process, they further undermined
the once traditional swidden (slash and burn) agriculture which had sustained the
jumma habitat for centuries and so confirmed the settler’s utter dependency on the
military in order to be protected and fed.38
Again we can see standard ‘Business as Usual’ elements at work but one we now
have no choice but to set against an entirely new and exponential order of stress
provided by anthropogenic climate change. In this context, the question one must
starkly pose can only be: ‘if the delta is inundated within the next century, as the
climate science now seems to consider inevitable, where will its people go?’ The
issue is hardly an academic one: whole areas of the cyclone-buffeted Sundarbans
are already disappearing very fast, leaving the country’s capital ever more heaving
with the inflow of environmental refugees. But then the crisis is more than simply
an internal one. India, already chastened by previous experiences of millions of
refugees fleeing from Bangladesh—not to say its own ongoing internal sequence of
climate-related disasters—is busily constructing a more than 2000 mile long fence
along the international border. The signal to Dhaka is blunt: its future travails will
not be Delhi’s responsibility.39 In such circumstances, is it entirely absurd to
imagine a last, mad, desperate struggle for Bangladesh’s survival played out
37 Levene, ‘Chittagong Hill Tracts’, 354–56.

38 See Kabita Chakma and Glen Hill, ‘Thwarting the Indigenous Custodians of Biodiversity’ in
Philip Gain (ed.), Bangladesh: Land, Forest and Forest People, Dhaka: Society for Environment and
Human Development, 1995), 123–137.
39 See ‘Time runs out for islanders on global warming’s front line’, Observer, 30 March 30 2008; also
http://www.independent.co.uk/news/world/asia/special-report-bangladesh-is-set-to-disappear-under-the-
waves-by-the-end-of-the-century-850938.html
between the embattled military custodians of Dhaka’s residual, sinking state and its
equally embattled hill peoples?
S C E NA R I O 2: B U S I N E S S A S U S UA L O V E RW H E L M E D
B Y G LO BA L WA R M I N G
................................................................................................................
To raise such scenarios seems grotesque where not gratuitously apocalyptic. One
can try and temper them with the argument that Bangladesh’s situation is a unique
one,40 that its particular circumstances are unlikely to be replicated elsewhere; or
alternatively, that global warming is not as dire as many of the climate scientists are
predicting.
This author, however, would argue that on both counts the contrary is true.
The cumulative radicalization of Bangladesh’s woes currently in train are no more,
nor less, than a harbinger of the wider global crisis. All the evidence, moreover, is
stacking up to suggest earlier climate predictions radically underestimated the rate
at which CO2 is building up in the atmosphere, leading to much more serious
earth system feedbacks and hence producing much steeper as well as more immi-
nent average temperature rises than previously thought possible. What the partic-
ular circumstances of Bangladesh-CHT provide is insight into how an interaction
between rapid climate change and the vagaries of political geography could lead to
contours of extreme, including genocidal violence along two possibly parallel
trajectories. In the first, states will practise triage against those parts of its citizen
or subject population considered least savable or, more cynically put, most super-
fluous. The specific conditions of climate catastrophe, however, raise the possibility
of exclusion from a universe of obligation being practised across borders, and even
applying to whole populations of perhaps, once sovereign states. In contemporary
international law, a polity must have a defined territory to exist as a state and so
enter into relations with other states.41 The case of an increasingly submerged
Bangladesh thus poses questions not only about what leverage its twilight leader-
ship might have within the world community but also what status its surviving
citizens would enjoy as fleeing for safety they are confronted with the reality of
India’s fence. In these circumstances, the possibility of genocide, whether at first or
40 See Astri Suhrke, ‘Environmental degradation, migration and the potential for violent
conflicts’, in Nils Petter Gleditsch et al. eds, Conflict and the Environment (Dordrecht: Kluwer
Academic, 1997), 257.
41 Mabey, Delivering, 87; Helen Fein, ‘Genocide: A Sociological Perspective’, Current Sociology, 38:1
(1990), 1–126 for more on the universe of obligation concept as a tool of genocide studies.
654 mark levene
second remove, becomes a function of a still extant state repudiating any notion
of obligation to those from a neigbouring one who ipso facto have become stateless.
Climate change realities in fact are pushing all manner of states towards radical
measures designed to deny entry to those so dispossessed. Indeed, it would appear
to be the richest amongst such states who are most exercised about the environ-
mental refugee ‘threat’. In a recent climate change war game, for instance, con-
ducted under the auspices of the Centre for New American Security (CNAS), game
players placed migration-prevention as the number one priority in any long-term
framework agreement on climate change, with an emphasis on the repatriation of
climate refugees to their country of origin as the necessary outcome. The proposed
agreement stated non-coercive repatriation as the ‘preferred’ method towards this
purpose, though one might be inclined to ask how exactly that would be accom-
plished for peoples from low-lying Pacific island nations such as Tuvalu or Kiribati
who are already threatened with early inundation?42 In fact, the implied policy
recommendations offered in the CNAS game are consistent with the general thrust
of US ‘security’ thinking dating back at least to the 2004 Pentagon-commissioned
report on ‘abrupt’ climate change. Then, as now, the whole emphasis has been not
on humanitarian assistance to states or societies reeling from climate catastrophe
but rather on shoring up ‘fortress America’ against waves of anticipated environ-
mental refugees. Behind such thinking too, are major Department of Defense
research and development (R&D) programmes whose purpose is to develop a
range of hi-tech weapon systems designed to interdict and immobilize ‘perimeter’
intruders. Proclaimed to be non-lethal, what damage such tazers, projectiles, ‘calm-
ative’ chemicals, as well as heat and noise weapons would actually do to masses of
human beings in the event of a major ‘emergency’ is entirely uncharted territory.43
The Oxford Research Group has aptly described this sort of thinking as that of a
‘control paradigm’ or more exactly ‘liddism’: a situation where leading states
instead of attempting to address the causes of the problem of which they, as
major carbon emitters are at the root, instead place their emphasis on preserving
the status quo, primarily through military means.44 Liddism, as policy, is clearly
both illogical and redundant. It cannot resolve the problem because the climate
change threat embraces all humanity and so can only be mitigated by an interna-
tional cooperation aimed at an overall planetary reduction of GHG emissions to
zero, in an already carbon-saturated atmosphere. Nor can liddism hope to save the
42 CNAS, Climate Change Wargame’, 28–30 July 2008, http://www/cnas.org/ClimateWarGame/

Thanks to Marc Hudson for alerting me to this exercise.
43 See Dave Webb, ‘Thinking the Worst, The Pentagon Report’; Steve Wright, ‘Preparing for Mass
Refugee Flows, the Corporate-Military Sector’, respectively chapters 2 and 3 of David Cromwell and
Mark Levene, eds. Surviving Climate Change, The Struggle to Avert Global Catastrophe (London: Pluto
Press, 2007).
44 Chris Abbott, Paul Rogers and John Sloboda, Global Responses to Global Threats, Sustainable
Security for the 21st Century (Oxford: Oxford Research Group, 2007), 28.
rich fossil-fuel dependent economies themselves, through some sort of security

isolation in the shorter-term, not least as their heavily populated but low-lying or
deltaic metropolitan regions are swept by an increasing frequency of climate-driven
storm surge and or, flooding, in part as a consequence of polar ice-melt.45
Again as a further empirical example of what this may actually mean consider at
the other end of the spectrum conflict-ridden Central Asia, most obviously
Afghanistan. At the time when this chapter was being drafted in late October
2008, the Royal United Services Institute (RUSI) announced that eight million
people in that country could now be threatened with winter-time starvation, as
much as a consequence of global warming-induced drought and soaring world
food prices as due to the ongoing Taliban insurgency. With the retreat of Himala-
yan glaciers and hence the further deterioration of already stressed irrigation
systems in a region hugely dependent on the careful husbanding of exactly this
limited, or seasonal water resource, complete societal breakdown is conceivable.46
Numerous expert studies have considered how governments in environmentally
challenged parts of the world when combined with demographic pressures and
weak undeveloped economies, are the most likely to default on delivery of basic
services to their populations and so most likely to pay the price through increased
militancy (jihadist or otherwise), insurgency, and warlordism.47 But what we are
contemplating here are countries that have not simply ‘failed’ per se in the Western
lexicon of what constitutes ‘success’ and failure but ones that may actually ‘disap-
pear’ off the modern sovereign state map altogether. After all, great civilizations of
the past, famously along the Silk Route, did exactly this as the wells and oases dried
up and the raiders closed in.48 The question for us is what happens—as in past
times, happened—to the peoples of these polities in the face of such calamities?
Left to their own devices do they, for instance, fight it out among themselves in
some Hobbesian zero-sum game, as the food and water resource itself diminishes
to zero? This would be a truly post-genocidal landscape in which atrocity is not
simply the norm as perpetuated by the simple conditions of extreme scarcity but
45 See IPPC 4th assessment report (AR4), 2007 www.ipcc.ch/ipccreports/ar4-syr.htm, 10, for
examples of anticipated regional impacts.
46 Afghanistan ‘Preventing an Approaching Crisis’, RUSI briefing note, 31, 31October 2008 http://
www.rusi.org/downloads/assets/RUSIAfghanBriefingNotepdf.pdf/. Also Stephan Harrison, ‘Climate
Change, Future Conflict and the Role of Climate Science,’ Royal United Services Institute Journal, 150: 6
(2005), 18–23.
47 See amongst others, Colin H. Kahl, States, scarcity, civil strife in the developing world (Princeton
and Oxford; Princeton University Press, 2006); James D. Fearon and David D. Laitin, ‘Ethnicity,
insurgency, civil war’, American Political Science Review 97, no. 1 (2003), 75–90; Jon Barnett and W. Neil
Adger, ‘Climate Change, Human Security, and Violent Conflict’, Political Geography 26 (2007), 639–
655.
48 See Rob Johnson, ‘Climate change, resources and future war: the case of Central Asia’, in Mark
Levene, Rob Johnson, Penny Roberts, eds., History at the End of the World? History, Climate Change
and the Possibility of Closure (Penrith: Humanities E-Books, forthcoming 2010).
656 mark levene
one in which, without the state or even outside agencies to offer a calculus as to the
political purposefulness of violence, no one single group of actors can be blamed,
let alone held to account, for the resulting carnage.
For the substantial (tier two) bloc of states, in other words those who seek to stay
afloat as coherent political entities above this fray, the climate change threat
operates in political terms from two pincer-like directions. In the first, there is
the straightforward fear of being ‘swamped’ by environmental refugees from a
neighbouring state or states which have already fallen into the lower tier three
category, or may soon do so. In the second, the threat operates on the level of
finding oneself unable to resist other wealthier, more powerful and militarily
stronger—though not necessarily tier one—states, interfering with or directly
appropriating one’s own scarce resources, most obviously food, water, as well as
energy supply. The anxiety of having to navigate between these twin Scylla and
Charybodis—like perils, moreover, will be exacerbated for each state’s elite by a
historic sense of mission to carry their country forward to ever higher levels of
preferably carbon-fuelled, industrially based development in order to meet the
needs of a fiercely competitive global market. Climate change, of course, contra-
dicts this aspiration foursquare. But it does so not simply through its range of
growing physical stresses but in the psycho-cultural burden it imposes on those
who have imbibed nothing other than a telos of development.
A world replete with nuclear weapons, moreover, could turn a struggle for
diminishing resources into an altogether more deadly encounter involving whole
national populations. In normal conditions, leading tier two players, including
Russia, China and India might be looking forward to a political ascendancy on the
world stage without recourse to inter-state conflict let alone use of their nuclear
arsenals. But then how are they likely to respond to conditions in which collapsing
neighbours might use the threat of military force, including, where those states
have their own nuclear weapons, to punch their way out of encroaching turmoil?
Ecological fragility could be the final straw for an already embattled, increasingly
lawless, indeed fragmenting Pakistan which, nevertheless, still retains its nuclear
wherewithal. By the same token, how is a clearly tier one, nuclear-armed Israel set
amongst altogether more precarious, yet hostile tier two Middle Eastern neigh-
bours likely to react to a sustained regional water crisis? Or is it, actually, the other
way round: an Israel which has most to fear not least from the Palestinians of the
occupied territories as perhaps, they make one final, desperate subaltern attempt to
redress the ecological as well as political balance?
Finally though, where do the leading tier one states fit into this darkening scene?
Expert Western opinion generally grafts a map of already existing global economic
poverty onto any forward-looking plot of vulnerability to climate conflict.49 Yet the
49 See Dan Smith and Janani Vivekananda, A Climate of Conflict: The Links between Climate
Change, Peace, and War (London: International Alert, 2007), 18–19.
very fact that a very poor country like Ivory Coast, for instance, has taken in so
many environmental refugees could equally indicate that the less ‘developed’ a state
or society the more resilient it is to the most serious environmental or socio-
economic challenges, man or nature can throw at it.50 At least populations in such
countries (whether urban or rural) have direct relationships with land and water,
however degraded those basic elements have become. By contrast, it is in rich tier
one countries where such relationships are at their most tenuous and where,
arguably, fears of mass refugee ‘invasions’ are also at their most intense. It is a
truism that hierarchic, complex, city-centred societies are only three or four meals
away from anarchy. Catastrophic breakdown in other words, is quite conceivable in
the face of some all-embracing crisis, not least given these societies’ absolute
dependency on thin, often distant supply lines to provide basic services, including
water, food, heat and light.51 In circumstances in which standard front-line public
services find themselves overwhelmed or unable to cope, populations will not only
be unable to meet their own basic physical needs but also be seriously psychically
disturbed by the realities confronting them. It is in exactly such emergency condi-
tions that elites of tier one states might become the most obvious candidates to
make responses which in normal times would be deemed not only unthinkable but
unforgivable.
C O N C LU S I O N
................................................................................................................
In the course of this chapter, we seem to have come a long way from Lemkin or of
his vision of how genocide, through international law, might one day be ultimately
removed from the actions of human states and societies. Lemkin’s purport was not
only entirely honourable but was passionately fought for, largely single-handedly.
To cut across this aspiration with not only an entirely more dystopian forecast but
one which in key respects questions the long-term value of the term ‘genocide’
itself, seems both churlish and contradictory. Lemkin’s law attempted to achieve
not simply clarity on the subject but in the process a mechanism for making things
better. By contrast, we have posed that without a firm grip on the understanding of
the driving forces which determine the wider formation and organization of our
50 Mabey, Delivering, 119. See also ‘Climate change and displacement’, Forced Migration Review 31
(October 2008), for the current debate on third world responses to environmental refugees compared
with other migration and/or displacement factors.
51 See Deborah MacKenzie, ‘The End of Civilisation’, and MacKenzie, ‘Are We Doomed?’, New
Scientist, 5 April 2008; Joseph A. Tainter, The Collapse of Complex Societies (Cambridge: Cambridge
University Press, 1988), for the wider argument.
658 mark levene
present international system, implementation of the Convention will not only

remain piecemeal and inadequate but will be rapidly overtaken by forces which
render its fragile efficacy null and void.
There is something more which needs to be said here. While we are now
standing at the apex of a particular human trajectory, at the same time, we also
possess sufficient analytical tools and material evidence to survey the entire land-
scape of human history and experience which preceded it. Throughout the histori-
cal record, the struggle for human existence carried with it, repeated proclivities
towards strages gentium. What is distinct about this potentially final global epoch is
that the disparity between the material overreach and the limits of the planetary
carrying capacity are taking us all—tier one included—into a totalizing mode of
exterminatory behaviour. If one is thus looking for one single prediction it is this: it
will be mass self-violence, not climate change per se which will take us over the
abyss.
What is the antidote? On one level, it is a terribly simple one. Arnold Toynbee
that same great if now much forgotten historian of civilization—who also had so
much of prescient value to say on the subject of genocide—put it aptly just before
his death. Our mission must be to seek not ‘a material mastery’ over the non-
human environment, but for ‘a spiritual mastery’ over ourselves.52 As for Mahatma
Gandhi, that apostle, as well as arguably the greatest exponent of non-violence of
recent times, he put the case even more tersely on behalf of the peoples of this
overcrowded planet: there is ‘enough for everybody’s need but not for everybody’s
greed’.53
In short, for those who would seek to avoid genocide in the twenty-first century,
the task cannot somehow be reduced to Lemkin’s law. The phenomenon cannot be
contained within this box: it is too fundamental a by-product of a more general
dysfunction, not to say, even as it transmutes into persistent post-genocide, a key
indicator of a more all-encompassing Nemesis. To arrest the encroaching inevita-
bility of this trajectory will require, amongst other things, a thoroughly post-
Lemkian effort to recognize the false chimera of the globalizing project and with
it the necessity for a sufficiency and sustainability upon which the term oikono-
mia—economy—was originally founded. Such an approach will be geared towards
the values of human scale and with it of an entirely gentler and certainly more
heterarchic social and communal empowerment.
52 Arnold Toynbee, Mankind and Mother Earth, A Narrative History of the World (New York and
Oxford: Oxford University Press, 1976), 18.
53 Quoted in M.S. Dadage, ‘Science and spirituality’, http://www.mkgandhi.org/articles/sci.%
20and%20sprituality.htm. Significantly this translates exactly into Aubrey Meyer’s visionary yet
scientific Contraction and Convergence proposition for how humankind might still tackle climate
change. See Aubrey Meyer, ‘The Case for Contraction and Convergence’, in Cromwell and Levene,
Surviving, 29–56.
Abbott, Chris, Paul Rogers and John Sloboda, Global Responses to Global Threats: Sustain-
able Security for the 21st Century (Oxford: Oxford Research Group, 2007).
Cromwell, David, and Mark Levene (eds.), Surviving Climate Change, The Struggle to Avert
Global Catastrophe (London: Pluto Press, 2007).
Davis, Mike, Late Victorian Holocausts, El Nino Famines and the Making of the Third World
(London and New York: Verso, 2001).
Diamond, Jared, Collapse: How Societies Choose to Fail or Survive (London: Penguin, 2005).
Levene, Mark, Genocide in the Age of the Nation State, vol. 1: The Meaning of Genocide, vol 2:
The Rise of the West and the Coming of Genocide (London and New York: I.B. Tauris,
2005).
Meadows, Donatella, et al., The Limits to Growth (New York: Universe, 1972).
Smith, Dan, and Janani Vivekananda, A climate of conflict: The links between climate change,
peace and war (London: International Alert, 2007).
Tainter, Joseph A., The Collapse of Complex Societies (Cambridge: Cambridge University
Press, 1988).
Wasdell, David, ‘Radiative Forcing & Climate Sensitivity’, Tällberg Consensus Project,
‘The Tipping Points we cannot cross: Defining the Boundary Conditions for Planetary
Sustainability’, 25–26 June 2008, http://www.jimhadams.com/eco/RadiativeForcingEd3.
doc
Wallerstein, Immanuel, The Modern World System, 3 vols. (San Diego and New York,
Academic Press Inc.1974–1988).
I. . N DEX
...........
Abdul Hamid II, Sultan 50 Aly, Götz 89

Abkhazia 516 Amaru, José Gabriel Túpac 318, 319
Adomnán of Iona 276 America
Adorno, Theodor W. 199, 202–3, 204, 224 n. 37, see indigenous Americans; Latin America;
226 n. 46, 233–4 North America; United States of America
Negative Dialectics (ND) by 230–1 American Military Tribunal 620
see also Dialectic of Enlightenment Amin Dada, Idi 552, 554, 557, 564–5, 593, 611
Aeðilfrith, Northumbrian king 261–4, 272 Amnesty International 475, 587, 649 n. 30
Æthelred II, King 292 ancient genocide 249–50
Afewerki, Isseyas 540 evidence for 240–3
Afghanistan 513, 655 motivations for 251–6
see also Taliban nature of 243–7, 256–7
Africa 644–5 Anderson, Benedict 184
colonial rule in 347–8 Anderson, David 359 n. 40
colonial warfare in 354–60 Andreopoulos, George J. 7, 174
German East 28, 349, 356–7, 410 Angola 568
German South West 28, 29, 91–2, 106, 345, Annan, Kofi 595 n. 27, 603
354–6, 358–9, 362, 409, 410 anthropology 142–3
Scramble for 349, 361, 531 activist 189–90
violence in southern 350–4 and ethnography 183–4
see also forced migration; Great Lakes region; and genocide 186–8
Horn of Africa; and individual countries and refugees 185–6
African Union 608, 613 cultural critique 190–2
Agamben, Giorgio 222, 224 forensic 188–9
Agamemnon 239, 240–1, 247 scholarship 182–3
Ahtisaari, Martti 525 anti-Semitism 407–9, 411
Aidit, D. N. 451 see also Hitler, Adolf; Nazism
Akayesu, Jean-Paul 623 Antonescu, Marshal Ion 52
Akhavan, Payam 522 Arab Gathering, the 546
al Dahab, Gen. Abdel Rahman Suwar 543 Arbenz Guzmán, Jacobo 492, 493, 584
Albania Arbour, Louise 606, 628–9
see Hoxha, Enver Arendt, Hannah 207–8, 215, 346
al-Bashir, Omar 96, 99, 630 banality of evil concept of 203–4, 209, 221
Albright, Madeleine 13, 168 thought and work of 173, 201 n., 210,
Alexander the Great 244, 257, 287 222, 231 n. 59
Alexander II, Czar 388 Argentina 492 n., 497, 498, 504–5, 626
Alfonso XIII, King 357 the question of genocide in 500–3
Algeria 356 see also Conintes Plan
Allen, Beverly 70 Arkan
Allende, Salvador 496 see Ražnatović, Željko
All-Ethiopia Socialist Movement Armenia 55
(MEISON) 536 see also Nagorno-Karabakh
al-Qadir, Abd 356 Armenian genocide 50–1, 52–3, 372–3
662 index
Armenian genocide (cont.) Bede, the Venerable 259–68, 271–3, 274, 275,
death toll 382 276, 285
debate concerning the 365–6 Belgian Congo 362, 409, 410
implementation of the 376–7, 381 Belgium 448–9
rape and the 73, 377–8 Bellamy, Alex J. 597 n. 5, 598 n. 8
see also Committee of Union and Progress; Bemba, Jean-Pierre 556, 569
Ottoman Empire; Special Organization; Ben-Ari, Eyal 326
Talaat, Mehmed Benes, Eduard 58
Armenians 374–5, 515 Benjamin, Walter 228 n. 54
pre-1915 massacres of 368–9 Benjamites, the 261–2, 277
Arnold, Bettina 186–7 Bessone, Gen. Diaz 504–5
Arusha Accords 94, 107, 572 Bey, Manastirli Bahaeddin 371
Asch, Solomon 200, 205, 210 Bey, Shukru 379
ASEAN 475 Bible, the 240–2, 244, 261–2, 282, 285, 301
Asser, medieval historian 267 Biha, Léopold 555, 570
Assyria 249, 255 Bilu, Yoram 326
see also Sennacherib of Assyria; Binding, Karl 410
Tiglath-pileser III; Tiglat-Pileser I biopower 151, 222
Audacht Moraind 268 Bismarck, Otto von 92, 362
Aufranc, Gen. Alcides Lopez 504 Black Death, the 300
Augé, Marc 102 Blair, Tony 602
Auschwitz 65, 103–4, 233–4, 417, 424 Blaškić, Tihomir 522
Australia 40, 83, 115–17, 479, 481 Bloxham, Donald 8 n. 19, 36, 71, 90, 175,
Austria 111 180 n. 44, 198, 485 n. 55, 608
Avars, the 271, 273, 275, 284 Bobrinski, Count 390 n. 10
Awami League 468, 469–70 Boers, the 350–2, 353–4
Azerbaijan 55 Bogdanović, Dimitrije 518 n. 23
see also Nagorno-Karabakh Bohemia 291–2, 294
Bolivia 495
Babylon 246, 255 Borah, Woodrow 307
Bach-Zelewski, Erich von dem 420 n. 24 Bordaberry, Juan Maria 495–6
Badoglio, Italian marshal 358 Bormann, Martin 417
Baker, James 627 Bos, Pascale Rachel 64
Balaguer, Joaquin 494 Bosch, Juan 494
Balandier, Georges 346 Bosnia 114, 153, 507, 510, 521–3, 599
Balkan wars (1912–13) 49, 369 Bosnian Serb Army (VRS) 522–3
Bangladesh 74, 653 Boutros-Ghali, Boutros 629
see also Chittagong Hill Tracts; Brazil 494–5, 505
East Pakistan Brezhnev Doctrine 513
Banzer, Gen. Hugo 495 Brooks, Francis 307
Barbie, Klaus 495 Browning, Christopher R. 200 n. 8,
Bariatinskii, Prince 388 206 n. 23, 210
Barre, Siad 533, 540–1, 547 Brownmiller, Susan 71
Barta, Tony 117, 148 Brubaker, Rogers 22, 440
Bartelson, Jens 117 Buckley-Zistel, Susanne 114
Bartrop, Paul 92 Budak, Mile 519
Baskar, Bojan 526 Bulgaria 50
Bassiouni, Prof. M. Cherif 137 see also Committee of Union and
Bauer, Yehuda 4 n. 7 Progress
Bauman, Zygmunt 11–12, 89, 144, 152–3, Bundism 24
162, 224 bureaucracy
Beachler, Donald 471 and genocide 89–90, 152–3
Beaufre, André 492 n. Burger, Thomas 158
index 663
Burma Uyghurs in 426, 438, 440 n. 29

see state-sponsored violence see also political genocide; sinicization
Burnett, Californian governor 338 Chinese Communist Party (CCP) 436–8, 440,
Burundi 554, 557, 560–1 443, 454–6
genocide 562, 570–1 Chittagong Hill Tracts (CHT) 642, 649–52
violence in 553, 555, 556, 569–70, 571–2 choice problem, the 202, 205–6, 208, 215
Busch, Mario 495 Christopher, Warren 526
Bush, George H. W. 618 Churchill, Ward 330
Buyoya, Pierre 571–2 Churchill, Winston 32, 58, 357, 359, 361
Cicero 250–1, 253, 254
Caedualla, King 259, 261, 263, 264, 272, 274 Cigar, Norman 95, 518
Caedwalla, King 268, 271–2 classicide 156, 157
Caesar, Julius 244, 257 climate change 642, 648–9, 652, 656, 658
California 335–9 and genocide 653–4, 655
Callahan, Mary P. 473 n. 21 see also global warming; resource capacity
Cambodia 77, 106, 113, 192 Cohen, William 13, 168
see also Khmer Rouge; political Cohn, Norman 85
genocide; Pot, Pol Cold War 54, 94, 584–5, 624–6, 627
Canada 40 see also Brezhnev Doctrine; Warsaw Pact
Capron, Elisha Smith 336 Cole, Tim 66
Carpenter, R. Charli 61, 68 Colombia 498, 618
Casey, Conerly 194 colonialism 305, 410
Catari, Dámaso 317 and genocide 346–7, 349, 363
Catari, Nicolás 317 medieval 288–9
Catari, Tomás 317, 318, 319 settler 433, 438
Catari, Túpac 318 see also Africa; Horn of Africa; imperialism;
Catholic Church 281, 296, 482 Italy; Lemkin, Raphael; North America
in colonial Latin America 311–13, 315, 316, Čolović, Ivan 517
317, 318 Columbus, Christopher 305
Catiti, Alonso 317 Committee of Union and Progress (CUP) 50,
Cavell, Stanley 233 n. 69 99, 365–6, 375, 377, 383, 384–5
Césaire, Aimé 346 and Jews 371
Chad 546, 548, 568 and Ottoman Greeks and Bulgarians 369–70
Chalk, Frank 7, 85, 149–50, 160, 170, and ‘Turkification’ 375, 381
493 n. Communist Party of Kampuchea see Khmer
Charlemagne, king of the Franks 271, 273, Rouge
274, 276, 289 concentration camps
Charles II of Naples 298 in Africa 359
Charles IV of Luxembourg 292 in Latin America 495, 496, 497, 503, 505
Charny, Israel 7 see also Auschwitz
Chávez, Federico 493 Confino, Alon 118
Chavez, Hugo 617 conformity effect 205–6, 212–13, 215
Chea, Nuon 98 Congress of Berlin 367–8
Chechnya 510 Conintes Plan 491
see also population displacement Cook, N. D. 308
Chesterman, Simon 610, 611 Cook, Sherburne 307
Chile 496–7 Copelon, Rhonda 71
China 545, 611, 647 crimes against humanity
Mongols in 426, 432, 434–7, 439 and war crimes 620–1
rectification of names in 441–3 see also genocide
the Han in 428, 436, 438, 443 Croatia 108–9, 519–21
the Manchu in 428, 429–30, 432–5 see also Ustasha
Tibetans in 426–8, 438 Crosby, Alfred 325, 328
664 index
Crossley, Pamela 430 Dudayev, Dzhokhar 514

Crusades, the 285, 287, 296–8 Dulić, Tomislav 109, 519
cultural genocide 39–40, 134–5, 157, 193–4, 332 Durkheim, Emile 103
and forced labour in Africa 360–2 Dutch East India Company 351
in China 426–8 Duval, Kathleen 327
in Latin America 312 Duvalier, François 492, 494
in North America 336 Duvalier, Jean Claude 494
see also Lemkin, Raphael Duviols, Pierre 311
cultural identity 448 Dwyer, Leslie 196
Cultural Revolution, the 456–9
Čuruvija, Slavko 525 East Pakistan 593, 625
Cyprus 54 see also state-sponsored violence
Czechoslovakia 607 East Timor 68, 611
see also state-sponsored violence
Dalai Lama, the 427–8, 436, 438, 441 Eav, Kaing Guek 113
Daniels, Roger 324 Eck, Kristine 171, 172
Daniel-Wrabetz, Joana 70 Edward I, king of England 288, 291, 292, 301
Danilov, V. P. 394 n. Eichmann, Adolf 203–4, 421, 423
Darfur 2, 67, 74, 96–7, 138, 533, 595, 597, 613–14 trial 112, 129, 133
genocide in 530, 534, 548 Einhard, medieval biographer 271, 273
violence in 545–6 Einsatzgruppen 153, 420, 422, 423, 624
Darwin, Charles 84 Eisele, Manfred 591 n. 18
see also social Darwinism Ekkehard, German chronicler 287
Davidson, Donald 233 n. 68 Ekmečić, Milorad 517
Dayton Peace Accords/Treaty 59, 523, 526, 599 Eliot, John 324
de-Cossackization 392–3 elitocide 521
de-kulakization 393–4, 397 Elkins, Caroline 359 n. 40, 359 n. 41
Del Ponte, Carla 629 Elliot, Mark 430
Democratic Republic of Congo (DRC) 74–5, 115, Elliott, Jane 213
556, 558, 559 Eltringham, Nigel 145
resources in the 645–7 Endecott, John 332
violence in the 552–3 England 47, 288, 290, 292–3, 294
see also Belgian Congo; Kabila, Joseph; see also Northumbria; Saxons
Kabila, Laurent; Lumumba, Patrice; Enlightenment, the
Mobutu, Joseph-Désiré; Zaire and genocide 227–8
Deng, Francis 604, 613, 614 see also Dialectic of Enlightenment
Denikin, Gen. 392 enslavement
Denitch, Bogdan 109 in Latin America 309–11, 491
Dergue, the 536 in medieval Europe 276–7
Desmond, first earl of 290 in the ancient world 240, 244–5,
Dialectic of Enlightenment (DE) 228–30, 231–2 249–50, 251–2
Dietrich of Niem 289 in the General Government 419
diffusion 27–8 see also cultural genocide
Diner, Dan 111 Ensler, Eve 74
‘dirty war’ concept 492 Eritrea 529, 532, 539, 540
Dominican Republic 494 violence in 535–6, 537
Douglas, Mary 191 Eritrean People’s Liberation Front (EPLF)
Downes, Alexander 179 537, 539, 547
Drechsler, Horst 363 Ermath, Elizabeth Deeds 12 n. 24
Drinnon, Richard 324, 332 n. 43 Ershad, Gen. 650, 652
Dubnow, Simon 24 Etchecolatz, Miguel Osvaldo 500 n., 501
Duch, Comrade Ethiopia 348, 359, 529, 547
see Eav, Kaing Guek Red Terror in 533, 534, 536, 548
index 665
violence in 530–2, 538–9 Freeman, Michael 332–3

see also Selassie, Emperor Haile Freeport 478
ethnic cleansing 135, 286, 419 Frei, Norbert 627
and ethnic conflict 57–9 Fritzsche, Peter 118
and genocide 45–6, 405
and international law 46 Gaita, Raimond 4 n. 7
defining 43–4, 130 Gamzatov, Razul 513
history of 47–9 Gandhi, Mahatma 658
modern 49–55 Garang, John 544–5
nature and development of 42–3 Garzón, Judge Baltazar 500, 502
sociology of 154–7 Gellately, Robert 6
see also Horn of Africa; Mann, Michael; Gellner, Ernst 24
Russia; South Ossetia; Soviet Union; gender
state, the; Yugoslavia and genocide 66, 67–9, 78–9
ethnicity 446–9 and the Holocaust 63–5
in colonial Latin America 318–19 definition of 62
in medieval Europe 269–70, 286–7, 288, see also genocide; rape
290–1, 293, 301–2 gendercide 68, 157
see also Great Lakes region; violence General Government 418–19
ethnocide 39, 157, 187, 194, 220, 311, 313, 315, 360 General Plan East (Generalplan Ost) 51, 420
see also cultural genocide; Latin America genocidal intent 75, 78, 87, 136
eugenics 409–10 and sociology 148–50, 155
Evans, Gareth 613 in colonial Latin America 309–10, 320
Ezhov, Nikolai 91 in North America 323
philosophical perspective on 220–1
Fackenheim, Emil 218 n. 3 genocide
Falkenberg, Johann 298 and crimes against humanity 125–8, 135
Fanon, Frantz 346 and masculinity 76–7
Fein, Helen 7, 62, 84–5, 148–50, 155, 159, and perpetrator behaviour 201–4, 208–9
160, 170, 533 and regime types 173–4
Ferdinand, Archduke Franz 108 and social theory 278
Finzsch, Norbert 333 and truth, memory and representation 194–5
First Arab-Israeli War 57 and war 25–6, 84–5, 92, 157, 170, 176–8
First Punic War 250 as a legal concept 123–4, 141
Fisher, Siobhan 71–2 as an intentional crime 138–9
Fitzmaurice, Andrew 26 attempted 136
forced migration 44–5, 47, 58, 249 auto- 463
in Africa 358, 537–8, 539, 545 biological 305, 314, 332
in Latin America 313–14 collateral 320
in medieval Europe 291–2, 301 definitions of 32–5, 89, 128, 131, 148–9,
in Nazi Europe 51–2 160, 534 n. 12
see also Ottoman Empire; Russia; delayed 416
Soviet Union origin of the term 22
Ford, Gerald 481 physical 305, 310, 313, 426
Forges, Alison des 99 prosecuting 631–5
Foucault, Michel 12, 151–2, 212 n. 46, scholarship and studies 2–5, 6–9, 14, 20, 61,
213–14, 222 67, 445–6, 471, 639
Fouracre, Paul 265 terms 157–8
Fowler, Jerry 605 n. 33 see also individual entries
France 47, 283, 293, 609, 612 Geoffrey of Monmouth 284
Frazer, Sir James 24–5 Geoffrey of Paris 292
Frederick II of Hohenstaufen 291 Georgia 598
Fredrickson, George 324 see also South Ossetia
666 index
Geraci, Bob 388 n. 2 Greece

Gerald of Wales 285, 289, 295 ancient 248–9, 252, 253–4
Geras, Norman 219 n. 9 see also Committee of Union and
Gerlach, Christian 8 n. 19, 86–7, Progress; Cyprus
485 n. 55 Greenawalt, Alexander 105
Germany 283, 300, 589 n. 12, 624–5 Gregory of Tours 259
and the ‘stab in the back’ concept 412 Gregory VII, Pope 281
beliefs and practices of pre—Nazi 410–11 Grotius, Hugo 26
homosexuals and ‘Gypsies’ under Nazi 415, groups and groupism 22–3, 40, 160–1
416–18, 424 Grunfeld, Tom 428
sterilization in Nazi 414, 417 Guatemala 194, 195, 493, 500, 505, 584–5, 612
the handicapped under Nazi 415–16, see also Arbenz Guzmán, Jacobo
417, 424 Guatemalan Forensic Anthropology
see also Africa; Historikerstreit; Hitler, Foundation 189
Adolf; Lebensraum; Nazism Guevara, Ernesto ‘Che’ 495
Gervase of Tilbury 288, 289 Guibert of Nogent 299
Geyer, Michael 110 Gurianov, A. 401
Gildas, sixth century British scholar 265, Gurr, Ted 171, 446, 533
266–7, 274–5 Gutman, Roy 69
Glatzer, Wolfgang 143
global warming 641, 651, 653, 655 Habyarimana, Juvénal 99, 554, 556,
Goebbels, Joseph 110 558, 574
Goffman, Erving 213–14 Hacking, Ian 211, 214
Goldenberg, Myrna 65 looping effect of 212–13
Goldhagen, Daniel Jonah 202 n. 14, Hague Convention (1907) 124
206 n. 23 Hague, Euan 77
Goldstone, Richard 628 Haiti 494
Gorbachev, Mikhail 512 see also Duvalier, François
Gordon, Gen. Charles 356 Halbwachs, Maurice 103
Goremykin, Prime Minister 390 Hale, Charles R. 189
Gotovina, Gen. Ante 520 Halsall, Guy 263
Goulart, João 494 Hammarskjold, Dag 566
Grand Mufti of Jerusalem 423 Hannibal, Carthaginian general 250, 254
Gray, Thomas 287 Hardman, Anna 64
Graziosi, Andrea 396 Harff, Barbara 85–6, 88, 170–1, 172, 173, 176, 177,
Great Britain 284, 287, 580, 582, 644–5 431 n., 446, 493 n., 533
see also England; medieval genocide; Harris, Richard Cole 324, 341
Scotland; Wales Hayden, Robert M. 109
Great Famines, the 395–8 Hazan, Pierre 627
Great Lakes region Hegel, Georg Wilhelm Friedrich 226 n. 44
authoritarian regimes in the 557–8 Heidegger, Martin 218, 225
balance-of-power in the 560–1 Heidenrich, John 84
disease in the 562, 564, 566, 568, 646 Heim, Susanne 89
ethnicity in the 562–3 Helmold of Bosau 289, 298
insurgencies in the 559 Henry of Huntingdon 290, 292
natural resources in the 558 Henry of Livonia 297
refugees in the 561–2 Henry the Lion of Saxony 290
regime transitions in the 554–7 Henry II, king of England 288
see also Burundi; Democratic Republic of Herder, Johann Gottfried von 23–4
Congo; population displacement; Herero, the 91, 345, 354–5, 358, 410
Rwanda; Uganda Herman, John 432
Great Leap Forward, the 454–5 Hersh, Seymour 471 n. 11
Greco-Turkish war 51, 383 Heydrich, Reinhard 422, 423
index 667
Hilal, Musa 96–7 imperialism 409

Hilberg, Raul 64, 66, 89, 152 American settler 333–4, 336
Himmler, Heinrich 51, 66, 415, 417, and genocide 25–9
418–19, 420, 421–2 India 56–7, 470, 652
Hinton, Alexander 143, 185, 190–2, 194 indigenous Americans
Hirsch, Herbert 144, 146 and disease 328–30, 335
Historikerstreit 116–17 ethnic cleansing of 48
Hitler, Adolf 51, 52, 89, 98, 99, 110, 143, 415–16, see also Pequot War
418–19, 421–3, 424, 607 Indochina War
aims of 109 First 460
anti-Semitism of 412–13 Second 459, 460
Ho, Ping-ti 429 Indonesia
Hobsbawm, Eric 447 see political genocide; state-sponsored
Hobson, J. A. 27 violence
Hoche, Alfred 410 Indonesian Communist Party (PKI) 450–4, 477
Hollenbach, David 340 Indonesianization 476, 478, 481
Holocaust, the 407 Innocent IV, Pope 300
and ethnic cleansing 52 International Court of Justice (ICJ) 69, 131,
concepts of 152–3 604, 623, 629
scholarship on 3–5, 7, 63–5 Nicaragua rulings 592
uniqueness of 219 Yugoslavia rulings 94, 129–30, 132, 140, 606
see also Jews International Criminal Court (ICC) 2, 72, 111,
Homer 239 610, 618, 623, 634
see also Iliad Rome Statute of the 129, 130–1, 140,
Honduras 497 589, 636
Horkheimer, Max 231 weakness of the 620, 629–31
see also Dialectic of Enlightenment International Criminal Tribunal for Rwanda
Horn of Africa (ICTR) 72, 129, 133–4, 623–4, 628–9,
colonialism and decolonization in the 531–2 633, 634
ethnic cleansing in the 541 International Criminal Tribunal for the former
question of genocide in the 533–4, 548 Yugoslavia (ICTY) 72, 130, 138,
violence in the 532–3, 535 140, 634
see also Eritrea; Ethiopia; Somalia; Sudan establishment of the 522, 627–8
Horowitz, Irving Louis 85, 89, 144, 173 on national and ethnic groups 133–4
Howard, John 115–16 Srebrenica rulings 46, 68–9, 136–7, 518
Hoxha, Enver 517 weakness of the 628–9, 633
Huguenots 47 International Crisis Group 612
Hull, Isabel 91–2, 410 International Military Tribunal (IMT) 36, 92,
Hultman, Lisa 171, 172 125–8, 620–21, 624
human rights 586–8 international security and intervention
see also United Nations 590–95, 627
Human Rights Watch 475, 588, 628 see also military intervention
Hurtado, Albert 335 Iran
Hussein, Saddam 56, 633 see Khomeini, Ayatollah
Hutu Iraq 88, 591 n. 16, 603
refugees 185–6, 562, 567, 575 Anfal campaign in 56
-Tutsi relations/divisions 348, 556, 557–8, war 594, 598, 614
561, 563 Ireland 270, 271, 273, 276, 277, 288, 290,
-Tutsi violence 553, 555, 569–72, 573–5 295, 302
Huyssen, Andreas 76 Islam 531
fundamentalist/militant 513, 515, 527–8
Iliad, the 239, 241, 244, 245, 247, 255 in Sudan 544, 545, 546
see also Trojan War see also Mujahideen
668 index
Isle of Wight 267–8, 269, 271 Khan, Gen. Yahya 469

Israel 117, 584, 656 Khmer Rouge 77, 93, 95, 98, 106, 459–63, 471, 506
ancient 241–2, 252, 256, 263 Khoikhoi, the 351
see also Benjamites; First Arab-Israeli War Khomeini, Ayatollah 510
Italy Kiernan, Ben 6–7, 83, 106, 341, 533
as colonial power 357–8 Kikuyu, the 359
Izetbegović, Alija 521, 522, 527 Kimmerling, Baruch 431 n.
Ki-moon, Ban 604
Jäckel, Eberhard 412 King, Henry 127
Jackson, Justice Robert 125–6, 623 Kiribati 654
Jacob, biblical figure 261–2 Kissinger, Henry 470, 481
Jacobs, Steven L. 20 n. 4 Kitchener, Lord 359, 361
Jaimes, M. Annette 330 n. 31 Kobulov, Bogdan 404 n. 56
Janhunen, Juha 440 Kony, Joseph 565
Janjaweed/janjawiid 96–7, 546 Kosovo 517–18, 523–5, 593, 628
Japanese Army Kosovo Liberation Army 524
Second World War behaviour of 75 Kraft, Claudia 20 n. 3
Jashari, Adem 524 Krain, Matthew 171, 172, 173–4, 176
Jennings, Francis 323, 332 n. 43 Kroeber, Alfred 187
Jerusalem 248, 296–7 Krstic, Gen. Radislav 46
see also Grand Mufti of Jerusalem Kuper, Leo 7, 20 n. 5, 86, 148, 169–70, 175
Jesus Christ 299, 300 Kuperman, Alan 598
Jews Kupperman, Karen 330
medieval persecution of 299–301 Kurds 56, 367–8, 377, 381, 383, 590
Nazi genocide of the 420–4
persecution of by imperial Russia 389–90 labelling theory 213
Soviet persecution of 401, 402 Lang, Berel 221 n. 18
see also anti-Semitism; Committee of Langer, Lawrence 64
Union and Progress language
Jèze, Gaston 26 racist and genocidal 294–6, 299
Jonassohn, Kurt 7, 85, 149–50, 160, 170 see also linguicide
Jones, Adam 61, 67–8, 73, 75, 533 Las Casas, Bartolomé de 26–7, 305
Jordan, Winthrop 324 Latin America 625
Jünger, Ernst 76 concept of war in 499
Just War tradition 602, 604 disease in 305–6, 308–10
ethnocide in 311–13, 315
Kabila, Joseph 568, 569 population levels 306–8
Kabila, Laurent 553, 567–8, 645 question of genocide in 505–6
Kaganovitch, Lazar 397 repression in 498
Kaishek, Chiang 454 resistance and subaltern genocide in 310–11,
Kalyvas, Stathis 179 315–19
Kansteiner, Wulf 104, 115 violence in independent 490–1
Kant, Immanuel 203, 226 n. 44, 234, 602, 632 see also forced migration; genocidal intent;
Karadžić, Radovan 72, 521 National Security Doctrine; School of the
Kardelj, Edvard 517, 524 Americas; and individual countries
Karen, the 472–6 Lausanne Peace Conference/Treaty 51, 383
Karim, Abdel 357 law
Kasa-Vubu, Joseph 566 and politics 618–19
Katz, Steven 329, 332 international 131–2, 617–18, 636–7
Kayibanda, Gregoire 558, 573 return of 627
Kazakhstan 395–7, 399, 400, 402, 404 see also ethnic cleansing; genocide;
Kennan, George 583 International Court of Justice;
Kenya 74, 359, 534, 539, 542 International Criminal Court;
index 669
International Criminal Tribunal for MacKinnon, Catherine 70, 71

Rwanda; International Criminal Tribunal Madagascar 52
for the former Yugoslavia; International Plan 422
Military Tribunal; Lemkin, Raphael; Madley, Benjamin 337 n. 66
Nuremberg trials; United Nations; United Maier, Charles 8
Nations Genocide Convention Malinowski, Bronislaw 24–5, 27, 29, 35 n. 61
Lazar, Knez 108 Malkki, Liisa 185–6
League of Nations 22, 25, 27, 30, 580–1 Mamdani, Mahmood 347 n. 7
Lebensraum 412–13, 418–20 Mann, Michael 8, 43, 87–8, 93–4, 144, 159, 170,
Lemarchand, René 174 173–4, 176, 180 n. 44
Lemkin, Raphael 2, 10, 39, 47, 83, 145, 160–1, 167, on ethnic cleansing 154–8
429 n. 11, 639, 657–8 Mariam, Col. Mengistu Haile 534, 536, 537, 539
and biological genocide 314 Marr, Wilhelm 409
and colonialism 322, 346–7, 362 Martin, Terry 396–7
and cultural genocide 37–8, 305, 311–12, 324, Marx, Karl 214
336, 360 Marxism 408, 413, 461
and international law 30–1, 41, 123–4, 192–3 Matilda, countess of Tuscany 281
and physical genocide 310, 313 Matthäus, Jürgen 198
and sociology 146–8 Maxwell-Fyfe, Sir David 125
and the Nuremberg trials 36–7, 92, 125, 127 Mazzini, Giuseppe 23
and the United Nations 37–8, 133 medieval genocide
Axis Rule in Occupied Europe by 22, 26, 31–5, accounts of 284–5
82, 124, 142, 193, 305, 322 elites and 271–2
genocide concept of 45, 149–50, 157–8, 323, evidence for 260–1, 264
445–6, 447, 534 n. 12, 621 nature of 262–4
reputation of 19–20 question of 266–8, 269, 274–6, 277, 286
thought of 21, 22–9 religion and 272
Léopold II, King 362, 558 Medvedev, Gen. 393 n. 23
Leutwein, Governor 92 Mehmed II, Sultan 48
Levene, Mark 8 n. 19, 9, 87–8, 110, 485 n. 55, Melson, Robert 169–70, 176, 177
598, 649 memory
Levi, Primo 198 as fuel for genocide 106–10
Levinas, Emmanuel 226 n. 46 conflicts over 116–18
Lewin, Kurt 200 studies 103–4
Lewy, Guenther 417 traumatic 112–15
Liberia 608 see also philosophy
Libya 568 Méndez, Aparicio 496
liddism 654 Menelaus 252, 255
linguicide 440 n. 29 Menelik II, Emperor 531
Liulevicius, Vejas 410 Merkulov, Vsevolod 400
Livy 244 Merlin 289
London Charter 620–1 Mesha, king of Moab 242
London Conference 125–6 Meštrović, Stjepan 518, 526
Lopez Jr., Donald 427 Mexico 29, 307
Lord’s Resistance Army (LRA) 552, 555, 559, 565–6 Meyer, Aubrey 657 n. 53
Louis XIV, King 47 Micombero, Michel 570
Lowther, Gerard 369 Midlarsky, Manus 170, 177–8
Lumumba, Patrice 552, 556, 566 Migration
Luxembourg 34–5 see forced migration
Milgram, Stanley 200, 205–7, 208–10
McDougal, Governor John 338 military intervention
Mace, James 388 n. 1 and ending genocide 599–601
Macedonia 50 as a duty 602–9
670 index
military intervention (cont.) Neier, Aryeh 71

inhibitors to 609–15 Nennius, Welsh monk and scholar 289
see also Responsibility to Protect principle Neo-Assyrian Empire 47
Miller, Arthur G. 206 n. 23 Netherlands, the 448–9
Milošević, Slobodan 77, 94, 95–6, 419, 510, 519, Neurath, Konstantin von 125
524–5, 526, 632 New England 324–5, 331
indictment and trial of 628–9 New Zealand 40
nationalism of 108, 518 Newman, Leonard 209–10
Mladić, Gen. Radko 68, 72, 514, 522, 523 Ngendandumwe, Pierre 555, 570
Mobutu, Joseph-Désiré 553, 556, 558, 560, Nicaragua 593
566–8, 645 Nicholson, Michael 521
modernity 154, 191 Nietzsche, Friedrich 12, 102, 115, 225, 226
thesis 152–3 Nigeria 194–5, 625, 647
Molotov, Vyacheslav 397 nihilism 218
Mommsen, Hans 208 Nikolić, Tomislav 523
Mongolian People’s Republic (MPR) 435–6, 439 Nimeiri, Jaafar 543
Monroe Doctrine 584 Nithard, Carolingian historian 266, 274
Montenegro 526 Nixon, Richard 470
Morgan, Gen. Mohamed Said Hersi 540 Nizkor 502
Moscow Declaration 620 Nkunda, Laurent 563, 569
Moses, A. Dirk 72, 83, 109–10, 485 n. 55 NKVD 90–1, 399, 400–1, 403, 404
Mujahideen 513, 514 Nol, Gen. Lon 460
Mukti Bahini 468, 469 non-interference principle 604, 610
Museveni, Yoweri 552, 560, 565 Nora, Pierre 103
Mussolini, Benito 357, 359 North America 48, 83
mustard gas 357 colonial and post-colonial violence in
Mutlu, Servet 382 n. 70 330–2, 334–5, 338
Mwambutsa, Mwami 570 colonialism in 323, 324–6
Myanmar historiography of colonialism in
see Burma 327–8, 330
race in post-colonial 334, 335–7
Nagorno-Karabakh see also California; imperialism; indigenous
violence and population displacement in Americans; New England
512–13 Northumbria 259, 261, 262–3, 264
Naimark, Norman 7, 43, 88, 157, 405, 526 Norton, Henry Kittredge 339
Najibullah, Mohammad 527 Norway 36
Nama, the 358, 410 Notker the Stammerer 274, 276
Namibia 568 Novick, Peter 4
nasty neighbour problem 207–8, 215 Nuba Mountains 544, 599
national identity 446, 449, 508, 511 Nuremberg trials 125, 624–5
National Security Doctrine 493–4, 495, 497, 584 see also American Military Tribunal;
development of the 489–90 International Military Tribunal; Lemkin,
political repression under the 491–2 Raphael
NATO 524–5, 583, 593, 599, 602, 608, 628–9 Nürnberg (Nuremberg) Laws 415, 421
Nazism 412 Nyerere, Julius 611
and anthropology 187
and archaeology 186–7 Obote, Milton 552, 554, 560, 564–5
anti-Semitic nature of 109–10 OGPU 394–5
misogynistic nature of 65–6 omnicide 641
see also Jews; Nürnberg Laws O’Neill, Kevin Lewis 194
Ndadaye, Melchior 94, 553, 555–6, 572 Operation Condor 493–4, 495
Ndayizeye, Pres. 572 Opium War, first 433
Ndizeye, Charles 555, 570 Organia, Gen. Juan Carlos 497
index 671
Organisasi Papua Merdeka (OPM) 477, 478 political genocide 449–50, 464
Osiel, Mark 635–6 in Cambodia 450, 459–63
Otto IV, emperor of Germany 288 in China 450, 454–9
Ottoman Empire in Indonesia 450–4, 459
deportations in the 366, 368, 370, 371, 373–4, in the Soviet Union 463
375–6, 378–80 political identity 446–9
ethnic cleansing in the 49–50 political science
forced migration in the 48 comparative method in 168–70, 172
policies of the 366–8, 369 rationalist approach in 174–5
see also Committee of Union and Progress scholarship and genocide 163–4, 165–8, 171–2,
Owen, David 522 173, 179–80
Ozick, Cynthia 64 see also law
politicide 170–1, 446, 540
Pakistan 656 see also sinicization
see also East Pakistan; West Pakistan Polybius 243 n. 12
Palestine 57 Popé, Pueblo Revolt leader 316–17
Palyi, Melchior 142, 144 population displacement
Pan Germans 78 in Bangladesh 652
Panchen Lama, the 438 in Chechnya 514–15
Paraguay 493–4 in Darfur 546
see also Stroessner, Gen. Alfredo in the Great Lakes region 562, 566, 567, 568,
Paris, Matthew 287, 299 569, 571, 573, 575, 646
Parson, William S. 7 in the Soviet Union 510
Pasha, Djemal 99 in Yugoslavia 510, 520, 523
Pasha, Enver 99 see also Nagorno-Karabakh; South Ossetia
Pasha, Talat 99 Porter, Jack Nusan 85
Patterson, David 226 n. 47 Pot, Pol 93, 98, 99, 106, 593
Pavelić, Ante 109 Potsdam conference 54, 58
Pearce, Roy Harvey 324 Powell, Colin 534, 605
Pella, Vespasian 133 Power, Samantha 13
Pequot War 331–3 Přemysl Otakar II, king of Bohemia 291
Perón, Juan 497 Princip, Gavrilo 108
Persia Prussia 297
ancient 249, 255 Psychology
Peru 308, 313–14, 317, 318 see social psychology
Peter the Venerable 299 Pufendorf, Samuel 26
Philip II, king of Macedonia 257 Putin, Vladimir 515, 516
Philip IV, king of France 292, 301
philosophy Qiu, Jin 457 n. 23
and genocide 225–7, 231 Quautlatas, Tepehuan leader 316
and trauma, memory and representation 223–4
scholarship and genocide 217–18 Rae, Heather
see also Dialectic of Enlightenment; pathological homogenization concept
genocidal intent; social death of 88
pillarization 448–9 Rahman, Sheikh Mujibur 469
Pinochet, Gen. Augusto 496 Ranger, Terence 447
Pizarro, Francisco 306 Ranković, Aleksandar 524
Plato 247 rape
Pogge, Thomas 602 n. 18 genocidal 71–2, 73–4
Poland 291, 511–12 in East Pakistan 468
Nazi-occupied 35, 418 in Rwanda 70–1
Soviet-occupied 401–3 in Sudan 544
see also General Government in the ancient world 245–6
672 index
rape (cont.) Rwandan genocide 556, 561, 562, 598, 609, 612–13
in the Great Lakes region 560 memory and the 112–13, 114
in Yugoslavia 69–71, 77, 527 origins and nature of the 107, 153, 191, 574
see also Armenian genocide see also International Criminal
Rappaport, Emil S. 30 Tribunal for Rwanda
Rawski, Evelyn 429
Rawson, Harry 361 Saakashvili, Mikheil 516
Ražnatović, Željko 95–6, 520 St Bernard of Clairvaux 297–8
Red Guards, the 456–8 St Brice’s Day massacre 292
Reingold, Iosif 392 St Columba 274
Renan, Ernest 102 St Patrick 277
Renner, Karl 24 St Paul 262
resource capacity 643–4 Salbi, Zainab 62
see also Democratic Republic of Congo Samnite War
Responsibility to Protect principle 597–8, Second 254
604–6, 608, 611 Third 250
Ribes, Champetier de 105 Sanford, Victoria 189
Rice, Condoleeza 527 Santa Cruz massacre 481, 483
Richter, Daniel 328 Sapor II, Persian king 255
Ricoeur, Paul 102–3, 104 Sarkozy, Nicolas 516
Riffian Berber rebellion 357 Sartre, Jean-Paul 133, 201–3, 346
Ringelblum, Emmanuel 63 Saul, king of Israel 242, 261, 263, 278
Ringelheim, Joan 64 Sautman, Barry 428
Robertson, Geoffrey 624 Saxons 273, 275, 284, 289
Rockefeller Foundation 187 Anglo- 260, 264, 265, 267–8, 272, 276
Rodney, Walter 347 Scales, Len 259, 263, 264, 270
Rogers, William 338 Schabas, William 446
Rohrbach, Paul 355, 361 Schafft, Gretchen E. 142, 187
Romania Scheper-Hughes, Nancy 153, 187
ethnic cleansing in 52–3 Schleunes, Karl 89
Rome School of the Americas 489–90
ancient 248, 250, 251–2, 253–4 Schrodt, Philip 171, 172
Rommel, Erwin 423 Scilingo, Adolfo 500, 502 n. 15
Roosevelt, Franklin D. 620, 625 Scotland 267, 268, 284, 288
Roosevelt, Theodore 597 Seeskin, Kenneth 217 n.
Rorty, Richard 231–2, 233 n. 68, 234 n. 75 Selassie, Emperor Haile 357, 536, 537
Roseman, Mark 207–8 Sémelin, Jacques 170, 176, 177, 180 n. 44, 535
Rudd, Kevin 115–16, 117 Sennacherib of Assyria 47, 246, 248, 250, 253, 255
Rugova, Ibrahim 524 Serbia 108–9, 129, 140, 593, 606
Rummel, Rudolph 158, 171, 172, 173 see also Milošević, Slobodan
Russia 611, 626 Serbian Democratic Party (SDS) 520, 521–2
deportations and ethnic cleansing in Serbian Radical Party 524
388–91, 392–3 Šešelj, Vojislav 95–6
forced migration in imperial 48, 49, 50 Seven Years War 49
violence in 387 Shaka, Zulu king 353
see also Chechnya Shapoval, Juri 396
Russian Civil War 391–2 Shaw, Martin 8, 175, 176–8
Rwagasore, Louis 555, 570 Shawcross, Sir Hartley 36
Rwanda 558, 592, 603, 648 Shea, Jamie 628
violence in 553, 555, 556, 573–4, 575 Shirer, William L. 142
see also Habyarimana, Juvénal; Hutu; rape; Sierra Leone 75
Tutsi Sigismund, Bohemian king 295
index 673
Sihanouk, Prince Norodom 98, 106, 460 Stanford Prison Experiment (SPE) 205 n. 21,
Simpson, Gerry 617 206–7, 209
sinicization 443 Stannard, David 330
and politicide 429–36 state, the
through Nationality Regional Autonomy responsibility of to prevent genocide 139–40
(NRA) 436–41 role of in genocide, ethnic cleansing and
Sitglmayer, Alexandra 70 mass violence 55–6, 85–8, 89–90, 91, 97,
Slavs 289, 391, 424 99–100, 155, 163
smile problem, the 202, 205–7, 208, 210, 215 state-sponsored violence 484–5
Smith, Roger W. 77, 174 as genocide 486–7
social Darwinism 355, 360, 408–9, 650 in Burma 472–6
social death 222–3 in East Pakistan 467–71
social psychology in East Timor 479–83
situationist 199–200, 209–10 in West Papua 476–9
see also choice problem; conformity effect; Staub, Erwin 170
nasty neighbour problem; smile problem; Stein, Stuart 68
Stanford Prison Experiment Stephen, king of England 290
sociology 166 Stephen of Ripon 275
and history 158–9 Stevenson, E. A. 336
Nazi-era 143 Stockholm Forum 111
scholarship and genocide 144, 148–50, 160 Stolypin, P. A. 106
see also ethnic cleansing; genocidal intent; Stone, Dan 6
Lemkin, Raphael Stone, John 331–2
Solomon, biblical King 531 Straus, Scott 86, 176, 178, 180 n. 45, 192,
Solomon, Charles 26 208, 485 n. 53
Somalia 529, 532, 533, 540–2 Stroessner, Gen. Alfredo 492, 493–4
Somaliland 541 Su, Yang 458
Sorabji, Cornelia 114 Suazo, Hernán Siles 495
South Ossetia 510, 527 Sudan 356, 359, 361, 529, 547–8, 568,
population displacement and ethnic cleansing 630–1, 647
in 515 South 532, 534, 545
Soviet Union 419–20, 422, 455 violence in 531–2, 534, 542–5
and the United Nations 583 see also Darfur; Islam; Janjaweed
composition of the 511 Suharto, Pres. 450–1, 481
deportations and ethnic cleansing in the 53, Sukarno, Pres. 450–2
394–5, 398–405
minorities in the 387, 400, 403 Talaat, Mehmed 371, 373, 374, 379,
Stalinist 90–1, 93, 106 380–1, 383, 384–5
see also Cold War; de-kulakization; Great Taliban 527
Famines; political genocide; population Tanzania 561, 565, 593, 611
displacement; Russia Taylor, Christopher C. 76, 191–2
Sovietization 400–1 Tec, Nechama 65
Spain 47, 115, 246 Tejada, Luis Garcı́a Meza 495
Special Organization (TM) 365–6, 371, Tertullian, Christian theologian 447
372, 377, 383 Thornton, Russell 329
Srebrenica 54, 59, 136–7, 592 Thucydides 247, 249, 253, 254
massacre 46, 68–9, 129, 518, 523, 599 Tianhui, Huang 458
see also International Criminal Tribunal for Tiglath-pileser III, king of Assyria 47
the former Yugoslavia Tiglat-Pileser I, king of Assyria 240
Stalin, Josef 53, 98, 99, 393–4, 396–8, 419, 455, 511 Tigray People’s Liberation Front (TPLF)
see also Soviet Union 537, 547
Stanford, Californian governor 337 Tito, Josip Broz 517, 524
674 index
Tocqueville, Alexis de 356 and the United Nations 580, 583, 594
Torres, Gen. Juan Jose 495 see also Cold War; National Security
Total 474 Doctrine; North America
Totten, Samuel 7, 533 universal jurisdiction principle 589
Toynbee, Arnold 657 Unocal 474
Transcaucasus 510 Urban II, Pope 296
ethnic cleansing in the 54–5 urbanization 283, 293–4
Transdniester 511 Uruguay 495–6
Trojan War 239–40, 246, 247, 255 Ustasha/Ustaša, the 53, 495, 518, 519
Trotha, Gen. Lothar von 91–2, 354–5 Uzbekistan 589 n. 12
Trujillo, Rafael 494
Tudjman, Franjo 109, 514, 518, 519–21, 527 Valentino, Benjamin 98, 171–2, 173–6, 177
Turkey 10 Van Dyke, Walter 337
see also Armenian genocide; Committee Vance, Cyrus 522
of Union and Progress; Cyprus; Vanuatu 479
Greco-Turkish war; Ottoman Empire Vassiliev, Vassili 396
Tutsi 554, 560 Vattel, Emeric de 26
see also Hutu Verwey, Wil 608, 609
Tuvalu 654 Videla, Gen. Jorge Rafael 497, 503–4
Tyrell, Ian 340 Vietnam 93, 95, 460, 471 n. 10, 593, 609, 610–11
violence
Uganda 569 medieval 296–8
violence in 552, 555 medieval inter-ethnic 292–4
see also Amin Dada, Gen. Idi; Lord’s patterns of 640
Resistance Army; Obote, Milton relational 72–5
Ukraine, the 396–8 see also state-sponsored violence; and
Underhill, John 331 individual entries
United Kingdom 40, 481, 646 Vitalis, Oderic 290
see also Blair, Tony; Great Britain Vitoria, Francesco de 26–7
United Nations (UN) 47, 482 Volkov, Shulamit 411
and international law 586–7
Charter 581–2, 590–2, 603, 605, 607, 610, Waal, Alex de 613–14
621, 630 Walder, Andrew G. 458
Declaration on the Rights of Indigenous Wales 267, 291, 294, 302
Peoples 39–41 Wallerstein, Immanuel 347 n. 8, 640–41
origins and development of the 580–1 Wangenheim, German captain 357
resolution on genocide (1946) 1, 37, 621 War on Terror 479, 498, 614
Security Council 44, 581–2, 583, 585, Warren, Mary Ann 68
590–3, 594, 603, 604–8, 610, 611, Warsaw Pact 583
612–13, 625–6, 628, 630 Washington, George 324
Universal Declaration of Human Rights 586 Waxman, Zoë 64
see also League of Nations; Weber, Max 143, 145, 152, 158–9, 161, 633
Lemkin, Raphael Weiss, Thomas 603
United Nations Genocide Convention Weiss-Wendt, Anton 21 n. 6
(UNGC) 10–11, 78, 105, 124, 131, 585 Weitz, Eric D. 7, 170, 176, 485
content of the 45, 72, 92, 138, 147, 604 Weller, Californian governor 338
development of the 82–3, 127, 133, 134–5, Wendish Crusade 298
193–4, 621–3 Wenskus, Reinhard 269–70
nature of the 86, 128–9 Wernich, Christian Von 500 n., 501
shortcomings of the 112, 130–1, 319–20, West Pakistan 468–9
405, 445 Wheeler, Nicholas J. 601 n. 12, 602 n. 14, 609
United States of America 13, 40, 470, 481–2, Whitaker, Benjamin 136–7
614, 633–4, 654 White, Richard 330
index 675
Wiesel, Elie 224 Yugoslav People’s Army/Yugoslavian National

Wilhelm II, Kaiser 91–2, 355 Army (JNA) 95, 520, 521–2, 524
Wilson, Sir Ronald 116 Yugoslavia 36
Windschuttle, Keith 116 ethnic cleansing in 43–4, 54, 520
Winter, Jay 93 genocide in 516–18, 526–7
Winthrop, John 329, 333 see also Bosnia; Croatia; International Court
Wolfe, Patrick 326–7, 330, 334, 340 of Justice; International Criminal Tribunal
Wolff, Christian 26 for the former Yugoslavia; Kosovo;
Woolf, Alex 275 Montenegro; population displacement;
Wright, Mary 429 rape; Serbia; Srebrenica
Yugoslavian Army (VJ) 524–5
Xhosa, the 352–3
Xiaowen, Chinese emperor 432 Zaire 556, 563
Zedong, Mao 436, 439, 454, 455–8
Yagoda, Genrikh 91 Zimbabwe 568
Yandarbijev, Zelimkhan 515 Zimbardo, Philip 200, 205–7,
Yanushkevich, Gen. 389, 391 209–10, 213
Yeltsin, Boris 514–15 Zuckerman, Michael 328
Young, James E. 112, 117 Zulu, the 348, 350, 353

The Oxford Handbook of Genocide Studies (PDFDrive) PDF

Uploaded by

Copyright:

Available Formats

The Oxford Handbook of Genocide Studies (PDFDrive) PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Oxford Handbook of Genocide Studies (PDFDrive) PDF

Uploaded by

Copyright:

Available Formats

the oxford handbook of

Editor’s Introduction: Changing Themes in the Study of Genocide 1

1. Raphael Lemkin, Culture, and the Concept of Genocide 19

2. ‘Ethnic Cleansing’ versus Genocide? 42

3. Gender and Genocide 61

4. The State and Genocide 81

5. Genocide and Memory 102

PART II: INTERDISCIPLINARY PERSPECTIVES

6. The Law and Genocide 123

7. Sociology and Genocide 142

8. Political Science and Genocide 163

9. Anthropology and Genocide 182

10. Social Psychology and Genocide 198

11. Philosophy and Genocide 217

PART III: PREMODERN AND EARLY

12. Genocide in the Ancient World 239

14. Central and Late Medieval Europe 280

15. Colonial Latin America 304

16. Rethinking Genocide in North America 322

PART IV: GENOCIDE IN THE LATE MODERN WORLD

17. Genocide and Mass Violence in the ‘Heart of Darkness’:

18. Genocide at the Twilight of the Ottoman Empire 365

19. Mass Deportations, Ethnic Cleansing, and Genocidal

20. The Nazi Empire 407

21. Twentieth-Century China: Ethnic Assimilation and Intergroup Violence 426

22. Political Genocides in Postcolonial Asia 445

23. State-Sponsored Violence and Secessionist Rebellions in Asia 466

25. Genocide and Population Displacement in Post-Communist 509

26. Genocidal Warfare in North-East Africa 529

PART V: THE CONTEMPORARY WORLD:

29. Military Intervention 597

30. Punishment as Prevention? The Politics of Prosecuting Génocidaires 617

31. From Past to Future: Prospects for Genocide

Alex J. Bellamy is Professor of International Relations and Executive Director of

Dominik J. Schaller is a lecturer and researcher at the Ruprecht-Karls-Universität

T H E S TAT E OF GENOCIDE STUDIES

1 E.g. Online Enyclopedia of Mass Violence: http://www.massviolence.org

Another strand of scholarship pointed in a different direction. If one term

have been at least as adept as historically oriented scholars in devising increasingly

one-genocide-per-chapter empirical studies and broad thematic/conceptual stud-

reference to European settler societies in colonial contexts, he debunks the spurious

scientiﬁc scholarship that all human communities of identity are constructed to

GENOCIDE AND THE W O R L D T O DAY

inﬂammatory as well as paciﬁcatory, has been hailed as a potential panacea to

For this reason, Lemkin tended to associate ‘destruction’—a word he preferred to

of international humanitarian law, which is dominated by the discourse of human

only an anachronistic return to ‘medieval organic imagery’ or fundamental confu-

Undergirding the protection of group existence against extermination, then, is the

ethnography invented in the service of nationalism, which had practised “going to

Consequently, he concluded, ‘the destruction of cultural symbols is genocide.’ To

Conquest, Occupation, and Cultural Change

Because genocide so often occurred in contexts of conquest and occupation,

tradition of natural law-based critique of imperialism rather than solely in Lemkin’s

Consequently, Lemkin was disturbed by occupations like German colonial rule

Criticisms that Lemkin’s conception is imprecise or incoherent, and that therefore

31 Lemkin, ‘The Concept of Genocide in Anthropology’.

A sentence later, however, he seems to restrict genocide to extermination,

So does genocide mean exterminating or ‘crippling’ a people? He begins Chapter

We do not seem closer to a clear answer.