LONDON SCHOOL OF ECONOMICS

AND POLITICAL SCIENCE

Whose Kurdistan?
Class Politics and Kurdish Nationalism
in the Middle East, 1918-2018

Nicola Degli Esposti

A thesis submitted to the Department of International Relations of the

London School of Economics and Political Science for the degree of Doctor
of Philosophy.
London, 13 September 2020
Declaration
I certify that the thesis I have presented for examination for the MPhil/PhD
degree of the London School of Economics and Political Science is solely
my own work other than where I have clearly indicated that it is the work of
others. The copyright of this thesis rests with the author. Quotation from it
is permitted, provided that full acknowledgement is made. This thesis may
not be reproduced without my prior written consent. I warrant that this
authorisation does not, to the best of my belief, infringe the rights of any
third party.
I declare that my thesis consists of 98,640 words.

2
Abstract

This thesis is a study of the different trajectories of Kurdish nationalism in

the Middle East. In the late 2010s – years of momentous advance for Kurdish
forces in Turkey, Iraq, and Syria – Kurdish politics was deeply divided into
competing movements pursuing irreconcilable projects for the future of the
Kurdish nation. By investigating nationalism as embedded in social conflicts,
this thesis identifies in the class basis of Kurdish movements and parties the
main reason for their political differentiation and the development of
competing national projects. After the defeat of the early Kurdish revolts in
the 1920s and 1930s, Kurdish nationalism in Iraq and Turkey diverged along
ideological lines due to the different social actors that led the respective
national movements. In Iraq, the Kurdish national movement that emerged
in the early 1960s was largely dominated by the tribal and landowning elite
and primarily interested in preserving existing social hierarchies, while
middle-class progressive nationalists were systematically sidelined.
Conversely, in Turkey, as the Kurdish traditional elite had been co-opted
into the Kemalist state in the 1950s, the nationalist struggle was resumed in
the late 1970s by a generation of Kurdish students of peasant extraction
that framed their project as an anti-colonial struggle and that violently
opposed both Turkish security forces and Kurdish landlords. The thesis
argues that these opposing trajectories of Kurdish nationalism were
influenced by the evolving class structure of the newly established states
of Turkey and Iraq, and by their location in the international state system
and within global capitalism. The analysis of the social origins of competing
‘nationalisms’ provides a novel approach to the study of nationalism based
on Historical Sociology and Political Economy and offers a materialist
reading of the history of Kurdish politics.

3
Acknowledgements

First of all, I need to thank my supervisor Dr Katerina Dalacoura who, in the

best LSE tradition of theoretically engaged and empirically grounded
scholarship, tried to teach me for the past four years how to think as a social
scientist.
I am indebted to my friends and colleagues at the IR Department, the Middle
East Centre, and the Contentious Politics Workshop, for all their valuable
feedback and advice. Andrew Delatolla, my PhD buddy, was of great
support to my early disoriented months on campus. I am particularly grateful
to my cohort mates, Eleonore Heimsoeth, Alessandro Guasti, and Jackie
Majnemer because of what I learnt from them and, most importantly, for
never letting me panic alone.
My fieldwork in Kurdistan was a greatly enriching experience and I would
like to thank all the people I met and interviewed there. Particularly, Mera
Jasm Bakr for his help organising and translating my interviews, and for his
friendship.
I was very lucky to find, as I landed in London, great old friends like Lorenzo
D’Agostino and Marwan Shehadi, and to immediately meet new ones, like
Luke Rees, Tom Seal, and a handful more. Good friends do not grow on trees
and I could only make it through this PhD because they let me forget about
it once in a while. I am also grateful to the friends and comrades left in
Faenza, Bologna, Leiden, Ankara, Istanbul, and Sulaymaniyah, for the
countless nights of smoke, wine, and politics which somehow led me here.
I would like to thank my family for their support, my sister Federica, her
partner Lanfranco, and my new-born niece Emilia. My parents Donatella and
Piero worked days and nights to give us the opportunities they never had.
They passed on to us their values, taught us to be curious about the world
and to stand up for our ideas.
Finally, the biggest thanks of all goes to Evelyn Pauls, the most important
encounter I had during this PhD journey. Her intelligence pushes me every
day to question what I know and what I think, and her support makes me
believe a little more in myself.

This thesis is dedicated to my grandmother Lucia who passed away while I

was in London. I could have never imagined that this PhD on Kurdish
nationalism would make me think so much about her life and my own family
history of urbanised landless peasants. That she could be proud of me,
despite the oceanic distance separating what I do from her long life of toil
in the field, the factory, and the house, is for me the greatest honour and
motivation to continue.

4
Table of content

Declaration .................................................................................................................................................. 2
Abstract ....................................................................................................................................................... 3
Acknowledgements .................................................................................................................................... 4
Table of content .......................................................................................................................................... 5
List of images .............................................................................................................................................. 8
Maps ........................................................................................................................................................ 8
Tables ...................................................................................................................................................... 8
List of Abbreviations ................................................................................................................................... 9
Kurdish Organisations and Institutions ................................................................................................. 9
Non-Kurdish Organisations and Institutions ....................................................................................... 10
Chapter 1 Introduction ............................................................................................................................. 12
The Purpose and Questions of this Thesis ......................................................................................... 12
Kurdish Nationalism and the Kurdish Question .................................................................................. 15
The Kurds and Kurdistan...................................................................................................................... 18
The Literature on Kurdish Nationalism ................................................................................................ 26
Methodology and Methods .................................................................................................................. 33
The Structure of the Thesis ................................................................................................................. 38
Chapter 2 Nationalism and Class Politics ............................................................................................... 41
Introduction ........................................................................................................................................... 41
Diffusionism and Eurocentrism ............................................................................................................ 42
Is Nationalism an Autonomous Force? ................................................................................................ 49
Culture, Ideology, and the Struggle for State Power ......................................................................... 53
Studying Class ...................................................................................................................................... 59
Sketches of Nationalisms..................................................................................................................... 65
Classical Bourgeois Nationalism ..................................................................................................... 66
Middle-Class Nationalism ................................................................................................................ 68
Feudal Nationalism ............................................................................................................................71
Nationalism and the Subaltern Classes .......................................................................................... 73
Conclusion............................................................................................................................................. 76
Chapter 3 Feudal Nationalism in Kurdistan (1918-1946) ....................................................................... 78
Introduction ........................................................................................................................................... 78
Kurdistan in the Late Ottoman Period ................................................................................................. 79
Defining Feudal Nationalism ................................................................................................................ 84
Kurdish revolts in Kemalist Turkey ...................................................................................................... 90
Kurdish revolts in Hashemite Iraq ....................................................................................................... 96
Conclusion........................................................................................................................................... 102
Chapter 4 Land Reform and Kurdish Revolt in Postcolonial Iraq (1946-1991) ................................... 106
Introduction ......................................................................................................................................... 106
Class and Politics in Colonial Iraq ...................................................................................................... 107
The Iraqi Revolution of 1958 ............................................................................................................... 114
Land reform and Kurdish Revolt ......................................................................................................... 117
5
 The Cold War, and the Collapse of the Kurdish Revolution ............................................................ 126
Iraq’s Two Kurdish Nationalisms........................................................................................................ 129
War, Genocide, and Liberation .......................................................................................................... 135
Conclusion........................................................................................................................................... 138
Chapter 5 The Class Structure of Kurdish Self-Rule in Iraq (1991-2014) ............................................ 141
Introduction .......................................................................................................................................... 141
Liberation and Civil War ..................................................................................................................... 142
Urbanisation and Dependence .......................................................................................................... 146
Warlords and Chiefs: the Formation of a New Ruling Class ............................................................ 149
The 2003 Invasion of Iraq and the unification of the KRG .............................................................. 156
The Kurdish Economic Boom and Its Contradictions ....................................................................... 159
The Promises of Kurdish Crude ......................................................................................................... 164
Subaltern Classes and Opposition in a Rentier (Quasi-)State ......................................................... 167
Conclusion............................................................................................................................................ 171
Chapter 6 The Crisis of Kurdish Nationalism in Iraq (2014-2019) ....................................................... 173
Introduction ......................................................................................................................................... 173
War and Economic Crisis ................................................................................................................... 174
Protest and Political Crisis ................................................................................................................. 176
The Evolving Power Structure of the KRG ........................................................................................ 180
External Relations and Domestic Rule ............................................................................................... 184
The 2017 Independence Referendum ............................................................................................... 189
The Crisis of Kurdish Nationalism ..................................................................................................... 193
Conclusion........................................................................................................................................... 197
Chapter 7 Underdevelopment and the Kurdish Question in Cold-War Turkey (1946-1987) ............ 201
Introduction ......................................................................................................................................... 201
The Kurds in the Early Kemalist Republic ......................................................................................... 202
The ‘Agha-State’ Alliance and the Incorporation of the Kurdish Elite ............................................. 204
The Underdevelopment of the Kurdish Region ................................................................................ 209
Kurdish Nationalism and Turkish Socialism ...................................................................................... 212
The Kurdish Left in the 1970s ............................................................................................................ 216
Preparing the Kurdish Revolution ...................................................................................................... 223
The PKK Insurgency and the Kurdish Peasantry .............................................................................. 227
Conclusion........................................................................................................................................... 229
Chapter 8 The PKK Insurgency and the Transformation of Turkish Kurdistan (1987-1999) ............ 231
Introduction ......................................................................................................................................... 231
Military Rule and Neoliberal Reforms in Turkey ............................................................................... 232
The Destruction of Kurdish Rural Society ......................................................................................... 236
The Kurdish Insurgency in the 1990s ................................................................................................ 239
The ‘Urbanisation’ and Ideological Transformation of the PKK .......................................................243
The Kurdish pro-Democracy Movement and its Social Basis ......................................................... 245
Kurdish Diaspora and Exile in Europe................................................................................................ 250
The Capture of Abdullah Öcalan ....................................................................................................... 254
Conclusion........................................................................................................................................... 256
Chapter 9 The Political Economy of the New Kurdish Movement (2000-2019) ................................ 258
Introduction ......................................................................................................................................... 259

6
 From the PKK to the New Kurdish Movement .................................................................................. 260
Öcalan’s New Ideological Paradigm .................................................................................................. 262
Gender, Class and the Kurdish Women’s Movement ....................................................................... 266
Hegemonic Struggle in Turkish Kurdistan ........................................................................................ 272
The Rojava Experiment ...................................................................................................................... 279
The Political Economy of the Kurdish Movement ............................................................................ 282
Conclusion........................................................................................................................................... 287
Chapter 10 Conclusion ........................................................................................................................... 289
Introduction ......................................................................................................................................... 289
Comparing Kurdish Nationalism in Iraq and Turkey ......................................................................... 290
Social Classes and Kurdish Nationalism ........................................................................................... 296
Feudal Nationalism ......................................................................................................................... 297
Middle-Class Nationalism ..............................................................................................................300
Nationalism and the Subaltern Classes ........................................................................................ 302
Concluding Remarks........................................................................................................................... 305
List of References ................................................................................................................................... 307
Books, Journal Articles and Edited Volumes .................................................................................... 307
Newspapers and Magazines Articles ................................................................................................324
Reports ................................................................................................................................................ 329
Unpublished Documents, Webpages and Other Items ....................................................................332
List of Interviews .....................................................................................................................................335

7
List of images

Maps
Map 1 – The Kurds at the End of the Ottoman Empire………………………………………………………………………11
Map 2 – Kurdish Region of Iraq and Major Tribes………………………………………………………………………….105
Map 3 – Kurdish Autonomy in Iraq………………………………………………………………………………………………..140
Map 4 – The Kurdish Insurgency in Turkey………………………………………………………………………………….200
Map 5 – Kurdish Politics in the 2010s………………………………………………………………………………………….258

Tables
Table 1 – Estimates on the Size of the Kurdish Population…………………………………………………………….26
Table 2 – Iraq’s Largest Landowning Families in 1958…………………………………………………………………..111
Table 3 – Pro-Kurdish Political Parties in Turkey…………………………………………………………………………249

8
List of Abbreviations

The acronyms for political organisations are used in the language they are
most commonly known. Normally, names of organisations that are originally
written in the Latin script (for example, Turkish or Kurdish Kurmanji) keep
the original acronym while names originally written in the Arabic script (for
example, Arabic, Persian, Sorani Kurdish) take the acronym of their English
translation. The following list specifies the country in which the organisation
is predominantly active unless the name of the country is already present in
the organisation’s name or they are transitional organisations.

Kurdish Organisations and Institutions

BDP Peace and Democracy Party (Turkey)
DBP Democratic Regions Party (Turkey)
DDKO Revolutionary Cultural Societies of the East (Turkey)
DEHAP Democratic People's Party (Turkey)
DEP Democracy Party (Turkey)
DTK Democratic Society Congress (Turkey)
DTP Democratic Society Party (Turkey)
HADEP People’s Democracy Party (Turkey)
HDP Peoples' Democratic Party (Turkey)
HEP People’s Labour Party (Turkey)
KCK Kurdistan Communities Union
KDP Kurdistan Democratic Party (Iraq)
KDPI Kurdistan Democratic Party of Iran
KDPS Kurdistan Democratic Party of Syria
KRG Kurdistan Regional Government (Iraq)
KUK National Liberators of Kurdistan
PDKT Kurdistan Democratic Party of Turkey
PJAK Kurdistan Free Life Party (Iran)
PKK Kurdistan Workers’ Party (Turkey)
PSK Kurdistan Socialist Party (Turkey)
PUK Patriotic Union of Kurdistan (Iraq)
PYD Democratic Union Party (Syria)

9
Non-Kurdish Organisations and Institutions
AKP Justice and Development Party (Turkey)
ANAP Motherland Party (Turkey)
CHP Republican People’s Party (Turkey)
DP Democrat Party (Turkey)
DYP True Path Party (Turkey)
ECHR European Court of Human Rights
ICP Iraqi Communist Party
IMF International Monetary Fund
ISIL Islamic State of Iraq and the Levant
MHP Nationalist Action Party (Turkey)
NDP National Democratic Party (Iraq)
NGO Non-governmental organisation
OHAL Governorship of Region in State of Emergency (Turkey)
RP Welfare Party (Turkey)
SCP Syrian Communist Party
SHP Social Democratic Populist Party (Turkey)
TİP Workers’ Party of Turkey
UN United Nations

10
Chapter 1
Introduction

The Purpose and Questions of this Thesis

This thesis was conceived and written in years of momentous change for
the Middle East and the Kurds. After the defeat of the Islamic State of Iraq
and the Levant (ISIL) in Kobanî in January 2015, a large Kurdish-controlled
region emerged in north-eastern Syria, alongside the one that had existed
in northern Iraq since 1991. In the meantime, almost one hundred
municipalities in south-eastern Turkey were ruled by a pro-Kurdish party.
Even if the Kurds were still a people without a state, in the mid-2010s, they
reached a degree of self-rule without precedent in modern times. Yet,
Kurdish politics was fragmented and divided. The border between the
Kurdish-controlled regions of Iraq and Syria was closed, preventing much-
needed supplies from reaching the ISIL frontline. In Syrian Kurdistan, the
Kurdish opposition linked to the Iraqi Kurds was silenced and repressed
while forces sympathising with the Kurdish struggles in Syria and Turkey
were banned in Iraqi Kurdistan. The Kurdish Regional Government (KRG) of
Iraq suspended the regional parliament in summer 2015 and Kurds took to
the streets on a daily basis protesting their own Kurdish rulers. The Kurdish
movement in Iraq, on the one hand, and those in Turkey and Syria, on the
other, are divided by profound ideological differences and opposite political
projects for the future of the Kurds. The ruling Kurdish parties in Iraq are
pursuing, through semi-authoritarian means, a classic nation-building
project aimed at secession from Iraq and the integration of Iraqi Kurdistan
into the global economy. The dominant Kurdish forces in Turkey and Syria
have developed a project of community-based stateless democracy that
rejects nationalism and yet there are significant doubts on its compatibility
with a basic level of political pluralism. Rooted in divergent historical paths
of development dating back to the partition of the Kurdish land after the
dissolution of the Ottoman Empire, this divide seems far stronger than the
feeling of pan-Kurdish solidarity. Pursuing opposite nation-building projects,
the dominant Kurdish forces are locked in a political and military competition
that places them on the opposite fronts of the zero-sum geopolitics of the
Middle East.

These internecine conflicts fit well into the Orientalist picture of a Middle
East dominated by primordial allegiances and tribal, ethnic, or sectarian
identities – an idea that still holds ground in the academy and dominates the
policy circles informing great-power strategy on the region. In the Orientalist
narrative produced in the West and developed in parallel to colonial
expansion, nomadic and tribal people were depicted as noble warriors
whose tradition of freedom allowed them to resist Oriental despotisms and
yet condemned them to a primitive and violent existence. Lawrence of
Arabia crystallised this view in his famous tirade against the Bedouins: “So
long as the Arabs fight tribe against tribe, so long they will be a little people,
a silly people, greedy, barbarous, and cruel.”1 These ideas were deployed to
justify colonial tutelage and yet were not alien to the pre-colonial Middle
East. The nomadic inhabitants of imperial and state borderlands – like the
Kurds but also the Bedouins or the Berbers – were often portrayed by the
neighbouring settled peoples who dominated them as unruly tribal warriors
incapable of developing advanced forms of politics and society. In the
1950s, Iraqi Prime Minister Nuri al-Said claimed that, to quell the revolt of a
Kurdish tribe, he simply had to “send a bag of gold to a neighbouring chief.”2
These stereotypical readings of Middle Eastern politics feed culturalist
understandings of the region as the seat of ancestral conflicts that resist
modern and rational modes of politics and prevent the spread of democracy
and the free market. By locating the sources of contemporary conflicts in
the unchangeable character of the Kurds and their region, these readings
erase politics.

1
David Lean (dir.), Lawrence of Arabia (Columbia Pictures, 1962).
2
Cited in David McDowall, A Modern History of the Kurds (London: I.B. Tauris,
2004), 10.
13
To counter this depoliticising narrative, this thesis seeks an alternative
explanation to the present divide within Kurdish nationalism capable of
reconnecting contemporary Kurdish politics with the history and political
economy of the Middle East region. As this study will show, all Kurdish
nationalist groups aim at the self-determination of the Kurdish people and
hold a similar understanding of the Kurdish nation – who the Kurds are and
where Kurdistan is. Despite these apparently fundamental commonalities,
the relationship among these groups is conflictual to the point that
competition overshadows forms of cooperation. This contradiction begs
questions on the origins and resilience of the present conflicts within
Kurdish nationalism: What determines the political divides within a national
movement? This question can be further unpacked to explain the current
state of Kurdish politics as well as to engender broader theoretical
considerations: What is the origin of conflicts among Kurdish nationalists
and why are they so resilient? What is the origin of the alternative and
competing nation-building projects they pursue? What is nationalism if it can
be associated with the most diverse set of ideological propositions? What
determines the political content of nationalist movements?

By adopting an approach based on Historical Sociology, this study dissects

the competing Kurdish nationalist projects that developed in the twentieth
century and finds their roots in class politics. In a century of deep change
that altered Kurdish society beyond recognition, social classes were
transformed by the long-term spread of capitalist social relations, the
encroachment of colonialism, the Cold War, as well as the geopolitical
partition of the Kurdish lands and their incorporation into the four nation-
states of Turkey, Iraq, Iran, and Syria. Changing relations of power within
Kurdish society triggered the development and diversification of alternative
nationalist projects whose influence depended on their capacity to give
political expression to existing class interests. Class struggle is by no means
the only driver of politics but the history of Kurdish nationalism shows that
the power of political movements goes hand in hand with their capacity to
give voice to and draw their strength from material grievances and

14
demands. Drawing these considerations from the history of the Kurdish
national movement and scholarly debates on nationalism, I suggest that we
cannot understand the competing forms of Kurdish politics in the Middle
East without understanding their class origins and basis.

This introductory chapter explains the purpose of this study by highlighting

the issues of contention concerning the Kurdish Question, the Kurds and
Kurdistan, as well as the scholarly interpretations of the rise of Kurdish
nationalism. After suggesting an appropriate methodology based on
Historical Sociology, I explain the choice to select the history of the Kurdish
movements in Turkey and Iraq as the case studies. The introduction ends
with an outline of the organisation of the thesis around seven pivotal
moments in the history of Kurdish nationalism in the past century.

Kurdish Nationalism and the Kurdish Question

This thesis investigates the class politics that lies behind the competing
Kurdish nationalist projects that developed over the past century. It is a
study of Kurdish nationalism that treats nationalism as a historical product
the meaning of which shifted depending on which social group raised the
national flag. The next section highlights the complexity of pinning down the
Kurds and Kurdistan but also shows that we can adopt some viable
definitions without reproducing the nationalist paradigm that presents Kurds
and Kurdistan as natural or perennial entities. In these terms, this is a study
of Kurdish nationalism and not a national history of the Kurds, a distinction
reflecting a fundamental methodological concern that drives the research
and shapes its scope.

National histories, defined as “one of the most successful exports of Europe

in the imperial age,”3 inherently create and reproduce national teleologies
that present the nation as a natural fact, 4 endowed by history with unique

3
Stefan Berger and Chris Lorenz, The Contested Nation : Ethnicity, Class, Religion
and Gender in National Histories (New York: Palgrave Macmillan), 2.
4
In these terms, natural does not only mean biological in the sense of informed by
some form of pseudo-scientific racism. Naturalising the nation means also
15
features and often a unique mission that distinguishes it from all other
nations. Nationalism thus becomes the historically inevitable process of the
‘awakening’ of an objectively existing and yet dormant nation. Nationalist
historians “read national histories backward” to find fundamental dates that
mark the stages of national formation and the awakening of national
consciousness.5 In these terms, national histories are not only controversial
from a political perspective, as a form of historiography that is in itself the
product of nationalism and that is deployed as a tool of nation-building.
They are also problematic in analytical terms. By treating nationalism as an
autonomous force only caused by the very existence of the nation, they
conflate explanandum and explanans, producing a circular narrative in
which the awakening of national identity is both the object of investigation
and its explanation.

In this thesis, I try to do the opposite. While acknowledging the existence of

the Kurdish nation as a historical and not as a natural fact, Kurdish
nationalism is studied here with a sociological lens to identify the historical
contexts of its development and the reasons for its ever-shifting meaning in
the material interests it serves and legitimises. The different and competing
forms that Kurdish nationalism took in over a century of history are treated
as alternative political projects rather than stages of the same process. The
best antidote to the idealism inherent in national histories is a materialist
approach that allows us to discuss Kurdish society as a locus of conflicts
and opposing interests rather than an organic community whose threats can
only come from outside or from an immature national consciousness.

The first consideration that follows the commitment to avoid writing a

national history is that being a Kurd implies by no means being a Kurdish
nationalist. At no point in history have Kurdish nationalist movements
represented the totality of the Kurds. The past century of struggles is a

identifying its origins in a remote past that far preceded the awakening of modern
nationalism.
5
Stefan Berger, Mark Donovan, and Kevin Passmore, eds., Writing National
Histories: Western Europe Since 1800 (London and New York: Routledge, 1999),
10.
16
history filled with uncountable examples of Kurds who, because of tribal
rivalry, class antagonism, or ideological differences, fought against the
national movement or that simply identified with the states from which they
were supposed to be liberated. If these considerations apply to virtually all
examples of national liberation movements, the contested nature of
nationalism in the Kurdish case is particularly significant for two reasons. In
the first place, the dominant role of tribal structures in the Kurdish region
led to the inevitable overlapping of any new form of political divide with
older tribal rivalries. This meant that in virtually all cases in which a Kurdish
tribe took part in the national movement, rival tribes were driven to the
opposite side regardless of ideological considerations. Secondly, the
partition of Kurdistan between Turkey, Iraq, Iran, and Syria after World War
I put these states in the position to generously reward Kurdish loyalism at
home while supporting Kurdish nationalism abroad to destabilise their
neighbour with the consequence of deepening the divide between
nationalist and non-nationalist Kurds.

The next chapter will outline the theoretical framework of this study by
extensively reviewing established theories of nationalism. The aim is to find
avenues that avoid treating nationalism as an autonomous force and allow
studying it as embedded in political and social conflicts. Ernest Gellner
defined nationalism as “a political principle that holds that the political and
the national unit should be congruent.”6 The next chapter will problematise
this popular definition and particularly its implications on the relationship
between nationalism and culture. However, Gellner’s take on nationalism is
a useful starting point for it suggests that nationalism always involves a re-
definition of political boundaries and not simply the expression of a cultural
identity. This means that the existence of a collective cultural identity, that,
in the Kurdish case, dates back to literary work of the sixteenth century, is
a necessary but insufficient condition for nationalism to develop.
Nationalism comes into being only with the idea that the borders the nation

6
Ernest Gellner, Nations and Nationalism (Ithaca, NY: Cornell University Press,
1983), 1.
17
and the borders of the state – must coincide. It can, therefore, be expressed
in both the demand for independence or regional self-government whereas
a mere call for cultural recognition does not per se constitute a nationalist
programme.

In these terms, nationalism is an eminently modern phenomenon and

nationalist discourse had, among the Kurds, little or no political significance
before the collapse of the Ottoman Empire in 1918.7 The partition and
incorporation of Kurdistan into the newly-established states of Turkey, Iraq,
Iran, and Syria, the assimilationist nation-building process these states
pursued, and the uneven socio-economic development that this
incorporation engendered for the Kurdish regions are all historical reasons
for the rise of Kurdish nationalism. As a result, this rise can be freed from
the teleology inherent in the idea of the national awakening and treated as
a framework, made possible by the abovementioned historical
circumstances, within which Kurdish political actors could frame their claim
to power over the region.

These reflections on nationalism and on the risks involved in writing national

histories circumscribe the scope of this study clarifying what this thesis is
and is not about and enable some viable working definitions of the Kurds
and Kurdistan.

The Kurds and Kurdistan

As this thesis is about Kurdish nationalism in its historical expressions, the

definition of the Kurds and Kurdistan is not its subject of investigation,
except insofar as these terms are historically framed by Kurdish nationalists.
However, to avoid blindly reproducing the nationalist discourse, it is
important to outline the problematic nature of these questions to anticipate
the way the words ‘Kurd’ and ‘Kurdistan’ will be used in this thesis. The most
common characterisations of the Kurds as “a people without a country” and

7
See Chapter 3.
18
“the world’s largest stateless nation”8 suggest that defining the border of
the Kurdish nation – both in terms of land and members – is an inherently
problematic task or, better, an eminently political decision. However, even
if the Kurdish nation does not have a state to issue passports and set the
boundaries of its membership, millions of people around the world identify
as Kurds, regardless of their Turkish, Iraqi, or German passports or even the
lack thereof. It is precisely the resilience of Kurdish national identity and its
persistent political significance in the absence of a state that make Kurdish
nationalism such a compelling subject.

The Kurds are normally identified with the speakers of a number of Iranic
languages hailing from the region that straddles the borders of Turkey, Iran,
Iraq and Syria. The self-identification of these linguistic groups within a
collective Kurdish identity is the result of a long historical process. Kurdish
nationalist historians and political organisations have defined these
boundaries by attributing a common ancestry to these groups and traced
the origins of the Kurds in the Medes, an ancient Iranic people that settled
in present-day north-western Iran around the seventeenth century BC.9 In
the words of Mustafa Barzani – arguably the single most prominent Kurdish
nationalist leader of the twentieth century – “deeply rooted in history since
before 3000 BC, our people, with distinct characteristics, undeniably
inhabited the Zagros mountains.”10 This idea served the political purpose of
constructing a common past for people speaking languages that are only in
part mutually intelligible even if the ‘Median hypothesis’ has been challenged
on philological grounds.11 Kurdish scholar Mehrdad Izady does not go that

8
Gerard Chaliand, ed., A People Without a Country: The Kurds and Kurdistan
(London: Zed Books, 1993); ‘Who Are the Kurds?’, Al Jazeera, 1 October 2017,
<https://www.aljazeera.com/news/2017/05/kurds-170516072934887.html>.
9
See, for example: Taufiiq Wahbi, The Origins of the Kurds and Their Language
(Oxford: Clarendon Press, 1966).
10
Mustafa Barzani, ‘Speech Presented to the Congress of the Kurdish Exiles in the
Soviet Union: Baku, January 19, 1948’, The International Journal of Kurdish Studies,
XI.1–2 (1997), 35.
11
David MacKenzie argues that the Kurdish languages belong to the south-western
Iranic group, unlike ancient Median, that belongs to the north-western group. See,
19
far back and yet claims that the Kurds were a defined ethnic group before
the Islamic conquest in the sixth century and that since then the “Kurds are
a multi-lingual, multi-religious, multi-racial nation, but with a unified,
independent, and identifiable history and culture.”12 In historical records,
Kurdistan – literally the land of the Kurds – was first mentioned in the work
of early Arab geographers and, in the twelfth century, the Seljuk Sultan
Sanjak established a Koordistan province in present-day north-western
Iran.13 In the early sixteenth century, the term became of common usage to
define the mountainous borderlands between the Ottoman and the Persian
empires along a line that still roughly coincides with the current north-
western borders of Iran.

Calling the region Kurdistan did not imply, however, the demarcation of
defined ethnic boundaries. In a region historically dominated by tribalism
and nomadism, Kurdish tribes coexisted with Turkish and Arab tribes,
merged and split over the centuries: the tribal confederations that ruled the
area in the fifteenth century – the Aq Qoyunlu and Kara Qoyunlu, the White
and the Black Sheep – were mixed Turkish and Kurdish.14 The ‘Turkish’
Safavid dynasty was probably originally Kurdish and was Turkified first, and
then Persianised, while at the same time claiming an ‘Arab’ descent via the
Prophet Muhammed. When the region was ruled by the Ottomans (1516-
1918), the use of the word Kurd often had a social rather than a linguistic
connotation since it was used to collectively identify the tribal and nomadic
people of the region as opposed to the sedentary population. In the early
1920s, Turkish sociologist Ziya Gökalp observed this dynamic at play in his
native Diyarbakır where the distinction between a Turk and a Kurd largely

David N. MacKenzie, Kurdish Dialects Studies I-II. (London: Oxford University

Press, 1961).
12
R. Mehrdad Izady, The Kurds: A Concise Handbook (Washington: Taylor &
Francis, 1992), 185.
13
T. Maria O’Shea, Trapped between the Map and Reality: Geography and
Perceptions of Kurdistan (London and New York: Routledge, 2004), 230.
14
McDowall, A Modern History, 9.
20
overlapped with that of an urban and a rural dweller. 15 It was the
development of nationalism in the region in the early twentieth century – the
very subject of this thesis – that forced people with different and fluid
identities “to opt for an unambiguous [Kurdish] ethnic identity.”16

Historical evidence seems to dispel the idea of a primordial ethnogenesis

of the Kurds as a distinct group. In the course of the twentieth century, the
Kurds were increasingly identified with the speakers of the Kurdish
language as the trait that most significantly distinguished them from their
neighbours. This characterisation based on language – albeit more solid
than other features discussed below – comes with two problems. First, there
is hardly such a thing as a Kurdish language. Kurdish people speak several
dialects, that are only in part mutually intelligible,17 but the absence of a
Kurdish state prevented the rise of one of them to a unified national Kurdish
language. As modern sociolinguists taught us in the past century, “a
language is a dialect with an army and navy” and distinguishing between a
national language and a vernacular dialect is always a political statement.18
The two most widely spoken Kurdish dialects are Kurmanji – predominant
among the Kurds of Turkey, Syria and the Dohuk province in Iraq – and
Sorani – spoken by most Iraqi and Iranian Kurds. Kurmanji and Sorani are
both official languages in the autonomous Kurdish Region of Iraq and use
respectively the Latin and Arabic scripts. There are many other Kurdish

15
Martin van Bruinessen, ‘Kurds and the City’, in Joyce Blau: l’éternelle Chez Les
Kurdes, ed. by Hamit Bozarslan and Clémence Scalbert-Yücel (Istanbul: Institut
Français d’études anatoliennes, 2018)
<https://books.openedition.org/ifeagd/2199?lang=en>.
16
Martin van Bruinessen, ‘Kurdish society, ethnicity, nationalism and refugee
problems’, in The Kurds: A Contemporary Overview, ed. by Philip G. Kreyenbroek
and Stefan Sperl, The Kurds, 35-37.
17
Iranologist Philip Kreyenbroek argues that the two most-spoken Kurdish ‘dialects’
Kurmaji and Sorani differ from one another as much as German and English. Philip
G. Kreyenbroek, 'On the Kurdish Langauage', in The Kurds, ed. Kreyenbroek and
Sperl, 53-56. In my experience, this statement exaggerates the difference and
speakers of the two languages seem to communicate fairly easily. However, the
growing mutual intelligibility might also be a recent phenomenon driven by the rise
of the Kurdish-language mass media since the early 1990s.
The quote is widely attributed to Yiddish linguist Max Weinreich. Cited in Ronald
18

Wardhaugh, An Introduction to Sociolinguistics (Oxford: Wiley-Blackwell, 1997), 28.

21
dialects such as Zaza in Turkey, as well as Gorani and other southern
dialects in Iraq and Iran. The second problem with a linguistic definition of
the Kurds is that not all Kurds speak Kurdish. The troubled history of the
Kurds over the past century, the policies of assimilation imposed on them,
and the mass forced and spontaneous migration meant that it is fairly
common to meet people who identify as Kurds even if they can barely – or
not at all – speak any of the Kurdish dialects.

In religious terms, most of the Kurdish speakers are also Sunni Muslims of
the Shafi'i school, a trait that distinguishes them from both the neighbouring
Arab and Turkish Sunnis, who follow the Hanafi school, and the Persian and
Azeri Twelver Shias.19 However, many Kurdish speakers are Alevi in Turkey
and Twelver Shia in Iran, and the two religious minorities of the Yazidis and
the Yarsanis use respectively Kurdish Kurmanji and Gorani as their liturgical
language. Islam in Kurdistan has also been historically characterised by the
dominance of Sufism in particular of the Naqshbandi and Qadiri orders.
Religion, however, only played a prominent role in the first phase of Kurdish
nationalism, during the Kurdish revolts of the interwar period discussed in
Chapter 3. After that phase, Kurdish nationalist movements tended to be
secular and to define the Kurdish nation in terms of language to mark a
clearer distinction from the (fellow Muslim) neighbouring peoples.

An important aspect of Kurdish society is tribalism. Kurdish rural society

was, up to the mid-twentieth century largely organised in tribes. In the
richest scholarly account of Kurdish tribal society, Martin van Bruinessen
describes the Kurdish tribe as a

socio-political and generally also territorial (and therefore economic)

unit based on descent and kinship, real or putative, with a
characteristic internal structure. It is naturally divided into a number
of sub-tribes, each in turn again divided into smaller units: clans,
lineages, etc.20

19
Veli Yadirgi, The Political Economy of the Kurds of Turkey: From the Ottoman
Empire to the Turkish Republic (Cambridge: Cambridge University Press, 2017),
24-25.
20
Martin van Bruinessen, Agha, Shaikh and State: The Social and Political
Structures of Kurdistan (Chicago: University of Chicago Press, 1992), 51.
22
Van Bruinessen emphasises the importance of the two social and political
roles in tribal society. First, the agha, the tribal chief, who seats on top of
each unit of the tribal hierarchy – clan, sub-tribe, tribe – and commanded a
degree of political authority over his fellow tribespeople. Second, the
shaykh, the leader of a Sufi brotherhood, who does not belong to any tribe
and thus maintains a super-tribal mediatory authority. Kurds were never all
tribal and, up to the twentieth century, tribes were the dominant form of
social organisation among nomadic Kurds. Settled Kurdish peasants were
generally not part of any tribes and, along with most Christians living among
the Kurds, were unfree and politically subjected to Kurdish aghas or
shaykhs.21

Van Bruinessen goes a long way to explain that we should move beyond the
traditional dichotomy between tribe and state, as conceptual opposites, and
instead look at tribes as shaped by the policy of the states they gravitate
around to the point that they “can even be seen as creations of the state.”22
In the history of the Kurds, the tribal policy of the Ottoman Empire was to
have long-lasting consequences: from the creation of Kurdish tribal
confederacies and their elevation to borderland princedoms in the sixteenth
century to their destruction during the administrative centralisation of the
nineteenth century and the state-sponsored tribal revival of the closing
decades of the Empire. During the twentieth century, tribes suffered a
gradual decline of their social function accelerated by the process of
urbanisation. Particularly, the role of the agha and the shaykh changed
dramatically. With the spread of capitalist relations in the Kurdish
countryside, aghas and shaykhs were in the position to register in their
name land that was customarily owned by the tribe or by the Sufi lodge and
gradually became large landowners while their fellow tribespeople became
waged labourers. This process was heavily supported by the central states
who kept relying on these traditional powerholders to maintain order in the
Kurdish periphery. This transformation heavily affected the traditional

21
Ibid., 50.
22
Ibid., 134.
23
kinship-based solidarity that held tribes together and tribal loyalties
gradually waned. It was by no means an even process and many tribes
continue to play an important social role.

Highlighting the complexity of these transformations is important because,

in this thesis, I almost exclusively speak about the ‘tribal elite’ rather than
tribes. That is because the class stratification that occurred in tribal society
is by far its most relevant aspect in regard to the development of
nationalism. Where ‘tribes’ are mentioned, it is to emphasise episodes in
which people mobilise as members of the tribe regardless of the position
they hold in it. As will become clear, these episodes became rarer and rarer
in the twentieth century.

As mentioned before, the land of Kurdistan is normally identified with a

region that, straddling state borders, includes south-eastern Turkey, the far
north and north-east of Iraq, north-western Iran, and north-eastern Syria.
Kurdish nationalists make a point of stressing the unity of the Kurdish land
and use the cardinal directions to identify each of the regions under foreign
occupation. So, the Kurdish regions of Turkey, Iraq, Iran, and Syria are called
Northern, Southern, Eastern, and Western Kurdistan, or simply by using the
Kurdish for north (Bakur), south (Başur), east (Rojhelat) and west (Rojava).

This form, however, does not help define the ‘external borders’ of the
Kurdish nation, that is to say, to identify which areas are predominantly
inhabited by Kurds. This predominance is problematic to assess for a
number of reasons, the first of which is the region’s cultural diversity. Before
the indigenous Christian population was virtually wiped out during the First
World War, the eastern-Anatolian homelands of the Kurds and the
Armenians largely overlapped. Even today, there are no defined linguistic
borders between Kurdish, Turkish, Persian, and Arabic speakers, and large
areas, as well as important cities, have mixed populations, while linguistic
enclaves exist throughout the region. Secondly, in the twentieth century,
the Kurds were victims of Arabisation policies in Iraq and Syria and forced
assimilation in Turkey that changed the demographic balance of many
mixed areas. In addition to that, the state of semi-permanent warfare, the
24
scorched-earth and mass deportation strategies deployed by the states to
quell the various Kurdish revolts, as well as poverty and structural
underdevelopment, forced millions of Kurds to leave their homes. Forced
and voluntary migrations contributed to the Kurdification of previously
mixed cities and the creation of large Kurdish diaspora communities in non-
Kurdish cities. The third problem with defining predominantly Kurdish areas
is that Turkey, Iraq, Syria, and Iran systematically prevented large-scale
statistical research on the ethnic composition of the region, making a
reliable assessment of the Kurdish population impossible. To overcome
these shortcomings, I refer to the predominantly Kurdish areas as the
Kurdish regions of Turkey, Iraq, Iran, and Syria, or Turkish, Iraqi, Iranian, and
Syrian Kurdistan with unspecified borders. The terms south-eastern Turkey,
northern Iraq, north-eastern Syria, and north-western Iran will be used when
the Kurdish regions are referred to from the perspective of the central
states. To make these terms meaningful, they will be associated, where
possible, to clear topographical reference points and to the administrative
divisions of the states, such as the Turkish provinces (iller) and the Iraqi
governorates (muhafazat). The borders of the predominantly Kurdish areas
drawn in Map 1 are a rough approximation and must be used only to
geographically place the events analysed in the following chapters.

Due to the abovementioned scarcity of statistical data on the ethnic

composition of these states, estimates of the size of the Kurdish population
are often loaded with political significance. Moreover, when such surveys
did exist, state authorities had an interest in ‘correcting’ the number while
interviewees might have been reluctant to disclose their Kurdish identity to
state officials. For example, in the last of such surveys conducted by Turkey
in 1965, the people who declared Kurdish as their mother tongue were 2.3
million people or 7.7 per cent of the country’s population,23 far below the
estimates of most scholars. Even by using the same data, Servet Mutlu has
recalculated the 1965 Kurdish population as 3.13 million and 10 per cent of

23
Servet Mutlu, ‘Ethnic Kurds in Turkey: A Demographic Study’, International
Journal of Middle East Studies, 28 (1996), 517–41.
25
the total population.24 Table 1 collects estimates of the Kurdish population
worked out by scholars. Those numbers can serve as a reference point to
inform the historical analysis but, given the lack of reliability, will not be cited
here to support a particular argument. For the purposes of this study, Kurds
are those who identify as such.

The next section surveys the existing literature on Kurdish nationalism

showing how it often failed to depart from the national-history paradigm.
However, it also discusses the works that can inform the study of Kurdish
nationalism as embedded in social conflicts.

The Literature on Kurdish Nationalism

The study of Kurdish politics is a relatively recent scholarly enterprise.

Before the events of the Kurdish insurgencies in Turkey and Iraq in the late
1980s and early 1990s attracted a great deal of international attention,
scholars of Kurdish politics had to rely on few works of historical and
anthropological scholarship.25 Even after the 1990s, when the literature on
the subject grew exponentially, interest in the Kurds has followed the pace
of events and has been dominated by works of political science and its
subdisciplines, and, to a lesser extent, sociology and anthropology. A lot of

24
Ibid.
25
To mention some of the most influential: Vladimir Minorsky, 'Les origines des
Kurdes', Actes du XXe Congres Internationale des Orientalistes, Louvain (1940),
143-152; Fredrik Barth, Principles of Social Organization in Southern Kurdistan
(Oslo: Universitet i Oslo, 1953); Vasily Nikitin, Les Kurdes: Etude Sociologique et
Historique (Paris: Imprimerie Nationale, 1956); Derk Kinnane, (1964), The Kurds and
Kurdistan (London: Oxford University Press, 1964); Abdul Rahman Ghassemlou,
Kurdistan and the Kurds (London: Collets, 1965); Vanly, Survey.
26
this literature tended to age fast hindering the creation of a body of
cumulative knowledge on Kurdish politics. This is a common tendency in
political science – especially when event-driven – and in the case of Kurdish
studies, this short-sightedness of the literature is also due to the objective
scarcity of background work in Kurdish history that could provide social
scientists with well-informed historical analyses.26 A symptom of this issue
is the proliferation of edited volumes that, even when insightful, tend to offer
fragmented knowledge.27 Moreover, a considerable amount of literature on
Kurdish politics belongs to the field of security studies written from a
profoundly ideological counterterrorism or counter-insurgency angle and
aimed directly or indirectly at policy-makers.28

Within this vast literature, this section reviews some of the most influencial
works on Kurdish nationalism, briefly surveying the literature that
reproduces a culturalist and identity-based explanation and more
extensively discussing works that give space to socio-economic factors and
that were thus more relevant to the present study on class and nationalism.
Before delving into the analysis of the literature, it is important to note that
the relationship between nationalism and social classes is largely absent
from literature on the Kurds. This is due mostly to the scarcity of specialist

26
For a discussion on this, see Jordi Tejel, ‘New perspectives on writing the history
of the Kurds in Iraq, Syria and Turkey: A history and state of the art assessment’,
in The Kurdish Question Revisited, ed. by Gareth Stansfield and Mohammed
Shareef (London: Oxford University Press, 2017), 3-16.
27
To name just a few titles published solely in 2019: Michael Gunter, ed., Routledge
Handbook on the Kurds (London and New York: Routledge, 2019); A. Faleh Jabar,
and Renad Mansour, eds., The Kurds in a Changing Middle East: History, Politics
and Representation (London: I.B. Tauris, 2019); Anwar Anaid, and Emel Elif Tugdar,
eds., Iraqi Kurdistan’s Statehood Aspirations: A Political Economy Approach
(London: Palgrave Macmillan, 2019); Mehmet Gürses, David Romano, and Michael
Gunter, eds., The Kurds in the Middle East: Enduring Problems and New Dynamics
(Lanham: Rowman & Littlefield, 2019).
28
For example: İhsan Bal and Sedat Laçiner, Ethnic Terrorism in Turkey and the
Case of the PKK: Roots, Structure, Survival, and Ideology, (London: Frank Cass
2004); Mitchel P. Roth and Murat Sever, ‘The Kurdish Workers Party (PKK) as
Criminal Syndicate: Funding Terrorism through Organized Crime, A Case Study’,
Studies in Conflict & Terrorism, 30.10 (2007), 901–20; Mustafa Cosar Ünal, ‘The
Kurdistan Workers’ Party (PKK) and Popular Support: Counterterrorism towards an
Insurgency Nature’, Small Wars & Insurgencies, 23.3 (2012), 432–55.
27
political economy literature and the virtual absence of Marxist scholarship
on the Kurds – with some exceptions discussed below. Therefore, this
overview of the literature, rather than discussing the missing alternative
arguments on the role of social classes, focuses on analysing the space
given by the literature to the role played by socio-economic structures in
the rise of Kurdish nationalism. Finally, this section discusses the works on
Kurdish politics in Iraq and Turkey that most significantly influenced the
arguments I developed on the two case studies of this thesis.

As already described earlier in this chapter, many Kurdish nationalists – like

most nationalists in general – tend to see the Kurdish nation as a perennial
entity that has existed since far before nationality became a principle of
state organisation. The development of nationalism in the twentieth century
was thus a natural consequence of the oppression of the Kurds by other –
Turkish, Arab, Persian – nations. This discourse is reproduced by Kurdish
and non-Kurdish scholars who attribute the development of nationalism first
and foremost to the suppression of Kurdish identity. For Wadie Jwaideh,
after World War I:

The Kurds, now in a resentful mood, were rendered even more

restive and unmanageable by the heightened impact of Western
civilization […] a development they felt threatened to undermine
their way of life. The Kurdish masses, with the encouragement of
their leaders, were determined to resist this influence. The various
Kurdish rebellion, besides being violent manifestations of Kurdish
nationalist sentiment, were also waged in defence of the Kurdish
way of life.29
The Kurdish revolts of the interwar period are framed by Jwaideh in cultural
terms, as a movement of the entire Kurdish nation in defence of its identity.
In Ofra Bengio’s triumphalist account of Kurdish autonomy in Iraq, the story
takes an epic tone:

The amazing story of the Kurdish revival in Iraq, following the

genocidal war of 1988–1989, very much resembles that of the Jews
following the Holocaust. Within the first four years after that war, the
Kurds managed to launch an ambitious project for Kurdish nation

29
Wadie Jwaideh, The Kurdish National Movement: Its Origins and Development
(Syracuse: Syracuse University Press, 2006), 292.
28
 building and state building […]. And although the Kurdish national
project encountered enormous internal and external obstacles […]
its success was […] achieved through the accumulated fruit of
eighty-five years of ongoing struggle.30
These works refuse to problematise Kurdish nationalism by erasing its
contested nature and the conflicts within Kurdish society. They are
problematic both in analytic terms, because they provide a simplistic
historical picture, and in political terms, as they reproduce a narrative that
legitimise the power claim of the Kurdish elite, especially in the context of
Kurdish self-rule in Iraq.

A large part of the literature on Kurdish nationalism, however, emphasises

the fundamental role played by the socio-economic transformations of the
twentieth century – such as urbanisation, mass literacy, the role of modern
states – in the development of Kurdish nationalism. In some authors, these
premises are implicitly framed within an approach based on modernization
theory, a power-free way of analysing the political economy, that inevitably
leans towards culturalist explanations. For instance, the democratic deficit
of the Kurdish autonomous region in Iraq is then attributed to the resilience
of tribal social structures – as in Hussein Tahiri – or to the development of
an rent-driven ‘culture of dependence’ – as in Denise Natali.31 This line of
argument ends up depoliticising social conflicts and – even beyond the
intention of the authors – absolving the Kurdish elite by moving practices
such as corruption and patronage from the realm of politics to the one of
culture. Even more problematic is the case of a more recent strand of
literature produced within Iraqi-Kurdish universities in which mainstream
political economy is deployed to depoliticise the power relations of the
region.32 David Romano’s account of the rise of Kurdish nationalism in

30
Ofra Bengio, The Kurds of Iraq: Building a State Within a State (London: Lynne
Rienner Publishers, 2012), 315;
31
Hussein Tahiri, The Structure of Kurdish Society and the Struggle for a Kurdish
State (Costa Mesa, CA: Mazda Publishers, 2007); Denise Natali, The Kurdish Quasi-
State: Development and Dependency in Post-Gulf War Iraq (Syracuse: Syracuse
University Press, 2010).
32
See, for example, Nyaz Noori Najmalddin, ‘The Failure of Economic Reform in the
Kurdistan Region of Iraq (1921-2015): The Vicious Circle of Uncivic Traditions,
29
Turkey offers some interesting insights – including the attention towards the
role of rural conflicts – and yet suffers from similar theoretical limitations.33
His rational-choice theoretical framework reduces political projects to
strategic decision failing to account for the alternative avenues – such as
assimilation, Islamism, the Turkish left – that were available to the Kurds.34

Martin van Bruinessen’s Agha, Shaikh and State and David McDowell’s A
Modern History of the Kurds need to be mentioned for their importance in
the field.35 Van Bruinessen’s anthropological study is one of the foundational
text on Kurdish politics. Despite its little engagement with nationalism, this
text remains an essential read on Kurdish politics and its interesection with
the traditional structures of Kurdish society. David McDowall’s book is by far
the most comprehensive history of the Kurds to date. Despite not always
complying to academic standards, McDowall provides a rich account on
Kurdish history in the twentieth century with a high degree of attention for
the socio-economic transformation of the region. Even if neither of these
texts provide a particularly significant interpretation of nationalism and lack
systematic analysis of its relation to the class structure, they both
immensely contributed to this author’s understanding of Kurdish history and
politics.

A number of specific studies on the Kurds of Iraq and Turkey constituted

the most important sources of empirical material and theoretical reflections
on the two case studies of this thesis. Of the less sizeable literature
published on the Iraqi Kurds – compared to those of Turkey – two books

Resource Curse, and Centralization’, British Journal of Middle Eastern Studies, 45.2
(2018), 156–75, as well as the various contributions included in Anaid and Tugdar,
eds., Iraqi Kurdistan’s Statehood Aspirations; Aram Rafaat, ‘The Fundamental
Characteristics of the Kurdish Nationhood Project in Modern Iraq’, Middle Eastern
Studies, 52.3 (2016), 488–504.
33
David Romano, The Kurdish Nationalist Movement: Opportunity, Mobilization and
Identity (Cambridge: Cambridge University Press, 2006).
34
For a more extensive critique along this line, see Cengiz Güneş, The Kurdisn
National Movement in Turkey: From Protest to Resistance (Milton Park, Abingdon,
Oxon: Routledge, 2012), 19-24.
35
Van Bruinessen, Agha, Shaikh and State; McDowall, A Modern History;
30
stand out for their attention to the political-economic trajectory taken by the
Kurdish region of Iraq in the 1990s, Gareth Stansfield’s Iraqi Kurdistan:
Political Development and Emergent Democracy and Denise Natali’s The
Kurdish Quasi-State.36 Despite Stansfield’s focus on governance rather than
nationalism and leaving aside Natali’s abovementioned moralistic
conclusion, these two books offer a compelling description of the politics
and political economy of the region that deeply influenced the analysis of
Iraqi-Kurdish nationalism in this thesis. These are by far the most
comprehensive studies on Kurdish self-rule in Iraq after 1991, and yet,
despite their attention paid to the political economy of the region, neither of
them gives any space to the radical transformation of its class structure and
the consequent transformation of Kurdish politics and of the social function
of Kurdish nationalism.37

The two texts – both about the Turkish Kurds – that most heavily influenced
this study, especially in terms of the development of the argument, are
Cengiz Güneş’s The Kurdish National Movement in Turkey (2012) and Vali
Yadirgi’s The Political Economy of the Kurds of Turkey (2017).38 The fact that

36
Gareth Stansfield, Iraqi Kurdistan: Political Development and Emergent
Democracy (London and New York: Routledge, 2003). Natali, The Kurdish Quasi-
State.
37
Two recent articles do emphasise the function played by Kurdish nationalism in
the power structure of the Kurdish region of Iraq even if they are not focused on
the political economy. Dylan O’Driscoll & Bahar Başer’s article on the 2017
independence referendum and Andrea Fischer-Tahir’s chapter on the construction
of a collective memory emphasise the instrumental use of nationalism made by the
Iraqi-Kurdish elite reaching conclusion similar to those developed in Chapter 6 of
the present work. See Andrea Fischer-Tahir, ‘Searching for Sense: The Concept of
Genocide as Part of Knowledge Production in Iraqi Kurdistan’, in Writing the Modern
History of Iraq: Historiographical and Political Challenges, ed. by Jordi Tejel, Peter
Sluglett, Riccardo Bocco, and Hamit Bozarslan (Singapore: World Scientific, 2012),
227-43; Dylan O’Driscoll, and Bahar Başer, ‘Independence Referendums and
Nationalist Rhetoric: The Kurdistan Region of Iraq’, Third World Quarterly, 40.11
(2019), 2016-2034.
38
Yadirgi, The Political Economy; Güneş, The Kurdish National Movement. I need
to mention also the work by Joost Jongerden on the relocation of the Kurdish
peasantry in his Joost Jongerden, The Settlement Issue in Turkey and the Kurds:
An Analysis of Spatial Policies, Modernity and War (Leiden: Brill, 2007), as well as
the work he co-authored with Ahmed Akkaya on the origins and trasformation of
the ideology of the Kurdish movement in Turkey (see bibliographical references in
Chapter 8 and 9).
31
Güneş focuses on the political discourse of the Kurdish movement and
Yadirgi on the political economy of the region makes their books
complementary for research on nationalism and class. Güneş deploys post-
Marxist Discourse Theory as developed by Ernesto Laclau and Chantal
Mouffe to stress the multiple historical articulations of Kurdish nationalism
discourse.39 As nationalism is analysed as an empty shell, devoid in itself of
ideological content, a discourse analysis of the primary sources reveals the
ideological content of each historical articulation of Kurdish nationalism.
Thus Güneş can place the expressions of Kurdish nationalism in twentieth
century’s Turkey in their particular historical context and explain that
nationalism was articulated “initially within the Islamist-conservative
discourse (the early 1920s), as a modernist discourse (1920s and 1930s),
underdevelopment (1960s), Marxist-Leninism (1970s and 1980s), and,
finally democracy (1990 onwards).”40 Even if Güneş is not directly concerned
with political economic issues, his analysis of the evolution of the political
discourse produced by Kurdish nationalists makes it a fundamental read to
understand the development and diversification of nationalist projects.

On the contrary, Veli Yadirgi’s work represents a rare case of recent

scholarship on the political economy of the Kurds. Since the 1970s, Turkish
and Kurdish Marxists came to see the relationship between the Turkish state
and the Kurdish region through the prism of colonialism, building on the
theoretical premises of Dependency Theory. While acknowledging its
merits, Yadirgi departs from this tradition criticising its “static
conceptualisation of the relation between ‘powerless and peripheral’
Kurdish areas and the ‘all-powerful and dominant’ Turkish state” and its
economist premises.41 Yadirgi re-politicises this literature by framing the
underdevelopment of the Kurdish region as an example of ‘de-development’
resulted from the policies of nation-building pursued by Turkish nationalist

39
See Ernesto Laclau and Chantal Mouffe, Hegemony and Socialist Strategy :
Towards a Radical Democratic Politics (London: Verso, 1985) and Ernesto Laclau,
New Reflections on the Revolution of Our Time (London: Verso, 1990).
40
Güneş, The Kurdish National Movement, 11.
41
Yadirgi, The Political Economy, 53-57.
32
governments since the Young Turk Revolution (1908). Deportations,
dispossession, and cultural suppression contributed to forestall the
development of Kurdistan not just by simply maintaining the region poor but
actively rescinding “the prospect of autonomous indigenous existence”. 42
By combining the study of the different ideological trajectories developed
within the Kurdish national movement with a thorough analysis of the
political economy of the region, this thesis builds upon the existing literature
to provide a novel account of the relationship between the evolving class
structure and the nationalist projects.

The initial motivations for this study, outlined earlier in this introduction,
were strengthened by the analysis of the literature on Kurdish nationalism.
Despite the great deal of interest in the subject among social scientists, our
understanding of Kurdish nationalism remains fragmented and the debates
around its origin are still quite underdeveloped. Although many of the
authors do not show any particular ideological hostility towards the study of
class structure and give a great deal of weight to socio-economic dynamics,
none of the abovementioned works provide a systematic analysis of the role
of classes in Kurdish politics and their relationship to Kurdish nationalism.

Methodology and Methods

This investigation of Kurdish nationalism was conducted as a work of

Historical Sociology based on the comparison between the history of the
Kurdish national movement in Iraq and Turkey. Comparative historical
analysis has traditionally been the best candidate in the social sciences to
answer the “questions about large-scale outcomes that are regarded as
substantially and normatively important by both specialists and
nonspecialists.”43 This focus on the ‘big questions’ drive historical
sociologist to use the work of historians with a sociological end, to formulate

42
Ibid., 58-59.
43
Dietrich Mahoney James and Rueschemeyer, ‘Comparative Historical Analysis in
the Social Sciences’ (Cambridge: Cambridge University Press, 2003), 7.
33
theoretical arguments on long-term processes. It tends to rely thus on
secondary sources as its subject is too broad to be based on primary
research. In this thesis, events of war and insurgency, the rise and fall of
organisations and leaders, and the emergence of different expressions of
Kurdish nationalism are played against the backdrop of the longue-durée of
the evolution of the region’s class structure. Alternating these two
dimensions is particularly important to avoid economic determinism that
treats political events as mere epiphenomenal expressions of structural
processes. Even though between the 1980s and 1990s the class structure
of Iraqi Kurdistan changed substantially, the Kurdistan Democratic Party
(KDP) remained the political actor that best represented the interest of the
Kurdish ruling class. It remained so through the structural transformation of
the ruling class but was also the actor that most significantly contributed to
the reshaping of the relations of power in the 1990s.

The form of comparison used in this thesis is that of the ‘method of

difference’ needed to explain different outcomes – the different forms of
Kurdish nationalism in Iraq and Turkey – to similar initial circumstances –
Kurdistan at the time of its partition.44 This form of comparative work
showed its great explanatory power in some of the classics of Historical
Sociology that originally inspired this study.45 However, this thesis follows
the Marxist tradition in studying social reality in its unity and resisting the
isolation of individual factors and the reduction of broad transformation to
a single explanatory variable. In these terms, this thesis remains a ‘soft
comparison’ to avoid letting the methodology determine the conclusion and
to maintain an open approach that allows for multi-causal explanations.
That is to say that the diversification of the balance of class forces in the

44
As opposed to the ‘method of agreement’ which serves to explain why different
circumstances lead to similar outcomes.
45
Such as Barrington Moore, Social Origins of Dictatorship and Democracy: Lord
and Peasant in the Making of the Modern World (Boston, MA: Beacon Press, 1966);
Theda Skocpol, States and Social Revolutions: A Comparative Analysis of France,
Russia, and China (Cambridge: Cambridge University Press, 1979); Dietrich
Rueschemeyer, Evelyne Huber Stephens, and John D. Stephens, Capitalist
Development and Democracy (Cambridge: Polity Press, 1992).
34
two Kurdish regions – the main explaining factor – is analysed in its
interactions with local and contingent specificities, the unevenness of
economic development, as well as an ever-changing international
environment.46

However, this study tries to take seriously the criticism advanced by non-
Marxist scholars to study not only “very long-run economic development”
but to account for the role played by “immediate transformations that occur
in the structure and functions of state organizations” and in “the relations
between the state and social classes.”47 If the balance of power among
social classes is the condition for the rise of a certain form of nationalism or
certain organisations, it is by no means the only determinant of the political
outcome. To give an example, the Kurdish revolt in Iraq of 1961-1975 had its
origins and found its strength in the power of the traditional landowning
class. Yet, the reason it could last for so long and constitute such a
challenge to Baghdad is that several foreign powers were – due to
exogenous Cold War dynamics – interested in weakening the Iraqi
government. When foreign support was withdrawn, the revolt abruptly
collapsed. In a similar vein, the counter-insurgency strategy adopted by Iraq
in the 1980s and Turkey in the 1990s brought Kurdish rural society to violent
and rapid destruction and radically transformed the class structure of the
two Kurdish regions with durable political consequences.

Besides its methodological advantages, the choice to carry on a

comparative work is partly imposed by the Kurdish reality. Incorporated into
four different states since the early 1920s, the ‘histories’ of the Kurds
diversified to a point that it is impossible not to follow a country-by-country
narration. This is particularly true for the political economy – the focus of
this thesis – of each Kurdish region, shaped by the policies imposed by the
state centres.

46
Simon Bromley, Rethinking Middle East Politics: State Formation and
Development (Cambridge: Polity Press, 1994), 119-121.
47
Skocpol, States and Social Revolutions, 35.
35
There are three reasons for the choice of the Kurdish movements in Iraq and
Turkey as the case studies. First of all, Kurdish organisations from Iraq and
Turkey have dominated Kurdish politics for the past few decades, exercising
their influence in a competitive way over the Kurds of Iran and Syria. The
comparison is thus almost imposed by the circumstances. Moreover, the
Kurdish insurgencies in Turkey and Iraq in the twentieth century were the
longest and their protagonists shaped transnational Kurdish politics in a far
more profound way than any organisation from Iran and Syria. In these
terms, telling the stories of Kurdish nationalism in these two countries is the
closest way of telling the whole story of Kurdish nationalism. Kurdish
national forces in Syria and Iran had enough power to be politically
significant in the two countries only in a few historical moments, namely
1945-46 and 1979-80 in Iran, and after 2012 in Syria. Those are exactly the
periods in which their stories most significantly intersects with that of the
Kurds in Iraq and Turkey and this is when they feature in this thesis
(respectively in Chapters 3, 4, and 9).

Secondly, Kurdish nationalism in Iraq and Turkey is also particularly

significant because of the strikingly different forms of its development. The
facts that they started from very similar structural conditions – in the Late-
Ottoman and interwar periods – and that they were always fighting for the
same Kurdish nation – in terms of who the Kurds are and where Kurdistan is
– makes it particularly significant that these nationalist movements radically
differed along ideological lines. As Kurdish nationalist movements with
different class bases developed two politically opposite national projects,
their comparison constitutes a vantage point to investigate the relationship
between class and nationalism. In these terms, Iranian and Syrian Kurdistan
did not lend themselves to such a ‘clean’ comparison. Iranian Kurdistan was
not part of the Ottoman Empire and was separated from the other parts of
Kurdistan since the sixteen century and – despite important commonalities
– presented different starting conditions that would have compromised the
value of the comparison. Since the Kurdish population in Syria was always
smaller and territorially discontinuous constituting a majority in only a few

36
limited areas, chosing Syria as as a case would have led to an ‘unbalanced’
comparison.

Finally, the choice to exclude the history of Kurdish nationalism in Iran and
Syria from the comparison also had a practical reason. The available
literature on the Iraqi and Turkish Kurdish movements is far wider and
provides enough material for a historical sociological work covering a
century of Kurdish history. The literature on the Kurds of Iran and Syria is
much more limited and focusing on those to cases would have forced me to
change the nature of this study and to rely more heavily on primary
research. Considering the difficulties of carrying on fieldwork research in
Iranian and Syrian Kurdistan, this was not a feasible option.

In terms of methods, this thesis is largely based on the analysis and

comparison of secondary sources as is generally the case in historical
sociological analyses covering a long time span and aimed at answering
theoretical questions. Different sets of literature were combined in a novel
vein, especially the histoire événementielle of the Kurdish movement, and
the literature on the economic and social history of Iraq and Turkey. In
regard to Iraq, these fields were especially disconnected with scholars of
Iraq as a whole disregarding the Kurdish region, and students of the Kurds
often failing to link Iraqi Kurdish politics to that of the rest of the country.
The combination of these two strands of literature was particularly fruitful.
For example, reading the history of Iraqi Kurdish nationalism against the
backdrop of the history and economic history of Iraq allowed for a much
deeper understanding of the motivations and dynamics of the 1961 Kurdish
revolution.48 The secondary sources also include literature produced by
international organisations, government bodies, or think tanks, and
especially political economic reports.

Primary sources were not central in the development of this thesis mostly
due to the long historical period covered – over a century – that did not leave
much space to analyse events in detail. For the most recent period – the

48
For a discussion on these issues, see Chapter 4.
37
2000s and 2010s, analysed in Chapters 6 and 9 – newspaper articles
constituted an important source of raw information that informed the
analysis. Moreover, the scarcity of historical literature on the Kurds, and its
frequent gaps, necessitated complementing the secondary sources with 29
semi-structured interviews with Kurdish politicians, former combatants,
journalists, and intellectuals. The interviews were conducted in two long
trips to the Kurdish region of Iraq (in summer 2018 and spring 2019), in short
trips to Brussels, Berlin, and Paris (winter 2019-2020), as well as in London
and over the telephone. They were conducted in English, Kurdish, Turkish,
Arabic, Dutch, and German, and most of them with the help of a translator.
The information collected in this form helped illuminate some obscure
historical passages. However, and most importantly, the contribution given
by these conversations – and by the many more I had during my fieldwork
trips – to my understanding of Kurdish politics and to the reflections that led
to the formulation of the argument goes far beyond the space that
interviews occupy in the final draft of this thesis.

The Structure of the Thesis

This thesis is built around seven chapters on the history of Kurdish

nationalism. The first chapter presents a broad picture of Kurdistan in the
interwar period, while the following six are divided by the two cases with
three chapters on Iraq and three on Turkey.

The first part of the thesis sets the stage for the study of Kurdish nationalism
in Iraq and Turkey. This introductory chapter is followed by the theoretical
framework (Chapter 2) which assesses modernist approaches to the study
of nationalism, highlighting how they tend to treat nationalism as an
autonomous force. After having problematised the Eurocentric and idealist
tendencies of this literature, the chapter suggests setting aside abstract
theorising on the (a)historical nature of nationalism. It proposes instead to
study nationalist movements as context-specific expressions of the
struggle for state power embedded in class conflicts.

38
Within this theoretical framework, Chapter 3, the first historical chapter,
discusses the earliest politically significant expressions of Kurdish
nationalism in interwar Turkey and Iraq as a unitary phenomenon. The
Kurdish ‘feudal’ revolts that took place in both countries in the 1920s and
1930s were the results of the reluctance of the Kurdish traditional elite to
be integrated into the newly established states of Turkey and Iraq after the
collapse of the Ottoman Empire.

After the defeat of these revolts, however, Kurdish nationalism in Turkey

and Iraq took two opposite trajectories shaped by the different outcomes of
the feudal revolts but even more importantly by the different nation-building
strategies deployed by the new states, their different degrees of
development and integration into the capitalist world-system, as well as the
different geopolitical status of the two countries – initially an autonomous
middle power the former and a colonial subject the latter, later on the
opposite sides of the Cold War divide. All these factors conspired to shape
and diversify the class structure of the two Kurdish regions. In the following
six chapters, the history of Kurdish nationalism in Iraq (Chapters 4, 5, and
6) and in Turkey (Chapters 7, 8, and 9) is reconstructed through pivotal
historical moments in which Kurdish nationalist movements rose and fell
against the backdrop of the evolving political economy of the region. The
two cases run parallel chronologies, to emphasise the different trajectories
of Kurdish nationalism in (roughly the same) historical periods: Iraq and
Turkey during the Cold War (Chapters 4 and 7), the 1990s and early 2000s
(Chapters 5 and 8), and the increasingly fragmented and unstable Middle
East of the 2010s (Chapters 6 and 9). However, each chapter is also a story
on its own, in which the evolution of the social and economic structures of
the region as well as the agency of the Kurdish political actors change the
class basis and politics of the Kurdish national movement.

The concluding chapter of the thesis (Chapter 10) brings together the two
cases in a comparative manner, analysing the patterns of political behaviour
of the Kurdish social classes. Building on previous chapters, it shows how
the diverging structural contexts of Iraqi and Turkish Kurdistan drove the

39
same social classes – the landowning class, the middle classes, and the
peasantry – to develop forms of Kurdish nationalism with different strategies
and ideologies. This final chapter connects the historical analysis to the
theoretical framework of the thesis and shows that the study of the Kurdish
national movement constitutes an extraordinary vantage point to make
wider considerations on nationalism and class politics.

40
Chapter 2
Nationalism and Class Politics

Introduction

The term ‘nationalist’ is commonly used to describe a wide variety of political

movements. The fact that a Marxist anti-colonial movement in southern
Africa and a neo-Nazi group in western Europe can be both and
unproblematically called nationalist reflects the ambiguity of the term.
Historians and social scientists have grappled with the issue since the
nineteenth century, but it was only in the last quarter of the twentieth
century that a blossoming of publications on the subject created the
interdisciplinary field of Nationalism Studies. Scholarly interest was driven
by an apparent revival of nationalist identities and a parallel demise of class
as the principal source of political divide. As workers in the Global North
turned their back to social democratic politics and the national liberation
movements of the Global South gave way to authoritarian regimes and
ethnic conflicts, scholars identified in the nation a source of identity stronger
than any ‘rational’ approach to politics based on material interests or
universal values. Even if most students of nationalism profess the modernist
belief that the nation is a historical construct, their explanations often fall
into the circular argument whereby nationalism is important because it is
important.

This circularity derives primarily from the idea that the nation-state had its
origin in nineteenth-century Europe where it arose out of long-term
historical developments and then diffused to the rest of the world. The
failure to explain nationalism outside the north Atlantic region as anything
but an imported idea reflects a Eurocentric bias of the field but it also reifies
the idea that nationalism is an autonomous force whose relationship to
contextual political struggles is only contingent. Conceptualised in this way,

41
nationalism exists regardless of other social and political conflicts. For this
reason, the theories of nationalism have little to say about the actual political
content of nationalist struggles and the evident diversity of nationalist
movements, the progressive character of some and the reactionary
character of others. Ultimately, they tell us very little about the politics of
nationalism. To re-politicise the study of nationalism, this thesis follows
critical geographer James Blaut in studying nationalist struggles ultimately
as struggles for state power embedded in class conflicts.49 Nationalist
movements struggling for state power – to win power in an existing state or
to create a new state – are always embedded in conflicts between classes,
class fractions, and class coalitions outside which we cannot understand
their ideological outlook and political programme. Placing nationalist
movements within their society’s specific set of class relations, shaped by
local history, geopolitics and their location in the global division of labour,
opens the study of nationalism to empirical investigation. Eschewing grand
theories which inevitably compare any expression of nationalism to an
original (European) standard, this chapter proposes an alternative way to
study the politics of nationalist movements.

Diffusionism and Eurocentrism

The issues outlined above already suggest that the theoretical approach
built in this chapter excludes the existence of any primordial root of
nationalist feelings. As Roger Brubaker observes, “no serious scholar today
holds the view that is routinely attributed to primordialists in straw-man
setups, namely that nations or ethnic groups are primordial, unchanging
entities.”50 The primordialism held by many real-world nationalists who
naturalise and de-historicise the nation has been long demystified by
modernist scholars of nationalism. Modernist authors such as Ernest Gellner

49
James M. Blaut, The National Question: Decolonizing the Theory of Nationalism
(London and New Jersey: Zed Books, 1987).
50
Roger Brubaker, Nationalism Reframed: Nationhood and the National Question in
the New Europe (Cambrdige: Cambridge University Press, 1996), 15.
42
turned upside down the primordialist notion that nationalism is the process
of awakening of dormant – yet pre-existing – nations. On the contrary, they
successfully demonstrated that nations are the product of nationalist
movements that create them by setting boundaries and identifying their
essential cultural features.51 Accepting the validity of the modernist critique
of nationalism, this chapter stays within this tradition and proposes solutions
to overcome its limits.

Going through some of the foundational works of nationalism studies written

by modernist and Marxist scholars, the reader will notice an undertone of
contempt towards nationalists. With a certain arrogance, Gellner claims

that the prophets of nationalism were not anywhere near the First
Division, when it came to the business of thinking (…) It is rather that
these thinkers did not really make much difference. If one of them
had fallen, others would have stepped into his place (…) The quality
of nationalist thought would hardly have been affected much by
such substitutions. Their precise doctrines are hardly worth
analysing.52
For Eric Hobsbawm, “no serious historian of nations and nationalism can be
a committed political nationalist,” implying that real existing nationalisms
can only be primordialist.53 Tom Nairn – a (Scottish) nationalist himself! –
thinks that the contradictions of modernity found their solution in “the
crudity, the emotionalism, the vulgar populism, the highly-coloured
romanticism of most nationalist ideology (all the things intellectuals have
always held their noses at).”54 As Nairn, like Hobsbawm, is also Marxist, his
statements are imbued with a profound pessimism for the prominence of
irrational – even deplorable – nationalist feelings over universalist and
emancipatory politics. He adds:

Nationalism [worked] because it actually did provide the masses

with something real and important—something that class

51
Gellner, Nations and Nationalism.
52
Ibid., 124.
53
Eric J. Hobsbawm, Nations and Nationalism since 1780: Programme, Myth,
Reality (Cambridge: Cambridge University Press, 1992), 12-13.
54
Tom Nairn, ‘The Modern Janus’, New Left Review, 1.94 (1975), 22.
43
 consciousness could never have furnished, a culture which however
deplorable was larger, more accessible, and more relevant to mass
realities than the rationalism of our [emphasis added] Enlightenment
inheritance.55
The apparent demise of class politics is, for Nairn, the evidence for the
strength and autonomy of nationalist feelings. In this famous essay, Nairn
describes nationalism as a ‘Modern Janus’ that, like the ancient Roman god
with two faces. is, with no exceptions, “both healthy and morbid […] both
progress and regress” while “fascism tells us far more about nationalism
than any other episode.”56

These considerations can only make sense if nationalism, as an idea, is

given an autonomous force and can be theorised upon without connection
to material reality. In these accounts, nationalism has an autonomous
existence which means that it can reproduce itself in any context and that
the presence of nationalism is enough to explain nationalism away. The next
section will delve more deeply into the problems that this conceptualisation
carries as well as the circularity of the argument that it engenders. But first,
it is important to explain the origin of the idea that nationalism is not only
autonomous from but also stronger than universal values and class-based
politics. To be clear, this thesis has no intention to dismiss the work of
Gellner, Hobsbawm, Nairn, or Benedict Anderson whose research provided
an invaluable contribution to our understanding of nationalism. On the
contrary, the point is to highlight the limits of their approaches, including the
theoretical issues that hinder our capacity to investigate the politics and
ideologies of contemporary nationalist movements, especially outside the
Euro-Atlantic area.

The most consequential outcome of treating nationalism as an autonomous

force is the diffusionism that this approach engenders. Modernist authors
explain the rise of nationalism and the nation-state in its original European
or American cradle with highly sophisticated arguments and paint

55
Ibid., 22.
56
Ibid., 17.
44
convincing pictures of the process in those specific historical contexts. After
its original emergence, however, nationalism simply spreads to the rest of
the world, pursued by local elites eager to follow the European way of
success. This ‘diffusionism’ significantly undermines the explanatory
capabilities of these theories, relegating most contemporary nationalist
movements to the realm of irrationality and primordial allegiances. In other
words, it denies their politics. This attitude fuels the abovementioned
contempt that these authors show towards real-world nationalist struggles:
if one thinks that nationalism was merely the result of historical processes
taking place in nineteenth-century Europe, then, any expression of
nationalism occurring later or elsewhere will inevitably be autonomous from
those historical processes. It will simply be an idea that spreads across the
world.

For Tom Nairn, the diffusion of the idea of nationalism from Europe to the
rest of the world is central. Nairn sees nationalism as a catching-up strategy
that, rather than developing in the most advanced countries – Britain,
France, and the United States – originated in late-coming Italy and
Germany.57 It was the Italian and German elite, frustrated by the
underdevelopment of their countries and preoccupied with catching up with
their western neighbours that invented nationalism as a strategy of
development. It was the Italian and German nationalist movements that
provided the blueprint for the elites of other latecomers such as Japan and
south-eastern Europe first and the colonised peoples later in the twentieth
century. That the Italian and German unification became points of reference
for other national movements is certainly true but not that consequential.58
Nairn’s account presents these frustrated nationalist elites as a given, failing
to notice how the Italian and German national movements were diverse and

57
Ibid., 14.
58
As Nazih Ayubi explains, the influence exercised by Hegelian idealism of the
German and Italian schools in the Arab world was motivated by the search for a
form of organicist society as opposed to Anglo-Saxon individualism. It was not a
model of nationalism. Nizah N. Ayubi, Over-Stating the Arab State: Politics and
Society in the Middle East (London: I.B. Tauris, 1995), 17-21.
45
conflictual. In both countries, the national movement was initially led by the
urban middle and lower classes that found the only way to obtain liberal or
even democratic constitutions in the liberation from direct and indirect
Austrian domination. In both cases, the Royal Houses of Savoy and
Hohenzollern and the landed interests associated with them were less
motivated by national feelings than traditional dynastic ambitions and were
often in conflict with the respective national movements. The nature of
Italy’s and Germany’s nation-building process was the result of the context-
specific compromise between different forces – and the exclusion of others
– rather than the desire of an unspecified elite to catch up with mighty
Britain. Nairn’s argument is even more problematic when applied to colonial
contexts especially considering the almost universal opposition of national
liberation movements to the indigenous comprador elites compromised with
the colonial power.

In Benedict Anderson’s 1983 Imagined Communities, the blueprint of

nationalism, composed of “nation-states, republican institutions, common
citizenships, popular sovereignty, national flags and anthems, etc.” spread
from the Atlantic Revolutions of the late eighteenth and early nineteenth
century that made the model “available for pirating.”59 Particularly, Anderson
gives centrality to ‘Creole nationalism’ in Latin America showing how the
political modern vocabulary of the French revolution was deployed to create
a regional system of functionally equal and sovereign constitutional
republics.60 These reinterpreted concepts then travelled back to Europe and
provided the basis for nineteenth century nationalism. However, the spread
of these political norms to Europe intertwined with the existence of
vernacular communities created by print capitalism that did not coincide
with the existing polyglot polities of feudal origin – which was a problem
unknown to the Spanish-speaking creoles fighting Spanish colonialism. This
specific European problem generated a form of nationalism built upon the

59
Benedict Anderson, Imagined Communities: Reflecting on the Origin and Spread
of Nationalism (London: Verso, 1983), 81.
60
Ibid., 47-66.
46
centrality of vernacular languages in setting the boundaries of nations.61 In
Europe, nationalism triumphed to such a degree that even the most
reactionary royal houses, previously fierce opponents of the principle of
nationality, had to come to terms and compromise with the new standard of
political legitimacy. This process generated yet another autochthonous form
of nationalism, its most reactionary expression, that Anderson calls ‘Official
Nationalism’ and that was to lead to scientific racism and the tragedies of
the early twentieth century.62 Anderson’s superb description of these
processes and, especially, of how political movements were capable of
adopting ideas developed elsewhere and reinterpreting them according to
their particular predicament, is lost when the book turns to nationalism in
Africa and Asia, where

[a] blend of popular and official nationalism has been the product of
anomalies created by European imperialism: the well-known
arbitrariness of frontiers, and bilingual intelligentsias poised
precariously over diverse monoglot populations. One can thus think
of many of these nations as projects the achievement of which is still
in progress.63
The punctual and evocative terms (creole, vernacular, official), so
effectively deployed by Anderson to stress the originality of each
expression of nationalism in Europe and the Americas, have no
corresponding term to describe the anticolonial of movements of the
peoples of Africa and Asia. The anti-colonial movements of the twentieth
century are instead defined merely as “the last wave” and described as only
able to ‘pirate’ concepts from previous forms of nationalism.

Ernest Gellner’s account of nationalism is heavily informed by Max Weber’s

distinction between traditional and modern societies. Nationalism is the
response to a qualitatively more complex social and economic system:

The level of literacy and technical competence, in a standardized

medium […] which is required of members of this society if they are
to be properly employable and enjoy full and effective moral

61
Ibid., 67-82.
62
Ibid., 83-112.
63
Ibid., 124.
47
 citizenship, is so high that it […] can only be provided by something
resembling a modern 'national' educational system, a pyramid at
whose base there are primary schools, staffed by teachers trained
at secondary school, staffed by university-trained teachers, led by
the products of advanced graduate schools. […] The monopoly of
legitimate education is now more important, more central than is the
monopoly of legitimate violence.64
The most significant marks of nationalism are thus, in Gellner, the political
pursuit of a standardised language and the extension to the masses of the
high and abstract culture that was previously reserved for the elite and that
is now needed for society to function. Gellner’s account illuminates the
process of nation-building in many European countries in the nineteenth
century, convincingly stressing the role played by the standardisation of
culture and mass education. Yet, his theories tell us virtually nothing beyond
that context. Even when accepting that all modern societies need to reach
a degree of cultural unity and a standardised medium of communication, this
process is by no means necessarily associated with the construction of the
nation. One needs just to think about the numerous countries in which the
colonial language kept playing that role – officially or not – after
independence: from the Latin American, to the British white-settler colonies,
to vast polyglot societies such as India, Congo, or Nigeria. Gellner’s stress
on language is revealing of a deep Eurocentric bias. It is not a coincidence
that his Ruritania, the fictional nation used in the book with an illustrative
purpose, is evidently inspired by a handbook case-study of small-nation
nationalism from nineteenth-century Mitteleuropa.

As geographer James Blaut explains, most theories of nationalism follow a

common diffusionist model that fails to provide a materialist explanation for
the spread of nationalist movements in the Global South:

[Nationalism Theory describes nationalism as] a concrete,

observable, social or socio-political process, but its cause, or
source, or mainspring, or motor, is an idea or ideology. This idea is
itself uncaused; or rather it sprang forth in France and Britain 200
years ago as simply the logic of advancing civilization, of creating a
modern nation state; and then the idea diffused to the rest of Europe

64
Gellner, Nations and Nationalism, 34.
48
 and eventually the colonies. Note, therefore, that the idea is
primordial; save perhaps in the original West European ‘homelands’
it arises for no local geographical or historical reason, no reason of
economic impoverishment, political oppression or whatever. It
results only from the diffusion of an idea.65
This diffusionism – a form of idealism of Hegelian origin – pervades most
modernist approaches regardless of the geographical area of the Euro-
Atlantic region – different in each author – where nationalist ideas are
thought to have originated first.

The problem with Nairn’s, Anderson’s, or Gellner’s account is not that

political ideas did not travel back and forth across the Atlantic or that
nationalists around the world did not learn from each other. The problem
with these accounts is rather that they depict nationalism as emerging out
of objective material forces but then spreading purely at the level of political
discourse. While Europeans and Creole elites can incarnate – in Hegelian
terms – the spirit of history, the colonised can only imitate. Nationalism, after
its original emergence, is therefore given an autonomous force that needs
not to relate to contextual material dynamics and conflicts with the result of
de-politicising nationalist struggles.

Is Nationalism an Autonomous Force?

As outlined in the previous section, the most popular scholarly works on

nationalism tend to fall into the trap of diffusionism: nationalism emerged in
a specific context due to specific historical circumstances but then it simply
spreads and can be adopted by political actors even in the absence of those
originating circumstances. This approach seems to be inevitable to authors
trying to build a ‘grand theory’ of nationalism able to grasp this elusive
concept across time and space with a parsimonious and elegant theoretical
construction.

Within such a theoretical framework, nationalism can only survive the

structural context in which it developed due to its intrinsic autonomy from

65
Blaut, The National Question, 18.
49
those conditions. Then nationalism becomes an autonomous force that
exists regardless of contextual factors. Treating nationalism as autonomous
is analytically problematic as it conflates explanandum and explanans,
producing a circular narrative in which national identity is both the object of
investigation and its explanation (nationalism matters because it matters).
However, this approach is also extremely problematic in political terms, as
it depoliticises nationalism concealing the social and political conflicts in
which nationalist struggles are always embedded. Commenting on Nairn’s
book, Blaut observes that foreign domination engendering nationalist
movements

seems to consist in nothing worse than a denial to the elite classes

in the dominated society of the opportunities for greater wealth (…)
Nationalism, then, emerges as a psychological frustration-reaction
on the part of the elites of backward countries to the trauma of
uneven development.66
Presenting nationalism as an autonomous force not only hides social
hierarchies and dismisses conflicts by legitimising the conservative claim
that nationalism is an expression of transclass solidarity. But, even more
problematically - especially for Marxists – it reifies the fictitious separation
between the economic and political spheres that is typical of capitalist
modernity. In pre-capitalist societies, the economic realm is embedded in
cultural and political structures of social reproduction.67 Economic
structures and political, cultural, social (super-)structures are thus
indistinguishable and it is meaningless to speak of a separate economic
realm. As Ellen Meiksins Wood explains, in advanced capitalist societies the
political nature of the appropriation surplus is mystified, and capitalist
reproduction is presented as happening only in the economic sphere and as
compatible with a democratic framework. For example, workers’ struggles
for higher wages “may be perceived as merely 'economic'” while we would
not think that about “the rent struggle waged by medieval peasants, even

66
Blaut, The National Question, 79-80.
67
For a classic exposure of this view, see Karl Polanyi, The Great Transformation:
The Political and Economic Origins of Our Time (Boston: Beacon Press, 2001).
50
though the issue in both cases is the disposition of surplus labour and its
relative distribution between direct producers and exploiting
appropriators.”68 This separation is reproduced by the positivist social
sciences, first of all, through the creation of a separate discipline of
economics based on technical knowledge and ‘apolitical’ general laws.

The fictitious separation of the two spheres is the reason why mainstream
social science can erect boundaries and divide our knowledge of the social
world into separate fields of studies or ‘autonomous’ disciplines. If
international politics follows its own laws that are not dependent on
domestic dynamics, we will need the autonomous discipline of International
Relations. If nationalism is an autonomous force existing regardless of the
way politics and society ‘normally’ work, we will need an autonomous field
of Nationalism Studies. Obviously, a certain degree of disciplinary division
within the social sciences is indispensable to limit the scope of research and
to organise professional academic life. Yet, the strict adherence to
disciplinary boundaries fragments our understanding of the social world and
has long been rejected by all strands of critical scholarship from Marxism to
post-structuralism, from feminism to post-colonialism.

The reproduction of this separation in modernist accounts of nationalism is

most evident in the assumption that capitalism needs no boundaries and
that there is an intrinsic contradiction between the spread of state borders
– engendered by nationalist movements – and the globalising tendency of
capitalism. The multiplication of independent states since the mid-twentieth
century is for Nairn “anachronistic” and in contradiction with the spread of
capitalism and, therefore, evidence that nationalism is a force autonomous
from both class and capitalism.69 Hobsbawm’s entire account of the history
of nationalism, while extremely rich, is based on the idea that nationalism
was rational in the nineteenth century when unified national markets were
conducive to the development of capitalism, and irrational in the twentieth

68
Ellen Meiksins Wood, Democracy against Capitalism: Renewing Historical
Materialism (Cambridge: Cambridge University Press, 1995), 44-45.
69
Nairn, The Modern Janus, 24.
51
century when nation-states were increasingly in contradiction with a
globalising capitalism.70 The idea that capitalism does not need the state
and that tension between the two is ever-present was particularly popular
in the last quarter of the twentieth century and had a prominent role in the
development of the theory (and ideology) of globalisation. On the contrary,
again following Meiksins Wood, it is the separation of the political and the
economic spheres that makes the state even more necessary under
capitalism as “the powers of surplus appropriation and exploitation do not
rest directly on relations of juridical or political dependence” such as that
between masters and slaves or lords and serfs:

Absolute private property, the contractual relation that binds

producer to appropriator, the process of commodity exchange - all
these require the legal forms, the coercive apparatus, the policing
functions of the state. Historically, too, the state has been essential
to the process of expropriation that is the basis of capitalism. 71
The proliferation of states engendered by nationalist movements is thus,
rather than in contradiction to, functional to the spread of capitalism. This
becomes even more evident in the age of global capitalism when huge profit
is made precisely through the exploitation of different wage structures
across countries and labour can be disciplined under the threat of
delocalising production. In these terms, the spread of nationalism and the
spread of capitalism are by no means autonomous from each other.

These reflections can introduce a way to investigate nationalist movements

as embedded in and mutually reinforced by context-specific social conflicts.
The separation of the political and the economic is typical of the advanced
capitalist society and is absent not only from pre-capitalist societies but also
from societies in transition towards capitalism as well as from societies that
occupy a peripheral position in the world economy. Outside the core,
capitalism often maintains a brutally extractive nature and the appropriation

70
At the end of the book, Hobsbawm even concedes to an ‘End of History’-type of
argument (it was published in 1990!) suggesting, through the metaphor of
Minerva’s owl, that we might be at the end of the history of nationalism. Hobsbawm,
Nations and Nationalism, 192.
71
Meiksins Wood, Capitalism against Democracy, 29-30.
52
of surplus is performed via extra-economic – that is to say, political – means,
an arrangement that hinders the establishment of formal democracy and
favours authoritarian solutions. This distinction is extremely relevant to the
study of nationalism because the presence or absence of the formal
separation between the political and economic and, more generally, a
country’s location in the core or periphery of the global economy shape its
class structure and political institutions and, in turn, class conflict and
potential coalition. These considerations can provide an initial explanation
to the reason why nationalist movements against colonialism and neo-
colonialism tend to be progressive and often socialist politics because
distinguishing between political oppression and economic exploitation is
meaningless.

Culture, Ideology, and the Struggle for State Power

Before delving more deeply into the relationship between class and
nationalism, it is necessary to get rid of the proverbial elephant in the room.
As discussed above, modernist accounts tend to attribute an autonomous
force to nationalism. The result is that they end up unintentionally
legitimising the culturalist discourse of real-world nationalists – that is to
say, that national identity is stronger than any other political divide – a result
that defeats the purpose of the modernist critique itself. The relationship
between nationalism and culture is clearly a very complex one. This section
will discuss first the case of nationalisms which stress the role of cultural
homogeneity in setting the boundaries of nations, and then the case of
nationalisms developing within pre-existing borders of pre-national states
or colonies. The comparison of these two historical expressions of
nationalism shows that nationalism comes very close to coincide with the
struggle for state power.

The tendency to conflate nationalism with its nineteenth century European

manifestation leads scholars and laymen to overemphasise the importance
of cultural uniformity in setting the boundaries of nations. This tendency is
most evident, again, in Gellner and in the popularity of his definition of
53
nationalism as the principle demanding the congruity between cultural and
political borders (each nation a state).72 As discussed before, this
conceptualisation has little or no applicability to the anti-colonial
movements of both the nineteenth and the twentieth century where cultural
unity was simply neither the issue of contention nor an objective to pursue.
But even in Europe, where language remains the most common cultural
criterion to determining national boundaries, many nations are multi-
linguistic (such Belgium, Switzerland, Spain) while some linguistic
communities are divided into separate nations (such as Germany and
Austria; Serbia, Montenegro, Croatia, and Bosnia). But even assuming than
the abovementioned are all exceptions and that nations fundamentally
coincide with linguistic communities, there were always countless more
languages than there are nations.

The apparent arbitrariness of the cultural criteria defining the nation is in

reality the result of historical contingency, and of the different conflicts and
divides running through a given society. Working within the modernist
tradition, Paul Brass presents the adoption of specific cultural criteria as the
result of the conscious politicisation of culture by the local elite. Nation-
building, is for Brass essentially

the process by which elites and counter-elites within ethnic groups

select aspects of the group’s culture, attach new value and meaning
to them, and use them as symbols to mobilize the group, to defend
its interests, and to compete with other groups.73
In the struggle for power and resources, elites mobilise objective cultural
features that characterise their ethnic group but that were previously
unfixed and apolitical and turn them into political symbols with a fixed
character.74 These cultural features – language, religion, tradition, collective

72
Gellner, Nations and Nationalism, 1.
73
Paul Brass, Ethnicity and Nationalism: Theory and Comparison (London: SAGE
Publications, 1991), 74-75.
74
For Brass, an ethnic group is “any group of people dissimilar from other peoples
in terms of objective cultural criteria and containing within its membership, either
in principle or in practice, the elements for a complete division of labor and for
reproduction forms an ethnic category.” The internal division of labour
54
memory – are objective in the sense that they really exist but their meaning
in pre-national and pre-modern societies was continuously shifting as
linguistic boundaries were blurry and cultural contamination and religious
syncretism were the rule rather than the exception. In the process of ethnic
and national identity formation, elites select the cultural features that best
serve their political purpose, freeze their meaning, and set the boundaries
of the ethnic community. In doing so, social elites politicise culture and “the
ethnic community or nation created in this way does not necessarily
constitute an entirely new entity but one that has been transformed, whose
boundaries have in some ways been widened, in other confined.”75

Brass identifies several sources of elite conflict that constitute opportunities

for elites to politicise culture: the resistance of a local aristocracy against an
‘alien’ conqueror, the struggle between competing religious elites, and
between religious elites and local or alien aristocracies. 76 The variety of
situations that drive elites to stimulate ethnic and national identities requires
a definition of elites that only partly overlaps with that of classes. If in
agrarian societies the political elite is often the landed aristocracy and thus
coincides with a social class, the class extraction of the religious elite often
matters much less. In other cases, the elites can be just a fraction of a social
class, like in a modernising society where the political elite might be
constituted by the most educated fraction of the middle class as well as
military officers, bureaucrats, or the leadership of political parties and the
interests that they represent are less often defined by their class than by
their institution.77

By eschewing treating nationalism as an autonomous force and by

grounding it instead in the social conflicts that generated it, Brass offers

distinguishes ethnic groups from “non-self-sufficient” forms of social categories

like class, gender, or age. Ibid., 19.
75
Ibid., 244.
76
Ibid., 26.
77
Ibid., 14.
55
“contextual rather than teleological explanations”78 and yet gives us more
applicable theoretical insights than each grand theory of nationalism.
Brass’s approach allows us to look at the construction of national identities
as an eminently political process through which new and old elites can
legitimise their claim to power and mobilised subordinate classes belonging
to the same ethnic group. Their claim to cultural homogeneity is, in these
cases, used to hide opposed class interests. This approach is particularly
effective to understand nationalist movements that aim at independence
and investigate the particular interest that lay behind the call to national
awakening.

In many other cases, nationalism is actively promoted by the elites of states

that evolved from pre-modern and pre-national polities – China, France,
Afghanistan, Thailand, Iran, Russia, Britain, Japan, etc. – whose boundaries
were initially determined with no concern for cultural homogeneity. In these
cases, the nation is an ex-post construct, imposed over borders previously
set by dynastic wars of expansion. Even in those cases, nationalism is often
promoted from the top to play down class antagonisms and stress trans-
class solidarity. This is the case, for example, of the ‘official nationalism’ –
as Anderson calls it – of the late nineteenth century that grew especially in
opposition to the rise of organised labour. As Anderson explains, the link
between nationalism and pre-existing political borders is also fundamental
in the case of colonialism. Administrative colonial borders – as arbitrary as
they are – proved to be extremely resilient, limiting the possibilities of
nation-building available. Anderson cites the example of the creole
republics in Latin America where the borders of the provinces of the Spanish
empire became – roughly – the borders of the new states.79 This aspect is
even more evident in the current borders of most of the post-colonial world
– especially Africa, South-East Asia and the Middle East – which coincide
with administrative colonial divisions. Most attempts to overcome these
artificial borders to build culturally-homogenous nation-states – the Patria

78
Ibid., 244.
79
Anderson, Imagined Communities, 47-66.
56
Grande in Spanish America, the United Arab Republic, the Indonesia Raya –
failed.

What all these different forms of nation-building have in common is that they
all represent forms of struggle for state power: either to gain political
independence or to gain power within an existing state. However, this
argument cannot tell us much about the reason why nationalist movements
adopt such different ideologies spanning from fascism to Black liberation. If
nationalism is essentially the struggle for state power, what determines the
programmatic content of a nationalist movement, its progressive or
reactionary character?

This idea that nationalism is an empty shell devoid of ideological content is

shared by a wide range of political theorists. Liberal scholar Michael
Freeden, for example, claims that “in order to be a distinct ideology, the core
of nationalism, and the conceptual patterns it adopts, will have to be unique
to itself alone” because “ideologies compete over the ‘correct’ meanings of
political concepts.”80 However, the core concepts of nationalism are vague
and empty as they do not provide “answers to the political questions that
societies generate.”81 The core principles of nationalism82 “are too vacuous
[…] to provide interpretations of political reality and plans for political action”
and each of them “logically contains a number of possible meanings” and
answers “to questions of social justice, distribution of resources, and
conflict-management which mainstream ideologies address.”83 In the first
half of the twentieth century, the period that we normally associate with ‘the
apogee of nationalism’,84 all the major political ideologies were framed within
the principle of nationality: from Wilsonian liberal internationalism to the

80
Michael Freeden, ‘Is Nationalism a Distinct Ideology?’, Political Studies, 46.4
(1998), 750.
81
Ibid., 750.
82
For Freeden: (a) the priority given to the nation; (b) its positive valorisation; (c)
the desire to give to it institutional expression; (d) the role of time and space in
determining national identity; (e) the role of sentiment and emotion. Ibid., 751-752.
83
Ibid., 751.
84
Hobsbawm, Nations and Nationalism.
57
Soviet Marxist revision of the ‘national question’ in Lenin and Stalin; from the
racial nationalism of European fascisms to the national liberation
movements in the colonial world. The Marxist Benedict Anderson presents
the principle of nationality, rather than as a modern ideology, as something
akin to religion or dynasticism, “taken-for-granted frames of reference.”85
Nationalism can then assume, as Anderson shows in his Imagined
Communities, radically different political forms.

These considerations lead to the central claim of this theoretical chapter

that, rather than another grand theory of nationalism, is essentially a
methodological proposal. If we consider nationalism as the expression of
the struggle for state power devoid, in itself, of programmatic political
content, and we reject – as discussed earlier in the chapter – the idealistic
and Eurocentric premises of treating nationalism as an autonomous force,
then we need to turn to the context-specific, material conflicts that
accompany nationalist struggles. Studying the class structures and
trajectories of conflicts within a given society can tell us a great deal about
the opposing material interests at stake and thus the ideological tenants
that each party is more likely to profess. In these terms, investigating the
class character of a nationalist movement – the class that leads it, the
classes that take part in it, the classes it opposes – can tell us far more about
the politics of nationalism than each of the grand theories developed within
the field. As James Blaut suggests, maintaining an empirically open
approach is vital if we are to treat nationalism as a form of struggle for state
power embedded in class conflicts:

[W]e should not expect [nationalism] to be associated with one

specific ideology, because each class or class combination in each
kind of nationalist struggle would have an ideological position of its
own and these would moreover differ for different historical epochs
and geographical circumstances.86
In these terms, studying nationalism as embedded in class struggles does
not mean ultimately reducing it to an expression of the conflict between

85
Anderson, Imagined Communities, 12.
86
Blaut, The National Question, 33.
58
producers and appropriators. On the contrary, it means analysing the
conflicts that take place in a given society and the material interest of social
classes, class fractions, and class coalition which determine the allies and
enemies a nationalist movement will choose for itself. Earlier in this section,
Brass was mentioned to show an effective approach to empirically study
how nationalism intertwines with and carries existing local conflicts. We can
thus turn to sociological approaches that can help investigate the political
posture adopted by class actors.

Studying Class

Studying class does not mean disregarding other societal divides such as
gender, culture, or race. On the contrary, class formation is a historical
process and the historical expressions of class identity – and including its
political manifestations – are imbued with contextual norms. The study of
class thickens our understanding of other forms of identity – including
national identities. On the one hand, it helps us make sense of them within
the social context in which they develop and identify their function in the
reproduction of society and its hierarchies. On the other hand, studying
normative and cultural structures in relations to material life guards us
against shallow forms of cultural essentialism. These considerations are
important in order to avoid the forms of economic determinism suffered by
orthodox strands of Marxist theory and class analysis in the twentieth
century that contributed to discrediting the study of class in the social
sciences.

The theoretical reflections elaborated in this chapter hint at an

understanding of class as embedded in power relations. Weberian and
Marxist scholars developed approaches to class analysis that, albeit in
different ways, focus on the centrality of power relations in understanding
class structures. Neo-Weberian scholars tend to focus on the concept of
‘opportunity hoarding’ that see relations among classes as determined by

59
socially-imposed mechanisms of mobility closure.87 Accessing high-income
jobs and powerful positions are restricted to individuals satisfying certain
conditions that vary according to contextual social norms – from boasting a
prestigious patrilineal lineage to holding certain university degrees. The
restrictions placed to limit access to these position of privilege are what
make classes sticky and the mobility across classes limited, allowing for
continuity in their reproduction over time. This approach is particularly
effective in identifying the role of cultural, religious, and racial discrimination
in reproducing class hierarchies. Racializing classes has been a particularly
effective way of limiting social mobility in forms as varied as legal
mechanisms such as Jim Crow or Apartheid or pre-emptively dismissing job
applicants with Muslim names. Focusing on opportunity hoarding
acknowledges the relations of power embedded in class structures
highlighting that “the economic advantages people get from being in a
privileged class position are causally connected to the disadvantages of
people excluded from those class positions.”88

The opportunity hoarding approach to the study of class introduces a

relational element. However, this relation is only unidirectional: the privilege
of the few is guaranteed by the exclusion from privilege of the many. In
Marxist and neo-Marxist work the relational nature of class is based on
interdependency. What constitutes the ruling class is the capacity to
appropriate surplus created by – the capacity to exploit – the working class.
Erik Olin Wright illustrates the difference between the Marxist relational
approach and the Weberian mobility-closure approach through two
“classical cases”:

in the first, large landowners seize control of common grazing lands,

prevent peasants from gaining access to them, and reap economic
benefits from having exclusive control of that land for their own use.
In the second, the same landowners, having seized control of the
grazing lands and excluded the peasants, then bring some of those

87
This is, for instance, the approach taken by Charles Tilly in his famous study on
the persistence of social inequality. Charles Tilly, Durable Inequality (Berkeley:
University of California Press, 1998).
88
Erik Olin Wright, Understanding Class (London: Verso, 2015), 18.
60
 peasants back onto the land as agricultural labourers. In this second
case, the landowners not only gain from controlling access to the
land (opportunity hoarding), they dominate the farm workers and
exploit their labour. This is a stronger form of relational
interdependency than in the case of simple exclusion, for here there
is an ongoing relationship between not only the conditions but also
the activities of the advantaged and disadvantaged.89
In these terms – as claimed by Wrights himself – the Weberian and Marxist
approaches to the study of class are not only compatible but they can be
even seen as complementary due to their emphasis on two different ways
power shapes relations among classes.90

These approaches to the study of class focus on its ‘objective dimension’ as

a social structure observable in all modern societies. But identifying the
configuration of the class structure of society does not in itself tell us much
about the expressions of class politics. The distinction between the
objective and the subjective dimensions of class is present in both Marx –
respectively ‘class in itself’ and ‘class for itself’ – and in Weber, through his
distinction between economic classes and status groups.91 While classes as
an objective phenomenon have objective interests rooted in the conflictual
relations between each other, the actual identity developed by class groups
is inevitably socially constructed and historically contingent. Classes are not
social agents but structures. Saying that political movements are driven by
material interests does not imply that they are the direct political expression
of a class and that they represent the interests of a class in its totality. Class
identity is always shaped by the most diverse range of social phenomena
and must be studied on empirical grounds.

89
Ibid., 10.
90
Erik Olin Wright already highlighted this compatibility claiming that “inside every
leftist neo-Weberian is a Marxist struggling to stay hidden”. Combining the qualities
of both approaches was actually the intellectual aim of Wright in the last part of his
career. Approaches to Class Analysis, ed. by Erik Olin Wright (Cambridge:
Cambridge University Press, 2005), 1-18, 27.
Max Weber, From Max Weber: Essays in Sociology, ed. by Hans Gerth and C.
91

Wright Mills (New York: Routledge, 1948), 180-195.

61
Two elements affecting class identity are particularly central to the
development of class consciousness. Firstly, the transnational spread of
political ideas that shapes class consciousness. Secondly, and relatedly, the
process of political organisation. However, recognising that class interests
are socially and historically constructed does not mean denying that there
needs to be some degree of consistency between culturally loaded
expressions of politics and class interests. Most often, political conflicts are
the expression of narrower material interests and see fractions of the same
class on opposite sides: the national bourgeoisie versus the comprador elite;
the ‘native’ working class versus migrant workers; the international financial
capital versus the domestic market-oriented capital. Such groups are
historical and context-specific expressions of the class structure but do not
strictly overlap with social classes. On the contrary, it is common to see
coalition-building happening among fractions of different classes around
narrow objectives.

These theoretical considerations on the study of class help us identify the

class motives that lead nationalist movements. If nationalist movements are
the expression of classes, class fractions, and class coalitions in their
struggle for state power then the political – programmatic – content of these
movements will reflect a political project for the state they want to conquer
or establish. This approach makes the study of nationalism – as nation-
building projects – quite similar to the study of political regimes as the result
of a certain set of class relations.

The study of the class origins of political regimes and particularly that of the
class configuration that is more likely to lead to democratic and authoritarian
regimes has been for long a central research question within Historical
Sociology. Pioneered by Barrington Moore with his 1966 classic The Social
Origins of Dictatorship and Democracy,92 this approach brought together,
on the one hand, a structuralist study of the origins of social cleavages
rooted in the institutional history and political economy of the context and,

92
Moore, Social Origins.
62
on the other hand, a more actor-oriented approach that looks at how social
actors pursue their strategy through conflict, competition and coalition
building. In this literature, class dynamics occupy a central role in social and
political conflicts as political interests are largely derived from the common
material interest of the members of the group. Within a framework that
explains regime outcomes through the relative distribution of power among
social classes, Moore claimed a positive causal relationship between the
strength of the bourgeoisie and the development of democracy. Building
upon his historical and comparative methodology, a number of scholars
have criticised the liberal bias underpinning Moore’s conclusion and
enriched the field by bringing in the role of the modern state and of the
international system on the one hand, and a problematisation of the role of
the capitalist class in economic development.93

Dietrich Rueschemeyer, Evelyne Huber Stephens, and John Stephens have

most convincingly built upon these critiques in their 1992 book Capitalist
Development and Democracy.94 Rueschemeyer et al. expanded Moore’s
class-based approach to include the role of the state apparatus and
transnational economic processes. However, class remains the central
element “to understanding the social structuring of interests and power in
society”.95 Through their ‘relative class power’ model, Rueschemeyer et al.
survey democratisation processes in the capitalist North Atlantic core, the
Caribbean, and South America to reach the opposite conclusion to that of
Moore. They show that, in virtually all cases, it was the working class that
drove democratic development and that “the classes that benefitted from

93
For example, in her critical review of Moore’s work, Theda Skocpol introduced
some of the elements that will be central to the development of her on work on
social revolution. See, Theda Skocpol, ‘A Critical Review of Barrington Moore’s
Social Origins of Dictatorship and Democracy’, Politics and Society, 4.1 (1973), 1–
34.
94
Rueschemeyer and others, Capitalist Development. Interestingly enough, the
conclusions qualitatively reached by this book have been recently tested and
confirmed quantitatively through a large-n study. See Sirianne Dahlum, Carl Henrik
Knutsen, and Tore Wig, ‘Who Revolts? Empirically Revisiting the Social Origins of
Democracy’, Journal of Politics, 81.4 (2019), 1494–99.
95
Rueschemeyer and others, Capitalist Development, 5.
63
the status quo nearly without exception resisted democracy.”96 Despite
reaching the opposite empirical conclusion, the authors are confident in the
validity of Moore’s framework and particularly in the combination of
historical and comparative research. The two dimensions are seen as a way
to avoid the voluntaristic nature of single-case historical work on the one
hand and the overly structural analysis of comparative research on the
other:

to construct a framework of enquiry that is in principle equally well

attuned to the study of process and to the recognition of structural
constraints. […] The voluntaristic bias of case-oriented research is
counterbalanced by comparison. Even in single-case studies
comparative awareness and especially a longer time span of
investigation can […] make the structural conditions of different
event sequences more visible. It is, however, actual comparison of
cases featuring different structural conditions that really turns things
around.97
While this approach offers a convincing framework to study political actors
as social forces as well as the structures constraining their action, it also
calls for a clearer definition of the unit of analysis. The need to
operationalise the concept of class in social research in order to study
political action also as class action requires moving away from a dichotomic
division of capitalist societies between the two classes of those who sell
and those who buy labour. Building upon the Weberian tradition,
Rueschemeyer et al. define class through the concepts of mobility closure –
the tendency of moving between similar class positions – and of interaction
closure – the tendency to interact mostly with members of the same class.
This definition opens up social classes and brings a more complex picture
to the surface, in which different fractions of the same class might have
opposing interests:

96
If capitalist development favours democracy it is not due to but, if anything,
despite of, the capitalist class. According to Rueschemeyer et al. what enhances
the chances of democracy is the structural transformation brought by capitalism
that weakens the landowning pre-capitalist elite and strengthens the urban
working classes. Ibid., 41-47.
97
Ibid., 33-34
64
 With these analytical tools […] we can distinguish the owners of
capital who employ labor on a sizeable scale – the bourgeoisie
proper – from the urban petty bourgeoisie. We can identify the lower
non-manual employees – such as clerical workers and sales clerks
without much of a supervisory role – as a class distinct from middle-
level managers and professional experts outside the chain of
command.98
This approach allows for a much more flexible reading of class dynamics
and is able to account for intra-class conflicts and processes of coalition
building between fractions of different classes.

The methodology deployed by historical sociologists aimed at explaining

the social origins of political regimes as grounded in relations between social
classes, can help us investigate the political content of nationalism.
Nationalist movements – like all social movements – reflect a class coalition
that shapes their political content and programme and that determines what
kind of nation they aspire to build. Acknowledging that the political posture
– in this case, the approach to nationalism – of social classes in each specific
context must be assessed on empirical grounds, the next section sketches
the historical patterns of political behaviours adopted by social classes
which can guide the empirical analysis to follow.

Sketches of Nationalisms

The last two sections of the chapter discussed the role of existing material
conflict in shaping nationalist movements as well as the political posture of
social classes. In both cases, I emphasised the context-specific nature of
every expression of nationalism and the necessity to study nationalist
movements on empirical grounds. However, after acknowledging that the
“political posture of class actors cannot be read off the underlying class
structure in any one-to-one fashion,” Rueschemeyer et al. claim that there
“are not infinitely variable either. […] we expected classes to exhibit
definitive central political tendencies.”99 If class conflicts are one of the

98
Ibid., 51-53.
99
Ibid., 5.
65
fundamental drivers of politics, we can make use of history to identify
general patterns of behaviour for class actors in different contexts.

Even if social classes always find their political expression in a subjective

form, shaped by cultural norms and local history, classes are ultimately
made of objective social relations that present similarities across different
contexts: we talk about a Bengali peasantry and a Salvadoran peasantry
because we identify a number of similarities such as a particular relation of
property to the land. Class analysis is particularly effective to carry
comparative research on nationalist movements precisely because it allows
us to pin down differences and similarities across different cultural systems
that, with different approaches, would remain incommensurable. Every
social class tends to develop a discourse about the nation and to develop a
form of nationalism loaded with grievances and demands that reflect the
understanding that class has of itself and its material interests.

The following section identifies historical patterns in the political posture

developed by social classes towards nationalism. It shows that identifying
the structural location of each class and the material interests that derive
from it can tell us a great deal about the political programme of a nationalist
movement, its ideology, and its progressive or reactionary character.

Classical Bourgeois Nationalism

The ‘classical’ form of nationalism is the one expressed by the capitalist

bourgeoisie and deeply associated with the nineteenth-century transition
from the Ancien Régime to the modern nation-state in Europe. In pre-
modern Europe, political power was restricted to the aristocratic and clerical
castes and society was regulated through a complex net of historically
sedimented privileges. The bourgeoisie that started acting according to a
capitalist logic found its rise restricted by insurmountable legal and political
obstacles. It was against this system that the bourgeoisie developed its
nationalism as a political project to transform the state at its own image.
Bourgeois nationalism was thus anti-feudal – aiming at replacing particular

66
privileges with universal rights enshrined in a liberal constitution – and
secular – aiming at breaking the alliance of ‘throne and altar’. It aimed at
replacing parochial traditions and local dialects with a uniformed high
culture and standardised language. All these political ideas have an obvious
connection to the material interests of the class that produced them. This
national project created centralised states and unified markets in which
capital could develop and labour was ‘freed’ from feudal and corporatist
constraints. Even for Karl Marx and Friedrich Engels, witnesses to this
process, the creation of this form of state was the historical mission of the
capitalist bourgeoisie.

This is the form of nationalism that was of interest for most of the academic
discipline of Nationalism Studies and that can be prototypically exemplified
by the national movements in Italy and Germany in the nineteenth century.
If we limited our interest in nationalism to this historical expression, Tom
Nairn’s Modern Janus, criticised above, would become significantly more
relevant. The industrial bourgeoisie was indeed a class deeply concerned
by the competition of more advanced economies. The cases of Japan, Italy,
and Germany, cited by Nairn, were all characterised by an economic elite
determined to shelter local production and in all these cases nationalism
came to be associated – although to a different extent – with protectionist
trade policy.

The political content of bourgeois nationalism varied depending on the

specific weight gained by this class in each context as well as the alliances
it developed. In mid-nineteenth century Europe – especially during the 1848
revolutions – European bourgeoisies largely participated in the revolutionary
movements led by the urban working and middle classes on a platform of
liberal nationalism, although ambivalent towards their democratic
tendencies. However, when the European bourgeoisie had consolidated its
power in the second half of the century, bourgeois nationalism became the
strongest support of the monarchy and was used to justify militarism and
colonial expansion – Anderson’s ‘official nationalism’ – as well as to keep in
check the rising power of organised labour.

67
Bourgeois nationalism was, in these terms, a largely European phenomenon
deeply connected to the rise of capitalism in the continent and, as noted
earlier, the tendency to subsume all forms of nationalism under this is the
result of the Eurocentric bias of the field of Nationalism Studies. In the
decolonising countries of Asia and Africa in the twentieth century, the
bourgeoisie was nowhere close to being a hegemonic class, also due to the
constraints imposed by colonial domination on capitalist development.
While bourgeois nationalism was in itself a virtually insignificant
phenomenon, the posture of the bourgeoisie towards nationalism most
often depended on its economic relation with the colonial power. In these
contexts, the most relevant distinction is that between the national
bourgeoisie, with economic interests in the domestic market and thus
nationalist and anti-colonial, and the comprador bourgeoisie, economically
dependent on the colonial power and thus opposed to independence.

Middle-Class Nationalism

Another ideal-typical form of nationalism is the one developed by the middle

class. However, setting the boundaries of this class is more difficult. By
‘middle classes’ we mean a diverse set of intermediate social positions that
are not directly related to the process of production and that include both
the civil society – professionals such as lawyers and doctors, white collars,
traders – and the state apparatus – bureaucracy, army officers and teachers.
Nationalism is particularly appealing to these groups because they are most
likely to provide the staff for a newly-independent or strengthened state
due to their education and social status, but also their location ‘outside’ the
production process that allows the bureaucracy to claim a mediatory role in
conflicts between capital and labour. Schematically speaking, in cases in
which the natives are excluded by the administration and the security
forces, middle-class nationalists are most often interested in replacing
foreign staff. In the case of colonial states, the anti-colonial nationalism of
the native bureaucrats and officers most often reflects their ambition to shift
from a position of executors of colonial policies to that of the ruling elite of
68
an independent state. On the one hand, middle-class nationalism opposes
the traditional ruling class by proposing a progressive ideology – liberal or
socialist – whereby some sort of meritocracy takes the place of the privilege
of birth as the main mechanism of social mobility. On the other hand, it tends
to assume a state-centric, developmental and technocratic character that
limits the involvement of the popular masses in the political process.

These considerations are consistent with the findings by Rueschemeyer et

al. about the historically “ambiguous” posture of the middle classes towards
democracy:

They pushed for their own inclusion but they are most in favor of full
democracy where they were confronted with intransigent dominant
classes and had the option of allying with a sizeable working class.
However, if they started feeling threatened by popular pressures
under a democratic regime, they turned to support the imposition of
an authoritarian alternative.100
With regards to nationalist movements like in the democratisation process,
the intermediate position of the middle classes allows for a variety of
coalition-building strategies. On the one hand, fascism developed often a
middle class reaction – driven by frustration and fear of social downgrading
– to the rise of the labour movement and resulted in an alliance with the
dominant classes. On the other hand, decolonisation was often the product
of a progressive coalition led by the middle classes and enpowered by the
support of the organised working classes or peasantry.

The latter case is particularly significant because in many postcolonial

societies, army officers state bureaucrats ended up becoming – thanks to
their control of the state – the dominant social group. In the words of
Guinean anticolonial leader Amilcar Cabral, colonialism – due to its
extractive nature – imposes severe limits on the integration of the educated
middle class into the colonial elite:

a feeling of bitterness or, a frustration complex is bred and develops

among the indigenous petite bourgeoisie. At the same time, they are
becoming more and more conscious of a compelling need to

100
Ibid., 8.
69
 question their marginal status, and to re-discover an IdentIty. […]
Thus, they turn to the people around them, the people at the other
extreme of the socio-cultural conflict: the native masses.101
In this sense, the limits posed on their upward mobility is the strongest
material motive that drives the native middle classes against the colonial
power that heavily contributed to their own formation as a class.

Colonialism tends to shape public institutions according to the extractive

interests of the colonial power and to rely on traditional pre-capitalist elites.
On the hand, this generates security-oriented colonial states in which the
army and the bureaucracy tend to be oversized and state officials constitute
an educated and connected mass, concentrated in the major urban centres
and easy to mobilise. On the other hand, the weakness or sheer absence of
a capitalist bourgeoisie whose development is actively hindered by the
traditional elite in power allows the middle classes to claim the leadership of
the national liberation. This position of leadership in such contexts is shown
by the capacity of middle-class nationalists to incorporate the struggles of
the subaltern classes. An exemplary case is that of Arab nationalism in the
1950s and 1960s. Largely led by military officers and state officials, Arab
nationalists in Egypt, Iraq, and Algeria conquered state power thanks to their
capacity to integrate workers’ and peasants’ struggle within the national
liberation movement.

As Aijaz Ahmad explains, the apparently contradictory ideological

expressions of these classes in power have all in common

a certain fetishization of the State, and the creation of a whole range

of disparate and mutually contradictory ideologies – e.g. Western-
style developmentalism, the ‘socialism’ of the radical-nationalists
with its emphasis on ‘nationalizations’, the ethno-religious fascism of
the Khomeini variety – which are none le less united in viewing the
state as the principal agency of social transformation.102

Amilcar Cabral, Return to the Source: Selected Speeches (New York and London,
101

Monthly Review Press, 1973), 63.

102
Aijaz Ahmad, ‘Class, Nation, and State: Intermediate Classes in Peripheral
Societies’ in Dale L. Johnson (ed.) Middle Classes in Dependent Countries (Beverly
Hills and London: Sage Publications, 1985), 44-45. It could be argued that the
statolatry of European fascisms comes from the same class dynamics.
70
 Feudal Nationalism

In the classical forms described above, nationalism remains largely a ‘Tiers

État’ phenomenon. The feudal classes are, in the conventional narrative, the
defenders of the old order and the most strenuous opponents of the
egalitarian principles inscribed in the idea of the nation. Yet, history is
punctuated with examples of landed aristocracies raising the national
banners and it is so especially in nineteenth-century Europe, the classical
locus of nationalism. As eastern Europe was dominated by the multi-national
Austrian, Russian, and Ottoman empires, nationalism often emerged from
the local landlords concerned with the distribution of land to ‘alien’
aristocrats from the imperial core, the exclusion from public offices, or the
abolishment of ancient privileges. In Hungary and Poland, the nationalism of
the powerful feudal class was directed against the absolutist tendencies of
the Austrian and Russian empire. In the latter case, the memory of the
‘Golden Liberty’ the collective aristocratic rule of pre-partition Poland that
excluded both the towns and serfs, motivated the anti-Russian feelings of
the landed class:

Polish nationalism remained, even after the revolutionary crisis of

the seventeen-nineties, essentially an affair of the landed gentry as
well as of those social strata which, like the emerging inteligencja,
originated from, and were still linked with, the traditional feudal elite
[that] failed to inject into the Polish bourgeoisie and the peasant
masses a feeling of national solidarity strong enough to outlast the
loss of political independence.103
But this form of nationalism is by no means limited to eastern Europe and
similarities can be spotted in the example of the role played by planter class
in the national movements of Latin America mentioned before in this
chapter. Benedict Anderson offers an interesting and original account on
the deep motives of the Hispano-American nationalisms and of the use that
the creole elite made of the ideas of self-determination and

103
Michael G. Muller, ‘Poland’ in Nationalism in the Age of the French Revolution,
ed. by Otto Dann and John Dinwiddy (London and Ronceverte: The Hambledon
Press, 1988), 113-114.
71
constitutionalism. While in France, these ideas were used to take down the
Ancien Régime, in Latin America the same ideas were used by the creole
landowning class to oppose the metropole when Madrid became more and
more inclined to make concessions towards the indigenous peasants and
the Black slaves and less and less willing and able to protect the local elite
against major revolts.104 The deployment of the same set of ideas had
radically different outcomes in France, where it helped dismantle feudal
structures, and in Latin America, where it helped preserve or even
strengthen the local racialized class hierarchies.

For Paul Brass, this ‘landed-class nationalism’ is a common case in contexts

characterised by cultural differences – linguistic or religious – between the
imperial centre and the local aristocracy which legitimise the claim to local
power of the latter.105 Feudal nationalism is thus often triggered by the
imperial centre’s attempts to replace the local elite with loyal nobles from
the core, or by the alliance between the imperial centre and other local
classes – the burghers or even the peasantry – to bypass local powerholders
and promote state centralisation. It is thus a form of nationalism
characterised by reactionary – in the literal sense – demands. Even in regard
to the feudal class, the findings of historical sociological work on
democratisation point to a consistently reactionary posture of this class. In
Barrington Moore’s classic study, “labour-repressive” agriculture – such as
feudalism or plantation slavery – that “require political methods to extract
surplus, keep the labour force in place and make the system work” is
characterised by a landowning class that is the most irreducible opponent
of democratisation as the emancipation of agricultural labourers would
inevitably bring the end of their power.106

104
Anderson mentions the ‘Indian revolt’ in Peru, led by Tupac Amaru in the early
1780s and the Haitian Revolution of 1791-1804. Anderson, Imagined Communities,
48-49.
105
Brass, Ethnicity and Nationalism, 26-27.
106
Moore, Social Origins, 434.
72
In decolonising societies, the ‘feudal class’ was generally identified with the
traditional landed elite opposed to the national movement. One of the
almost107 universal tendencies of colonialism is to rely on the existing
traditional elite to sustain colonial domination. As in the abovementioned
case of the comprador bourgeoisie, these classes are empowered by the
foreign rulers – and even revitalised where they were in decline – giving
them a stake in the continuation of colonial rule. They thus tend to oppose
national liberation movement. The case of the Arab world is one of the most
significant in this sense. British colonialism actively turned traditional
notables and tribal leaders in Egypt, Jordan, and Iraq, in a powerful class of
landowners which became the most strenuous defenders of the monarchies
backed or even imposed by Britain.108 As mentioned earlier, Arab nationalism
and Pan-Arabism in the Cold War era found their most significant supporters
in the urban middle classes and their direct target in the traditional elites
compromised with colonialism.

Nationalism and the Subaltern Classes

The term ‘subaltern classes’ begs a definition. In Antonio Gramsci, the

subaltern groups coincide collectively with the dominated masses, defined
in terms of their exclusion from the political process but fragmented along
the lines of different forms of exclusion constituted by relations of
exploitation, race and nationality, gender, religion.109 Exclusion and
fragmentation, imposed by the dominant classes, are the reasons why
Gramsci’s subalterns do not constitute a class for itself and live “on the

107
Of course, except for the cases of settler colonialism where the indigenous
population – elite and commoners – is subject to complete exclusion and
suppression.
108
Ayubi, Over-Stating the Arab State, 86-134.
Marcus E. Green, ‘Rethinking the Subaltern and the Question of Censorship in
109

Gramsci's Prison Notebooks’ Postcolonial Studies, 14.4 (2011), 387.

73
margins of history” or, as in postcolonial scholarship, they “cannot speak.” 110
It is through the process of political organisation that the subalterns
overcome their fragmentation and gain political agency and autonomy as a
class and thus are able to create an alternative social bloc and “become
state.”111 The relevance of this concept to the present discussion on
nationalism is that through political organisation the subaltern classes, like
all the other classes, can speak, as they can develop their own discourse on
the nation and launch their own bid for state power. Due to their
subalternity, their position of social subordination, subaltern politics will
have an intrinsic emancipatory character and will tend to develop in
combination with a progressive ideology.

In the mid-nineteenth century, the movements that led to the pan-European

1848 Revolution had a strong working-class component that was both
patriotic and radical democratic or even proto-socialist and no contradiction
was seen between the two. It was only towards the end of the century that
Nationalism and socialism were often presented as the competing products
of mass politics in the industrial era respective expressions of the middle
classes and the working class. As an expression of working-class politics,
socialism was characterised, since the early nineteenth century, by a strong
internationalism that claimed a commonality of interest among workers
beyond national borders. This allowed right-wing nationalists to present
socialists and labour organisers as ‘anti-national’ and saboteurs when they
rejected chauvinism or colonial expansion. Yet working-class social
democratic parties in western Europe, the closer they got to political power,
the keener they were to reject this accusation and to frame their claim to
power in national terms: In France, Germany, and Britain, they ultimately
supported their countries’ war effort in 1914-1918, and developed an
ambiguous approach towards colonialism proving their willingness to

110
Gayatri C. Spivak, ‘Can the Subaltern Speak?’ in Marxism and the Interpretation
of Culture, ed. by Cary Nelson and Lawrence Grossberg (Urbana and Chicago:
University of Illinois Press, 1988).
Antonio Gramsci, Selections from the Prison Notebooks, ed. By Quintin Hoare
111

and Geoffrey Nowell Smith (New York: International Publishers, 1971), 53.
74
support even an expansionist form of nationalism. At the time of the Russian
revolution, both communists and social democrats debated intensively the
‘National Question’ and Bolsheviks proclaimed their support for the
liberation of the colonised world.

In the course of the twentieth century, the working classes and the
peasantry became the backbone of virtually the entirety of the global
movement for decolonisation to the point that Marxism became the – almost
universally – shared language of the struggle against colonialism. The
subaltern classes, where politically mobilised, were more consistently anti-
colonial than the other classes as they had to bear the heaviest brunt of the
transformation imposed by colonialism of traditional social relations and
property regimes. In semi-colonies such as Egypt and Iraq in the 1950s, the
subaltern classes were part of wider political coalitions in favour of
independence, and workers’ and peasants’ mobilisation were decisive in
giving the final shove that allowed the nationalists to win. In China, the
communist mobilised the working class and the peasantry to fight a
liberation war against the Japanese invaders, parallel to the one waged by
the ruling-class nationalists of the Kuomintang. After the war, the two sides
fought for state power proposing radically alternative national projects.

Thought through in theoretical terms, subaltern class nationalism primarily

aims at a restructuring of the power relations within the state from which
the subordinate has much more to gain than they have from external
expansion. Foreign rule and colonial domination are most often
characterised by hyper-exploitation and extractive practices which push
the subordinate classes to support independence and to lead
independentist movements. If nationalism is defined in class terms, as a
struggle for independence and state power, it is also hardly in
contradistinction with the internationalism historically developed by
working-class movements. Internationalism literally presupposes the
division of the world into nations prescribing fraternity among them. That is
the reason why we see no contradiction in the expressions of solidarity

75
towards the struggle for self-determination of oppressed nations, such as
the Palestinians or the Kurds.

Conclusion

This chapter proposed an alternative way to study nationalist movements.

Established theories of nationalism present sophisticated accounts of the
origins of nationalism explaining its development as the result of objective
historical circunstances and placing its origin in a specific time and space –
most often nineteenth century Europe. However, after its ‘rational’ origin,
nationalism simply diffused to the rest of the world even in the absence of
the original conditions of its emergence. Nationalist struggles in the colonial
world appear to be driven by the autonomous force of the principle of
nationality disconnected from pre-existing social and political conflicts.
Nationalism is thus depoliticised as the political content of nationalist
movements – their progressive or reactionary character – is only given
secondary importance.

However, if both academic and public discourse can use the term
nationalism to describe movements as far apart as a neo-Nazi group and a
Marxist national liberation movement, the term inevitably loses meaning and
analytical usefulness. To shed light on the ideological content and
programme – that is to say, on the actual politics – of nationalist movements,
this chapter proposed to study nationalism as the expression of the struggle
for state power embedded in class conflict and coalition building. Studying
the class politics behind a nationalist movement allows us to understand
their ideological positions and political posture in the interests it represents
and serves within society. This means that nationalist movements must be
ultimately studied on empirical grounds in light of the specific class structure
and relations of power out of which they developed.

The chapter showed how the study of class origins of political regimes
within Historical Sociology can provide a point of reference to identify
historical patterns of the political behaviour of social classes. The last

76
section of the chapter sketched these historical patterns surveying the
forms of nationalism developed by social classes across history and
formulating theoretical considerations that can guide the study of nationalist
movements on empirical grounds, which is the subject of the following
chapters.

77
Chapter 3
Feudal Nationalism in Kurdistan
(1918-1946)

Introduction

This chapter reconstructs the social origins of Kurdish nationalism in the

interwar period showing that the first phase of the Kurdish national
movement was essentially a class project promoted by the tribal landowning
elite. Until the collapse of the Ottoman Empire in 1918, the Kurdish traditional
elite largely identified with the Ottoman state and legitimised its position of
power through tribal and religious sources. With the establishment of the
Middle East state-system after World War I, the Ottoman successor states
of Turkey and Iraq experienced a series of revolts whereby the Kurdish tribal
elite attempted to renegotiate their power over the tribal periphery
threatened by the centralising ambitions of the new states. Kurdish identity,
only elaborated in cultural terms in the Ottoman context, was thus
politicised in the form of a feudal nationalism, loaded with the conservative
demands of the tribal elite and deployed to legitimise their claim to power
over the region. The pre-nationalist Kurdist discourse that had developed in
the previous decades, was thus emptied of its initial progressive and
modernising character and filled with political demands that reflected far
narrower class interests of the Kurdish traditional ruling class.

The Kurdish revolts that took place in Turkey, Iraq, and, to a lesser extent,
Iran, between 1919 and 1946, followed a well-established pattern of state-
tribe relationship and presented a number of shared features that allow us
to study them as a unitary phenomenon. Despite their narrow class basis,
the feudal-nationalist revolts – and particularly the short-lived Kurdish
Mahabad Republic – assumed a central role in the nationalist narrative and
the development of a pan-Kurdish identity. This step is particularly
78
important because the different extent of the defeat of feudal nationalism –
and of the Kurdish traditional class – in Iraq and Turkey engendered two
divergent paths of development for Kurdish nationalism.

Kurdistan in the Late Ottoman Period

Kurdish nationalism developed in a time of great transformation. Between

the early sixteenth century and the First World War, Kurdistan constituted
the border region between the Ottoman and the Persian empires. Given the
remote and mountainous nature of Kurdistan, the two empires allowed a
higher degree of autonomy for the local elite and the region was organised
in a system of vassal emirates functioning as border marches. The Kurdish
emirs were paramount chiefs ruling over a tribal confederation and their
legitimacy rested on their vassalic relationship with the Ottoman Sultan or
the Persian Shah. Their courts thrived on their location along the Silk Road
and allowed for the first development of several Kurdish languages in the
written form.

The power of the Kurdish emirs on the Ottoman side of the border ended in
the mid-nineteenth century when the modernisation project promoted by
the Porte in Istanbul required – and allowed for – a stronger presence of the
central authority in the periphery of the empire. One by one, the emirates
were suppressed through a series of military campaigns. However, the
actual presence of the Ottoman government was all but ineffective and the
local elite, composed by aghas, the tribal chiefs, and shaykhs, the heads of
Sufi brotherhoods (generally Qadiriyah and Naqshbandiyyah) maintained a
strong political role. As a matter of fact, tribal leaders were empowered by
the replacement of the paramount authority of the emirs with the distant
rule of Istanbul. The suppression of the emirates led to increasing inter-tribal
violence. Unrelated to any of the local tribes, the shaykh traditionally held
the position of mediator between aghas and, with the spread of tribal

79
conflicts in the late nineteenth century, the shaykhly caste experienced a
political golden age.112

The transformation of the region was accelerated by the inclusion of

Kurdistan into the imperialist projects of the European powers through the
disruptive effects of the Christian missionaries active among the religious
minorities and the looming threat of Russian expansion. Moreover, the
Kurdish areas were severely impoverished by the decline of the land routes
of long-distance trade weakened by the growth of European maritime
power. With trade, the Kurds also lost a significant part of their rich
manufacturing production, especially textile, due to the competition of
cheap industrial products from Europe. Distance from the sea denied
Kurdish agriculture the opportunities brought to farmers of coastal Anatolia
by the increasing European demand for agricultural goods. If between the
1830s and 1912 Istanbul and Izmir almost tripled their population, the
Kurdish Diyarbakır lost a quarter of its inhabitants.113

Part of the long-term process of modernisation and centralisation of the

Ottoman Empire – known as Tanzimat (reorganisation) period – the Land
Code of 1858 had a deep impact on the Kurdish region. The Land Code
promoted the private property of the land with the aims of boosting
agricultural production and creating an easily taxable class of landowners.
In Kurdistan and other peripheral regions, the project had also the objective
of sedentarising nomadic and semi-nomadic tribes by turning herders into
farmers. The year of 1858 should be seen as the starting point of a long
historical process of transformation, since the actual enforcement of the
Land Code took decades and, in some areas, was only implemented by the
states that succeeded the Ottoman Empire. The Kurdish tribal elite,

For more on the instability created by the suppression of the emirates, see van
112

Bruinessen, Agha, Shaikh and State, 175-181; McDowall, A Modern History, 38-53;
Hamit Bozarslan, ‘Tribal Asabiyya and Kurdish politics: a socio-historical
perspective’ in The Kurds: Nationalism and Politics, ed. by Faleh A. Jabar and
Hosham Dawod (Beirut: Saqi, 2006), 134-138.
113
Charles Issawi, The Economic History of Turkey, 1800-1914 (Chicago: University
of Chicago Press, 1980), 34-35.
80
empowered by the suppression of the emirates, greatly benefitted from this
long-term process of agrarian reform. In his work on the tribal structure of
Kurdish society, anthropologist Martin van Bruinessen explains that Kurdish
rural life gradually lost its communal features.114 The aghas and – and even
more frequently – the shaykhs took advantage of their position to register
communal, tribal, and religious lands under their name and then to enforce
their property rights.

This process of class stratification of Kurdish society was accelerated in

1891 when the Ottoman Sultan Abdülhamid II tried to incorporate the Kurdish
tribes into the Ottoman state and army through the institution of the
Hamidiye, a corps of irregular cavalry modelled after Russia’s Cossacks. By
distributing weapons directly to the tribal chiefs, the Hamidiye increased the
power of the aghas and brought about a sort of tribal revival. It was never
fully controlled by the Ottoman authority and was often involved in sectarian
violence contributing to the collapse of the fragile intercommunal
relations.115 The establishment of the Hamidiye cavalry increased the
coercive power of the tribal leadership and their capacity to grab communal
lands through violence.116

The transformation of Kurdish tribal society induced by the suppression of

the emirates, the Land Code, and the institution of the Hamidiye led to a
deep restructuring of the class structure of the region. As the cultivators
were deprived of their traditional collective rights, the tribal elite gradually
evolved into a class of large landowners. Most of their non-tribal serfs, as
well as many tribespeople, became their sharecroppers or waged labourers,
a process that gradually reduced the difference between the two groups
and that enormously increased the power of the tribal elite over the rest of

114
Van Bruinessen, Agha, Shaikh and State, 181-184.
115
Janet Klein, The Margins of Empire: Kurdish Militias in the Ottoman Tribal Zone
(Stanford: Stanford University Press, 2011).
Yadirgi, The Political Economy, 160-120. According to Robert Olson, in 1910, the
116

Hamidiye cavalry enlisted 53.000 men. Robert Olson, The Emergence of Kurdish
Nationalism and the Sheikh Said Rebellion, 1880-1925 (Houston: University of
Texas Press, 1989), 10-11.
81
the Kurdish rural population. This process of social stratification, that was
still ongoing in the early decades of the twentieth century, is essential to
understanding the social context in which Kurdish nationalism developed
and spread. The modernising reforms of the Ottoman sultans resulted in a
radical transformation of the Kurdish tribal elite with many leaving the
countryside and turning into a class of absentee landlords.117 However, their
tribal lineage continued to serve as a source of legitimacy to the power that
they exercised over their constituencies, often their former fellow tribesmen
turned peasants.118

In the rapidly changing social context of the Ottoman Kurdish provinces,

nationalist discourse had, up to World War I, an almost insignificant
presence. The Kurdish elite kept drawing its power from traditional sources
of legitimacy and therefore they neither challenged that of the Ottoman
sultan nor imagined themselves outside the boundaries of Ottoman
citizenship and Muslim identity. Only at the end of this period, the Young
Turk Revolution of 1908 boosted the development of ethnic nationalism in
the Empire and especially that of Turkism.119 The Young Turks, even though
officially supportive of Ottoman unity,120 gave way to the development of a
specific Turkish identity. As the term Kurd was at that time still commonly
associated with a tribal condition and not a national identity, most members
of the urban Kurdish elite saw themselves as Ottoman citizens and even

117
Van Bruinessen, Agha, Shaikh and State, 81.
118
The process described in this paragraph regards largely the Muslim population.
Most of the autochthonous Christians living in the region (Armenians, Syriacs,
Assyrians) had always been politically subjected to the Kurdish (or Turkish) tribes.
However, the condition of the Christian population is less relevant to this study
since most of the Christians in the region were killed, assimilated or forced to leave
during World War I.
119
In Erik Zürcher’s account of Turkish history, the Young Turks Revolution on 1908
represents the decisive moment that divides Late-Ottoman and modern Turkish
history. Erik J. Zürcher, Turkey: A Modern History (London and New York: I.B.
Tauris, 1997), 1-8.
120
For a definition of the late Ottoman political identity see Zürcher’s concept of
Muslim nationalism. Eirk J. Zürcher, ‘The Vocabulary of Muslim Nationalism’,
International Journal of the Sociology of Language, 137 (1999), 81–93.
82
embraced Turkish nationalism.121 Nevertheless, the idea of ‘Kurdishness’ as
a distinctively cultural, rather than social, identity started to rise in the years
between the 1908 revolution and the First World War. The first manifestation
of Kurdish national identity was still expressed within the framework of
Ottomanism and its promoters called for autonomy for the Kurds within the
Ottoman state, rather than independence. Kurdish national consciousness
was in that sense neither particularly different nor particularly late,
compared to Turkish or Arab nationalisms. In general terms, promoters of
national identity among the Muslim peoples of the empire did not imagine
themselves out of the framework of a reformed – for Turks – and
decentralized – for Arabs and Kurds – Ottoman state.

The first Kurdist clubs and papers were established in Istanbul in the early
twentieth century to demand administrative, economic, and educational
reforms for the Kurdish provinces. They were promoted by a cohesive group
of aristocratic Kurds mostly working for the Ottoman state as public
servants and “connected to one another through kinship ties or the
Naqshbandi network”.122 A prominent role was assumed by once-powerful
Kurdish princely families, such as Bedirkhans and Babans, exiled to Istanbul
after the suppression of the emirates and their following was largely
constituted by Kurds hailing from leading tribal families but based in the
capital. They share an urban lifestyle and cosmopolitan education but also
a deep disconnect with Kurdish tribal and rural society which they viewed
with a “kind of paternalism” as composed by infants in need of their
leadership to enter the modern world.123 The efforts to spread their views in
the Kurdish provinces met little enthusiasm and, the few branches of the
Kurdist clubs that were opened in Kurdish cities like Diyarbakır or Bitlis,

A case in point is that of Ziya Gökalp (1876-1924), a half Kurd from Diyarbakır
121

who became one of the most enthusiastic ideologues of Turkish nationalism. See:
Heper, The State and Kurds, 52-57.
122
Hakan Özoğlu, Kurdish Notables and the Ottoman State (Albany: State
University of New York Press, 2004), 87.
123
Keith Hitchins, ‘Kurdish Elites and Nationhood in Anatolia, 1890s-1938’, in Joyce
Blau, ed. by Bozarslan and Scalbert-Yücel.
83
“became, in effect, local organizations, where, in the absence of
intellectuals of the sort active in Istanbul, local elites, espousing more
traditional ideas, used the clubs to further their own interests.” 124 Alienated
from their lands and with a narrow social base, these early pioneers of
Kurdish identity had very few followers in the Kurdish provinces and even
fewer among the masses.125 As Janet Klein argues, describing these early
Kurdist organisations as proto-nationalists fits into an Orientalist and
Eurocentric historical narrative that portraits the Ottoman Empire as
destined to collapse as the result of the inevitable spread of opposing
nationalisms.126 On the contrary, as long as the Ottoman Empire existed,
nearly all political expressions of Kurdish identity remained within the
framework of the Ottomanist ideology. It was only after the end of the
empire that Kurdish identity came to express a more unambiguous demand
for separation.

Defining Feudal Nationalism

This section conceptualises the first stage of Kurdish nationalism from 1918
to 1946 showing that it largely constituted a unitary phenomenon
transcending the political fragmentation of the region after the fall of the
Ottoman Empire. In the few years following the Armistice of Mudros
(October 1918), the Ottoman Middle East was permanently partitioned into
several new states,127 which divided the Kurdish lands and deprived the
Kurdish elite of their main source of political identity, the Ottoman Muslim
citizenship. Moreover, the centralising ambitions of the new states

124
Hitchins, Kurdish Elites.
125
Hamit Bozarslan, 'Some Remarks on Kurdish Historiographical Discourse in
Turkey (1919-1980)' in Essays on the Origins of Kurdish Nationalism, ed. by Abbas
Vali (Costa Mesa, CA: Mazda Publishers, 2003), 27.
126
Janet Klein, ‘Kurdish Nationalists and Non-Nationalist Kurdists: Rethinking
Minority Nationalism and the Dissolution of the Ottoman Empire, 1908-1909’,
Nations and Nationalism, 13.1 (2007), 135–53
127
Turkey, Iraq and Syria for what the Kurds are concerned, but also Jordan,
Lebanon and Palestine.
84
threatened the local power of Kurdish tribal landowners who feared a more
invasive presence of the central authorities on their lands. The response
was a form of ‘feudal nationalism’ due to the dominant role played by the
Kurdish traditional ruling class in both contexts.

The rest of the chapter will discuss the events in Turkey and Iraq showing
how similar uprisings resulted in different outcomes loaded with
consequences for the development of Kurdish nationalism in the two
countries. Although the focus is on the developments in Turkey and Iraq,
Iran was the theatre of similar events. In the interwar period, Iran was going
through a process of modernisation partly inspired by Turkey128 and the
reaction of the Kurdish tribal elite can be read within the same framework
of feudal nationalism.129 What is particularly significant about Iran in this
period is that the experience of the short-lived Kurdish Republic of Mahabad
of 1946 can be seen as the final act of this first phase of the Kurdish
movement and as a decisive moment for construction of the transnational
dimension of Kurdish politics.

The concept of feudal nationalism was used in reference to the Kurds by

Amir Hassanpour in his book Nationalism and Language in Kurdistan (1992),
to describe the first literary expressions of Kurdish cultural distinctiveness
in the seventeenth and eighteenth centuries.130 The term is intentionally
oxymoronic, as Hassanpour meant to point out the historical contradiction
inherent in the development of national feelings among a tribal aristocracy,

128
Erik Zürcher and Touraj Atabaki’s comparative work on the projects of
modernisation in Turkey and Iran in the 1920s emphasises the learning process
between the two regimes: “The denial of a Kurdish identity after 1928 in Iran echoes
that in Turkey after 1926. The influence of the Kemalist example seems to have
grown after the shah’s 1934 state visit to Turkey.” Touraj Atabaki and Erik J.
Zürcher, Men of Order: Authoritarian Modernization under Ataturk and Reza Shah
(London and New York: I.B. Tauris, 2004), 110.
For example, the revolt led by the tribal chief Simko Shikak between 1918 and
129

1922, and again in 1926.

Amir Hassanpour, Nationalism and Language in Kurdistan, 1918-1985 (Lewiston,
130

NY: Mellen Research University Press, 1992).

85
rather than a national bourgeoisie, due to the absence of the latter.131
Despite Hassanpour’s problematic application to the context of the Kurdish
emirates, the concept to feudal nationalism greatly captures the peculiar
class nature of the Kurdish revolts in the interwar period. Rather than a
contradiction, Kurdish ‘feudal nationalism’ is a convincing refutation of the
standard account of nationalism as the historical expression of the
nineteenth-century European bourgeoisie. As argued in Chapter 2, if
nationalism is freed from a teleological and diffusionist perspective and
framed as a struggle for state power embedded in class conflicts, the term
feudal nationalism assumes a much more consistent meaning. Framed in
those terms, the nationalism of the traditional landowning classes is a fairly
common phenomenon in rapidly modernising societies characterised by a
degree of ethnic differentiation. Modernising states, with their bureaucracy,
aim at establishing a direct presence in peripheral areas which was
previously mediated by local elites. This process implies a significant loss of
political power for the local elite and, if a cultural difference between the
centre and periphery exists, the local elite is likely to try to politicise the local
culture – promoting nationalism – to mobilise the population against the
central state.132

The term ‘feudal’, in the Kurdish context, identifies the tribal aristocracy –
aghas and shaykhs – that by the end of the nineteenth century had
transformed into a class of large landowners through the accumulation of
land that was previously the collective property of the tribes or mortmain
(waqf). Like in a pre-capitalist feudal system, the appropriation of the
agricultural surplus was pursued by the tribal elite via extra-economic
means thanks to their military prowess and their position of power within

131
Hassanpour, Nationalism and Language, 56-57. For a discussion on the use of
this term in the context of the Kurdish emirates, see Abbas Vali, 'Genealogies of
the Kurds: Constructions of Nation and National Identity in Kurdish Historical
Writing' in Essays, ed. by Vali, 87-97.
132
See, Paul Brass, Ethnicity and Nationalism, 272-275.
86
the Ottoman state.133 This transformation took place in a context in which
Christian and non-tribal Kurdish peasants were already politically subjected
to Kurdish aghas. Van Bruinessen compares the conditions of the Kurdish
peasantry to that of the serfs in medieval Europe: “Their lords considered
them as their private property, owned in the same way as their sheep and
mules”.134 A British traveller of the early nineteenth century described the
tribal elite and the peasantry as “totally distinct races” adding that

A tribesman once confessed to me that the clans conceived the

peasants to be merely created for their use; and wretched indeed is
the condition of the Koordish cultivators [that] much resembles that
of a Negro slave in the West Indies.135
The expansion of the private ownership of the land in the second half of the
nineteenth century and the land-grabbing practices of the tribal elite drove
many tribesmen into a similarly subordinate condition. As they lost their
customary rights over the land, they became sharecroppers or waged
labourers in the estates of their agha or shaykh.

Given this context, the Kurdish elite had no reason to dream of a Kurdish
nation-state. In the Ottoman Empire, a polity based on dynastic and religious
legitimacy in which the Sultan was both head of the imperial Ottoman house
and Caliph of the Muslims, the Kurdish elite never challenged its place within
an Ottoman society that was hierarchically divided along religious lines. The
members of the Kurdish elite were either of prestigious tribal lineage or the
heads of the Sufi brotherhoods, the most powerful religious organisations
in the region. When conflicts between the local elite and the central
government arose, it was nearly always to negotiate the degree of local
autonomy rather than to claim independence. Kurdish revolts in the Ottoman
Empire must be thought of within the mechanism that Şerif Mardin calls the

For a definition of feudalism in these terms, see Perry Anderson, Lineages of the
133

Absolutist State (London: Verso, 1974), 401.

134
Van Bruinessen, Agha, Shaikh and State, 50.
135
Cited in Jwaideh, The Kurdish National Movement, 27.
87
Ottoman ‘tacit contract’.136 As Hamit Bozarslan explains, although the tacit
contract did not legitimise open revolt against the Sultan, “the Ottoman
state tradition conceived of rebellion, or at least resistance, as a means of
bargaining and negotiation by the subordinate peripheral groups for
improving their status within the state.”137 The frequent tribal revolts in the
Ottoman Empire – as well as Qajar Persia – must be understood within this
framework of state-tribe relationship. However, the end of the empire left a
number of smaller states each pursuing a nation-building project led by its
Turkish, Arab, or Persian majority and each of them promoting
modernisation and centralisation policies that challenged the political power
of traditional elites. Most important of all, the new states appeared
determined to claim the monopoly of violence – thus to disarm local elites –
and the administration of justice taking away from the tribal elites their most
fundamental sources of power over their tribesmen and peasants. That was
the context in which the Kurdish elite adopted a nationalist discourse to
legitimise its revolts against the new authorities.

As mentioned earlier, the promotion of Kurdish identity had been initiated

by Istanbul-based intellectuals in close association with a project of
modernisation, albeit in a top-down and paternalistic fashion. Even though,
after the fall of the empire, these Kurdist groups largely adopted a more
defined nationalist agenda, they were incapable of starting a social
movement of their own due to their lack of a power base in the Kurdish
region. When the Kurdish tribal elite revolted against the modernisation
projects promoted by the new states, these nationalist political
organisations had to join the movements from a subordinate position and to
give up on the progressive character of their nationalism. Examples of this
phenomenon can be drawn from each of the Kurdish regions: the Azadi
Committee in Turkey that supported Shaykh Said’s rebellion (1925) but also
the Iraqi-Kurdish Hiwa (1939) and the Iranian-Kurdish Komala (1942) that

136
Şerif Mardin, “Freedom in an Ottoman Perspective,” in State, Democracy, and
the Military: Turkey in the 1980s, ed. by Metin Heper and Ahmet Evin (Berlin: De
Gruyter, 1988), 23-35.
137
Bozarslan, Some Remarks, in Essays, ed. by Vali, 185-186.
88
had to surrender the leadership to two tribal landowners, respectively Mam
Vasta Hilmi and Qazi Muhammad. Moreover, the first modern Kurdish
political parties, the Kurdistan Democratic Party of both Iran (1945) and Iraq
(1946) elected as presidents two religious leaders, Qazi Muhammad and
Mullah Mustafa Barzani. The support lent by urban intellectuals provided a
degree of national legitimacy to the feudal revolts and allowed to inscribe
these early episodes within the nascent narrative of a Kurdish national
movement. However, these revolts were overwhelmingly led and fought by
members of the tribal elite and virtually all led by shaykh, due to their
traditional inter-tribal authority.

Moreover, the Kurdist intellectuals utterly failed to serve as a link between

the feudal nationalist leaders and the popular masses. One of the other
prominent features of the phase of feudal nationalism was precisely the lack
of popular participation and the apathy, if not outright hostility, of the
Kurdish rural masses. As Van Buinessen points out:

Contemporary reports […] suggest that the subject peasantry, even

if they had vague nationalist feelings, were more strongly motivated
by resentment against their landlords. Indeed, in the later Kurdish
risings in Iraq, which were more widespread than Shayk Said's
revolt, the non-tribal peasants did not participate in any significant
scale, but they did rise against their landlords several times.138
From the perspective of the peasantry, as Bruinessen continues, these
revolts were “not directed against their own exploiters, but against a
government that promised to curtail the power of these exploiters”. 139 The
passive attitude of the peasantry is relevant as it confutes the
characterisation, by contemporary Turkish nationalists, of the Kurdish
revolts as a Turkish Vendée, in which the Kurdish masses were deceived
into rebellion by their backward shaykhs.140

Van Bruinessen, Agha, Shaikh and State, 294. A similar point is made by Olson,
138

The Emergence of Kurdish Nationalism, 98.

139
Van Bruinessen, Agha, Shaikh and State, 294.
140
Mesut Yeğen cites, for example, the 1925 tribunal that sentenced to death
Shaykh Said: “Everybody must know that as the young Republican government will
definitely not condone any cursed action like the incitement and political re action
[…]. The poor people of this region who have been exploited and oppressed under
89
The point at stake is that the Turkish effort to deny any national character
to the feudal revolts of the interwar period pushed authors sympathetic with
the Kurds to over-emphasise it. Robert Olson, in his detailed account of the
1925 Shaykh Said Revolt, goes to great lengths to show the rebels’
commitment to the Kurdish national cause. Despite his acknowledgement of
the narrow class basis of the rebellion, the sheer lack of involvement of the
urban population, the material and religious grievances driving the tribal
elite, Olson defines it as “the first large-scale nationalist rebellion of the
Kurds” and even a “proto-type of a post-World War I nationalist rebellion.”141
This discussion is meaningful only if framed within an understanding of
nationalism as a completely autonomous force and national identity as
sufficient motivation for action. On the contrary, if the feudal revolts are
placed within the material reality of interwar Kurdistan, with its social and
political conflicts, then there is no reason to doubt their national character.
Feudal nationalism is the natural outcome of the grievances and demands
of the Kurdish ruling class whose reproduction as a class was threatened by
the new states and who thus understandably turned towards separatism to
establish their own state. The next two sections show how the events of
Turkey and Iraq fit into the feudal nationalism paradigm.

Kurdish revolts in Kemalist Turkey

When, in September 1914, the Ottoman Empire joined the Great War, its
north-eastern provinces became the front-line of the war against Czarist
Russia. The Caucasus front was characterized by an increasingly sectarian
dimension of the conflict, in which the local Christian population was seen
as a ‘fifth column’ of the enemy and, episodes of ethnic cleansing multiplied.
Tribal Kurds, enrolled as irregular forces in the Hamidiye cavalry,

the domination of sheikhs and feudal landlords will be freed from your incitement
and evil, and they will follow the efficient paths of our Republic which promises
progress and prosperity.” Mesut Yeǧen, ‘Turkish Nationalism and the Kurdish
Question’, Ethnic and Racial Studies, 30.1 (2007), 128.
141
Olson, The Emergence of Kurdish Nationalism, 153-164.
90
participated in the massacre and deportation of the Armenian population
moved by the fear of future Christian domination but also, more prosaically,
to take over their properties. When, in 1919, after the surrender of the
Ottoman Empire, sectors of the army led by Mustafa Kemal (later known as
Atatürk) rose against the post-war settlement that had imposed a colonial
partition of Turkey, most of the Kurdish tribes joined Atatürk’s rebellion. The
Turkish War of Independence (1919-1923) was largely fought under the
banner of Islam, to defend the Sultan-Caliph, and to protect the Ottoman
land from the Christian powers. Kurdish tribal leaders were particularly
concerned by the expansion of the newly-established and internationally-
backed Armenian state in eastern Anatolia and feared the consequences of
their involvement in the Armenian genocide.142

These international dynamics frustrated the aspiration of part of the Kurdish

nationalist intelligentsia who had seen in the collapse of the empire a brief
window of opportunity for Kurdish statehood. Most Kurdish tribes answered
the appeal to Islamic brotherhood and sided with the Turkish forces. In
December 1919, the Kemalist army defeated the Armenians and asserted its
control over eastern Anatolia, including most of the Kurdish areas. If, during
the war of independence, the stress had been placed on Muslim
brotherhood, the more the new Ankara-based Kemalist government
consolidated its power, the more the official ideology shifted towards
Turkish nationalism, modernisation, and secularism. The victory in the
liberation war had given Mustafa Kemal a degree of power and prestige that
allowed him to move towards the implementation of a far more radical
programme. With the abolition of the sultanate (1923) and the caliphate
(1924), the source of legitimacy chosen for the new republic was the
Turkishness of its people. The promise to establish an autonomous Kurdish
region in south-eastern Anatolia, approved in February 1922 by the national

142
Hamit Bozarslan, 'Kurds and the Turkish State'. In The Cambridge History of
Turkey, Volume 4, ed. by Reşat Kasaba (Cambridge: Cambridge University Press,
2008), 335–337.
91
assembly but never implemented,143 was put aside and assimilation became
the only response to Turkey’s ethnic diversity.

The denial of Kurdish identity at the inception of the Turkish Republic can
be seen as the historical beginning of the Kurdish question in Turkey.
However, explaining the outbreak of the Kurdish revolts of the interwar
period solely as a response to the ideological shift of the Kemalist republic
towards Turkish nationalism and secularism is problematic. On the one
hand, this argument reproduces a teleological perspective that sees in the
rise of nationalism the endpoint of Ottoman history and Kurdish nationalism
as its last – and unsuccessful – manifestation. This is the case of Robert
Olson’s reading of the 1925 Shaykh Said Revolt as the culmination of a four-
stage Kurdish national awakening that started in the closing decades of the
nineteenth century.144 As pointed out before, the traditional elite of the
Kurdish provinces had not yet shown any particular manifestation of
national feelings and tended to see itself as part of the Ottoman Muslim elite.

On the other hand, this argument obscures the material dynamics that seem
to be the actual trigger of the Kurdish revolts. After the abolition of the
monarchy and the caliphate, the Kemalist project proved to be antithetical
to the traditional sources – tribal and religious – upon which the local power
of the Kurdish elite rested. Moreover, the abandonment of any project of
regional autonomy revealed the ambition of the new Turkish state to impose
its much more intrusive presence in the periphery of the country. This
process became increasingly apparent to the Kurdish elite as, with the end
of the war, Ankara started filling the administration of the predominantly-
Kurdish south-east with loyal officials coming from western Turkey.145 While
these tangible threats to their power were the prime reason for the Kurdish
revolts, the assimilationist policy of the Kemalist was almost exclusively a
concern for the small circles of urban-based Kurdists who joined the revolts

143
Heper, The State and the Kurds, 118-123.
144
Olson, The Emergence of Kurdish Nationalism, 1-25.
145
Mcdowall claims that all the senior and half of the junior administrative posts in
the Kurdish areas were filled by Turks. McDowall, A Modern History, 191.
92
in a subordinate position. The fact that the culturalist explanation had such
a long-lasting impact is likely due to a choice of historical narrative. After
the failure of the feudal revolts, the history of the genesis of the Kurdish
national movement was written by the intellectual vanguard in exile rather
than the tribesmen who actually led the revolts.146

The Kurdish revolts in interwar Turkey must be understood within the

framework of a long period of tribal agitations in the periphery of the new
republic. In the two decades following the establishment of the state,
Yadirgi counts twenty-seven episodes of Kurdish revolts while seventeen
of the eighteen military campaigns conducted by the Turkish army between
1924 and 1938 took place in Kurdish areas.147 These numbers signal that the
inquietude of the tribal elite was widespread even though only a few major
uprisings made use of Kurdish nationalist discourse.

The most significant episode that showed the potential of nationalism to

give voice to growing tribal anxiety in the Kurdish areas was the 1925
rebellion. Led by Shaykh Said, this revolt mobilised the Kurdish tribes to the
point that it constituted a real threat to the stability of the Kemalist regime.
The abolition of the caliphate and the imposition of Turkish as the sole
national language of the Republic in 1924 were seen as affronts to the
Kurdish elite. Besides their immediate religious and cultural meaning, these
initiatives were politically loaded and put in doubt the idea that all Muslim
elites would be treated equally by the new republic as they were in the old
empire. Moreover, the wave of Turkish civil servants sent to the Kurdish
region threatened the prerogative of the Kurdish elite to enforce their will
on their peasants and to administer justice among their tribesmen. In the
same years, the Kemalist government discursively showed its hostility
towards the backwardness of the tribal elite, even threatening to
expropriate and redistribute their land.148

146
Van Bruinessen, Agha, Shaikh and State, 266-267.
147
Yadirgi, The Political Economy, 168.
148
Despite the government’s threatening statements, a very limited form of land
redistribution was implemented in 1929 (Law 1505) with only negligible effects on
93
Shaykh Said of Palu, a prominent Naqshbandi master and powerful landlord,
planned and organise the revolt with the Azadi (freedom) committee. Heirs
of the late-Ottoman Kurdist clubs, the founders of Azadi had a background
similar to the previous organisations – urban and state employees – but it
was predominantly formed by action-oriented Ottoman army officers.149 In
early 1925, in a climate of rising tensions in the Kurdish provinces, Shaykh
Said declared the restoration of the caliphate and called for a member of
the Ottoman house to become the King of Kurdistan. The revolt did not last
for more than a few months and exhausted its force around the siege of
Diyarbakır which remained loyal to Ankara. Notwithstanding its limited
duration, Ankara had to organise a massive military intervention and guerrilla
activities by tribal forces continued for years. The Ararat revolt (1927-1930)
can be seen as a continuation of Shaykh Said’s, also due to a similar
interaction between nationalist and tribal elements.150 Moreover, the social
engineering project aimed at dispersing the Alevi Kurdish population of the
Dersim region through mass deportation provoked a popular revolt in 1936
that was led by Sufi leaders and answered with the mass extermination of
rebels and civilians.151

The Shaykh Said rebellion is the most significant episode of Kurdish feudal
nationalism in Turkey. The rapid adoption of nationalist jargon by the tribal
elite had been a consequence of the reforms that directly threatened its
power in the region. This discourse had been borrowed by nationalist
organisations that saw Kurdish nationalism as a progressive force for the
re-birth and modernisation of the south-east. But its employment by Shaykh
Said and the other rebel leaders reflected more closely traditional sources
of legitimacy or at least a mixture of modernity and tradition. The nationalist
nature of the proclamation of a Kingdom of Kurdistan, for example, is

the rural structure of the Kurdish region. Ugur Ümit Üngör, The Making of Modern
Turkey: Nation and State in Eastern Anatolia, 1913-1950 (Oxford: Oxford University
Press, 2011), 145.
149
Van Bruinessen, Agha, Shaikh and State, 279-281.
150
Ibid., 265.
151
Bozarslan, The Kurds and the Turkish State, 341.
94
contradicted by the invitation of an Ottoman Turkish prince to take the
throne. As Hamit Bozarslan points out, the Azadi Committee was filled with
intellectuals and former Ottoman officers of Kurdish extraction but educated
in Western-style schools and academies and often with a background in the
Young Turks’ and Kemalist movements. This urban elite “considered the
tribal chiefs and religious brotherhoods to be mutegallibiyya (usurpers) or
obstacles preventing the Kurds from accessing ‘civilisation’. They rejected
the state mainly because it was a Turkish – i.e. non-Kurdish – state”.152
However, the lack of social bases of this group forced them to lend their
national credentials to the tribal forces “which initially rejected the state not
because it was a Turkish state but simply because it imposed and militarised
borders.”153

The defeat of the Shaykh Said rebellion convinced the Kemalist elite that
the militarization of the Kurdish region was the only way to avoid the
resurgence of reactionary forces. The wave of repression, executions and
deportation employed by the Turkish state in the following two decades
transformed the relations of power in the region. The implementation of the
Kemalist reforms was even accelerated by the Kurdish revolts. For example,
in 1925, Atatürk took advantage of the Shaykh Said revolt to close down all
the Sufi brotherhoods of the country, seen as representative of a backward
and irrational form of Islam but also as powerhouses for supporters of the
previous regime.154 The repression vertically hit the whole society of the
Kurdish region and particularly curbed the power of the tribes which, by the
1940s, had lost all their capability of military mobilization. Prominent tribal
families were decimated, and the survivors had to come to terms with the
Kemalist regime. The outcome of this process was the transformation of the
Kurdish elite, a process completed with its gradual re-integration into the
political system in the 1950s. Tribal dynamics gradually gave way to a new

152
Ibid., 339-340.
153
Ibid., 339-340.
154
Zürcher, Turkey, 191.
95
form of social stratification in which the members of the elite were less and
less tribal and religious leaders and more and more absentee landlords.

Despite their national character, these revolts can still be read within the
framework of the Ottoman ‘tacit contract’ to negotiate power between the
centre and the periphery and, according to Bozarslan, that is probably the
mindset with which the majority of the participants joined the revolts. 155
However, unlike the Ottoman Empire, the new Turkish state had no intention
to negotiate sovereignty over its border areas with the local elite. The result
of this process was that the Kurdish elite had to abandon their allegiance to
both traditional forms of power and to any political form of Kurdish identity.
However, as Chapter 7 will show, they soon realised that adopting a mild
version of Kemalism – generally that of the conservative parties that
dominated Turkish politics from the 1950s – was enough to be integrated
into the national political system and to have the Turkish security apparatus
protecting their property from a more and more turbulent peasantry. In this
context, feudal nationalism disappeared from the Turkish context and
Kurdish nationalism only re-emerged in a radically different form as the
instrument of a new generation of Kurds of peasant extraction. Turkey’s
Kurdish nationalists in the 1960s and 1970s will draw upon different
ideological sources and will direct their actions against both the Turkish
state and the very same Kurdish landowning class that led the feudal revolts
in the interwar period.

Kurdish revolts in Hashemite Iraq

The situation of the Kurds who remained south of the Turkish border was
quite different. After the surrender of the Ottomans in October 1918, British
forces recognised the authority of the Kurdish tribal leaders who controlled
the mountainous eastern and northern parts of Mosul province. Since the
British were promising autonomy for the Kurdish areas, Kurdish chiefs
tended to recognise British rule and seemed to be much more worried by

155
Bozarslan, Some Remarks, in Essays, ed. by Vali, 185-186.
96
the possibility of being incorporated into the Arab state centred in Baghdad.
Their first concern was to keep state officials as far away as possible from
their estates and distant British rule could have allowed them the same
degree of autonomy they had enjoyed under the Ottoman Empire. In these
years and in those that followed the establishment of the British mandate of
Iraq in 1921 under the Arab Hashemite monarchy, the Kurdish tribes rose
numerous times against the intrusiveness of the new state and its colonial
patron and demanded the fulfilment of the early promises of local autonomy.
Britain repressed the early revolts and, once Iraq was given partial
independence in 1932, ultimately sided with Baghdad as the stability of the
Hashemite state was regarded as strategically important to its wider
colonial interests. The Kurdish revolts that took place in the interwar period
in Iraq present the traits of feudal nationalism. They were led by powerful
religious leaders and the Kurdish nationalist urban intelligentsia was absent
at least until the late 1930s. Townspeople tended to be hostile to the tribal
leaders often welcoming British and Iraqi repression while the peasantry
was indifferent to the revolt when not in revolt itself against their abusive
Kurdish landowners.

In the last months of 1918, British authorities started making deals with
Kurdish tribal leaders to stabilise the region after the withdrawal of Ottoman
forces. Among them, Shaykh Mahmud Barzanji was appointed as governor
of Sulaymaniyah, the biggest Kurdish town in Iraq. Mahmud Barzanji hailed
from a centuries-old family of Qadiri shaykhs and was a large landowner
who held great authority over the tribes of the Sulaymaniyah hinterland.156
The recognition of local authority by British officers was sought by these
tribal leaders as a form of external legitimation that did not imply any actual
control since, in this early period, the British had little or no military presence
in the region. Shaykh Mahmud from his position of governor of
Sulaymaniyah claimed to be the ruler of the entire Kurdish region, a position
that the tribal leaders who had been assigned different districts had no

Michael Gunter, Historical Dictionary of the Kurds (Lanham: Scarecrow Press,

156

2011), 25.
97
intention to acknowledge.157 Moreover, Shaykh Mahmud was extremely
unpopular among the townspeople of Sulaymaniyah who were concerned
by his authoritarian and violent methods.158

In spring 1919, the British attempts to mediate among the various actors
pushed Shaykh Mahmud to start a revolt that British troops took a few
months to quell. Despite the references made by the rebels to the Kurdish
nationalist character of the revolt, this first episode of feudal nationalism in
Iraq was extremely localised and both the people of Sulaymaniyah and many
Kurdish tribes of the area remained neutral when not openly siding with the
British. The revolt was followed by a long period of tribal disorders. Shaykh
Mahmud’s nationalist claims must be put into perspective. On the one hand,
his contacts with Mustafa Kemal, intent on fighting Turkey’s independence
war, show that he had not yet abandoned the ideal of Muslim unity. On the
other hand, Shayk Mahmud’s revolt was part of the wider tribal agitations
against the British that affected both Arab and Kurdish provinces of Iraq in
the immediate aftermath of the war and that escalated in the 1920 country-
wide revolt.

As Chapter 4 will discuss more extensively, British colonial authorities were

extremely keen to appease tribal leaders. To restore order, they recalled
Shaykh Mahmud from his exile in southern Iraq in November 1922 but, at
the beginning of the following year, the shaykh was again in revolt. Iraqi and
British forces were able to retake Sulaymaniyah only in May 1924 and
Shaykh Mahmud kept raiding the area until 1932 when he was defeated. In
the meantime, a new revolt had been initiated in 1931 by another shaykh,
Ahmed Barzani,159 who claimed his right to collect revenues among his tribe
and refused to accept an Iraqi garrison in his lands. Baghdad suppressed
the revolt only in 1933 and Ahmed Barzani was sent into exile.

157
Tahiri, The Structure of Kurdish Society, 55-56.
For Mahmud Barzanji’s rule of Sulaymaniyah and the revolts discussed below,
158

see Mcdowall, A Modern History, 151-183.

Despite the assonance, there is no relation between the Barzanji and the Barzani
159

families.
98
The feudal nationalist revolts of the 1920s and 1930s in Iraqi Kurdistan did
not have any real followers apart from a small number of tribal landowners
and their fellow tribesmen. Despite the early usage of a nationalist
vocabulary, the urban middle class of the towns kept fearing the
abusiveness of the tribal leaders and refused to bend to their rule.
Moreover, the Kurdish towns, located in the valleys, were economically
connected to the Mesopotamian plain far more than they were with each
other and only Sulaymaniyah, the biggest of them, demanded a degree of
administrative autonomy.160

It was precisely in Sulaymaniyah that a group of middle-class intellectuals

established, in 1922, the first Kurdish nationalist organisation in Iraq, the
Association for the Independence of Kurdistan.161 The association,
established by former Ottoman officer Mustafa Pasha Yamulki strongly
opposed Shaykh Mahmoud’s rule over the city due to his violent methods
and was closed down by the shaykh. Urban-based Kurdish nationalism
experienced significant growth in the 1930s. The nationalist circles in
Sulaymaniyah and Kirkuk, along with Kurdish students in Baghdad and
Mosul, found several nationalist organisations almost exclusively of
progressive, socialist, and – more and more – anti-colonial orientation.162 In
addition to the aforementioned hostility of the townspeople to the tribal and
landowning elite, this urban intelligentsia was pushed to the left by the Iraqi
Communist Party (ICP) that, established in 1934, was the only national party
officially in favour of Kurdish self-determination.163 Like in Turkey, Kurdish
nationalist organisations in Iraq initially strongly opposed the power of the
tribal landowners whom they saw as a backward force that prevented the
modernisation of the region. However, as the events unfolded in the 1940s,

160
McDowall, A Modern History, 166-167.
161
Tahiri, The Structure of Kurdish Society, 57.
162
Yaniv Voller, The Kurdish Liberation Movement in Iraq: From Insurgency to
Statehood (London and New York: Routledge, 2014), 44-45.
Johan Franzén, Red Star Over Iraq: Iraqi Communism Before Saddam (New York:
163

Columbia University Press, 2011), 57-63.

99
the lack of military prowess and manpower of the urban intelligentsia forced
them to cooperate with the tribal leaders from a subordinate position.

In 1943, Ahmed Barzani’s brother Mullah Mustafa started a new revolt. 164
Despite the small size and limited geographical reach of the revolt, the
mountainous nature of the terrain helped Barzani keep the Iraqi army at bay
for two years.165 Due to the relative success of the revolt, elements of the
emergent urban national movement became more and more aware of the
necessity of collaboration with tribal forces. Some groups started
supporting the revolt providing national legitimation to what would have
otherwise been a merely tribal uprising. Barzani was simply demanding
amnesty for those involved in the 1931 revolt and the possibility to settle
again in Barzan, his ancestral land. Kurdish historian Fared Assasard claims
that, when Sulaymaniyah-based nationalists asked him to include
administrative autonomy in his demands, he could not understand what they
meant.166 When Barzani’s forces were overpowered by the Iraqi army in late
1945, he and large part of his men were forced to cross the Iranian border.

North-western Iran had been under the occupation of the Soviet Union since
1941 and nationalist elements of both the Azeri and the Kurdish minorities
had set up their own Soviet-backed provisional governments. In December
1945, local Kurdish nationalists proclaimed a Kurdish republic in the town of
Mahabad and elected Qazi Muhammad as its president. The fact that Qazi
Muhammed was a member of a prestigious religious family167 shows that the
Kurdish movement in Iran was following similar dynamics of those in Turkey
and Iraq, where the weak urban nationalists had to compromise with the
traditional elite.168 The arrival of Mustafa Barzani from Iraq gave the

164
McDowall, A Modern History, 290-293.
165
Kerim Yıldiz, The Kurds in Iraq: The Past, Present and Future (London: Pluto
Press, 2005) 169-170.
166
Interview with Fared Assasard (Sulaymaniyah, 2018)
167
Gunter, Historical Dictionary, 169-170.
168
This dynamic of interaction is particularly interesting in the context of the
Mahabad Republic, as Kurdish forces had the chance to experiment with a degree
of self-government. According to Hamit Bozarslan, the Mahabad experiment
resulted in a mix of modernising reforms and tribal and religious politics. Hamit
100
Mahabad Republic a pan-Kurdish dimension and strengthened its nationalist
credentials. In summer 1946, Qazi Muhammed promoted the establishment
of a unified political party, the Kurdistan Democratic Party (KDPI) and was
elected as its president. This example was followed by Mullah Mustafa who
sent a letter to most of the Kurdish organisations in Iraqi Kurdistan asking to
unite in a single Kurdish national party.169 Despite the widespread hostility
towards the tribal leadership, most Iraqi-Kurdish nationalists followed
Barzani’s request and founded the Iraqi Kurdistan Democratic Party (KDP).170
The congress also elected Mullah Mustafa as the president of the party. The
Mahabad Republic did not survive the year.171 As the Red Army withdrew,
Iranian troops entered Mahabad in December 1946 and sent Qazi
Muhammed to the gallows. Mullah Mustafa was able to flee and, with a
handful of his men, found shelter beyond the Soviet border. Ibrahim Ahmed
– who was to become one of the most prominent Kurdish nationalists in Iraq
– claimed that Mahabad taught them to “never again let tribal leaders lead
their national liberation revolution and to take a new enlightened true
liberation course that will match the changing world condition and
progress.”172

The relations of power on the ground had forced Kurdish urban nationalists
in Iraq to accept the alliance with the tribal elite and to surrender the
leadership of the emerging national movement. However, the experience of
Mahabad marked also the beginning of complex and troublesome relations
between urban and tribal Kurdish nationalists in Iraq. Tensions between

Bozarslan, '"Being in Time": The Kurdish Movement and Universal Quests' in The
Kurdish Question Revisited, ed. by Stansfield and Shareef, 67. For a similar reading
of the relationship between Kurdish nationalism and traditional social structure in
Iran, see Farideh Koohi-Kamali, The Political Development of the Kurds in Iran:
Pastoral Nationalism (London: Palgrave Macmillan, 2003), 165-196.
169
Voller, The Kurdish Liberation Movement, 46.
170
Tahiri, The Structure of Kurdish Society, 105-106.
For a detailed, albeit dated, account of the Mahabad Republic, see: William
171

Eagleton, The Kurdish Reoublic of 1946 (London: Oxford University Press, 1963).
172
Ibrahim Ahmed, ‘The Republic of Kurdistan: A Personal Memoir’, The
International Journal of Kurdish Studies, 11.1/2 (1997), 31–32.
101
these two components and their conflicting class bases were to become a
constant feature of the Kurdish national movement in Iraq. As Chapter 4 will
show, the continuous conflicts within the KDP will finally lead to the split of
the urban and leftist component that, in 1975, established the Patriotic
Union of Kurdistan (PUK).

Conclusion

After a brief introduction to the social and political context of the Kurdish
region between the nineteenth and twentieth centuries, this chapter
discussed the emergence of the Kurdish national movement in the interwar
period. The Kurdish revolts that took place in Turkey and Iraq after World
War I had some common characteristics that allow us to group them under
the label of feudal nationalism. In these countries, the Kurdish traditional
elite revolted against the centralisation project promoted by the new nation-
states that succeeded the Ottoman Empire. These revolts had a
predominantly tribal character and Kurdish urban nationalists, unable to
take up arms against the state on their own, joined in a subordinate position
despite their ideological opposition to the conservative tribal elite. The form
of nationalism that this process generated was filled with the particular
demands of the Kurdish tribal elite which emptied it of its initial progressive
character. As the chapter argued, denying the national character of these
revolts, due to the prominent role played by the tribal aristocracy,
reproduces a narrow understanding of nationalism as a bourgeois and
European phenomenon. On the contrary, reducing the Kurdish revolts to a
selfless expression of nationalist feelings triggered by the denial of Kurdish
identity erases their conflictual nature and fails to explain the lack of
participation of the Kurdish masses. This chapter thus deployed the concept
of feudal nationalism to explain the undisputable national character of these
revolts – as their goal was the establishment of a Kurdish state – but also
their extremely narrow social basis. Rather than a contradiction of history,
feudal nationalism was the consistent response of the Kurdish landowning
class to the modernisation process of the new states that, by challenging
102
their power over the Kurdish periphery, threatened their reproduction as a
class.

The era of feudal nationalism shaped the subsequent developments of the

Kurdish national movement in two important ways. First, the creation of a
Kurdish transnational political space. Each revolt, even if not directly linked
to the others, was part of a single phenomenon that presented similar
characteristics in all the Kurdish regions. The Kurdish leaders of these years
took inspiration from the Kurdish revolts in the neighbouring countries and
many Kurds started looking at Kurdistan as a single trans-border political
space. Cooperation among the Kurdish rebels had its highest moment in
1946, when Mustafa Barzani and his tribesmen from Iraqi Kurdistan
participated in the short-lived Kurdish Republic of Mahabad in Iran.
Mahabad acquired a central role in the narrative of Kurdish nationalism. The
Kurdistan Democratic Party founded in Mahabad established a sister party
in each of the Kurdish regions and, despite never achieving operative unity,
became the first Kurdish transnational party.173

The second long-term consequence of the feudal revolts was the

diversification of the Kurdish movement. This chapter sets the stage for the
development of the different trajectories of Kurdish nationalism in Turkey
and Iraq as rooted in the different outcomes of the initial and unitary ‘feudal
phase’. The Kurdish feudal elite that led the revolts of the interwar period
underwent different processes of incorporation into the new states of
Turkey and Iraq. As Chapter 4 shows, the Kurdish revolts of the 1920s and
1930s in Iraq must be placed within the broader context of the tribal
agitations that took place throughout the country. The response of the
British colonial authority was to appease the tribal elite by legally legitimising
their land-grabbing tendencies and turning them into a powerful landowning
class in support of the Hashemite monarchy. The empowerment of the Iraqi-
Kurdish traditional elite made it the best candidate to raise the Kurdish
national flag when – with the end of the monarchy in 1958 – the revolutionary

For the establishment of the Turkish and Syrian KDPs, see, respectively,
173

Chapters 7 and 9.
103
Iraqi regime directly threatened their power over the land. Chapter 7 shows
how, in Turkey, the crushing defeat of the Kurdish revolts of the interwar
period forced the Kurdish elite to accept a position of subordination into the
new Turkish state and to give up the political meaning of their Kurdish
identity. Given this context, Kurdish nationalists in the 1970s turned to the
peasantry to challenge both the Turkish state and the Kurdish tribal
landowners. In these terms, the present chapter showed the origins of the
divergence for the unfolding of two – opposite in their class dimension –
trajectories of Kurdish nationalism.

104
105
 Chapter 4
Land Reform and Kurdish Revolt in
Postcolonial Iraq (1946-1991)

Introduction

This chapter covers the history of Kurdish nationalism in Iraq from the
development of the anti-colonial movement in the 1940s to the
establishment of Kurdish self-rule in 1991. After the failure of the feudal
revolts of the interwar period, Kurdish nationalism became temporarily a
marginal force in Iraq as the Kurdish tribal and landowning elite was
integrated into the country’s ruling class. This process was actively
promoted by Britain, Iraq’s colonial overlord, according to the project of
creating a class of large landowners in support of the Hashemite monarchy.
As the chapter shows, the history of Iraq and its Kurdish provinces in the
1940s and 1950s can be told with virtually no reference to Kurdish
nationalism. In these decades, politics in the Kurdish region was, as in the
rest of Iraq, dominated by the growing movement against the monarchy and
its colonial protector as well as by the anti-landlord mobilisation of the
impoverished peasantry. The resurgence of Kurdish nationalism in the early
1960s must thus be placed within the history and political economy of
postcolonial Iraq and the restructuring of its class structure under British
rule that tremendously empowered tribal landowners. Kurdish nationalism
only emerged as an important political force after the Iraqi Revolution of
1958 that brought an end to the Iraqi monarchy and to British influence
threatening the interests of the Kurdish elite. The Kurdish landowners
revolted in response to the project of land redistribution and taxation
promoted by the post-revolutionary Iraqi government, giving the Kurdish
revolt that started in 1961 and lasted until 1975 a strong class character.

106
The 1961 uprising was triggered by the spontaneous revolt of tribal
landowners against the empowerment of the peasantry and the
implementation of the land reform, rather than by Kurdish nationalist forces
who only reluctantly lent their support. Mullah Mustafa Barzani, the leader
of the Kurdish revolts of the 1930s and early 1940s, became the natural
point of reference of the uprising thanks to his status in Kurdish tribal
society and the justified perception that his nationalist project was
compatible with the class interest of the revolting aghas. The urban-based
and leftist nationalists leading the KDP, unable to link their national claims
to the wider struggle for democracy in Iraq, were forced – like in the interwar
period – to follow Barzani from a subordinate position. The KDP provided
nationalist legitimacy to the tribal revolt of the 1960s and put aside its
demands for social transformation. The situation only changed in the
second half of the 1970s, when the defeat of Barzani’s revolt and the growth
of the Kurdish left – fuelled by a tumultuous process of urbanisation –
allowed for the establishment of the PUK, a Kurdish nationalist force
alternative to Barzani’s KDP. This chapter explains the social origins of the
Kurdish revolt of 1961-1975 in the set of class relations imposed on Iraq by
colonial rule. The relations of power between the peasantry, the urban
middle classes, and the tribal landowners shaped the development of
Kurdish nationalism in Iraq and imposed the conditions for long-term
dynamics such as the division between KDP and PUK. Moreover, the events
of the 1960s were central to the development of Kurdish nationalism in Iraq
as they determined a set of power relations between the different class
actors that gave the Kurdish national movement in the country a
predominantly conservative character.

Class and Politics in Colonial Iraq

The participation of Iraqi-Kurdish leader Mullah Mustafa Barzani in the

Kurdish Republic of Mahabad in 1946, was a critical moment for the history
of Kurdish nationalism in Iraq. The establishment of the KDP in the same
year constituted the culmination of the Kurdish revolts that followed the
107
creation of the Iraqi state in the 1920s and 1930s. This momentous year
marks, however, also the end of the first phase of Kurdish nationalism in
Iraq. As Mullah Mustafa took refuge in the Soviet Union, the KDP went
underground and remained marginal until its re-emergence after the Iraqi
revolution of 1958 when Brigadier ‘Abd al-Karim Qasim overthrew the
Hashemite monarchy.

The forms in which the Kurdish movement re-emerged after the 1958
revolution were shaped by the structural transformation that Iraq underwent
in its first decades of existence. Largely the ‘invention’ of British colonial
administrators, the mandate-state of Iraq was created in 1920 by the
merging of the Ottoman vilayets of Mosul, Baghdad, and Basra according to
the broader geopolitical interests of the British Empire. The new state
included an extremely diverse population, with a clear Arab Shia majority
and two significant Arab and Kurdish Sunni components as well as
numerous smaller groups including Sunni and Shi’a Turkmen, Christians of
different denominations, Kurdish-speaking Yazidis, and a sizeable Jewish
community that largely fled to Israel in the 1950s. The dominant role that the
Sunni minority had played in Ottoman times was preserved by the British
who imposed on the country a ‘foreign’ Sunni king, Faisal of the Hashemite
family from Hejaz that had supported them during World War I. Apart from
drawing borders in disregard of ethnic and religious identities, the projects
of social and institutional engineering promoted by the colonial overlord
shaped the history of the country for decades to come. The relations
between Britain and Iraq were regulated by a series of Anglo-Iraqi treaties
the first of which, in 1922, established the mandate state of Iraq. The 1930
treaty – revised and renewed in 1948 – recognised Iraq’s nominal
independence – proclaimed in 1932 – though ensuring wide British influence
in the form of military presence and control over the country’s foreign policy
and oil resources. Until the new revolutionary government in 1958
repudiated the Anglo-Iraqi Treaty, the relationship between the two
countries was, in all but name, that of semi-colonial rule.

108
Britain’s approach to Iraq was largely shaped by two contradictory
objectives.174 On the one hand, the colonial power had to make sure that the
new institutional arrangement would not threaten British monopoly over the
oil of Iraq and would preserve the colony’s geopolitical function of protecting
wider British interests in the Persian Gulf. On the other hand, to reduce its
own financial burden, Britain needed a viable Iraqi government seen by its
people as legitimate enough to maintain public order. This contradiction was
ultimately resolved in favour of Britain’s colonial interests and at the expense
of the King’s legitimacy and popularity and the early Iraqi state was thus
based on a “duality of power”, as “cabinet ministers, as well as provincial
governors, district executives, and city mayors, were assigned British
‘advisors’ whose views were expected to be taken “into careful
consideration” and who deeply shaped the early stages of state-building.175

British colonial rule had long-term consequences on the process of class

stratification of Iraq, particularly through its tribal policy. The social and
economic history of this period has been masterfully captured by Hanna
Batatu’s monumental work on the social structure of monarchical Iraq. In his
1978 The Old Social Classes and the Revolutionary Movement of Iraq, Batatu
provides a wealth of data on the structural transformation of the first four
decades of Iraqi history.176 He describes in length the process through which
British administrators empowered tribal leaders177 - both in Arab Iraq and in
the Kurdish areas – in order to create a ruling class of conservative
landowners that would be inherently pro-British. Already in 1916, in the

174
Adeed Dawisha, Iraq: A Political History from Independence to Occupation
(Princeton: Princeton University Press, 2009), 8-14.
175
Ibid., 18.
Hanna Batatu, The Old Social Classes and the Revolutionary Movements of Iraq:
176

A Study of Iraq’s Old Landed and Commercial Classes and of Its Communists,
Baʻthist, and Free Officers (Princeton: Princeton University Press, 1978).
177
When Batatu speaks of shaykhs, he broadly refers to the traditional elite of Iraq’s
tribal areas including large parts of the Shi’a south, of the Sunni west and of Kurdish
north. As Chapter 3 shows, the tribal elite of the Kurdish region was composed by
the shaykhs, heads of the sufi brotherhoods, and the aghas, the leaders of the
tribes. In the Arab context, however, shaykh is used to address both tribal
chieftains and religious authorities.
109
midst of the war, the British instituted the Tribal Criminal and Civil Disputes
Regulation – confirmed by royal decree in 1924 – that made the tribal chiefs
responsible for administrating justice among their tribesmen, giving them
tremendous power and little accountability in the process. The gradual
process of de-tribalisation which Iraq had been undergoing since the end of
the nineteenth century due to its gradual integration into the global markets
was suddenly reversed, “the progress of villages toward independence from
surrounding tribes forbidden, and the escape of peasant tribesmen from the
shaykhs’ lands prevented.”178

The process of privatisation of communal lands initiated by the Ottoman

Land Code (1858) was greatly accelerated by the Land Settlement Laws of
1932 and 1938, “which facilitated the transfer into [the shaykhs’] hands of
vast expanses of state and customary tribal land.”179 This process took place
evenly in both Arab and Kurdish Iraq: in 1958, five of the twenty largest
landowners of the country were Kurdish (or Arabized Kurds), including the
single largest one (Table 2). With the intent of creating a solid class of
landowners, the British rigidly classified tribes and subtribes and coercively
imposed the shaykhs over their fellow tribesmen, through a process of
social engineering that “decisively transformed the shaykh’s place in Iraqi
society and the character of his political role.”180

178
Batatu, The Old Social Classes, 94-95.
179
Ibid., 46-47.
180
Relying on archival material, Toby Dodge shows that the colonial officers were
driven by the romantic and Orientalist view of a “pre-modern and rural” Iraq
“untainted by the negative and destabilizing effects of capitalism” in which “the
Shaikh and his tribe were therefore ‘naturally’ the dominant institutions through
which British policy aims were to be realized.” Toby Dodge, Inventing Iraq: The
Failure of Nation Building and a History Denied (New York: Columbia University
Press, 2003), 83-84.
110
The new power of the tribal elite was sealed in the political institutions of
the new state. When Iraq had sent its representative to the Ottoman
parliament in 1914, none of them was a tribal leader. By contrast, among the
99 members of the Iraqi Constitutional Assembly elected in 1924, “no fewer

111
than 34 were shaykhs and aghas.”181 Before the inaugural session of the
assembly, these tribal chiefs took a public oath “to support the [Anglo-Iraqi]
Treaty and not to take any action without common consent,” to expand the
Tribal Disputes Regulation, and to prevent the government from alienating
the land they had recently appropriated.182 The oath showed that, at the
very moment of the establishment of the Iraqi state, this group already
demonstrated a significant degree of class consciousness that transcended
ethnic and sectarian divides as well as the awareness that their class
interests were best served by British colonial rule. As Batatu points out, the
decaying power of the shaykh was resuscitated by colonialism as “life was
pumped into it artificially by an outside force that had an interest in its
perpetuation.”183 The consolidation of the shaykhly class as the dominant
economic and political group made direct colonial rule redundant: Britain
recognised Iraq’s nominal independence in 1932 while maintaining control
of its oil and foreign policy.

The concentration of a large part of the county’s arable lands in the hands
of a few – largely tribal – families who were also the holders of political power
prevented economic development and fuelled social conflicts. Even if
agriculture constituted the largest sector of the economy, landowners were
barely taxed and the state’s revenues had to rely on unequal indirect
taxation. Moreover, as long as the landlords could use their political power
to expand their estates at the expenses of small farmers and uncultivated
lands, they had no incentive to invest capital in the modernisation of
agriculture. The system was therefore characterised by a very low level of
productivity and accumulation was sustained through land-grabbing –
favoured by the collusion of the state apparatus – and the increasing
exploitation of the peasantry.184 Low productivity and exploitation drove the

181
Batatu, The Old Social Classes, 95.
182
Cited in Ibid., 95.
183
Ibid., 99.
184
Samira Haj, The Making of Iraq, 1900-1963: Capital, Power, and Ideology
(Albany: State University of New York Press, 1997), 36-38.
112
peasants towards forms of passive and active resistance. Land desertion
was the most common reaction: As hungry peasants fled the countryside,
the urban population of Iraq increased from 30 per cent in the 1930s to 42
per cent in 1958 and Baghdad’s population doubled. However, one of the
most important characteristics of the two decades preceding the 1958
revolution was the spread of rural conflicts. Batatu points out that the
enrichment of the shaykhs at the expenses of their fellow tribesmen
undermined tribal loyalties which made the shaykhs “simultaneously rising
as a class and decaying as a traditional status group.”185 From 1947 to 1958,
nine major peasant revolts broke out in Iraq, three of which in the Kurdish
region.186 The first of these revolts, in 1947, took place in the countryside of
Sulaymaniyah and was directed against Shaykh Latif, the largest landowner
of the area and son of Mahmud Barzanji who had led the allegedly Kurdish
nationalist revolts of the 1920s.

This overview of the structural transformations of Iraq between 1920 and

1958 shows that the early history of the country can be told with no mention
of the Kurdish movement. The tribal revolts that took place in the Kurdish
province of Sulaymaniyah in the early 1920s and that were led by Shaykh
Mahmud Barzanji, when placed within the history of Iraq, do not look much
different than the tribal revolts that took place in the rest of the country in
the same period. In the immediate aftermath of the First World War, a time
of wide geopolitical transformation and redefinition of borders and
hierarchies, tribal shaykhs and aghas were in the position to mobilise sizable
military forces and to renegotiate their power vis-à-vis the new imperial
rulers. The nation-wide tribal agitations of 1920 – Including Shaykh
Mahmoud’s Kurdish revolt in Sulaymaniyah – was a decisive moment that, in
the words of Samira Haj, convinced “colonial officers [to take] systematic

185
Batatu, The Old Social Classes, 6.
186
Ibid., 467-468.
113
measures to legitimize the power of the shaykhly class and their claims to
the land.”187 As Hanna Batatu’s central argument goes:

The tribal rebellions of the first decades of the monarchy […] appear
in retrospect as the gasps of a tribal world approaching its end. The
rural rebellions of the last decade of the monarchy were of an
entirely different character. They were rebellions not under shaikhs
but against them, and were made by tribesmen whose customary
ideas and norms of life had been shaken to their foundation. […] The
old, patriarchal, life-furthering relationship which once tied the
tribesmen to their shaikh had given way to an overlord-quasi-serf
relationship which chained them to distress and privation, and the
idea now sank into them that this was not an unalterable state of
things. The idea was, of course, spread by Communists.188
This social arrangement was bound to generate the opposition of the
exploited peasantry. The next section shows how the mobilisation of the
peasants gave the chance to other class actors such as the industrial
workers, the urban middle classes and the – thin yet existent – non-tribal
bourgeoisie to express their grievances against a state organised around
the interests of the tribal landowning class and of the colonial power. Even
in the context of growing opposition and unrest of the 1950s, Kurdish
nationalism maintained a marginal role and social and political struggle in
the Kurdish region was characterised by anti-landlord and anti-colonial
mobilisation just as in the rest of Iraq.

The Iraqi Revolution of 1958

Rising tensions in rural areas, as well as the increasing presence of

impoverished peasants in the largest cities, constituted fertile ground for
the growth of the Iraqi Communist Party (ICP) since its foundation in 1934
and throughout the 1930s and 1940s. The communists’ demand for land

187
Haj, The Making of Modern Iraq, 30. As already mentioned in Chapter 3, the
Kurdish revolts led by Mustafa Barzani in 1931-32 and 1943 had a much more
idiosyncratic character and saw the participation of pretty much his own tribe only.
Despite their importance for the developments of the Kurdish movement, these
‘later’ tribal revolts were not significant to the history of Iraq.
188
Batatu, The Old Social Classes, 46.
114
reform resonated among the landless peasants and the party built a strong
base in the Kurdish north thanks also to its official commitment toward
Kurdish self-determination. Jalal Jawhar remembers the political
atmosphere in the Kurdish village where he grew up:

when communism spread in the 1930s and 40s, there was a lot of
support in Iraq and Kurdistan for the party. They were against the
aghas and tribal leaders whom they accused of collaborating with
the imperialists. They were telling the people that [the ICP] would
bring back all the land that the aghas and the tribal leaders had taken
from them.189
The ICP grew even stronger among the workers of the small industrial sector
and, in the 1940s, came to dominate a rapidly growing labour union
movement.190 Despite the fierce repression faced by the labour movement,
the clandestine work of the ICP to organise and mobilise the urban and rural
masses provided the opposition to the monarchy with a significant degree
of popular support. The other two major components of the opposition were
the National Democratic Party (NDP), expression of the liberal-minded and
non-tribal bourgeoisie, and the Ba’ath Party, on pseudo-socialist and pan-
Arabist positions, strong among the urban middle class.191

The reluctance of the other opposition parties to co-operate with the

communists was only slowly overcome thanks to the national uprisings of
1948, 1952, and 1956. During the 1940s, all opposition forces came to blame
Iraq’s underdevelopment on the combined power of the dominant shaykhly
class and British imperialism sealed in the Hashemite monarchy. The first
uprising took place on the occasion of the renewal of the Anglo-Iraqi Treaty
in 1948, that confirmed all the privileges previously awarded to Britain. The
second uprising, in 1952, was partly inspired by the nationalisation of the oil

189
Interview with Jalal Jawhar (Sulaymaniyah, 2019).
In the early 1950s, there were 12,000 industrial workers in the country. The
190

Economic Development of Iraq (International Bank for Reconstruction and

Development, 1952), 133. In the same years, twelve of Iraq’s sixteen legal labour
unions were led by the ICP. Joel Beinin, Workers and Peasants in the Modern
Middle East (Cambridge: Cambridge University Press, 2001), 128.
For an analysis of the of the social bases of Iraq’s main opposition party in the
191

monarchical period, see: Haj, The Making of Modern Iraq, 85-98.

115
industry in Iran (March 1951) and the Egyptian Revolution (July 1952) that
exposed the weakening of Britain’s position in the Middle East. The third
uprising, in 1956, broke out in response to the Baghdad Pact (1955) – that
locked Iraq in an anti-communist regional alliance sponsored by Britain – and
in solidarity with Egpyt, under attack during the Suez Crisis. All of these
revolts were preceded by waves of rural uprisings and workers’ strikes, and
the social and anti-colonial characters of the opposition to the monarchy
gradually welded together. After the 1956 uprising, the opposition forces
formed the United National Front – this time with the inclusion of the ICP –
that prepared the ground for the revolution of 1958.192

The coup d’état that overthrew King Faisal in July 1958 was led by Brigadier
‘Abd al-Karim Qasim and supported by a heterogenous republican front that
included all the major parties. Qasim’s first major challenge came from within
this coalition. The pan-Arabist fringes of the NDP and the Ba’athists pushed
for Iraq to merge into the United Arab Republic, the federation between
Egypt and Syria established in February 1958 and led by Egyptian President
Gamal Abdel Nasser. Qasim feared that a union with Egypt would have
turned him into Nasser’s lieutenant while the National Democrats felt
anxious about the competition of Egypt’s relatively more advanced industry
and stronger financial sector. Qasim relied heavily on the ICP to garner
enough popular support to resist the pan-Arabist sympathies of the urban
masses. The alliance with the communists helped Qasim crush his pan-
Arabist enemies but turned the ICP into the most powerful political
organisation of the country. Allowed to operate legally and often supported
by the government, the ICP experienced, in the first year of the republic, the
apex of its strength and popularity.193

192
Decisive to the success of the revolution was the fact that the Iraqi Army was in
great expansion and, from the 1950s, started to be packed with officers
sympathetic to the nationalists, the Ba’athists and, to a lesser degree, the
communists. See, Marion Farouk-Sluglett and Peter Sluglett, 'The Social Classes
and the Origins of the Revolution', The Iraqi Revolution of 1958: The Old Social
Classes Revisited, ed. by Robert A. Fernea and William Roger Louis (London: I.B.
Tauris, 1991), 130-131.
193
Dawisha, Iraq, 176-179.
116
The growth of the ICP was met with apprehension by National Democrats in
power. Tensions came to the surface when, after the purge of the Ba’athists
in late 1958, the communists resumed the issue of land redistribution, one
of the crucial demands of the revolution against the monarchy. In
September 1958 the government had passed a limited land reform.194 The
law compensated large landowners for their losses and allowed the
peasants to buy the confiscated land. However, only rich farmers had the
means to benefit from the reform whereas most of the landless peasants
were unable to access the credit necessary to buy the land.195 The land
reform was supposed to take place in the span of five years, but its
implementation proved to be very slow. By the end of 1958, the communists
were pushing for more radical land redistribution and started mobilising the
peasantry to occupy large estates. Qasim was, however, determined to
resist the demands of the communists. In summer 1959 the government
started to systematically dismantle popular organisations such as trade
unions and peasant societies and to arrest communist cadres
indiscriminately. In September 1959, Qasim reinstated Martial Law and the
ICP received an organisational blow from which the Iraqi communist
movement would never completely recover.

Land reform and Kurdish Revolt

When the KDP resumed its legal activities after the revolution, the ICP was
the leading political actor in the Kurdish region. In the preceding decades,
the political trend of Iraqi Kurdistan had been consistent to that of the whole
of the country: The Kurdish tribal elite took advantage of their position to
appropriate communal land and tied their destiny to that of the Arab
landowning class and the monarchy. The communists were making gains

194
The land reform imposed a limit to personal land ownership at 1000 dunums
(250 hectares) for the irrigated lands and 2000 dunums for the less productive
rainfed areas. Batatu, The Old Social Classes, 836-837. The latter largely include
the Kurdish valleys.
195
Haj, The Making of Modern Iraq, 120-121.
117
among the peasantry as the land-grabbing and exploitative practices of the
Kurdish elite loosened tribal and religious loyalties. Kurdish nationalism and
the KDP survived only thanks to the support of non-tribal middle-class
Kurds living in the towns – Sulaymaniyah, Kirkuk, but also Baghdad – playing
a role that in terms of class representation was similar to that played by the
Arab nationalist Ba’ath Party in the rest of Iraq. However, their influence was
limited by the small size and little political significance of the Kurdish towns.
Moreover, the Kurdish nationalists were facing the competition of the ICP
that, unlike the other Iraqi political parties, acknowledged the ethnic
specificity of the Kurdish region. Between 1946 and 1958, when Mullah
Mustafa Barzani and his tribesmen were in exile in the Soviet Union, the KDP
was led by Ibrahim Ahmed, a lawyer from Sulaymaniyah, and his left-leaning
supporters. As most of these Kurdish nationalists were Marxist and opposed
tribalism and colonialism, their political discourse often overlapped with that
of the ICP.196

Following the fall of the monarchy, the KDP was allowed to resume its
activities and, in October 1958, Mullah Mustafa returned from exile. At the
time, the Kurds were natural allies of Qasim’s attempt to avoid the merger
of Iraq with the United Arab Republic that would have turned them into an
insignificant minority within a larger Arab polity. The KDP particularly
benefitted from its close relationship with the communist party, at the time
the key ally of the government and in November 1958 KDP and ICP signed
a Covenant of Cooperation.197 In this period, Mullah Mustafa’s role as
chairman of the KDP was largely symbolic and reflective of his prestige
among the wider Kurdish population rather than of his influence over the
party itself that was run by secretary-general Ibrahim Ahmed. Returning
from a 12-year-long exile, Mullah Mustafa was determined to maintain a low
political profile vis-à-vis the government and to regain his place within

196
Jwaideh, The Kurdish National Movement, 267-272.
197
Franzén, Red Star Over Iraq, 116.
118
Kurdish tribal society. He toured the region, meeting aghas and shaykhs,
building alliances but also reactivating dormant tribal rivalries.

The fall of the King and the advent of a new revolutionary government had
made the traditional elite countrywide increasingly anxious. A monarchy that
recognised and legitimised traditional forms of power had been replaced by
a government that spoke a language of modernisation and that painted
tribal leaders and shaykhs as forces of the past. Moreover – and more
importantly – the land reform represented a direct threat to their position
within the class structure of the country and the alliance of Qasim with the
communists indicated that the state was no longer going to protect them
from revolting peasants. In the Kurdish region, the inquietude of the tribal
chiefs did not only have the social dimension of a landowning class scared
by the rise of the communists and their cooperation with the KDP. The return
of Mullah Mustafa and his good relationship with Baghdad had also a
disrupting effect on inter-tribal relations, making the tribes historically
hostile to the Barzanis and their associates increasingly nervous. It is in this
context of a rapid deterioration of the established relations of power that
the tribal agitation of the post-revolutionary periods must be understood. In
the spring of 1959, the Kurdish tribes of the Baradost area took up arms
against the government. The suppression of the Baradost uprising was a
joint effort by the Iraqi Army, the peasant societies backed by the
communists and the KDP, and Barzani’s tribesmen.198 For Kurdish historian
Wadie Jwaideh, the revolt of the Kurdish Baradosts was, like that of the Arab
Shammar in Mosul just a couple of months earlier, a “belated reaction of the
conservative and feudal elements against the July revolution.”199

In the course of 1959, however, alliances began to shift again. As Qasim

gradually turned against the ICP, Barzani pressured the KDP leadership to
break with the communists and, in November 1958, the KDP suspended its

198
Jwaideh, The Kurdish National Movement, 283-284. See also Avshalom H.
Rubin, ‘Abd Al-Karim Qasim and the Kurds of Iraq: Centralization, Resistance and
Revolt, 1958-63’, Middle Eastern Studies, 43.3 (2007), 364.
199
Jwaideh, The Kurdish National Movement, 284.
119
cooperation with the ICP.200 Several decades of work of the left-leaning
intellectuals based in Sulaymaniyah to link the Kurdish question to the
broader struggle of the subaltern classes were swept away as Ibrahim
Ahmed and his followers were marginalised. As the Kurdish movement re-
emerged as a political force in post-revolutionary Iraq, the relations of power
within it proved to be the same as in the interwar period. The urban and
progressive elements had to capitulate once again to the preponderant
power of their tribal allies. Barzani had cared little about the KDP since his
return, he had independently pursued his tribal diplomacy and developed a
direct relationship with Baghdad. But when the policies and alliances of the
KDP did not match his own, he imposed his view on the party. For the
moment, Ibrahim Ahmed remained secretary-general of the party but the
centre of power of the Kurdish movement was no longer the politburo of the
KDP but Mullah Mustafa and his men.

By the end of 1959, Qasim had eliminated his pan-Arabist rivals, significantly
curbed the power of the communists, and was growing increasingly wary of
the dominant position built up by Mullah Mustafa Barzani in the Kurdish
region. The tribes hostile to the Barzanis were aware that Qasim’s concerns
matched their own and, in November, the chiefs of the Zebari, Surchi, and
Raikani tribes turned to Baghdad for protection.201 Clashes between
Barzanis and Zebaris resumed in spring 1960 – the climate of the Kurdish
valleys discouraged wintertime warfare – and, this time, Baghdad started to
supply the latter with weapons. Throughout 1960, Qasim and Barzani
deployed their tribal diplomacy to gain the support of the Kurdish tribes,
including the ones that they had fought against just one year before.202 This

200
Johan Franzén, ‘From ally to foe: The Iraqi communist party and the Kurdish
question, 1958-1975’, British Journal of Middle Eastern Studies, 38.2 (2011), 171-
172. Edgar O’Ballance, The Kurdish Revolt: 1961-1970 (London: Faber and Faber,
1973), 68-69.
201
Rubin, Abd al-Karim Qasim, 365.
202
Avshalom Rubin describes the dynamics of this tribal diplomacy: “In the spring,
Barzani toured northern Iraq, meeting with leaders of the Harki, Baradost, and
Pizhdar tribes. […] Qasim received a delegation of the Jaf tribe in March, Harki and
Surchi delegations in May, and Baradost and Khoshnaw delegations in August.
120
conflict by proxy had nothing to do with Kurdish nationalism and followed
the traditional pattern of state-tribe relations: As a tribal chief became too
powerful, the central government armed the rival tribes to restore the
balance of power in the region and to avoid the creation of alternative
centres of power in the country’s periphery.

In 1961, since tribal warfare was resumed after the winter pause, Barzani’s
forces made significant progress against their foes and, in summer, the
Zebaris were increasingly isolated as their tribal allies, one by one, left the
anti-Barzani coalition. Qasim was still reluctant to confront Barzani directly
in the hope that they could eventually reach an agreement. 203 However,
different dynamics were to precipitate the situation. Since 1960, the
government had started implementing the Land Reform in the Kurdish areas
and, in the spring of 1961, had imposed a land tax.204 In June 1961, a
delegation of tribal chiefs from the southern part of the Kurdish region
travelled to Baghdad to petition Qasim to abolish the and amend the reform.
As Qasim refused to meet them, the tribal chiefs returned to the Kurdish
region committed not “to pay the tax or to allow the implementation of the
agrarian reform.”205 The rebellion spread rapidly, easily gaining the support
of the Kurdish landowning class. In the course of the summer, the rebels
were increasingly in contact with Barzani who was fighting his tribal enemies
further north. In September, the revolt escalated when tribal forces in the
rebel areas attacked a column of Iraqi troops.206 Although the revolting
landowners naturally looked to Barzani for help, Mullah Mustafa was still
determined to avoid open warfare with Baghdad and tried to leave with his
men for Syria. However, the situation was decided by Qasim bombing

When the Barzanis appealed to the Baradost for aid that autumn, Qasim gave
Shaykh Rashid 60,000 dinars to turn them down.” Ibid., 365-366.
203
Ibid., 366-368.
204
Rony Gabbay, Communism and Agrarian Reform in Iraq (London: Croom Helm,
1978), 108-122.
Saʻad Jawad, Iraq and the Kurdish Question, 1958-1970 (London: Ithaca Press,
205

1981), 77-78.
206
O’Ballance, The Kurdish Revolt, 76.
121
Mustafa’s position and preventing him from crossing the border.207 The cold
war between Barzani and Qasim turned into open conflict and Barzani
became the leader of a wider Kurdish revolt.

In the Kurdish nationalist narrative, this moment is seen as the beginning of

the Kurdish Revolution in Iraq both because the fight involved, for the first
time after the revolution, the Iraqi Army but also because, for the first time,
a significant proportion of the tribes – yet never all of them – joined the
forces led by Mustafa Barzani. Interviewed on the subject, the current head
of the Jaf tribe in the Dukan area Ibrahim depicts the revolt as a national
revolution:

It was a national movement. The first step was taken by the tribes,
then the local police, the KDP […] followed. My dad was a tribal
leader in the movement. […] my uncle was the first martyr in the
movement along with three other people. They were the first martyrs
of the Kurdistan movement, killed on September 11, 1961, near
Chamchamal. They were killed when they confronted an Iraqi unit
that was heading to Sulaymaniyah and Dukan. They wanted to stop
them.208
Insisting that the Land Reform played a very small role in the uprising, he
emphasises the chaos brought about by the “communist rule” over the
country that followed the fall of the monarchy as one of the drivers of the
tribal revolt revealing its strong class dimension:

The communist party at that time was influential all over Iraq,
including Kurdistan. The KDP was weak. […] the shaykhs and aghas
were assaulted, sometimes even clashes broke out, and people
were killed! […] The communist chaos was very strong in 1959 and
1960. […] It continued until 1961 and then tensions between aghas
and peasants ended because […] in all of the region, the aghas, the
shaykhs, the bags, the mullahs united. This is how the revolution
started.209
As the Barzani forces were increasingly coordinating their war efforts with
those of the anti-Qasim tribes, the KDP was torn apart by the contradiction

207
Interview with Fared Assasard (Sulaymaniyah, 2018).
208
Interview with Ibrahim Jaf (Dukan, 2019).
209
Interview with Ibrahim Jaf (Dukan, 2019).
122
between its socialist and its nationalist identities. The tribal agitation in the
south of the region had an evident class and tribal character that
contradicted the commitment of the KDP to land redistribution and the party
initially condemned the revolt as reactionary.210 Moreover, the revolting
tribes were adamant that they were only going to accept the leadership of
a fellow tribal chief such as Barzani.211 However, the success of Barzani’s
tribesmen and the rapid territorial spread of the violence was turning it more
and more into a Kurdish revolt and the KDP could not afford to be excluded
by a potentially national revolution. Again, Qasim preferred to choose his
own enemies and decided on their behalf. On September 23 rd, two weeks
after his first attack on Barzani’s forces, Qasim banned the KDP, forcing the
entire leadership to join the revolt.212 The support of the KDP was a priceless
gift to the tribal elite allowing its leadership to present the revolt as a
national revolution and to hide its original class motives. On the contrary,
the party gained very little. Several left-leaning intellectuals left the KDP
claiming that the party had capitulated to the reactionary aghas. 213
Moreover, the party leadership was aware that their contribution to the
military effort was going to be insignificant and that they risked becoming a
mere instrument in the hands of Barzani. To maintain a certain degree of
independence, in December 1962, the KDP established its own military
forces, the peshmerga, (‘those who face death’) and set up its headquarters
in Mawat, near Sulaymaniyah and far away from Barzani’s northern
heartlands.214

After the winter pause, Barzani started attacking Iraqi troops in March 1962.
Over the course of the year, the rebel tribal forces gradually developed a
more defined command structure and the semblance of an army as tribal
fighters were reinforced by more and more Kurdish deserters from the Iraqi

210
Jawad, Iraq, 80.
211
Interview Qadir Haji Ali (Sulaymaniyah, 2019).
212
O’Ballance, The Kurdish Revolt, 78-79.
213
Jawad, Iraq, 82.
214
Ibid., 82.
123
Army.215 Unable to pursue the Kurds in the mountains, the Iraqi Army fought
a defensive war and made large use of the air forces. The indiscriminate
bombing and raiding of Kurdish villages by the army contributed
significantly to the growing popular support for the uprising.216 With the ICP
gone and the KDP on the side of the uprising, the anti-landowning
sentiments of the Kurdish rural masses gave way to the terror of the Iraqi
army. As Jalal Jawhar – a rare Kurdish politician with a peasant background
– recalls:

[the Iraqis] saw no difference between a tribal leader and a peasant,

between a teacher and a student. […] I was a kid in 1963 when they
looted and burned down our village […] they saw no difference
between the aghas, the shaykhs, and the people […] This is why
people started to think that national oppression is more important
than the oppression of the aghas. Because national oppression
meant displacement and death [whereas] the oppression of the
aghas meant giving up a third of your harvest. It is not the same as
being displaced and killed.217
In these conditions, the Kurdish revolt gradually assumed a trans-class
character though at the cost of framing away any demands for democracy
and redistribution and accepting the leadership of the conservative aghas.

The Kurdish forces that were still on the government’s side – and that the
rebels derogatorily called jash, ‘little donkey’ – were declining vertiginously.
By the end of summer, most of the anti-Barzani tribes had given up,
switched to the rebels or became neutral.218 Moreover, the rebels affiliated
to the KDP were operating in the Kurdish cities of Kirkuk, Erbil, and,
especially, Sulaymaniyah, where they had virtual control of the streets at
night.219 The inconclusiveness of Qasim’s efforts to quell the Kurdish revolt
eventually contributed to his downfall. On February 8th 1963, Ba’athist
officers seized power and executed Qasim. Mullah Mustafa and the KDP

215
O’Ballance, The Kurdish Revolt, 82-85.
216
Ibid., 88-91.
217
Interview with Jalal Jawhar (Sulaymaniyah, 2019).
218
Rubin, Abd al-Karim Qasim, 374.
219
O’Ballance, The Kurdish Revolt, 91-92; Rubin, Abd al-Karim Qasim, 371.
124
welcomed the provisional government and agreed on a ceasefire. 220
However, negotiations between Baghdad and the Kurds failed due to
Barzani’s demands to establish a large Kurdish autonomous region including
all the ethnically mixed areas and especially the oil-rich Kirkuk as well as a
separate Kurdish army with its own air forces.221 The talks collapsed in early
June 1963 and the Iraqi Army attacked again Kurdish positions.

The war continued until November 1963, when another military coup
overthrew the government in Baghdad. The new Nasserite faction in power
agreed on a ceasefire and, after secret negotiations, Barzani and the
government reached an agreement on February 10th, 1964, that only listed
vague commitments towards Kurdish rights and did not mention territorial
autonomy. The KDP leadership, excluded from the negotiations, protested
the agreement and refused to comply. Tensions rose between Barzani and
the KDP throughout the spring and, in July, Barzani unilaterally convened a
party congress. Only a handful of delegates loyal to Ibrahim Ahmed made it
to the Congress and were arrested by Mullah Mustafa’s men while the entire
faction was expelled. In mid-July, Barzani’s forces marched on the KDP
headquarters in Mawat and forced Ibrahim Ahmed and his followers to flee
to Iran.222 Soon after Mullah Mustafa had ridden himself of his Kurdish rivals,
relations with Baghdad soured and, in April 1965, an incident precipitated
the war.

Mullah Mustafa was, at this point, the undisputed leader of the Kurdish
revolt, recognised by a large part of the Kurdish tribes but also by what was
left of the KDP. His rapid turnaround in the negotiations with Baghdad – from
demanding virtual independence in 1963 to accepting vague and limited
promises one year later – showed that, by 1964, he felt strong enough to
take a break from the war with the government and to focus on his internal
rivals. From its inception as the uprising of a group of disgruntled
landowners, in just five years, the Kurdish revolt had expanded to the towns

220
Jawad, Iraq, 107-113.
221
Ibid., 133-135.
222
Ibid., 169-172.
125
and even acquired a certain degree of peasant support. The KDP had
sacrificed its progressive programme and given nationalist legitimacy to the
revolt only to be forcibly taken over by Barzani’s forces while its leftist
leadership was purged. The expulsion of Ibrahim Ahmed’s faction from the
KDP formalised the historical fracture between the tribal and urban strands
of the Kurdish movement beyond repair.

The Cold War, and the Collapse of the Kurdish Revolution

After a new ceasefire in 1966, the war continued only via proxies, with
Baghdad arming the pro-government tribes hostile to the Barzanis.
Moreover, the government invited the KDP expellees to return from exile
and started subsidising them as a counterweight to Barzani’s forces. Ibrahim
Ahmed, whose leadership of the Kurdish left was increasingly shared with
his son in law and rising star Jalal Talabani, re-organised his military force
and clashed with Barzani’s men on numerous occasions.223

In July 1968, the Iraqi government was removed by a new coup. The Ba’ath
Party, in power again, showed an unprecedented will to approach the Kurds
and, between August and September, a series of decrees introduced
education in the Kurdish language, established a Kurdish Cultural Academy
and a university in Sulaymaniyah, and recognised Newroz, the Kurdish new
year, as a national holiday. The main author of this advanced Kurdish policy
was Iraq’s vice-president Saddam Hussein, a Ba’athist officer, who was
rapidly emerging as the strongman of the new regime. The Ahmed-Talabani
faction was, at this time, the main Kurdish partner of the government. The
Ba’athists had a clear interest in delegitimising Barzani but the partnership
with Ibrahim Ahmed and Talabani also had an ideological underpinning. Both
groups linked their nationalism to a broader project of modernisation and
socialist policies and viewed Barzani as a reactionary feudal leader and as
the fifth column of imperialism due to his growing relationships with all of

223
Ibid., 205.
126
Iraq’s enemies such as Iran, Israel, and the United States. 224 In autumn,
clashes between Kurdish factions led to a new phase of direct – if low-
intensity – fighting.225

Rising tensions with Iran convinced Baghdad to reopen negotiations with

Barzani and a peace agreement was signed in March 1970. The agreement
promised the institution of an autonomous Kurdish region and the Kurds
gained ministerial posts as well as the governors of the Kurdish provinces
of Sulaymaniyah, Erbil, and Dohuk.226 By the end of the year, the government
had amended the constitution, took steps towards the economic
development of the Kurdish north, and, while the pro-government jash were
disbanded, Barzani’s peshmerga started receiving a salary from Baghdad.227

After this initial phase of optimism, however, the relationship between

Barzani and Baghdad gradually deteriorated. On the one hand, the Kurds
accused the government of promoting an Arabisation policy to change the
demographic balance of Kirkuk. On the other hand, Baghdad was aware that
Barzani had never stopped smuggling weapons from Iran. Emboldened by
Tehran’s supply of heavy weapons and anti-aircraft systems, Barzani
demanded Kirkuk as the capital of the Kurdish region. Negotiations lingered
on until the beginning of 1974 when, in March, the government decided to
promulgate a unilateral and limited Autonomy Law for the Kurdish region.228
As Barzani rejected the plan, war with Baghdad broke out again.

The escalation of the first half of the 1970s was closely related to broader
regional and international dynamics. As the Shah of Iran was growing more
assertive – and more pro-active in his support for Barzani – Baghdad
became inclined to make unprecedented concessions to the Kurds.
However, neither side was willing to compromise on the question of Kirkuk

224
McDowall, A Modern History, 323-324.
225
Jawad, Iraq, 242-246.
226
Cited in Tahiri, The Structure of Kurdish Society, 353-357.
227
McDowall, A Modern History, 327-328.
228
For the text of the 1974 Autonomy Law, see Tahiri, The Structure of Kurdish
Society, 359-365.
127
that, at the time, provided a large part of the country’s oil revenues. The
control of Kirkuk would have made the Kurdish region virtually independent
which was what Barzani ultimately wanted. In 1972, the signature of the
Iraqi-Soviet Treaty of Friendship and the nationalisation of the oil industry
in Iraq added new reasons for the United States and Iran to push the Kurds
to revolt. In 1973, combined financial efforts from the United States, Iran,
and Israel provided the Kurds with one million dollars per month, a figured
that was further increased in 1974.229 The new phase of the Iraqi-Kurdish
war reached unprecedented intensity, due to the heavy weaponry at
Barzani’s disposal and the direct assistance of Iranian special forces.230 On
the other hand, the Iraqi Army had been reinforced by Soviet supplies and
by the much greater resources made available by the nationalised oil
industry.

However, the deeper involvement of foreign powers in the war was also the
reason for its sudden end. On March 6th, 1975, Saddam Hussein and the
Shah met in Algiers and signed an agreement that settled their border
dispute in favour of Tehran. On the same day, Iran and its allies ceased to
support the Kurds in all forms and the entire front collapsed. About 100.000
Kurdish fighters and civilians took refuge in Iran, many others surrendered
to the Iraqi Army. Complying with the iron laws of the Cold War, the United
States made clear that their interest in the Kurdish cause went only as far
as it was compatible with the foreign policy of their regional ally.231 Mullah
Mustafa spent the following four years under strict surveillance between

Henry Kissinger, Years of Renewal (New York: Simon & Schuster, 1999), 584-
229

585.
230
McDowall, A Modern History, 337-338.
231
In a letter to the Shah, US Secretary of State Henry Kissinger made his order of
priorities very clear: “With respect to the Kurdish question, […] this is obviously a
matter for Your Majesty to decide in the best interests of your nation.” Kissinger,
Years of Renewal, 594. According to Marianna Charountaki, US support to the
Kurds only had the function to wear out Iraq while the prospect of Kurdish self-rule
would have been an unacceptable outcome to key US allies such Iran and Turkey.
Ultimately, “Kissinger and the Shah both hoped that their clients – the Kurds –
would not prevail.” Marianna Charountaki, The Kurds and US Foreign Policy:
International Relations in the Middle East since 1945 (London: Routledge, 2011),
138.
128
Tehran and Washington, where he died of cancer in March 1979, leaving the
leadership of his movement to his sons Idris and Masoud.232

With the collapse of the Kurdish forces in 1975, Baghdad restored its
authority over the region. However, Mullah Mustafa’s exile allowed different
Kurdish forces to emerge and compete for the leadership of the national
movement. Within the context of the rapid urbanisation and social
transformation of the country, the urban and leftist strand of Kurdish
nationalism took the chance for an unexpected comeback.

Iraq’s Two Kurdish Nationalisms

The expulsion of Ibrahim Ahmed and Jalal Talabani from the KDP in 1966
marks a turning point in the history of the Kurdish nationalism in Iraq. The
Kurdish revolt of 1961 had followed the script of previous Kurdish uprising
since the interwar period: The progressively minded and left-leaning
nationalist intellectuals had to surrender the leadership of the national
movement to tribal chiefs due to the far superior military means of the latter.
Ibrahim Ahmed and his followers in Sulaymaniyah and other Kurdish towns
had tried to give Kurdish nationalism a modernising mission. Like
progressive intellectuals in the rest of Iraq and in much of the decolonising
world, they viewed traditional social structures such as tribalism as deeply
intertwined with colonialism. The ties between the KDP and the communist
party in the 1940s and 1950s had been based on a strikingly similar agenda
that combined the recognition of Kurdish national rights in a federal and
democratic Iraq and social progress. However, the social and economic
structure of the Kurdish region did not play in their favour. The peripheral
location of the Kurdish lands in the broader economy of Iraq reinforced its
rural character. The Kurdish towns remained small and economically
marginal and the urban middle classes did not have any link with the
overwhelmingly rural masses. Unable to rally popular support, the
leadership of the KDP was forced, in the early 1960s, to follow the tribal elite

232
Idris Barzani died in 1987
129
in a rearguard struggle against land redistribution, out of fear of siding with
Baghdad against fellow Kurds. However, this contradiction was doomed to
explode and, when the Ahmed-Talabani faction was expelled from the KDP,
they became an alternative political force.

Rather than being the promoters of a specific political ideology, Mullah

Mustafa Barzani and his followers were characterised by a lack thereof.
During his decade-long exile in the Soviet Union, between 1946 and 1958,
Barzani vaguely adopted some Marxist rhetoric yet strongly rejected class
struggle. In a speech delivered at the Congress of the Kurdish Exiles in Baku
in 1948, Barzani seemed to perfectly understand that, as a member of the
tribal elite, he had no interest in talking the language of class politics:

Our Party defends the interests of all Kurdish classes including

chieftains, mercantilists, workers, small landowners, skilled workers,
farmers and intelligentsia. It brings all these together under the
banner of national liberation of the homeland and defending the joint
interests of all classes. Under the banner of this party, class struggle
in Kurdistan is not appropriate.233
This language set Barzani apart from most of the national liberation
movements hosted and supported by the USSR in those years, forces that
emphasised the link between anti-colonial struggle and social revolution.

Mustafa Barzani’s political project is more easily discernible through the

study of his practices and his approach to power. Since the beginning of his
political career in the 1930s, Barzani seemed to give little importance to the
role of political organisation.234 He seems to have regarded his chairmanship
of the KDP more as a prestigious title that gave him a degree of super-tribal
authority rather than an office within the organisational structure of the
political party. For all his life, Barzani pursued his personal tribal diplomacy,
ignoring the KDP as a decision-making body and distributing party offices
to please his tribal allies. This attitude reflected a mindset by fellow tribal

233
Barzani, Speech Presented, 46.
234
This aspect was heavily stressed by both Fared Assasard and Kamran
Karadaghi. Interview with Fared Assasard (Sulaymaniyah, 2018) and Kamran
Karadaghi (London, 2018).
130
chiefs for whom ideological differences had little or no weight in determining
patterns of alliances. Modern political ideologies have been mostly used to
retroactively justify decisions that had been made according to the relations
of power at a certain time. The Zebari tribe, for example, had consistently
opposed Mullah Mustafa throughout the 1960s, but, after the 1970 peace
agreement and when Barzani seemed triumphant, they suddenly became
Kurdish nationalists. Hoshyar Zebari, who was also the brother-in-law of
Mullah Mustafa, joined the Kurdish movement after 1970 and presents a
telling picture of Barzani’s attitude:

Well, it's a long story… I was brought up more or less in Mosul, I

completed my high school in Mosul. At the time there were some
tribal differences between us the Zebari tribe and the Barzanis. But
when Mullah Mustafa and the KDP signed the 11 of March 1970
autonomy deal with Iraq, there was a boom of Kurdish nationalism.
[…] that was the beginning of change or reconciliation and Mullah
Mustafa was a true statesman, a historical leader. Really… with all
those Kurds who had opposed him, […] he pardoned everybody.235
Hoshyar Zebari rapidly rose to prominence in the KDP and, in the 1980s, he
had already become one of its most powerful leaders.

When the KDP was established by Sulaymaniyah-based intellectuals in

1946, Barzani accepted to become its chairman under the condition that two
powerful landowners and members of the tribal elite were elected as vice-
chairmen: Muhammad Kaka Ziyad, an agha from Koya and Latif Barzanji, a
shaykh.236 The ideological shallowness of these personalities is well
represented by Latif. Son of Mahmoud Barzanji, the feudal nationalist leader
of the 1920s, Shaykh Latif was, like his father, a land-grabbing shaykh and
became the target of a major peasant revolt in 1947.237 Latif’s official role in
the KDP did not prevent his brother Baba Ali from becoming a minister in

235
Interview with Hoshyar Zebari (Pirmam, 2018).
236
Interview with Fared Assasard (Sulaymaniyah, 2018)
237
Batatu, The Old Social Classes, 467-468. Anecdotal information collected in
Sulaymaniyah says Latif is remembered less for his role in the KDP than for his
land-grabbing and exploitative habbits. When I asked about Latif, people told me
that he would show up at a village with his thugs, shoot his rifle, and claim all the
land to the point the bullet touched the ground.
131
Baghdad several times: first under the King, then under Abd al-Karim Qasim,
and once again under the Ba’ath.238 In the same way, during the Kurdish
revolts in the 1960s, Barzani relied on local tribal leaders as commanders of
the section of the front where they had their land.239 This strategy created
a direct and quasi-feudal relation between the paramount leader and the
local chiefs. But, even more importantly, this strategy demonstrated to the
tribal elite that for them the Kurdish revolt was the best way to preserve
their power over the land, threatened by a mobilised peasantry and by the
land reform. Consistent with this strategy, Barzani’s demand for regional
autonomy lost all the elements of development and modernisation and
revealed its nature as a project of personal rule underpinned by the
traditional structures of Kurdish society. It is then not surprising that in the
early 1970s, the survival of the Kurdish revolt depended exclusively on the
active support of Iran and the United States that systematically opposed
progressive movements and regimes throughout the Middle East.

Mullah Mustafa’s power over the Kurdish national movement in Iraq was a
direct consequence of the structural weakness of its urban middle-class
component. The birth of the KDP in 1946 had followed the pattern of class
relations that had characterised the Kurdish movement in the interwar
period. Intellectuals living in the main Kurdish towns saw nationalism as a
project of modernisation and development. Mostly drawn from the urban
middle-classes – lawyers, doctors, army officers, government employees –
they viewed the tribal structure of Kurdish society as an obstacle to
progress. However, unable to confront the government militarily, they had
to surrender the leadership of the movement to the powerful tribal chiefs.
The presence of part of the tribal elite in the KDP is one of the reasons why
the party was unable to take advantage of the period of mass mobilisation
against the monarchy in the 1950s. At the time, the Kurdish region was torn
apart by rural conflicts and the KDP failed to lead and even struggled to take

Edmund A. Ghareeb and Beth Dougherty, Historical Dictionary of Iraq (Lanham:

238

The Scarecrow Press, 2004), 349-359.

239
Interview with Qadir Haji Ali (Sulaymaniyah, 2019).
132
part in the mass movement leaving a vacuum that was filled by the
communist party. Behind the reluctance of the KDP leadership to link their
progressive nationalism to the agrarian struggles there was their
unwillingness to break with the sector of the tribal elite with whom they had
allied despite the fact that they were exploitative and reactionary
landowners, as the example of Latif Barzanji showed. The ICP became the
natural point of reference for progressively minded Kurds in the 1950s and
the KDP remained a marginal force in the struggles against the landlords
and the monarchy.240

When the Kurdish revolt of 1961 broke out and Barzani allied with the tribal
landowners threatened by land reform, the KDP lost all room for political
initiative and was forced into a rearguard struggle against the Iraqi
government. Unable to link their action to that of other progressive forces
in the country, the KDP put itself at the mercy of Mullah Mustafa, and as
soon as a disagreement between the two sides emerged, the latter could
simply take over the party with his armed men. Expelled by the KDP in 1966,
Ibrahim Ahmed and his followers were reduced to an instrument in the
hands of the Iraqi government.

However, the Kurdish uprising that started in 1961 and that continued
intermittently until 1975 had more far-reaching consequences for the social
structure of the region. On the one hand, Barzani’s movement gradually
acquired a mass dimension as the repressive measures and scorched-earth
tactics of the Iraqi Army were driving a wider portion of the Kurdish
population towards the rebels. On the other hand, these practices created
continuous waves of rural refuges to the Kurdish towns that added to the
ongoing structural process of urbanisation. In 1977, the urban population of

240
One example of this trend is Kamran Karadaghi, a Kurdish student in Baghdad
in the 1950s, for whom the ICP represented the only option for progressive
activism. Karadaghi fell victim to the anti-communist repression of 1961 and fled to
the Soviet Union where he completed his studies. Upon his return to Iraq, in the
early 1970s, the ICP was no longer a competitive political actor and Karadaghi
became active in the Kurdish national movement. Interview with Kamran Karadaghi
(London, 2018).
133
the Kurdish region reached almost 50 per cent.241 Once they had moved to
an urban environment, rural refugees tended to lose their tribal identity. As
Kurdish historian Fared Assasard explains in reference to the Sulaymaniyah
area, “In the 1970s tribalism wasn’t a significant force, […] the heads of the
tribes were respected, but people would not particularly listen to them.” 242
The rural migrants would be more easily politicised over issues of
employment, basic services or the brutality of the security forces. These
demands were answered by the far-leftist political organisations that were
mushrooming in the late 1960s and early 1970s. The growth, mostly among
high-school and college students, of radical leftist groups was a global
phenomenon but, in Iraq, it was strengthened by the continuous
government crackdowns that targeted the communist party, systematically
weakening the strongest force on the left. More generally, there was a
widespread belief among urbanised Kurds that the 1975 defeat of the
movement had been largely due to its tribal and reactionary character. Qadir
Haji Ali, who at the time was a high school student and far-left activist in
Sulaymaniyah, recalls: “our problem with the [KDP] was that it was a tribal
party, the aghas and the shaykhs were with them, and [we] were against
tribalism and the bourgeoisie!”243

These transformations constituted the underlying social conditions for the

establishment of a Kurdish force alternative to the Barzani-controlled KDP.
When the Kurdish revolt collapsed in 1975, Jalal Talabani and other Kurdish
leaders of the left took refuge in Syria, while the old Ibrahim Ahmed went
into exile in London. On June 1st, the Patriotic Union of Kurdistan (PUK) was
established as the federation of three groups: Jalal Talabani and his
followers, Komala, a well-organised Maoist group led by Nawshirwan
Mustafa, and the Socialist Movement of Kurdistan (Bezutnawa) led by Ali
Askari and Rasul Mamand. The party was formed as “a broad democratic

241
53.2 per cent in Erbil, 47.2 in Sulaymaniyah, 42.9 in Dohuk. Iraq Population
Situation Analysis Report 2012 (Iraq National Population Commission and UNFPA,
2012) <http://iraq.unfpa.org/publications/cat_view/1-documents-english>, 97.
242
Interview with Fared Assasard (Sulaymaniyah, 2018).
243
Interview Qadir Haji Ali (Sulaymaniyah, 2019).
134
and patriotic front that allows the fighting unity and coexistence of the
different progressive tendencies under the leadership of a Kurdish
revolutionary vanguard.”244 They ascribed the collapse of the revolt to “the
inability of the feudalist, tribalist, bourgeois rightist and capitulationist
Kurdish leadership” and proclaimed the PUK's commitment to autonomy for
the Kurds and democracy for Iraq.

War, Genocide, and Liberation

The establishment of the PUK gave new vigour to the Kurdish movement.
While the KDP was still reorganising its forces inside Iran, the PUK, thanks
to its cells active in the Kurdish towns, was able to start low-key military
operations in the region. Between 1976 and 1980, the KDP had little or no
presence in the Kurdish region but clashed several times with the PUK in the
border areas. The Iranian Revolution in 1979 triggered events with far-
reaching consequences for the Kurdish movement. Hoping to take
advantage of the country’s post-revolutionary instability, Saddam Hussain
invaded Iran in September 1980. The Iran-Iraq War, that lasted for much of
the 1980s, was one of the bloodiest and most destructive conflicts of the
Cold War era and allowed Kurdish forces to resume large-scale military
operations in Iraq.

During the following years, the Iran-Iraq war was mostly concentrated in
southern areas on the Persian Gulf far from the reach of Kurdish groups. The
situation changed in the mid-1980s when Iran started pushing on the
northern side of the front with the support of KDP fighters. As the war
expanded to the Kurdish region, Saddam started mass recruitment of tribal
Kurds into irregular ‘jash’ forces and Talabani found himself squeezed
between the Iraqi Army and the advancing Iranian and KDP forces.245 He had
no choice but to reach out to his Kurdish enemies. Sponsored by Tehran,

244
Revolution in Kurdistan: The Essential Documents of the Patriotic Union of
Kurdistan (New York: PUK Publications, 1977), 6.
245
Mohammed Malek, ‘Kurdistan in the Middle East Conflict’, New Left Review,
1.175 (1989), 88-91.
135
the KDP and PUK agreed to join efforts against Baghdad. In May 1987, Jalal
Talabani and Masoud Barzani246 established the Kurdistan Front with a
unified command and a clear division of the frontline. The KDP operated in
the north of the Iraqi-Kurdish region – the provinces of Dohuk and Erbil –
and the PUK in its southern part – in Sulaymaniyah and Kirkuk.247

Saddam perceived the unity of the Kurds with the backing of Iran as a mortal
threat and responded by appointing his cousin Ali Hasan al-Majid as
governor of the north with absolute power. In the spring of 1987, the Iraqi
Army started using chemical weapons against Kurdish villages, deporting
the villagers and often executing all adult men. These genocidal practices
were extended to the entire Kurdish region when, in January 1988, the army
initiated the infamous Anfal Campaign. On March 16th, about 5,000 people
were killed by Iraqi chemical weapons in the town of Halabja. The Anfal
campaign continued even when, in August 1988, Iran and Iraq agreed on a
ceasefire. By June 1989, between 50,000 and 100,000 Kurds had been
killed, thousands of villages had disappeared, and their populations
massacred or moved to concentration camps to be finally resettled in
newly-built towns where the government could much more easily exert
control.248 Hundreds of thousands of refugees found shelter in Iran and
Turkey. The Kurdish revolt, once again, collapsed. As former PUK
peshmerga Jalal Abdullah Hamarahim recalls, “small groups remained active
but, once the villages were eradicated, the bulk of the insurgency could no
longer survive without their support and we had to withdraw from Iraq.”249

In August 1990, Saddam’s invasion of Kuwait triggered the intervention of

the United States that easily defeated and pushed back the Iraqi Army. In
February 1991, American President George H. W. Bush invited the people of

246
His brother Idris had died in January 1987 leaving Masoud as the undisputed
leader of the KDP.
247
McDowall, A Modern History, 351-352.
248
Genocide in Iraq: The Anfal Campaign against the Kurds (Human Rights Watch,
1993) <www.hrw.org/legacy/reports/1993/iraqanfal>.
249
Interview with Jalal Abdullah Hamarahim (Sulaymaniyah, 2018).
136
Iraq “to take the matters into their own hands.”250 On March 1st, most of the
Shiite south revolted forcing the army to withdraw. On March 4th, the Kurdish
town of Ranya rose against the government and, in a matter of days, Iraqi
troops were expelled from the entire Kurdish region. As KDP and PUK had
maintained only a very marginal presence in Kurdish areas since the end of
the Iran-Iraq war, the revolt was largely spontaneous. While the leaders of
the Kurdistan Front hesitantly joined the rebels, the withdrawal of Iraqi
troops was most often negotiated by local tribal chiefs and jash
commanders.251 On March 19th, Kurdish forces took Kirkuk.

However, both the US and the rebels had underestimated the resilience of
Saddam’s military capability. By the end of March, the government had
recaptured most of the south and, on March 28th, Iraqi troops retook Kirkuk.
The Iraqi Army launched a massive air-bombing campaign and swiftly
recovered much of the Kurdish region. One and a half million Kurdish
civilians tried to leave the region under the bombs of the Iraqi air forces. The
brutal repression of the Kurdish uprising became the justification for direct
American intervention in Iraq.252 On April 5th, the United Nations (UN)
Security Council passed a resolution condemning the repression and, on the
following days, the US, Britain and France imposed a no-fly zone over
northern Iraq. Unable to use his air forces, Saddam struggled to contain the
Kurdish peshmerga. In July, the Kurdistan Front re-took control of Erbil and
Sulaymaniyah and, in October, the Iraqi Army again withdrew from the
Kurdish areas. At the end of 1991, the KDP and PUK were in control of most
of the Kurdish region of Iraq, with the notable exception of Kirkuk.

‘Excerpts From 2 Statements by Bush on Iraq’s Proposal for Ending Conflict’,

250

The New York Times, 16 February 1991

<https://www.nytimes.com/1991/02/16/world/war-gulf-bush-statement-
excerpts-2-statements-bush-iraq-s-proposal-for-ending.html>.
251
McDowall, A Modern History, 371.
252
However, it seems that Turkey’s pressure played a central role in the decision.
Fighting its own Kurdish conflict, Ankara had proven unwilling to let in Kurdish
refugees from Iraq. Michiel Leezenberg, ‘Humanitarian Aid in Iraqi Kurdistan’,
Cahiers d’Études Sur La Méditerranée Orientale et Le Monde Turco-Iranien, 29
(2000), 33-34.
137
Conclusion

The American intervention of 1991 marked a turning point in the history of

Iraq. After decades of struggle, Kurdish forces took control of large part of
the Kurdish-inhabited areas in the country to which the Iraqi Army no longer
had access. Kurdish autonomy, though finally achieved via international
intervention, was the result of a long liberation struggle that this chapter has
presented as rooted in the class relations of the region. The overview of the
structural transformations of Iraq during the colonial and monarchical
periods showed that the resurgence of the Kurdish movement in the 1960s
cannot be disentangled from the broader social and political dynamics of
the country. Most of the literature on the Kurdish movement in Iraq follows
the path of military campaigns and negotiations and overlooks the structural
dynamics that constitute the conditions for the Kurdish movement to
emerge and to become a political force able to challenge the central
government.253 Historians of Iraq have instead paid much more attention to
the role of colonialism and class stratification in the process of state
formation but, in their work, the political events of the Kurdish movement
rarely take up more than a few footnotes.254 By comparing these two sets of
literature, a more complex picture of the Kurdish movement emerges. In
particular, the comparison reveals the centrality of the colonial period in the
process of class formation that sets the conditions for the Kurdish revolts
of the 1960s. The strength of the tribal landowning class became an element
of primary importance that forced the other strand of the Kurdish movement
– the urban and left-leaning nationalists – to a position of subordination. The
analysis in this chapter showed that the main issue of contention in Kurdish
politics from the 1940s to the 1970s was that of the ownership of land, as in
any other largely agrarian society. The anti-landlord mobilisation of the

253
See, for example, Jwaideh, The Kurdish National Movement, 147-289; Jawad,
Iraq; O’Ballance, The Kurdish Revolt; Bengio, The Kurds of Iraq; Kerim Yıldiz, The
Kurds in Iraq.
This chapter especially relied on Batatu, The Old Social Classes. See also
254

Dawisha, Iraq; Haj, The Making of Iraq; Dodge, Inventing Iraq,

138
peasantry in the 1940s and 1950s, in Kurdistan as in the rest of Iraq, gave
an essential contribution to the success of the Iraqi national movement
against the monarchy and colonialism that culminated in the 1958
Revolution.

The Kurdish tribal uprising of 1961 was first and foremost a reaction against
the empowerment of the peasantry and, therefore, an episode of class
struggle. The strength of the tribal landowning class forced the urban
middle-class nationalists to join the revolt in a position of subordination that
allowed its framing in terms of national liberation. Within this conflict – and
in conjunction with the repression suffered by the communists countrywide
– the peasantry lost its political agency and villages were forced to pick one
side and face retaliation from the other. The result of continuous warfare in
the Kurdish region was the gradual destruction of Kurdish rural society. As
shown by this chapter, the first consequence of this process was the growth
of the Kurdish towns and the emergence of an alternative force of Kurdish
nationalism expressed by the establishment of the PUK in 1975. The next
chapter will trace the long-term consequences of this process of forced
urbanisation, that war and genocide in the 1980s dramatically accelerated.
This new ‘war society’ created a radically different class structure and
gradually eroded the ideological differences between KDP and PUK that,
since 1991, imposed their power duopoly over the – finally ‘liberated’ –
Kurdish region of Iraq.

139
140
 Chapter 5
The Class Structure of Kurdish
Self-Rule in Iraq (1991-2014)

Introduction

The social structures and class relations out of which the Kurdish national
movement in Iraq developed in the 1960s and 1970s – the subject of Chapter
4 – were radically transformed by the conflict that ravaged the Kurdish
region in the 1980s. When the KDP and the PUK, the two dominant Iraqi-
Kurdish parties, gained control of the region in 1991, the – largely rural –
Kurdish economy had been devastated. In the absence of any source of a
productive economy, the two parties competed to coopt the local elite and
to gain control of smuggling routes and foreign humanitarian aid. This
chapter argues that the result of this competition was the development of a
new class structure in which the political leaders and the members of the
old elite gained military control over the few sources of accumulation
available in the region keeping the population unproductive and dependent
on handouts. This set of class relations established in the 1990s was
strengthened by the recognition of Kurdish autonomy and the establishment
of a unified Kurdistan Regional Government (KRG) after the 2003 US-led
invasion of Iraq. As the security apparatus remained under the command
structure of the two ruling parties, the Kurdish political elite was able to
appropriate most of the wealth generated by the sale of the region’s oil and
had no interest in investing in the development of industry and agriculture.
With no alternative available, a large part of the Kurdish population was
forced to survive on government handouts tying them to the patronage
network of the political elite. This chapter shows that the political structure
of the KRG characterised by the absence of the rule of law and the partisan
control of security forces and public institutions cannot be understood
141
without reference to the specific class structure that emerged in the region
in the 1990s.

Liberation and Civil War

The withdrawal of Saddam Hussein’s army from the Kurdish region of Iraq
in October 1991 marked a turning point in the history of the Iraqi Kurds.
Protected by the no-fly zone imposed by the United States over northern
Iraq, the peshmerga militias of the two major Kurdish parties – the KDP and
the PUK – gained complete control over the predominately Kurdish
governorates of Dohuk, Erbil, and Sulaymaniyah. The economic and social
situation of the region was dire: After more than a decade of destructive
fighting and the international sanctions imposed on Iraq in 1990, the Kurdish
region faced an additional internal blockade put in place by Saddam upon
the withdrawal of his army.

One of the first measures taken by the Kurdistan Front dominated by the
KDP and PUK was a general amnesty for all the jash, the Kurdish tribal forces
that had fought alongside the Iraqi Army and that numbered in the hundreds
of thousands. Some of the jash had been forcibly recruited by the regime
and had maintained a rather passive attitude during the war and in some
cases even kept active communications with the Kurdish insurgency. Many
jash leaders had played a major role in the 1991 uprising that triggered the
liberation of the region. However, the intent of pacifying Kurdish society
went hand in hand with the determination of the KDP and PUK to incorporate
vast chunks of the population in their constituency by negotiating directly
with their tribal leaders.

The liberation of the Kurdish region brought with it a geopolitical dilemma.

Neighbouring Turkey, Iran, and Syria – each hosting their ‘own’ Kurdish
minority – were extremely uncomfortable with the idea of Kurdish self-rule
in Iraq while the United States remained committed to the territorial integrity
of the country. At the same time, the internal blockade meant that Baghdad
had stopped appointing and paying civil servants in the Kurdish

142
governorates and the region was in dire need of restoring a functioning
administration. To reassure the neighbouring powers that they did not aim
at secession, the Kurdistan Front decided to unilaterally implement the old
Autonomy Law issued by Iraq in 1970, that had established a regional
parliament based in Erbil and organised regional elections in May 1992.
Since no other party passed the 7 per cent threshold the KDP and PUK
obtained respectively 51 and 49 seats. 255 While all participants accused
each other of fraud, tensions were running high between the two parties
which had only come together in 1986, after two decades of rivalry and
intra-Kurdish fighting. However, keen to maintain international support the
KDP and the PUK went through a few days of intense negotiations that
ushered in a comprehensive power-sharing agreement.

The Kurdistan Regional Government (KRG) created by the two-party

agreement established a ‘50:50 system’ that Gareth Stansfield – in the most
extensive study of the Kurdish region of Iraq in the 1990s – describes as an
equal division of “all executive and legislative positions […] with real power
being unofficially vested in the political bureaux of the KDP and PUK.”256 The
legitimacy of the new system was significantly undermined by the absence
of the two party leaders, Masoud Barzani and Jalal Talabani who did not
take any direct role in the KRG government and preferred to exercise their
power through lesser political figures showing their half-hearted
commitment to the agreement. In 1993, the relationship between the two
parties was already deteriorating. The KDP accused the PUK of using the
office of the KRG Prime Minister to control the regional institutions and to
increase its influence over the Kurdish capital Erbil. Moreover, as both
parties had only partly given up their sources of revenues to the
government, the PUK was growing wary of the rapid increase in wealth of
the KDP due to the latter’s control of the Ibrahim Khalil border crossing with
Turkey that constituted by far the greatest source of revenue in the region.

Ruud Hoff, Michiel Leezenberg, and Pieter Muller, Elections in Iraqi Kurdistan
255

(May 19, 1992): An Experiment in Democracy (Amsterdam: Netherlands-Kurdistan

Friendship Society, 1992), 16, 29.
256
Stansfield, Iraqi Kurdistan, 145.
143
The peshmerga, the party militias meant to become the armed forces of the
KRG, were never unified and remained under the separate command
structures of each political party.

In a matter of months, tensions escalated into full-blown civil war. In the

winter of 1993-1994, the PUK became more and more aware that the KDP’s
close relations with Iran and the revenue from Ibrahim Khalil were rapidly
changing the balance of power in the region. In May 1994, while incidents
between followers of the two parties spread, the PUK took over the Kurdish
parliament in Erbil definitively breaking the power-sharing agreement.257
The lack of minimum trust between the two parties made the 50:50
agreement so fragile that the first perceived change in the balance of power
was fatal. Moreover, this zero-sum mentality was fed by the new political
economy of the region, which made territorial control essential to access
the only available sources of revenues, namely the border-crossings and
the distribution of international aid.

For about two years, the Kurdish civil war that started in spring 1994 took
the form of a low-intensity conflict between the KDP controlling Dohuk and
the PUK holding Erbil and Sulaymaniyah. Despite its historical alliance with
Iran, the KDP had also been able to build a friendly relationship with
Baghdad especially thanks to their cooperation on cross-border
contraband. This development inevitably drove Iran and the PUK closer,
turning the Kurdish conflict into a proxy war that contributed to its
significant escalation. In July 1996, the PUK allowed the Iranian
Revolutionary Guards to enter the Kurdish region and attack the Iranian-
Kurdish guerrilla that had found shelter in the town of Koya. In response, the
KDP attacked Erbil with the direct support of the Iraqi Army and took the
Kurdish capital on August 31. The Iraqi-backed KDP offensive continued to
the point that the PUK was forced to leave Sulaymaniyah until the end of
the year when the heavy weapons provided by Iran allowed Talabani’s
forces to recover the city. The fighting of 1996 heavily involved the civilian

257
Michael Gunter, The Kurdish Predicament in Iraq: A Political Analysis (New York:
St. Martin’s Press, 1999), 76. Tahiri, The Structure of Kurdish Society, 276-277.
144
population forcing up to 70.000 followers of the two parties to leave their
homes as the frontline moved.258 The year 1997 saw a series of failed
negotiations as well as the direct intervention of another foreign power,
Turkey, into the conflict. Since the beginning of the 1990s, Ankara’s
involvement in the region had been a spillover of its own Kurdish war against
the Kurdistan Workers’ Party (PKK) that had set its bases in the Qandil
Mountains north of Erbil. In May 1997, Ankara carried out a large operation
against the PKK and kept some of its forces in the region. 259 Given their
geographical proximity, Turkey started building a close relationship with the
KDP which brought the PUK closer to the PKK but also increased its
dependence on Iran’s support. When, in October, the PUK launched an
offensive to retake Erbil, the KDP managed to thwart the attack with the
help of Turkey’s air force.260

The failed PUK-offensive of October 1997 marked the end of hostilities and,
in November 1997, the KDP and PUK signed a truce. The United States,
worried about the growing influence of Baghdad and Tehran – its regional
foes – over the warring Kurdish parties, took the lead. In September 1998,
the negotiations ushered in the Washington Agreement that committed the
two parties to re-establish the KRG and to solve all pending issues, such as
the sharing of revenues and the freedom of movement within the region,
with an American guarantee against the intervention of the regional powers.
However, none of those issues was resolved. Each party established its own
administration and thousands of followers had to move to the area
controlled by their party. The region remained divided into two security
areas named after the respective party colours: a KDP-controlled Yellow
Zone in the governorates of Dohuk and Erbil and a PUK-controlled Green-
Zone in Sulaymaniyah.

Tahiri, The Structure of Kurdish Society, 278; Gunter, The Kurdish Predicament,
258

86-88.
259
The 1997 incursion was Turkey’s fourth and largest operation in Iraq in the
1990s, involving up to 50.000 troops. Bill Park, Turkey's Policy towards Northern
Iraq: Problems and Perspectives (London: IISS, 2005), 19-20.
260
Gunter, The Kurdish Predicament, 88-89.
145
Urbanisation and Dependence

The economic situation and class relations of Iraqi Kurdish society in the
1990s was primarily the result of the destruction brought by a decade and
a half of warfare. The Iran-Iraq War and the genocidal Anfal Campaign were
followed by the Kurdish uprising of 1991 and by the Kurdish Civil War
between 1994 and 1997. These tragic events largely destroyed the rural
society of Iraqi Kurdistan and led to a massive process of urbanisation. In
the 1990s, the Kurdish region was also subjected to a dual blockade brought
by the combination of the UN-imposed sanctions on Iraq and the domestic
embargo imposed by Saddam Hussain on the Kurds. Moreover, the
withdrawal of the I raqi army allowed for the flow of a large amount of
humanitarian aid that, while providing indispensable relief, further
undermined the productive capacity of the Kurdish economy. All these
factors together contributed to turning Iraqi Kurdistan into a rentier society
militarily controlled by the two major parties.

The long cycle of violence of the 1980s and 1990s drove many Kurdish
peasants to leave their villages and find refuge in the Kurdish towns.
Baghdad’s counter-insurgency campaigns – that ultimately led to the
genocidal Anfal Campaign (1986-1989) – were primarily aimed at removing
the sources of supply for the Kurdish guerrilla by evacuating thousands of
villages and depopulating the countryside. This policy resulted in the
displacement of hundreds of thousands of peasants that “were relocated
into mujamma'at, crude new settlements located on the main highways in
army-controlled areas of Iraqi Kurdistan”.261 The Iraqi Army cleared several
miles of territory along the Iranian border of villages and forest and entire
areas of the countryside were rendered unviable with defoliant chemicals,
landmines, deforestation, and the destruction of wells and irrigation
systems. The result of this process was the devastation of Kurdish
agriculture and a massive process of forced urbanisation. Rural migrants

261
Genocide in Iraq, 6.
146
swelled the Kurdish cities which grew exponentially. Between 1977 and
1997, the urban population of the three Kurdish of governorates (Dohuk,
Erbil, and Sulaymaniyah) grew from below 50 per cent, to over 70. The
average growth of 23 per cent in the urban population of the Kurdish
provinces contrasts starkly with the rest of Iraq that experienced virtually
none.262

The situation became catastrophic in the early 1990s, when the Iraqi
government withdrew from the Kurdish region, stopping the payment of civil
servant salaries as well as food subsidies and other forms of public
assistance while unemployment rose to 70-80 per cent.263 The end of
financial support from Baghdad was part of the economic embargo put in
place by Saddam Hussein against the newly-established KRG. The domestic
embargo was an additional burden on the Kurdish provinces already hit by
the international sanctions on Iraq that were having a devastating effect on
the country’s economy. The hardship imposed by this dual blockade was
mitigated by a wave of humanitarian assistance. Given the restrictions
posed by the Iraqi regime on aid delivery in the rest of the country, between
1991 and 1996 the Kurdish region received two-thirds of total aid delivered
to Iraq worth over $1 billion.264 The aid provided essential goods and
services that allowed for the survival of the most vulnerable sectors of
society at a time where prices skyrocketed and food and energy had
become inaccessible to a vastly unemployed population.

In her well-informed study on the topic, Denise Natali identified two issues
with humanitarian aid that structurally affected the Kurdish region. First,
despite the sanctions, the UN remained committed to the territorial integrity
of Iraq and, to be able to operate in the country, could not directly interact
with the Kurdish Government that Baghdad deemed illegal. Therefore,
humanitarian aid could not be channelled through the KRG. 265 As most of the

262
Iraq Population Situation Analysis, 97.
263
Stansfield, Iraqi Kurdistan, 53-54.
264
Natali, The Kurdish Quasi-State, 30.
265
Ibid., 35.
147
essential goods and services were delivered directly by non-governmental
organisations (NGOs), the KDP and PUK had no interest in hiring civil
servants to run the ministries, setting up government agencies, and
developing a public sector. The two parties remained military organisations
focused on security rather than political machines working within a
parliamentary system. As Natali explains “without the ability to implement
its own tax programs, acquire capital or foreign exchange, or negotiate with
a foreign government to attract investments, the regional economy
experienced no real structural change.”266 The redundancy of the KRG
institutions meant that the KDP and PUK had little incentive to work within
shared institutions and much to gain from territorial control, a dynamic that
ultimately precipitated the outbreak of the civil war in 1994.

The second issue with humanitarian aid highlighted by Natali is the

structural undermining of the productive sectors of the economy.
Humanitarian assistance was rarely aimed at capability building, a tendency
reinforced by the prohibition to work directly with Kurdish institutions. For
example, when the electricity broke down in the Dohuk governorate in 1993,
the UN, rather than repairing the grid or funding the creation of an
alternative source of energy, “distributed generators that provided
temporary but costly power dependent upon imported diesel.”267 Still, in
2002, the UN purchased four massive generators for the city of Erbil, each
of them costing $2.5 million and requiring 360 oil barrel per month, instead
of repairing the Bakhma dam for a total estimated cost of $2 million.268

The role of foreign aid in generating dependence was strengthened by the

so-called oil-for-food programme instituted in May 1996 through an
agreement between the UN and Baghdad that allowed Iraq to sell oil in order
to buy foodstuff and medicines despite international sanctions. The KRG
institutions were excluded from the agreement that explicitly stated that the
distribution of supplies in Erbil, Dohuk and Sulaymaniyah was to be

266
Ibid., 44.
267
Ibid., 42.
268
Ibid., 60 n21.
148
“undertaken by the United Nations […] with due regard to the sovereignty
and territorial integrity of Iraq”.269 Compared to the $1 billion worth of aid
received during the relief phase, the share of oil revenues destined for the
Kurdish governorates (13 per cent) amounted to almost $10 billion. 270
However, as in the first phase of humanitarian aid, the programme failed to
invest in development. The most devastating consequences of this
approach were experienced by local agriculture. In the early 1990s, Kurdish
agriculture was gradually recovering thanks to the reconstruction of 2.800
villages271 and, especially, to the effects of the sanctions and the blockade
on food prices.272 In 1996, with the oil-for-food programme, food aid flooded
the region and prices dropped, forcing thousands to abandon the
countryside and go – sometimes back – to the cities. Rather than buying
wheat from local farmers, the UN purchased it from Australia, spending half
of the allocated budget in transportation.273

Warlords and Chiefs: the Formation of a New Ruling Class

With the withdrawal of the Iraqi Army in 1991, the KDP and PUK gained
permanent military control of the Kurdish region. As mentioned before, one
of the first issues that the two parties had to face was that of the jash, the
mostly tribal Kurds who had fought alongside the Iraqi Army as irregular
forces. Rather than carrying on a punitive policy towards the former
collaborators of the regime, the Kurdish parties issued a general amnesty
and entered into a fierce competition to gain the support of their previous
enemies. The result of this process was the creation of a new ruling class,
formed by the integration between the old jash elite and the peshmerga

269
‘IRAQ-UNITED NATIONS: MEMORANDUM OF UNDERSTANDING ON THE SALE
OF IRAQI OIL, IMPLEMENTING SECURITY COUNCIL RESOLUTION 986 (1995)’,
International Legal Materials, 35.4 (1996), 1100.
270
Natali, The Kurdish Quasi-State, 53.
271
Stansfield, Iraqi Kurdistan, 53.
From 1990 to 1995, in the Kurdish region, wheat price increased by 50 per cent
272

and the area of cultivated land by 52 per cent. Natali, The Kurdish Quasi-State, 32.
273
Stansfield Iraqi Kurdistan, 42; Natali, The Kurdish Quasi-State, 60.
149
commanders. Thanks to its military orientation, this new elite was in the
position to take advantage of the two major economic activities available –
the management of foreign aid and the control over smuggling – to amass
great wealth.

The jash leaders were more than military commanders. Almost exclusively
the members of leading Kurdish tribal families and often historical rivals of
the tribes that had joined the Kurdish liberation movement, these figures
had been greatly empowered by the policies promoted by Saddam Hussein
in the 1980s. In 1980, the Kurdish economy was still largely based on
agriculture and produced a third of Iraq’s barley and nearly half of its
wheat.274 However, the devasting consequences of the war combined with
the depressing effects of oil revenues over all the non-oil sectors of the
economy triggered a rapid decline in profits in agriculture. The decline of
agriculture and the growing availability of state funding contributed to a
gradual shifting of the traditional elites from landownership towards other
more profitable activities. In Iraq as a whole, tribal leaders loyal to the
government – along with Ba’ath Party officers – were the main beneficiaries
of the privatisation policies promoted by the regime during the Iraq-Iran
War.275 Moreover, the government rewarded political loyalty by assigning
lucrative contracts for state-funded infrastructural projects. Gradually, this
process led to the transformation of declining tribal landowners into a class
of regime-backed entrepreneurs whereby capital – rather than land –
became the main source of power.276

M. Sajjadi, ‘State of Economy in Kurdistan’, in Kurdistan: Political and Economic

274

Potential, ed. by Maria O’Shea (London: Geopolitics and International Boundaries

Research Centre SOAS, 1992), 50.
275
For the extent and impact of Saddam Hussein’s 1980s privatisation policies, see:
Kiren Aziz Chaudhry, ‘On the Way to Market: Economic Liberalization and Iraq’s
Invasion of Kuwait’, MERIP Middle East Report, 170 (1991), 14–23.
276
It is however questionable whether this can be called, as Leezenberg does, a
‘capitalist class’. Michiel Leezenberg, ‘Urbanization, Privatization, and Patronage:
The Political Economy of Iraqi Kurdistan’, in The Kurds, ed. by Jabar and Dawod,
158.
150
In the Kurdish region, tribal chiefs were also those with enough social power
to set up the jash units – officially National Defence Battalions – the loyalist
forces raised among pro-government Kurdish tribes to fight the insurgency.
The jash leaders were directly funded by the government to set up their own
units by recruiting fellow tribesmen. This practice did not only provide the
tribal elite with another opportunity for enrichment, but it also strengthened
their power over their constituencies. Joining the jash was often the only
opportunity for rural Kurds to receive a decent salary and avoid being
enlisted in the army and sent to the Iranian front. Moreover, this system
provided tribal leaders with semi-private armies that could be used for
private purposes.277 The attachment of this ruling class to the Ba’athist
government was, however, a marriage of convenience and, when Kurdish
forces returned to the region in 1991, they were quick to change side.

The competitive nature of the relationship between the KDP and the PUK
forced the two parties to battle to coopt the jash commanders. In the words
of Hoshyar Zebari, a powerful KDP leader, the general amnesty was a
triumphant process of “Kurdish national reconciliation” whereby the old
enemies “were integrated [into the political parties], their properties
protected, all cases against them were dropped.”278 Some leaders hailing
from traditionally anti-Barzani tribes made peace with the KDP, others
joined the PUK, while those who resisted were brutally murdered. 279 The
amnesty allowed them to keep their forces and they often became local
warlords enjoying a direct relationship with the party leadership and even
switching side whenever it suited them.280 Not only did the regime change

277
Leezenberg, Urbanization, Privatization, and Patronage, 158-159. See also,
Michiel Leezenberg, ‘Refugee Camp or Free Trade Zone? The Economy of Iraqi
Kurdistan since 1991’, in Iraq’s Economic Predicament, ed. by Kamil A. Mahdi
(Reading: Ithaca Press, 2002), 291-292.
278
Interview with Hoshyar Zebari (Pirmam, 2018).
279
This is the case for example of Hussein Agha, of the traditionally anti-Barzani
Surchi tribe, killed with most of his family by KDP forces in 1996. Patrick Cockburn,
‘Kurdish Chief’s Death Brings Civil War Nearer’, The Independent, 6 July 1996
<https://www.independent.co.uk/news/world/kurdish-chiefs-death-brings-civil-
war-nearer-1327460.html>.
280
Leezenberg Urbanization, Privatization, and Patronage, 167.
151
involve no loss of power or status for the old Kurdish elites, it actually
strengthened them. The replacement of the overpowerful Ba’ath regime
with the quarrelsome Kurdish parties gave them more leverage and
autonomy. The result was the creation of a new elite composed of former
jash and peshmerga commanders turned politicians both – although to a
different degree – tied to the political parties and whose military power gave
them the possibility to benefit from the economic opportunities offered by
the Iraqi Kurdish context of the 1990s. As some of these personalities, as
collaborators of Saddam Hussein, had also been involved in the mass killing
of Kurdish civilians in the late 1980s, they desperately needed the protection
of the KDP and PUK, as in the case of Qasim Aghay Koya who was involved
in the Anfal campaign and who joined the KDP after 1991.281

The integration of these personalities impacted on the ideological positions

of the two dominant parties which lost any trace of difference. The PUK,
given its defined leftist orientation, was to be more deeply changed by this
process than the tribal and traditionalist KDP. During the liberation war, the
PUK was still a federation of nominally independent parties. In 1992, the
three parties decided to merge into the same organisation to run for the
regional elections. Jalal Jawhar, at the time, one of the rising cadres in
Komala, the most organised and ideologically defined component of the
PUK, remembers his and his comrades' reluctance to accept Talabani’s
leadership of the new unified PUK:

we suggested to change the name of the party but not the beliefs
and structure of the party. […] At the time, I had the belief that we
needed to change our direction because the Soviet Union was
collapsing. We were willing to renew our ideology. […] but in terms
of leadership, Komala was different […] shaykhs or [tribal leaders]
did not exist in our party.282
Locked within a wider and more fluid structure centred around Talabani’s
personality and network, the more-ideologically defined Komala activists

Interview with Kamal Chomani (personal communication, 2019). See also Rebwar
281

Karim Wali, ‘Qasim Agha, Other Collaborators of Saddam Hussein’, Ekurd Daily, 31
October, 2010 https://ekurd.net/mismas/articles/misc2010/10/state4329.htm
282
Interview with Jalal Jawhar (Sulaymaniyah, 2019).
152
lost their political weight. At its founding congress in 1992, the PUK
abandoned its radical socialist positions and announced a ‘social-
democratic turn’, gradually assuming the ideological shallowness that was
typical of the KDP.283 In its first manifesto as a unified party, the PUK erased
any reference against tribalism and the traditional structures of Kurdish
society.284 As KDP Hoshyar Zebari could proudly claim, the KDP and PUK
were by then as different as “Pepsi and Coca-Cola.”285 With the primary aim
of coopting the old elite, the 1992 electoral campaign was centred around
the leaders’ personalities with little or no difference in the parties’
programmes.286 However, thanks to this re-branding of the party, the PUK
was able to co-opt powerful personalities such as Shaykh Muhammad
Kasnazani, a Sufi leader and long-term Saddam collaborator who in the
1970s was even using his pro-government militias against Talabani’s
forces.287

As discussed before, the flow of foreign aid to the Kurdish region in the
1990s offered essential emergency relief but also negatively impacted the
economic structure of the region. The UN and other organisations could not
deal directly with the KRG institutions but they did hire local staff and had
to rely on local NGOs for the delivery of aid and on local construction
companies to build essential infrastructure. Given the level of militarization
of the region, local warlords – both members of the old elite and peshmerga
commanders – and the two dominant political parties could easily take
control of the process. The KDP and PUK “set up, or gained control over,
numerous local NGOs, which were quickly perceived as lucrative sources of

283
Hoff and others, Elections in Iraqi Kurdistan, 18.
284
Patriotic Union of Kurdistan (PUK), ‘The Program & Political Platform of the
Adopted in the First PUK Congress 1992’, PUKmedia, 2015
<https://www.pukmedia.com/EN/EN_Direje.aspx?Jimare=34632>.
285
Interview with Hoshyar Zebari (Pirmam, 2018).
286
Hoff and others, Elections in Iraqi Kurdistan, 22.
287
Martin van Bruinessen ‘Tribes and Ethnic Identity’, in The Kurds in a Changing
Middle East, ed. by Jabar and Mansour, 157.
153
income, and as powerful instruments of clientelization.”288 On the one hand,
the politicisation of aid delivery strengthened the ongoing process of class
formation by increasing the opportunities for enrichment for the local elite.
On the other hand, as the political affiliation of NGOs, evident to the public,
reinforced the idea that the provision of essential services was the task of
the political parties and not of public institutions like the KRG, crystallising a
pattern of party-state relations that was already familiar to the people who
had lived under the Ba’athist regime. Within these relations of power, the
large amount of aid brought by the oil-for-food programme from 1996
onwards multiplied the opportunities for private enrichment triggering “an
exponential increase in personal wealth for local entrepreneurs” holding the
contracts for the sale of crude and the delivery of food.289

Stimulated by the international sanctions and the domestic blockade, illegal

trade constituted the other lucrative economic activity of the region.
Smuggling took place on a vast scale on all borders of the Kurdish region
and became the biggest source of revenue for the political parties and a
source of enrichment for the dominant personalities within them. The
Ibrahim Khalil border crossing with Turkey near Dohuk, mentioned earlier in
this chapter, was by far the most lucrative with an estimated annual $750
million revenues from illicit trade,290 and its control by the KDP contributed
to the outbreak of the civil war. The embargo stimulated the smuggling of
all sorts of products into the region and created a “privatised oil trafficking”
with Iraqi oil illegally sold in Turkey based on the cooperation of high-ranked
KDP and Ba’athist officers in violation of both the international sanctions and
the internal blockade.291 Moreover, the dire conditions of the civilian
population created a market for the smuggling of passports, exit visas, and

288
Leezenberg, Humanitarian Aid, 40.
289
Natali, The Kurdish Quasi-State, 57-58.
290
Ibid., 44.
Michiel Leezenberg, ‘Iraqi Kurdistan: Contours of a Post-Civil War Society’, Third
291

World Quarterly, 26.4–5 (2005), 637-638. Also cited in Kenneth Pollack, The
Threatening Storm: What Every American Needs to Know Before an Invasion in Iraq
(New York: Random House, 2002), 81.
154
actual people who were asked to pay up to $5000 to cross the Turkish
border.292

The lack of stable revenues, combined with the fact that the peshmerga
forces continued to respond to the command structures of the political
parties, undermined the autonomy of the KRG institutions making them
somewhat redundant. As the economic and military sources of power
remained in the hands of the political parties and their allies, the competition
between the KDP and PUK for the co-optation of the old elite and that for
the control of territory and border crossings was bound to create conflicts
that ultimately escalated into the civil war of 1994-1997. With the failure of
the PUK offensive of 1997, the war front stalled along a line that roughly
divided the governorates of Erbil and Sulaymaniyah. The consolidation of
the dividing line coincided with the exponential growth in foreign aid brought
by the oil-for-food programme that gave the local elite greater opportunities
for accumulation and contributed to the end of hostilities.293 Given the
importance of territorial control for these strategies of accumulation, it is not
surprising that the Washington Agreement of 1998, which included the
commitment to re-unify the KRG, remained a dead letter.

According to Gareth Stansfield, the two Kurdish statelets – each dominated

by one of the political parties – that emerged from the civil war, represented
a more efficient and stable system than the previously unified KRG. 294 The
50:50 arrangement of 1992 had pushed the two parties to use the KRG
offices against each other and it was precisely the perceived change in the
balance of power that pushed the PUK to break the agreement in 1994. This
argument is also based on the assumption that the post-civil war single-
party statelets were more suitable to the context than a multi-party system,
given the legacy of Ba’athist rule that had made the population accustomed
to the identification of the state with a political party. However, what
Stansfield downplays is that the biggest incentive to maintain the separate

292
Leezenberg, Iraqi Kurdistan, 638.
293
Ibid., 639.
294
Stansfield, Iraqi Kurdistan, 6.
155
administrations was the opportunities for accumulation that that system
offered. Unchecked by their rival, each political party was able to
consolidate the grip over its respective sub-region especially when the start
of the oil-for-food programme promised unprecedented profit. Under the
tight military control of their peshmerga forces, the KDP and PUK created a
system ruled by the party politburos and for the benefit of their dominant
families and political and economic allies. Unlike the context of a limited
political plurality of the early 1990s, the new single-party statelets were also
far more capable to repress any form of dissent and political opposition.

This process of accumulation had long-term effects on the structuring of

class relations in the Kurdish region. In a region devastated by violence and
dependent upon humanitarian aid, the Chamber of Commerce of
Sulaymaniyah could boast about a thousand millionaires.295 At the same
time, over half of Kurdish families survived with an income of $25 a month
supplemented by a food basket worth $50 a month funded by the oil-for-
food programme.296 Deprived Iraqi Kurds, made dependent on international
aid by the destruction of agriculture and other productive activities, were
forced into the condition of clients of a burgeoning patronage system.

The 2003 Invasion of Iraq and the unification of the KRG

The system of governance of Iraqi Kurdistan established in the 1990s and

based on two separate single-party administrations came to a sudden end
with the US-led invasion of Iraq of 2003. The fall of the Ba’athist regime
forced the KDP and the PUK to re-establish a unified KRG framework now
finally recognized by Baghdad. However, the class structure developed in
the 1990s was strengthened by this new institutional arrangement.

The participation of Kurdish forces in the invasion of the country was a

condition for the KDP and PUK to secure the recognition of Kurdish
autonomy by the Americans and the Arab forces in post-Ba’athist Iraq.

295
Cited by Natali, The Krudish Quasi-State, 100.
296
Leezenberg, Iraqi Kurdistan, 640.
156
Months before the invasion, in October 2002, the Kurdish parliament in Erbil
was convened – for the first time since 1994 – with the presence of both
KDP and PUK members. In February 2003, the two parties reopened each
other’s offices in the respective areas.297 The KDP and PUK then formed the
Kurdistan Alliance and, in January 2005, ran together in the first post-war
Iraqi parliamentary elections. The widespread boycott of the elections in the
Arab Sunni areas ravaged by the anti-occupation uprising left the country’s
Arab Shi’a majority and the Kurds in a strong position to determine the future
of the country. The PUK leader Jalal Talabani was elected President of Iraq,
starting an unwritten norm that would assign the presidency to the Kurds.
The country’s new constitution was drafted in a very short time and
approved by a referendum in October 2005. The constitution recognized
Kurdish as the second official language of Iraq and the governorates of Erbil,
Dohuk and Sulaymaniyah as a federal Kurdish region with its parliament and
government.298 However, the rush to approve a constitution left several
issues unsolved. Kirkuk and other mixed Arab-Kurdish areas remained
under the central government waiting for a referendum to be held by 2007.
The referendum never took place and the issue of the disputed territories
kept haunting the relationship between Baghdad and Erbil.

The elections for the Kurdish Parliament – the first since 1992 – were held in
conjunction with the federal elections of January 2005 and saw the KDP-
PUK alliance winning almost 90 per cent of the votes. This triumph was a
secure starting point for a new power-sharing agreement. Masoud Barzani
was elected President of the Kurdish region and, in January 2006, the two
parties signed the Unification Agreement meant to restore the unified
Kurdish administration. The agreement, signed by Barzani and Talabani in
their role as party leaders, explicitly detailed the distribution of the major
political offices between the two parties including a plan of for the following

297
Bengio, The Kurds of Iraq, 276.
‘Iraq’s Constitution of 2005’, Constitute Project,
298

<https://www.constituteproject.org/constitution/Iraq_2005.pdf?lang=en>
157
parliamentary term.299 Forced by the new conditions to find a synthesis, the
two leaders established a permanent line of communication to run the
region and to confirm their parties’ duopoly over the new institutions. As
Kamran Karadaghi – then the spokesperson for President Talabani – recalls:

When Talabani became president of Iraq his attitude changed, that

was his ultimate aim. Talabani recognized that Barzani was more
suitable to become president [of the KRG]. This created a trust
between the two and I witnessed it. They solved things together.300
The 2006 agreement marked the beginning of a new system, similar to the
50:50 power-sharing deal of 1992, in which all the KRG institutions were
monopolized by the two major parties. Compared to the 1992 attempt, the
new system had a stronger territorial character since the two parties had
been in control of the respective areas – the KDP in Erbil and Dohuk and the
PUK in Sulaymaniyah – for over a decade. This territorial dimension was
strengthened by the organisation of the region’s armed forces that were
never unified and remained subject to the command structures of the
respective political parties. The KDP-controlled Yellow Zone and the PUK-
controlled Green Zone continued to be divided by an actual border with
checkpoints flying the respective party flags. The Ministry of Peshmerga
Affairs had no real control over the peshmerga making these military units
effectively party militias.301 The fact that the loyalty of the peshmerga forces
went primarily to the political parties – or even personalities within the
parties302 – rather than to the region’s institutions, constituted a major

299
‘Kurdistan Regional Government Unification Agreement’, Kurdistan Regional
Government, January 23, 2006
<http://cabinet.gov.krd/a/d.aspx?r=223&l=12&a=8891&s=02010100&s=010000>.
300
Interview with Kamran Karadaghi (London, 2018).
301
Wladimir van Wilgenburg and Mario Fumerton, Kurdistan’s Political Army: The
Challenge of Unifying the Peshmerga Forces (Carnegie Middle East Center, 2015)
<https://carnegieendowment.org/files/ACMR_WilgenburgFumerton_Kurdistan_En
glish_final.pdf>, 1.
302
“PUK leader Kosrat Rasul Ali, for example, has his own protection brigade called
Hezekani Kosrat Rasul, which is made up of between 2,000 and 3,000 peshmerga
fighters. Similarly, the PUK’s Bafel Talabani (one of Jalal Talabani’s sons)
commands his own antiterror force that is not controlled by any ministry. The KDP’s
Nechirvan Barzani, the prime minister of the KRG, also has his own personal
security force […]. In addition to these assorted units, there are two PUK
158
obstacle to the establishment of an effective rule of law in the KRG. As all
the opposition parties claim “as long as they [PUK and KDP] control the
peshmerga, after the elections, they can do what they want.”.303 Moreover,
this system represented an extraordinary source of patronage for the
dominant parties which, by distributing stipends and pensions to the
peshmerga and their families, secured the loyalty of hundreds of thousands
of Kurds.304

However, the partisan use of the KRG institutions was not the only
consequence of the ruling parties’ control over the military. As Hoshyar
Omar claims “they can't control the economy - especially the oil - if they
don't have the military.”305 Even after the reunification of the KRG, the power
of the Kurdish elite kept resting on the control exercised by the two parties
over the region’s resources. The wider opportunities for accumulation in
post-2003 Iraqi Kurdistan reinforced the power and cohesion of the ruling
class that had emerged in the 1990s.

The Kurdish Economic Boom and Its Contradictions

In the years following the 2003 invasion of Iraq, the Kurdish region
experienced a period of rapid economic growth. The end of the blockade
and sanctions that suffocated the Kurdish economy in the 1990s allowed
the local elite to open up the Kurdish economy which led to a massive flow
of international investment. The KRG, ruled by the KDP-PUK duopoly,
particularly benefitted from a high level of political stability in comparison to

presidential peshmerga brigades personally controlled by Jalal Talabani or his wife,

Hero Ibrahim Ahmed”. Van Wilgenburg and Fumerton, Kurdistan's Political Army, 5.
303
Interview with Aram Saeed (Sulaymaniyah, 2018).
304
The peshmerga were said to be 120.000 in 2014. However, all the security
forces – including the police and the internal and external intelligence agencies –
are controlled by the KDP and PUK, making the number of Kurds working in the
security sector on behalf of the political parties far bigger than that. Van
Wilgenburg and Fumerton, Kurdistan's Political Army, 3.
305
Interview with Hoshyar Omar (Sulaymaniyah, 2018).
159
the rest of Iraq that was ravaged by the insurgency against the American
occupation and the spread of sectarian violence.

Given the recognition of Kurdish autonomy by the 2005 constitution, the

KRG could finally count on a relatively stable source of revenue. Based on
the estimate of the population of the three Kurdish governorates (3.9 million
in 2003),306 the regional government was entitled to 17 per cent of Iraq’s
budget almost exclusively derived from the export of oil. In the period
between 2006 and 2014, the budget transfer from Baghdad amounted to
roughly 80 per cent of the KRG revenues,307 while revenues from taxes never
reached 5 per cent of the total.308 This flow of revenues combined with the
political stability of the region allowed for a tumultuous economic growth.
Despite the absence of reliable GDP data separate from that of Iraq, the
economy of the KRG was estimated to have grown from $18 billion in 2008
to $27 billion in 2012.309

Rather than using these resources to rebuild the Kurdish economy through
the development of its productive sectors, the Kurdish elite adopted a
model of development based on the oil-producing monarchies of the

306
Regional Development Strategy for Kurdistan Region 2012-2016 (KRG -
Ministry of Planning, 2011) <http://www.mop.gov.krd/resources/MoP Files/PDF
Files/gd_ps/regional_development_strategy.pdf>, 55.
The Kurdistan Region of Iraq: Assessing the Economic and Social Impact of the
307

Syrian Conflict and ISIS (World Bank Group, 2015)

<https://doi.org/10.1192/bjp.111.479.1009-a>, 5.
The Kurdistan Region of Iraq: Reforming the Economy for Shared Prosperity
308

and Protecting the Vulnerable (World Bank Group, 2016)

<http://documents.worldbank.org/curated/en/672671468196766598/Main-
report>, XI.
309
‘Kurdistan Region Facts & Figures’, Invest in Group (IIG), 2019
<https://investingroup.org/country/kurdistan/facts-figures/>. The Rand
Corportation estimated the Gross Regional Product (GRP) excluding natural
resources because of the lack of reliable data. According to their report, the GRP
of the KRG amounted to 20,954 billion Iraqi Dinars in 2008 and to 27,381 billion
Iraqi dinars in 2012. Calculating the Gross Regional Product of the Kurdistan
Region - Iraq (KRG and Rand Corporation, 2016)
<https://doi.org/10.7249/rr1405.3>, xii, 40.
160
Persian Gulf.310 In July 2006, just a few months after the reunification of the
KRG, the Kurdish Parliament approved a very generous investment law and
established a Board of Investment. The law, defined by the KRG Foreign
Minister Falah Mustafa Bakir as “amongst the most investor-friendly laws in
the wider region”,311 is largely modelled after the ones in place in the Gulf
countries but it allows an even higher degree of freedom and capital returns
to foreign investors who are allowed to repatriate their profits in full, to buy
land, and enjoy a 10-year non-custom tax break.312 The KRG encouraged
partnership with local businessmen and companies and the Board of
Investments has a high degree of discretion in issuing the licences largely
favouring economic actors affiliated with the KDP and PUK.313 Moreover, the
KRG established a visa regime that circumvents Iraqi immigration law,
allowing most foreigners to obtain an entry visa upon arrival.

These policies paid off. Between 2006 and 2012, the Kurdish region
received $22 billion in investment, 21 per cent of which were foreign direct
investment or joint ventures with local partners.314 In 2012, investment in the
KRG amounted to 55 per cent of total investment in Iraq. 315 However, given
the extremely favourable fiscal conditions enjoyed by foreign corporations,

Nicky Woolf, ‘Is Kurdistan the next Dubai?’, The Guardian, 2010
310

<https://www.theguardian.com/commentisfree/2010/may/05/kurdistan-next-
dubai-iraq>;
311
‘“KRG’s Investment Law Is amongst the Most Investor Friendly Laws in the Wider
Region” – Minister Falah Mustafa Bakir’, The Kurdistan Tribune, 20 November 2011
<https://kurdistantribune.com/krgs-investment-law-amongst-most-investor-
friendly-laws-wider-region-minister-falah-mustafa-bakir/>.
‘Investment Law’, KRG Department of Foreign Relations, 2006
312

<https://dfr.gov.krd/p/p.aspx?p=69&l=12&s=050200&r=377>.
313
This dynamic was so widely recognised that “even Prime Minister [Nechirvan]
Barzani acknowledged to [US diplomats] in a 14 August breakfast meeting that
there was a perception by foreign investors that they had to partner with either
the KDP or the PUK a phenomenon he said he "can't say is not true."”.
‘Entrenched Corruption in Kurdistan Region of Iraq’, Baghdad Embassy, Wikileaks
Cables: 08BAGHDAD2731_a, 25 August 2008
<https://wikileaks.org/plusd/cables/08BAGHDAD2731_a.html>.
314
The Kurdistan Region of Iraq (2016), 6.
315
‘Kurdistan Region - Determined to Grow: Economy, Kurdistan Region, Review’,
Invest in Group (IIG), 2013 <https://investingroup.org/review/236/determined-to-
grow-economy-kurdistan/>.
161
the effect of this flow of capital on the KRG revenues has been negligible.316
Moreover, without substantial limitations on the hiring of foreign staff and in
combination with the lax visa regime, the KRG investment policy has
generated huge profit for the foreign companies and their local partners and
yet has had hardly any impact on the overall employment and the wider
population.317

The flow of foreign capital had its largest and most visible effect in the
construction boom which, in just a few years, transformed the skyline of the
main Kurdish cities. Private investments built large housing complexes,
residential compounds, luxury hotels, and huge shopping malls. In the city
of Erbil, the price for 250-square meter plot rose from $5,000 in 1996 to
$153,000 in 2010 and the cost of housing reached $423 per meter making
large areas of the cities unaffordable for most locals.318 The country that
benefitted the most from the KRG’s investment-friendly policy in the
construction sector was Turkey that in the mid- and late 2000s was
experiencing a tumultuous economic growth. In 2012, Turkish investment in
Kurdish real estate reached $4.3 billion and Turkish construction companies
played a central role in restoring essential infrastructure in the Kurdish
region.319

The end of the sanction regime as well as the following reduction or outright
cancellation of customs allowed for the recovery of international trade, even
though it did not eliminate smuggling. The Kurdish region of Iraq gained a
role as a transit area but, most significantly, as a market for foreign
products. The Kurdish unproductive rent-based economy became an
extraordinarily important market for Turkey’s burgeoning economy as well
as for Iranian manufacture and agricultural products. The territorial

316
The Kurdistan Region of Iraq (2016), 11.
317
Christina Bache, ‘Mutual Economic Interdependence or Economic Imbalance:
Turkish Private Sector Presence in the Kurdistan Region of Iraq’, Middle East
Critique, 27.1 (2018), 68.
318
Natali, The Kurdish Quasi-State, 99.
319
Kurdistan Region - Determined to Grow.
162
distribution of this trade partly overlapped with the political ties between
the two dominant Kurdish parties and the neighbouring countries, with a
predominance of Turkish products in the KDP-controlled area and a
significant role of Iranian exports in the area controlled by the PUK. Turkey,
however, had the lion’s share. Already in 2007, Turkish exports to the KRG
amounted to almost $1.5 billion and constituted roughly 50 per cent of its
total exports to Iraq. In 2013, the total value of Turkish exports was worth
over $8 billion and constituted 67 per cent of its exports to Iraq. 320 This
tumultuous increase meant that in 2013 approximately 80 per cent of all
goods sold in the region were made in Turkey, making daily life in the KRG
completely dependent on its political relationship with Turkey and the state
of the Turkish economy.321

These trade and investment policies had the most damaging effects on the
productive sectors of the Kurdish economy. Cheap products coming from
Turkey, Iran, and East Asia flooded the Kurdish markets undermining the
recovery of local production. In the first phase of the Kurdish economic
boom between 2006 and 2010, industry and agriculture received a low
share of the capital invested in the region, respectively 12.08 per cent and
1.89 per cent.322 The result of this paucity of investment is clearly shown by
the insignificance of these sectors in the Kurdish economy. According to
data produced by the Rand Corporation, in 2012, industry amounted to 6.1
per cent of the non-oil value-added of the KRG, while agriculture – once the
driving sector of the Kurdish economy – to a staggering low of 3.1 per
cent.323 Even if the rainfall was mostly in or above the average in the

320
Bache, Mutual Economic Interdependence, 71.
321
Kurdistan Region - Determined to Grow.
322
Regional Development Strategy, 41.
323
Rand’s estimates exclude the natural-resources sector which means that the
actual size of industry and agriculture in the whole of the Kurdish economy is even
less significant than what these figures suggest. Calculating the Gross Regional
Product of the Kurdistan Region, xiii.
163
2000s,324 the lack of investments and the flow of cheaper products from
abroad hit hard. Between 2000 and 2007, agricultural production nearly
halved and the number of Kurds who lived off agriculture dropped from 35
to 23 per cent.325

The Promises of Kurdish Crude

Since the reunification of the KRG in 2005, oil was the sector that most
attracted attention from foreign and local actors. Besides the interests that
large oil reserves normally mobilise, oil has, in this case, the particular
significance of providing the conditions of viability for an independent
Kurdish state but also to immensely strengthen the autonomy of the ruling
elite vis-à-vis the rest of Kurdish society.

In 2006, the oil sector in the Kurdish region was extremely underdeveloped
due to the internal blockade and the international sanctions over Iraq. The
2005 Iraqi constitution states that the country’s oil and gas reserves belong
to the Iraqi people as a whole and assigns their management to the federal
government in exchange for the equal distribution of its revenues – the 17
per cent share of the KRG – but only explicitly refers to “the present
fields.”326 Estimates that were available when the constitution was being
drafted and that was confirmed by geological surveys described the Kurdish
region of Iraq as sitting on a massive unexploited 41 billion barrels of oil and
1.5 trillion cubic meters of natural gas.327 Despite the vastity of these oil and

324
Drought Characteristics and Management in North Africa and the Near East
(Food and Agriculture Organization, 2018)
<http://www.fao.org/3/CA0034EN/ca0034en.pdf>.
325
Natali, The Kurdish Quasi-State, 96; Regional Development Strategy, 78.
326
See Iraq’s Consitution of 2005.
327
Robin Mills, ‘Under the Mountains: Kurdish Oil and Regional Politics’, The Oxford
Institute for Energy Studies, OIES Paper WPM 63 (January 2016)
<https://www.oxfordenergy.org/wpcms/wp-content/uploads/2016/02/Kurdish-
Oil-and-Regional-Politics-WPM-63.pdf>, 16-17. In 2017, proven reserves for the
whole of Iraq amounted to 148 billion barrels of oil and 3.5 trillion cubic meters of
natural gas. BP Statistic Review of World Energy – June 2018, 67th Edition (BP,
2018) <https://www.bp.com/content/dam/bp/business-
164
gas resources, in 2006 the Kurdish region was still importing all of its
electricity.328

It is not surprising then that the development of the oil and gas sector
became the top priority for the local political elite. Kurdish policy in this field
has been largely dominated by the personality of Ashti Hawrami. Appointed
Minister for Natural Resources by the KDP in 2006 and – reconfirmed in
every cabinet until 2019 – Hawrami promoted an extremely investor-friendly
policy determined to quickly develop a Kurdish independent oil sector even
at the cost of confrontation with Baghdad. In 2007, the Kurdish parliament
approved an oil and gas law that authorised the regional government to sign
independent contracts for exploration and production, despite strong
opposition from Baghdad. The new law accorded singularly favourable
conditions to foreign companies. Unlike the rest of Iraq, where oil companies
are treated merely as contractors and are normally entitled to $1-2 per
barrel, the KRG granted to private companies co-ownership of the field in
exchange of a royalty of 10 per cent of their gross income. Attracted by
these favourable conditions, several medium-sized corporations started
operating in the region between 2007 and 2011.329

These early operations convinced larger actors to move in. In November

2011, the KRG assigned exploration blocks to the giant ExxonMobil and, in
2012, to the Emirati TAQA, the French Total, the American Chevron, and the
Russian Gazprom Neft.330 Strengthened by the geopolitical significance of
these partnerships, the KRG started exporting oil independently via trucks
through Turkey and Iran, rather than using the Iraqi pipeline infrastructure.
In the meantime, the KRG started building its own pipeline to Turkey that
became operative in late 2013. The arrival of these oil giants meant that the

sites/en/global/corporate/pdfs/energy-economics/statistical-review/bp-stats-
review-2018-full-report.pdf>.
328
Kurdistan Region - Determined to Grow.
329
Gülistan Gürbey and Caner Yildirim, ‘Perspectives of an Independent Energy
Export Policy of Iraqi Kurdistan’, in Iraqi Kurdistan’s Statehood Aspirations, ed. by
Anaid and Tugdar, 62-63.
330
Mills, Under the Mountains, 11.
165
central government could no longer downplay the Kurdish energy policy and
inevitably perceived it as a threat to the country’s territorial integrity. In
response, Baghdad began international legal and diplomatic action against
the KRG. As oil started flowing to Turkey through the new Kurdish pipeline
the Iraqi government decided to retaliate and, on February 2014, blocked all
its budget transfer to the KRG.331

Baghdad’s strong opposition to the Kurdish oil policy was motivated by both
domestic and international concerns. Domestically, the new contracts
signed by the KRG undermined the central government’s control over the
most important economic and geopolitical asset of the country. Erbil’s
moves were observed with interest by other oil-producing governorates
such as Nineveh and Basra, raising the spectre of a dangerous
fragmentation of the country and putting at risk its long-term viability as a
unitary state.332 This process would undermine the redistributive capabilities
of the central government creating huge disparities between oil-producing
and non-oil-producing governorates. Moreover, the decentralisation of
Iraq’s oil policy would inevitably create local power centres significantly
more susceptible to foreign interference. Turkey’s strong support for the oil
policy of the KRG conveyed a clear message in this direction to the Iraqi
government. Turkey is a neighbouring country with a long history of
interference in northern Iraq and the development of the Kurdish oil sector
took place in a context of growing economic influence and warming ties
between Ankara and the KDP. The new KRG pipeline has strengthened this
relationship, tying the Kurdish region even more strongly to the growing and
energy-thirsty economy of Turkey. In the virtual absence of a Kurdish
banking sector, the revenue from the export of oil was transferred to the

331
Ibid., 12.
332
Ibid., 36.
166
KRG through the Turkish state-owned Halk Bankası, de facto putting Ankara
in control of the process.333

Political independence through oil has been the project consistently

pursued by the elite of the KDP in charge of the sector, even more
systematically, since Nechirwan Barzani, nephew of Masoud, took office as
KRG prime minister in 2012. The development of the oil sector and the
growing ties with Turkey were pursued with the hardly-hidden intention of
building the conditions – independent revenues and geopolitical support –
to full autonomy or even independence from Iraq. However, this project is
not primarily about Kurdish self-determination as it is often presented by
both local politicians and foreign experts.334 The deep relationship between
oil and political power will be more extensively discussed in Chapter 6, as it
will become more evident with the post-2014 events. However, in the early
2010s, oil had already become an extremely lucrative business for powerful
Kurds. Moreover, more economic and geopolitical autonomy from Baghdad
meant even less accountability and more opportunity for accumulation for
the Kurdish elite over the territory that they militarily control. Since its very
beginning, the Kurdish oil policy has been, first of all, an instrument of class
power that accelerated the transformation towards a rentier class structure
initiated in the 1990s.

Subaltern Classes and Opposition in a Rentier (Quasi-)State

The economic growth of the Iraqi Kurdish region after 2003 took place
within the set of the class relations established in the 1990s and discussed
earlier in this chapter. Despite the unification of the KRG institutions in 2006,
the KDP and PUK retained territorial control over the respective areas
through the party-affiliated security forces. This allowed the political elite to

333
Ali Ünal, ‘Turkey’s Halkbank Collects the Revenue of Kurdish Oil’, Daily Sabah,
24 June 2014 <https://www.dailysabah.com/energy/2014/06/24/turkeys-
halkbank-collects-the-revenue-of-kurdish-oil>.
334
See, for example: Yaniv Voller, ‘Kurdish Oil Politics in Iraq: Contested
Sovereignty and Unilateralism’, Middle East Policy, XX.1 (2013), 68–82.
167
maintain nearly absolute control over the economy and the increased
opportunities for accumulation offered by the oil rent after 2006.

Even in the 1990s, the revenues of the local institutions and political parties
as well as the sources of private accumulation came largely via streams of
rent, namely the fees generated by the border-crossings and the
management of foreign humanitarian aid. In this context, the
institutionalisation of the KRG in the 2000s contributed to completing its
transition towards a rentier economy. Given its complete control of the KRG
institutions and military apparatus, the ruling class formed in the 1990s was
in the position to siphon off a significant share of the revenues from the
regional budget. Rather than investing in productive activities, the Kurdish
ruling class followed a consistent strategy of accumulation based on the
appropriation of public wealth through corruption, misappropriation, and the
assignment of public contracts – namely in the construction sector – to
companies controlled by political leaders or their allies. This strategy was
enabled by the lack of transparency and the partisan control of the KRG
institutions. As mentioned before, looking at these practices in terms of
class relations sheds a different light on the role of the independent oil
policy promoted since 2007. Unlike the budget transfer from Baghdad, the
creation of revenues directly within the border of the KRG exponentially
increased the amount of money that could be directed to private pockets as
well as multiplied the ways through which this process could take place.
Maybe: Despite the impossibility to determine exact numbers, but
independent investigations locate the numbers in the order of billions of
dollars.335

This predatory strategy of accumulation went hand in hand with the

redistribution of a large part of the KRG budget among the Iraqi Kurdish
population in the form of civil servants’ salaries, pensions and subsidies. Due

335
‘Over $1 Billion from Iraqi Kurdistan Oil Revenue Missing in 1st Quarter of 2016:
MP’, Ekurd Daily, 7 April 2016 <https://ekurd.net/billion-iraqi-kurdistan-oil-missing-
2016-04-07>; ‘Iraqi Kurdistan Oil Revenue $1 Billion per Month: Source’, Ekurd
Daily, 2 November 2016 <https://ekurd.net/kurdistan-oil-revenue-billion-2016-11-
02>.
168
to the weakness or sheer absence of a productive economy, most of the
population was made dependent on government handouts, strengthening
the patronage nature of the system and increasing the cost for public
expressions of political dissent.

In 2013, more than 50 per cent of the KRG budget was spent on salaries for
public servants and pensions and roughly 1.4 million Iraqi Kurds were on the
regional government payroll out of a population of just over 5 million.336
Between 2007 and 2012, more than 80 per cent of the 750.000 new jobs
created in the KRG were in the public sector that, in 2014, employed 53 per
cent of the region’s working population.337 A significant proportion of public
employees earned $150-200 per month338 and, in many cases, an entire
family relied on one salary making it completely dependent to the local KDP
or PUK powerholders in control of public employment. Through this system,
a large part of the population was forced into the patronage networks of the
two parties and their leading politicians. Even if one salary barely kept a
family above the poverty line, public employment remained a far better
option than the uncertainty of the private sector. According to official – likely
conservative – estimates, unemployment stood at almost 18 per cent in
2009.339 Moreover, working conditions in the private sector were far from
inviting: according to the World Bank, in 2012, a great majority of the
employed poor population worked for private employers.340 Kurdish
businessmen in construction, agriculture and in the service sector
increasingly relied on migrant labour from Asia and Africa but often also on
seasonal workers from Iranian Kurdistan and the rest of Iraq.

336
The Kurdistan Region of Iraq (2016), CXI.
337
Ibid., 6.
338
Anwar Anaid, 'The Nature of Political Economy Challenges of the Kurdistan
Region of Iraq', in Iraqi Kurdistan’s Statehood Aspirations, ed. by Anaid and Tugdar,
25.
339
Regional Development Strategy for Kurdistan Region, 16.
340
“73% of employed urban poor (primarily construction, commerce and retail,
transport, storage and communication) and 78% of employed rural poor (primarily
agriculture and construction) worked in the private sector”. The Kurdistan Region
of Iraq (2016), 6.
169
The KDP and PUK were able to limit the expression of public dissent towards
this system thanks to their control of the security apparatus and the
dependence created through public employment and other limited forms of
redistribution. However, the uneven nature of the KRG development and
extremely unequal distribution of power and wealth in the region was bound
to generate opposition. The first significant expression of political
opposition to the KDP-PUK duopoly came in the form of the rise of Islamist
forces. The Kurdistan Islamic Group and the Muslim Brotherhood-affiliated
Kurdistan Islamic Union combined the critique of corruption and nepotism
with that of secularism and, until the late 2000s, were the strongest
opposition forces in the Kurdish region and particularly in the rural areas
where the control by the KDP and PUK was less tight.341

However, the most significant challenge to the ruling parties came from
within their ranks. In 2009, Nawshirwan Mustafa, one of the most popular
veterans of the liberation struggle and the leader of the left-wing of the PUK,
split from Talabani’s party and established the Gorran (‘change’) Movement.
Campaigning against corruption and for political liberalisation, Gorran
boomed in the July 2009 elections, winning 23 per cent of the votes and
beating the PUK in Sulaymaniyah. The rise of Gorran has shaken the Kurdish
political establishment and generated widespread enthusiasm among the
public but it has also shown the democratic limits of the KRG system. As
journalist Kemal Chomani explains, even if Gorran won the majority of the
Sulaymaniyah council in 2013, they had to reach an agreement with the PUK
as they soon realised that controlling the local administration was in vain as
long as the security apparatus, the civil servants, and most economic
activities kept responding to the command structure of the PUK.342

Although Gorran’s electoral success did not substantially change the

relations of power within the regional politics, it was certainly the sign of a
wider erosion of the political legitimacy of the local political elite. In February

341
See Bengio, The Kurds of Iraq, 210-220, 277.
342
Interview with Kamal Chomani (personal communication, 2019).
170
2011, encouraged by the popular revolts that were shaking the Middle East,
thousands gathered in Sulaymaniyah to protest corruption and demand
political reform.343 Despite the violent reaction of the local security forces,
protests continue for the following two months and five demonstrators were
killed and 150 injured. The repression of the protests was accompanied by
the action of masked armed men who attacked the demonstrations as well
as the offices of the opposition parties and independent media stations.344
Major protests in Erbil and Dohuk were avoided by preemptive actions by
the KDP and by the exemplary brutality of the response in Sulaymaniyah.
The 2011 protests were a watershed moment in the history of the region by
significantly undermining the relationship between the ruling parties and
large chunks of the population. The images of Kurdish security forces
shooting peaceful protests made many realise that the rule of fellow Kurds
could be as oppressive as that of any non-Kurdish dictator.

Conclusion

The establishment of a unified KRG, within the framework of the 2005 Iraqi
Constitution, crystallised a set of class relations in the Kurdish region that
had taken shape in the 1990s. After the liberation of the region in 1991, the
KDP and PUK competed for the co-optation of the old Kurdish elite. This
process led to the creation of a new ruling class, born from the alliance
between the KDP and PUK commanders turned politicians and the local elite
composed of former Ba’athist collaborators, such as tribal leaders, regime-
backed entrepreneurs, and former commanders of the counter-insurgency

343
Marina Ottaway and Anas Danial Kaysi, ‘Iraq: Protest, Democracy, and
Autocracy’, Canregie Endowment for International Peace, 28 March 2011
<https://carnegieendowment.org/2011/03/28/iraq-protest-democracy-and-
autocracy-pub-43306>.
344
‘Iraqi Kurdistan: Prevent Attacks on Protesters’, Human Rights Watch, 7 March
2011 <https://www.hrw.org/news/2011/03/07/iraqi-kurdistan-prevent-attacks-
protesters>; ‘Criminal Raid Prevents Kurdistan’s First Independent TV Station from
Covering Sulaymaniyah Unrest’, Reporters Without Borders, 22 February 2011
<https://rsf.org/en/news/criminal-raid-prevents-kurdistans-first-independent-tv-
station-covering-sulaymaniyah-unrest>.
171
jash battalions. Through the control of the security forces, this new ruling
class was able to appropriate large parts of the wealth created by various
sources of rent, particularly the export of oil after 2005.

As this chapter argued, the inability of the two Kurdish parties to establish
functioning institutions and a competitive democracy can be explained by
the strategy of accumulation employed by a ruling class largely
characterised by its military background – either in the Kurdish insurgency
or in the former pro-government forces. The direct control of the armed
forces – which were never allowed to become a politically neutral regional
army – has enabled the Kurdish elite to appropriate the wealth created by
different forms of rent: smuggling and foreign humanitarian aid in the 1990s
and the sale of oil after 2005. In turn, the reliance of the Kurdish subaltern
classes on different forms of public handouts is not the result of the lack of
a work ethic or a culture of dependence but rather the only strategy of
subsistence available in that context. The Kurdish ruling class showed no
interest in investing in the development of a productive economy. By
distributing parts of the region’s wealth through public salaries, pensions,
and subsidies, the political elite forced a large part of the population into
their patronage network, increasing the cost of political dissent. This system
gradually led to the erosion of the legitimacy of the KDP and PUK but the
combination of military coercion and economic dependence allowed the
ruling class to survive the rise of political opposition in the late 2000s.

172
Chapter 6
The Crisis of Kurdish Nationalism in
Iraq (2014-2019)

Introduction

The structure of power and class hierarchy that developed in the Kurdish
region of Iraq in the 1990s and 2000s were deeply shaken by the crisis that
started in 2014. Under attack by ISIL and with the oil price falling to record
lows, the KRG suddenly became unable to pay public salaries throwing
hundreds of thousands of Kurds to the edges of poverty. The austerity
policies implemented by the KRG broke the mechanism of redistribution on
which the Kurdish rentier system was based. To counter mounting popular
rage – manifested in frequent and tense public protests – the ruling KDP and
PUK deployed a combination of coercion and persuasion. On the one hand,
force was the most immediate response. While the protestors were met with
fire by the security forces, the government shut down the parliament for two
years (2015-2017) and illegally extended Masoud Barzani’s presidential
term. On the other hand, the Kurdish elite invested in aggressive nationalist
rhetoric that led to the decision of holding an independence referendum, in
September 2017, and to the heavy retaliation from Baghdad and
neighbouring powers. The combination of these two strategies within the
context of the KRG political and economic crisis reveals the class nature of
Iraqi Kurdish nationalism and of the separatist project pursued by the
political elite. Among the Kurdish subaltern classes, who had to pay the
highest price of the economic crisis, the frustrated aspirations to political
change gave way to a widespread sense of disillusion and a crisis of Kurdish
nationalism as a source of legitimacy for political power in the region.
Besides reconstructing the events surrounding the crisis and the
173
referendum, this chapter also delves into the evolution of the power
structure of the region in the 2010s and the growing importance of
neighbouring powers. The function of Kurdish nationalism in Iraq is thus
revealed as upholding and legitimising the political and class hierarchies of
the region. However, the more the nationalist credentials of the ruling elite
weakened, the more they had to rely on coercive means to stay in power.

War and Economic Crisis

When the political and economic crisis hit the KRG in 2014, the Kurdish
region of Iraq was experiencing a period of rapid growth. Per-capita wealth
was almost twice as high as the national average and, despite rapidly
growing inequalities, only 3.4 per cent of Kurds were living in acute poverty
compared to 13.3 per cent of Iraqis.345 Poverty remained largely
concentrated in the rural areas contributing to a process of continued
urbanisation.346 This period of growth consolidated a set of class relations
that had started forming in the early 1990s when the KDP and the PUK took
control of the region. As we saw in Chapter 5, the party leaders and their
affiliates were able to appropriate most of the wealth generated by the
export of oil and kept the rest of the population in a position of subordination
and dependent on public handouts. However, this system started crumbling
in 2014 due to a combination of political and economic events.

The rift between Erbil and Baghdad over the independent export of oil by
the KRG that Baghdad deemed unconstitutional escalated in February 2014
when the Iraqi government blocked all its budget transfer to the Kurdish
region. Compared to the IQD 14.3 trillion received in 2013 – 77 per cent of
the region’s total revenues – in 2014 the KRG only received IQD 1.1 trillion

345
Iraq Human Development Report 2014: Iraqi Youth Challenges and
Opportunities, (UNDP, 2014) <http://hdr.undp.org/sites/default/files/undp-
iq_iraqnhdr2014-english.pdf>, 128-129.
From 1997 to 2012, the urban population grew from 74.2 to 81.7 per cent in KRG
346

and only from 67.3 to 69.2 per cent in the rest of Iraq. Iraq Human Development
Report, 140, 145.
174
from Baghdad.347 Given its very narrow tax base, the KRG could rely only on
its independent oil export. As the Iraqi government threatened the KRG with
war, a military confrontation was likely avoided only due to the sudden turn
of events.348 In early June, the Islamic State of Iraq and the Levant (ISIL)
invaded northern Iraq and, while the Iraqi army collapsed, the Islamists took
control of Mosul and pushed southwards to Baghdad and eastwards to the
Kurdish region. ISIL, the evolution of al-Qaeda in Iraq, had thrived in the
context of the Syrian Civil War also thanks to its vast use of performative
brutality against non-Sunni civilians and prisoners of war that significantly
curtailed the morale of the overwhelmingly-Shia Iraqi army. While most of
ISIL’s pressure was directed towards Baghdad, the Kurdish peshmerga
managed to take control of long-disputed Kirkuk after the Iraqi Army
evacuated the city. However, when in August 2014 ISIL attacked the areas
of the Nineveh Governorate inhabited by the Yazidi Kurds, the KDP-
affiliated peshmerga withdrew. As ISIL initiated a genocidal campaign of
slaughtering and enslavement of this religious minority, 50.000 Yazidis
under siege on Mount Sinjar were only rescued by intervention from Syria
and Turkey of fighters affiliated with the PKK.349

The war on ISIL imposed a 1000-km long frontline on the KRG forcing Erbil
to significantly increase its defence spending. A flow of displaced people
from Syria and northern Iraq fled to the Kurdish region increasing the local
refugee population from the already-present 250.000 to 1.5 million by the

347
DeWeaver, Mark, The State of the Economy: Economic Issues in Iraq and the
Kurdistan Region (IRIS, March 2017) < https://auis.edu.krd/iris/publications/iris-
booklet-series-state-economy>.
348
Gareth Stansfield, ‘The Kurdish Experience in Post-Saddam Iraq’, in The Kurdish
Question Revisited, ed. by Stansfield and Shareef, 355-374.
349
The Kurdish-speaking Yazidis practice a syncretic monotheistic religion
autochthonous to the Mesopotamian plain. ISIL sees them as worshipers of the
devil which in their eyes gives them right to kill all Yazidi males and enslave the
women. Between 2,000 and 5,500 Yazidis were killed and more than 6,000 were
kidnapped by ISIL in August 2014. Valeria Cetorelli and others, ‘Mortality and
Kidnapping Estimates for the Yazidi Population in the Area of Mount Sinjar, Iraq, in
August 2014: A Retrospective Household Survey’, PLoS Medicine, 14.5 (2017)
<https://doi.org/10.1371/journal.pmed.1002297>.
175
beginning of 2015 and inflicting a huge burden on the KRG’s budget. 350
Moreover, as ISIL advanced to just a few kilometres from Erbil, thousands
of foreign staff left the region and tourism, trade and custom revenues
collapsed.351 Licenced investment fell from $12.4 billion in 2013 to just $4.4
billion in 2014.352 However, the heaviest blow to the region’s finances was
struck by the rapid drop in oil prices. The price per barrel fell from $108 in
June 2014 to $49 in January 2015 – and continued dropping to reach a low
of $29 in February 2016.353 Since Baghdad had cut its budget transfer to the
Kurdish region, the KRG had been keeping its finances from collapsing
through the independent sale of oil but the sudden price drop dashed
expected earnings putting the KRG on the brink of bankruptcy.

The financial crisis further highlighted the limits of Kurdish autonomy in Iraq:
Unable to print money or issue debt,354 the only option available to the KRG
was cutting expenses. However, public spending – in the form of salaries,
pensions and subsidies – was a political pillar of the Kurdish rentier system.
Political stability was maintained only through the redistribution of part of
the oil-generated rent to the wider population while the ruling class
appropriated a large part of it. As the KRG started introducing austerity
measures that threw ordinary Kurds into poverty, the dormant popular
opposition to the political elite exploded.

Protest and Political Crisis

The combination of the war effort and its economic consequences, and the
sudden drop in the price of oil brought the region’s finances on the brink of

350
The Kurdistan Region of Iraq (2016), 2.
351
The Kurdistan Region of Iraq (2015), 36-40.
352
The Kurdistan Region of Iraq (2016), 2.
353
‘Imported Crude Oil Prices’, US Energy Information Administration, 2019
<https://www.eia.gov/outlooks/steo/realprices/>.
354
An attempt made in June 2015 to issue KRG bonds was met with little
enthusiasm by international investors despite the high remuneration (11-12 per
cent) offered. Mills, Under the Mountains, 27.
176
collapse. The KRG halted the payment to contractors and suppliers causing
the bankruptcy of numerous companies in particular in the construction and
oil sectors and slashing the confidence of domestic and international
investors. At the end of 2014, the KRG owed contractors $12.5 billion which
amounted to almost its entire 2013 revenues.355 As these measures proved
to be insufficient to cover the spiralling budget deficit, the Kurdish
government proceeded to cut salaries and allowances. In 2015, the
government came short of paying four months of salaries.356 In March 2016,
all public employees received cuts for an average of 60 per cent of their
wages.357 As noted in the previous chapter, more than half of the KRG’s
workforce was employed by the state and hundreds of thousands of families
relied on a single public salary.

The effects of these measures on the life of ordinary Kurds were

devastating. An independent study estimated a jump in the poverty rate
from just above 3 per cent to 12 per cent in 2015 while average meat
consumption dropped from 45 kg to just 23 kg a year.358 Due to the collapse
of the private sector unemployment rose particularly among the youth.359
Total consumption in the region declined by 14 per cent in 2014 and 24 per
cent in 2015.360 The salary cuts of politicians and high-ranked officials did
not produce a sense of collective national sacrifice given the widespread
awareness that their salaries were little compared to what they gained
through corruption and misappropriation. For example, in January 2015, just

In Best of Times and Worst of Times: Addressing Structural Weaknesses of the

355

Kurdistan Region’s Economy (MERI, 2016) <http://www.meri-k.org/wp-

content/uploads/2016/01/MERI-Economic-Report-January-2016-2.pdf>, 15.
356
In Best of Times, 15.
357
The Kurdistan Region of Iraq (2015), 11.
This information comes from observants from the region and was collected in
358

DeWeaver, The State of the Economy, 3-4.

On this, the data is contradictory. For the year 2014, the KRG estimated youth
359

unemployment at 17.6 per cent while for the UNDP it was as high as 36.7 per cent.
See Capacity Building at the Kurdistan Regional Statistics Organization Through
Data Collection (KRG and Rand Corporation, 2014) <https://doi.org/10.1214/07-
EJS057>, 27-28; and Iraq Human Development Report, 53.
360
The Kurdistan Region of Iraq (2016), 4.
177
before implementing harsh austerity measures, the KRG covered $375
million of taxes owed to the Iraqi government by Korek a telecommunication
giant widely known to be indirectly owned by Prime Minister Nechirvan
Barzani.361 The lavish lifestyle of the Kurdish elite was more and more in
sharp contrast to the impoverishment of most of the region’s population.

In October 2015, after months of delayed and frozen salaries, protests broke
out in the Sulaymaniyah province targeting mostly the offices of the KDP.
As the party-affiliated security forces responded with fire, five protestors
were killed.362 The deadly outcome of the demonstrations discouraged
protests for a while. However, when the KRG extended the austerity
measures to the peshmerga forces – dangerously increasing the rate of
desertion on the ISIL front –363 and to the police, popular rage exploded.
Protests took place again in February and then intermittently for the whole
of 2016 and 2017.364 While ordinary Kurds were draining their savings and
taking loans to make ends meet, popular opposition to the ruling KDP and
PUK reached an unprecedented level. It is impossible to determine the exact
class composition of these protests but the austerity measures severely hit
the vast majority of the population. Besides the public employees – over half
of the workforce –365 the crisis had devastating effects on the poorest
components of Kurdish society especially those relying on daily salaries that
found themselves without an income overnight.

The austerity policies adopted by the KRG in the aftermath of the 2014 crisis
brought about the collapse of the rentier system established in the 2000s.

361
Michael Rubin, 'The Continuing problem of KRG Corruption', Routledge
Handbook on the Kurds, ed. by Gunter, 331-332.
‘Iraqi Kurdistan: Ruling Party Forces Fire on Protesters’, Human Rights Watch, 21
362

October 2015 <https://www.hrw.org/news/2015/10/21/iraqi-kurdistan-ruling-

party-forces-fire-protesters>.
363
Florian Neuhof, ‘Unpaid Peshmerga Are Voting with Their Feet’, Deutsche Welle,
2 February 2016 <https://www.dw.com/en/unpaid-peshmerga-are-voting-with-
their-feet/a-19020152>.
364
‘Protests Intensify in Iraqi Kurdistan amid Economic Crisis’, Human Rights Watch,
9 February 2016 <https://www.reuters.com/article/us-mideast-crisis-iraq-
protests-idUSKCN0VI11X>.
365
See Chapter 5.
178
As discussed in Chapter 5, the rapid economic growth of the previous year
had allowed the Kurdish ruling class to appropriate huge amounts of wealth
mostly through corruption and non-transparent budgetary practices.
However, the political implications of this process of accumulation at the top
had been mitigated by the distribution of part of the oil rent to the larger
population mostly through public handouts. Due to the weakness of the
private sector, most Iraqi Kurds were forced to rely on public employment
distributed by the ruling parties. By keeping people’s income dependant on
their loyalties to the parties, the political elite was able to silence the
expressions of popular opposition against their predatory practices.
However, as soon as the effects of the salary cuts in the public sector kicked
in, this political arrangement broke down and protests started taking place
regularly. The response of the ruling KDP and PUK was to use the security
apparatus to crack down on dissent and to unlawfully shut down the Kurdish
parliament where the opposition could give political expression to popular
anger.

When the economic crisis started, the KDP and PUK invited the opposition
parties to form a national unity government to have them share the political
burden of the crisis. The Gorran Movement – which had boomed in the 2013
regional elections on an anti-corruption platform – joined the government in
a short-lived atmosphere of national unity driven by the threat posed by
ISIL.366 Gorran received important ministerial posts as well as the speaker of
the parliament. Their experience in power was, however, rather
disappointing. Their promises of fighting corruption and imposing
transparent practices in the administration were frustrated by the resistance
posed by the KRG bureaucracy filled with personnel affiliated to the KDP
and the PUK that kept responding to their political parties rather than the
ministers. For the Minister of Finance, it proved impossible to track down
revenues and expenditure to construct a transparent budget as high-ranked
bureaucrats kept hiding the size and directions of each stream of revenues.

366
The Islamist parties Kurdistan Islamic Union and Kurdistan Islamic Group also
joined the national unity government.
179
Moreover, Gorran’s plan to transform the peshmerga forces from party
militias into a regional army was destined to fail as the KDP and PUK made
clear they were never going to give up their most powerful instrument of
social control.367 The control of the armed forces proved to be essential to
the KDP when its position of power was directly threatened.

The national unity government lasted for little more than a year. In summer
2015, Masoud Barzani’s presidential term was coming to an end. As
mounting popular anger was mostly directed against the KDP and the
Barzani family, the other parties were resolute to block Masoud’s re-
election. However, when public demonstrations exploded in October 2015,
Prime Minister Nechirvan Barzani accused Gorran of being behind the
protests and sacked their ministers. Thanks to its military control of Erbil,
the KDP could just shut down the parliament and prevent the speaker from
entering the capital.368 Despite an attempt to mediate, the United States,
leading the anti-ISIL coalition, never withdrew its support to the Barzanis
providing a degree of legitimacy that allowed the KDP to keep the
parliament shut for two years without any repercussion on their international
standing.369

The Evolving Power Structure of the KRG

The economic collapse and political unrest that started in 2014-2015

accelerated the crisis of legitimacy of the Kurdish political elite and, by
extension, threatened the position of the entire ruling class. To understand

367
Interviews with Hoshyar Omar (Sulaymaniyah, 2018) and Kamal Chomani
(personal communication, 2019).
368
Mohammed A. Salih, ‘KRG Parliament Speaker: Barzani’s Term Extension
“against the Law”’, Al-Monitor, 28 August 2015 <https://www.al-
monitor.com/pulse/originals/2015/08/kurdish-parliament-speaker-challenge-
barzani-legitimacy.html>.
369
Veteran peshmerga commander Kaka Hama claims that “American and British
representatives […] told us [that] if Kurds distract themselves with internal issues,
they won’t have the support of the UK and the US in fighting the Islamic State.”
‘Kaka Hama: Parties Warned by US Officials to Keep Barzani President’, Rudaw, 18
August 2015 <https://www.rudaw.net/english/kurdistan/180820156>.
180
why popular protests represented such a danger, it is important to step back
and discuss the evolution of the Kurdish ruling class between the 2000s and
2010s and its relationship to the political elite. The suspension of the Kurdish
parliament on October 2015 represents just the most blatant example of the
authoritarian nature of the two-party rule and particularly of the increasing
reliance on coercion that followed the rise of a meaningful challenge with
the establishment of Gorran in 2009 and the 2011 anti-corruption
protests.370

In these years, the function of the KDP and PUK and the power relations
within them changed. As discussed earlier, the massive amount of wealth
accumulated at the top strengthened the hierarchical structure of these
organisations. The long period of relative peace that started with the end of
the Kurdish civil war in 1997 reduced the importance of party cadres in
charge of the peshmerga units vis-à-vis that of the higher-ranking leaders
controlling the sources of rent and hence its redistribution downwards. The
power of the dominant families increased substantially to the point that the
organisational bodies of the parties – such as the politburo and the party
congress – lost their political relevance and were replaced by informal
relations of clientelism between powerful families and the middle party
cadres. Political bargaining at the regional level has been less between
political parties than between families that even built cross-party alliances.
The KDP and PUK have maintained their role as political institutions that run
the affairs of the region but the power relations lie in informal power chains
that run downwards from a dominant politician through loyal party cadres
and further down to the ordinary constituents in need of employment and
protection.371

The KDP has been centred around the Barzani family since its foundation
and the weakening of the decision-making bodies of the party was less

370
See Chapter 5.
371
For an analysis along these lines, see After Iraqi Kurdistan’s Thwarted
Independence Bid (International Crisis Group, 2019)
<https://d2071andvip0wj.cloudfront.net/199-after-iraqi-kurdistan.pdf>.
181
dramatic. The leadership has been undisputedly in the hands of the lineage
of the ‘immortal leader’372 Mullah Mustafa Barzani with his son Masoud as
the head of the family. The line of succession informally proceeds via
seniority with Nechirvan, the son of Masoud’s prematurely-dead elder
brother Idris, as the second in line and Masoud’s oldest son Masrour as the
third. Consequently, when Masoud Barzani was the KRG president,
Nechirvan was the Prime Minister and when Masoud resigned in 2017,
Nechirvan succeeded him as the president and Masrour became Prime
Minister. The KDP does little to conceal its quasi-monarchical structure and
individuals within the family are treated as public officials even if they do
not hold any office.373

These practices have become dominant also within the PUK but in a context
of increasing fragmentation. Until 1992, the PUK was a federation of formally
independent parties and, under Jalal Talabani’s leadership, the party always
contained independent powerful individuals. The encroaching of nepotistic
practices and corruption was the main cause of the 2009 split by
Nawshirwan Mustafa and his followers and their new party Gorran became
a powerful challenger to the PUK in Sulaymaniyah in a way that would be
unimaginable in the KDP-controlled areas. However, as the most vocal
opponents left with Gorran, the power shift from the PUK party structures
to family politics inevitably accelerated. The PUK did not hold any party
congress between 2010 and 2019 despite its constitution requiring one
every four years.374 Especially since the death of Jalal Talabani in October
2017, the party has been divided into two factions: One led by Talabani’s
widow Hero Ibrahim Ahmed and their sons and the other by Kosrat Rasul,

372
See ‘Congresses of Kurdistan Democratic Party (1946-1999)’, KDP,
<https://www.kdp.se/index.php?do=congress>
373
In April 2019, Masrour Barzani’s teenage son was photographed at a sport event
rewarding athletes with medals side by side with the French Consul in Erbil. See
Kamal Chomani (@KamalChomani), ‘Son of Masrour Barzani, could be around 15
yrs old (…)’, Twitter, · 28 April 2019, 17:05
<https://twitter.com/KamalChomani/status/1122532218298097664>.
374
After Iraqi Kurdistan’s Thwarted Independence Bid, 6.
182
the acting leader of the PUK after Talabani’s death.375 In the years of the
economic and political crisis, the PUK was unable to pursue a coherent
strategy. On the one hand, they tried to recover some of the electoral
ground lost in favour of Gorran by blaming the region’s problems on the KDP
and joining the other parties to stop Masoud Barzani’s re-elections in 2015.
On the other hand, they have kept a higher level of negotiations with the
KDP and resorted – as described later in the chapter – to even more virulent
electoral fraud to regain the positions lost in Sulaymaniyah.

The rise of a new generation of Kurdish leaders hailing directly from the
dominant families heavily tarnished the legitimacy of the two ruling parties.
Whereas Jalal Talabani and Masoud Barzani could still enjoy popularity and
respect due to their prominent role in the liberation struggle, 376 the rise of
their sons and nephews to power reveals the real nature of the KDP and
PUK as family-run enterprises. The consequence of this process on the two
ruling parties has been a dramatic loss of their credibility as the champions
of Kurdish nationalism in Iraq and the weakening of their legitimacy of the
rule over the region.377

This was the context in which the collapse of the Kurdish economy and
political system occurred in 2014-2015. The refusal of the two ruling parties
to address the public’s demands for a more equal distribution of the – now
scarce – KRG resources cannot be explained solely with the greed of an elite
accustomed to a lavish lifestyle. The appropriation of the KRG’s public
revenues was an indispensable component of their rule. Without siphoning
money off the public budget and maintaining control over the provision of
public contracts, the Kurdish leaders could not keep feeding the pyramid of

375
Ibid., 8.
376
In Sulaymaniyah, Jalal Talabani’s figure is still surrounded by an aura of purity.
Corruption and nepotism within the PUK are then often attributed to the greed of
his family – his wife Hero Ibrahim particularly – who took charge while Talabani was
based in Baghdad and then fell ill.
377
For an analysis of the increasingly dynastic nature of Iraqi Kurdish politics, see
Kamal Chomani, ‘Kurdistan Region at a Time of Crown Princes’, 1001 Iraqi Thoughts,
28 December 2018 <https://1001iraqithoughts.com/2018/12/28/kurdistan-region-
in-a-time-of-kurdish-crown-princes>.
183
interests that sustain their power. The network of businessmen, high-
ranked civil servants, private security companies but also tribal leaders in
the rural areas on top of which high-level politicians sit would simply
collapse if the stream of money and contracts were interrupted and these
local power holders would just migrate their loyalty to some political
competitor. In these terms, corruption and misappropriation are not solely
bad practices but also an indispensable tool that allows the political elite to
feed the wider ruling class of the region and maintain cohesion at the top of
the class structure. This cohesion became increasingly important as the KDP
and PUK were losing popular support.

External Relations and Domestic Rule

The erosion of the legitimacy of the KDP and PUK went hand in hand with
their increasing reliance on the external support of, respectively, Turkey and
Iran. As all opposition forces claim, the Department of Foreign Relations of
KRG held a purely ceremonial function and external relations were the
exclusive domain of the two ruling parties, also due to their direct control
over the security apparatus.378 The close relationship built up by KDP and
PUK with their neighbours had a dual function for the Kurdish elite. On the
one hand, the economic penetration of the region by Turkish and Iranian
actors is an extremely beneficial business for politically-connected Kurdish
companies. On the other hand, the KDP and PUK tied their geopolitical
survival in an increasingly unstable Middle East to the military might of
Turkey and Iran, also gaining insurance against the possibility that domestic
unrest escalated into a full-blown revolution.

However, such a close partnership with two foreign powers which

themselves oppress their Kurdish minorities heavily undermines the
nationalist credentials of the KDP and PUK. Kurdish nationalism is inherently

378
Interviews with Mustafa Abbas Abbas (Sulaymaniyah, 2018), Aram Saeed
(Sulaymaniyah, 2018), Hoshyar Omar (Sulaymaniyah, 2018), Muhammad Hakim
Jabar (Sulaymaniyah, 2018), and Muthanna Amin (Sulaymaniyah, 2018).

184
irredentist as the Kurds are spread as minorities across four countries. The
feeling of solidarity among Kurdish speakers across Iraq, Turkey, Iran, and
Syria made the KRG a natural haven for Kurdish opposition groups from
neighbouring countries. In the 1990s, both the PKK from Turkey and a
constellation of armed Iranian-Kurdish organisations settled within the
borders of the Kurdistan Region of Iraq giving Turkey and Iran a permanent
excuse to violate the KRG territory.379 The KDP and PUK have had to juggle
their close relationship with Turkey and Iran with the widespread popular
sympathy toward Kurdish opposition in those countries.

The strategic partnership between Iran and the PUK dates back to the
Kurdish civil war in the 1990s and was largely a response to the KDP’s
growing ties with Baghdad and Ankara.380 While the PUK was at war with the
KDP and under blockade by Baghdad, the Iranian border became the only
window to the world for the Sulaymaniyah-based PUK. But Iran’s support
did not come for free and, in 1996, Talabani allowed Iranian troops to cross
the border and attack the Iranian Kurdish fighters of the KDPI hosted in the
PUK-controlled area.381 Since then the PUK was forced into the
uncomfortable position of having to police the activities of the KDPI on
behalf of Tehran.382 Despite the paralysis of the Iranian-Kurdish party, Iran
kept hitting its enemy across the KRG as shown by targeted assassinations
of opposition leaders,383 and the 2018 bombing of the KDPI bases that killed

379
Tahiri, The Structure of Kurdish Society, 276-299.
380
Gunter, The Kurdish Predicament, 111-112.
According to Asso Hasanzadeh, the son of the then-leader of the KDPI Abdullah
381

Hasanzadeh, Jalal Talabani warned the Iranian Kurds so that “when [the Iranians]
bombed here, only our peshmerga were left here, our families were sent to the UN
camps in Erbil.” Interview with Asso Hasanzadeh (Koya, 2018).
382
Loghman Ahmedi, a young member of the KDPI leadership claims that “we made
a very strategic mistake in the early 1990s when we decided to leave our bases in
the mountains and came down here. […] We halted some of our activities becuase
[…] we wanted to give the KRG an opportunity to stabilise and create their own
institutions and so on. But that weakened our party.” Interview with Loghman
Ahmedi (Koya, 2019).
Fazel Hawramy, ‘Assassinations Mount as Iranian Kurdish Militants Clash with
383

Tehran’, Al-Monitor, 7 March 2018 <https://www.al-

185
15 militants.384 Due to its reliance on Iranian support, the PUK is in no position
to effectively protest these violations.

The political ties between Tehran and the PUK parallel the economic
penetration of the PUK-controlled areas by Iranian business. The level of
economic influence Iran exercises over Sulaymaniyah extends beyond the
area’s dependence on Iranian-made consumer goods and food supply. The
Kurdish political elite – in a more or less transparent way – partners with
Iranian firms and takes a stake in the profits made by Iranian investment,
while the PUK benefits directly from the control of the border-crossings.385
Generally speaking, the multi-layered influence exercised by Tehran over an
important part of the KRG is one of the reasons why Iran supported the
establishment of the autonomous Kurdish region in 2003. This fits in very
well with Iran’s broader objective of keeping Iraq fragmented and
decentralized, to avoid the resurgence of the strong enemy they faced in
the 1980s.

Iran’s influence over Sulaymaniyah is, however, surpassed by Turkey’s sway

over the KDP-controlled area of the KRG. Since the PKK established its
bases in the Qandil mountain in the north-eastern corner of the KRG in 1991,
the Turkish-Kurdish conflict intertwined with Iraqi Kurdish politics. The PKK
took advantage of the remoteness of this area to wage attacks on Turkish
positions, pushing Turkey to carry on a series of costly but unsuccessful
cross-border operations on Iraqi soil. The proximity of the KDP-controlled
area to the Turkish border and Barzani’s interest in limiting the influence of
the PKK deepened Turkish-KDP relations in the 1990s. However, open ties
could only develop in the late 2000s, after the Islamist Justice and
Development Party (AKP) led by Recep Tayyip Erdoğan consolidated its

monitor.com/pulse/originals/2018/03/iran-kdpi-kurdish-opposition-iraq-
assassinations-rahmani.html#ixzz5bkPrbUIC>.
384
‘Rockets Hit Iranian Kurdish Opposition Offices in Iraq’s Koya’, Al Jazeera, 8
September 2018 <https://www.aljazeera.com/news/2018/09/rockets-hit-iranian-
kurdish-opposition-offices-iraq-koya-180908090605503.html>.
‘Iran to Establish Goods Trading Center in Iraq’s Sulaymaniyah’, Mehr News
385

Agency, 29 October 2018 <https://en.mehrnews.com/news/139170/Iran-to-

establish-goods-trading-center-in-Iraq-s-Sulaymaniyah>.
186
power in Turkey and could overcome the resistance of the viscerally anti-
Kurdish nationalist establishment. These efforts culminated in the opening
of a Turkish consulate in Erbil in 2010. Since then, the rate of Turkish
economic investments in the KRG grew exponentially, and cross-border
trade flourished. As in the case of Iranian economic penetration, this
economic partnership largely benefitted the local Kurdish elite.386 Profits
increased exponentially when the KRG started selling oil to Turkey. Ankara’s
thirst for cheap fuel offered the KDP leadership the opportunity to bypass
Baghdad’s constitutional monopoly over the export of Iraqi oil by selling it
directly to Turkey. In the early 2010s, the oil partnership with Turkey became
one of the most significant points of contention between Iraq and the KRG
increasing Erbil’s bargaining power vis-à-vis Baghdad but reducing its
autonomy from Ankara.387

The relationship between Turkey and the KDP held also a significant political
dimension. Engaged in a long-term campaign against the pro-PKK parties
for electoral hegemony over Turkey’s Kurdish region, Erdoğan used his
relationship with Barzani to demonstrate the compatibility between the AKP
values and Kurdish identity. This attempt culminated in Barzani’s
participation, in November 2013, in a mass rally in Diyarbakır – Turkey’s
largest Kurdish-majority city – where he shared the stage with Erdoğan in
an unprecedented celebration of Turkish-Kurdish brotherhood.388 This
relationship raised Barzani’s standing as a pan-Kurdish leader who could
provide Turkey’s Kurds with a conservative alternative to the militant PKK.
However, the Turkey-KDP political partnership suffered a blow in 2015 with
the onset of the parliamentary alliance between Erdoğan’s AKP and the
Turkish far-right characterised by violently anti-Kurdish positions. With little

386
See Chapter 5.
See Bill Park, ‘Turkey, the US and the KRG: Moving Parts and the Geopolitical
387

Realities’, Insight Turkey, 14.3 (2012), 109–25.

388
Çandar, Cengiz, ‘Erdogan-Barzani 'Diyarbakir encounter' milestone’, al-Monitor,
20 November 2013 <https://www.al-
monitor.com/pulse/originals/2013/11/erdogan-barzani-kurdistan-diyarbakir-
political-decision.html>.
187
or no opposition from the KDP, Turkey resumed airstrikes over the PKK
camps in Iraqi Kurdistan that regularly kill local Kurdish villagers. The
violation of the KRG territory and the multi-front war waged by Erdoğan
against the PKK, the civilian Kurdish opposition as well as the Kurds of
Syria,389 exposed the contradictions between the KDP’s nationalism and its
relations with Ankara undermining its credibility among the Kurds. 390
Nevertheless, the KDP’s economic and geopolitical dependency on Turkey
remained intact as shown by the ban on pro-PKK political activities,391 and
the opening of two Turkish military bases on KRG soil.392 Turkish military
presence stirred popular resentments manifested in violent
demonstrations.393

When looked at from the standpoint of domestic politics, it becomes clear

why the benefits of the foreign partnerships of the KDP and the PUK
outweigh their cost in terms of political legitimacy. Besides the profits
brought by partnering with foreign companies, the KDP and PUK enjoy the
backing of a regional power vis-à-vis Baghdad but also their own
populations. The stronger their ties with Turkey and Iran, the less
accountable they feel to domestic dissent increasing their authoritarian
control over the region’s politics and economy. When the 2014 crisis broke
out, the mounting pressure from below pushed the KDP and the PUK to
invest in a much more aggressive nationalist rhetoric in open contradiction

389
See Chapter 9.
390
‘Iraqi Kurds Protest against Turkish “Genocide”’, Euronews, 8 February 2016
<https://www.euronews.com/2016/02/08/iraqi-kurds-protest-against-turkish-
genocide>.
391
The KDP-controlled KRG Electoral Committee banned from the regional
elections of September 2018 the pro-PKK Tevgar Azadi despite the fact that the
party had been allowed to participate in the Iraqi federal elections of May.
Interviews with Aram Saeed (Sulaymaniyah, 2018), Abbas Mustafa Abbas
(Sulaymaniyah, 2018), and Tara Muhammed (Sulaymaniyah, 2018).
392
‘KRG Spokesperson Statement on Turkish Forces Presence in Bashiqah’,
Kurdistan Regional Government, 6 October 2016
<http://previous.cabinet.gov.krd/a/d.aspx?s=040000&l=12&a=54994>.
393
Fehim Taştekin, ‘Storming of Base in Iraq a Grave Signal for Turkey’, Al-Monitor,
1 February 2019 <https://www.al-monitor.com/pulse/originals/2019/02/turkey-
iraqi-kurdistan-storming-of-base-grave-signal.html>.
188
with their deep ties to Turkey and Iran. However, this contradiction
backfired in the occasion of the 2017 independence referendum.

The 2017 Independence Referendum

As mass anti-austerity demonstrations started taking place regularly since

late 2015, the KDP and PUK realised that they were standing on increasingly
shaky ground. As meeting public demands was not an option, the KRG
leadership responded with a carrot-and-stick strategy combining political
repression with heightened nationalist rhetoric against Baghdad. After
showing no hesitation to fire on protesters, they suspended the parliament
and expelled the other political forces from the government leaving a KDP-
PUK cabinet to run the affairs of the region without any check while Masoud
Barzani remained president even after his term expired in 2015. Parallel to
repression, the ruling parties revamped their aggressive nationalist rhetoric
to raise the tensions with Baghdad in an attempt to generate a ‘rally-‘round-
the-flag’ effect.

The attempt to invest in nationalism and to divert public attention from the
economic crisis was aided by the mass popularity of the peshmerga
deployed on the ISIL front and by the occupation by Kurdish forces of most
of the areas disputed by Baghdad and Erbil. In particular, the Kurdish
takeover of the oil-rich city of Kirkuk from ISIL inflamed nationalist feelings,
given the city’s highly symbolic value for the Kurds. Riding popular
sentiments, the KRG president Masoud Barzani raised the issue of Kurdish
independence as soon as the war started.394 In spring 2017, as Iraqi and
Kurdish forces liberated Mosul from ISIL while – at the same time – protests
in the KRG intensified, a high-level meeting between the KDP and PUK
resulted in the decision to hold a referendum by the end of that year.395

394
‘Iraq Kurdistan Independence Referendum Planned’, BBC News, 1 July 2014
<https://www.bbc.co.uk/news/world-middle-east-28103124>.
395
‘Kurdistan Will Hold Independence Referendum in 2017, Senior Official’, Rudaw,
2 April 2017 <https://www.rudaw.net/english/kurdistan/020420171>.
189
The KDP heavily invested in the referendum to shift the debate away from
the political and economic crises and the suspension of the parliament.
Masoud Barzani presented the referendum as about “the destiny” of the
Kurdish people and therefore as above “any other political framework, or
any political parties, or any political problems within the party system.” 396
The KDP managed to present a solid and unified position maximizing their
appeal to nationalist voters. The situation was more complicated for the PUK
that was going through a period of fragmentation especially since Jalal
Talabani fell terminally ill before dying in October 2017. Moreover, the
referendum was bound to generate tensions in the ethnically-mixed Kirkuk
and jeopardise Kurdish control of the city which had traditionally been a PUK
stronghold. The PUK leadership in Kirkuk, supported by the Talabani family,
was against holding the referendum in the city. However, Kirkuk governor
Najmaddin Karim appealed to the PUK acting leader Kosrat Rasul who sent
3,000 peshmerga to the city to enforce the holding of the referendum.397

The nationalist frenzy promoted by the KDP forced the opposition to

attenuate their positions. While recognizing the right of the Kurds to self-
determination, Gorran questioned the timing of the referendum and
demanded to have the regional elections – due in November 2017 – in
conjunction with it. When the government convened the parliament on
September 15th, after two years of forced closure, just to ceremonially
proclaim the referendum date on September 25th, Gorran and other
opposition forces, as well as part of the PUK boycotted the meeting.
Observers describe the divide running through the region with the KDP-
controlled areas covered with pro-independence propaganda and nothing
of the kind in the PUK-controlled areas.398 On the day of the referendum, 93

396
Campbell MacDiarmid, ‘Masoud Barzani: Why It’s Time for Kurdish
Independence’, Foreign Policy, 15 June 2017
<https://foreignpolicy.com/2017/06/15/masoud-barzani-why-its-time-for-
kurdish-independence/>.
397
Bill Park, Joost Jongerden, Francis Owtram, and Akiko Yoshioka, ‘Field Notes:
On the Independence Referendum in the Kurdistan Region of Iraq and Disputed
Territories in 2017’, Kurdish Studies, 5.2 (2017), 208.
398
Ibid., 207.
190
per cent of voters supported independence. However, the huge differences
in the participation rate present the picture of a very divided region with 91
per cent turnout in KDP-controlled Dohuk and just 55 per cent in
Sulaymaniyah.399

One of the reasons why many were sceptical of the referendum was the
sheer lack of international support needed to make Kurdish statehood
viable. The US opposed the referendum and Baghdad threatened retaliation
while Ankara and Tehran, fearing the effects of Kurdish statehood on their
own Kurdish minorities, restated their commitment to the territorial integrity
of Iraq. In the aftermath of the referendum, the KRG leadership found itself
completely isolated on the international stage and had to face the retaliatory
actions of the neighbouring powers.400 Turkey and Iran showed their
muscles by organising a joint military exercise on the border. Supported by
the two neighbours, Baghdad immediately halted international flights to the
KRG while Tehran and Ankara closed their airspace to all flights to and from
the region. On October 15th, Iraqi troops marched on Kirkuk without
encountering much resistance and possibly in agreement with the PUK
faction that had opposed the referendum in the first place. 401 In just a few
days, most of the disputed territories gained by the Kurds during the war on
ISIL were lost. The repercussions of the referendum surpassed the most
pessimistic expectations and, on November 1st, Masoud Barzani resigned as
KRG president.

The size and weight of the punitive measures taken against the KRG
triggered a debate over the choice to pursue the referendum despite the
virtually unanimous international hostility. The independence of a
landlocked Kurdish state without the support of any one of the neighbouring
powers would inevitably turn into a geopolitical nightmare. Most scholars

399
Turnout figures are estimates. Ibid., 205.
400
See, Bill Park, ‘Explaining Turkey’s Reaction to the September
2017Independence Referendum in the KRG: Final Divorce or Relationship Reset?’,
Ethnopolitics, 18.1 (2019), 46-60.
401
Park and others, Field Notes, 208.
191
and observers explained Barzani’s choice as the result of miscalculations or
of genuine nationalist aspirations that go beyond any material
considerations.402 These interpretations ‘orientalise’ the Kurdish leadership
depicting them as patriotic warriors too naïve to understand geopolitical
power games and legitimise Barzani’s and the KDP’s victimhood narrative.
More importantly, these readings dismiss the contested nature of the
referendum and the opposition of important sectors of the Kurdish society.

More convincingly, the analyses by Palani and others (2019) and, O’Driscoll
and Başer (2019) point to domestic political dynamics within the KRG
explaining the referendum as a political move by the KDP. 403 However,
presenting the referendum solely as Barzani’s gamble to stay in power
overlooks the structural context that forced the KDP to invest in an
increasingly aggressive nationalist stance as their last source of political
legitimacy. The referendum must be seen in continuity with this process as
well as tied to the increasing reliance on repression and authoritarian
methods. The referendum was, in these terms, the ultimate tool of the
strategy of survival of the Kurdish ruling class and it is not coincidental that
it was promoted and supported by the entire KDP leadership and by a
significant part of the PUK.

Even if Barzani made some gross miscalculations, he was aware that a

parliamentary majority to keep the presidency for himself was simply not
there. Yet, the chaos generated by the referendum allowed the KDP to
postpone the regional elections for another year. Ultimately, the KDP –
unlike the Iraqi Kurds as a whole – did not lose anything from the referendum
and, as the next section shows, they even came out stronger. Masoud

402
Just to give some examples: Fahrettin Sumer and Jay Joseph, ‘The Paradox of
the Iraqi Kurdish Referendum on Independence: Contradictions and Hopes for
Economic Prosperity’, British Journal of Middle Eastern Studies, 45.4 (2018), 574–
88; Morgan L. Kaplan, ‘Foreign Support, Miscalculation, and Conflict Escalation:
Iraqi Kurdish Self-Determination in Perspective’, Ethnopolitics, 18.1 (2019), 29–45 .
403
O’Driscoll and Başer, Independence Referendums; Kamaran Palani, Jaafar
Khidir, Mark Dechesne, and Edwin Bakker, ‘The Development of Kurdistan’s de
Facto Statehood: Kurdistan’s September 2017 Referendum for Independence’,
Third World Quarterly, 40.12 (2019), 2270-2288.
192
remained the head of the Barzani family and arguably the most powerful
individual in the region.

Kirkuk was undoubtedly a great symbolic and material loss but the KDP
managed to shift the blame on the PUK forces that garrisoned the city. The
traditional hold of the PUK over Kirkuk meant that its inclusion in the
upcoming regional elections would have significantly boosted the party in
the Kurdish parliament while the control of the Kirkuk oilfields could have
allowed the PUK to bridge the gap with the KDP in terms of economic power.
However, independent studies show that the oil from Kirkuk will not
constitute a significant part of the KRG production for more than a few years
and, with the expansion of the oil sector to the region’s vast unexploited
fields, it will likely become marginal in the course of the 2020s.404 In the long-
term, Kirkuk’s oil output would not make a huge difference for the viability
of a Kurdish state while its inclusion in the KRG would certainly change the
internal relations of power and constitute a permanent source of tension
with Baghdad. Ultimately, even if it was not a calculated move, losing Kirkuk
was a convenient unintended consequence for the KDP.

The result of the Kurdish referendum was a renewed political paralysis.

Under the threat of invasion from Baghdad and the neighbouring powers,
the KDP managed to postpone the regional elections for another year,
showing that the resignation of President Barzani was not going to bring any
step towards the democratisation of the region. The ruling elite had lost its
legitimacy in the eyes of most of the Iraqi Kurdish population, yet the
militarization of the region seemed to leave no room for political alternatives.

The Crisis of Kurdish Nationalism

The 2017 independence referendum marks a turning point in the history of

the Kurdish region of Iraq. Despite the overwhelming victory, the
independence of the region was never proclaimed while the KRG had to pay
the heavy consequences of the vote. The result was a widespread feeling

404
Mills, Under the Mountains, 24.
193
of resentment towards the ruling elite which had lost its legitimacy in the
eyes of a large part of the Kurdish population, a feeling manifested in the
frequency and width of public protests. However, the continuous use of
coercion by the dominant party and the gradual co-optation of Gorran into
the ruling establishment generated a pervasive sense of disillusionment for
any prospects of political change within the system. The ultimate victim of
these dynamics was Kurdish nationalism as the fundamental source of
legitimacy for political power in the region.

The hope that the independence referendum would trigger a ‘rally-round-

the-flag’ effect was dashed in just a few days. At the announcement of
Masoud Barzani’s resignation, a spontaneous crowd stormed the Kurdish
parliament in Erbil.405 Large anti-austerity protests resumed in the fall of
2017 and took place frequently throughout 2018. In December 2017, the
police fired on the protestors killing six.406

Despite growing popular resentment, the Iraqi federal elections of March

2018 proved to be a success for the KDP and PUK. The KDP triumphed in
Dohuk and Erbil, while the PUK became again the first party in
Sulaymaniyah. Gorran, after two consecutive electoral victories, dropped
from 39 per cent in 2014 to just 23 per cent. Contributing to Gorran’s fall
was the establishment of the New Generation Movement, a populist party
founded by businessman Shaswar Abdulwahid Qadir that managed to
attract part of the protest vote. However, the most significant
characteristics of these elections were the dramatic drop in voters’ turnout
and the allegations of an unprecedented degree of electoral fraud.407 The

405
‘Protesters Storm Kurdistan Parliament after Barzani Announces Resignation’,
Reuters, 29 October 2017 <https://www.reuters.com/article/us-mideast-crisis-
iraq-kurds-protests/protesters-storm-kurdistan-parliament-after-barzani-
announces-resignation-idUSKBN1CY0PI>.
406
Gul Tuysuz, Hamdi Alkhshali, and Muwafaq Mohammed, ‘At Least 6 Killed during
Violent Protests in Iraqi Kurdistan’, CNN, 19 December 2017
<https://edition.cnn.com/2017/12/19/middleeast/iraq-kurdish-
protests/index.html>.
407
According to Abbas Mustafa Abbas, Professor at Sulaymaniyah University,
“more than 100,000 dead people are still in the [electoral] lists. Also, they are
duplicating the names. For example, the peshmerga are 300,000 people and they
194
turnout averagely fell by more than 20 per cent compared to the 2014
elections, as only half of the registered voters showed up at the ballot. In
KDP-controlled Erbil, turnout was as low as 43 per cent.408 The opposition
parties were heavily penalised by this low level of participation driven by the
widespread sense of disillusionment. A feeling that was well justified. The
perception that KDP and PUK had manipulated the results – more heavily
than usual – came not only from civil society groups and the opposition
parties but also from the Iraqi government and several international
organisations.409 Unsurprisingly, a very similar picture emerged from the
Kurdish regional elections held just a few months later on September 2018.
Both the KDP and PUK increased their seats in the regional parliament while
Gorran lost almost half. The turnout was as low as 58.4 per cent (-16 per
cent from 2013) but even this figure might have been manipulated.410

The elections of 2018, the first after the beginning of the crisis and popular
protests, showed that KDP and PUK were no longer willing to allow any
margin to political challengers even at the cost of turning the electoral
process into a farce. In this context measuring the actual remaining popular
support to the two ruling parties is futile but the need to resort to even more
blatant electoral fraud suggests that it is very low. These elections also had
a profound impact on Gorran. Gorran’s control of the provincial council of
Sulaymaniyah and their participation in the regional government in 2014-
2015 already showed the narrow margins for change even when its
members when in executive positions. The widespread disillusionment that
followed the 2017 referendum heavily affected Gorran especially as the KDP
and PUK blocked the electoral path to change. With the death of Gorran’s

can vote both before the elections and during the elections. It means 30 seats!”.
Interview with Abbas Mustafa Abbas (Sulaymaniyah, 2018).
408
Renad Mansour and Christine Van Den Toorn, The 2018 Iraqi Elections: A
Population in Transition?, (LSE Middle East Centre and IRIS, July 2018)
<http://eprints.lse.ac.uk/89698/7/MEC_Iraqi-elections_Report_2018.pdf> 16.
409
Ibid., 18.
410
Kristina Bogos, Analysis: Looking Back on the 2018 Kurdish Elections (IRIS,
March 2019) <https://auis.edu.krd/iris/sites/default/files/Analysis_Looking Back on
the 2018 Kurdish Elections_March 2019.pdf>.
195
founder Nawshirwan Mustafa in May 2017, the party visibly fell under the
influence of his sons Nma and Cheya, a dynamic that generated great
opposition within the party, with cadres accusing the leadership of turning
the party into a family business like the KDP and PUK.411 While much of the
youth, as well as veteran peshmerga commanders and early followers of
Nawshirwan Mustafa, left the party,412 the so-called ‘pragmatist wing’ gained
strength. The argument put forward by the now-dominant figures in Gorran
is that the elections showed that the only way for the party to be competitive
is to be in power where they can distribute jobs and handouts among their
followers and build a solid support base similar to those of the KDP and
PUK.413 These considerations reflect the widespread belief that the space
for a mode of politics alternative to that of the KDP and PUK was closed. In
the year that followed the regional elections of September 2018, the KDP
led the negotiation to form a government with both the PUK and Gorran until
an agreement was found in May 2019. Gorran joined the cabinet and voted
with the KDP to elect Nechirvan Barzani as President of the region and
Masrour Barzani as the Prime Minister.

411
Gorran’s Hoshyar Omar explains that the party properties were registered under
Nawshirwan Mustafa’s name – and are now controlled by his sons – to avoid the
risk of having them seized in case Gorran’s licence was revoked. “For that purely
legal requirement this [Zargata] hill was not transferred to the party.” Interview with
Hoshyar Omar (Sulaymaniyah, 2018).
412
One of them was Qadir Haji Ali, a well-respected veteran of the liberation war
and early follower of Nawshirwan Mustafa. Interview Qadir Haji Ali (Sulaymaniyah,
2019). For the flight of activists and cadres, see Fazel Hawramy, ‘Iraqi Kurdistan’s
Movement for Change Faces Rebellion from Within’, Al-Monitor, 9 July 2018
https://www.al-monitor.com/pulse/originals/2018/07/change-movement-
nawshirwan-mustafa-kurdistan-iraq.html>
413
Independent journalist Kamal Chomani attributes Gorran’s shift to the
opportunistic tendency expressed largely by the parliamentary factions and the
upper cadres that fear that are ones likely to gain ministerial posts and cabinet
positions when Gorran joins the government. Interview with Kamal Chomani
(personal communication, 2019). For the debate within Gorran, see Zmkan Ali
Saleem and Mac Skelton, ‘Protests and Power: Lessons from Iraqi Kurdistan’s
Opposition Movement | Middle East Centre’, LSE Middle East Centre Blog, 10
November 2019 <https://blogs.lse.ac.uk/mec/2019/11/10/protests-and-power-
lessons-from-iraqi-kurdistans-opposition-movement/>.
196
The co-optation of Gorran into the political establishment marked the end
of the hopes for a political change within the framework of the KRG. The
collapse of the electoral turnout shows that the appeal to the nationalist
feeling is no longer effective. In the meantime, political opposition was more
and more expressed by public demonstrations. Interestingly enough,
Shaswar Abdulwahid’s New Generation Movement – the biggest opposition
party since Gorran joined the government – radically shifted away from
nationalism and blamed the KRG’s lack of cooperation with Baghdad for the
problems of the region. New Generation’s Aram Saeed explains:

it's very important to have partners in Iraq. […] We can help Basra,
Basra can help us… we think the nationalism of the KDP and PUK is
a very bad tradition. We discussed with [Muqtada al-]Sadr and
[Haider al-]Abadi, we don't want to make a coalition of one
nationality against other nationalities, or Shi'as against Sunnis. We
need to pass this kind of thinking.414
Before establishing his highly personalistic and self-funded party, Shaswar
rose to political prominence at the time of the referendum when he launched
a No, for Now campaign heavily pushed by his own TV channel NRT.

Moreover, the crisis of legitimacy of the ruling parties went hand in hand
with the rising popular sympathy, particularly among the youth, towards the
more militant PKK and its Syrian-sister party the Democratic Union Party
(PYD) due to their emphasis on “social justice, freedom, and women’s
liberation”.415 The PKK is particularly effective at conveying the image of the
guerrillas’ frugal life in the mountains in stark contrast with the opulence of
the Iraqi Kurdish politicians.416

Conclusion

The KRG elections of September 2018 and the solid victory of the KDP and
PUK suggest that the Kurdish elite has been able to survive the economic

414
Interview with Aram Saeed (Sulaymaniyah, 2018).
415
Interview with Kamal Chomani (personal communication, 2019).
416
Interview with Kamal Chomani (personal communication, 2019).
197
and political crisis that started in 2014 and that triggered the greatest threat
to their power since the liberation of the region in 1991. Despite their size
and frequency, popular protests have not been able to evolve into a political
alternative to the ruling party. Both the ferocity of the KRG-led repression
and the weakness of the opposition parties contributed to this failure. The
KDP and PUK are still in control of the KRG institutions and its security forces
while their greatest political challenger Gorran has been weakened and co-
opted. In March 2019, as his last act as Prime Minister, Nechirvan Barzani
announced the end of the austerity measures imposed after 2014.417 The
current relations of power in the region seem to allow the Kurdish ruling
class to continue its strategy of accumulation based on corruption and
misappropriation of public resources. The political elite that organises the
interests of and mediates within the ruling class defused the challenge
coming from the parliamentary opposition though repression and cooptation
ensuring the survival of the system. However, the crisis and the referendum
have changed the economic picture of the region. The crisis left the KRG
with a huge debt towards foreign corporations including $3 billion owed to
oil companies.418 The pre-crisis plans to build a strong oil sector independent
from Baghdad has been dashed by Turkey’s opposition to the referendum
and its cooperation with the Iraqi government. The KRG has to send its oil
to Baghdad and to rely again on the constitutional budget transfer, now
reduced from 17 per cent to just 12.5 per cent, with the addition of the oil
that the Kurds manage to smuggle beyond the border.419

All these conditions raise doubts about the KRG’s ability to rebuild the rentier
system that was in place until 2014. With less cash available for

Kosar Nawzad, ‘Kurdistan Ends Unpopular Austerity Measure, Will Pay Public
417

Salaries “in Full”’, Kurdistan 24, 8 March 2019

<https://www.kurdistan24.net/en/news/34b6f15b-f457-4a18-a76f-
01531dab5475>.
418
Lawk Ghafuri, ‘Before Sending Oil, Kurds Want Baghdad to Clear Their Debts’,
Rudaw, 23 September 2019
<https://www.rudaw.net/english/kurdistan/230920193>.
419
Omar Sattar, ‘New Pipeline in Works to Transport Iraqi Oil to Turkey’, Al-Monitor,
18 September 2019 <https://www.al-monitor.com/pulse/originals/2019/09/iraq-
turkey-pipeline-kirkuk-ceyhan.html>.
198
redistribution and the same needed to keep the power structure of the
region intact, the KRG will struggle to recover a degree of political stability
especially since the strategies deployed by the KDP and PUK have heavily
undermined their legitimacy to rule in the eyes of the wider Kurdish public.
A long-term victim of the crisis seems to be Kurdish nationalism itself as the
ideological source of political legitimacy in the Iraqi Kurdistan.

199
Chapter 7
Underdevelopment and the Kurdish
Question in Cold-War Turkey
(1946-1987)

Introduction

After the defeat of the Kurdish feudal revolts of the interwar period, the
political dynamics and class structure of the Kurdish regions of Turkey and
Iraq diversified substantially as they were shaped by the different nation-
building projects promoted by the two countries. This chapter discusses the
development of the Kurdish movement in the Turkish republic from the
suppression of the interwar-period revolts to the rise of the PKK in the
1980s. After the defeat of the feudal revolts of the 1920s and 1930s, Kurdish
nationalism in Turkey was brought to silence while the Turkish state worked
to erase any expression of Kurdish identity. With the beginning of multi-
party democracy in 1950, the traditional elite of the Kurdish region was
gradually re-integrated into Turkey’s political system. Their authority over
rural areas allowed them to control the vote of vast peasant and tribal
constituencies, making their contribution precious to nationwide
conservative parties. In these years, the economic subordination of the
Kurdish region to the more developed western Anatolia, combined with the
mechanisation of agriculture, transformed Kurdish society, increasing
landlessness and rural migration.

The alliance between the Kurdish aghas and the Turkish state via the
dominant conservative parties was the principal reason why Kurdish political
identity re-emerged within the Turkish left in the 1960s and 1970s. Kurdish
activists started framing the relationship between Turkey and its Kurdish
region in terms of colonialism and the Kurdish elite as the agent class of
Turkish colonialism. This was the decisive step towards the development of
an autonomous Kurdish left with the PKK as its most successful expression.
Unlike its competitors, the PKK developed a strategy aimed specifically at
winning the support of the Kurdish peasantry through violent actions
directed against both the Kurdish landowners and the Turkish security
forces. After the 1980 military coup swept away most of the Turkish and
Kurdish left, the PKK started an insurgency that had a strong base of
support among the peasantry and turned into the biggest military challenge
faced by the Turkish republic since its foundation.

The Kurds in the Early Kemalist Republic

In the years between the proclamation of the Republic in 1923 and the first
democratic elections of 1950, Turkey was ruled by Mustafa Kemal Atatürk’s
Republican People’s Party (CHP) and underwent a process of authoritarian
modernisation. As Chapter 3 showed, the suppression of the feudal revolts
of the 1920s and 1930s became the occasion for Ankara to impose a form
of tight military control over the Kurdish areas of south-eastern Turkey420
allegedly to liberate the region from feudalism and tribalism. This brutal
repression was accompanied by a project of demographic engineering
aimed at diluting and Turkifying the Kurdish population defined by the
Kemalist regime as ‘tribal populace that do not speak the Turkish
language’.421 In the 1930s, over 25.000 Kurds were deported to western
Turkey and replaced by several thousand non-Kurdish settlers.422 With the
aim of destroying the tribal structures of Kurdish society, many of the aghas
and shaykhs involved in the revolts were deported. Obsessed by the threat

420
As explained in the introduction of this thesis, defining the predominantly
Kurdish provinces with precision is an impossible task. Map 4 shows a rough
correspondence between Kurdish presence and Turkish administrative divisions.
421
Cited in Yadirgi, The Political Economy, 181.
422
Ibid., 180-185.
202
of separatism, Ankara was resolute to erase Kurdish identity and, in 1924,
the public use of the Kurdish language was prohibited.

Repression and deportations took place in a region already devasted by war.

Eastern Anatolia had been the frontline of the war against Russia, a theatre
of the First World War that had especially heavy repercussions for the
civilian population. The perception of the Christian element as an internal
enemy pushed the Ottoman government to undertake a systematic
campaign of massacres and deportation of the local Christians which
developed into the genocide of one to two million Armenians and other
Ottoman Christians. The effect of wartime devastation and depopulation on
the economy and social fabric of the region were aggravated by the fact
that the autochthonous Christian population who had been deported and
massacred represented a significant proportion of the artisans and traders
of Eastern Anatolia.

Underpopulated and ruled by martial law, the Kurdish region of Turkey –

home in 1927 to roughly two million people –423 was unable to benefit from
the country’s post-war economic recovery. As explained by Veli Yadirgi, in
the years of Kemalist single-party rule (1923-1950), the gap between
western and eastern Turkey increased enormously. The policy of
industrialisation promoted by Ankara was largely directed at the west of the
country and the share of manufactures located in the south-east dropped
from 17.8 per cent in 1927 to only 7.7 per cent in 1955 while agricultural
production was heavily affected by the lack of machinery and infrastructure
and the scarcity of labour. In 1943 – when Turkish agriculture was booming
– the income for a hectare of land in south-eastern Turkey was half of the
national average.424

The yawning east-west gap was not only due to the uneven distribution of
state resources but also to the effects of the same economic policies
applied to two different contexts. The rural structure of western and central

423
Ibid., 189.
424
Ibid., 188-191.
203
Anatolia was based on small independent farmers and agriculture had been
commercialised since the nineteenth century thanks to a relatively high
degree of infrastructural development. In eastern Anatolia, the
underdeveloped infrastructure and the roughness of the terrain rarely
allowed for agricultural products to be exported outside the region.
Ownership patters were not uniform, with the least productive – generally
mountainous – areas characterised by small farms hardly able to produce
beyond subsistence and the most fertile lands – such as the Diyarbakır
plains – often dominated by large estates.425

The violent quelling of the Dersim rebellion in 1938 marked the end of the
political agitations of the interwar period in the Kurdish region and the
beginning of a period characterised by the virtual absence of expressions
of Kurdish political identity that lasted until the early 1960s. Historian Hamit
Bozarslan offers three reasons for the beginning of this ‘period of silence’ of
Kurdish nationalism. First, the Kemalist repression of the early Kurdish
revolts had been violent enough to discourage open dissent. Second, the
sudden end of the Iranian Kurdish Republic of Mahabad in 1946 showed that
no regional or global power was willing to support Kurdish separatism.
However, the most important reason given by Bozarslan is the partial
opening of the Turkish political system from the mid-1940s. In particular, the
rise of the Democrat Party (DP) and its electoral victory in 1950 allowed the
aghas and shaykhs – the Kurdish tribal elite – to be integrated into the
Turkish political system.426

The ‘Agha-State’ Alliance and the Incorporation of the Kurdish Elite

Turkish sociologist İsmail Beşikçi, who spent cumulatively seventeen years

in prison for his research on Kurdish society, explains that:

once all the focal points of rebellion in Kurdistan had been done
away with, the state presented the “Kurdish ruling classes” with two
alternatives: they were either to take the side of the state, or they

425
Çağlar Keyder, State and Class in Turkey (London: Verso, 1987).
426
Bozarslan, Kurds and the Turkish State, 343-344.
204
 would be sent to the gallows like Sheikh Said and Seyid Rıza. 427
Under these circumstances, the sheikhs, landlords, and tribal chiefs,
in other words, the “Kurdish ruling classes” turned into agents for
the Turkish government.428
This passage from Beşikçi’s International Colony Kurdistan is a vivid
summary of the process of the incorporation of the traditional elite of the
Kurdish region into the Turkish state that took place in the 1940s and 1950s.
The dominant class of the Kurdish region described by Beşikçi had gone
through a process of transformation in the turbulent previous decades and
assumed more defined characteristics at the moment of its integration into
the Turkish political system in the early 1950s.

Most often, large landowners still belonged to the traditional ruling class of
Kurdish tribal society: aghas and shaykhs who did not participate in the
feudal revolts of the interwar period due to individual inclinations, local
rivalries, or geographical distance from the centres of the revolts. The
Kemalist state’s repression was fierce and at times indiscriminate but it by
no means hit all Kurds in the same way. While many aghas and shaykhs were
executed and exiled, many other powerful individuals in the regions were
rewarded for their loyalty to Ankara. One of these individuals mentioned by
Beşikçi was Shaykh Ibrahim Arvas, a member of a prominent Kurdish family
from Van and parliamentarian for the entire single-party period (1920-
1950).429 Beside these representatives of the traditional elite, there were
also Kurds of peasant extraction who had been able to take advantage of
both the deportation of the Armenians and the repression of the Kurdish
revolt to acquire properties left behind and become large landowners

427
The leaders of, respectively, the Shaykh Said revolt (1925) and the Dersim
rebellion (1938). See Chapter 3 for the background and outcome of the interwar-
period Kurdish revolt.
İsmail Beşikçi, International Colony Kurdistan (London: Gomidas Institute, 2015),
428

56.
Volkan Tunç, ‘Van Milletvekili İbrahim Arvas’ın Biyografisi ve Türkiye Büyük Millet
429

Meclisi’ndeki Faaliyetleri’, Akademik Tarih ve Düşünce Dergisi, 6.2 (2019), 388–432.

See also, Beşikçi, International Colony, 77.
205
themselves.430 Nevertheless, the authoritarian nature of the single-party
rule and the militarisation of the south-east did not allow for the full
integration of the Kurdish elite into the political system, except for the
appointment of high-profile personalities like Arvas to parliament. The real
turning point was in 1950 when the establishment of a multi-party system
made the south-east electorally attractive to Turkish parties.

In 1950, forced by domestic and international pressure to hold free

elections, the CHP was defeated by the DP that largely represented the
interests of Turkish farmers gradually alienated by the CHP’s industrial
policies. Adnan Menderes, the leader of the party and the new Prime
Minister was himself a large landowner.431 Without fully breaking with the
secularist tradition of the Republic, the DP appealed to the most
conservative sectors of Turkish society by promising a more liberal religious
policy.432 The new government intuited the electoral significance of the
Kurdish region where tribal structures, as well as the ownership of the land,
gave the local elite the power to control sizable packages of votes. Several
Kurdish tribal leaders like Mustafa Remzi Bucak, Halis Öztürk, Edip Altınakar,
Yusuf Azizoğlu, Ziya Serefhanoğlu were elected with the DP in 1950.
Moreover, the Kurdish agha and shaykhs who had been exiled after their
participation in the interwar-period revolts were allowed to return to the
south-east and re-claim their properties, quickly regained their power and
status. In 1957, the DP elected to parliament Abdülmelik Fırat, the grandson
of Shaykh Said of Palu, the leader of the Kurdish revolt of 1925.433 This policy

430
Examples of these self-made landowners are present in several accounts
written at the time. See Nur Yalman, ‘On Land Disputes in Eastern Turkey’, in Islam
and Its Cultural Divergence: Studies in Honor of Gustave E. von Grunebaum, ed. by
Girdhari Tikku (Urbana: University of Illinois Press, 1971), 180-217 and Vanly,
Survey, 36.
431
Zürcher, Turkey, 212.
432
Ibid., 210-218.
433
Abdülmelik Fırat was to be elected again in 1991 in the ranks of the True Path
Party (DYP), another conservative force. However, he later adopted an open
position in favour of Kurdish rights that cost him the expulsion from the DYP and
even a brief period in prison in 1996. Abdülmelik Fırat, Fırat Mahzun Akar (Istanbul:
Avesta Kitabevi, 2006).
206
marked the beginning of a de-facto alliance between the Turkish state and
the Kurdish elite that was to last far beyond the period of DP rule. From the
1950s onwards, prominent Kurdish families provided most of the members
of parliament elected from the Kurdish provinces and supported the various
conservative majorities that dominated Turkish politics since then. From this
position, the Kurdish elite was able to have their say on Ankara’s policies
towards the south-east and control the flow of state funding and contracts.
This power was used not only to prevent social change – including an
effective land reform – but also to revitalise the traditional social structure
that sustained their power.

A major price to pay for the Kurdish elite to enjoy the benefit of the alliance
with the state was to renounce to any public – not to mention political –
expression of Kurdish identity.434 South-eastern politicians had to be
extremely careful: The accusation of pursuing a secret pro-Kurdish agenda,
even without any evidence, could end a successful political career, as in the
cases of Health Minister Yusuf Azizoğlu in the early 1960s and Kamran İnan
in the 1970s.435

In Beşikçi’s work since the late 1960s, this class is defined as an ‘agent class’
due to its relationship with the Turkish state. The Kurdish elite – not only
composed of landowners but also of the big merchants and high-ranked
civil servants of the region – was not a ruling class of its own but rather an
agent class: unable to determine the wider economic policies that affected
Turkey’s south-east, their role was that of implementing it and they were
rewarded by the state through “various types of credit, as well as licenses
to distribute consumer goods, open petrol stations and so forth, so that
these agents may increase their influence in the respective areas.” 436
Beşikçi’s idea, influenced by dependency theory, was decisive in the
development of the approach within the political left that identified

434
Vanly, Survey, 36.
435
Güneş Murat Tezcür, ‘Kurdish Nationalism and Identity in Turkey: A Conceptual
Reinterpretation’, European Journal of Turkish Studies, 10 (2009), 10.
436
Beşikçi, International Colony, 56.
207
Kurdistan as an internal colony of Turkey. The Kurdish region then
constituted a periphery of the Turkish economy supplying agricultural
products, raw materials, and cheap labour to the growing industry of
western Anatolia. In The Order of the East,437 first published in 1969, Beşikçi
rejected the official Kemalist narrative that presented the ‘feudal structure’
of eastern Turkey as the cause of its underdevelopment. On the contrary,
Beşikçi’s sociological work showed that it was precisely the policy of the
Turkish state that systematically strengthened the traditional social
structures of the region with the multiple purposes of preventing the
emergence of Kurdish separatism, serving the economic interests of the
Turkish core, and providing electoral support to the ruling conservative
parties.

Despite the strengths of this argument, describing the Kurdish elite as an

‘agent class’ – in dependency-theory terms a ‘comprador’ class – is
problematic. Rather than the agent of a foreign colonial power, the Kurdish
elite was incorporated in the ruling social bloc dominated by the industrial
bourgeoisie of western Turkey and the Kemalist bureaucracy.438 Even if the
Turkish state was primarily organised around the interests of those groups,
the ruling class of the Kurdish region benefitted from its position as part of
the ruling bloc. They guaranteed the subordinate position of the south-east
within the country’s political economy and provided flocks of conservative
MPs. In exchange, they enjoyed the protection of the state apparatus which
was willing to turn a blind eye on the violence exercised over the subaltern
classes. This is why the ‘State-Agha’ alliance survived well beyond the 1960
coup that removed the DP from power.

Beşikçi’s early work was decisive in exposing the role of the Turkish state in
maintaining the south-east underdeveloped: a state that, while preaching
modernisation, actively empowered the reactionary elite of its Kurdish

437
İsmail Beşikçi, Doğu Anadolu’nun Düzeni: Sosyo-Ekonomik ve Etnik Temeller
(Ankara: Yurt Kitap-Yayın, 1992).
438
For the reconfiguration of Turkey’s ruling bloc in the DP era see Keyder, State
and Class, 117-140.
208
periphery. This approach to the political economy of the Kurdish region as
an internal colony of Turkey was to be widely popularised by the Turkish
and Kurdish left in the 1970s.

The Underdevelopment of the Kurdish Region

The alliance between the central government and the local elite had deep
transformative consequences on the relations of power of the Kurdish
provinces. When the Democrat Party took office in 1950, most Kurdish
peasant families still owned small plots of land and lived off subsistence
agriculture. However, where the land was more productive, such as in the
Diyarbakır plains, more than a third of the peasant families were landless
and worked as sharecroppers (yarıcı) or agricultural labourers in large
estates.439 The DP government had an ambitious programme for agricultural
development based on the provision of credit to large farmers, the
distribution of state-owned land and the maintenance of high prices for
agricultural products.440 These policies paid off: from 1947 to 1953, the
country’s agricultural output more than doubled.441 However, the DP’s
agricultural policies were explicitly aimed at rewarding large, highly-
productive, and export-oriented farms. The previous – hardly enforced –
landownership limit of 500 dunums (50 hectares) was increased in 1950 to
5000 dunums (500 hectares) which allowed the Kurdish elite to expand their
estates significantly while limiting the amount of productive land available
for redistribution.442

Within this context of growing production and gradual commercialisation of

agriculture, the introduction of the tractor had a great impact on the Kurdish
countryside. In the 1950s, the government used Marshall Plan aid to

439
Van Bruinessen, Agha, Shaikh and State, 22-23.
440
Yadirgi, The Political Economy, 194. See Resat Aktan, ‘Problems of Land Reform
in Turkey’, International Journal of Middle Eastern Studies, 20.3 (1966), 317–34.
Sevket Pamuk, ‘Economic Change in Twentieth Century Turkey: Is the Glass
441

More than Half Full?’, The Cambridge History of Turkey, ed. by Kasaba, 266–300.
442
Yadirgi, The Political Economy, 195.
209
subsidise the purchase of tractors by farmers: there were 2,014 tractors in
the Kurdish region in 1963 and 47,861 by 1982.443 The mechanisation of
Kurdish agriculture was a greatly disruptive process. While almost
exclusively large landowners were able to afford a tractor, most small
farmers were forced to hire the machine and – overwhelmed by its hiring
cost – eventually to sell their lands. In addition to this new wave of landless
peasants, the tractor-owning landowners needed much less labour to run
their farms as a single tractor could replace over fifty peasants working with
oxen and ploughs.444 Expiring sharecropping contracts were not renewed,
and a large portion of rural labourers found themselves unemployed.445
Throughout Turkey, the share of peasant families who did not own land
increased from less than 6 per cent to over thirty per cent between 1950
and 1962.

This process was even more devastating in the Kurdish provinces where the
land was already less evenly distributed while lower infrastructural
development made it more difficult for small farmers to commercialise their
products. By the end of the DP era, each of the south-eastern provinces had
a higher share of landless peasants than the national average with peaks of
55 per cent in Şanlıurfa and 47 per cent in Diyarbakır.446 As late as 1983, an
Agence France-Presse journalist described the life of a poor Kurdish hamlet
in the Mardin province. Located twenty kilometres away from the nearest
road, “the inhabitants are cut off from the world, without a school, without
electricity, without a road, without even a transistor”. The village had no
contact whatsoever with the Turkish administration and the only visible
source of authority was that of the landowning agha. The entire active
population of this and the neighbouring hamlets worked eleven hours a day

443
Very low numbers compared to the national total of 51,781 in 1964 and 491,004
in 1982. Cemal Aladağ, Milli Mesele ve Kürdistan’da Feodalite Aşiret (Frankfurt:
Komkar Yayınları, 1981), 138.
444
Yalman, On Land Disputes, 198.
445
Yadirgi, The Political Economy, 194.
446
Mustafa Sönmez, Kürtler: Ekonomik ve Sosyal Tarih, Doğu Anadolu’nun Hikayesi
(Istanbul: Arkadaş Yayınevi, 1992), 144.
210
for $1-2 in the cotton fields of the 1,600-hectares estate owned by the
agha.447

The transformation of the country’s rural life triggered a tumultuous process

of urbanisation. Turkey’s urban population overall grew from around 18 per
cent in 1950 to 40 per cent in 1980. The eastern provinces present a similar
picture (from less than 15 per cent to over 40 per cent) but, due to their
higher birth rates, also greatly contributed to the urbanisation of western
Turkey.448 In the large cities of the Marmara and Aegean coast, these waves
of Turkish and Kurdish rural migrants constituted a new class of urban poor
and provided cheap labour to their growing industry.

In the 1950s, despite the growth of the agricultural output, the Kurdish
region remained largely underdeveloped and lagged far behind the rest of
the country. During the 1960s and 1970s, when following Turkish
governments promoted import-substitution industrialisation, a large part of
the state investments and nearly all private investments were directed to
other regions of the country. In the Kurdish region, most investments were
directed towards the mining sector and the construction of hydroelectric
power plants even though the raw materials extracted and the energy
produced were largely exported overseas or to western Turkey. The share
of national income of the seventeen eastern and south-eastern provinces –
including non-Kurdish areas – dropped from 10.39 per cent in 1965 to 8.17
per cent in 1979.449

The underdevelopment of the Kurdish region in the 1960s and 1970s was
partly a consequence of the alliance between the central state and the
Kurdish ruling class initiated in the 1950s and continued by all successive

447
Jean-Jacques Cazeaux, ‘Turquie: La Grande Misère Des Kurdes’, Le Monde, 16
June 1983 <https://www.lemonde.fr/archives/article/1983/06/16/turquie-la-
grande-misere-des-kurdes_3076818_1819218.html>. Cited also in McDowall, A
Modern History, 421.
448
Yeşim Arat and Sevket Pamuk, Turkey between Democracy and
Authoritarianism (Cambridge: Cambridge University Press, 2019), 28-54.
449
For a description of the process of ‘de-development’ of the Kurdish provinces
in the 1960s and 1970s see Yadirgi, The Political Economy, 205-213.
211
governments. The growing concentration of land and the mechanisation of
agriculture increased the frequency and violence of rural conflicts in the
region. However, the attempts by the landless peasantry to occupy lands
and prevent the introduction of tractors were systematically frustrated by
the landowners’ hired thugs and the Turkish gendarmerie.450

Kurdish Nationalism and Turkish Socialism

In the 1950s and 1960s, the transformation of Turkey in terms of capitalist

development, urbanisation – but also mass schooling and political pluralism
– led to a gradual re-emergence of the ‘Kurdish question’. Explicit references
to Kurdistan or the Kurds were still a taboo in mainstream Turkish politics
and the political elite violently reacted even to the softest expressions of
Kurdish identity. In response to the poor performance of the economy in the
late 1950s and Adnan Menderes’ authoritarian turn, the Turkish military
overthrew the government in 1960, inaugurating its role as the self-
appointed guardian of the Kemalist tradition. The constitution approved in
1961 – when the junta allowed for new elections – introduced an
unprecedented level of civil liberties. With Menderes and his party gone,
most of the Kurdish elite migrated to other mainstream parties and in
particular to Süleyman Demirel’s Justice Party (AP), the conservative
successor of the DP.

The debates on the state of the Kurds within the Turkish republic took place
outside the framework of mainstream Turkish politics and in two different
political spaces on the right and the left. On the one hand, a minority of the
Kurdish elite was inspired by the development of the Iraqi Kurdish
movement led by a traditional figure such as Mullah Mustafa Barzani and
framed the Kurdish issue in exclusively nationalist terms. On the other hand,

Yadirgi, The Political Economy, 204. Numerous examples of land conflicts in this
450

period are provided by van Bruinessen, Agha, Shaikh and State, 313-315; Yalman,
On Land Disputes, 199-213; Kendal Nezan, ‘Kurdistan in Turkey’, in A People
Without a Country: The Kurds and Kurdistan, ed. by Gerard Chaliand (London: Zed
Books, 1993), 91.
212
the emergence of a Turkish socialist left created a space for a new
generation of Kurds to link the state’s anti-Kurdish policies to the
reactionary forces that ruled the Kurdish region.

With the repression of the Kurdish revolts of the interwar period, many of
the Kurdish ‘feudal’ nationalists that had animated the first Kurdist
organisations went into exile. Within Khoybûn (independence), an
organisation founded in Lebanon in 1927, exiled Kurdish aristocrats like the
Bedirkhan brothers, Nuri Dersimi, and İshan Nuri found a space to write
about and ‘to imagine’ the Kurdish nation. Despite the limited political
relevance of these figures, their writings created the canon of modern
Kurdish nationalism including “a map; a unified historical narration; a flag; an
idea of martyrdom and glorification of martyrs; the myth of Kawa, liberator
of the Kurds; the notion of Mesopotamia as the cradle of the Kurdishness”.451
Influenced by the primordialist approach of contemporary Turkish
nationalist literature,452 these writers in exile naturalised the Kurdish nation
creating an intellectual legacy that was to influence the development of the
Kurdish national movement.

It was in continuity with this tradition that the years of relative liberalisation
that followed the 1960 coup saw the blossoming of Turkish-language
publications on Kurdish history, literature, and poetry – but also more timidly
on politics and society – such as in the liberal Turkish magazine Barış
Dünyası (World Peace) in 1962. Attempts to publish journals in Kurdish were
however frustrated by state censorship as shown by the short life of the
bilingual monthly Dicle-Fırat (Tigris-Euphrates) and the newspaper in
Kurdish Deng (Voice) and Riya Newe (New Path). Despite state repression,
these attempts showed a widespread renewed interests in Kurdish culture.
As Kendal Nezan, at the time a student in Ankara, recalls

From time to time there were cultural evenings, like 'Bitlis Cultural
Evening', people would go, listen to the music and to some

451
Bozarslan, Kurds and the Turkish State, 344.
452
An influence evident in Celadet BedirKhan’s choice to follow the footstep
Atatürk’s Turkish language reform and create a Kurmanji Latin alphabet.
213
 speeches, they were playing Kurdish songs and it was a way to bring
people coming from the same places together.453
It was within this intellectual context that the first explicitly – and yet
clandestine – Kurdish nationalist party of modern Turkey was established in
1965. Inspired by the Iraqi Kurdish leader Mullah Mustafa Barzani and his
KDP, Faik Bucak established the Kurdistan Democratic Party of Turkey
(PDKT). Bucak was an agha from Siverek – whose brother Mehmet Tevfik
Bucak had been a member of parliament for the DP – and the new party
aimed at the traditional Kurdish elite. For Derwich Ferho, whose brother was
the founder of the PDKT in Mardin, “the PDKT was the expression of the
class of the aghas and they had many followers among the mullahs and
imams.”454 However, despite its conservative nature, the PDKT served as a
training ground for more radical Kurds especially after the assassination,
under unclear circumstances, of Faik Bucak in 1966.455 The leader of the
left-wing of the PDKT, Dr Şivan moved to Iraqi Kurdistan in a very early
attempt to organise a guerrilla movement but was killed in 1971.

Dr Şivan’s attempt must be understood in relations to the wider

development of the political left. The 1961 constitution authorised the
creation of independent trade unions and the first socialist party, the
Workers’ Party of Turkey (TİP). At the 1965 elections, the TİP won 15 seats
mostly from the industrial areas in the west but three of them also from the
Kurdish south-east. Without explicitly referring to the Kurds, the TİP
emphasized the so-called ‘Eastern Question’ (Doğu Sorunu) and the
oppression faced by the people living in the east of Turkey.456 This position
won the sympathy of leftist-minded Kurds that were critical of the Kurdish

453
Interview with Kendal Nezan (Paris, 2019).
454
Interview with Derwich Ferho (Brussels, 2019). According to Bozarslan, the
PDKT “essentially brought together some urban notables, craftsmen and Kurdish
ulama” Hamit Bozarslan, ‘Political Aspects of the Kurdish Problem in Contemporary
Turkey’, in The Kurds, ed. by Kreyenbroek and Sperl, 98.
455
Michael Gunter, The Kurds in Turkey: A Political Dilemma (Boulder: Westview
Press, 1990), 16.
456
In its first congress in 1964, the party recognised that the people living in the
east “who speak Kurdish or Arabic” were discriminated. Cited in Güneş, The
Kurdish National Movement, 59.
214
nationalism promoted by the PDKT. The active presence of Kurds within the
TİP – known as ‘Easterners’ (Doğulular) – gradually pushed the party to
adopt an even more advanced position. In their 1970 congress documents,
the TİP stated that “there is a Kurdish people in the East of Turkey” that “the
fascist authorities representing the ruling classes have subjected […] to a
policy of assimilation and intimidation which has often become a bloody
repression.”457 This resolution cost the party dearly and in 1971 the TİP was
accused of supporting separatism and banned by a Turkish court.

Both sides of this emerging interest in the Kurdish question in Turkey

participated in the mass movement that developed in the Kurdish region in
the late 1960s. In 1967, a series of demonstrations called the Eastern
Meetings (Doğu Mitingleri) took place in Diyarbakır and other Kurdish cities
in which the themes of underdevelopment and ethnic oppression welded
together. In 1969, following this first phase of mass mobilisation, Kurdish
students in Ankara and Istanbul established the radical Revolutionary
Cultural Societies of the East (DDKO), the first ‘easterner’ organisation with
the aim of a socialist revolution, and that soon opened branches across the
south-east. The DDKO seems to have been composed of a wide range of
people including cadet members of the landowning class who had turned
leftist during their university years.458 In their publications, they denounced
the poverty of the south-east, the oppression of the peasantry by the
traditional Kurdish elite and the violence of the Turkish state in the region.459

All of these developments were temporarily interrupted by a new military

intervention in 1971. With the alleged aim of restoring order, state repression
heavily targeted the political left – banning both TİP and DDKO – and the
trade unions. The growing and more explicit interest of Turkey’s left in the

457
Vanly, Survey, 53-54.
458
“Many of the people who were influential in the democratic and revolutionary
movement developing in Northern Kurdistan in the 1960s were the younger
generation of the landowning classes […] who were definitely opposed to the kinds
of relationships their fathers and grandfathers had formed with the Turkish state.”
Beşikçi, International Colony, 84.
459
Gunter, The Kurds in Turkey, 18.
215
Kurdish question had the effect of combining the traditional anti-
communism of the Kemalist elite with the fear of ethnic separatism. This
attitude was visible in the far more indulgent treatment accorded to the
violent Turkish nationalist groups that were assertively imposing their
presence on the streets.460 The 1971 intervention was, thus, the moment in
which the rightward orientation of the military – in the form of chauvinism
and anti-separatist paranoia – came to coincide with the interest of an
emerging industrial capital in western Turkey determined to put down a
growing and emboldened labour movement.461 The most significant
consequence of these events was the realisation, by many Turkish and
Kurdish leftists, that no solution to their problems could come from the
Kemalist and capitalist state.

The Kurdish Left in the 1970s

The military intervention of 1971 was mostly directed against the rising
street violence and political instability and was successful in limiting the civil
liberties granted by the 1961 constitution and in breaking the labour
movement. Nevertheless, rather than a return to political stability, 1970s
Turkey experienced a much more intense wave of street violence and an
economic crisis that the twelve weak coalition governments that took office
between 1971 and 1980 were unable to address. The policy of import-
substitution industrialisation generated a chronic lack of foreign currency
reserves that was aggravated by the spike in the oil price after 1973. The
government responded by imposing import restrictions while inflation and
unemployment were on the rise for the entire decade. The process of mass
urbanisation accelerated in the 1970s increasing the number of urban poor
often living in shantytowns – called gecekondu – at the outskirts of the main

460
Zürcher, Turkey, 258-260.
461
Membership to unions in Turkey increased from 296,000 in 1963 to 1.2 million
in 1971, while the number of working days lost for strikes rose from 12.255 to
295.950. Sosyalizm ve Toplumsal Mücadeleler Ansiklopedisi, Volume 5 (Istabul:
İletişim Yayınları, 1988), 2147.
216
cities. These informal settlements became an ideal recruiting ground for the
radical organisations of the left as well as for the rising Islamist and far-right
nationalist groups. At the same time, mass education allowed more and
more Turks with a working-class background to access high-school and
university degrees and to seek public employment.

The suppression of the TİP and the strongest unions strengthened the
radical character of the student and labour movement that was now even
more proactive in trying to violently win the control of the streets and the
campuses. Turkey’s new left “was characterised foremost by its fecund
parthenogenesis and the resulting sectarianism” that in the 1970s led to the
multiplication of groups with virtually identical – Leninist, Maoist, Guevarist
– ideological orientations.462 Their radicalism was fuelled by the admiration
for the revolutions taking place in the Global South and particularly by “the
Palestinian struggle [which] exercised a magnetic appeal to all revolutionary
groups.”463 This radicalism, however, prevented their interactions with the
established political system. The left was opposed by a much less
fragmented far-right youth organised around Alparslan Türkeş’ pseudo-
fascist Nationalist Action Party (MHP) and the Islamist National Salvation
Party led by Necmettin Erbakan. Unlike the radical left that lacked any
interlocutor in institutional politics, the parties of the extreme right joined
Süleyman Demirel’s AP in several coalition governments and used their
ministerial posts to pack the civil service with their youth. The colonisation
of the security forces by the ultra-nationalist MHP made the police passive
or even complicit in neo-fascist street violence.464 Along with the Turkish
left, Kurdish activists and religious minorities were natural targets.465

As mentioned earlier, the 1971 military intervention exposed the democratic

limits of the Kemalist state and thus had a radicalising effect on the Kurdish

462
Keyder, State and Class, 210.
463
Interview with Cengiz Çandar (personal communication, 2020).
464
Zürcher, Turkey, 262-263.
In 1978, more than one hundred Alevis – Turkey’s Shia minority – were killed by
465

neo-fascists and Islamists militants in a pogrom in Kahramanmaraş.

217
youth. The establishment of the DDKO in 1969 had already represented a
decisive step in the process of ‘autonomisation’ of the Kurdish movement.466
Socialist Kurds were less and less satisfied with the position of the Turkish
left that, though acknowledging the oppression of the Kurds, subordinated
their liberation to the socialist transformation of Turkey. While maintaining a
Marxist framework, the Kurdish leftist groups of the 1970s identified in the
‘colonial condition’ of the Kurdish region a specific form of oppression to
which the Kurds in Turkey were subject and the Turkish working class and
peasantry were not. As Kemal Burkay, founder of the Kurdistan Socialist
Party (PSK), wrote in the early 1970s, “in Kurdistan, the feudal relations have
not been defeated and a bourgeois democratic revolution has not occurred.
Therefore, the main contradiction for the Kurdish people is national.”467
Burkay rejected the idea – dominant among Turkish leftists – that Kurdish
nationalism could only be the expression of the Kurdish feudal class and
proposed an anti-colonial nationalism as a progressive and democratic
force.468

This line of thought became the dominant discourse among Kurdish leftist
after the suppression of the TİP and the DDKO in 1971. Burkay’s PSK was
still trying to move within the margins of official politics and, at the 1977
Diyarbakır mayoral elections, helped the tailor Mehdi Zana defeat all the
notables running for the Turkish mainstream parties.469 However, very much
like the Turkish left, more radical Kurdish groups multiplied in the mid-1970s
despite sharing a similar understanding of the colonial condition of
Kurdistan, inspiration in the anti-colonial struggles, and sympathy for
Maoism. Among these groups, Rizgari (liberation) evolved from a pro-

466
Marlies Casier and Olivier Grojean, ‘Between Integration, Autonomization and
Radicalization. Hamit Bozarslan on the Kurdish Movement and the Turkish Left’,
European Journal of Turkish Studies, 14 (2012), 6.
467
Cited in Güneş, The Kurdish National Movement, 72.
468
For an in-depth discussion of this ideological shift, see Güneş, The Kurdish
National Movement, 66-74.
469
Gilles Dorronsoro and Nicole F. Watts, ‘Toward Kurdish Distinctiveness in
Electoral Politics: The 1977 Local Elections in Dı̇yarbakir’, International Journal of
Middle East Studies, 41.3 (2009), 457–78.
218
Kurdish publishing house in 1975. The Democratic Cultural Association
founded in 1974 established discussion groups throughout Turkey out of
which Kawa (1976) and the National Liberators of Kurdistan (KUK) emerged.
Active in both Turkey’s major cities and the Kurdish region, by the second
half of the 1970s, some of these groups started planning guerrilla activities
against the Turkish state.

The proliferation of far-leftist Kurdish groups was the breeding ground for
the establishment of the Kurdistan Workers’ Party (PKK), the organisation
that ultimately became hegemonic within the Kurdish movement in
Turkey.470 The group gathered in the early 1970s in Ankara around university
drop-out Abdullah Öcalan. In 1975, Öcalan and his followers established
their presence in the Kurdish region to create a network of support in
preparation of the liberation war but it was only in April 1979 that they
officially adopted the name PKK.471 By then, Öcalan’s followers had already
been involved in some low-key military actions that were symptomatic of
the broader characteristics of this pre-uprising phase. First, already in 1977,
they were involved in a feud with another revolutionary group named
Tekoşin which speaks to the clannish attitude of these organisations.

470
The group was initially known as Apocular, ‘the followers of Apo’, Öcalan’s
Kurdish nickname. Ahmed Akkaya and Joost Jongerden, in their writings on the
origins of the PKK, claim that Öcalan’s group was ‘born from the left’ in the sense
that they were less the product of the Kurdish movement than they were of the
Turkish left. In the early to mid-1970s –before the establishment of the PKK in 1978-
1979 – Öcalan and his followers were part of the Ankara Democratic Higher
Education Association (ADYÖD) rather than in the numerous radical Kurdish groups
active at the time. Joost Jongerden and Ahmet Akkaya, ‘Born from the Left: The
making of the PKK’, in Nationalisms and Politics in Turkey: Political Islam, Kemalism
and the Kurdish Issue, ed. by Marlies Casier and Joost Jongerden (London and
New York: Routledge, 2011), 123-142.
471
Mehmet Ali Birand, APO ve PKK (Istanbul: Milliyet Yayinlari, 1992), 85-96. The
long gestation of the PKK is explained by Jongerden as motivated by the need to
overcome the ideological and organizational weakness of the Turkish and Kurdish
left and to go through a long period of preparation before starting the insurgency.
Joost Jongerden, ‘A Spatial Perspective on Political Group Formation in Turkey
after the 1971 Coup: The Kurdistan Workers’ Party (PKK)’, Kurdish Studies, 5.2
(2017), 134–56.
219
Öcalan – who at the time could count on 250-300 followers –472 became
increasingly known among other groups for his brutal treatment of his
rivals.473 The elimination of rival Kurdish groups was part of the strategy to
become the only point of reference for the Kurdish masses. But to win a
popular base, the PKK had to direct its attention towards the exploitative
components of Kurdish society. In 1978, Öcalan’s followers successfully
mobilised people from the town of Hilvan against their landowners
belonging to the pro-government aghas of the Süleymanlar tribe. The
success of this action convinced the leadership to insist on this strategy. To
announce the foundation of the party, the PKK decided to perform a highly
spectacular action, in July 1979, attempting to kill Mehmet Celal Bucak, a
powerful landowning agha from Siverek famous for his brutality against the
peasants and the local leftists. Bucak survived the attack but the action
resonated widely among the Kurdish peasantry.474

Like the other leftist Kurdish groups, the PKK adopted the ‘colonial thesis’
as a starting point for their analysis and the development of their
revolutionary strategy. The pre-capitalist feudal structure of Kurdish society
was described as functional to Turkish colonialism as “tribal-feudal interests
are being watched and forced to live to deepen social fragmentation” and
prevent the development of Kurdish identity.475 As the agha class was
complicit with the Turkish state and a structural component of colonial
domination, the Kurdish working class and peasantry were the only actors
able to conduct a revolution that was at the same time national and

Aliza Marcus, Blood and Belief: The PKK and the Kurdish Fight for Independence
472

(New York: New York University Press, 2007), 38-39.

Aliza Marcus particularly insist on this aspect and supports her argument with
473

numerous interviews. Ibid., 40-42.

474
The Bucak family already appeared twice in this chapter and it is a well-known
example of leading tribal family that could express members of both pro-state
aghas like Mustafa Remzi Bucak, DP parliamentarian in the 1950s, or Faik Bucak, a
Kurdish nationalism, though of conservative orientation. This position in between
the state and the nationalists is very similar to that of the Barzanji in Iraqi Kurdistan,
as discussed in Chapter 4.
475
Abdullah Öcalan, Kürdistan Devriminin Yolu (Cologne: Weşanen Serxwebun,
1993), 63-64.
220
socialist.476 This approach was strengthened through the ‘lesson learned’ by
the failure of the Barzani revolution in Iraq in 1975. 477 In Öcalan’s reading of
these events, the Kurdish traditional elite had only been able to formulate
an autonomist project and, unable to mobilise the subordinate classes,
became dependent on the support of imperialist powers. 478 However, in his
approach to colonialism, Öcalan went further due to the influence of
Antillean philosopher Frantz Fanon. The struggle against the coloniser
becomes, as in Fanon, also a struggle against oneself as a colonised
subject.479 After a long digression on Kurdish history, the 1978 PKK
Manifesto argued that violence was the only way through which the Kurdish
people could free themselves from their history of enslavement by foreign
rulers.480

The radicality of Öcalan’s Fanonian approach was not the only specific
feature of the PKK. Derwich Ferho, a member of a prestigious tribal family
and a student in Midyat (Mardin) when the PKK started operating in the area,
claims that the PKK was unique among the Kurdish organizations:

The PDKT was the expression of the class of the Aghas and they
had many followers among the Mullahs and Imams. But you can say
that all the [Kurdish] organizations had the same problem. The
DDKO, DDKD, the KUK... all of them! The leaders of all these groups
were from the upper classes. Except for the PKK who were university
students but - like a friend of mine from Midyat who could never go
to university – also just simple people.481
Öcalan and his followers were characterized by a particularly humble
background. Observing the development of the movement in the mid-
1980s, anthropologist Martin van Bruinessen noticed that the PKK was “the
only organisation whose members were drawn almost exclusively from the
lowest social classes — the uprooted, half-educated village and small-town

476
Güneş, The Kurdish National Movement, 86-91.
477
For the 1961-1975 Barzani revolution, see Chapter 3.
478
Güneş, The Kurdish National Movement, 89.
479
Frantz Fanon, The Wretched of the Earth (New York: Grove Press, 1991).
480
Öcalan, Kürdistan, 20-29.
481
Interview with Derwich Ferho (Brussels, 2019).
221
youth who knew what it felt like to be oppressed, and who wanted action,
not ideological sophistication.”482 Öcalan was born to a peasant family in
Ömerli, a village in the Şanlıurfa province, an hour away from the closest
school. He managed to attend a vocational school in Ankara, in 1966, that
allowed him to graduate high school, work for the civil service, and even
enter Ankara University in 1971. That, in the 1970s, Öcalan’s followers had
largely a similar background is confirmed by both Aliza Marcus’ book, based
on a wealth of interview material, and by Güneş Murat Tezcür who has
collected biographical information of PKK fighters from the obituaries
published on the party’s magazine Serxwebûn. The 142 ‘martyrs’ of the early
stages of the PKK – up to 1980 – paint the picture of an organisation largely
reflecting the class location of its leader. Like Öcalan, most of the killed
militants share a plebeian or rural background and yet held high-school or
college degrees.483 Öcalan attracted a group of young Kurds of peasant
origins who had the chance to partly lift themselves through education and
were largely “university and teacher’s school students or drop-outs.”484

The specificity of their background in the peasantry distinguished the PKK

from the rest of the Kurdish movement in Turkey. The left-ward turn of
Kurdish activists since the 1960s had allowed them to win vast popular
support in the Kurdish towns and cities as evidenced by Mehdi Zana’s
victory in Diayarbakır in 1977. The leaders of these groups were either
cadets of the Kurdish elite or members of the urban working classes – Zana
was a tailor. However, the Kurdish countryside remained the realm of feudal
landowners, where the local agha controlled the votes of the peasants living
in the villages he literally owned. The peasant origins of Öcalan and the early
PKK cadres provided them with a specific sensitivity absent in the rest of
the movement as well as the language to speak with the peasantry. The
language was not solely the Kurdish language to win the villagers to the

Martin van Bruinessen, ‘Between Guerrilla War and Political Murder: The
482

Workers’ Party of Kurdistan’, MERIP Middle East Report, 153 (1988), 41.
483
Güneş Murat Tezcür, ‘Violence and Nationalist Mobilization: The Onset of the
Kurdish Insurgency in Turkey’, Nationalities Papers, 43.2 (2015), 256-257.
484
Marcus, Blood and Belief, 37.
222
national cause but more importantly the language of performative anti-
landlord violence. The decision to celebrate the establishment of the PKK
with the highly-visible attack on Bucak was intended to speak directly to the
peasantry, showing them that, besides the sophisticated theories on
colonialism and nationalism, the PKK was willing and able to act against the
people directly responsible for their misery.

In the same days of the Bucak attack, in July 1979, the PKK decided to
relocate to Syria to organise for a long-term and large-scale insurgency in
Turkey. Öcalan was in Damascus when the Turkish military, in September
1980, carried out another coup d’état that swept away anything resembling
leftist politics in the country. The violence of the repression against the
Turkish and Kurdish left eliminated most of PKK’s rivals. As Remzi Kartal,
who had been a member of Dr Şivan’s PDKT, explains “the coup was like a
bulldozer: it eliminated all the political organisations, only the PKK survived.
After that, we started looking at the PKK.”485

Preparing the Kurdish Revolution

The military coup of September 1980 marks a watershed moment in Turkish

history. The Turkish military’s declared purpose was to end street violence
and political instability as well as to ostensibly rescue the Kemalist republic
from the combined threats of communism, Islamism, and Kurdish
separatism. As the next chapter will show, the 1980 coup also had the
function of implementing austerity measures deemed indispensable to
overcome the crisis that Turkey’s economy faced in the 1970s. The military
showed their clear intention to break with the previous political system,
disbanding the political parties and arresting the leading politicians. In the
twelve months following the coup, 122,600 people were arrested, and street
violence decreased to a negligible level.486 Even if, like in the 1971 coup, the

485
Interview with Remzi Kartal (Brussels, 2019).
486
Zürcher, Turkey, 279.
223
political left was by far the biggest target,487 the Kemalist elite tended to
present the coup as a last-resort defence of the secular republic against the
Islamist threat that had just overwhelmed Iran. However, while the Turkish
left and labour movement were to never fully recover from the 1980 coup,
both the Islamists and the ultra-nationalists re-emerged in the late 1980s
and 1990s with the same leaders – Erbakan and Türkeş – and the
comparable electoral strength of the 1970s.

Alongside the Turkish left, the other principal target of the military
repression was the Kurdish movement. The military took unprecedented –
even for Turkish standards – measures to repress Kurdish identity: the
public and private use of the Kurdish language was banned and thousands
of topographic names of Kurdish origins were changed. 488 The military
influence over education even allowed them to give new life to the old
Kemalist theory that the Kurds were, rather than a people on their own, just
‘Mountain Turks’ (dağ Türkler).489

The PKK had relocated to Syria in 1979 and after the military coup, most
Turkish and Kurdish leftist groups followed suit. At the time, Hafez al-
Assad’s Ba’athist regime acted as a patron to leftist and nationalist
organisations active against Syria’s neighbours, and particularly to the
Palestinian resistance.490 Most of the leftist groups from Turkey were

487
Ibid., 269-270.
488
See: Senem Aslan, ‘Incoherent State: The Controversy over Kurdish Naming in
Turkey’, European Journal of Turkish Studies, 10 (2009).
489
For example, Şükrü Seferoğlu’s 1982 book titled: ‘The first Turkish inhabitants
of Anatolia: the Kurds’. See Şükrü Kaya Seferoğlu, Anadolu’nun İlk Türk Sakinleri:
Kürtler (Ankara: Ayyıldız Matbaası, 1982). See also: Şukrü Kaya Seferoğlu and Halil
Kemal Türközü, 101 Soruda Türklerin Kürt Boyu (Ankara: Türk Kültürünü Araştırma
Enstitüsü, 1982); Bayram Kodoman, Sultan II. Abdülhamit’in Doğu Anadolu Politikası
(Ankara: Ankara Üniversitesi Basımevi, 1987).
490
In 1975, the Iraqi-Kurdish PUK was established in Damascus (see Chapter 4).
Asad also allowed Syrian Kurds to join foreign organisations with the purpose of
ridding the country of potential troublemakers. Polad Jan, from the Syrian Kurdish
movement explains that “the Syrian regime approved that active and smart Kurds
went to the mountains. […] Syria and Turkey were enemies and the Syrians wanted
Turkey to have problems and conflicts. The youth who could create conflicts and
troubles [in Syria] could go to the mountains [in Turkey] and get killed there.”
Interview with Polad Jan (Sulaymaniyah, 2019).
224
extremely demoralised by the complete lack of resistance to the military
coup and showed little interest in preparing for an insurgency. One by one,
the PKK’s competitors left Syria, generally to Europe. Öcalan’s determination
to build up the PKK’s military prowess allowed the party to attract militants
from the other declining organisations and was decisive to establish a
strong bond with the Syrian regime and the Palestinian resistance. Already
in early 1980, Öcalan convinced left-wing Palestinian groups to train PKK
members in their camps in Lebanon’s Beqaa Valley, at the time under Syrian
occupation.491 In 1986, as the number of PKK recruits increased the PKK
opened its own training camp in the Beqaa, the Mahsum Korkmaz
Academy.492 Several of the first post-1980 PKK ‘martyrs’ died alongside the
Palestinians in the 1982 Israeli invasion of Lebanon.493 While the Palestinian
connection enriched the PKK with invaluable guerrilla training, a deal struck
between Öcalan and Barzani’s KDP allowed the PKK to freely move through
the KDP-controlled areas of northern Iraq. Unlike the flat Syrian-Turkish
border, the mountainous nature of the Iraqi-Turkish border made it an ideal
location from which to start guerrilla activities, and in 1983 the PKK sent its
first reconnaissance to enemy territory.494

At the time of the coup, the PKK was already emerging as the strongest
Kurdish leftist organisation. After the 1980 coup, three-quarters of all people
tried for secessionism – virtually all Kurdish activists – belonged to the PKK
and the PKK had the second-highest number of militants under trial, second
only to the People's Liberation Party-Front of Turkey (THKP-C).495 In the
following year, the PKK’s growth in Syria and Lebanon was reflected in the
increasing visibility of the organisation among the Kurds in Turkey partly due

491
Interviewed by Aliza Marcus Palestinian leader Abu Layla remembers: “we had
met other Turkish Kurds and they didn’t seem to be very reliable. […] We thought
that [the PKK] was the most serious group in Turkish Kurdistan. This is why we
kept them.” Marcus, Blood and Belief, 54-58.
492
Ibid., 57.
493
van Bruinessen, Between Guerrilla War and Political Murder, 44.
494
Gunter, The Kurds, 72-74.
495
Feryal Matbaacılık, State of Anarchy and Terror in Turkey (Ankara, 1983).
225
to the complete disappearance of the rival organisations. The PKK inmates
in the Diyarbakır prison organised highly performative actions of resistance
against the regime of torture to which they were subject.496 After one of the
PKK founders, Mazloum Doğan killed himself on March 21st, 1982, eight more
senior cadres self-immolated in a ‘death fast’ in the following months.497 The
PKK propaganda was very effective in spreading information about these
events and in linking heroic acts of resistance to the myth of Kawa the
blacksmith and the celebration of Newroz. March 21st – the day of Doğan’s
suicide – is the first of the Persian calendar and is celebrated by the Kurds
as well as in the wider Iranic world and central Asia. One of the myths on the
origins of Newroz is based on the figure of Kawa the blacksmith, who led
the Medes – the putative ancestors of the Kurds –498 to revolt against their
Assyrian overlords and celebrated the victory by lighting a bonfire on a hill.
Kurdish nationalists politicised the myth of Kawa and the celebration of
Newroz – with bonfires – as symbols of the Kurdish resistance.499

As in its first performative actions in the late 1970s, the PKK continued to
deploy symbolic politics to build its hegemony over the Kurdish movement
and wider Kurdish society. After this long period of preparation, the PKK’s
liberation war officially started, on August 15, 1984, with two coordinated
attacks in the towns of Eruh and Şemdinli, almost 400 kilometres apart. The
Kurdish guerrillas took over the towns for about an hour to announce the
beginning of the insurgency.

496
For a description of the regime of torture in the Diyarbakır prison, see Torture in
the Eighties (London: Amnesty International Publications, 1984), 217-220.
497
Güneş, The Kurdish National Movement, 98-99.
498
See Chapter 1.
499
See Cengiz Güneş, 'Mobilization of Kurds in Turkey during the 1980s and 1990s',
in The Kurdish Question Revisited, ed. by Stansfield and Shareef, 187-198; Delal
Aydın, 'Mobilizing the Kurds in Turkey: Newroz as a Myth' in The Kurdish Question
in Turkey: New Perspective on Violence, Representation, and Reconciliation, ed. by
Cengiz Güneş and Welat Zeydanlıoğlu (London: Routledge, 2014), 68-88.
226
The PKK Insurgency and the Kurdish Peasantry

After the August 1984 attacks, the PKK insurgency proceeded slowly due to
the militants’ inexperience as well as Turkey’s increased military presence in
the region. However, in the years 1984-1986, PKK fighters started to
become a known presence to the local peasants. The PKK propaganda
specifically targeted the villagers presenting the organisation as a force that
opposed the Turkish security forces as well as the exploitative landowners.

In the first period of the insurgency, the PKK expanded its base of support
among the Kurdish peasantry, although it is impossible to measure its
extent. Güneş Murat Tezcür’s database of killed PKK militants paints a
picture – though incomplete – of the background of the first wave of PKK
‘martyrs’. In the period between 1984 and 1989, 31 per cent of the killed PKK
militants were peasants prior to recruitment, while 38 per cent were low-
skilled workers, and 24 per cent were students. However, 99 per cent of
them were born in a village.500 Considering that village-born students
constituted the biggest group of PKK members before 1984, it seems that,
after the beginning of the insurgency, the PKK largely recruited peasants
and rural labourers. Another indication of the class background of the early
PKK is a survey conducted by Turkish police among the 262 PKK-affiliated
inmates in Ankara in 1996 – thus including people captured throughout the
1980s – indicating that 21 per cent of the militants were illiterate or semi-
literate and 39 per cent only finished primary school.501

The PKK’s increasing ability to recruit the Kurdish peasantry and to use the
villages as shelter and logistical support for its guerrilla activity is also
suggested by the response of the Turkish state. In 1985, the Turkish
government established the Village Guards (Köy Korucuları), a paramilitary
militia recruited among the pro-government Kurdish tribes. In a way very
similar to the one employed by Baghdad against the Kurdish insurgency, 502

500
Güneş Murat Tezcür, ‘Ordinary People, Extraordinary Risks: Participation in an
Ethnic Rebellion’, American Political Science Review, 110.2 (2016), 247–64.
501
Cited in Bal and Laçiner, Ethnic Terrorism in Turkey, 29-30.
502
See Chapter 4.
227
the Turkish government and military authorities negotiated directly with
tribal chiefs and landowners who received the funding to pay the guards’
salary, strengthening their power over their fellow tribesmen and the
peasants of the villages they owned.503 For example, McDowall reports that,
in 1992, the head of the Alan tribe in Van Sadun Şeylan raised 500 guards
from his 26 villages receiving $115,000 monthly from the government. 504
Lightly armed, the military function of the Village Guards was to prevent the
guerrilla from finding shelter in isolated hamlets where the Turkish army had
hardly any access. Moreover, the village guards served the purpose of
countering the PKK recruitment activity, by arming pro-government villages
but also to retaliate against the villagers and families with known sympathy
for the guerrillas. Like in Iraq, the system left wide room for abuses: as the
guards' commanders were subject to little supervision and even less
accountability, the Village Guards became an instrument to settle pre-
existing disputes and to serve the private interests of the commanders. The
institution of the Village Guards increased the civil war (Kurds versus Kurds)
dimension of the PKK insurgency and divided Kurdish rural society forcing
peasants to pick a side while massacres of civilians became more frequent.
This dynamic ultimately favoured the PKK, vindicating Öcalan’s opposition
to the Kurdish elite as a collaborationist class.

In the second half of the 1980s, the PKK insurgency intensified exponentially
and, with it, Turkish repression. In 1986, a report drafted by a group of
opposition parliamentarians who visited the Kurdish region described the
south-east as an open-air concentration camp ruled through torture and
brutality.505 In 1987, in a move that publicly acknowledged the condition of
war in the Kurdish south-east, the Turkish government imposed the state of
emergency over the Kurdish provinces of Bingöl, Diyarbakır, Elazığ, Hakkari,
Mardin, Siirt, Tunceli and Van. The area took the name of Governorship of

503
According to Marcus, the monthly salary of the village guards amounted to
35,000 lira (about $70). Marcus, Blood and Belief, 97.
504
McDowall, A Modern History, 424.
505
‘SHP’li Canver’in Güneydoğu İzlenimleri: İskence Ayibi Hepiminiz’, Cumhuriyet,
12 February 1986.
228
the Region in State of Emergency (OHAL) ruled by a military super-governor
based in Diyarbakır and entrusted with martial law powers. Within this legal
framework, the military had the authority to relocate entire communities
anywhere they could serve as sanctuaries for the PKK guerrillas. By 1989,
around 400 Kurdish villages had been evacuated.506 As the next chapter will
discuss in length, the evacuation of Kurdish villages was to be pursued on a
massive scale in the early 1990s triggering a process of forced urbanisation
that transformed the Kurdish region of Turkey.

Conclusion

The chapter discussed the economic transformation of the Kurdish region

of Turkey since its integration into the newly-established Turkish Republic
and the re-emergence of a Kurdish national movement. After the feudal
revolts of the interwar period were crushed, a large part of the Kurdish
landowning elite and tribal aristocracy was incorporated into the Turkish
political system in the 1950s. Ruling conservative parties nurtured the
Kurdish elite which, in exchange, provided electoral support thanks to the
power they held over the peasantry. Unlike the Kurdish elite in Iraq that,
from 1958, felt its position threatened by the post-revolutionary regime,
Kurdish landowners in Turkey had no interest in challenging the central
government. The re-emergence of a Kurdish political identity took place
among urbanised Kurds – both in the Kurdish region and in the rest of Turkey
– and found its natural breeding ground and interlocutor in the Turkish left.
However, most of the Turkish and Kurdish left was swept away by the 1980
military coup.

Abdullah Öcalan’s PKK not only managed to survive the coup but was also
able to start, in 1984, a Kurdish insurgency in Turkey that, in just a few years,
assumed a massive scale. This chapter showed that the success of the PKK
was largely due to its ability to mobilise the Kurdish peasantry against both
their Kurdish landowners and the Turkish state. The development of the

506
McDowall, A Modern History, 426.
229
uprising in the second half of the 1980s drove large parts of the Kurdish
countryside into a state of civil war. The Turkish government responded to
the increasing involvement of the peasantry with the institution of the
Village Guards, mobilising pro-government tribal chiefs and landowners.
This process led to a higher involvement of the civilian population but also
legitimised the discourse of the PKK that presented the Kurdish landowning
elite as a collaborationist class. The next chapter describes the deep
transformation of the class structure of the Kurdish region driven by both
Turkey’s neoliberal turn and by a counterinsurgency strategy aimed at
depopulating the Kurdish countryside. The virtual destruction of Kurdish
rural society severely affected the class bases of the uprisings but also
contributed to extending the hegemony of the PKK far beyond the
countryside.

230
Chapter 8
The PKK Insurgency and the
Transformation of Turkish
Kurdistan (1987-1999)

Introduction

This chapter discusses the transformation of the Kurdish movement in

Turkey in the 1990s, highlighting the profound interaction between the
political events and the structural transformation that occurred in the
country and its Kurdish region. In the 1980s, Turkey was politically
dominated by the military which initiated far-reaching neoliberal reforms
while the Kurdish south-east saw the rise of the PKK-led insurgency centred
around the mobilisation of the peasantry. In the early 1990s, Ankara
deployed a strategy of depopulation of the Kurdish countryside by removing
the peasant population with the aim of destroying the breeding ground of
the Kurdish insurgency. This counterinsurgency strategy had profound
consequences for the social structure of the region and turned a large part
of its peasant population into a class of urban poor. As the character of the
region shifted from predominantly rural to predominantly urban, so did the
Kurdish movement. In the 1990s, the PKK vastly expanded its social base
beyond the peasantry to the urban working and middle classes and became
the hegemonic force of a much wider Kurdish movement. Over the course
of the decade, this process allowed pro-Kurdish parties to win a large part
of the local administrations of the region. In tracing these processes, this
chapter emphasises the interaction between the political events and the
structural transformation taking place in the Kurdish region. Turkey’s
counterinsurgency policy and economic reforms changed the class
structure of the Kurdish region which in turn led to a deep ideological

231
renovation of the PKK. The outcome of this effort was the creation of a new
– and wider – class coalition that allowed the Kurdish movement to gain a
hegemonic position in the region.

Military Rule and Neoliberal Reforms in Turkey

The transformation of the socio-economic structure of the Kurdish region

and the growing influence of the Kurdish movement in the 1990s must be
understood against the backdrop of the military rule and neoliberal reforms
that characterised Turkey in the 1980s.

The 1980 military coup was not only a response to the instability and political
violence of the previous decade but also to the worrying condition of the
country’s economy. The import-substitution industrialisation strategy
promoted by Ankara since the 1960s – as by much of the developing world
– was based on the protection of national industry to foster local production
and the domestic market. This model allowed Turkey to enjoy an average
annual GDP growth of over 3 per cent and to double the size of its economy
between 1950 and 1980.507 As capital accumulation was ensured by state
subsidies and by the state-induced oligopolistic structure of the market, the
large industrialists of western Turkey were willing to grant concessions to
the trade unions. Their factory workers – a small minority of the Turkish
working class – enjoyed relatively high salaries. The rest of the country –
particularly the Kurdish south-east – was largely excluded from these
developments except for its function as a source of migrant labour.

The main shortcoming of import-substitution industrialisation was the

constant balance of payment deficit – due to the subsidised imports – that
was severely aggravated by the four-fold increase of Turkey’s energy bill
following the 1973 oil crisis. Trying to address the endemic shortage of
foreign currency reserve, the government imposed severe import controls

507
Şevket Pamuk, Uneven Centuries: Economic Development of Turkey since 1820
(Princeton: Princeton University Press, 2018), 222-226.
232
causing, for instance, continuous power cuts even in mid-winter.508 In 1979,
inflation was running at a 90 per cent rate while the country’s public debt
had increased five times from the beginning of the decade.509 The
continuous growth in size and militancy of the labour movement – from one
million workdays lost to strikes in 1973-1976, to 3.7 million in 1976-1980 –510
combined with the poor performance of the economy, started to
significantly affect the rate of profit and created the consensus among the
Turkish ruling class for abandoning import-substitution industrialisation and
for an authoritarian solution to the economic crisis.511 In July 1979, the
Turkish government reached an agreement with the International Monetary
Fund (IMF) and the World Bank for a $1.8 billion loan dependent on severe
austerity measures.512 The unpopularity of these measures and a new wave
of strikes and factory occupations made the implementation of the package
impossible for Süleyman Demirel’s minority government.

The economic crisis of the late 1970s – along with the threats to the
established order coming from socialists, Islamists, and Kurdish separatists
–513 led to the military coup of the 12 September 1980, the third in twenty
years. Turgut Özal, a conservative technocrat and former World Bank
employee,514 was appointed deputy Prime Minister in the military-led cabinet
and was entrusted with the implementation of the IMF-led stabilisation
programme. Özal went beyond these early measures and, as Prime Minister
from 1983 to 1989, contributed to the dramatic transformation of Turkey

508
Zürcher, Turkey, 267.
509
Cited in Yadirgi, The Political Economy, 215.
510
Cited in Caglar Keyder, State and Class in Turkey (London: Verso, 1987), 192.
511
For an account of the class dimension of the import-substitution industrialisation
strategy and of its crisis, see Caglar Keyder, State and Class in Turkey (London:
Verso, 1987), 165-196.
512
Zürcher, Turkey, 268.
513
For the political instability leading to the coup, see Chapter 7.
514
Özal had ties with the conservative Naqshbandi order and, in 1977,
unsuccessfully run in the list of the Islamist National Salvation Party. His time at the
World Bank (1971-1973) made him familiar with the emerging Washington
Consensus policies.
233
into a neoliberal export-oriented economy. Under military rule – with the
unions banned and most labour activists in jail – Özal greatly devaluated the
Turkish Lira, liberalised the trade regime, eliminated subsidies and price
controls, and promoted foreign capital.515 These reforms – that turned
Turkey into a model country for the IMF – led to a radical restructuring of
the country’s class structure in favour of capital. The share of wages in the
national income dropped from 35 per cent in 1978 to 20 per cent in 1986
and wage suppression favoured the export-oriented shift of the country’s
economy.516

After imposing a new constitution in 1982, the military allowed for a return
of power to civilian hands and Özal’s new Motherland Party (ANAP) won a
landslide in the 1983 elections. In the decade in which Özal dominated
Turkish politics,517 the exports’ share of the GDP rose from 2.3 per cent in
1979 to 8.6 per cent in 1990. In 1989, Özal further liberalised the exchange
rate and removed restrictions on capital movement, making Turkey’s
economy significantly more susceptible to external shocks. Even if Özal was
elected President of the Republic in 1989, the ANAP lost its majority in the
1991 elections and the return to unruly coalition governments marked the
end of fiscal discipline. Turkey’s economy in the 1990s, vulnerable to sudden
outflows of foreign capital, became increasingly dominated by financial
instability and hyperinflation.518

As described in Chapter 7, the Turkish economy was characterised by

extremely uneven development and by the subordination of the
predominantly Kurdish south-east. The import-substitution industrialisation
period had only a marginal impact on central and eastern Anatolia, as state
subsidies were directed towards the large industries based in western
Turkey. Manufacture in the rest of the country remained marginal, small-

515
Pamuk, Uneven Centuries, 248-249.
516
Cited in Keyder, State and Class, 225.
In 1980-1983 as the Deputy Prime Minister entrusted with the economic reforms,
517

and in 1983-1989 as the Prime Minister.

518
See Pamuk, Uneven Centuries, 249-257.
234
sized, and heavily labour-intensive.519 Turgut Özal’s reforms hit agriculture
which in the Kurdish region represented the source of livelihood for most of
the population. The 1980s were the beginning of a long process of
transformation that will see the share of agriculture in Turkey’s labour force
decreased from 50 per cent in 1980 to 25 per cent in 2015 and the sector’s
output drop from 25 to 8 per cent of the GDP.520 As Özal ended agricultural
subsidies and price control, subsistence agriculture virtually disappeared,
forcing many small farmers to sell their land and increasing the unequal
distribution of land ownership. In 1990, more than half of the agricultural
land of south-eastern Turkey was owned by only 8 per cent of the farming
families, while 38 per cent owned smallholdings between 10 and 50 dünüms
(1-5 hectares), and 38 per cent were landless peasants.521

Officially, the Turkish government and military identified the region’s

underdevelopment as the main cause of the PKK uprising. However, this
underdevelopment was attributed – rather than to Ankara’s policies – to the
backward landownership structure and survival of tribalism. 522 As Turkish
general İlker Başbuğ later explained

The PKK is essentially a ‘peasant movement’. [its] energy comes

from the fact that the PKK has mobilised hundreds of years of anger
of the Kurdish peasants against their emirs, aghas, and beys. This
enormous peasant energy represented by the PKK [was not
generated by] Turkish state, it had rather accumulated over the
centuries against its own rural structures. In fact, the first armed
actions of the PKK started in Siverek, against the head of the Bucak
tribe.523
Yet, Turkey’s neoliberal reforms only increased the concentration of land
ownership and worsened the condition of the peasantry. But, even more
consequentially, the counterinsurgency policies deployed by Ankara

519
Keyder, State and Class, 175.
520
Pamuk, Uneven Centuries, 280.
521
Yadirgi, The Political Economy, 257-258.
522
Ibid.
523
Cited in Cengiz Çandar, Mezopotamya Ekspresi: Bir Tarih Yolculuğu (Istanbul:
Iletisim, 2012), 38.
235
strengthened the grip of the traditional landowning elite over their peasants.
The Turkish military and government showed that maintaining the support
of the Kurdish landowners remained a priority in the fight against the PKK,
in particular given their capacity to contribute to the military effort by setting
up their own semi-private Village Guards battalion.524 With the complicity of
the military authority that ensured their impunity and generously rewarded
for the contribution to the anti-PKK operations, these tribal landowners
turned warlords thrived.

The Destruction of Kurdish Rural Society

The neoliberal reforms described in the previous section were not the only
reason for the structural transformation of the Kurdish region of Turkey.
From 1984 onwards, Turkey’s south-east was the theatre of the war
between the Turkish army and the PKK that, by the end of the 1980s, had
assumed a mass character in the rural areas. Military rule was re-imposed
in 1987 over nine Kurdish provinces in the form of the OHAL regime, that
gave exceptional powers to a military super-governor based in Diyarabakır.

The further empowerment of the tribal and landowning elite was by no

means the only repercussions of the war on the social structure of the
region. As discussed in the previous chapter, the intensification of the PKK
insurgency in the late 1980s was due to the growing popularity enjoyed by
the organisation among the Kurdish peasantry that provided manpower and
logistical support. It was becoming impossible for the Turkish army to
maintain a permanent presence in the mountain areas, especially at night.
In response to this challenge, the military imposed curfews and food
embargos over communities suspected of aiding the PKK but also started
to evacuate villages by physically removing their inhabitants and preventing
their return.525 This practice, made possible by the OHAL legal framework

524
See Chapter 6.
525
Jongerden, The Settlement Issue, 78.
236
was initially used as a form of collective punishment against communities
seen as siding with the guerrilla.

By 1990, the evacuation of villages increased and assumed the character of

a counterinsurgency strategy aimed at depopulating the countryside. In his
comprehensive study of the spatial dimension of Turkey’s anti-PKK
strategy, Joost Jongerden explains that the evacuation of villages
“constituted a concerted attempt by the Turkish military to bring about a
conclusive transformation of the regional settlement structure” and the
“destruction of rural society was simply considered part of the solution.” 526
The figures of evacuated villages and villagers remain highly contested as
the military rulers of Turkey’s south-east were well aware to be acting in
violation of the Geneva Convention. Not only did journalists have little
access to the region, but the military also prevented Prime Minister Tansu
Çiller and Deputy Prime Minister Murat Karayalçin from visiting the region in
1993.527 Jongerden compared the contradictory figures coming from Turkish
official sources with those provided by non-governmental organisations.
While there is general agreement that the settlements evacuated in the
course of the 1990s were around 3,000, the estimates on the number of
displaced people vary: 384,000 according to official sources while
independent sources claim 1-1.5 million and some even 3-4 million.528 The
available data shows that the evacuation and destruction of Kurdish villages
became systematic only from 1991 onwards when this strategy was
implemented in 109 villages. The number rose in 1992 (295) and 1993 (874)
to culminate in 1994 when 1.531 rural communities were destroyed, and
their inhabitants displaced.529

526
Ibid., 43-44.
527
Ibid., 93.
528
Ibid. Veli Yadirgi compares the same sources arguing that “because of such
discrepancies, it is reasonable to assume that at least 1 million people were
removed from their homelands.” Yadirgi, The Political Economy, 224-225.
529
Cited in Jongerden, The Settlement Issue, 82.
237
The military’s objective of permanently depopulating the Kurdish
countryside was implemented through the material devastation that
accompanied the evacuation of villages and that prevented a future return
of the villagers. Houses were destroyed and livestock was slaughtered up
to an estimated loss of $2.3 billion in the 1990s. 530 A large portion of the
region’s forest was burned. A 2007 study based on satellite images shows
that 7.5 per cent of all the forested areas in the province of Tunceli – the
Kurdish Dersim – were destroyed.531 Forests surrounding villages were
heavily targeted to affect more directly the livelihood of the local population
and of destroying orchards and crops.532 The construction of dams also
contributed to mass displacement forcing 200-350,000 people to leave
their homes between the 1980s and the early 2000s.533 Presented as a
developmental solution to the Kurdish issue, hydroelectric plants also
served the same purpose as village evacuations and forest burning:
depopulate the countryside and resettle allegedly (actual or potential) pro-
PKK peasants to a more easily controllable urban environment.

The result of Turkey’s counter-insurgency strategy was a rapid and

disorderly process of urbanisation. Between 1990 and 1997, the rural
population in the region under the OHAL regime decreased from 54.2 per
cent to 41.9.534 This 12 per cent drop is particularly striking when compared
to the data from the rest of Turkey (non-OHAL provinces) where the rural
population decreased only by 5.4 per cent. Moreover, the latter figure does
not only show Turkey’s structural urbanisation process but also includes the
many displaced Kurds who moved to cities outside the OHAL region, in
central and western Anatolia and thus increased the urban population of

530
This massive loss contributed to Turkey’s shift from a meat-exporting country
to a net importer in the 1990s. Ibid.
531
Joost Jongerden, Hugo de Vos, and Jacob van Etten, ‘Forest Burning as
Counterinsurgency in Turkish-Kurdistan: An Analysis from Space’, The International
Journal of Kurdish Studies, 21.1&2 (2007).
532
Ibid.11-12.
533
Yadirgi, The Political Economy, 256.
Matthew A. Kocher, ‘Human Ecology and Civil War’ (unpublished doctoral thesis,
534

University of Chicago, 2004), cited in Jongerden, The Settlement Issue, 86.

238
these provinces. The population of the major towns and cities of the Kurdish
region increased exponentially in the 1990s. According to independent
estimates, between 1991 and 1996, Diyarbakır’s population increased
almost four times: from 380,000 to 1.3 million people.535 Displaced Kurds
swelled Turkey’s shantytowns joining the already existing underclass of
urban poor. They also provided cheap labour to export-oriented and hyper-
exploitative industries in central Turkish cities – the so-called Anatolian
Tigers (Anadolu Kaplanları) – that had been booming since the 1980s. This
process of forced urbanisation had dramatic political consequences that
were contrary to the intention of the Turkish government and military. As
observed by Martin van Bruinessen, these “large urban Kurdish population
concentrations with a strengthened sense of ethnic identity […] proved
more easy to politically mobilize […] than villagers”536 and the PKK took
advantage of the new situation to expand its geographical and social reach.

The Kurdish Insurgency in the 1990s

The shift in Turkey’s counterinsurgency strategy was the result of the PKK’s
increasing ability to move freely across vast areas of the Kurdish region.
Even if the guerrilla presence was still largely limited to the rural and
mountainous areas, the PKK started to make inroads into the Kurdish towns
and cities and the Kurdish communities in non-Kurdish areas. Information
about the insurgency would easily spread from the villages where the PKK
operated to the towns. Most small Kurdish towns existed in symbiosis with
the surrounding rural environment as markets for agricultural products that
villagers would visit frequently and in which they often had family ties. In the
larger urban centres of the south-east and among the Kurdish diaspora in
western Turkey, Kurdish leftist nationalism had been a powerful force since

535
Hakkari from 35,000 to 80,000, Batman from 150,000 to 250,000, Van from
153,000 to 500,000. David McDowall, The Destruction of Villages in South-East
Turkey: A Report by Medico International and the Kurdish Human Rights Project
(KHRP) (London: Medico International and KHRP, 1996), 19.
536
Van Bruinessen, Kurds and the City.
239
the 1970s and especially among the working and lower-middle classes.537
While underdevelopment, cultural oppression, police brutality – the
grievances that motivated the radical Kurdish politics of the 1970s – were
even more accentuated after the 1980 coup, the military had eliminated all
the alternatives to the PKK making it the only possible outlet for Kurdish
discontent. Moreover, the brutal post-coup repression of the movement,
that evenly hit all Kurdish groups regardless of their militancy, validated
Öcalan’s especially uncompromising stance in the eyes of many Kurds. From
1989 onwards, the PKK received a wave of recruits from urban areas,
including many university students.538

Due to the limited information that filtered through the region, the Turkish
public became aware of the extent of the PKK’s popularity only in 1990. In
March, in the small town of Nusaybin, the funeral procession for a PKK
activist turned into a mass protests where thousands of locals opposed the
security forces and one protestor was killed. As the news of the revolt
spread, protests broke out in Cizre, another small Kurdish town, leaving five
dead. Signs of solidarity with the protests were visible throughout the
south-east where shops were shut down for days. These mass
demonstrations that turned into riots took the name of serhildan – literally
‘raising one’s head’ – and made explicit references to the First Palestinian
Intifada (1987-1993). The serhildans became more frequent in 1991 and
1992 and tended to coincide with Newroz, the Kurdish new year on March
21st, or the anniversary of the establishment of the PKK on November 10 th.539

These often spontaneous revolts convinced the PKK leadership that the
time was ripe to move beyond guerrilla tactics and towards forms of
conventional warfare that would show the PKK’s capability to hold

537
See Chapter 7.
538
Alize Marcus goes a great length to describe the efforts of these urban recruits
to adapt to a guerrilla’s lifestyle. Aliza Marcus, Blood and Belief: The PKK and the
Kurdish Fight for Independence (New York and London: New York University Press,
2007), 134-139.
For more on the serhildans, see Paul White, ‘The March 1990 Uprising in Turkish
539

Kurdistan & Its Effects on Turkish Politics’, Kurdish Times, 4.1/2 (1991), 97–106;
Marcus, Blood and Belief, 134-140;
240
conquered ground and encourage a widespread popular uprising. In August
1991, when Turkey attacked the guerrilla camps in northern Iraq, the PKK
fighters maintained their positions and – despite heavy losses – repelled the
Turkish army.540 In August 1992, the town of Şirnak became the theatre of
an urban battle that displaced virtually its entire population – 20,000 out of
25,000 – and left 40 gendarmes, 85 PKK fighters, and 22 civilians dead. 541
These displays of strength by the PKK were meant to signal the shift to a
phase of mass popular uprising allowing the PKK to deploy its forces into
the open and to challenge a demoralised Turkish army. With the liberation
wars of Algeria and Vietnam in mind, Öcalan hoped that the Turkish military
would eventually lose public support and be forced to withdraw from the
region.542 However, this shift never materialised. Rather than withdrawing,
by 1993, the Turkish army had deployed roughly half of its land forces to
the region – 185,000 soldiers – that combined with the gendarmerie and the
village guards amounted to more than 300,000 men. This was a force that
could not be challenged in terms of conventional warfare.543

If Öcalan had underestimated the determination of the Turkish military, his

intuition about the growing gap between the military and the civilian
authorities proved to be partially correct. As mentioned before, the military
allowed little civilian supervision over the Kurdish war and maintained firm
control over the public narrative. Yet, some Turkish politicians made timid
attempts to challenge the military’s monopoly over the Kurdish issue and
signs of this new attitude became more frequent in the early 1990s. In
February 1991, the government lifted the ban on using the Kurdish language,
and in December, Prime Minister Süleyman Demirel announced that Turkey
“acknowledged the Kurdish reality.” Turgut Özal – who had become
President in 1989 – spoke favourably about the need to find a political

540
Jongerden, The Settlement Issue, 61-63.
541
The Kurds of Turkey: Killings, Disappearances and Torture (New York: Human
Rights Watch, March 1993)
<https://www.hrw.org/sites/default/files/reports/TURKEY933.PDF>
542
Jongerden, The Settlement Issue, 61-63.
543
Cited in Jongerden, The Settlement Issue, 64.
241
solution.544 In the meantime, Öcalan appeared on a series of public
interviews to improve the PKK’s image among the Turkish public. 545
Interviewed by Turkish journalist Doğu Perinçek, in 1990, Öcalan announced
that the PKK was no longer demanding independence and was willing to
negotiate on a platform of regional autonomy.546

The conditions for negotiations briefly materialised in spring 1993. Between

February and early March, the Iraqi-Kurdish leader Jalal Talabani acted as a
mediator meeting both Öcalan in Damascus and Turgut Özal in Ankara.
Encouraged by this climate, on March 17th, Öcalan proclaimed a unilateral
ceasefire ordering his followers to keep Newroz quiet and his commanders
to suspend their military operations. However, no official response arrived
from the Turkish government. While Özal was lobbying the state apparatus
to create the conditions for a political response, the military took Öcalan’s
move as a sign of weakness. In early April 1993, the Army launched a
massive operation in the countryside of Diyarbakır with indiscriminate
bombings. While many PKK cadres were increasingly critical, the remaining
hope for a political solution was shattered by Özal’s sudden death on April
17th. According to Cengiz Çandar, at the time special advisor to Özal, the
conditions for peace were simply not there, regardless of the president’s
“goodwill” or his sudden death:

He was a lonely President encircled by many opponents, including

the Prime Minister. […] Secondly, the military […] was totally against
a political and peaceful resolution to the Kurdish issue. They wanted
to go on in the conventional way and suppress them as bandits and
terrorists. […] Ideologically, the PKK was also not enough mature to
reach a settlement. It was Öcalan’s fantasy. We are talking about the
year 1993, they were still more or less a pro-independence Marxist
movement. When we look back, conditions were not mature to bring

544
Michael Gunter, ‘Turgut Özal and the Kurdish question’, in Nationalisms and
Politics, ed. by Casier and Jongerden, 85-100.
545
Popular journalist Mehmet Ali Birand’s 1988 interview to Öcalan – later published
as a book – contributed significantly to ‘humanise’ the PKK leader. See Birand, APO
ve PKK.
546
Doğu Perinçek, Abdullah Öcalan Ile Görüşmeler (Istanbul: Kaynak Yayınları,
1990).
242
 about a resolution. I can attest how much Özal was committed […].
But in politics, that's not sufficient.547
Turkey’s official response to the PKK ceasefire arrived on May 25 th, when
the National Security Council refused negotiations and offered a partial
amnesty in exchange for unconditional surrender. On the same day, a PKK
unit captured a military bus and executed 33 unarmed recruits, an action
that horrified and angered the Turkish public.548 Two weeks later Öcalan
abandoned the ceasefire.

The ‘Urbanisation’ and Ideological Transformation of the PKK

Even if the intensity of the war continued to grow in the mid-1990s, the PKK
remained committed to a political solution and, both in 1995 and 1998,
Öcalan proclaimed new unilateral ceasefires that received no answer from
the Turkish side. On the part of the PKK, there was a degree of war fatigue
and a growing awareness that the disproportion of means made a
continuous escalation of the conflict unviable. However, the ceasefires were
also the result of a gradual ideological transformation of the organisation.
The crisis and collapse of Soviet communism made the PKK’s project of
establishing a socialist Kurdistan look anachronistic and less appealing while
liberal democracy was increasingly perceived as the only legitimate form of
politics. Moreover, the fall of the South African Apartheid regime (1991), the
Oslo Accords between Israelis and Palestinians (1993 and 1995) and the
peace process in Northern Ireland (1994-1998) created a widespread
optimism that the end of the Cold War provided the conditions for the
resolution of decades-old conflicts. In the mid-1990s, Öcalan was explicitly
taking distance from communism and re-defined the PKK’s programme as
“humanistic in essence”:

547
Interview with Cengiz Çandar (personal communication, 2020). For a more
detailed account of these events see Cengiz Çandar, Turkey’s Mission Impossible:
War and Peace with the Kurds (Lanham: Lexington Books).
548
Martin van Bruinessen, Kurdish Ethno-Nationalism versus Nation-Building
States: Collected Articles (Istanbul: ISIS, 2000), 33-37.
243
 We are dedicated to a philosophy that is based on democracy and
pluralism, not on the power of the state. We favour the synthesis of
capitalism and socialism, an economic structure in which individuals
will freely develop to their fullest potential. We are against all
ideologies that defend absolute authority for the state at the
expense of individual freedom.549
These positions were officially adopted by the PKK in its Fifth Congress in
1995, where the hammer and sickle were removed from the party flag. This
transition was not particularly dramatic. Despite its allegiance to Marxism-
Leninism, class politics was never a particularly central element of the PKK’s
discourse that was instead largely based on the issue of ‘Turkish
colonialism’. In these terms, the PKK was really an heir of the Turkish New
Left and the student movement of the 1970s. In the cultural context where
the PKK was originally established, the working class – partly incorporated
by the state within the import-substitution industrialisation compromise –
was absent and mostly evoked out of theoretical necessity. 550 Even in its
most ‘rural’ phase, in the mid-1980s, when the main constituency of the PKK
was – by far – the Kurdish peasantry, the socialisation of land was not a
central aspect of its programme.551 More than promising land, the PKK won
a large part of the peasantry by targeting oppressive landowners and tribal
chiefs, ‘collaborationist aghas’, that the PKK considered agents of Turkish
colonialism. As Remzi Kartal explains

the PKK did not carry out actions against those aghas who were
national aghas, who were patriotic, who stood by the cause of the
Kurdish people. They acted against those aghas who were regime
collaborators.552
This distinction between patriotic and collaborationist aghas reflects the
classical distinction of Marxist anti-colonial thought between the ‘national

549
PKK: Program ve Tüzüğü (Cologne: Weşanên Serxwebûn, 1995).
550
Keyder, State and Class, 209.
551
Interview with Hamit Bozarslan (Paris, 2019).
552
Interview with Remzi Kartal (Brussels, 2019).
244
bourgeoisie’ – with which the working class and peasantry should be allied
– and the native ‘comprador bourgeoisie’ working for the colonisers.553

Moreover, in the 1990s the PKK expanded its presence in the urban
contexts, thanks to its military successes and its effective propaganda that
exponentially increased its appeal on urbanised Kurds of diverse class
background. The ideological transformation of the PKK was parallel to and
mutually reinforced by its growing strength among social groups other than
the Kurdish peasantry. Precisely when the PKK was winning sections of the
Kurdish working and middle classes, millions of Kurdish peasants – many of
whom from PKK-supporting communities – were moving into the cities as a
result of the evacuation of villages and other war-related reasons but also
the damming of valleys and the country’s structural urbanisation. These
parallel processes gave the PKK unprecedented presence and popularity in
both the urban centres of the Kurdish region and among the Kurdish
communities in central and western Turkey and the European diaspora. An
indication of the expansion of the PKK’s class base can be found in the rapid
growth of pro-Kurdish legal parties and their links to the Kurdish insurgency.

The Kurdish pro-Democracy Movement and its Social Basis

The 1980 military coup was particularly ruthless in the repression of Kurdish
leftist forces including the ones that, in the 1970s, had tried the path of
electoral politics rather than armed struggle. The socialist and Kurdish
nationalist mayor of Diyarbakır Mehdi Zana, elected in 1977, was arrested
and tortured by the putschist generals and jailed until 1991.554 Turkey’s
legislation forbade the establishment of parties with an explicit ethnic or
regional identity, and, during the 1980s, the south-east again became a

553
For a classical example, possibly known to Öcalan himself, see Mao Zedong, On
the Question of the National Bourgeoisie and the Enlightened Gentry (March 1,
1948), (Bejing: Foreign Language Press, 1969).
554
His memoirs, translated into English and French, significantly contributed to the
increasing awareness in Europe of the Turkish Kurds’ plight. Mehdi Zana, Prison
No.5: Eleven Years in Turkish Jails (Watertown: Blue Crane Books, 1997).
245
terrain of competition for nation-wide mainstream parties. In the 1987
elections, the Social Democratic Populist Party (SHP) became the main
opposition party with 24 per cent and performed particularly well in the
Kurdish areas thanks to its commitment to undoing the political legacy of
the 1980 coup. In 1989, however, the party leadership came under
increasing pressure and eventually expelled several Kurdish MPs that had
publicly spoken about the oppression of the Kurds in international venues.555
The expulsion pushed six more Kurdish MPs and the chairs of twelve south-
eastern SHP branches to resign.

The Kurdish MPs were joined by a number of non-Kurdish socialists to

establish the People’s Labour Party (HEP) in June 1990. This new party
proposed a civilian constitution for Turkey as well as a resolution of the
Kurdish issue through the democratisation of the country. The HEP located
itself on the left aspiring to be the party of

the workers, the unemployed, the rural people, the civil servants, the
teachers, democrats, the intellectuals of social democratic and
socialist persuasion, the small businesses and artisans, the masses
who have been subjected to oppression and exploitation and above
all everyone who supports democracy.556
The HEP was unable to run for the snap elections of October 1991 due to
technicalities but managed to strike a deal with SHP that was fearing an
electoral collapse in the south-east. Thanks to this arrangement, 22 HEP
members were elected in the SHP list. The HEP was becoming a political
space in which Kurdish activists felt free to express their identity
unapologetically. At the same time, the Turkish centrist and right-wing
forces, as well as the media, led an incessant campaign to depict the HEP
as the political wing of the PKK. Under these conditions, the non-Kurdish

555
In early 1989, Ibrahim Aksoy, parliamentarian for Malatya, was expelled by the
SHP for a speech he delivered at the European Parliament. In October, seven SHP
parliamentarians were expelled by the party for attending an international
conference organized by the Kurdish Institute in Paris. At the conference, Aksoy
harshly criticised the SHP calling it “racist and chauvinistic”. For Aksoy’s full
speech, see International Paris Conference 14-15 October 1989. The Kurds: Human
Rights and Cultural Identity (Paris: Institut Kurde de Paris, 1992), 53-60.
556
Cited in Güneş, The Kurdish National Movement, 157.
246
HEP members were increasingly uneasy with the composition of the party
and many of them resigned, in turn accelerating the ‘Kurdification’ of the
party.557

The alleged links to the PKK and the accusation of separatism haunted the
HEP until the Constitutional Court shut down the party in 1993. This act
inaugurated a pattern whereby every time a pro-Kurdish party was banned,
it was immediately re-opened under a different name. The HEP was
replaced by the Democracy Party (DEP) (1993-1994), in turn, closed down
and replaced by the People’s Democracy Party (HADEP) (1994-2003).
These parties became increasingly ‘more Kurdish’, as Nicole Watts observes
in her book on pro-Kurdish legal activism, “police harassment and state
coercion radicalised the party and its leadership, which isolated it from
mainstream parties and promoted the politics of polarization and
difference.”558 On its part, the PKK was certainly interested in influencing
electoral politics especially since the HEP was established roughly at the
time in which Abdullah Öcalan started calling for negotiations with the state.
This can be noticed in the shift from the 1987 to the 1991 elections: In 1987,
the SHP performed well in the south-east but was nowhere close to being
the strongest party. In the 1991 elections, when the Kurdish candidates on
the SHP lists were largely HEP members and the PKK was openly supportive,
the SHP boomed in the Kurdish region. Despite its nation-wide 6 per cent
decline – a loss of almost one million voters – the SHP/HEP experienced a
spectacular increase in areas of strong PKK presence: From 16 to 71.7 per
cent in the Diyarbakır 2 constituency, from 27.8 to 54 per cent in Mardin,
from 17.4 to 39.7 in Siirt.559 These results show the overwhelming popularity
reached by the PKK in significant parts of the Kurdish region but also

557
Ibid., 162-163.
Nicole F. Watts, Activists in Office: Kurdish Politics and Protest in Turkey (Seattle
558

and London: University of Washington Press, 2010), 65.

559
‘Seçim Arşivi’, T.C. Yüksek Seçim Kurulu <http://www.ysk.gov.tr/tr/secim-
arsivi/2612>.
247
contributed to the Turkish public’s perception of the HEP as a mere political
extension of the PKK.

Watts explains that, despite not being a creation of the PKK, “the parties’
most active rank-and-file membership and much of its voter base supported
pro-Kurdish parties because they viewed them as sympathetic to or as a
surrogate of the PKK.” Within the HEP, DEP, and HADEP, there was always
a degree of tension between “those who maintained close communication
with the PKK leadership and indeed viewed the party as an unofficial front
for the PKK” and “those who sought to maintain some distance and
autonomy from the guerrilla organization.”560 The creation of these parties
at a time in which the PKK had reached a level of popularity and strength
that was unprecedented in Turkey’s Kurdish movement meant that the room
for an alternative Kurdish politics was very limited. The gradual breakaway
of Turkish leftists was, according to Hamit Bozarslan, virtually inevitable, as
the success of the PKK insurgency triggered “a total reconfiguration of the
Kurdish political space, as a distinct space from the Turkish political
space.”561 In earlier stages of the insurgency, Kurdish nationalists and leftists
could still maintain critical positions on the PKK’s ideology and strategy. By
the early 1990s, the PKK had become so powerful that it was increasingly
able to determine the boundaries of ‘legitimate’ Kurdish politics. Within this
context, the HEP and its successors had little room to escape the PKK’s
narrative that each Kurd could only be either with the PKK or with the
Turkish state.

560
Watts, Activists in Office, 14.
561
Cited in Casier and Grojean, Between Integration, Autonomization and
Radicalization, 11.
248
The PKK’s involvement provided these parties with wide support among the
landless peasantry and the urban poor, especially in the mid-1990s, when a
large part of the former was becoming the latter. Since this was the social
base that allowed the pro-Kurdish parties to obtain such spectacular results
in the Kurdish south-east, the party programme remained quite radical:
“land reform and redistribution, programs to eradicate regional economic
disparities, continued public ownership of state-owned economic
enterprises, and greatly expanded services to aid the poor and
unemployed.”562 The party leadership, however, was largely filled by the
urban middle classes and in particular professionals – lawyers, teachers,
engineers – and people who had gained prominence through civil society
activism.563 Relying on this wide class base, by the mid-1990s, the pro-
Kurdish parties had largely displaced the Turkish mainstream parties and
were the dominant electoral force in the south-east. The main challenge to
their electoral hegemony came from the rising Turkish Islamist movement.
The Islamist Welfare Party (RP) provided conservative Kurds with an
alternative to the ‘opposed nationalisms’ and emphasised Turkish-Kurdish
brotherhood under the banners of (Sunni) Islam. During the 1990s, the
Kurdish south-east became the arena for the competition between the
Islamists and the pro-Kurdish forces while the other Turkish parties became
marginal in the region. The RP, however, shadowed its rivals among both

562
Watts Activists in Office, 70.
563
Ibid., 71.
249
Turkish and Kurdish slum dwellers in cities such as Istanbul, where the
Islamists’ network of charity organisations won them vast support.564

The pro-Kurdish parties faced a hostile judicial and political system. The
1982 military constitution imposed a nation-wide 10 per cent – the world’s
highest – electoral threshold, designed to ensure strong parliamentary
majorities but also to make it difficult for ‘regional’ – read ‘Kurdish’ – parties
to enter parliament. For these reasons, the pro-Kurdish parties had to opt
for an electoral alliance – like in 1991 – or to run their candidates as
independents. The success of the pro-Kurdish parties made them a target
of the wave of extra-judicial violence that, in the 1990s, hit Kurds as well as
Alevi, leftists, liberals, and secular intellectuals. The hundreds of murders
that took place in this decade revealed the links between far-right
organisations – like the Grey Wolves565 and Turkish Hizbullah566 – the Turkish
mafia, and the Turkish ‘deep state’.567 Between 1991 and 1994, more than
fifty DEP/HEP activists were murdered including two parliamentarians. 568
These crimes largely remained unpunished and took place in a context in
which the political mainstream publicly painted elected Kurdish officials as
terrorists and traitors. After the DEP was banned in 1994, five MPs were
given sentences up to 15 years for their alleged ties with the PKK while six
more went into exile before their trials began.

Kurdish Diaspora and Exile in Europe

One of the reasons behind the PKK’s growing strength between the 1980s
and the 1990s was its ability to mobilise the Kurdish diaspora in Europe.

564
Henri Barkey, ‘The People’s Democracy Party (HADEP): The Travails of a Legal
Kurdish Party in Turkey’, Journal of Muslim Minority Affairs, 18.1 (1998), 129–38.
565
The Gray Wolves (Bozkurtlar) is a violent neofascist group affiliated to the MHP.
566
Known also as ‘Kurdish’ Hizbullah, due to its strength in the south-east, this
Sunni organisation is unrelated to the more famous Lebanese Hizbollah.
567
Zürcher, Turkey, 321-323.
Güneş, The Kurdish National Movement, 163-164; Marcus, Blood and Belief,
568

208.
250
Germany hosted the biggest Kurdish community in Europe, roughly 20 per
cent of the 650.000 Turkish nationals who emigrated there as guest workers
since the 1960s.569 With the waves of Kurdish refugees escaping Turkey,
Iran, and Iraq in the 1980s, by the mid-1990s, the Kurdish population in
Europe amounted, to approximately 850.000 individuals, half a million of
whom in Germany.570 Up to the 1970s, the Kurdish population in Germany
was overwhelmingly made up of migrant workers with a low degree of
politicisation.571 It was the great influx of Kurdish political refugees after the
1980 coup – roughly 30.000 – that helped rapidly politicise the Kurdish
community.572 Activist Devris Çimen left Turkey with his family in 1978,
escaping anti-Alevi persecution in southern Turkey and developed his
Kurdish identity in Europe:

I was a child basically when they sent us to Switzerland… I was

asking 'why am I here?'. I didn't like it, my roots were in Maraş. […]
The Kurdish movement was active in Switzerland. As a child, I
thought 'why would I fight for Kurdistan? I'm from Elbistan 573!'. We
were suppressed to such a degree that we didn't even know that we
were Kurds. And when we arrived in Switzerland we realised that the
issue was much broader than we ever thought.574
In the early 1980s, the PKK sent cadres to Europe with the specific purpose
of organising the Kurdish communities and, by the end of the decade, the
diaspora had become a major source of funding and recruitment. The PKK
set up the Weşanên Serxwebûn publishing house in Cologne and organised
community life through sectoral organisations – for women, workers, youth,
but also religious groups (Sunnis, Alevis, Yazidis).575 These groups organised

Birgit Ammann, ‘Kurds in Germany’, in Encyclopedia of Diasporas, ed. by Melvin

569

Ember, Ian A. Skoggard (Boston, MA: Springer, 2005), 1011.

Osten Wahlbeck, Kurdish Diasporas: A Comparative Study of Kurdish Refugee
570

Communities (New York: St. Martin’s Press, 1999).

571
Ammann, Kurds in Germany, 1011 .
572
Martin van Bruinessen, ‘Shifting National and Ethnic Identities: The Kurds in
Turkey and the European Diaspora’, Journal of Muslim Minority Affairs, 18.1 (1998),
45.
573
A district in Kahramanmaraş Province.
574
Interview with Devris Çimen (Brussels, 2019).
575
Güneş, The Kurdish National Movement, 109-111.
251
rallies and protests in solidarity with the insurgency in Turkey but also music
festivals, dance contests, the Newroz celebrations with the aim of
politicising Kurdish culture. According to police reports, in the mid-1990s,
the PKK-affiliated organisations in Germany could count on approximately
7,500 activists and were able to mobilise up to 50,000 German Kurds for
protests and cultural events.576 In 1995, The pro-PKK MED-TV started
broadcasting from Europe and, in just a few months, was being viewed by
millions of people around the world.577 The successful organisation and
mobilisation of Europe’s Kurds were aimed at creating a new pool of recruits
for the insurgency but, even more importantly, a reliable flow of revenues.
This topic is highly contested because of Turkey’s incessant campaign to
present the PKK’s financing as based on extortion and the control of major
drug trafficking routes.578 Even if the idea that the PKK exercises direct
control on the flow of drugs to Europe has no actual foundation, the PKK
certainly ‘taxes’579 legal and illegal commercial activities in areas under its
influence.580 The PKK’s social control over Kurdish diaspora communities is
manifested in both the large extent of voluntary donations but also in the

576
Van Bruinessen, Shifting National and Ethnic Identities, 51.
577
Amir Hassanpour, ‘Satellite Footprints as National Borders: Med‐tv and the
Extraterritoriality of State Sovereignty’, Journal of Muslim Minority Affairs, 18.1
(1998), 53–72; Bilgin Ayata, ‘Kurdish Transnational Politics and Turkey’s Changing
Kurdish Policy: The Journey of Kurdish Broadcasting from Europe to Turkey’,
Journal of Contemporary European Studies, 19.4 (2011), 523–33.
578
Despite the poverty and dubious origin of the available sources, this narrative
often makes it into academic and pseudo-academic literature, especially in the
field of Terrorism Studies. See for example Roth and Sever, The Kurdish Workers
Party (PKK); Ahmet Pek and Behsat Ekici, ‘Narcoterrorism in Turkey: The Financing
of PKK-KONGRA GEL from Illicit Drug Business’, in Understanding and Responding
to the Terrorism Phenomenon: A Multi-Dimensional Perspective, ed. by Ozgur
Nikbay and Suleyman Hancerli (Washington: IOS Press, 2007), 140–52.
579
Voluntary donations resembling taxes – regular and somehow proportionate to
the donor’s income – are an almost universal characteristic of diaspora support to
liberation struggles in the home country: from the Irish to the Tamils or the
Basques.
For a critical discussion on the alleged criminal activities of the PKK, see Paul
580

White, Primitive Rebels or Revolutionary Modernizers?: The Kurdish National

Movement in Turkey (London and New York: Zed Books, 2000), 191-200.
252
use coercion to persuade more reluctant members of the Kurdish
communities to contribute to the cause.

The PKK’s presence in Europe also served the purpose of internationalising

Turkey’s Kurdish question. Since the failure of the 1993 ceasefire attempt,
Öcalan started planning to establish a wider Kurdish national assembly in
exile to gain a degree of international legitimacy that the PKK alone was
unable to achieve. The Kurdistan Parliament in Exile held its first session in
The Hague in April 1995, soon after the DEP was banned in Turkey. Öcalan
was unable to convince any major party or organisation from the other
Kurdish regions to join and the Kurdish parliament was largely composed by
current and former DEP deputies, civil society activists, and some
independent personalities. As Remzi Kartal, one of the exiled DEP
parliamentarian and founder of the Parliament in Exile, explains

we came to Europe in 1994 because we had escaped [Turkey]. Then,

as a delegation of the [DEP] MPs in exile, we went to Damascus and
met Öcalan. Öcalan said that the most important thing for us now
was to establish a national congress to bring all these political
parties together. And then we worked for that, we spoke with
everybody, but the [Iraqi-Kurdish] KDP and PUK said that they had
interests with Turkey and, for that reason, they did not come. So we
established the Kurdish parliament in exile that was limited to
northern [Turkish] Kurdistan. It wasn't a national parliament. But we
worked in that direction.581
The inability to convince other major groups weakened the parliament and
in 1999 it transformed into the slightly more representative Kurdistan
National Congress.582

However, that severe limits to Kurdish political activism existed also in

Europe became clear already in the early 1990s. In June 1993, after Kurdish
activists had attacked the Turkish consulate in Munich, German authorities
complied with Turkey’s request to outlaw the PKK, and France followed suit

581
Interview with Remzi Kartal (Brussels, 2019).
582
Tahiri, The Structure of Kurdish Society, 219-221.
253
by the end of the same year.583 Even if these bans have not been
consistently enforced, they create a sense of precariousness as pro-Kurdish
organisations and events are occasionally shut down and convey the
message that the US and European government are ultimately always going
to side with Turkey.

The Capture of Abdullah Öcalan

The Kurdish insurgency in Turkey was suddenly interrupted by the capture

of the PKK’s leader Abdullah Öcalan in February 1999. Since 1979, Öcalan
had been based in Damascus as a guest of the Ba’athist regime of Hafez al-
Asad. In the 1970s and 1980s, Asad offered safe heaven to leftist
revolutionary and separatist groups active in neighbouring and hostile
countries such as leftist Palestinians, the insurgency-aspiring Turkish New
Left, Kurdish nationalists from Iraq, or Armenian militants. This policy was
aimed at destabilising neighbouring Turkey, Israel, and Iraq – each of them
stronger than and hostile to Syria.584 The reason why Damascus was able to
provoke these countries without facing retaliation was that, from the 1970s,
Syria enjoyed the protection of the Soviet Union as its most stable Middle
Eastern ally. Syria’s support to the PKK was also specifically aimed at
destabilising Turkey’s south-east and delaying the construction of large
infrastructural projects in the upper Euphrates valley that could seriously
affect the volume of water reaching eastern Syria.585 Rather than direct
military and logistical assistance, Syria mostly offered the PKK the freedom
to move and organise inside its territory.

583
Alynna J. Lyon and Emek M. Uçarer, ‘Mobilizing Ethnic Conflict: Kurdish
Separatism in Germany and the PKK’, Ethnic and Racial Studies, 24.6 (2001), 938-
939.
584
This policy certainly had also an ideological dimension and the organizations
supported by Syria were largely socialist in orientation. However, in the 1980s, with
the decline of leftist forces throughout the Middle East, Asad pragmatically started
to support rising Islamist groups such as Hamas and Islamic Jihad in Palestine or
Hezbollah in Lebanon.
585
See Mark Dohrmann and Robert Hatem, ‘The Impact of Hydro-Politics on the
Relations of Turkey, Iraq and Syria’, The Middle East Journal, 68.4 (2014), 567–83.
254
With the end of the Cold War, the balance of power shifted in favour of
Turkey. In just a few years, the end of Soviet military aid to Syria as well as
the increasing military cooperation between Turkey and Israel made
Ankara’s threat of military invasion much more credible.586 In October 1998,
Syria expelled Öcalan and shut down the PKK’s offices and training facilities.
Öcalan fled first to Athens and then Moscow but both countries refused him
asylum. In November, Öcalan made it to Italy. Under Turkish and American
pressure, the Italian government did not grant Öcalan refugee status, and
yet it refused extradition on the grounds that Turkey could apply the death
penalty.587 There would have probably been safer choices for Öcalan – such
as Libya or North Korea – than Italy, a NATO member with strong economic
ties to Turkey. The PKK leader explained the move as a political choice:

To explain the reality of [the Kurds], and if possible to find a political

solution, I found it inevitable that I should go to Rome […] to explain
the merciless truth to European public opinion and to seek the
chance to create a political opening.588
As observed by Paul White, Öcalan’s choice was aimed at breaking the
political isolation of “a guerrilla chief hiding away in a so-called ‘pariah state’”
and at bringing the Kurdish issue directly to the core of Europe.589 His hope
was not well placed. In February, Öcalan was pressured to leave Italy for
Greece where he was eventually flown to Nairobi in the hope that he could
apply for asylum in South Africa thanks to the long-standing ties between
the PKK and the ruling African National Congress. Once in Nairobi, Öcalan
was captured by Turkish special forces aided by American and Israeli
intelligence.

586
For an analysis of the 1998 Turkish-Syrian crisis in terms of geopolitical and
military development, see Meliha B. Altunışık and Özlem Tür, ‘From Distant
Neighbors to Partners? Changing Syrian-Turkish Relations’, Security Dialogue, 37.2
(2006), 229–48.
587
Alessandra Stanley, ‘Italy Rejects Turkey’s Bid For the Extradition of Kurd’, The
New York Times, 21 November 1998
<https://www.nytimes.com/1998/11/21/world/italy-rejects-turkey-s-bid-for-the-
extradition-of-kurd.html>.
588
Cited in Tahiri, The Structure of Kurdish Society, 235.
589
White, Primitive Rebels, 183.
255
Öcalan’s capture led to an explosion of public rage in Turkey’s south-east
but also in Iraqi Kurdistan and among the Kurdish diaspora. In Europe, PKK
sympathizers attacked Greek and Israeli embassies and, in Germany,
clashes between Turks and Kurds took place. Held two months after
Öcalan’s capture, the local and legislative elections of April 1999 turned out
to be a show of strength by the pro-Kurdish HADEP that won Diyarbakır and
six other provincial capitals. With nearly 5 per cent, the HADEP was far from
passing the 10 per cent threshold and yet it came first in eleven of Turkey’s
81 provinces. In June 1999, Öcalan was sentenced to death by a military
tribunal although the sentence was commuted to life imprisonment after
Turkey abolished the capital punishment in 2002.590

Conclusion

From the early 1990s, the PKK became the only significant Kurdish national
force in Turkey and the unavoidable point for reference of the Kurdish left,
including the legal pro-Kurdish parties. Chapter 7 showed the immediate
consequences of the 1980 coup and particularly how the elimination of the
PKK’s political rivals by the putschist military contributed to the
unforeseeable success of the early PKK insurgency (1984-1990). This
chapter described the long-term consequences of the 1980 military coup.
Through political repression and neoliberal reforms, the military regime
profoundly altered the class structure and political landscape of the country.
Banning the unions and forbidding strikes, the military laid the conditions for
Turkey’s transformation into an export-oriented industrial country whose
competitiveness was based on the contraction of salaries and workplace
discipline. By creating a docile and fragmented working class, this strategy
disarmed the political left eliminating the historical Turkish interlocutor for

590
In 2005, the European Court of Human Rights (ECHR) declared that Öcalan’s
trial had not been fair and that he had not been granted his right to appeal. ‘CASE
OF ÖCALAN v. TURKEY (Application No. 46221/99)’, ECHR, 12 May 2005
<https://hudoc.echr.coe.int/eng#%7B%22itemid%22:[%22001-69022%22]%7D>.
256
Kurdish activists and created an unbridgeable gap between Turkish and
Kurdish politics in Turkey.

In the Kurdish region, this transformation went parallel, in the mid-1990s, to

the coercive depopulation of the countryside through the forced
displacement of millions of Kurdish villagers – the social base of the PKK –
that were turned into urban poor. In this context of social fragmentation, the
PKK had to rearticulate its political project abandoning the idea of a socialist
and independent Kurdistan. As class mobilisation was made impossible by
the authoritarian and neoliberal reforms and socialist politics had been
discredited worldwide by the fall of the Soviet Union, the PKK and the pro-
Kurdish parties re-oriented their political project towards the resolution of
the Kurdish issue through the democratisation of Turkey. Without
abandoning social demands and a clear leftist stance, the re-framing of the
Kurdish issue in terms of democratic representation allowed for the building
of a wide social coalition in support of the PKK and the pro-Kurdish parties.
Chapter 9 will show how this re-orientation of the PKK continued after the
capture of Öcalan in 1999, developing into an alternative political project for
the Kurds.

257
Chapter 9
The Political Economy of the New
Kurdish Movement (2000-2019)

Introduction

This chapter discusses the expansion of the social basis of Turkey’s Kurdish
movement in relation to the process of ideological and organisational
transformation that followed the capture of PKK leader Abdullah Öcalan in
1999. The transition of the PKK from Leninism and nationalism to radical
democracy, the growth of the Kurdish women’s movement, and the
experience of Kurdish self-government in Rojava have attracted an
unprecedented degree of scholarly and media interest in the Kurds. Also
due to the nearly complete abandonment of Marxist language by the
movement itself, discussions of Kurdish politics are centred on issues of
democratic representation, minority rights, and women’s liberation. The
political-economic context in which the new Kurdish movement operates as
well as the conflictual class dynamics existing within the movement itself
are most often not part of the discussion.

After discussing the new ideological paradigm developed by Öcalan and

centred around the critique of nationalism, this chapter explores the three
most significant aspects of the new movement in the 2000s and 2010s: the
Kurdish women’s movement; the growth of legal pro-Kurdish parties in
Turkey; and the establishment of Kurdish-controlled political entity in Syria.
Each section touches upon relevant issues of political economy that
affected these developments as well as the class dynamics at play, while
the final section draws broader considerations on the political economy of
the movement. While the movement’s strategy seems to be oriented
towards building a wide social coalition with the inclusion of the middle and
even upper classes, its electoral and military strength remains rooted in the
mass support the movement enjoys among the subaltern classes.

From the PKK to the New Kurdish Movement

In August 1999, from the island prison of İmralı and with a pending death
sentence, Öcalan ordered his guerrillas to withdraw from Turkey and
proclaimed another unilateral ceasefire. The PKK fighters retreated under
Turkish fire to their bases in the northern Iraqi Qandil Mountains. The
decision to suspend the war and the highly conciliatory statements by
Öcalan to the Turkish government and public in these years drove many
militants to abandon the organization.591 However, the PKK largely complied
with the instructions, passed on by Öcalan through his lawyers, proving that
the leader was still in control.

The years following Öcalan’s capture were a period of transition for the PKK.
The organisation changed its name twice: into Kurdistan Freedom and
Democracy Congress (KADEK) in April 2002 and again into Kurdistan
People’s Congress (Kongra-Gel) in late 2003. These were attempts to
renovate the public image of the PKK and to avert the consequences of
being designated terrorist.592 However, the name changes were not purely
cosmetic operations. On the contrary, in the early 2000s, the PKK was going
through intense ideological debates that had been triggered by the defeat
of 1999. The congress documents of these years are increasingly focused
on the critique of nationalism as the main source of the Kurdish Question:

591
Around 1,500 militants according to White. Paul White, The PKK: Coming Down
from the Mountains (London: Zed Books, 2015), 31.
592
The PKK had been banned as a terrorist organisation by Germany and France in
1993 and by the US State Department in 1997. That adding the PKK to terrorist lists
was a political decision which had little to do with the PKK’s actions became clear
to all Kurds in the early 2000s, when both the United Kingdom (2000) and the EU
(2002) banned the PKK despite the latter had withdrawn from Turkey and was
calling for a political solution to the conflict. On the PKK’s response to the listing,
see Marlies Casier, ‘Designated Terrorists: The Kurdistan Workers’ Party and Its
Struggle to (Re)Gain Political Legitimacy’, Mediterranean Politics, 15.3 (2010), 393–
413.
260
“the system of the 20th Century which is based on nationalism, division,
denial and destruction is not the right solution but in fact the reason behind
the problem.”593 These discussions were triggered and encouraged by the
intellectual inputs coming from Öcalan’s prison cell. At his numerous trials, 594
the Kurdish leader opted for a political rather than a legalistic defence and
used his court appearances as a platform to propose a new political project.
His defence texts, presented before Turkish and international courts, were
later published and officially adopted by the PKK and constituted the basis
for a paradigm shift. At its 9th Congress, in spring 2005, the organisation
reformed its structure and officially adopted Öcalan’s new paradigm.

Öcalan turned his own imprisonment into the opportunity to accelerate the
process of transformation that the PKK had started in the 1990s. 595 By the
late 2000s, the PKK had developed a new radical democratic project and
rejected the nation-state as the political trajectory for the Kurdish
movement. At the same time, the Kurdish movement that openly refers to
Öcalan’s thought grew much wider than the PKK itself, both in terms of
geographical reach and diversification of activities. From this transformation
onwards, describing the complex of organisations following Öcalan’s
thought as ‘the PKK’ is not only empirically incorrect as it is politically
problematic. Even if the PKK leadership’s influence over the wider
movement is stronger than they admit, it would be impossible for a guerrilla
group to directly control the galaxy of Öcalan-inspired organisations in
every field of society and spread around Turkey, the Middle East, and
Europe. Moreover, describing the wider movement as PKK-affiliated
legitimises the narrative of the Turkish government aimed at criminalising
every expression of Kurdish politics by labelling it as terrorism.

593
‘Final Resolution on the 8th Congress of the PKK’, Management Committee of
KADEK, 15 April 2002. Se also, Final Declaration of the Foundation Conference of
the People's Congress of Kurdistan, 17 November 2003, (Cologne: International
Initiative Freedom for Ocalan).
594
The Imrali court that sentenced him to death and the Court of Appeals in Ankara
in 1999, a local court in Şanlıurfa and the ECHR in 2001, a court in Athens in 2003,
and the Grand Chamber of the ECHR in 2004.
595
See Chapter 8.
261
With Marxism, the movement also abandoned the language of class politics,
and the ideological transformation was accompanied – and arguably
mutually reinforced – by the expansion and diversification of the
movement’s class base. The next section describes the PKK’s ideological
and organisational transformation in more detail to introduce the discussion
of the political-economic programme and class basis of the new Kurdish
movement.

Öcalan’s New Ideological Paradigm

This section summarises the main features of Democratic Confederalism,

the PKK’s new paradigm, as a coherent political project as it appears in the
2010s. However, it must be noted that Öcalan’s capture did not constitute a
clear cut-off point between the PKK’s Leninist and nationalist past and its
democratic and autonomist present. Even if the leader’s arrest in 1999
became the occasion to accelerate the shift, the PKK’s transformation is
doubtlessly in continuity with its evolution in the 1990s and particularly with
Öcalan’s developing critique of the nation-state and the patriarchal society.
As supporters proudly claim, the PKK’s resilience throughout the decades is
the result of its capacity to continuously evolve. In the words of Remzi
Kartal:

The desire for change never dies in the PKK. We change

continuously and according to the needs, not only of the Kurds but
of the people of the Middle East and of humanity. We have to change
to meet the demands and needs of the people. […] the dogmas that
have damaged the international left and that block the development
of the Middle East […] have to be broken to make change possible 596
Öcalan’s analysis, drawing heavily on the writings of American anarchist
Murray Bookchin, starts from a critique of the nation-state as a source of
oppression.597 The early democratic thinking of the European enlightenment

596
Interview with Remzi Kartal (Brussels, 2019).
597
Within academia, Joost Jongerden and Ahmet Akkaya are the authors of the
most comprehensive and thorough work of interpretation of Öcalan’s writings and
of the PKK’s new paradigm and this section largely relies on their publications on
the topic. See Joost Jongerden and Ahmet Akkaya, ‘The PKK in the 2000s:
262
was defeated by the rise of nationalism that translated into the pursuit of
culturally homogenous states – like the Turkish Republic – where all minority
groups had to be suppressed and forcefully assimilated. The democratic
solution for the Kurdish question cannot, therefore, be found in the
establishment of a Kurdish nation-state, which would be as oppressive as
the existing nation-states of Turkey, Iraq, Iran, and Syria where the Kurds
were denied their freedom. A Kurdish nation-state would need to create a
homogenous Kurdish people and to erase the regional, cultural, and
linguistic diversity of the land of Kurdistan. As an alternative, Öcalan
proposed the establishment of a ‘truly’ Democratic Republic in Turkey:

I offer the Turkish society a simple solution. We demand a

democratic nation. We are not opposed to the unitary state and
republic. We accept the republic, its unitary structure and laicism.
However, we believe that it must be redefined as a democratic state
respecting peoples, cultures and rights. On this basis, the Kurds
must be free to organize in a way that they can live their culture and
language and can develop economically and ecologically. This would
allow Kurds, Turks and other cultures to come together under the
roof of a democratic nation in Turkey.598
With continuous reference to Atatürk’s early speeches, Öcalan claimed that
granting Kurdish rights in Turkey would not undermine the founding
principles of the republic but strengthen its democracy.599 The PKK would,
therefore, support political parties that not only defend the Kurds but also

Continuity through breaks?’ in Nationalisms and Politics, ed. by Casier and

Jongerden, 143-162; Ahmet Akkaya and Joost Jongerden, ‘Reassembling the
Political: The PKK and the Project of Radical Democracy’, European Journal of
Turkish Studies, 14 (2012); Ahmet Akkaya and Joost Jongerden, ‘Democratic
Confederalism as a Kurdish Spring: The PKK and the Quest for Radical Democracy’,
in The Kurdish Spring: Geopolitical Changes and the Kurds, ed. by Michael Gunter
and Mohammed Ahmed (Costa Mesa, CA: Mazda Publishers, 2013), 163–86; Ahmet
Akkaya and Joost Jongerden, ‘Confederalism and Autonomy in Turkey: The
Kurdistan Workers’ Party and the Reinvention of Democracy’, in The Kurdish
Question in Turkey, ed. by Güneş and Zeydanlıoğlu, 186-204; Joost Jongerden,
‘The Kurdistan Workers’ Party (PKK): Radical Democracy and the Right to Self-
Determination Beyond the Nation-State’ in The Kurdish Question Revisited, ed. by
Stansfield and Shareef, 245-257.
598
Abdullah Öcalan, War and Peace in Kurdistan (Cologne: International Initiative
“Freedom for Abdullah Öcalan – Peace in Kurdistan”, 2008), 39.
599
Güneş, The Kurdish National Movement, 137.
263
promote the recognition of the cultural rights of all ethnic and religious
groups in Turkey as well as the secularists and other discriminated groups
such as the LGBT+ community.

The project of Democratic Confederalism goes even beyond that. Aware

that changing the current Middle East borders – in order to create a Kurdish
state – could only lead to bloodshed and displacement, the Öcalan proposes
to build democracy by ignoring, rather than dismantling, state borders.
Communities would autonomously organise on the basis of democratic
councils in a confederal relation to each other. In such a system, the need
to establish a Kurdish state would simply disappear. When framed in these
terms, the status of the Kurds as the world’s ‘largest stateless nation’ is
turned into the opportunity for an alternative political project and a true
democratisation of the region. In Öcalan’s project, a “Middle Eastern
democratic Confederalism […] will reunite those whose free lives were
destroyed by the nation-state wars imposed on the former mosaic of the
Middle Eastern peoples”.600

Following these ideological premises, the 9th Congress of the PKK in 2005
restructured the whole organisational framework of the group rejecting its
previous “state-like hierarchical structure” now deemed in “dialectic
contradiction to the principles of democracy, freedom and equality”. 601 The
PKK itself became – at least formally – just a component of a wider
confederal body called the Kurdistan Communities Union (KCK). The KCK, a
confederation of local councils and organisations, was constructed as a
direct alternative to the nation-state to collect all political forces that
support the project of Democratic Confederalism under the same
coordinating body. Within this framework, democracy takes the form of
people’s power, rather than that of representative politics. The PKK became
then the KCK-affiliated party representing the Kurds of Turkey and was

Abdullah Öcalan, Manifesto for a Democratic Civilization, Volume 1, Civilization:

600

The Age of Masked Gods and Disguised Kings (Porsgrunn, Norway: New Compass
Press, 2015), 39.
601
Öcalan, War and Peace, 28.
264
joined by a sister party for each of the Kurdish regions: the Kurdistan
Democratic Solution Party (PÇDK) in Iraq, the Democratic Union Party (PYD)
in Syria, and the Kurdistan Free Life Party (PJAK) in Iran.602 Moreover, the
KCK included the military wings, the women’s branches, as well as Kurdish
diaspora associations. After 2005, what used to be known as the PKK
became, rather than a party, a party-complex,603 organised in a less
centralised structure.

This transformation did not necessarily imply a significant change of the

leadership and most of the powerful positions in the KCK, the PKK, and the
military wings were taken by PKK veterans and members of Öcalan’s inner
circle.604 However, the new structure created a plenitude of organisations,
committees, and councils, empowering a generation of cadres who had
joined the Kurdish struggle during the mass growth of the PKK in the 1990s.
As described later in this chapter, the growth in strength and political
significance of other formally and informally affiliated organisations, such as
the women’s movement, the legal pro-Kurdish political parties in Turkey or
the Syria-based PYD, inevitably created alternative centres of power which
can only partly be subject to the acting leadership in Qandil. It is probably
not coincidental that this more polycentric structure of the PKK/KCK party-
complex suits the interest of a jailed leader unable to run the day-to-day life

602
While the PÇDK only had very marginal role in the crowded Iraqi-Kurdish political
environment, the PJAK has become an important competitors to the declining
Iranian Kurdish parties. The PYD is discussed more in details later in this chapter.
603
Jongerden and Akkaya, The PKK in the 2000s, 147;
604
In particular, Murat Karayılan and Cemil Bayık, two founding members of the
PKK, have maintained leading positions throughout this period. In 2013, Karayılan
left the leadership of the KCK to Bayık and became the head of the PKK’s military
unit. ‘Interview with Hozat, Karayılan and Bayık’, Peace in Kurdistan, 18 July 2013
<https://peaceinkurdistancampaign.com/2013/07/18/interview-with-hozat-
karayilan-and-bayik/>. The fact that senior commanders freely switch freely
switch positions between formally-independent organisation applies also to the
PYD and PJAK whose leaders were chosen among Syrian and Iranian PKK officials.
Mazloum Kobani, that in the late 2010s became an internationally-renown symbol
of the PYD-led Kurdish struggle in Syria was previously PKK representative to
Europe and guerrilla commander in Turkey. Interview with Cengiz Çandar (personal
communication, 2020).
265
of the organisation and worried to be sidelined by his own acting
commanders.

In his writings, Öcalan dedicated very little time to issues of class and
political economy. Whereas capitalism is at the centre of Öcalan’s critique
of Western modernity, this critique is historically oriented and driven by
moral considerations on individualism and oppression. The analysis of the
political economy of the Kurdish regions, of Turkey’s neoliberal
transformation, of the role of oil in Iraqi Kurdistan do not feature in Öcalan’s
writings, nor does the analysis of the evolving class structure of these
societies.605 Political economy was never the main focus of Öcalan’s
intellectual attention and certainly, after his capture, he could no longer
access the sources required for its study. The following sections of this
chapter cover the most important aspects of the PKK’s new paradigm as
well as the attempts at its practical implementation in Turkey and Syria
centring the analysis on the way they are shaped by the class structure and
political economy of the region.

Gender, Class and the Kurdish Women’s Movement

One of the most important aspects of the PKK/KCK’s new paradigm – and
certainly the one that has resonated most among non-Kurds – is its critique
of patriarchy and its emphasis on women’s liberation. Rather than attributing
this transformation solely to Öcalan’s prison writings, feminist scholars
showed that the centrality of women within the Kurdish movement is largely
the result of women activists and militants organising to offset the
patriarchal tendencies of the movement itself.

The PKK was founded in the 1970s as a Leninist organisation that supported
gender equality but subordinated women’s emancipation to national
liberation and socialism. In the late 1980s, Öcalan started to gradually revise

605
Michiel Leezenberg, ‘The Ambiguities of Democratic Autonomy: The Kurdish
Movement in Turkey and Rojava’, Southeast European and Black Sea Studies, 16.4
(2016), 676.
266
the PKK’s stance on women that was previously informed by classical
Marxist writings.606 Hamdan Çağlayan explains that Öcalan was particularly
successful at shifting the meaning of the concept of namus, honour, from
the male’s guarantee over the female body and its chastity to the defence
of the motherland violated by Turkish colonialism. The shift had the double
purpose of calling on men to defend their honour by joining the revolution
but also of removing “the namus barrier preventing women’s participation in
the same fight.”607 This revised attitude also led Öcalan to make a strong call
for the establishment of women-only military units and organisations. The
left-wing ideological framework of the PKK and Öcalan’s attention to the so-
called ‘woman question’ allowed for the inclusion of Kurdish women within
a masculinist project of nation-building “keeping in mind that they were
primarily invited by the leader.”608 However, it was only after his capture that
Öcalan placed the oppression of women at the origin of the history of
civilisation, as the first act of enslavement, and their liberation at the centre
of his revolutionary strategy:

State and power centres gave the father-man within the family a
copy of their own authority and had them play that role. Thus, the
family became […] the fountainhead of slaves, serfs, labourers,
soldiers and providers of all other services needed by the ruling and
capitalist rings. That is why they set such importance to family, why
they sanctified it.609
Within these premises, “[l]iberating life is impossible without a radical
women’s revolution which would change man’s mentality and life”.610

Çağlayan makes a point to explain that these theoretical discussions served

as “the preliminary steps for enabling gender egalitarian environments” but

606
Öcalan’s 1980s writings on the topic were collected in Abdullah Öcalan, Kadın
ve Aile Sorunu (Istanbul: Melsa Yayınları, 1992).
607
Handan Çağlayan, ‘From Kawa the Blacksmith to Ishtar the Goddess: Gender
Constructions in Ideological-Political Discourses of the Kurdish Movement in Post-
1980 Turkey’, European Journal of Turkish Studies, 14 (2012), 10-11.
608
Çağlayan, From Kawa the Blacksmith, 18.
609
Abdullah Öcalan, Liberating Life: Woman’s Revolution (Cologne: International
Initiative 'Freedom for Abdullah Öcalan – Peace in Kurdistan', 2013), 37.
610
Öcalan, Liberating Life, 51.
267
that the women’s experience was decisive in turning theory to practice.611 A
similar argument is central to the work of Nadje al-Ali and Latif Tas who
claim that Öcalan’s intellectual input “was as much inspired by the actual
experiences and struggles of Kurdish women’s rights activists as by outside
political philosophers” and only translated into gender-based equality
thanks to the continued struggle of Kurdish women within the wider
movement.612 It is important to properly historicised these theoretical
discussions. It is not by chance that Öcalan’s call for women participation in
the late 1980s and early 1990s coincides with the PKK’s entering the phase
of the people’s war that necessitated mass mobilisation. The growing
involvement of the civilian population in the Kurdish insurgency and Ankara’s
counterinsurgency campaigns, combined with a surge in women recruits for
the PKK, made women much more visible and present at every level of the
Kurdish movement.613

The need for separate women’s organisations was recognised by the PKK
in 1995, first at a women-only conference in March and then at the
organisation’s fifth congress in May which sanctioned the establishment of
women-only guerrilla units and the Kurdistan Women’s Freedom
Movement.614 In these years, women-only organisations combined with
gender quotas in the non-women organisations started playing a central role
in the strategy of the Kurdish women movement. As women members of
decision-making bodies were elected exclusively by women-only
organisations, they were accountable to their women-only constituency and
thus maintained their autonomy from the male-dominated leadership. The
principle of transitivity – as Çağlayan refers to it – “between women working
in the political parties’ women’s units and those with positions in decision-
making bodies” informed the strategy of both the women in the PKK and of

Handan Çağlayan, Women in the Kurdish Movement: Mothers, Comrades,

611

Goddesses (London: Palgrave Macmillan, 2020), 77.

612
Nadje Al-Ali and Latif Tas, ‘Reconsidering Nationalism and Feminism: The
Kurdish Political Movement in Turkey’, Nations and Nationalism, 24.2 (2018), 462.
613
For these events, see Chapter 8.
614
White, Coming Down, 121-123.
268
those in the pro-Kurdish political parties in Turkey.615 That the principle of
transitivity was strengthening women’s power within the PKK is made
evident by the reactions it triggered. In the immediate aftermath of Öcalan’s
capture, the male-dominated leadership of the PKK tried to curtail the
autonomy of the women’s divisions but was forced to backtrack by the
strong reactions of PKK women who also cut their hair in protest.616

The creation of a space for autonomous feminist politics in the pro-Kurdish

political parties in Turkey took more time. In the 1992 programme of the HEP,
gender issues took up a single paragraph in the 64-page document. In 1994,
the only woman elected to the 30-member party assembly of the HADEP
was the sister of one of the founders of the PKK.617 The great visibility
earned by Kurdish MP Layla Zana was not itself the sign of a deeper shift,
as explained by former Diyarbakır co-mayor Gültan Kışanak:

we had Leyla Zana as an example of an MP in 1991, but her role and

power did not come from women's organisations or representation.
She was the wife of a famous Kurdish politician, Mehdi Zana […]
When Leyla Zana was first elected, her social status and her being
the wife of somebody famous was the main reason for her to be
elected.618
Kurdish women activists challenged these practices by claiming spaces
through the creation of women-only branches and by demanding change in
the party’s structure. The HADEP adopted a women’s quotas of 25 per cent
in 2000 and its successor raised the quotas to 35 per cent in 2003 and 40
per cent in 2005.

From 2004, the Kurdish movement in Turkey adopted the so-called ‘co-
chair system’ that became a worldwide known symbol of the achievements

615
Çağlayan, Women in the Kurdish Movement, 126.
616
Cited in Joost Jongerden, ‘Learning from Defeat: Development and Contestation
of the “New Paradigm” Kurdistan Workers’ Party (PKK)’, Kurdish Studies, 7.1 (2019),
85.
617
Çağlayan, Women in the Kurdish Movement, 101-102.
618
Latif Tas, Nadje Al-Ali, and Gültan Kişanak, ‘Kurdish Women’s Battle Continues
against State and Patriarchy, Says First Female Co-Mayor of Diyarbakir. Interview
’, OpenDemocracy, 12 August 2016 <https://www.opendemocracy.net/en/kurdish-
women-s-battle-continues-against-state-and-patriarchy-/>.
269
of the Kurdish women’s movement. All the single-person positions in the
Kurdish movement were ‘doubled’, to include both a man and a woman. The
system was then extended to the local administration and the pro-Kurdish
parties appointed co-mayors in the cities they won. Moreover, the growing
strength of women within the parties gradually changed the profile of the
Kurdish female politicians. As long as elected women were used as tokens,
female politicians tended to be either famous women or wives of famous
men. As the women’s movement grew stronger, women from more diverse
“economic, social, ethnic and religious backgrounds” were elected to party
positions or public offices.619 As Kışanak – a co-mayor herself – explains,
female co-mayors were initially perceived “as assistants” to their male
colleagues. Yet, with time and political labour, their role was gradually
acknowledged by the movement’s base.620

The achievements of the Kurdish women since the early 2000s is often
attributed to Öcalan’s prison writings and especially to his
reconceptualization of Kurdish feminism into a ‘science of women’. The term
Jineology, coined by Öcalan, defines a “new science” that “criticizes the
connection of hegemony, oppression and science” and “the hegemony of
men on history.”621 As Al-Ali and Tas observe, Öcalan – normally an avid
reader of critical theory – fails to acknowledge that, long before him, critical
feminist scholars of different theoretical strands have denounced the
patriarchal nature of the social sciences as well as the links between liberal
and white feminisms and imperialism.622 This remark on the dubious
originality of Öcalan’s feminist writings is particularly relevant because the
imprisoned leader is often the only theoretical reference explicitly
mentioned by Kurdish women’s activists, at times, in a rather dogmatic

619
Tas and others, Kurdish Women's Battle.
620
Tas and others, Kurdish Women's Battle.
621
‘Jineolojî’, International Vrije Vrouwen Stichting, 2019
<https://www.ifwf.nl/2019/03/15/jineoloji/>.
622
Al-Ali and Tas, Reconsidering Nationalism and Feminism, 467.
270
way.623 This is problematic because it erases the role of the Kurdish women’s
movement and attributes its success to the intellect of the male leader.
Moreover, Jineology ethnicises Kurdish feminism, reinforcing an idea of
Kurdish exceptionalism that negatively affects its capacity to create
alliances with non-Kurdish feminists across the region. If Öcalan’s
overarching presence as a symbol and theoretical presence seems to deny
women’s authorship over their own achievements, it can also be the result
of a patriarchal bargain made by the Kurdish women.624 The appeal to the
(male) supreme leader’s intellectual authority would be, in these terms, an
indispensable device employed by the women activists to give legitimacy to
feminist ideas and practices across a male-dominated Kurdish movement.

As observed above, the autonomisation of women’s politics within the

Kurdish movement through the establishment of separate women’s
organisation expanded the social basis of Kurdish female politicians.
Especially since the mid-2000s, elected Kurdish women reflected the more
diverse class base of the movement that set them apart from the profile of
the female politician in Turkey, most often middle or upper class and highly
educated.625 If before, women were handpicked by the male leadership to
serve as tokens, the creation of the women-only organisations
‘democratised’ the women’s movement giving a more powerful voice to the
largely working-class base of the Kurdish movement. This change can be
observed in the social policies implemented by Kurdish-controlled
municipalities in south-eastern Turkey hiring women in public administration
– including male-dominated jobs like bus drivers – setting up free laundry
services and supporting the establishment of women’s co-operatives.626

623
Al-Ali and Tas, Reconsidering Nationalism and Feminism, 470.
624
As women’s strategizing “within a set of concrete constraints” imposed by
patriarchy. Deniz Kandiyoti, ‘Bargaining with Patriarchy’, Gender & Society, 2.3
(1988), 274–90.
625
Çağlayan, Women in the Kurdish Movement, 121-122.
626
See the report written by the German solidarity organisation TATORT Kurdistan
Kampagne after the observation tour in the region in 2011. TATORT Kurdistan,
Democratic Autonomy in North Kurdistan (Porsgrunn, Norway: New Compass
Press, 2013).
271
These initiatives, aimed at tackling women’s unemployment and poverty, are
clearly aimed at the movement’s working-class constituency and shows the
capability of women-only bodies to reflect the social need of inner-city
women. However, this attention is not necessarily reflected in the political
discourse which more often associates womanhood with Kurdishness – as
in Öcalan’s writings – as two interlocked sources of oppression leaving little
theoretical room for the structural role of class hierarchies. In the
movement’s theoretical analysis, class is more often conflated with
gender,627 and, by treating women as a class, class difference and inequality
between women inevitably lose political significance.

Hegemonic Struggle in Turkish Kurdistan

Öcalan’s prison writings had a great impact on the pro-Kurdish legal parties
in Turkey which were receptive to a less militant approach. In the 2000s,
pro-Kurdish parties took advantage of a relatively less oppressive political
environment that allowed for their sustained electoral growth. On the one
hand, the PKK had withdrawn from Turkey in 1999 and, even when fighting
resumed in 2004, the intensity of the insurgency did not reach the level of
the 1990s and left more room for electoral politics.628 On the other hand, the
2002 elections brought the Islamist Justice and Development Party (AKP)
led by Recep Tayyip Erdoğan to power. In the 2000s, especially thanks to
the country’s sustained economic growth, the AKP was able to absorb a
large part of the centre-right vote with their moderate and liberal platform
of reform. The AKP’s commitment to the accession of Turkey to the EU led
to a partial liberalisation of the political system and a slightly more open
debate on the Kurdish issue. Moreover, the AKP inherited the strength in the

627
Çağlayan, Women in the Kurdish Movement, 4, 120; Al-Ali and Tas,
Reconsidering Nationalism and Feminism, 468.
628
While in the mid-1990s the conflict cost thousands of life each year, casualties
throughout the 2000s were in order of hundreds. ‘Turkey: Kurdistan’, Uppsala
Conflict Data Program, Conflict Encyclopedia, Uppsala University
<https://ucdp.uu.se/additionalinfo/354/1>.
272
Kurdish region of its Islamist predecessor, the Welfare Party,629 and became
the main competitor of the pro-Kurdish parties. While these conditions
allowed for a more relaxed political environment relative to the 1990s, the
pro-Kurdish parties continued to face the heavy hand of the (largely)
Kemalist Turkish bureaucracy as “provincial governors, prosecutors,
security forces, and other central authorities retained considerable capacity
to circumscribe their activities through bureaucratic and legal
procedures.”630 Turkish courts continued to ban pro-Kurdish parties just a
few years after their establishment. The HADEP, banned in 2003, was
replaced by Democratic People's Party (DEHAP) itself banned in 2005 and
succeeded by the Democratic Society Party (DTP). The DTP was shut down
in 2009 and the Peace and Democracy Party (BDP) took its place.631

In the local elections of 1999, held only a few months after Öcalan’s capture,
the HADEP achieved quite a spectacular result winning seven provincial
capitals (Diyarbakır, Van, Batman, Mardin, Hakkari, Siirt, Ağri) and thirty
more municipalities across the south-east. However, in 2004, the DEHAP
faced a much stronger advance of the AKP. Despite winning some additional
seats across municipalities, the DEHAP lost Siirt, Van, Bingöl, and Ağrı to the
AKP. Breaking the traditional taboo of denying the existence of the Kurds in
Turkey, Erdoğan made continuous reference to Turkish-Kurdish
brotherhood especially in terms of shared religion and history. Moreover,
the AKP heavily invested in a narrative that depicted pious Turks and Kurds
as fellow victims of secularist oppression imposed by both the Kemalist elite
and the PKK terrorists.632 The fact itself that, in the 2000s, the AKP was
often in conflict with the Turkish military and Kemalist establishment made

629
As shown in Chapter 8, in the 1990s pro-Kuridish and Islamist forces largely
displaced Turkish mainstream parties in the south-east becoming the main
competitors for the electoral control of the region.
630
Watts, Activists in Office, 143.
631
See Table 3.
632
Ioannis N. Grigoriadis and Esra Dilek, ‘Struggling for the Kurdish Vote: Religion,
Ethnicity and Victimhood in AKP and BDP/HDP Rally Speeches’, Middle Eastern
Studies, 54.2 (2018), 293-295.
273
it easier for Kurds to sympathise with the new government.633 At the time,
the AKP could also count on the support of the Hizmet movement, led by
the Islamic preacher Fethullah Gülen, that built a massive network of charity
and educational institutions in Turkey’s south-east.634

However, the strength of the AKP in the Kurdish region cannot be

exclusively attributed to its ideology and discourse. The AKP’s success in
the south-east was also the result of its capacity to replace previous centrist
Turkish parties as the point of reference for the Kurdish traditional elite. In
the least urbanised provinces – as shown by Feryaz Ocaklı in his study on
Muş and Bingöl – where tribal ties maintained a strong social value, the AKP
inherited the network of support of the previous ruling parties among pro-
government tribal leaders.635 Regardless of the ideological positions
expressed by the AKP, siding with the ruling party was, for much of the
Kurdish ruling class, indispensable to maintain their social power over their
constituencies. The ‘state-Agha alliance’ – that sustained power
relationships in the south-east for a good half of the previous century –636
remained in place in the AKP era as economic growth, and particularly the
construction boom multiplied the opportunity for enrichment for local
notables – of tribal and non-tribal extraction –637 supportive of the AKP. In
the rural areas, the alliance continued to rest heavily on the existence of the

633
For example, in occasion of the 2007 presidential elections.
See Mustafa Gürbüz, Rival Kurdish Movements in Turkey: Transforming Ethnic
634

Conflict (Amsterdam: Amsterdam University Press, 2016).

635
Feryaz Ocaklı, ‘Politics in the Kurdish Periphery: Clan Networks and Local Party
Strategies in a Comparative Perspective’, Middle Eastern Studies, 53.4 (2017), 571–
84.
636
See Chapter 7.
637
As already discussed in Chapter 7, the ruling class of the Kurdish region only
partly coincides with the tribal elite. In many areas, especially where urbanisation
weakened tribal ties, the local ruling class is more likely composed by non-tribal
landowners, entreprenours owning factories and construction companies, and rich
professionals, and resembles more closely that of the rest of Turkey. However, in
rural areas, tribal leaders often continue to be powerholders allied to the state and
the ruling party. For example, Chapter 7 mentioned Sadun Şeylan as the head of
the Alan tribe and pro-state warlord in the 1990s. In 2019, his son and successor
Abdurrahman Şeylan was elected mayor of Çatak, the ancestral homeland of the
Alan tribe, for the AKP.
274
Village Guards that allowed pro-government tribal leaders to maintain their
semi-private petty armies at the expense of the state.638 The DEHAP’s
setback at the 2004 local elections was therefore not only the result of the
Islamists’ appealing discourse but also of the migration of the local elite from
the previous ruling parties to the now dominant AKP that restructured power
relations in the region.

Since 1999 and for most electoral challenges of the 2000s and early 2010s,
the pro-Kurdish parties increased their share of municipalities and
parliamentary seats. The parties’ control of municipalities across the
country’s south-east was used to challenge the Turkish state over issues
such as the language taboo, introducing the provision of municipal services
in Kurdish and other minority languages.639 But the pro-Kurdish parties and
municipalities embarked on a wider project, using their growing power and
popular support to implement the concept of Democratic Autonomy as
developed by Öcalan’s prison writings. In October 2007, the Kurdish
movement organized its parties, municipalities, and civil society
organizations into the Democratic Society Congress (DTK) with the purpose
of building a stateless democracy without directly challenging the existence
and borders of the Turkish state.640 The Kurdish movement promoted the
establishment of citizen councils which worked alongside public institutions
and took the form of neighbourhood assemblies based on direct democracy
anywhere pro-Kurdish politics was strongest.641

638
Despite the AKP government’s promise to abolish it as a part of the EU-
accession reforms, the Village Guards system remained in place. See ‘Turkey’s
Village Guard System – Still in Place, Still an Obstacle’, Kurdish Human Rights
Project, Briefing Paper, 22 March 2011 <http://www.khrp.org/latest-news/646-
turkeys-village-guard-system-still-in-place-still-an-obstacle.html>
Watts, Activists in Office, 152-153. Particular emphasis on minority languages
639

was given by the 2005 DTP charter. See Demokratik Toplum Partisi: Program ve
Tüzüğü, 2005.
640
‘Democratic Society Congress’, Demokratik Toplum Kongresi
<https://www.kcd-dtk.org/english-2/>.
641
TATORT, Democratic Autonomy, 25-28.
275
As the pro-Kurdish parties became the dominant force in many Kurdish
provinces – even though never across the entire region – they also tended
to become more inclusive. At the 2011 elections, the BDP led a coalition of
small Turkish left-wing groups but was also able to add to their list and elect
to parliament veteran Kurdish politicians coming from a conservative and
Islamic background such as Şerafettin Elçi and Altan Tan. The fact that such
personalities ran as independent candidates for the BDP was the sign that
the Kurdish movement was winning its hegemonic bid over the region as
political rivals were persuaded to cooperate from a position of
subordination. In cities like Diyarbakır where pro-Kurdish parties had been
continuously in power since 1999, a wider section of the affluent middle and
upper classes was increasingly orbiting around the movement. As pro-
Kurdish parties were now (locally) in power, expressing sympathy for the
movement for businessmen and professionals was not only less ‘socially
costly’, it could even open the doors to public contracts and high-level
positions in the local administration. This process, however, was limited to
certain areas and never threatened the alliance between the local ruling
class and the AKP especially because state repression continued to hit the
pro-Kurdish parties and the Kurdish-controlled administrations.

The growing strength of the Kurdish movement was, in fact, bound to

generate the reaction of the state both in terms of repression and political
opening. In 2009, Turkish security forces launched an operation to arrest
thousands and convict hundreds of Kurdish activists and elected officials
on charges of terrorism. The so-called ‘KCK trials’, that took place between
2010 and 2012, directly targeted the legal expression of the Kurdish
movement but also became the opportunity to hit sympathetic Turkish left-
wingers and trade unionists.642 While the state apparatus deployed a
repressive strategy, the AKP government made the first serious attempt to

642
Systematic information on the trials was published on the bulletin of the Human
Rights Foundation of Turkey collected and translated by their German support
group, the Democratic Turkey Forum. See ‘Backgrounder on the Union of
Communities in Kurdistan, KCK’, Democratic Turkey Forum,
<http://www.tuerkeiforum.net/enw/index.php/Backgrounder_on_the_Union_of_Co
mmunities_in_Kurdistan,_KCK#cite_note-bia201110-9>.
276
offer a political solution to the Kurdish issue. Between 2009 and 2011, the
government held secret negotiations with the PKK in Oslo. The failure of this
first round of talks led to a new wave of violence in 2011 and 2012, the
bloodiest years since fighting had resumed in 2004.643 However, in
December 2012, Erdoğan officially announced that Ankara was negotiating
with the PKK’s jailed leader and, in March 2013, Öcalan ordered the PKK to
end the armed struggle and to withdraw from Turkey. Even if the peace
process was short-lived and unsuccessful, it yet gave a boost of democratic
legitimacy to the Kurdish movement. As advised by Öcalan, in late 2013,
Kurdish activists launched the Peoples' Democratic Party (HDP) as a
coalition of the Kurdish BDP and minor socialist, environmentalist, feminist,
LGBT+, and ethnic and religious minorities groups.644 The growth of the HDP
was partly led by a section of the Turkish millennial generation who had their
first political experience in the anti-AKP 2013 Gezi Park protests in Istanbul.
The charismatic HDP co-chair Selahattin Demirtaş was particularly
successful in presenting the HDP as the only real alternative to the AKP and
its project of a radical pluralist democracy as the political antithesis of
Erdoğan’s AKP.645 The HDP’s narrative was strengthened by the fact that the
years of the peace process coincided with Erdoğan’s gradual authoritarian
and conservative turn.

In 2015, as the peace negotiations in Turkey reached a standstill, the

advance of the PKK-linked PYD in the Syrian Kurdish region and along the
Turkish border increased tensions in Turkey and Erdoğan’s increasingly
anti-Kurdish rhetoric was matched by a surge in extrajudicial violence
against Kurdish and leftist activists, as well as HDP offices.646 At the June

643
‘Turkey: Kurdistan’, Uppsala Conflict Data Program.
644
After the 2014 local elections, the BDP was replaced by the by Democratic
Regions Party (DBP) which is now complementary to the HDP. The HDP runs for
parliamentary and presidential elections and is concerned with national politics
while the DBP exclusively works on the local administrations in the Kurdish south-
east.
Zeynep Kaya and Matthew Whiting, ‘The HDP, the AKP and the Battle for Turkish
645

Democracy’, Ethnopolitics, 18.1 (2019), 92–106.

646
See Francis O’Connor and Bahar Başer, ‘Communal Violence and Ethnic
Polarization before and after the 2015 Elections in Turkey: Attacks against the HDP
277
2015 elections, the HDP reached a surprising result – 13 per cent – that cost
the AKP its parliamentary majority. In July 2015, the peace process
collapsed and the fighting between the PKK and the army resumed. As the
AKP recovered its majority with the November 2015 snap elections,
Erdoğan’s move to the right was sealed by the alliance with the fascistoid
and violently anti-Kurdish MHP. From summer 2015, the HDP faced an
unprecedented level of repression. In November 2016, HDP co-chairs
Demirtaş and Figen Yüksekdağ were imprisoned. Dozens of HDP/BDP
mayors elected in 2014 were removed by the government, replaced by
trustees, and removed again after being re-elected in 2019. Despite this
degree of repression, the HDP has maintained quite significant support.
Thanks to the alliance with progressive Turkish groups “in the main cities in
Turkey’s west, […] the HDP is also quite popular among the university
students and urban intelligentsia from the more affluent districts and
neighbourhoods.”647 However, the HDP’s project of building a wider
progressive coalition including larger sections of the non-Kurdish electorate
has been heavily undermined by the government’s criminalisation of the
whole of the Kurdish movement as well as by the reluctance of the
mainstream Turkish opposition to co-operate with them.

The war threatened the HDP/BDP also in its Kurdish heartland. The state
violence deployed in the Kurdish region has been directed at systematically
dismantling the social power built up by the Kurdish movement in the
previous decades. The vast social coalition set up to support the Kurdish
municipalities has been under strain and the party had to find a difficult
balance between the “conservative and increasingly affluent strata that
would otherwise vote AKP as they had done before and the impoverished
youths […] who showed a permanent potential for more radical political

and the Kurdish Population’, Journal of Southeast European and Black Sea, 18.1
(2018), 53–72.
647
Cengiz Güneş, ‘The rise of the pro-Kurdish democratic movement in Turkey’, in
Routledge Handbook on the Kurds, ed. by Michael Gunter, 263.
278
action.”648 The new phase of the conflict from autumn 2015 was marked by
semi-spontaneous uprisings in the small towns of the PKK heartland on the
Turkish border with Syria and Iraq. These armed revolts, largely led by poor
and unemployed youth, revealed a degree of dissatisfaction with the
movement’s moderate and legalistic stance.649

The Rojava Experiment

The democratic revolution that began in Syria in 2011 in the wider context
of the Arab uprisings became an unexpected opportunity to establish a
Kurdish autonomous entity in the country and to experiment with Öcalan’s
project on a larger scale. The roughly two million Kurds, who live in the north
of Syria and the major cities, have a deep historical relation to the Kurds of
Turkey with whom they share the Kurmanji dialect. The establishment of the
Syrian-Turkish border in 1923 separated tribes, clans, and families affecting
especially the nomadic Kurds that moved across the area. Thousands of
Kurds from Turkey migrated to Syria in the 1920s and 1930s after the failure
of the early Kurdish revolts.650 These movements provided a recurrent
excuse for Syrian Arab nationalists to present the Kurds as alien and, in
1962, 120.000 Syrian Kurds were stripped of their citizenship.651

While many Kurds, in the 1930s and 1940s, were involved in anti-landlord
struggles and in the workers’ movement up to the point of dominating the
Syrian Communist Party (SCP), the first Kurdish nationalist party was the
Kurdistan Democratic Party of Syria (KDPS) established in 1957 under the

Michiel Leezenberg, ‘The Rise of the White Kurds – An Essay in Regional Political
648

Economy’, in The Kurds in a Changing Middle East, ed. by Jabar and Mansour, 107.
649
Metin Gürcan, ‘PKK Looks to the Future with Creation of Youth Militias’, Al-
Monitor, 31 August 2015 <https://www.al-
monitor.com/pulse/originals/2015/08/turkey-kurds-pkk-armed-young-
militias.html>.
650
The feudal revolts of the interwar period (see Chapter 3).
651
Jordi Tejel, Syria’s Kurds: History, Politics and Society (London and New York:
Routledge, 2009), 1-7, 50-51. According to Alan Semo, by the 2000s, “the number
of stateless Kurds in Syria reached half a million.” Interview with Alan Semo
(personal communication, 2020).
279
influence of the Iraqi KDP.652 Like its Iraqi-Kurdish model, the KDPS was
ravaged by the very same “tensions between its left-wing (ex-SCP
members, young students, teachers and manual laborers) and its right-wing
(notables, religious leaders and landowners).” Since the 1960s, the KDPS
faced numerous splits that over time were characterised less by ideological
differences than by personal rivalries and, on the eve of the 2011 uprisings,
a dozen parties were claiming its legacy.653

The PKK was allowed to establish itself in Syria after the 1980 Turkish
military coup and to use the country for training, logistics, and
recruitment.654 Attracted by the militancy of the PKK and with the tacit
consent of Damascus, the number of Syrian Kurds in the PKK ranks grew to
the point that, in the late 1990s, nearly twenty per cent of PKK ‘martyrs’ were
Syrian.655 As Alam Semo a Syrian Kurdish student at the time explains

[The PKK] came with a new idea and a new project, a vision and a
clearer position for an independent and united Kurdistan. Syrian
Kurdistan was open to these ideas […]. The elite was linked to the
KDP in Iraq and at that time they had lost their bases and their
argument. […] Most of them were landowners and tribal leaders. The
PKK took advantage of the weakness of the classical nationalist
parties.656
After the expulsion of the PKK from Syria in 1998, its followers and
sympathisers reorganised and in 2003 established the PYD which grew a
strong clandestine network. Its organisational strength and links to the PKK
placed the PYD in the best position to take advantage of the outbreak of the
Syrian civil war in 2011. When, in summer 2012, the Syrian regime withdrew
from the Kurdish region in order to defend more vital areas, the PYD took
control. The PYD could also rely on a pre-existing political infrastructure

652
Tejel, Syria’s Kurds, 39-42.
653
Ibid., 87-88.
654
See Chapter 7.
655
Accoriding to Güneş Murat Tezcür’s martyrs data, 529 out of the 2,919 PKK
military casualties between 1995 and 1999 were Syrian. Güneş Murat Tezcür,
‘Ordinary People, Extraordinary Risks: Participation in an Ethnic Rebellion’,
American Political Science Review, 110.2 (2016), 247–64.
656
Interview with Alan Semo (personal communication, 2020).
280
constituted by a clandestine network of local councils dating back to 2007
and organised along the lines of Öcalan’s democratic confederalist
project.657 With the beginning of the uprisings, these councils emerged and
took “responsibility of organizing social life” including the provision of basic
services and security.658

The PYD established an autonomous administration in north-eastern Syria

widely known as Rojava.659 In late 2014, the Rojava administration repelled
the offensive of the ISIL in the city of Kobanî and, with American support,
gained control of virtually all of Syria east of the Euphrates by 2017. The role
of the Kurdish-led Syrian Democratic Forces (SDF) in the fight against ISIL
in north-eastern Syria triggered a great deal of international sympathy and
grassroots solidarity though little material support from state actors. The
important presence of female combatants’ units contrasted sharply with
ISIL’s misogynist brutality.

The system of government established in Rojava is based on a Social

Contract that was adopted by the councils in 2014 and amended in 2016
and that works as a constitution for the regional administration.660 The
diffusion of the council system and of forms of direct democracy greatly
expanded popular participation in the governance of the region in strong
contrast to the authoritarian nature of the Syrian regime. Despite this
emphasis on council democracy, many observers denounced the lack of
pluralism of the PYD rule and the repression of rival Kurdish parties.661 The
greatest and most widely recognised achievement of the Rojava

657
Interview with Khawla Mustafa (Sulaymaniyah, 2019).
658
Cited in Akkaya and Jongerden, Democratic Confederalism, 173.
659
The name Rojava, the Kurdish for ‘West’, is a common way for Kurds to refer to
the Syrian Kurdish region as the western part of Kurdistan.
660
Reader #5: Stateless Democracy, realized with Kurdish Women’s Movement,
New World Academy, 19 May 2015, < https://www.e-
flux.com/announcements/29509/new-world-academy-reader-5-stateless-
democracy-realized-with-kurdish-women-s-movement/>, 130-157.
661
See, for example, Fred Abrahams and Lama Fakih, Under Kurdish Rule: Abuses
in PYD-Run Enclaves of Syria (Human Rights Watch, 2014)
<https://www.hrw.org/report/2014/06/19/under-kurdish-rule/abuses-pyd-run-
enclaves-syria>.
281
administration has been in the direction of women’s liberation. The
introduction of the co-chair system and of separate women’s bodies is
having a deep impact on established gender relations in the region.662

The PYD has placed particular emphasis on the Mesopotamian or Kurdistani

– rather than Kurdish – identity of Rojava with an inclusive attitude towards
the ethnic and religious minorities living in Kurdish-majority areas such as
Arabs, Turkmens, Yazidis, and various Christian groups. Minorities have
been given highly visible positions in the region’s decision-making bodies
and were encouraged to establish their own political organisations and
military units. It remains, however, to be seen the extent to which council
democracy was extended to the Arab-majority areas of the governorates of
Raqqah and Deir Ez-Zor. In the Syrian desert local power was partly
delegated to tribal leaders who had opposed ISIL allowing for the
reproduction of authoritarian and patriarchal modes of politics in direct
contradiction with Rojava’s proclaimed democratic aspirations. The success
– or lack thereof – of the Rojava model in Arab-majority areas will be critical
to understanding whether Öcalan’s political project can be completely
dissociated from Kurdish ethnonationalism and implemented in a non-
Kurdish context. As of mid-2020, the Rojava administration has guaranteed
relatively peaceful coexistence in a very diverse region especially when
considering the extreme levels of sectarian violence reached within the
context of the Syrian Civil War.

The Political Economy of the Kurdish Movement

The Kurdish movement inspired by Öcalan’s thought moved away from its
initial socialist positions and adopted a new political project. The critique of
capitalism has remained part of the PKK/KCK’s ideology but is by no means
a central aspect and its intellectual foundations can be found, rather than in
Marxian analysis, in Murray Bookchin’s anarchism and anti-materialism. The

See Michael Knapp, Anja Flach, and Ercan Ayboğa, Revolution in Rojava:
662

Democratic Autonomy and Women’s Liberation in Syrian Kurdistan (London: Pluto

Press, 2016).
282
social and economic programme of the movement is indeed heavily
influenced by Öcalan’s reading of Bookchin and by the latter’s idea of the
organic society as the primordial form of human organisation.663 The
construction of a confederation of municipalities based on direct democracy
from the local level is, for Bookchin, an indispensable way to return to an
organic society which is interdependent, cooperative, and ecologically
sustainable. Very much like democracy, the economy needs to be localised
and decentralised.664 These ideas translated into an economic project that
is alternative to capitalism but also to both Soviet-style and Keynesian
centralised planning and that aims at building a decentralised and
sustainable ‘social economy’.

Even if economic transformation as a path for popular emancipation is

marginal in Öcalan’s writings, the idea of the organic society inspired the
anti-capitalist practices and policies enacted by the Kurdish movement.
Naturally, the very limited space for governance available to the Kurdish
movement in Turkey and the state of war in which the Rojava experiment
has taken place relegated discussions of social and economic policies to a
marginal place. In the case of Rojava, it is particularly difficult to distinguish
socialist policies from war-economy policies – as in the case of food
distribution and price control. However, the Kurdish movement in both
Turkey and Syria has maintained strong popular bases – among the
peasantry, the working class, and the urban poor – that shaped its
programme in the direction of redistribution and even collective ownership
in the form of cooperative enterprises. An academic study funded by the
Turkish police in 2014 shows that, despite its ideological transformation, the
PKK is still overwhelmingly composed of Kurds with a lower-class
background. According to their findings, 72 per cent of the militants only

663
Murray Bookchin, The Ecology of Freedom: The Emergence and Dissolution of
Hierarchy (Palo Alto: Cheshire Books, 1982).
664
Bookchin, The Ecology of Freedom, 344.
283
have an elementary school diploma and 78 per cent were unemployed
before joining the guerrilla.665

The establishment of cooperatives has become the most popular economic

policy promoted by the Kurdish movement in Turkey and Syria. In south-
eastern Turkey, cooperatives were set up in a very diverse set of fields, from
agriculture to housing to carpet production.666 In Rojava, cooperatives
dominate agriculture – the region’s dominant sector – but have also
mushroomed in urban contexts “for bread-baking, textile production,
sewing and alterations, cheese-making and other dairy production, growing
peanuts and lentils, and selling cleaning materials”.667 These cooperatives
are not allowed to be independent enterprises and, even if their leadership
is elected, they remain dependent on the democratic control of the local
councils which ensure that production meets the needs of society.668 The
local authority receives 20 per cent of the cooperative profit, while 50 per
cent is divided among the workers and the remaining 30 per cent is re-
invested in production.669 The creation of this model of a cooperative
economy, in combination with the state of war, brought the economy of
Rojava largely under council control leaving only a marginal and
subordinated role to profit-led enterprises.

The Kurdish cooperative movement has been particularly successful in the

field of agriculture which holds a central ideological role in the idea of the
organic society and in the movement’s critique of the city. With the ambition
to reach food self-sufficiency, the Rojava autonomous administration

665
This study does not use PKK ‘martyr’ data like others cited earlier but is largely
based on police records. Süleyman Özeren, Murat Sever, Kamil Yılmaz, and Alper
Sözer, ‘Whom Do They Recruit? Profiling and Recruitment in the PKK/KCK’, Studies
in Conflict & Terrorism, 37.4 (2014), 322–47.
666
The TATORT report dedicate ample spaces for examples supported by
interviews with cooperative workers. See TATORT, Democratic Autonomy, 34-35,
44, 82-86, 101, 109.
667
Knapp and others, Revolution in Rojava, 200.
668
Ibid., 204-205.
‘The Experience of Co-operative Societies in Rojava’, TEV DEM Economic
669

Committee, 19 January 2016, 4.

284
moved towards crop diversification as opposed to the Ba’athist policy of
centrally-planned monocrop production.670 Given that the Syrian state
owned more than 80 per cent of the land, the Rojava authority did not have
to collectivise privately-owned land – which often remains in the hands of
large landowners –671 and instead built rural cooperatives on state land. 672
By early 2016, 330.000 acres of land was farmed by cooperatives. 673 In
Turkish Kurdistan, land distribution remained extremely uneven. In 2001, the
2.7 per cent largest landowners owning more than 33.2 per cent of the land
while 56 per cent Kurdish peasants own less than 50 hectares accounting
for only 9.7 per cent of the land.674 The economic unviability of the smallest
farms pushed some Kurdish peasants to support the establishment of rural
cooperatives whereby they self-collectivised their own small plots of land.675

The critique towards the city as the centre and propagator of capitalist
modernity that underpins the Kurdish movement’s push for a return to the
‘organic’ life of the village, is directly inspired by Bookchin’s thought.
However, it is also extremely relevant to the life experience of a large part
of the movement’s base in Turkey which is effectively composed of
peasants violently uprooted from their villages and deported into city slums
in the 1990s.676 As a Kurdish activist in Turkey recalled in 2015:

Our villages were razed, forcing us to move into the cities. We don’t
know what city life is like or how you’re supposed to live here. […] Due
to the forced relocation into the cities and modern capitalism, an
extreme individualism prevails here. And we have to deal with it.677

670
Knapp and others, Revolution in Rojava, 192-194.
671
Interview with Alan Semo (personal communication, 2020).
672
Knapp and others, Revolution in Rojava, 199.
673
The Experience of Co-operative Societies, 1.
674
Yadirgi, The Political Economy, 258.
675
For example, in the Van province. Eliza Egret and Tom Anderson, Struggles for
Autonomy in Kurdistan & Corporate Complicity in the Repression of Social
Movements in Rojava and Bakur (London: Corporate Watch Cooperative, 2016)
<https://corporatewatch.org/wp-content/uploads/2017/09/Struggles-for-
autonomy-in-Kurdistan.pdf>.
676
See Chapter 8.
677
TATORT, Democratic Autonomy, 40.
285
The relevance of the environmental turn of the movement to the life
experience of millions of its supporters is quite a significant example of the
way the new paradigm of the PKK/KCK speaks directly to much of the
Kurdish masses. In these terms, the emphasis on the organic society and
the critique of the city are truly ideological expressions of the forcibly-
urbanised peasantry that represents the core of the class basis of the
movement – in terms of electoral support for the pro-Kurdish parties and
manpower for the insurgency.

As largely confined to the moral level, this critique of capitalism tends to

neglect the analysis of the material conditions of the Kurdish regions and
thus ends up concealing class conflicts and social hierarchies. In reference
to the Kurdish women’s movement, feminist scholar and activist Handan
Çağlayan observes that

The neoliberal transformation worsened the situation of poor Kurds,

both in rural areas and in the cities. […] in Istanbul, we saw that in
poor [Kurdish] families almost all the young women were working at
textile workshops with no social rights and social security. Their
wages were under minimum wage level. Among seasonal agriculture
workers who have been working under the worst working
conditions, there have been hundreds thousand of Kurdish women.
Still, it is difficult to state that social issues go beyond the rhetorical
level in the political discourse, political agenda and practices of
Kurdish women. […] One of [the reasons], I think, is the assumed
strict connection between class-based exploitation and deprivation
of Kurds from rights.678
By conflating exploitation with cultural oppression, social conflicts within
Kurdish society fade away. The virtual absence of class analysis in the
elaboration of the new Kurdish movement can be explained by the
worldwide advance of identity politics in the 1990s and 2000s that coincide
with the period of deepest transformation of the PKK. Those decades were
also characterised by a great expansion of the class basis of the PKK,
particularly towards the urban middle classes. As discussed earlier in this
chapter, activists with a middle-class background – and particularly

678
Interview with Handan Çağlayan (personal communication, 2020).
286
professionals – are more likely to gain prominent positions in the Kurdish
women’s movement and the legal parties. Middle-class cadres, due to their
individual class position, are more likely to be politicised by forms of
oppressions that they personally experienced, such as patriarchy and anti-
Kurdish racism, rather than economic exploitation.

Conclusion

This chapter discussed the political economy of the wide Kurdish movement
that follows Abdullah Öcalan’s new ideological paradigm. After Öcalan’s
capture in 1999, the PKK evolved into a complex confederation that goes
way beyond its military dimension and completed a long process of
ideological transformation that saw the organisation abandoning both
Kurdish nationalism and Marxism and replace them with a post-nationalist
and radical democratic project. This transformation can only be properly
explained if located in its historical context. The 1990s and 2000s were
decades of deep crisis for the political left when socialism and class politics
were erased by the political debate. Crushed by the fall of the Soviet Union
and the defeat of labour movements worldwide, the left developed along
fragmented lines and around environmentalist struggles, the anarchist-led
anti-globalisation movement, and identity politics. All these influences can
be spotted in Öcalan’s prison writings and the PKK/KCK’s new paradigm.
The ideological shift was accompanied and mutually reinforced by the great
expansion of the movement’s class basis especially towards the urban
middle classes.

In the new paradigm of the Kurdish movement in Turkey and Syria, the
critique of capitalism is based on the idealistic assumptions of Murray
Bookchin and his critique of the city and the industrial civilisation. Despite
the lack of discussion about the political economy of the Kurdish regions,
this line of critique resonates among the popular basis of the Kurdish
movement. The call for an ecologically sustainable society and the critique
of the city and the industrial society are extremely relevant to a great
number of supporters in Turkey who were deported from their villages to
287
city slums in the 1990s and who often had to accept exploitative conditions
in Turkey’s most labour-intensive industries. In this context, the most
significant expression of the popular basis of the new Kurdish movement is
the burgeoning cooperative movement for its particular attention towards
agriculture and, more generally, for its push towards collective ownership.

Even though this dimension directly reflects the interests and demands of
the forcibly-urbanised peasantry that supports the PKK/KCK, the
movement’s programme places little or no emphasis on social conflicts and
class hierarchies within Kurdish society. This aspect is particularly
significant because the most powerful element of the PKK’s rise in the 1980s
was its capacity to combine the struggle for the national liberation of the
Kurds with a frontal attack against the Kurdish landowning class that was
presented as both exploitative and supportive of Turkish colonialism. The
erasure of a critique of the class structure of the Kurdish region from the
new Kurdish movement’s discourse inevitably conceals any other social
conflict. Paradoxically, the ‘old’ nationalist PKK (up to the 1990s) – by
acknowledging and intellectually elaborating upon conflictual dynamics
within Kurdish society – was more capable to problematise Kurdish national
identity than the current post-nationalist PKK.

The contradictions of the new Kurdish movement are the inevitable

manifestations of its great expansion from the self-proclaimed revolutionary
vanguard that started the PKK insurgency in 1984 into a wide and plural
movement involving several million people. The transformation of this
movement was, as this and the previous two chapters showed, also the
result of its evolution from a peasant-based and led insurgency into a pro-
democracy multi-class coalition.

288
Chapter 10
Conclusion

Introduction

The concluding chapters of each of the two case studies (Chapters 6 and
9) present a picture of Kurdish nationalism in the 2010s. As Map 5 illustrates,
the fragmentation and weakening of Iraq and Syria as well as Turkey’s
authoritarian turn and regional ambitions significantly accelerated the pace
of Kurdish politics and assigned Kurdish forces an unprecedented centrality
in Middle East regional dynamics. For the first time in modern history, two
Kurdish-controlled political entities exist in Iraq and Syria, and pro-Kurdish
forces have been in power in almost one hundred municipalities in south-
eastern Turkey. Yet, the widespread revival of Kurdish national feelings in
all of the Kurdish regions did not translate into a unity of intent or action. On
the contrary, Kurdish actors are pursuing radically different social and
political projects and their relationship is conflictual to the point that they
often side with ‘foreign’ powers against each other. The ideological
differences between the Kurdish movements in Iraq – conservative and
ethnonationalist – and in Turkey and Syria – radical and post-nationalist –
seem to be far greater than the feelings of solidarity among Kurdish groups.
These considerations were the starting point for the formulation of a number
of empirical and theoretical research questions: What determines the
political divides within a national movement? What is the origin of conflicts
among Kurdish nationalists and why are they so resilient? What is the origin
of the alternative and competing nation-building project they pursue? What
is nationalism if it can be associated with the most diverse set of ideological
propositions? What determines the political content of nationalist
movements? The thesis covers over a century of Kurdish history and
identifies the origins of the present divide in class politics. The politically

289
significant and successful Kurdish nationalist organisations that emerged in
this long period were those that were capable of giving voice and
incorporating the interests, demands, and values of broad sections of
Kurdish society. Descending from abstract theorising to empirical
investigation reveals the historical manifestations of nationalism as
expressions of the struggle for state power between conflicting social
actors. Far from being an autonomous force, Kurdish nationalism developed
in different forms all embedded in social conflicts and each expressing the
bid for power of Kurdish historical classes and class coalitions. These
findings confirm the critique of nationalism theory formulated in Chapter 2,
showing that the ideological content – such as social values, or ideas
regarding the form of state and economic organisation – of a nationalist
movement cannot be autonomous from its class basis.

While the thesis reconstructs these developments in Iraq and Turkey in two
parts of three chapters each, this concluding chapter highlights the
elements of comparison across the two countries. First, this chapter runs
through the comparison chronologically, outlining how the integration into
the nation-states of Turkey and Iraq, characterised by different political and
class structures and geopolitical location, contributed to the diversification
of the Kurdish national movement allowing different class actors to emerge.
Second, these developments are analysed class-by-class, reconnecting
Kurdish nationalism to the theoretical discussion introduced in Chapter 2
and comparing the political posture and form of nationalism developed by
the same classes in each context. Finally, the chapter ends with concluding
remarks assessing the contribution to the field and the generalisability of
the findings, as well as identifying potential new avenues for research.

Comparing Kurdish Nationalism in Iraq and Turkey

This section summarises the findings of this thesis highlighting the elements
of comparison that emerged by the discussion of the two cases of Iraq
(Chapters 4-6) and Turkey (Chapters 7-9). It shows how a focus on class

290
politics significantly improves our understanding of certain crucial moments
of the history of Kurdish nationalism.

In both Turkey and Iraq, the first significant expression of Kurdish

nationalism occurred in the interwar period, when Ottoman Kurdistan was
divided into the newly-established states of Turkey, Iraq, and Syria. The
Kurdish revolts of the interwar period had a number of shared
characteristics that allowed us to treat them as part of a single phenomenon
defined as feudal nationalism. They largely involved the Kurdish tribal
aristocracy, the aghas, and were led by shaykhs whose religious role gave
them a degree of inter-tribal authority. The Kurdish tribal elite revolted
against the intrusiveness – enforcement of taxes and territorial control – of
the newly-established states of Turkey and Iraq and to preserve their
traditional power. The programme and discourse of these revolts mixed
Kurdish nationalist themes – borrowed by the early Kurdist organisations –
with religious and traditionalist demands. As Chapter 3 discussed in detail,
these revolts followed a traditional pattern of state-tribe relationship: tribal
agitations were typical of periods of transition when the tribal periphery
could seek to take advantage of instability at the centre to re-negotiate its
local power vis-à-vis the state. This was a pattern that the old decentralised
empires were more willing to accommodate than the newly-established
nation-states. Despite these shared characteristics, the size and
significance of these revolts in Turkey and Iraq were not even. The
modernising and secular zeal of the Kemalist government in Turkey
triggered both larger tribal revolts and a much more violent state-led
repression. On the contrary, the conservative Iraqi monarchy and its British
colonial patron were more accommodating towards tribal power and the
Kurdish revolts in Iraq were militarily less significant and more easily
forgiven. The different approach of the respective central states was
decisive in shaping the diversification of the trajectories of Kurdish
nationalism in the two countries.

These different approaches to the early feudal revolts were not simply the
result of the ideological positions of the two central governments but also

291
of the different location of Turkey and Iraq in the international system and
world economy. After resisting colonial encroachment in the early 1920s,
Turkey became a fully independent state and was welcomed into the US-
led Western block of the Cold War as a full member of NATO in 1952 and a
bulwark against communism. This geopolitical positioning allowed Ankara to
initiate a state-led programme of development, even though in a rather
uneven way, that industrialised western Turkey and left the predominantly
Kurdish east underdeveloped as a subordinate source of raw material and
cheap labour. After the harsh repression of the Kurdish revolts of the
interwar period, the Kurdish traditional elite had to abjure Kurdish identity
and, in the 1950s, was allowed to re-take its economic and social position,
entrusted with the task of maintaining order in the Kurdish countryside and
of providing large packages of votes to the successive conservative
coalitions in power in Ankara. This ‘state-agha’ alliance contributed to the
increasing impoverishment of the peasantry. Unlike Turkey, Iraq was
created as a British colony and remained substantially so until 1958. The
Kurdish tribal elite and their Arab counterpart highly benefitted from the
openly pro-tribal British policy that turned tribal leaders into a class of large
landowners who controlled the domestic politics of Iraq and provided an
indispensable social basis for the British-imposed Hashemite monarchy.
The land-grabbing and exploitative practices of the landowning elite
generated, in the 1940s and 1950s, a wave of rural struggles that affected
the entire country but that were particularly intense in the Kurdish north.
The power of the organised peasantry was decisive for the success of the
republican and anti-colonial Iraqi revolution of 1958. This differentiation in
the class structures that developed in the two Kurdish regions is key to
understanding the different trajectories taken by Kurdish nationalism in Iraq
and Turkey when it re-emerged in the early 1960s.

In Iraq, Kurdish nationalism survived in the KDP that, established in 1946,

was led by leftist professionals and intellectuals but was a marginal force in
a context dominated by landlord-peasant conflicts. The Kurdish revolt of
1961 started as the spontaneous uprising of tribal landowners against the

292
plan by the post-revolutionary government to redistribute and tax land as
well as a reaction against the general empowerment of the peasantry. The
revolting landowners legitimised their actions in nationalist terms, and the
progressive KDP, after its initial reluctance, was forced to join the revolt in
a position of subordination due to the far superior military means of the tribal
elite and to accept the leadership of the tribal Barzani family. While providing
ideological cover to the revolt, the KDP remained militarily marginal until the
contradiction between its socialist ideology and the class interests that led
the revolt exploded and the leftist leadership of the party was expelled in
1966. The relations of power between the tribal and the urban components
of Iraqi-Kurdish nationalism only changed in the following decade due to the
rapid process of urbanisation and to the establishment of the PUK. The KDP
and PUK fought a parallel war against Baghdad in the 1980s until they joined
forces and took control of the Kurdish region in 1991 after the withdrawal of
the Iraqi army. In a conflict led by different social classes, the demands of
the Iraqi Kurdish peasantry disappeared from the political landscape.

In Turkey, the absence of a revolutionary break allowed for the consolidation

of the state-agha alliance. Kurdish nationalism re-emerged in the 1960s in a
middle-class and radical form – the Eastern Meetings movement – that can
be compared to the urban strand of Iraqi-Kurdish nationalism but that was
virtually swept away by the 1971 military coup. It was only in the 1970s that
a new form of Kurdish nationalism branched out of the Turkish New Left. By
the mid-1970s, a generation of Kurdish intellectuals of peasant and working-
class extraction broke with the Turkish left – perceived as too compromised
with the Kemalist tradition – and sought to replicate the experience of Third-
World liberation struggles against what they identified as Turkish
colonialism. The PKK emerged as the most successful of these new groups
by pursuing a strategy that targeted exploitative Kurdish landowners even
before turning against Turkish security forces. By playing on the existing
rural conflicts, the PKK won, by the mid-1980s, large sectors of the Kurdish
peasantry and working classes to the point of constituting an
unprecedented military challenge to the Turkish state.

293
To counter the respective Kurdish uprisings, the governments of Iraq and
Turkey deployed similar and highly consequential counter-insurgency
policies. Both governments made wide use of loyalist tribal chiefs in setting
up Kurdish anti-guerrilla battalions. These irregulars were semi-private
armies that, besides their immediate military function, greatly enriched and
strengthened the social power of loyalist tribal leaders. Even more
significantly, the Iraqi and Turkish armies – respectively in the late 1980s
and early 1990s – engaged in the systematic cleansing of the Kurdish
countryside with the purpose of destroying the sources of supply and
recruitment of the insurgencies. Several million Kurdish peasants were
forced to leave their villages while their return was prevented by the
systematic destruction of their sources of subsistence – houses, cattle,
crops, woods, and wells. The virtual destruction of the peasantry
permanently changed the class structure of both Kurdish regions but also
their demographic and ecological outlook, depopulating the countryside and
creating a vast class of urban poor. Kurdish towns doubled or tripled their
population, mixed cities became predominantly Kurdish, and large Kurdish
communities appeared in non-Kurdish cities. In Turkey, displaced Kurdish
peasants constituted an immense ‘reserve army of labour’ that fed the
industrialisation of the 2000s largely based on labour-intensive production.
In Iraq, they constituted the dependent class of the burgeoning rentier
system.

These deep structural transformations had different political consequences

in the Kurdish regions of Iraq and Turkey. In Iraq, the US-led interventions
of 1991 and 2003 allowed first for the establishment of a Kurdish regional
administration – the KRG – controlled by the KDP and PUK and then for its
official recognition within post-Saddam Hussein Iraq. Due to the absence of
industry and the destruction of agriculture, the region survived on several
sources of rent, namely the control of smuggling routes and the
management of international aid in the 1990s, and the export of oil since the
mid-2000s. Locked in a competition that even led to a period of civil war
(1994-1998), the KDP and PUK strove to co-opt former regime collaborators

294
to expand their own constituencies. This process led to the creation of a
new ruling class made up of two components. First, the PUK and KDP
commanders of the liberation war turned politicians whose power came
from their control of the party militias regardless of their class background.
Second, the part of the Kurdish ruling class that had remained loyal to
Baghdad and fought against the Kurdish insurgency and that, with the
transformation of the Iraqi economy in the 1970s and 1980s, had turned
from landowners into regime-sponsored entrepreneurs. This new ruling
class used its military control of the region to appropriate the streams of
rent and to impose, since 1998, a KDP-PUK political duopoly. The vast
majority of the Kurdish population was forced into a position of dependence
on the patronage network of the two parties which dispensed public
employment, most often the only source of income available. In Turkey,
where Kurdish forces never gained control of the region, the destruction of
the PKK’s rural basis of support contributed to the gradual ‘urbanisation’ of
the Kurdish movement. In the 1990s, displaced peasants turned urban poor
became a political force that could be easily mobilised in support of the PKK
struggle. On the one hand, Kurdish poor and unemployed youth took over
towns and neighbourhoods in many occasions and for days. On the other
hand, displaced peasants socialised in the nationalist and socialist discourse
of the PKK became the electoral backbone of the rising pro-Kurdish parties
that – even if only partly associated with the PKK – took control of many
Kurdish municipalities.

In the late 2000s and early 2010s, the trajectories of Kurdish politics in Iraq
and Turkey diversified even further but also increased their degree of
interaction in terms of regional rivalry over a more and more interconnected
transnational Kurdish space. In the KRG, the stabilisation of an authoritarian
form of politics and oligarchic control over the economy led to the growth
of popular and parliamentary opposition. In the mid-2010s, the collapse of
the oil price brought the KRG rentier system to its knees and triggered
sustained popular protests. The response of the Iraqi-Kurdish elite was, on
the one hand, to violently crack down on dissent, and, on the other hand, to

295
invest in an increasingly aggressive nationalist rhetoric to regain political
legitimacy with the result of exacerbating tensions with Baghdad but also
with the neighbouring Turkey and Iran. In Turkey, the growing strength of
the PKK in the urban contexts and control by pro-Kurdish parties of the local
administrations contributed to a great expansion of the social basis of the
Kurdish movement, especially among the urban middle classes. Parallel to
this process – and mutually reinforcing – the PKK adopted a radical-
democratic and post-nationalist political project and increased its presence
in the other Kurdish regions. In the context of the Syrian Civil War, local
Kurdish forces linked to the PKK took over the Kurdish region of Syria and
established the Rojava autonomous administration that attempted to
implement the new PKK ideology on a regional scale. The war on ISIL (2014-
2017) was fought on both the Iraqi-Kurdish and Syrian-Kurdish fronts and
saw the involvements of Kurdish groups from all four regions, strengthening
popular feelings of pan-Kurdish solidarity. However, it also contributed to
an increasingly virulent rivalry between the two competing national projects
for the Kurds expressed by the Iraqi-Kurdish leadership and the PKK-Rojava
movement.

This thesis reconstructed the social origins of these projects highlighting

the role of class conflict and alliance building in the history of Kurdish
nationalism in Iraq and Turkey. The next section compares the nationalist
projects developed by the same classes in the two different contexts
showing common patterns but also the context-based differentiation.

Social Classes and Kurdish Nationalism

After building up the theoretical framework of this thesis, Chapter 2 ends by

outlining historical patterns and tendencies for the approach to nationalism
taken by different classes. Acknowledging the importance of the context-
specific genesis of each nationalist movement means ultimately leaving the
study of its class basis to the realm of the empirical investigation. Yet, social
classes do not develop infinitely variable forms of politics. Historical
patterns exist and are rooted in objective material interests that
296
characterise social classes as such. The following sections present the
findings of this thesis by comparing the historical posture taken towards
nationalism by the main Kurdish social classes in Iraq and Turkey against
the background of the theoretical considerations outlined in Chapter 2 in
reference to the same social classes. The purpose of this section is to
highlight common patterns and especially contextual differences in the
posture of the same classes showing the value of comparative analysis, but
also how the study of these two Kurdish cases can tell us something of
theoretical relevance.

The end of Chapter 2 sketched the forms of nationalism that have been
historically developed by four social classes in terms. The following survey
neither starts with nor includes the classical form of ‘bourgeois’ nationalism
that played such a pivotal role in nineteenth century Europe. In Kurdistan, a
class interested primarily in unifying the national market, rationalising
societal institutions, and fully separating the political and economic spheres
with the aim of promoting private capital, simply did not exist. On the
contrary, this section covers the other three forms of nationalism sketched
in Chapter 2 – feudal, middle-class, and subaltern nationalisms – as the
historical expressions of the three ‘social protagonists’ of the past century
of Kurdish history: the tribal and landowning elite, the urban middle classes,
and the peasantry.

Feudal Nationalism

The subject of Chapter 3 was the first historical expression of Kurdish

nationalism developed by the traditional ruling class of Kurdish tribal
society. The power of aghas and shaykhs rested on traditional – religious,
tribal, dynastic – sources of legitimacy that were no longer recognised by
the states that succeeded the Ottoman Empire. In the 1920s and 1930s, the
sudden adoption of Kurdish nationalist discourse by members of this class
and the outbreak of the Kurdish revolts were a response to the threat posed
by the centralising tendencies of the new states to the power traditionally
held by this class. All of the revolts that took place in Kurdistan in those
297
decades were harshly crushed by the newly-established Turkey and Iraq. In
different ways – depending on specific paths of state formation – the
traditional Kurdish elite largely renounced Kurdish nationalism and found a
comfortable place in the power structure of both countries. In Iraq, as an
integral part of the country’s ruling class created by British colonialism. In
Turkey, as a dependent elite entrusted by the central state with the
maintenance of order and social hierarchies in the Kurdish periphery.

In socio-economic terms, this class had been going through a long process
of transformation and, by the mid-twentieth century, its power rested on
large estates based on what Barrington Moore calls ‘labour-repressive
agriculture’.679 Rather than increasing productivity by investing in
improvements and technology, this class tended to secure higher surplus
through the hyper-exploitation of the peasantry and to rely on extra-
economic coercive means to prevent peasant organisation. In both Iraqi and
Turkish Kurdistan in the 1950s, the aghas, rather than employing villagers,
owned villages. In Moore’s account on the social origins of democracy,
labour-repressive landowners are the most strenuous opponents of political
democracy as the emancipation of the peasantry would inevitably lead to
the collapse of the form of appropriation of surplus they rely on. Their
approach to Kurdish nationalism was driven by similar concerns. Since its
incorporation into the new state, the Kurdish landowning class of Iraq
‘rediscovered’ Kurdish nationalism only when the post-revolutionary
government allied with the organised peasantry and directly threatened
their interests. The dominance of this class determined the historically
conservative character of Kurdish nationalism in Iraq, marginalising
progressive middle-class nationalism until the late 1970s, and blocking the
autonomous political organisation of the subaltern classes. In Turkey, in the
absence of a revolutionary break, the ‘state-agha’ alliance was confirmed
by all governments representing the conservative blocs that ruled Turkey
since the 1950s, including the Islamist government that came to power in
2002. This is why the ruling class of Turkish Kurdistan – unlike its Iraqi-

679
Moore, Social Origins.
298
Kurdish counterpart – largely opposed Kurdish nationalism and also why the
only significant expressions of Kurdish nationalism in Turkey developed out
of the political left.

It is important to notice that in neither of the cases the Kurdish ruling class
acted as a united class actor. The central states were always – except for
Turkey in the Kemalist period – willing to provide strong incentives for the
members of the traditional elite to remain loyal. In Iraq, many tribes remained
loyal to Baghdad throughout the conflict providing the government with the
counter-insurgency irregular forces, the jash, as Kurdish nationalists called
them. This position was often driven by historical tribal rivalries but also by
the fact that the anti-landlord zeal of the 1958 revolution was soon forgotten
and Baghdad generously rewarded tribal loyalism. In Turkey, where most of
the tribal elite sided with the government, there were always ‘patriotic
aghas’ who let the Kurdish guerrillas operate in their territories or even
supported them out of nationalist sympathy – though more likely with the
intent of weakening a neighbouring loyalist tribe or reducing state control
over their own land. Prestigious tribal families also hedged their bets by
keeping members on both sides. In this regard, I explored the cases of the
Barzanji in Iraq and the Bucak in Turkey. Mahmud Barzanji, a powerful
shaykh in Sulaymaniyah province, led the feudal nationalist revolt that
intermittently lasted from 1919 to 1932. His son Shaykh Latif became the
vice-chairman of the KDP while being at the same time the target of a major
peasant revolt in 1947. His brother Baba Ali was a loyalist politician and ‘a
man for all seasons’, appointed minister first under the King, then under Abd
al-Karim Qasim, and once again under the Ba’athists. In Turkey, the powerful
Bucak family from Siverek represented a similar trend. Mustafa Tevfik Bucak
was a conservative DP parliamentarian in the 1950s while his brother Faik
was the founder of the ‘feudal nationalist’ PDKT. This could be the result of
a strategic choice or conflicting individual choices, yet in both cases, it
signalled that the tribal landowning elite was confident to keep their social
position regardless of the forces in power and the historical record proves
that they most often did.

299
 Middle-Class Nationalism

Kurdish nationalism was never exclusively the expression of the Kurdish

feudal class. There was always a significant middle-class component and,
in the course of the twentieth century, several forms of middle-class
nationalism developed among the Kurds. The first politicised form of Kurdish
national identity developed in the early twentieth century among Ottoman
bureaucrats and army officers. As Chapter 3 explained, even if the Kurds
working for the Ottoman Empire often hailed from the Kurdish feudal class,
they had long rescinded their ties to the Kurdish provinces and were part of
the Ottoman state class. The first Kurdist organisations emphasised the
cultural specificity of the Kurds, demanded decentralisation and
administrative autonomy for the Kurdish provinces and were characterised
by a modernist and developmentalist ideology. When the Ottoman Empire
collapsed, these demands were re-articulated in terms of national
independence. In both cases, these demands reflected the ambition of
becoming the ruling elite of a Kurdish autonomous entity within the empire
or a Kurdish independent state. In the new nation-states of Iraq and Turkey,
the principle of ethnic exclusivity made it hard for the Kurds to attain public
offices and middle-class nationalism directly reflected the ambition to
replace Arabs and Turks – or Arabised and Turkified Kurds – in the public
offices. From the 1940s, Kurdish nationalist activism and organising were
largely expressed by professionals from the major towns. By the same
token, the re-birth of Kurdish nationalism in Turkey in the 1960s was a
largely urban and middle-class phenomenon.

Middle-class nationalism in the Kurdish case suffered from a structural

weakness due to the predominantly rural character of Kurdish society.
Kurdish towns were, up to the 1970s, small and largely dependent on their
rural surroundings as agricultural marketplaces. Even if intellectually prolific
and ideologically sophisticated, urban-based nationalists proved to be
utterly unable to challenge the state by their own means or to lead wider
social coalitions. Thus, they most often had to follow the lead of different
300
class actors. In the feudal revolts of the interwar period as well as the
Barzani Revolution in Iraq (1961-1975), urban and progressive nationalists
had to surrender the leadership of the national struggle to the traditional
tribal elite due to the latter’s superior military means and, in both contexts,
they provided nationalist legitimacy to the reactionary demands of the
traditional landowning class. In other cases, middle-class nationalists allied
from a subordinate position with mass-based movements. This was the
case in Iraq in the 1940s and 1950s when landlord-peasant struggles
dominated the political landscape of the Kurdish region and the progressive
leadership of the KDP worked in coordination with the stronger Iraqi
Communist Party. Even more significantly, this was the case in Turkey in the
1980s where the decimated Kurdish leftist groups of middle-class
extractions gradually lent their support to a PKK-led insurgency centred on
the mobilised peasantry.

In the 1970s, the rapid process of urbanisation changed the relative

importance of the towns vis-à-vis the rural areas and allowed for a degree
of autonomous action for middle-class nationalists. In Turkey, this process
manifested in the proliferation of Kurdish leftist organisations and an
intensification of the political struggle. However, the 1980 Turkish military
coup dealt a mortal blow to these groups. In Iraq, middle-class and
progressive nationalists gained their autonomy of action only in 1975 with
the establishment of the PUK. If observed from a purely class-based
perspective, the urban middle-class base of the PUK reached its objective
and, since 1991, has constituted an integral part of the ruling class of the
Kurdish region, or at least of the PUK-controlled part of the region. It must
be noted, however, that the power of the PUK rested, since 1991, on a
ruling-class pact with the KDP and the traditional elite that cost Talabani’s
party its ideological peculiarity and made it barely distinguishable from its
historical rival. In Turkey, the growth of pro-Kurdish parties, especially at
the local level, created a space for the participation of the urban middle
classes which gained a prominent position in the non-military wings – civil
society and electoral politics – of the Kurdish movement.

301
 Nationalism and the Subaltern Classes

One of the objectives of this work was to paint a fuller picture of the history
of Kurdish nationalism that goes beyond intra-ruling class politics – the
struggle between competing elites – and account for the role and
participation of the subaltern classes. One of the consequences of treating
nationalism as an autonomous force disconnected from social struggle is
that subaltern groups tend to fade away and be relegated to the background
of history. In the Kurdish context, the rural character of the region combined
with the late development (in Turkey) or virtual absence (in Iraq) of
industrialisation meant that the peasantry constituted by far the largest
subaltern group, the class producing the surplus that sustained the
landowning elite but also the urban classes. Between the late nineteenth
and the first half of the twentieth century, the Kurdish countryside went
through a process of concentration of land ownership and proletarianization
of tribesmen initiated by the Ottoman government and continued by the
successor states, particularly colonial Iraq, a process that Lisa Anderson
would call ‘the creation of a peasantry’.680

Peasants were by no means passive victims of this process. On the contrary,

Iraq’s 1940s and 1950s were characterised by intense rural struggles and
peasants organising – in both Arab and Kurdish areas – that fuelled the
growing political role of the ICP and the trade unions. The peasant
movement, despite being even stronger in the Kurdish region, did not
develop a Kurdish national project and is thus absent in scholarly accounts
of Kurdish nationalism. However, the Kurdish peasantry, along with the Arab
peasantry and the small but increasingly militant industrial working class,
formulated a powerful national project for Iraq condensed in the ICP’s
programme for a multi-national and socialist nation. Rather than
sympathising with their fellow Kurdish landowners, the Kurdish peasantry

Lisa Anderson, The State and Social Transformation in Tunisia and Libya, 1830-
680

1980 (Princeton: Princeton University Press, 1986), 28-29.

302
had a clear understanding of its commonality of interests with the Iraqi
workers’ movement beyond the linguistic divide. Iraq’s subaltern classes did
not only show their political agency through their autonomous action. They
also highlighted the indissoluble link between British colonialism and the
landowning class forcing reluctant Iraqi nationalists to acknowledge the
importance of the struggles at the point of production. It was not a
coincidence that, after more than a decade of struggles, the successful
revolution of 1958 occurred only after the inclusion of the ICP in the National
Front in 1956. The revolution led to a few years of unprecedented
empowerment of the peasantry in which their autonomous organising was
no longer violently obstructed by state authorities. It is in this pivotal period
that Chapter 4 identified the origins of the Kurdish revolt of 1961 in the
reaction of Kurdish landowners against the emancipating peasantry and a
state that was no longer on their side. In the long conflict that followed
(1961-1975), the Kurdish peasantry lost its political agency crushed
between their ancient exploiters and the scorched-earth strategy deployed
by an increasingly authoritarian and chauvinistic Iraqi government.
Moreover, in the course of the 1970s and 1980s, the peasantry lost its
centrality in the Iraqi economy as the increasing significance of oil exports
made agriculture a rather marginal sector. In the 1980s, the Iraqi-Kurdish
peasantry was virtually destroyed as a social class by the genocidal
counter-insurgency strategy deployed by Baghdad and aimed at
depopulating the Kurdish countryside. Dispossessed peasants became a
class of urban poor dependent on public handouts distributed by the
Kurdish forces that came to rule the region in 1991. As Chapter 5 and 6
emphasised, this position of subordination into the burgeoning Kurdish
rentier system was not passively accepted. Since the mid-2000s, several
waves of protests hit the Kurdish region of Iraq combining demands for
political freedom and more equitable distribution of the rent.

Contrary to Iraq, Turkey never experienced a countrywide peasant

movement due to a different rural structure dominated, especially in the
non-Kurdish areas, by small independent farmers. As Chapter 7 showed, in

303
Turkey’s Kurdish region, a much less even distribution of land and the
predatory practices of the landowning elite, led to an increase of rural
conflicts in the 1960s and 1970s, especially since the mechanisation of
agriculture significantly worsened the conditions of small farmers and
landless peasants. In the course of the 1970s, leftist Kurdish groups had
some degree of success at mobilising the popular classes in the Kurdish
towns but were unable to link their action to that of the rural masses. Only
the PKK, whose leadership largely hailed from the peasantry, deployed a
rural strategy based on military actions against the Kurdish landowning elite
and started building a support base among the peasantry. After the 1980
military coup eliminated most of its competition, the PKK emerged as the
dominant Kurdish group in Turkey. In the mid-1980s, the PKK was
successful at incorporating anti-landlord struggles in a project of national
liberation, creating the first Kurdish nationalist movement largely based on
the mobilisation of the peasantry. The extent of this support is suggested
by the overwhelming peasant background of the militants recruited by the
PKK in these years but, even more significantly, by the counter-insurgency
strategy deployed by Ankara. In the early 1990s, Turkey’s response to the
increasing strength of the insurgency was to forcibly evacuate thousands
of villages to destroy the base of supply and recruitment of the guerrillas.
Displaced peasants formed a vast class of urban poor and many Kurdish
towns doubled and tripled their population. The near-destruction of Kurdish
rural society had deep consequences on the development of the uprising
and Kurdish nationalism in Turkey. The urbanisation of hundreds of
thousands of peasants often already politicised by the PKK transferred the
centre of the Kurdish struggle to the urban areas and, in just a few years,
allowed pro-Kurdish forces – directly or indirectly linked to the PKK – to win
a large part of the local administrations of the Kurdish region. These
transformations led to a gradual expansion of the PKK’s base of support
among the urban working and middle classes, a process that had already
started between the late 1980s and early 1990s in the wake of the PKK’s
military successes. This class differentiation of the PKK led to a process of
deep ideological transformation from Marxism-Leninism to radical
304
democracy with less and less emphasis on class struggle. The PKK-led
Kurdish movement became much wider and, even though peasants and
urban poor kept constituting the strongest base of guerrilla recruitment, the
middle-class component gained a prominent role in the electoral and civil-
society organisations. As Chapter 9 showed, even if the movement
maintained a high degree of internal cohesion, tensions emerged at pivotal
moments between the part of the movement oriented towards electoral
politics and coalition building and the unemployed or under-employed urban
youth hailing from the displaced peasantry and more oriented towards
radical action.

Concluding Remarks

For reasons explained earlier, the story of Rojava, the Kurdish-controlled

region in north-eastern Syria established in 2012, is only briefly told in
relations to the developments of the Kurdish movement in Turkey. This
approach, imposed by the ‘economy’ of this thesis, runs the risk of
reproducing the idea that the Kurds of Syria have only recently come “out
of nowhere”681 and that their politics is a mere extension of that of the Kurds
of Turkey. In the same way, the rich history of the Kurds of Iran did not have
the space it deserves and only entered the story when it interweaved with
that of the Iraqi Kurds. However, Kurdish politics in Syria and Iran always
was and still is highly contested. A brief survey on the – much more limited
– literature, as well as the interviews conducted for this thesis with Syrian
and Iranian Kurdish activists, gave me the strong impression that Kurdish
politics in Syria and Iran was traversed by tensions similar to those that
shaped it in Iraq and Turkey and that conflicts and alliances between the
peasantry, the traditional landowning elite, and the middle classes were
decisive to the development of competing forms of Kurdish nationalism.
Primary research can provide the data we need to study the evolution of the

681
The title of Michael Gunter’s book on Syrian Kurdish history. Michael Gunter, Out
of Nowhere: The Kurds of Syria in Peace and War (London: Hurst & Company,
2014).
305
class basis of the Kurdish movement in Iran and Syria and particularly on the
extent of class conflicts that inevitably occur within trans-class movements.
The relationship between class politics and Kurdish nationalism in Iran and
Syria are thus fruitful avenues for future research.

However, this thesis aimed at providing a framework to study nationalist

movements beyond the Kurdish experience. Among the original motivations
for choosing this topic, my initial ‘empirical curiosity’ for the diverse
expressions of Kurdish nationalism and their historical roots were always
accompanied by a deep dissatisfaction for the way nationalist politics is
dismissed, especially in liberal academic environments, as the
unsophisticated outburst of an anachronistic identity. Violent and racist
nationalist projects have resurged in recent times in countries as far as the
United States and Turkey, Brazil and Israel, Italy and India. As these
movements threaten democratic institutions and collective values of
solidarity and peace, they cannot be rejected only on the ground of the
moral superiority of liberal multiculturalism or the post-political ideology of
technocracy and expertise. To counter the resurgence of nationalism, we
must investigate the social conflicts in which they are embedded, to discern
the legitimate grievances they represent from the violence they voice.
Distinguishing the anger generated by an unjust economic system from the
mere defence of entrenched racial and social privilege is the only way to
build an inclusive and just society and save the humanistic values of
freedom and solidarity.

306
List of References

Books, Journal Articles and Edited Volumes

Ahmad, Aijaz, ‘Class, Nation, and State: Intermediate Classes in Peripheral

Societies’ in Middle Classes in Dependent Countries, ed. by Dale L.
Johnson (Beverly Hills and London: Sage Publications, 1985)

Ahmed, Ibrahim, ‘The Republic of Kurdistan: A Personal Memoir’, The

International Journal of Kurdish Studies, 11.1/2 (1997), 9–32

Akkaya, Ahmet, and Joost Jongerden, ‘Democratic Confederalism as a

Kurdish Spring: The PKK and the Quest for Radical Democracy’, in The
Kurdish Spring: Geopolitical Changes and the Kurds, ed. by Michael
Gunter and Mohammed Ahmed (Costa Mesa, CA: Mazda Publishers,
2013), pp. 163–86

Akkaya, Ahmet, and Joost Jongerden, ‘Reassembling the Political: The PKK
and the Project of Radical Democracy’, European Journal of Turkish
Studies, 14 (2012)

Aktan, Resat, ‘Problems of Land Reform in Turkey’, International Journal of

Middle Eastern Studies, 20.3 (1966), 317–34

Aladağ, Cemal, Milli Mesele ve Kürdistan’da Feodalite Aşiret (Frankfurt:

Komkar Yayınları, 1981)

Al-Ali, Nadje, and Latif Tas, ‘Reconsidering Nationalism and Feminism: The
Kurdish Political Movement in Turkey’, Nations and Nationalism, 24.2
(2018), 453–73

Altunışık, Meliha. B., and Özlem Tür, ‘From Distant Neighbors to Partners?
Changing Syrian-Turkish Relations’, Security Dialogue, 37.2 (2006),
229–48

Ammann, Birgit, ‘Kurds in Germany’, in Encyclopedia of Diasporas, ed. by

Melvin Ember, Ian A. Skoggard (Boston, MA: Springer, 2005), pp. 1011–
307
 19

Anaid, Anwar, and Emel Elif Tugdar, eds., Iraqi Kurdistan’s Statehood
Aspirations: A Political Economy Approach (London: Palgrave
Macmillan, 2019)

Anderson, Benedict, Imagined Communities: Reflecting on the Origin and

Spread of Nationalism (London: Verso, 1983)

Anderson, Lisa, The State and Social Transformation in Tunisia and Libya,
1830-1980 (Princeton: Princeton University Press, 1986)

Anderson, Perry, Lineages of the Absolutist State (London: Verso, 1974)

Arat, Yeşim, and Sevket Pamuk, Turkey between Democracy and

Authoritarianism (Cambridge: Cambridge University Press, 2019)

Aslan, Senem, ‘Incoherent State: The Controversy over Kurdish Naming in

Turkey’, European Journal of Turkish Studies, 10 (2009)

Atabaki, Touraj, and Erik J. Zürcher, Men of Order: Authoritarian

Modernization under Ataturk and Reza Shah (London and New York:
I.B. Tauris, 2004)

Ayata, Bilgin, ‘Kurdish Transnational Politics and Turkey’s Changing

Kurdish Policy: The Journey of Kurdish Broadcasting from Europe to
Turkey’, Journal of Contemporary European Studies, 19.4 (2011), 523–
33

Ayubi, Nizah N., Over-Stating the Arab State: Politics and Society in the
Middle East (London: I.B. Tauris, 1995)

Bache, Christina, ‘Mutual Economic Interdependence or Economic

Imbalance: Turkish Private Sector Presence in the Kurdistan Region of
Iraq’, Middle East Critique, 27.1 (2018), 61-75

Bal, İhsan, and Sedat Laçiner, Ethnic Terrorism in Turkey and the Case of
the PKK: Roots, Structure, Survival, and Ideology, (London: Frank Cass
2004)

Barkey, Henri, ‘The People’s Democracy Party (HADEP): The Travails of a

308
 Legal Kurdish Party in Turkey’, Journal of Muslim Minority Affairs, 18.1
(1998), 129–38

Barth, Fredrik, Principles of Social Organization in Southern Kurdistan

(Oslo: Universitet i Oslo, 1953);

Barzani, Mustafa, ‘Speech Presented to the Congress of the Kurdish Exiles

in the Soviet Union: Baku, January 19, 1948’, The International Journal
of Kurdish Studies, XI.1–2 (1997), 33–48.

Batatu, Hanna, The Old Social Classes and the Revolutionary Movements
of Iraq: A Study of Iraq’s Old Landed and Commercial Classes and of
Its Communists, Baʻthist, and Free Officers (Princeton: Princeton
University Press, 1978)

Beinin, Joel, Workers and Peasants in the Modern Middle East (Cambridge:
Cambridge University Press, 2001)

Bengio, Ofra, The Kurds of Iraq: Building a State Within a State (London:
Lynne Rienner Publishers, 2012)

Berger, Stefan, Mark Donovan, and Kevin Passmore, eds., Writing National
Histories: Western Europe Since 1800 (London and New York:
Routledge, 1999)

Berger, Stefan, and Chris Lorenz, The Contested Nation: Ethnicity, Class,
Religion and Gender in National Histories (New York: Palgrave
Macmillan)

Beşikçi, İsmail, Doğu Anadolu’nun Düzeni: Sosyo-Ekonomik ve Etnik

Temeller (Ankara: Yurt Kitap-Yayın, 1992)

Beşikçi, İsmail, International Colony Kurdistan (London: Gomidas Institute,

2015)

Birand, Mehmet Ali, APO ve PKK (Istanbul: Milliyet Yayinlari, 1992)

Blaut, James M., The National Question: Decolonizing the Theory of

Nationalism (London and New Jersey: Zed Books, 1987)

Bookchin, Murray, The Ecology of Freedom: The Emergence and

309
 Dissolution of Hierarchy (Palo Alto: Cheshire Books, 1982)

Bozarslan, Hamit, and Clémence Scalbert-Yücel, eds., Joyce Blau

l’éternelle Chez Les Kurdes (Istanbul: Institut Français d’études
anatoliennes, 2018)
<https://books.openedition.org/ifeagd/2199?lang=en>

Brass, Paul, Ethnicity and Nationalism: Theory and Comparison (London:

SAGE Publications, 1991)

Bromley, Simon, Rethinking Middle East Politics: State Formation and

Development (Cambridge: Polity Press, 1994)

Brubaker, Roger, Nationalism Reframed: Nationhood and the National

Question in the New Europe (Cambrdige: Cambridge University Press,
1996)

Cabral, Amilcar, Return to the Source: Selected Speeches (New York and
London, Monthly Review Press, 1973)

Çağlayan, Handan, ‘From Kawa the Blacksmith to Ishtar the Goddess:

Gender Constructions in Ideological-Political Discourses of the Kurdish
Movement in Post-1980 Turkey’, European Journal of Turkish Studies,
14 (2012)

Çağlayan, Handan, Women in the Kurdish Movement: Mothers, Comrades,

Goddesses (London: Palgrave Macmillan, 2020)

Candar, Cengiz, Mezopotamya Ekspresi: Bir Tarih Yolculuğu (Istanbul:

Iletisim, 2012)

Çandar, Cengiz, Turkey’s Mission Impossible: War and Peace with the
Kurds (Lanham: Lexington Books)

Casier, Marlies, ‘Designated Terrorists: The Kurdistan Workers’ Party and

Its Struggle to (Re)Gain Political Legitimacy’, Mediterranean Politics,
15.3 (2010), 393–413

Casier, Marlies, and Joost Jongerden, eds., Nationalisms and Politics in

Turkey: Political Islam, Kemalism and the Kurdish Issue (London and
New York: Routledge, 2011)
310
Casier, Marlies, and Olivier Grojean, ‘Between Integration, Autonomization
and Radicalization. Hamit Bozarslan on the Kurdish Movement and the
Turkish Left’, European Journal of Turkish Studies, 14 (2012)

Cetorelli, Valeria, Isaac Sasson, Nazar Shabila, and Gilbert Burnham,

‘Mortality and Kidnapping Estimates for the Yazidi Population in the
Area of Mount Sinjar, Iraq, in August 2014: A Retrospective Household
Survey’, PLoS Medicine, 14.5 (2017)
<https://doi.org/10.1371/journal.pmed.1002297>

Chaliand, Gerard, ed., A People Without a Country: The Kurds and

Kurdistan (London: Zed Books, 1993)

Charountaki, Marianna, The Kurds and US Foreign Policy: International

Relations in the Middle East since 1945 (London: Routledge, 2011)

Chaudhry, Kiren Aziz, ‘On the Way to Market: Economic Liberalization and
Iraq’s Invasion of Kuwait’, MERIP Middle East Report, 170 (1991), 14–23

Dahlum, Sirianne, Carl Henrik Knutsen, and Tore Wig, ‘Who Revolts?
Empirically Revisiting the Social Origins of Democracy’, Journal of
Politics, 81.4 (2019), 1494–99

Dann, Otto, and John Dinwiddy, eds., Nationalism in the Age of the French
Revolution (London and Ronceverte: The Hambledon Press, 1988)

Dawisha, Adeed, Iraq: A Political History from Independence to Occupation

(Princeton: Princeton University Press, 2009)

Dodge, Toby, Inventing Iraq: The Failure of Nation Building and a History
Denied (New York: Columbia University Press, 2003)

Dohrmann, Mark, and Robert Hatem, ‘The Impact of Hydro-Politics on the

Relations of Turkey, Iraq and Syria’, The Middle East Journal, 68.4
(2014), 567–83

Dorronsoro, Gilles, and Nicole F. Watts, ‘Toward Kurdish Distinctiveness in

Electoral Politics: The 1977 Local Elections in Dı̇yarbakir’, International
Journal of Middle East Studies, 41.3 (2009), 457–78

Eagleton, William, The Kurdish Republic of 1946 (London: Oxford

311
 University Press, 1963)

Egret, Eliza, and Tom Anderson, Struggles for Autonomy in Kurdistan &
Corporate Complicity in the Repression of Social Movements in Rojava
and Bakur (London: Corporate Watch Cooperative, 2016)
<https://corporatewatch.org/wp-content/uploads/2017/09/Struggles-
for-autonomy-in-Kurdistan.pdf>

Fanon, Frantz, The Wretched of the Earth (New York: Grove Press, 1991)

Fernea, Robert A., and William Roger Louis, eds., The Iraqi Revolution of
1958: The Old Social Classes Revisited (London: I.B. Tauris, 1991)

Fırat, Abdülmelik, Fırat Mahzun Akar (Istanbul: Avesta Kitabevi, 2006)

Franzén, Johan, ‘From ally to foe: The Iraqi communist party and the
Kurdish question, 1958-1975’, British Journal of Middle Eastern
Studies, 38.2 (2011), 169-185

Franzén, Johan, Red Star Over Iraq: Iraqi Communism Before Saddam
(New York: Columbia University Press, 2011)

Freeden, Michael, ‘Is Nationalism a Distinct Ideology?’, Political Studies,

46.4 (1998), 748–65

Gabbay, Rony, Communism and Agrarian Reform in Iraq (London: Croom

Helm, 1978)

Gellner, Ernest, Nations and Nationalism (Ithaca, NY: Cornell University

Press, 1983)

Ghareeb, Edmund A., and Beth Dougherty, Historical Dictionary of Iraq

(Lanham, MD: The Scarecrow Press, 2004)

Ghassemlou, Abdul Rahman, Kurdistan and the Kurds (London: Collets,

1965)

Gramsci, Antonio, Selections from the Prison Notebooks, ed. By Quintin

Hoare and Geoffrey Nowell Smith (New York: International Publishers,
1971)

Green, Marcus E., ‘Rethinking the Subaltern and the Question of

312
 Censorship in Gramsci's Prison Notebooks’, Postcolonial Studies, 14.4
(2011), 387-404

Grigoriadis, Ioannis N., and Esra Dilek, ‘Struggling for the Kurdish Vote:
Religion, Ethnicity and Victimhood in AKP and BDP/HDP Rally
Speeches’, Middle Eastern Studies, 54.2 (2018), 289–303

Güneş, Cengiz, and Welat Zeydanlıoğlu, eds., The Kurdish Question in

Turkey: New Perspective on Violence, Representation, and
Reconciliation (London: Routledge, 2014)

Günes, Cengiz, The Kurdisn National Movement in Turkey: From Protest to

Resistance (Milton Park, Abingdon, Oxon: Routledge, 2012)

Gunter, Michael, ed., Routledge Handbook on the Kurds (London and New
York: Routledge, 2019)

Gunter, Michael, Historical Dictionary of the Kurds (Lanham: Scarecrow

Press, 2011)

Gunter, Michael, Out of Nowhere: The Kurds of Syria in Peace and War
(London: Hurst & Company, 2014)

Gunter, Michael, The Kurdish Predicament in Iraq: A Political Analysis (New

York: St. Martin’s Press, 1999)

Gunter, Michael, The Kurds Ascending: The Evolving Solution to the

Kurdish Problem in Iraq and Turkey, (London: Palgrave Macmillan,
2008)

Gunter, Michael, The Kurds in Turkey: A Political Dilemma (Boulder:

Westview Press, 1990)

Gürbüz, Mustafa, Rival Kurdish Movements in Turkey: Transforming Ethnic

Conflict (Amsterdam: Amsterdam University Press, 2016)

Gürses, Mehmet, David Romano, and Michael Gunter, eds., The Kurds in
the Middle East: Enduring Problems and New Dynamics (Lanham:
Rowman & Littlefield, 2020)

Haj, Samira, The Making of Iraq, 1900-1963: Capital, Power, and Ideology

313
 (Albany: State University of New York Press, 1997)

Hakan Yavuz, M., ‘Five Stages of the Construction of Kurdish Nationalism

in Turkey’, Nationalism and Ethnic Politics, 7.3 (2001), 1–24

Hassanpour, Amir, ‘Satellite Footprints as National Borders: Med‐tv and the

Extraterritoriality of State Sovereignty’, Journal of Muslim Minority
Affairs, 18.1 (1998), 53–72

Hassanpour, Amir., Nationalism and Language in Kurdistan, 1918-1985

(Lewiston, NY: Mellen Research University Press, 1992)

Heper, Metin, and Ahmet Evin, eds., State, Democracy, and the Military:
Turkey in the 1980s (Berlin: De Gruyter, 1988)

Heper, Metin, The State and Kurds in Turkey: The Question of Assimilation
(New York: Palgrave Macmillan, 2007)

Hobsbawm, Eric J., Nations and Nationalism since 1780: Programme, Myth,
Reality (Cambridge: Cambridge University Press, 1991)

Issawi, Charles, The Economic History of Turkey, 1800-1914 (Chicago:

University of Chicago Press, 1980)

Izady, R. Mehrdad, The Kurds: A Concise Handbook (Washington: Taylor &

Francis, 1992)

Jabar, A. Faleh, and Renad Mansour, eds., The Kurds in a Changing Middle
East: History, Politics and Representation (London: I.B. Tauris, 2019)

Jabar, Faleh A., and Hosham Dawod, eds., The Kurds: Nationalism and
Politics (Beirut: Saqi, 2006)

James, Dietrich Mahoney, and Rueschemeyer, ‘Comparative Historical

Analysis in the Social Sciences’ (Cambridge: Cambridge University
Press, 2003)

Jawad, Saʻad, Iraq and the Kurdish Question, 1958-1970 (London: Ithaca
Press, 1981)

Jongerden, Joost, ‘A Spatial Perspective on Political Group Formation in

Turkey after the 1971 Coup: The Kurdistan Workers’ Party (PKK)’,

314
 Kurdish Studies, 5.2 (2017), 134–156

Jongerden, Joost, ‘Learning from Defeat: Development and Contestation

of the “New Paradigm” Kurdistan Workers’ Party (PKK)’, Kurdish
Studies, 7.1 (2019), 72–92

Jongerden, Joost, ‘Village Evacuation and Reconstruction in Kurdistan

(1993-2002)’, Études rurales, 186 (2010),
<http://journals.openedition.org/etudesrurales/9241>

Jongerden, Joost, Hugo de Vos, and Jacob van Etten, ‘Forest Burning as
Counterinsurgency in Turkish-Kurdistan: An Analysis from Space’, The
International Journal of Kurdish Studies, 21.1&2 (2007)

Jongerden, Joost, The Settlement Issue in Turkey and the Kurds: An

Analysis of Spatial Policies, Modernity and War (Leiden: Brill, 2007)

Jwaideh, Wadie, The Kurdish National Movement: Its Origins and

Development (Syracuse: Syracuse University Press, 2006)

Kandiyoti, Deniz, ‘Bargaining with Patriarchy’, Gender & Society, 2.3 (1988),
274–90

Kaplan, Morgan L., ‘Foreign Support, Miscalculation, and Conflict

Escalation: Iraqi Kurdish Self-Determination in Perspective’,
Ethnopolitics, 18.1 (2019), 29–45

Kasaba, Reşat, ed., The Cambridge History of Turkey, Volume 4

(Cambridge: Cambridge University Press, 2008)

Kaya Seferoğlu, Şükrü, Anadolu’nun İlk Türk Sakinleri: Kürtler (Ankara:

Ayyıldız Matbaası, 1982)

Kaya Seferoğlu, Şukrü, and Halil Kemal Türközü, 101 Soruda Türklerin Kürt
Boyu (Ankara: Türk Kültürünü Araştırma Enstitüsü, 1982)

Kaya, Zeynep, and Matthew Whiting, ‘The HDP, the AKP and the Battle for
Turkish Democracy’, Ethnopolitics, 18.1 (2019), 92–106

Keyder, Çağlar, State and Class in Turkey (London: Verso, 1987)

Kinnane, Derk, (1964), The Kurds and Kurdistan (London: Oxford University

315
 Press, 1964)

Kissinger, Henry, Years of Renewal (New York: Simon & Schuster, 1999)

Klein, Janet, ‘Kurdish Nationalists and Non-Nationalist Kurdists: Rethinking

Minority Nationalism and the Dissolution of the Ottoman Empire, 1908-
1909’, Nations and Nationalism, 13.1 (2007), 135–53

Klein, Janet, The Margins of Empire: Kurdish Militias in the Ottoman Tribal
Zone (Stanford: Stanford University Press, 2011)

Knapp, Michael, Anja Flach, and Ercan Ayboğa, Revolution in Rojava:

Democratic Autonomy and Women’s Liberation in Syrian Kurdistan
(London: Pluto Press, 2016)

Kodoman, Bayram, Sultan II. Abdülhamit’in Doğu Anadolu Politikası (Ankara:

Ankara Üniversitesi Basımevi, 1987)

Koohi-Kamali, Farideh, The Political Development of the Kurds in Iran:

Pastoral Nationalism (London: Palgrave Macmillan, 2003)

Kreyenbroek, Philip G., and Stefan Sperl, eds., The Kurds: A Contemporary
Overview (London and New York: Routledge, 1992)

Laclau, Ernesto, and Chantal Mouffe, Hegemony and Socialist Strategy :

Towards a Radical Democratic Politics (London: Verso, 1985)

Laclau, Ernesto, New Reflections on the Revolution of Our Time (London:

Verso, 1990)

Leezenberg, Michiel, ‘Humanitarian Aid in Iraqi Kurdistan’, Cahiers d’Études

Sur La Méditerranée Orientale et Le Monde Turco-Iranien, 29 (2000),
31–49

Leezenberg, Michiel, ‘Iraqi Kurdistan: Contours of a Post-Civil War Society’,

Third World Quarterly, 26.4–5 (2005), 631–47

Leezenberg, Michiel, ‘The Ambiguities of Democratic Autonomy: The

Kurdish Movement in Turkey and Rojava’, Southeast European and
Black Sea Studies, 16.4 (2016), 671–90

Lyon, Alynna J., and Emek M. Uçarer, ‘Mobilizing Ethnic Conflict: Kurdish

316
 Separatism in Germany and the PKK’, Ethnic and Racial Studies, 24.6
(2001), 925–48

MacKenzie, David N., Kurdish Dialects Studies I-II. (London: Oxford

University Press, 1961)

Mahdi, Kamil A., ed., Iraq’s Economic Predicament (Reading: Ithaca Press,
2002)

Malek, Mohammed, ‘Kurdistan in the Middle East Conflict’, New Left

Review, 1.175 (1989), 79–94

Marcus, Aliza, Blood and Belief: The PKK and the Kurdish Fight for
Independence (New York: New York University Press, 2007)

McDowall, David, A Modern History of the Kurds (London: I.B. Tauris,

2004)

Minorsky, Vladimir, 'Les origines des Kurdes', Actes du XXe Congres

Internationale des Orientalistes, Louvain (1940), 143-152;

Moore, Barrington, Social Origins of Dictatorship and Democracy: Lord and

Peasant in the Making of the Modern World (Boston, MA: Beacon
Press, 1966)

Mutlu, Servet, ‘Ethnic Kurds in Turkey: A Demographic Study’, International

Journal of Middle East Studies, 28 (1996), 517–41

Nairn, Tom, ‘The Modern Janus’, New Left Review, 1.94 (1975), 3–29

Natali, Denise, The Kurdish Quasi-State: Development and Dependency in

Post-Gulf War Iraq (Syracuse: Syracuse University Press, 2010)

Nikitin, Vasily, Les Kurdes: Etude Sociologique et Historique (Paris:

Imprimerie Nationale, 1956)

Noori Najmalddin, Nyaz, ‘The Failure of Economic Reform in the Kurdistan

Region of Iraq (1921-2015): The Vicious Circle of Uncivic Traditions,
Resource Curse, and Centralization’, British Journal of Middle Eastern
Studies, 45.2 (2018), 156–75

O’Ballance, Edgar, The Kurdish Revolt: 1961-1970 (London: Faber and

317
 Faber, 1973)

O’Connor, Francis, and Bahar Başer, ‘Communal Violence and Ethnic

Polarization before and after the 2015 Elections in Turkey: Attacks
against the HDP and the Kurdish Population’, Journal of Southeast
European and Black Sea, 18.1 (2018), 53–72

O’Driscoll, Dylan, and Bahar Başer, ‘Independence Referendums and

Nationalist Rhetoric: The Kurdistan Region of Iraq’, Third World
Quarterly, 40.11 (2019), 2016-2034

O’Shea, Maria, ed., Kurdistan: Political and Economic Potential (London:

Geopolitics and International Boundaries Research Centre SOAS,
1992)

O’Shea, T. Maria, Trapped between the Map and Reality: Geography and
Perceptions of Kurdistan (London and New York: Routledge, 2004)

Ocaklı, Feryaz, ‘Politics in the Kurdish Periphery: Clan Networks and Local
Party Strategies in a Comparative Perspective’, Middle Eastern
Studies, 53.4 (2017), 571–84

Öcalan, Abdullah, Kadın ve Aile Sorunu (Istanbul: Melsa Yayınları, 1992)

Öcalan, Abdullah, Kürdistan Devriminin Yolu (Cologne: Weşanen

Serxwebun, 1993)

Öcalan, Abdullah, Liberating Life: Woman’s Revolution (Cologne:

International Initiative ‘Freedom for Abdullah Öcalan’ – Peace in
Kurdistan”, 2013)

Öcalan, Abdullah, Manifesto for a Democratic Civilization, Volume 1,

Civilization: The Age of Masked Gods and Disguised Kings (Porsgrunn,
Norway: New Compass Press, 2015)

Öcalan, Abdullah, War and Peace in Kurdistan (Cologne: International

Initiative “Freedom for Abdullah Ocalan – Peace in Kurdistan”, 2008)

Olson, Robert, The Emergence of Kurdish Nationalism and the Sheikh Said
Rebellion, 1880-1925 (Houston: University of Texas Press, 1989)

318
Özeren, Süleyman, Murat Sever, Kamil Yılmaz, and Alper Sözer, ‘Whom Do
They Recruit? Profiling and Recruitment in the PKK/KCK’, Studies in
Conflict & Terrorism, 37.4 (2014), 322–47
<https://doi.org/10.1080/1057610X.2014.879381>

Özoğlu, Hakan, Kurdish Notables and the Ottoman State (Albany: State
University of New York Press, 2004)

Palani, Kamaran, Jaafar Khidir, Mark Dechesne, and Edwin Bakker, ‘The
Development of Kurdistan’s de Facto Statehood: Kurdistan’s
September 2017 Referendum for Independence’, Third World
Quarterly, 40.12 (2019), 2270-2288

Pamuk, Şevket, Uneven Centuries: Economic Development of Turkey since

1820 (Princeton: Princeton University Press, 2018)

Park, Bill, ‘Explaining Turkey’s Reaction to the September

2017Independence Referendum in the KRG: Final Divorce or
Relationship Reset?’, Ethnopolitics, 18.1 (2019), 46-60

Park, Bill, ‘Turkey, the US and the KRG: Moving Parts and the Geopolitical
Realities’, Insight Turkey, 14.3 (2012), 109–25

Park, Bill, Joost Jongerden, Francis Owtram, and Akiko Yoshioka, ‘Field
Notes: On the Independence Referendum in the Kurdistan Region of
Iraq and Disputed Territories in 2017’, Kurdish Studies, 5.2 (2017), 199–
214

Park, Bill, Turkey's policy towards northern Iraq: problems and

perspectives (London: IISS, 2005)

Pek, Ahmet, and Behsat Ekici, ‘Narcoterrorism in Turkey: The Financing of

PKK-KONGRA GEL from Illicit Drug Business’, in Understanding and
Responding to the Terrorism Phenomenon: A Multi-Dimensional
Perspective, ed. by Ozgur Nikbay and Suleyman Hancerli (Washington:
IOS Press, 2007), pp. 140–52

Perinçek, Doğu, Abdullah Öcalan Ile Görüşmeler (Istanbul: Kaynak Yayınları,

1990)

319
PKK: Program ve Tüzüğü (Cologne: Weşanên Serxwebûn, 1995)

Polanyi, Karl, The Great Transformation: The Political and Economic Origins
of Our Time (Boston: Beacon Press, 2001)

Pollack, Kenneth, The Threatening Storm: What Every American Needs to

Know Before an Invasion in Iraq (New York: Random House, 2002)

Rafaat, Aram, ‘The Fundamental Characteristics of the Kurdish Nationhood

Project in Modern Iraq’, Middle Eastern Studies, 52.3 (2016), 488–504

Romano, David, The Kurdish Nationalist Movement: Opportunity,

Mobilization and Identity (Cambridge: Cambridge University Press,
2006)

Roth, Mitchel P., and Murat Sever, ‘The Kurdish Workers Party (PKK) as
Criminal Syndicate: Funding Terrorism through Organized Crime, A
Case Study’, Studies in Conflict & Terrorism, 30.10 (2007), 901–20

Rubin, Avshalom H., ‘Abd Al-Karim Qasim and the Kurds of Iraq:
Centralization, Resistance and Revolt, 1958-63’, Middle Eastern
Studies, 43.3 (2007), 353–82

Rueschemeyer, Dietrich, Evelyne Huber Stephens, and John D. Stephens,

Capitalist Development and Democracy (Cambridge: Polity Press,
1992)

Skocpol, Theda, ‘A Critical Review of Barrington Moore’s Social Origins of

Dictatorship and Democracy’, Politics and Society, 4.1 (1973), 1–34

Skocpol, Theda, States and Social Revolutions: A Comparative Analysis of

France, Russia, and China (Cambridge: Cambridge University Press,
1979)

Sönmez, Mustafa, Kürtler: Ekonomik ve Sosyal Tarih, Doğu Anadolu’nun

Hikayesi (Istanbul: Arkadaş Yayınevi, 1992)

Sosyalizm ve Toplumsal Mücadeleler Ansiklopedisi, Volume 5 (Istabul:

İletişim Yayınları, 1988)

Spivak, Gayatri C., ‘Can the Subaltern Speak?’ in Marxism and the

320
 Interpretation of Culture, ed. by Cary Nelson and Lawrence Grossberg
(Urbana and Chicago: University of Illinois Press, 1988)

Stansfield, Gareth, and Mohammed Shareef, eds., The Kurdish Question

Revisited (London: Oxford University Press, 2017)

Stansfield, Gareth, Iraqi Kurdistan: Political Development and Emergent

Democracy (London and New York: Routledge, 2003)

Sumer, Fahrettin, and Jay Joseph, ‘The Paradox of the Iraqi Kurdish
Referendum on Independence: Contradictions and Hopes for
Economic Prosperity’, British Journal of Middle Eastern Studies, 45.4
(2018), 574–88

Tahiri, Hussein, The Structure of Kurdish Society and the Struggle for a
Kurdish State (Costa Mesa, CA: Mazda Publishers, 2007)

Tejel, Jordi, Peter Sluglett, Riccardo Bocco, and Hamit Bozarslan, eds.,
Writing the Modern History of Iraq: Historiographical and Political
Challenges (Singapore: World Scientific, 2012)

Tejel, Jordi, Syria’s Kurds: History, Politics and Society (London and New
York: Routledge, 2009)

Tezcür, Güneş Murat, ‘Kurdish Nationalism and Identity in Turkey: A

Conceptual Reinterpretation’, European Journal of Turkish Studies, 10
(2009)

Tezcür, Güneş Murat, ‘Ordinary People, Extraordinary Risks: Participation in

an Ethnic Rebellion’, American Political Science Review, 110.2 (2016),
247–64

Tezcür, Güneş Murat, ‘Violence and Nationalist Mobilization: The Onset of

the Kurdish Insurgency in Turkey’, Nationalities Papers, 43.2 (2015),
248–66

Tikku, Girdhari, ed., Islam and Its Cultural Divergence: Studies in Honor of
Gustave E. von Grunebaum (Urbana: University of Illinois Press, 1971)

Tilly, Charles, Durable Inequality (Berkeley: University of California Press,

1998)
321
Tunç, Volkan, ‘Van Milletvekili İbrahim Arvas’ın Biyografisi ve Türkiye Büyük
Millet Meclisi’ndeki Faaliyetleri’, Akademik Tarih ve Düşünce Dergisi,
6.2 (2019), 388–432

Ünal, Mustafa Cosar, ‘The Kurdistan Workers’ Party (PKK) and Popular
Support: Counterterrorism towards an Insurgency Nature’, Small Wars
& Insurgencies, 23.3 (2012), 432–55
<https://doi.org/10.1080/09592318.2012.661610>

Üngör, Ugur Ümit, The Making of Modern Turkey: Nation and State in
Eastern Anatolia, 1913-1950 (Oxford: Oxford University Press, 2011)

Vali, Abbas, ed., Essays on the Origins of Kurdish Nationalism (Costa Mesa,
CA: Mazda Publishers, 2003)

van Bruinessen, Martin, ‘Between Guerrilla War and Political Murder: The
Workers’ Party of Kurdistan’, MERIP Middle East Report, 153 (1988),
40–46

van Bruinessen, Martin, ‘Shifting National and Ethnic Identities: The Kurds
in Turkey and the European Diaspora’, Journal of Muslim Minority
Affairs, 18.1 (1998), 39–52

van Bruinessen, Martin, Agha, Shaikh and State: The Social and Political
Structures of Kurdistan (Chicago: University of Chicago Press, 1992)

van Bruinessen, Martin, Kurdish Ethno-Nationalism versus Nation-Building

States: Collected Articles (Istanbul: ISIS, 2000)

Vanly, Ismet Cheriff, Survey of the National Question of Turkish Kurdistan

with Historical Background (Rome: Hevra, 1971)

Voller, Yaniv, ‘Kurdish Oil Politics in Iraq: Contested Sovereignty and

Unilateralism’, Middle East Policy, XX.1 (2013), 68–82

Voller, Yaniv, The Kurdish Liberation Movement in Iraq: From Insurgency to

Statehood (London and New York: Routledge, 2014)

Wahbi, Taufiiq, The Origins of the Kurds and Their Language (Oxford:
Clarendon Press, 1966)

322
Wahlbeck, Osten, Kurdish Diasporas: A Comparative Study of Kurdish
Refugee Communities (New York: St. Martin’s Press, 1999)

Wardhaugh, Ronald, An Introduction to Sociolinguistics (Oxford: Wiley-

Blackwell, 1997)

Watts, Nicole F., Activists in Office: Kurdish Politics and Protest in Turkey
(Seattle and London: University of Washington Press, 2010)

Weber, Max, From Max Weber: Essays in Sociology, ed. by Hans Gerth and
C. Wright Mills (New York: Routledge, 1958)

White, Paul, ‘The March 1990 Uprising in Turkish Kurdistan & Its Effects on
Turkish Politics’, Kurdish Times, 4.1/2 (1991), 97–106

White, Paul, Primitive Rebels or Revolutionary Modernizers?: The Kurdish

National Movement in Turkey (London and New York: Zed Books,
2000)

White, Paul, The PKK: Coming Down from the Mountains (London: Zed
Books, 2015)

Wood, Ellen Meiksins, Democracy against Capitalism: Renewing Historical

Materialism (Cambridge: Cambridge University Press, 1995)

Wright, Erik Olin, ed., Approaches to Class Analysis (Cambridge:

Cambridge University Press, 2005)

Wright, Erik Olin, Understanding Class (London: Verso, 2015)

Yadirgi, Veli, The Political Economy of the Kurds of Turkey: From the
Ottoman Empire to the Turkish Republic (Cambridge: Cambridge
University Press, 2017)

Yeǧen, Mesut, ‘Turkish Nationalism and the Kurdish Question’, Ethnic and
Racial Studies, 30.1 (2007), 119–51

Yıldız, Kerim, The Kurds in Iraq: The Past, Present and Future (London:
Pluto Press, 2005)

Zana, Mehdi, Prison No.5: Eleven Years in Turkish Jails (Watertown, Ma:
Blue Crane Books, 1997)

323
Zedong, Mao, On the Question of the National Bourgeoisie and the
Enlightened Gentry (March 1, 1948) (Bejing: Foreign Language Press,
1969)

Zürcher, Eirk J., ‘The Vocabulary of Muslim Nationalism’, International

Journal of the Sociology of Language, 137 (1999), 81–93

Zürcher, Erik J., Turkey: A Modern History (London and New York: I.B.
Tauris, 1997)

Newspapers and Magazines Articles

‘“KRG’s Investment Law Is amongst the Most Investor Friendly Laws in the
Wider Region” – Minister Falah Mustafa Bakir’, The Kurdistan Tribune,
20 November 2011 <https://kurdistantribune.com/krgs-investment-
law-amongst-most-investor-friendly-laws-wider-region-minister-
falah-mustafa-bakir/>

‘Criminal Raid Prevents Kurdistan’s First Independent TV Station from

Covering Sulaymaniyah Unrest’, Reporters Without Borders, 22
February 2011 <https://rsf.org/en/news/criminal-raid-prevents-
kurdistans-first-independent-tv-station-covering-sulaymaniyah-
unrest>

‘Excerpts From 2 Statements by Bush on Iraq’s Proposal for Ending

Conflict’, The New York Times, 16 February 1991
<https://www.nytimes.com/1991/02/16/world/war-gulf-bush-
statement-excerpts-2-statements-bush-iraq-s-proposal-for-
ending.html>

‘Interview with Hozat, Karayılan and Bayık’, Peace in Kurdistan, 18 July

2013 <https://peaceinkurdistancampaign.com/2013/07/18/interview-
with-hozat-karayilan-and-bayik/>

‘Iran to Establish Goods Trading Center in Iraq’s Sulaymaniyah’, Mehr News

Agency, 29 October 2018
<https://en.mehrnews.com/news/139170/Iran-to-establish-goods-

324
 trading-center-in-Iraq-s-Sulaymaniyah>

‘Iraq Kurdistan Independence Referendum Planned’, BBC News, 1 July

2014 <https://www.bbc.co.uk/news/world-middle-east-28103124>

‘Iraqi Kurdistan Oil Revenue $1 Billion per Month: Source’, Ekurd Daily, 2
November 2016 <https://ekurd.net/kurdistan-oil-revenue-billion-2016-
11-02>

‘Iraqi Kurdistan: Prevent Attacks on Protesters’, Human Rights Watch, 7

March 2011 <https://www.hrw.org/news/2011/03/07/iraqi-kurdistan-
prevent-attacks-protesters>

‘Iraqi Kurdistan: Ruling Party Forces Fire on Protesters’, Human Rights

Watch, 21 October 2015 <https://www.hrw.org/news/2015/10/21/iraqi-
kurdistan-ruling-party-forces-fire-protesters>

‘Iraqi Kurds Protest against Turkish “Genocide”’, Euronews, 8 February

2016 <https://www.euronews.com/2016/02/08/iraqi-kurds-protest-
against-turkish-genocide>

‘Kaka Hama: Parties Warned by US Officials to Keep Barzani President’,

Rudaw, 18 August 2015
<https://www.rudaw.net/english/kurdistan/180820156>

‘Kurdistan Will Hold Independence Referendum in 2017, Senior Official’,

Rudaw, 2 April 2017
<https://www.rudaw.net/english/kurdistan/020420171>

‘Over $1 Billion from Iraqi Kurdistan Oil Revenue Missing in 1st Quarter of
2016: MP’, Ekurd Daily, 7 April 2016 <https://ekurd.net/billion-iraqi-
kurdistan-oil-missing-2016-04-07>

‘Protesters Storm Kurdistan Parliament after Barzani Announces

Resignation’, Reuters, 29 October 2017
<https://www.reuters.com/article/us-mideast-crisis-iraq-kurds-
protests/protesters-storm-kurdistan-parliament-after-barzani-
announces-resignation-idUSKBN1CY0PI>

‘Protests Intensify in Iraqi Kurdistan amid Economic Crisis’, Human Rights

325
 Watch, 9 February 2016 <https://www.reuters.com/article/us-mideast-
crisis-iraq-protests-idUSKCN0VI11X>

‘Rockets Hit Iranian Kurdish Opposition Offices in Iraq’s Koya’, Al Jazeera, 8

September 2018 <https://www.aljazeera.com/news/2018/09/rockets-
hit-iranian-kurdish-opposition-offices-iraq-koya-
180908090605503.html>

‘SHP’li Canver’in Güneydoğu İzlenimleri: İskence Ayibi Hepiminiz’,

Cumhuriyet, 12 February 1986

‘Who Are the Kurds?’, Al Jazeera, 1 October 2017

<https://www.aljazeera.com/news/2017/05/kurds-
170516072934887.html>

Ali Saleem, Zmkan, and Mac Skelton, ‘Protests and Power: Lessons from
Iraqi Kurdistan’s Opposition Movement’, LSE Middle East Centre Blog,
10 November 2019 <https://blogs.lse.ac.uk/mec/2019/11/10/protests-
and-power-lessons-from-iraqi-kurdistans-opposition-movement/>

Çandar, Cengiz, ‘Erdogan-Barzani 'Diyarbakir encounter' milestone’, al-

Monitor, 20 November 2013 < https://www.al-
monitor.com/pulse/originals/2013/11/erdogan-barzani-kurdistan-
diyarbakir-political-decision.html>

Cazeaux, Jean-Jacques, ‘Turquie: La Grande Misère Des Kurdes’, Le

Monde, 16 June 1983
<https://www.lemonde.fr/archives/article/1983/06/16/turquie-la-
grande-misere-des-kurdes_3076818_1819218.html>

Chomani, Kamal, ‘Kurdistan Region at a Time of Crown Princes’, 1001 Iraqi

Thoughts, 28 December 2018
<https://1001iraqithoughts.com/2018/12/28/kurdistan-region-in-a-
time-of-kurdish-crown-princes>

Cockburn, Patrick, ‘Kurdish Chief’s Death Brings Civil War Nearer’, The
Independent, 6 July 1996
<https://www.independent.co.uk/news/world/kurdish-chiefs-death-
brings-civil-war-nearer-1327460.html>
326
Ghafuri, Lawk, ‘Before Sending Oil, Kurds Want Baghdad to Clear Their
Debts’, Rudaw, 23 September 2019
<https://www.rudaw.net/english/kurdistan/230920193>

Gürcan, Metin, ‘PKK Looks to the Future with Creation of Youth Militias’, Al-
Monitor, 31 August 2015 <https://www.al-
monitor.com/pulse/originals/2015/08/turkey-kurds-pkk-armed-young-
militias.html>

Hawramy, Fazel, ‘Assassinations Mount as Iranian Kurdish Militants Clash

with Tehran’, Al-Monitor, 7 March 2018 <https://www.al-
monitor.com/pulse/originals/2018/03/iran-kdpi-kurdish-opposition-
iraq-assassinations-rahmani.html#ixzz5bkPrbUIC>

Hawramy, Fazel, ‘Iraqi Kurdistan’s Movement for Change Faces Rebellion

from Within’, Al-Monitor, 9 July 2018 https://www.al-
monitor.com/pulse/originals/2018/07/change-movement-nawshirwan-
mustafa-kurdistan-iraq.html>

MacDiarmid, Campbell, ‘Masoud Barzani: Why It’s Time for Kurdish

Independence’, Foreign Policy, 15 June 2017
<https://foreignpolicy.com/2017/06/15/masoud-barzani-why-its-time-
for-kurdish-independence/>

Nawzad, Kosar, ‘Kurdistan Ends Unpopular Austerity Measure, Will Pay

Public Salaries “in Full”’, Kurdistan 24, 8 March 2019
<https://www.kurdistan24.net/en/news/34b6f15b-f457-4a18-a76f-
01531dab5475>

Neuhof, Florian, ‘Unpaid Peshmerga Are Voting with Their Feet’, Deutsche
Welle, 2 February 2016 <https://www.dw.com/en/unpaid-peshmerga-
are-voting-with-their-feet/a-19020152>

Ottaway, Marina, and Anas Danial Kaysi, ‘Iraq: Protest, Democracy, and
Autocracy’, Canregie Endowment for International Peace, 28 March
2011 <https://carnegieendowment.org/2011/03/28/iraq-protest-
democracy-and-autocracy-pub-43306>

Salih, Mohammed A., ‘KRG Parliament Speaker: Barzani’s Term Extension

327
 “against the Law”’, Al-Monitor, 28 August 2015 <https://www.al-
monitor.com/pulse/originals/2015/08/kurdish-parliament-speaker-
challenge-barzani-legitimacy.html>

Sattar, Omar, ‘New Pipeline in Works to Transport Iraqi Oil to Turkey’, Al-
Monitor, 18 September 2019 <https://www.al-
monitor.com/pulse/originals/2019/09/iraq-turkey-pipeline-kirkuk-
ceyhan.html>

Stanley, Alessandra, ‘Italy Rejects Turkey’s Bid For the Extradition of Kurd’,
The New York Times, 21 November 1998
<https://www.nytimes.com/1998/11/21/world/italy-rejects-turkey-s-
bid-for-the-extradition-of-kurd.html>

Tas, Latif, Nadje Al-Ali, and Gültan Kişanak, ‘Kurdish Women’s Battle
Continues against State and Patriarchy, Says First Female Co-Mayor
of Diyarbakir. Interview ’, OpenDemocracy, 12 August 2016
<https://www.opendemocracy.net/en/kurdish-women-s-battle-
continues-against-state-and-patriarchy-/>

Tastekin, Fehim, ‘Storming of Base in Iraq a Grave Signal for Turkey’, Al-
Monitor, 1 February 2019 <https://www.al-
monitor.com/pulse/originals/2019/02/turkey-iraqi-kurdistan-storming-
of-base-grave-signal.html>

Tuysuz, Gul, Hamdi Alkhshali, and Muwafaq Mohammed, ‘At Least 6 Killed
during Violent Protests in Iraqi Kurdistan’, CNN, 19 December 2017
<https://edition.cnn.com/2017/12/19/middleeast/iraq-kurdish-
protests/index.html>

Ünal, Ali, ‘Turkey’s Halkbank Collects the Revenue of Kurdish Oil’, Daily
Sabah, 24 June 2014
<https://www.dailysabah.com/energy/2014/06/24/turkeys-halkbank-
collects-the-revenue-of-kurdish-oil>

Wali, Rebwar Karim, ‘Qasim Agha, Other Collaborators of Saddam Hussein’,

Ekurd Daily, 31 October, 2010
<https://ekurd.net/mismas/articles/misc2010/10/state4329.htm>
328
Woolf, Nicky, ‘Is Kurdistan the next Dubai?’, The Guardian, 5 May 2010
<https://www.theguardian.com/commentisfree/2010/may/05/kurdista
n-next-dubai-iraq>

Reports

‘Turkey’s Village Guard System – Still in Place, Still an Obstacle’, Kurdish

Human Rights Project, Briefing Paper, 22 March 2011
<http://www.khrp.org/latest-news/646-turkeys-village-guard-system-
still-in-place-still-an-obstacle.html>

Abrahams, Fred, and Lama Fakih, Under Kurdish Rule: Abuses in PYD-Run
Enclaves of Syria (Human Rights Watch, 2014)
<https://www.hrw.org/report/2014/06/19/under-kurdish-rule/abuses-
pyd-run-enclaves-syria>

After Iraqi Kurdistan’s Thwarted Independence Bid (International Crisis

Group, 2019) <https://d2071andvip0wj.cloudfront.net/199-after-iraqi-
kurdistan.pdf>

Bogos, Kristina, Analysis: Looking Back on the 2018 Kurdish Elections (IRIS,
March 2019)
<https://auis.edu.krd/iris/sites/default/files/Analysis_Looking Back on
the 2018 Kurdish Elections_March 2019.pdf>

BP Statistic Review of World Energy – June 2018, 67th Edition (BP, 2018)
<https://www.bp.com/content/dam/bp/business-
sites/en/global/corporate/pdfs/energy-economics/statistical-
review/bp-stats-review-2018-full-report.pdf>

Calculating the Gross Regional Product of the Kurdistan Region - Iraq (KRG
and Rand Corporation, 2016) <https://doi.org/10.7249/rr1405.3>

Capacity Building at the Kurdistan Regional Statistics Organization

Through Data Collection (KRG and Rand Corporation, 2014)
<https://doi.org/10.1214/07-EJS057>

DeWeaver, Mark, The State of the Economy: Economic Issues in Iraq and

329
 the Kurdistan Region (IRIS, March 2017) <
https://auis.edu.krd/iris/publications/iris-booklet-series-state-
economy>

Drought Characteristics and Management in North Africa and the Near

East (Food and Agriculture Organization, 2018)
<http://www.fao.org/3/CA0034EN/ca0034en.pdf>

Genocide in Iraq: The Anfal Campaign against the Kurds (Human Rights
Watch, 1993) <www.hrw.org/legacy/reports/1993/iraqanfal>

Hoff, Ruud, Michiel Leezenberg, and Pieter Muller, Elections in Iraqi

Kurdistan (May 19, 1992): An Experiment in Democracy (Amsterdam:
Netherlands-Kurdistan Friendship Society, 1992)

In Best of Times and Worst of Times: Addressing Structural Weaknesses of

the Kurdistan Region’s Economy (MERI, 2016) <http://www.meri-
k.org/wp-content/uploads/2016/01/MERI-Economic-Report-January-
2016-2.pdf>

Iraq Human Development Report 2014: Iraqi Youth Challenges and

Opportunities, (UNDP, 2014)
<http://hdr.undp.org/sites/default/files/undp-iq_iraqnhdr2014-
english.pdf>

Iraq Population Situation Analysis Report 2012 (Iraq National Population

Commission and UNFPA, 2012)
<http://iraq.unfpa.org/publications/cat_view/1-documents-english>

Mansour, Renad, and Christine Van Den Toorn, The 2018 Iraqi Elections: A
Population in Transition?, (LSE Middle East Centre and IRIS, July 2018)
<http://eprints.lse.ac.uk/89698/7/MEC_Iraqi-
elections_Report_2018.pdf>

McDowall, David, The Destruction of Villages in South-East Turkey: A

Report by Medico International and the Kurdish Human Rights Project
(KHRP) (London: Medico International and KHRP, 1996)

Mills, Robin, ‘Under the Mountains: Kurdish Oil and Regional Politics’, The

330
 Oxford Institute for Energy Studies, OIES Paper WPM 63 (January
2016) <https://www.oxfordenergy.org/wpcms/wp-
content/uploads/2016/02/Kurdish-Oil-and-Regional-Politics-WPM-
63.pdf>

Reader #5: Stateless Democracy, realized with Kurdish Women’s

Movement, New World Academy, 19 May 2015, < https://www.e-
flux.com/announcements/29509/new-world-academy-reader-5-
stateless-democracy-realized-with-kurdish-women-s-movement/>

Regional Development Strategy for Kurdistan Region 2012-2016 (KRG -

Ministry of Planning, 2011) <http://www.mop.gov.krd/resources/MoP
Files/PDF Files/gd_ps/regional_development_strategy.pdf>

TATORT Kurdistan, Democratic Autonomy in North Kurdistan (Porsgrunn,

Norway: New Compass Press, 2013)

The Kurdistan Region of Iraq: Assessing the Economic and Social Impact of
the Syrian Conflict and ISIS (World Bank Group, 2015)
<https://doi.org/10.1192/bjp.111.479.1009-a>

The Kurdistan Region of Iraq: Reforming the Economy for Shared

Prosperity and Protecting the Vulnerable (World Bank Group, 2016)
<http://documents.worldbank.org/curated/en/672671468196766598/
Main-report>

The Kurds of Turkey: Killings, Disappearances and Torture (New York:

Human Rights Watch, March 1993)
<https://www.hrw.org/sites/default/files/reports/TURKEY933.PDF>

Torture in the Eighties (London: Amnesty International Publications, 1984)

Van Wilgenburg, Wladimir, and Mario Fumerton, Kurdistan’s Political Army:

The Challenge of Unifying the Peshmerga Forces (Carnegie Middle
East Center, 2015)
<https://carnegieendowment.org/files/ACMR_WilgenburgFumerton_Ku
rdistan_English_final.pdf>

331
Unpublished Documents, Webpages and Other Items

‘Backgrounder on the Union of Communities in Kurdistan, KCK’, Democratic

Turkey Forum
<http://www.tuerkeiforum.net/enw/index.php/Backgrounder_on_the_U
nion_of_Communities_in_Kurdistan,_KCK#cite_note-bia201110-9>

‘CASE OF ÖCALAN v. TURKEY (Application No. 46221/99)’, ECHR, 12 May

2005 <https://hudoc.echr.coe.int/eng#%7B%22itemid%22:[%22001-
69022%22]%7D>

‘Congresses of Kurdistan Democratic Party (1946-1999)’, KDP,

<https://www.kdp.se/index.php?do=congress>

‘Democratic Society Congress’, Demokratik Toplum Kongresi

<https://www.kcd-dtk.org/english-2/>

‘Entrenched Corruption in Kurdistan Region of Iraq’, Baghdad Embassy,

Wikileaks Cables: 08BAGHDAD2731_a, 25 August 2008
<https://wikileaks.org/plusd/cables/08BAGHDAD2731_a.html>

‘Final Resolution on the 8th Congress of the PKK’, Management Committee

of KADEK, 15 April 2002

‘Imported Crude Oil Prices’, US Energy Information Administration, 2019

<https://www.eia.gov/outlooks/steo/realprices/>

‘Investment Law’, KRG Department of Foreign Relations, 2006

<https://dfr.gov.krd/p/p.aspx?p=69&l=12&s=050200&r=377>

‘Iraq’s Constitution of 2005’, Constitute Project,

<https://www.constituteproject.org/constitution/Iraq_2005.pdf?lang=e
n>

‘IRAQ-UNITED NATIONS: MEMORANDUM OF UNDERSTANDING ON THE

SALE OF IRAQI OIL, IMPLEMENTING SECURITY COUNCIL
RESOLUTION 986 (1995)’, International Legal Materials, 35.4 (1996),
1095–1101

‘Jineolojî’, International Vrije Vrouwen Stichting, 2019

<https://www.ifwf.nl/2019/03/15/jineoloji/>
332
‘KRG Spokesperson Statement on Turkish Forces Presence in Bashiqah’,
Kurdistan Regional Government, 6 October 2016
<http://previous.cabinet.gov.krd/a/d.aspx?s=040000&l=12&a=54994>

‘Kurdistan Region - Determined to Grow: Economy, Kurdistan Region,

Review’, Invest in Group (IIG), 2013
<https://investingroup.org/review/236/determined-to-grow-economy-
kurdistan/>

‘Kurdistan Region Facts & Figures’, Invest in Group (IIG), 2019

<https://investingroup.org/country/kurdistan/facts-figures/>

‘Kurdistan Regional Government Unification Agreement’, Kurdistan

Regional Government, 2006
<http://cabinet.gov.krd/a/d.aspx?r=223&l=12&a=8891&s=02010100&s
=010000>

‘Seçim Arşivi’, T.C. Yüksek Seçim Kurulu <http://www.ysk.gov.tr/tr/secim-

arsivi/2612>

‘The Experience of Co-operative Societies in Rojava’, TEV DEM Economic

Committee, 19 January 2016

‘The Program & Political Platform of the Adopted in the First PUK Congress
1992’, PUKmedia, 2015
<https://www.pukmedia.com/EN/EN_Direje.aspx?Jimare=34632>

‘Turkey: Kurdistan’, Uppsala Conflict Data Program, Conflict Encyclopedia,

Uppsala University <https://ucdp.uu.se/additionalinfo/354/1>

Chomani, Kamal, (@KamalChomani), ‘Son of Masrour Barzani, could be

around 15 yrs old (…)’, Twitter, · 28 April 2019, 17:05
<https://twitter.com/KamalChomani/status/1122532218298097664>

Demokratik Toplum Partisi: Program ve Tüzüğü, 2005

Final Declaration of the Foundation Conference of the People's Congress

of Kurdistan, 17 November 2003, (Cologne: International Initiative
Freedom for Ocalan)

International Paris Conference 14-15 October 1989. The Kurds: Human

333
 Rights and Cultural Identity (Paris: Institut Kurde de Paris, 1992)

Kocher, Matthew A., ‘Human Ecology and Civil War’ (unpublished doctoral
thesis, University of Chicago, 2004)

Leam, David (dir.), Lawrence of Arabia (Columbia Pictures, 1962).

Matbaacılık, Feryal, State of Anarchy and Terror in Turkey (Ankara, 1983)

Revolution in Kurdistan: The Essential Documents of the Patriotic Union of

Kurdistan (New York: PUK Publications, 1977)

334
List of Interviews

Abbas, Mustafa Abbas, University of Sulaymaniyah and Coalition for

Democracy and Justice (Sulaymaniyah, September 10, 2018)

Ahmedi, Loghman, KDPI (Koya, August 6, 2018)

Amin, Muthanna, Kurdistan Islamic Union (Sulaymaniyah, August 7, 2018)

Arab, Sileman, PYD representative in Iraq and KRG (Sulaymaniyah, March

10, 2019)

Assasard, Fared, Research Institute of Kurdistan (Sulaymaniyah, August 5,

2018)

Bozarslan, Hamit, Ecole des Hautes Études en Sciences Sociales (Paris,

November 8, 2019)

Çağlayan, Handan, University of Bamberg (personal communication, April 8,

2020)

Çandar, Cengiz, journalist and writer (personal communication, April 14-28,

2020)

Chomani, Kamal, journalist (personal communication, September 9, 2019)

Çimen, Devris, HDP representative in Brussels (Brussels, November 6, 2019)

Ferho, Derwich, Kurdish Institute of Brussels (Brussels, November 7, 2019)

Haji Ali, Qadir, former PUK and former Gorran (Sulaymaniyah, February 26,
2019)

Hamarahim, Jalal Abdullah, former PUK peshmerga (Sulaymaniyah, July 25,

2018)

Hasanzadeh, Asso, KDPI (Koya, August 6, 2018)

335
Jabar, Muhammad Hakim, Kurdistan Islamic Group (Sulaymaniyah, August
8, 2018)

Jaf, Ibrahim, Head of Jaf Tribe in Dukan (Dukan, March 12, 2019)

Jamal Tahir, Karwan, KRG representative in London (London, October 14,

2019)

Jan, Polad, Syrian Democratic Forces (Sulaymaniyah, March 10, 2019)

Jawhar, Jalal, Gorran Movement (Sulaymaniyah, March 12, 2019)

Karadaghi, Kamran, journalist and former spokesperson for the President of

Iraq (London, January 22, 2018)

Kartal, Remzi, Kurdistan National Congress (Brussels, November 5, 2019)

Karwan, Abu, Kurdistan Communist Party - Iraq (Sulaymaniyah, September

9, 2018)

Muhammed, Tara, Tevgar Azadi (Sulaymaniyah, August 10, 2018)

Mustafa, Khawla, Rojava representative in the KRG (Sulaymaniyah, March 4,

2019)

Nezan, Kendal, Kurdish Institute of Paris (Paris, November 11, 2019)

Omar, Hoshyar, Gorran Movement (Sulaymaniyah, August 8, 2018)

Saeed, Aram, New Generation Movement (Sulaymaniyah, August 11, 2018)

Semo, Alan, PYD representative in London (personal communication, April

4, 2020)

Zebari, Hoshyar, KDP (Primam, August 2, 2018)

336