Archaeology and State Theory Subjects An

Archaeology and State Theory
DEBATES IN ARCHAEOLOGY
Series editor: Richard Hodges
Against Cultural Property John Carman

The Anthropology of Hunter-Gatherers Vicki Cummings
Archaeologies of Conflict John Carman
Archaeology: The Conceptual Challenge Timothy Insoll
Archaeology and International Development in
Africa Colin Breen & Daniel Rhodes
Archaeology and State Theory Bruce Routledge
Archaeology and Text John Moreland
Archaeology and the Pan-European Romanesque Tadhg O’Keeffe
Beyond Celts, Germans and Scythians Peter S. Wells
Combat Archaeology John Schofield
Debating the Archaeological Heritage Robin Skeates
Early European Castles Oliver H. Creighton
Early Islamic Syria Alan Walmsley
Gerasa and the Decapolis David Kennedy
Image and Response in Early Europe Peter S. Wells
Indo-Roman Trade Roberta Tomber
Loot, Legitimacy and Ownership Colin Renfrew
Lost Civilization James L. Boone
The Origins of the Civilization of Angkor Charles F. W. Higham
The Origins of the English Catherine Hills
Rethinking Wetland Archaeology Robert Van
de Noort & Aidan O’Sullivan
The Roman Countryside Stephen Dyson
Shipwreck Archaeology of the Holy Land Sean Kingsley
Social Evolution Mark Pluciennik
State Formation in Early China Li Liu and Xingcan Chen
Towns and Trade in the Age of Charlemagne Richard Hodges
Vessels of Influence: China and the Birth of Porcelain in Medieval
and Early Modern Japan Nicole Coolidge Rousmaniere
Villa to Village Riccardo Francovich & Richard Hodges
Subjects and Objects of Power
Bruce Routledge
Bloomsbury Academic
An imprint of Bloomsbury Publishing Plc
50 Bedford Square 1385 Broadway

London New York
WC1B 3DP NY 10018
UK USA
www.bloomsbury.com
First published 2014
© 2014 Bruce Routledge
All rights reserved. No part of this publication may be reproduced or transmitted

in any form or by any means, electronic or mechanical, including photocopying,
recording, or any information storage or retrieval system, without prior
permission in writing from the publishers.
Bruce Routledge has asserted his right under the Copyright, Designs and Patents
Act, 1988, to be identified as Author of this work.
No responsibility for loss caused to any individual or organisation acting on

or refraining from action as a result of the material in this publication can be
accepted by Bloomsbury Academic or the author.
British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library.
ISBN: epdf: 978-1-47250-410-4
Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress.
Typeset by Fakenham Prepress Solutions, Fakenham, Norfolk NR21 8NN

Contents
List of figures vii

Abbreviations ix
Orientations 1
1 After (neo-)evolution(ism) 9
2 Coercion and consent 27
3 Hegemony in action: The kingdom of Imerina in central
Madagascar 49
4 Beyond politics: Articulation and reproduction in Athens
and the Inca Empire 67
5 Spectacle and routine 101
6 Routine lives and spectacular deaths: The Royal Tombs of Ur 127
7 Conclusion: The hazardous necessity of comparison 157
Bibliography 167
Index 191
List of figures
3.1 Map of the Kingdom of Imerina 51

3.2 Radama I 59
3.3 Stone slab tomb, central Madagascar 62
4.1 Map of Classical Greece featuring the polis of Athens 75
4.2 Greek courtyard house, fifth century bc, Olynthus House
Avii4 86
4.3 Map of the Inca Empire 89
4.4 Acllaconas spinning wool within an acllawasi compound
overseen by a mamacona 93
4.5 Inca ceramic figurine of a man carrying an aríbalo of chica
on his back 97
5.1 Map of the Classic Maya heartland with sites mentioned in
text marked 114
5.2 Si(j)yaj K’in Chaak II impersonating the water-lily serpent.
Machaquilá Stele 4 116
5.3 Schematic cross-section of a Classic Maya convex run-off
water collection system 119
6.1 Map of Mesopotamia showing sites mentioned in text 128
6.2 General plan of the ‘Royal Cemetery of Ur’ 129
6.3 Tomb PG/789 showing distribution of hair adornment on
bodies 131
6.4 Tomb PG/1237 (‘The Great Death Pit’) showing
distribution of hair adornment on bodies 132
6.5 ‘Peace’ side of the Royal Standard of Ur showing a banquet
scene 140
6.6 Soundbox of lyre (U. 10556) from tomb PG/789 showing
a satirical banquet scene with animals and mythological
creatures 142
6.7 Early Dynastic seal from Ur featuring a satirical banquet
scene with animals 148
Abbreviations
AA American Anthropologist
AD Archaeological Dialogues
APAAA Archeological Papers of the American Anthropological

Association
ARA Annual Review of Anthropology
CA Current Anthropology
CAJ Cambridge Archaeological Journal
CQ Classical Quarterly
CSSH Comparative Studies in Society and History
ETCSL Electronic Text Corpus of Sumerian Literature

(http://etcsl.orinst.ox.ac.uk/)
JAA Journal of Anthropological Archaeology
JNES Journal of Near Eastern Studies
LAA Latin American Antiquity
SEH Social Evolution and History

Orientations
We cannot speak of the State-thing as if it was a being developing

on the basis of itself and imposing itself on individuals as if by a
spontaneous, automatic mechanism. The State is a practice. The
State is inseparable from the set of practices by which the State
actually became a way of governing, a way of doing things, and a
way of relating to government.
Michel Foucault (2007: 276–7)
Today, most of us live in large-scale political communities that

are at once real and unreal. Their reality is the reality of our daily
experience; serving as the subject of our conversations and the object
of our frustrations. Yet, these polities are also unreal in that they have
no distinct material existence. We understand them to be more than
just the sum of all of us, yet whenever we look closely at government
itself we find only other human beings getting on with their jobs
(more or less). As government, and more particularly as the state,
these collectives would appear to have very real material effects that
could not be executed by individuals as individuals – war, taxation,
legal adjudication and enforcement, infrastructure, social services,
mass education, etc. Yet, the relationship between this collective
agency and other nodes of association and social power – kinship,
religion, class, ethnicity, gender, market forces – is uncertain, shifting
in intimacy between contexts and issues. Despite the uncertainty,
we build lives on the traces left by this collective agency, such as
the residency permit in my passport that prevents me from being
deported. In this way we are both dominated and enabled, directed
and facilitated, all without the immediate human engagement we
would otherwise anticipate from persistent relationships of funda-
mental importance to our lives. This collective political experience
of virtual causes with material effects is familiar, interesting and
2 Archaeology and State Theory
conceptually difficult. In other words, it is something worth thinking

about.
Many aspects of this collective political experience seem profoundly
modern, but not all. Certainly war and taxation were not invented in
the eighteenth century, nor, for that matter, was the experience of being
‘administered’. Named polities in the past have served as the context
and justification for the concentration of populations, the extraction
of resources, the application of force, the execution of collective
projects, and even the re-imagining of the cosmos. At the same time,
we can also recognise that not all human beings have lived in such
large-scale communities, and indeed for most of human history no
one did. Hence, as has long been noted, there is a viable archaeo-
logical perspective on how and why such political communities exist.
However, these questions present a special problem for archaeologists
and ancient historians. This is not because we study the wrong side
of a magic line separating modern from premodern worlds. We are,
however, on the ‘wrong side’ of historical and disciplinary divides
that render problematic the study and conceptualisation of large-scale
politics in the more distant past.
Since at least the eighteenth century, political theory has usually
addressed collective politics, complex polities and large-scale political
domination under the more succinct, and mysterious, category of the
state. Indeed, abstract reflection on the category of the state is suffi-
ciently developed that in recent convention the phrase ‘state theory’
has come to refer specifically to this tradition of Western political
thought (e.g. Marinetto 2007). For archaeologists, however, the state as
a category presents several serious difficulties.
First is the problem of definition. Any archaeologist who has
actually followed Kent Flannery’s (1998: 14) directive and sought out
definitions of the state in social anthropology and political science
rather than archaeology is likely to have been dismayed by the result.
Even within state theory the state is a vexed term, definitions of which
decline in utility as they increase in precision. No one seems certain
as to whether the state is a set of institutions or a kind of society, a
Orientations 3
structure of governance or a structure of coercion, a social contract

or an instrument of class dominance. To define the state narrowly,
say in terms of the formal offices and institutions of government, is to
ignore the socially embedded nature of state power. To define the state
broadly, say as a kind of encompassing social system, is to ignore the
real limits of state power.
As Jens Bartelson (2001: 10–11) notes, the definition of the state is
not only ambiguous, it is this ambiguity that allows the state to play
a central role in political discourse. This centrality, in turn, means
that the state both gathers meanings to itself and ascribes them to
other terms (see Bartelson 2001: 11). Hence, words like sovereignty,
authority, legitimacy, territory or civil society, words that might play a
role in constructing or limiting a definition of the state, are themselves
partially dependent on the category of the state for their own meanings.
This is because the category of the state has been used both to describe
and to shape collective politics over the past several centuries. Talking
about the state helped to bring it about, and continues to help reproduce
it today. This active role for state discourse presents archaeologists with
their second problem, namely what to do in its absence. In other words,
can one speak of the state in contexts where no historical actors were
doing so themselves?
Finally, archaeologists face the special problem of their own disci-
plinary history. For much of the past 30 years the state in archaeological
discourse has suffered from an extended neo-evolutionary hangover.
For many, the very term state is synonymous with neo-evolutionism
to the extent that archaeologists either wilfully avoid its serious
discussion, or dissipate their energies flogging, resuscitating or geneti-
cally modifying a long-dead neo-evolutionist horse. Getting beyond
this impasse requires thinking about where we are after the critique
of neo-evolutionism. If the state is not what neo-evolutionists said it
was, then what is it? Is it possible for archaeologists to disembed the
state from social evolutionary trajectories? If so, what do we make
of differences in social inequality, hierarchy, scale and centralisation
across space and time? Can human political organisation be studied
comparatively and what are the political implications of engaging (or

not engaging) with such a project in relation to the state? In the end, is
such a project even worthwhile? Why not make a clean break and bury
the state along with all those other totalising concepts (e.g. race, nation,
culture) that serve to create what they purport to describe?
Such questions reflect an emerging recognition that archaeolo-
gists need to think again about politics writ large. In the early days
of post-processual archaeology a good start was made on thinking
critically about the problem of large-scale politics from an archaeo-
logical perspective. However, the anti-essentialist basis of this critique
presented an immediate problem. How could one address large-scale
politics in the past without reifying the very concepts (e.g. the state)
one was deconstructing in the present?
At least two solutions to this problem have been proffered in the
subsequent work of archaeologists over the past 30 years. One group,
often citing Michel Foucault, stated quite simply that one could not
construct characterisations of large-scale politics in the past that
were both critical and comparative, since critical inquiry hinged on
the refusal of all such ‘total’ characterisations of social life (Tilley
1990: 325–8). Instead, many post-processual archaeologists pursued
a strategy of turning inwards to the immediacy of their own experi-
ences, their own bodies, and their own senses as a foundation for
social and political analysis (for critical reviews see Barrett and Ko
2009; Brück 2005). In an age of torture and terror, of economic and
environmental crises, this inward retreat (much like George Orwell’s
‘life inside the whale’) seems limited, even quaint. Interestingly, it is
also not the strategy that Foucault himself pursued. In refusing to begin
from global categories, such as Nietzsche’s ‘cold monster’ of the state
(Foucault 2007: 109), Foucault relentlessly pursued a path to global
realities by other means (e.g. Foucault 2007; 2008: 76–8, 185–8). Most
particularly, these means were the specific and situated practices of
power; the actual acts, localities and dispositions by which power was
distributed and amplified.
In this sense, the second answer to the problem of how to critically
Orientations 5
engage with large-scale politics in the past is in fact closer to Foucault,

albeit more muddled in its self-understanding. Over the past 30 years
we have seen a proliferation of archaeological studies that appear
to leave the category of the state largely unquestioned, even as they
pursue detailed studies of particular practices (e.g. craft production,
feasting, public spectacle, fiscal policies, monument construction, etc.)
that served to constitute specific states (as noted already in Stein 1998:
3–5). One can characterise this trend as a shift in concern from what
the state was to what a state does; that is to say a shift from a concern
for entities with characteristics towards a concern for practices with
effects. Despite its implicit nature, this trend represents a profound
change in the conceptualisation of large-scale politics within archae-
ology. By focusing on the material practices that constitute particular
states, archaeologists have ceased to treat the state as a singular ‘thing’
whose existence and form is given in the mechanism of adaptation, and
have begun to address politics as a social process.
Essentially, this book is about mapping this second trend in archae-
ological analysis, making it explicit and giving it a little push in
certain directions with potential for fruitful archaeological research.
In keeping with Thomas Lemke’s succinct summation of Foucault’s
analytics of government, I want to emphasise ‘practices instead of
object, strategies instead of function, and technologies instead of insti-
tution’ (Lemke 2007: 58). At the same time, I want this concern for
what specific human beings do to lead us back to the material presence
and global effects of large-scale political domination.
I would hope that such research might be rigorous and productive,
but also that it might prove relevant to politics in the broadest sense of
thinking and acting collectively in the world.
While it may be idiosyncratic, this book is not radically new. I pick
up and develop ideas already in circulation (see Smith 2011), including
those that I myself have published elsewhere in a rather different form
(Routledge 2004). What this book does provide is a brief and relatively
accessible point of entry into what can be a difficult topic. In keeping
with the theme of the Debates in Archaeology series, I have written this
book as an extended argument, rather than as a general overview. It is

not, therefore, a textbook in any meaningful sense. I give little attention
to the history of state theory within archaeology and I focus on the
development of my own arguments at the expense of judiciously repre-
senting a range of views. While I acknowledge that state-formation
contributes to major social transformations at key historical junctures,
I give little attention to the initial catalysts for such transformations (e.g.
warfare, trade, population pressure, etc.). This is because, on one hand,
many (many!) archaeologists have already addressed such catalysts in
some detail, and on the other hand, a focus on initial catalysts seems to
engender the attitude that a bit of hunger and warfare is all that it takes.
In other words, under the right conditions states make themselves
and, once made, they stay made (until they collapse, that is). In this
book I want to make exactly the opposite point. Political authority is
constituted by daily effort and constant cultural, social and economic
production. Indeed, what we call the state has no existence save as such
an on-going process. If one needs a summary of my argument in this
book it is ‘forget the state; focus on state-formation’.
These rather bold statements require both elaboration and justifi-
cation; an endeavour that will take up much of the remainder of this
book. In Chapter 1, I will specify more carefully our subject of concern,
examining and modifying approaches to large-scale politics in archae-
ology premised on the critique of neo-evolutionism; approaches Adam
Smith (2011) has recently characterised as ‘archaeologies of sover-
eignty’. In Chapter 2 we will turn to consider why power should cohere
in the forms that it does under sovereignty, giving particular attention
to the problem of coercion and consent through an examination of the
political writings of Antonio Gramsci. In Chapter 3 we will consider
the nineteenth-century kingdom of Imerina in central Madagascar
in order to illustrate Gramsci’s understanding of hegemony in terms
relevant to a variety of archaeological and historical contexts. In
Chapter 4 we will ask how hegemonic strategies work to ‘hold things
together’ in complex polities by articulating materially interdependent
social elements and forces. In particular, the intersection of politics,
Orientations 7
gender, production and material culture will be examined in relation

to both Classical Athens and the Inca Empire. In Chapter 5 we will
consider more directly the practices and strategies of sovereignty along
a continuum defined by the extremes of spectacle and routine. Here the
practices and contexts of water management and water ritual in Classic
Mayan polities will serve as a case study illuminating how spectacle
and routine form two sides of a single, performative, coin. In Chapter
6 we will look at the routinised violence embedded in the funerary
spectacles of the Royal Tombs of Ur as a final case study linking
together the main themes of this book. Finally, our concluding chapter
will reflect on the necessity, potential and problems of a comparative,
politically engaged, archaeology of sovereignty.
1
After (neo-)evolution(ism)
… the state is neither a subject nor a thing. So how could the state
act as if it were a unified subject, and what could constitute its unity
as a ‘thing’? And how do social actors come to act as if the state were
a real subject or a simple instrument?
Bob Jessop (2007: 3)
In the epigraph hovering just above this line, Bob Jessop lays out our
central problem. If states are not things, but the effects of practices,
discourses and dispositions, how and why do these effects generate
an apparent collective agent or instrument? In this chapter I want to
explore this problem by first making it apparent and then specifying
it more closely.
One place to start is with the critique of neo-evolutionism, that
strand of processual archaeology where the state always seemed
best loved. Vigorous critique of neo-evolutionist state theory has
been on-going for at least 30 years, and indeed has increased even
as its target has withered (e.g. Chapman 2003; Lull and Micó 2011;
Pauketat 2007; Pluciennik 2005; Smith 2003; Yoffee 2005. For refor-
mulations of neo-evolutionism see Marcus 2008 and Prentiss et al.
2009). If one could find a single word to characterise the central
object of this critique it would probably be ‘unity’, or perhaps ‘totality’.
Despite its diversity, neo-evolutionist state theory generally presumed
that societies are wholes or totalities that can conform to a limited
number of social forms (baupläne or blueprints) defined by key
traits realised across time and space. Such social forms were also
presumed to resolve themselves into singular trajectories (e.g. band,
tribe, chiefdom, state), determined by adaptive responses to given
stimuli. For critics, this unitary view denied history, and especially
the history of colonialism, by categorising contemporary tribal and
peasant communities as fossilised ancestors (Fabian 1983; Pluciennik
2005). The presumed unity of state societies ignored variability
in their form and institutional structure (Chapman 2003: 88–100;
Pauketat 2007: 133–62), while masking class and gender distinctions
(Paynter 1989; Pyburn 2004), factional competition (Brumfiel 1992)
and heterarchical arrangements of power (Ehrenreich et al. 1995)
that seemed to be key dynamics in actually existing states. While it is
true that in the 1980s a focus on elite strategies, prestige goods and
repressive power did develop (e.g. Brumfiel and Earle 1987), the ‘total’
nature of elite power in these models meant that the state remained
unitary, if now cast as Hobbes’s Leviathan rather than Hegel’s divine
idea. The presumption of unitary states also ignored the actual,
discontinuous, spatial reality of complex polities (Smith 2003; Smith
2005). Critics have noted that political power is embedded in specific
practices, buildings, settlements and landscapes, rather than abstract
and homogenous territories (Smith 2003; Smith 2005). Political power
is also irregularly distributed, making states more akin to networks
than objects (Smith 2005).
Singular, universal trajectories, such as the necessity of chiefdoms
preceding states, were criticised as empirically unsupportable in certain
regions (Yoffee 1993; McIntosh 1999), not to mention teleological and
ahistorical in general (Kohl 1984; Pauketat 2007). For Darwinians, this
teleology showed that neo-evolutionism was not evolutionary enough
(Dunnell 1980; Leonard and Jones 1987). In particular, its directional
nature and its lack of a clear unit of inheritance, as well as clear mecha-
nisms of variation and inheritance, meant that neo-evolutionism
treated social evolution as something distinct from biological evolution
in its processes.
The critique of neo-evolutionist state theory has opened up the
state concept in archaeological analysis. In particular, it is no longer
presumed to be a single package, binding together political structure,
population, economic production, social differentiation and social
After (neo-)evolution(ism) 11
hierarchy. It is now commonly acknowledged that polities named as

states can show a considerable range of institutional arrangements
(e.g. Blanton 1998), that they are characterised by internal conflict
and stratagems, and dependent on the contingent social and cultural
practices of specific agents (e.g. Flannery 1999). However, disassem-
bling this package has also highlighted several unresolved conceptual
problems, always latent in neo-evolutionist state theory.
Most pressing is not so much the definitional question of what
the state is, as the ontological question of how states exist. While
recognising that the state is not simply, or obviously, a singular entity,
exactly how specific states are constituted by the practices, discourses
and strategies of specific human beings remains assumed, rather than
addressed, in most of the critiques of neo-evolutionism.
For example, Norman Yoffee’s ‘new rules’ of social evolutionary
theory do an admirable job of reinventing social evolution as a kind of
large-scale comparative history (Yoffee 2005: 180–95). His analytical
distinction between states as centralised political apparatuses and
civilisations as historical configurations of elite cultural production
(Yoffee 2005 17–19; Baines and Yoffee 1998; 2000) raises a number of
interesting questions regarding the relationship between class, political
power and cultural practices. Furthermore, Yoffee shows an acute
awareness of the diversity of states, both internally in terms of the
competing interests and conflicting roles they contain, and externally
in terms of their institutional arrangements and social configurations.
However, much as I have done in the preceding sentence, he continues
to treat the state as a unified subject (e.g. states do, or have, X, Y and Z)
without exploring how this subjectivity is constituted.
Similar is Robert Chapman’s (2003: 95–9, 193–6; see also Lull and
Micó 2011) attempt to cut the Gordian knot of the state by defining
it in classic Marxist terms (e.g. Lenin 1976 [1917]) as simply the legal
and coercive instrument of class dominance. Hence, for Chapman, the
state arises virtually in tandem with differences in access to the means
of production (e.g. ownership of land). While this approach turns
our attention very directly to concrete social relations, rather than
abstract taxonomies, it fails to explain just how the state is formed as an

instrument and deployed as a thing (i.e. as a tool of dominant classes).
Recent applications of network analysis have, to some extent,
‘fragmented’ the state by modelling premodern states as networks of
practices, power and resources that are concentrated in nodes linked
by pathways (or ‘edges’ in graph theoretic terms), rather than as
continuous sovereign territories (Smith 2005; Mizoguchi 2009). Yet
in the end such networks remain constituent of an entity (the state)
with little reflection on how this entity coheres and persists when
constituted by a latticework of distinct people, places, resources and
activities.
One might think that evolutionary archaeologists of a self-
consciously Darwinian bent, with their intense focus on the evolution
of specific cultural traits (e.g. O’Brien et al. 2010), would be well-
positioned to take up the challenge of displacing the state with the
analysis of practices, strategies and technologies. Yet here we encounter
some interesting problems. The reproduction of any complex cultural
effect, such as the state, cannot be analysed atomistically in terms of
an organism or a trait. Such phenomena are inherently relational and
hence dependent on the continued articulation of multiple, mutually
dependent, practices (cf. Kuper 2000; Rosenberg 2009). Furthermore,
these practices are reproduced in a niche of such intricate construction
that both the proximate and ultimate causes of selection are virtually
identical to specific historical sequences; essentially Yoffee’s social
evolution as comparative history (Yoffee 2005: 192–5). Such issues
may explain the paucity of studies approaching complex polities from
a ‘meme’s-eye view’. Indeed, the most interesting trait studies do not
concern themselves with state-formation per se; rather they deploy
concepts and methods derived from evolutionary biology to explore
the dynamics of given traits within a framework that already presumes
the state’s existence (e.g. Glatz et al. 2011).
Most commonly, when state-formation is directly addressed in
evolutionary terms, emphasis shifts from trait to ‘group’ or ‘multi-
level’ selection. In biology, multi-level selection examines conditions
under which traits that improve the reproductive fitness of a group

might spread via natural selection, even when such traits decrease the
reproductive fitness of specific individuals (Wilson and Sober 1994).
In this view, selection can occur at any level (gene, cells, organism,
population) depending upon the relationship between within-group
and between-group variation and competition.
Unfortunately, when multi-level selection is applied to complex
human polities, all the familiar mystifications seem to return. If groups
are treated as replicators (i.e. what is being reproduced and selected)
then a kind of neo-neo-evolutionism emerges, where the state as a
bauplan for political organisation is given coherence and substance as
the analogue of an organism or species (e.g. Rosenberg 2009; Spencer
and Redmond 2001). When the emphasis is placed on groups as vehicles
for the selection of what has been called ultrasocial human behaviour
(e.g. Richerson and Boyd 2001; Turchin 2011), human groups and their
organisational ‘levels’ are treated as already given, stable and unprob-
lematic, rather than emergent, contextual and problematic (see Latour
2005: 27–42). Similarly, human agents are modelled as behaviourally
one-dimensional and unusually consistent (e.g. altruists/prosocial
versus cheaters/‘free-riders’). None of this sheds much light on Jessop’s
problem of how and why the state becomes a non-material ‘thing’.
A key issue is the lack of coherent, theoretically encompassing
alternatives to the state when it comes to discussing collective politics,
complex polities and large-scale political domination. For those of
a post-structuralist bent this may be as it ought to be. The point of
deconstruction is not to replace one reductive, totalizing framework
with another, but to undermine such frameworks altogether. Hence,
the appropriate heir of a totalizing concept like the state is not an
alternative totality, but rather an infinite number of local possibilities.
Be that as it may, this shattering of the unitary state concept is not
what has actually happened over the past 30 years. Indeed, regardless of
the theoretical predilections of individual scholars, whenever archae-
ologists have turned their attention to large-scale political domination
the state, or something very much like it, has sneaked back in.
Even scholars with impeccable post-processual credentials, engaged in

projects with explicitly anti-essentialist objectives, still end up framing
their studies in terms of large-scale reified polities.
Lynn Meskell (2002), in her book Private life in New Kingdom
Egypt, must presume an entity called ‘New Kingdom Egypt’ in order to
imagine the space that makes possible the very concept of an ancient
Egyptian private life. In his book Art and the early Greek state, Michael
Shanks (1999) addresses the concept of the state more directly, but
he does so in order to hold it constant as a backdrop against which
are played out figurative and literary discourses on status, sexuality,
masculinity and violence. One could argue that in each of these cases
the authors are not perpetuating a universal social form, but merely
reproducing indigenous categories (i.e. the land of Egypt, or the Greek
polis). Yet this in itself implies some rather interesting questions. Why
has it been necessary for ancient authors, modern neo-evolutionists
and post-modern critics to imagine large-scale politics in terms of
reified entities that literally do not exist in any physical sense? More
to the point, we might echo Jessop (2007: 3) and ask how it is that
historical actors come to view such entities as if they did have a
physical existence. It seems that we face something of a paradox.
How do we account for the apparent coherence of large-scale political
domination without returning to some kind of unitary state?
Adam Smith’s (2003) approach to these problems is distinct and
worth considering for the manner in which he moves decisively
beyond post-structuralist refusal. Smith (2003: 84–7) follows contextu-
alist historians (e.g. Skinner 1989) in noting that the semantic meaning
of the state, as a doubly impersonal edifice distinct from both ruler
and ruled, has a distinctly modern history linked to political develop-
ments in Europe since the sixteenth century. The state, therefore, is a
historically specific concept presented as a natural object for political
ends. Smith argues that the retroactive application of this concept to
archaeological contexts provides an inadequate analytical account of
past political practices. At best, it turns the analysis of actual political
processes into a taxonomic exercise. At worst, projecting the state
into premodern contexts extends and naturalises asymmetrical power

structures in the present day. In this way archaeologists fall victim to
what Pierre Bourdieu (1999: 54) describes as ‘a state, which still thinks
itself through those who attempt to think it …’
Like many other scholars, Smith follows his principled rejection of
the state concept with the introduction of an alternative placeholder,
something he terms ‘Early Complex Polities’ (Smith 2003: 102–5).
However, the strength of Smith’s argument lies in his unwillingness to
leave the reification of Early Complex Polities unquestioned. Hence,
while he defines such polities loosely in terms of traits that are indis-
tinguishable from those of the ‘Archaic State’ (Smith 2003: 102–5) and
designates them as the ‘class of objects under examination’ (Smith
2003: 102), he immediately shifts his attention from traits to ‘the
relationships that constitute the political sphere of action’ (Smith 2003:
102).
For the purpose of analytical precision, Smith describes these
relationships in terms of four levels, moving from the relationships
between polities to the relationships between regimes, subjects and
institutions. However, if we step back from Smith’s organising scheme
we realise something fundamentally important about the analysis of
collective politics and complex polities. What is under examination is
not a class of objects named Early Complex Polities, nor is it even an
ensemble of distinct objects such as regimes, institutions or subjects.
What Smith examines are the relations, practices and discourses
that constitute these objects as effects. Taken together, these effects
constitute what Smith terms political authority, and it is this consti-
tution of political authority that he places at the heart of any analysis
of politics in early complex polities (Smith 2003: 105).
For Smith, political authority is authority, meaning the effect of a
‘posited, perceived, or institutionally ascribed asymmetry between
speaker and audience’ (Lincoln 1994: 4, as quoted in Smith 2003: 106),
extended to be the authority of last resort in a given territorial context.
The analysis of political authority, then, is the analysis of ‘… how, in
varying sociocultural formations, an authoritative political apparatus
came to gain varying degrees of ascendancy over all other social

relations’ (Smith 2003: 108).
As is well known, not all communities or societies have been struc-
tured by political relationships in which some person or institution
has the ‘last word’ in the sense of making decisions that are authori-
tative and binding for all members (see Clastres 1977; Graeber 2004).
Smith, therefore, continues to recognise that there are important
differences in the kinds of political relationships that exist and have
existed within human societies. Such differences can form the basis for
meaningful comparative study and even a renewed engagement with
social evolution. However, Smith’s focus on political relationships, and
the practices and settings through which they are constituted, also
implies a shift in how archaeologists are to approach such comparative
study. In particular, it turns our attention away from the supposedly
inherent traits of an entity or system and towards the actual socio-
spatial networks through which power is constituted (cf. Mann 1986:
1–18).
This said, political authority on its own does not provide all of the
tools that we need to understand large-scale political domination.
First, despite the centralising logic of political authority, no political
apparatus ever gains ‘varying degrees of ascendancy over all other social
relations’. As Michael Mann notes, if we understand power in terms of
socio-spatial networks of relationships, then these networks always
contain interstices (Mann 1986: 16). Whether in the form of rebellion,
criminality, token submission or the tacit recognition of domains
where it is never asserted, political authority always encounters limits
in its realisation (Hansen and Stepputat 2006). These limitations can be
practical and technological, but they are not accidentally or exclusively
so. Quite simply, the limits of political authority are inherent in the
possibility of difference and of resistance.
As the literature on heterarchy reminds us, even at its centre
political authority can be multi-centric, dispersed in multiple locales,
or alternatively strong or weak depending on the context and issues at
stake (e.g. Ehrenreich et al. 1995). What is significant about political
authority is not that a political apparatus does, or does not, ‘gain

ascendancy over all other social relations’, but rather that the promise
and potential of that ascendancy is produced and reproduced despite
its effective absence in specific contexts. This implies at least two
things: one is that political authority needs to be imagined or repre-
sented as continuous even though it is not; the other is that political
authority in the form of a political apparatus (be it one person or a set
of institutions) cannot stand alone, but must be linked to other social
forces, interests and orders in a complementary manner if it is to be
imagined as continuous (see Chapter 4).
One way of reproducing political authority at the limits of its reali-
sation is by means of force. Certainly, one must presume that the threat
of coercive force, be it physical or symbolic (see Bourdieu 1999: 56),
underlies the ‘last resort’ in Smith’s definition of political authority.
Relations of force are, of course, very much part of classic state theory.
Here we find two distinct emphases. On one hand, there is a concern
for force directed outwards via warfare, both as a state function
(Clausewitz’s ‘continuation of policy by other means’) and as a catalyst
to state formation (historically and organisationally). In this view, to
quote Charles Tilly (1975: 42), ‘wars made the state and the state made
wars’. On the other hand, there is a concern for force directed inwards
via legislative, judicial and police powers. This is Max Weber’s (1978:
54) state as a political organisation claiming a monopoly on legitimate
violence within a given territory. Both sorts of relations of force stand
at the limits of political authority (i.e. the point where consent falters)
and are therefore interesting for the ways in which they are kept
distinct. Put simply, warfare is distinguished from law and policing by
structures of identity (e.g. interior/exterior; us/them) and concepts of
legitimacy in the uses of violence.
Once linked to political authority, warfare is undeniably important
as a means of acquiring and defending resources. It can also serve as
a justification, even foundation, for authority itself (Agamben 2005).
However, as collective violence enacted across communal/identity
boundaries, warfare is not strictly dependent on political authority
in Smith’s terms. Warfare occurs between communities lacking any

authority of ‘last resort’ (Keeley 1996). Indeed, Pierre Clastres (1977)
went so far as to argue that, in the Amazon basin, continual warfare
between communities was a tactic that prevented chiefs from accumu-
lating resources and authority within communities. So, whereas
Thomas Hobbes claimed that life outside the state was a ‘warre of
every man against every man’ (Hobbes 1968 [1651]: 188), it seems
that the real problem is within the state, where warfare must be distin-
guished from the internal deployment or threat of coercive force. As
Weber recognised, this makes the question of legitimacy central to the
question of political authority. So, in addition to analysing the practices
and relationships that constitute political authority, we also need
to give attention to the practices and relationships of coercive force
(both physical and symbolic; see Bourdieu 1999) and the problems of
legitimacy these entail.
Smith (2011: 416) has in fact recognised the link between political
authority and coercive force in his more recent work, where he now
employs the term ‘sovereignty’ to refer to relations of political authority
grounded in violence (following Agamben 1998 and Hansen and
Stepputat 2006 in generalising Foucault’s understanding of premodern
sovereignty – on which, see pp. 21–2). However, one final problem
remains. While Smith clearly avoids the error of starting from reified
entities like the state, such entities are still constituted, both abstractly
in his analytical framework (e.g. Early Complex Polities) and histori-
cally in actual political formations (e.g. the kingdom of Urartu). Smith
gives ample attention to the spatial practices of this reification in his
discussion of polities (Smith 2003: 149–83). However, it is not clear
from his conceptual framework why such reification needs to occur
in the first place. Why must political authority, as the authority of last
resort, be experienced and analysed in these virtual terms?
De-centring the state
A starting point is to note that the modern state is one example of

political authority realised in virtual terms and that we can learn a
good deal from attempts to critically analyse its virtual nature. Smith
(2003: 96–7) correctly notes that in widely cited articles both Philip
Abrams (1988) and Timothy Mitchell (1999) have explored ‘the senses
in which the state does not exist’ (Abrams 1988: 82) as an entity unto
itself. However, it is important to note that the arguments of each
scholar go beyond the uses that Smith has made of them. Indeed, these
critical strands of state theory seek to de-centre the state: not simply
to note that it does not actually exist, but rather to examine how the
presumption of its existence has real material effects.
Abrams (1988: 73), for example, concedes that social cohesion is
one function of politics in class societies. However, he argues that if
one says that this cohesive function is what the state provides, one
has already assigned a misplaced concreteness to the state, thereby
simplifying and mystifying actual political practice. Instead, Abrams
distinguishes between the state idea and the state system. The state idea
refers to a discursive construct (‘the state’) that:
… presents institutionalised power to us in a form that is at once

integrated and isolated and by satisfying both these conditions it
creates for our sort of society an acceptable basis for acquiescence. It
gives an account of political institutions in terms of cohesion, purpose,
independence, common interest and morality without necessarily
telling us anything about the actual nature, meaning or functions of
political institutions. (Abrams 1988: 68)
The state idea, therefore, attributes ‘… unity, morality and independence

to the disunited, amoral and dependent workings of the practice of
government’ (Abrams 1988: 82). These practices of government are
what Abrams refers to as the state system and they are constituted by
the disparate ensemble of institutions, structures and agencies that
act to ensure the continued domination of key sectional interests in
modern capitalist societies. For Abrams the state idea is necessary

and real because without it the interested and illegitimate nature of
the domination achieved through the state system would be exposed.
Hence, his widely quoted statement: ‘In sum, the state is not the reality
which stands behind the mask of political practice. It is itself the mask
which prevents us seeing political practice as it is’ (Abrams 1988: 82).
Timothy Mitchell takes up Abrams’s argument, but notes that
distinguishing the state idea from the state system is not as easy as
Abrams implies. This is because, without the state idea, the practice of
government has no distinct location, being simply the social practices of
disciplinary power in general. For Mitchell, there is no point in distin-
guishing the real and the illusory aspects of the state, since the central
question for the modern state is how and why the mundane practices
of disciplinary power constitute the state idea as an abstract form;
dividing state, society and economy as if these were discrete structures,
agents or things. As Mitchell (1999: 77) notes: ‘the phenomenon we
name “the state” arises from techniques that enable mundane material
practices to take on the appearance of an abstract, nonmaterial form’.
Following on from Mitchell, we can see that there is an intimate
relationship between the state idea and the material practices of power
in modern states. As Mitchell suggests, the state idea is constituted
and given substance by repeated, and seemingly mundane, practices of
regulation, monitoring, definition and enforcement. At the same time,
the state idea does more than mask or justify the workings of power, as
it orients and articulates the state system in the moment of practice. By
orients I mean that the state idea allows people to act in the name of,
or with regard to, the state. This serves to remove their actions from the
realm of everyday human relations and place them in a virtual realm
of relations between citizens and the state. By articulating I mean that
the state idea provides the political cohesion necessary to align social
interests and forces in emergent configurations. In other words, the
state idea provides what Foucault (2007: 286) termed the ‘principle of
intelligibility’ for governmental practices.
It may seem odd to turn to Foucault in discussing the idea of the
state, especially in a book that is ultimately concerned with premodern

contexts. Foucault was well known for his apparent disinterest in state
theory, even stating: ‘I do, I want to, I must do without a theory of the
state, as one can and must forgo an indigestible meal’ (Foucault 2008:
76–7). Furthermore, for some, Foucault’s concept of ‘governmentality‘
as a series of techniques, or arts, of government predicated on the
management of populations as entities represents a radical break, both
conceptually in relation to classic state theory and historically in its
distinctly modern point of reference. However, the recent publication
(Foucault 2007) of the wider programme of lectures within which
Foucault developed his ideas on governmentality makes clear that it is
incorrect to suggest (as I did in Routledge 2004: 16–17) that he did not
engage with the concept of the state (see Jessop 2007: 140–54; Lemke
2007).
Foucault does argue that the modern era (especially from the late
eighteenth century) is marked by the recognition of human population
as something with its own mechanisms and qualities that can be
facilitated and managed through practices of governance that are not
limited to a central political apparatus (e.g. insurance, opinion polls,
public health, etc.). However, he does not oppose this governmentality
to the state; indeed, he argues that the reflexive recognition of the
state as an abstract category emerges in tandem with these new arts
of governing a population, serving as their principle of intelligibility
(Foucault 2007: 247–8, 276–8).
Importantly for our purposes, Foucault also acknowledges that the
idea of the state was ‘a way of thinking the specific nature, connec-
tions, and relations of certain already given elements and institutions’
(Foucault 2007: 286). In other words, these transformations in both
the arts of government and their principle of intelligibility occurred
within an already existing political context. Foucault, for example,
acknowledges that ‘the state as a set of institutions of sovereignty
has existed for millennia’ (Foucault 2007: 120) and that the problem
of sovereignty both predates the idea of governing a population and
continues alongside governmentality as a pressing concern (Foucault
2007: 106–8). For Foucault (2007: 11), sovereignty refers to the ability
to command and forbid in a binding manner within a given territory.
In short, it is very similar to Smith’s definition of political authority.
What changes between the late sixteenth and late eighteenth centuries,
therefore, is the techniques through which sovereignty is constituted
and manifest (population management as opposed to law and force
alone) and the principles through which these techniques are made
intelligible (e.g. the state and raison d’État as opposed to divine
right and God’s law). Stated throughout Foucault’s lectures, but left
undeveloped, is the fact that premodern political authorities and their
institutions of sovereignty were also dependent on specific techniques
and practices with their own principles of intelligibility.
To summarise; the modern state provides us with an example
of virtual political authority because it rests on specific practices of
governance that lend substance and effect to the concept of the state,
even as that concept renders these practices intelligible (by providing
orientation and articulation). Both the concept of the state and at least
some of these practices of governance may well have been inventions
of the modern era, but they were not de novo inventions. The modern
state as a doubly impersonal edifice, at once secular and transcendent,
displaced earlier constructions of political authority that were equally
transcendent, if neither impersonal nor secular (e.g. the divine right
of kings).
Think for a moment of the ancient Greek polis; while it could
be represented as an abstract category, it could never be viewed as
something distinct from the political community itself (i.e. it was
society and hence could not govern it – see Anderson 2009 and
Hansen 1998), and hence is very different from the state idea (see
Chapter Four). At the same time, the imagining and experiencing of
the collective will of the polis as a form of communal agency, one that
was constituted in articulation with key social interests and forces (e.g.
slavery and patriarchy), formed a kind of virtual political authority that
would both match the definitions of Smith and be amenable to analysis
along the lines of those proposed by Mitchell.
Admittedly, for most monarchies even this level of abstraction

was rare, with explicit political theory generally replicating Louis
XIV’s ‘l’état c’est moi’. Yet, even in the most personal forms of political
authority we find a virtual element that suggests that we might learn
something of general significance from the critical analysis of the
modern state.
Political authority in Pharaonic Egypt, for example, was quite clearly
and exclusively invested in the Pharaoh. However, both the Pharaoh
and his (or her) subjects were transformed in rather interesting ways by
a number of material practices. Best known, of course, is the deification
of the Pharaoh as the embodiment of Horus while alive and Osiris
when dead. In this role, the Pharaoh was charged with maintaining
ma’at (‘right order’) on earth, keeping the forces of chaos at bay
(Teeter 1997). Yet, while ancient Egyptians never articulated a formal
distinction between office and office-holder, it is clear that this deifi-
cation was not attached to the person of the Pharaoh as an individual,
but rather to his (and occasionally her) role as king (O’Connor and
Silverman 1995: xxv). Indeed, a large number of material practices
served to transform the nature of each Pharaoh on his (or her) ascent
to the throne. The title Pharaoh (pr ’3) itself means ‘the great house’
(i.e. palace), hence the king as Pharaoh was not simply an individual
but embodied in himself or herself the palace and its household estate
(O’Connor 1995). By the New Kingdom this title could be used in both
ritual and administrative documents without reference to the name of
the sitting Pharaoh, and hence could be ascribed an abstract quality,
especially within scribal and artistic genres that spanned multiple
centuries. Conventions for marking-off the Pharaoh’s throne name,
such as enclosing it in a cartouche or always following it with the
phrase ‘life, prosperity and health’, had this same historical quality such
that their repetition for successive Pharaohs over long periods of time
served to assimilate each of those individuals to the broader category
of Pharaoh. Indeed, even the adoption of throne names and titles itself
served to mark the transformation of the heir into something other
than merely human on ascension to the throne.
The subjects of the Pharaoh were also defined by material practices

through which they were collectively bounded and differentiated
from those that surrounded them. In contrast to the expectations of
scholars such as Anthony Giddens (1985), the Egyptian state seemed
to have a concept of its geographical boundaries. For example, in a
recently discovered inscription from Zawiyat er-Rahman, a fort on
ancient Egypt’s Libyan frontier, the commander Neb-Re takes the title
‘one who brings an end to the transgressors of his boundary’ (Steven
Snape, personal communication). As is well-known (e.g. Leahy 1995),
non-Egyptians are portrayed visually via a limited number of stereo-
typical features (hair and beard style, clothing, skin colour) that serve
to identify them with a limited set of foreign identities (e.g. Nubian,
Libyan, Syrian, Hittite, etc.) and contrast them with native Egyptians.
Usually, such foreigners appear as props in scenes of triumphant
xenophobia (e.g. head-smitings, bound prisoners, tribute bearers).
Even more subtle, however, is the visual nature of hieroglyphs, which
makes it impossible to write the words for male or female human being
without depicting a specifically Egyptian person (see Goldwasser 2002:
21–2).
These, and many more, material practices served to transform
political authority in ancient Egypt from an asymmetrical relationship
between persons (and hence subject to mutual recognition between
participants) into a virtual relationship between the Pharaoh and his
(or her) subjects. The virtual nature of this relationship meant that it
was defined categorically and a priori, rather than being dependent
on the nature, quality and context of the relationship between specific
participants. This provided a frame of reference within which political
authority was released from constraints that might otherwise limit its
ability to compel. For example, one did not need to interact directly
with any specific Pharaoh in order to know the power and authority
of the Pharaoh. More pointedly, it would release political authority
from cultural and social restrictions on the contexts in which, and
degree to which, physical and symbolic coercion could be enacted.
Certainly, relations of dominance, including physical violence, are
frequently sanctioned between genders and ages or within kin groups

in otherwise ‘egalitarian’ societies. However, such sanctions are made
possible by their limits, that is to say by the contexts in which open
coercion is denied and hence gender- or age-based identities are
differentiated and differentially valued (e.g. the oxymoron of ‘egali-
tarian’ patriarchal communities). The virtual relationships of political
authority are fundamentally different in this regard because they
recognise no such limits on the possibility of coercion in the last
instance (Agamben 1998; 2005).
Yet, even as transcendence frees authority from moral constraints
on violence, it also allows that authority to be delegated, responsive and
discontinuous without surrendering its claims to being centralised and
monolithic. For example, although Inca rulers were ostensibly the sole
leaders in war, religion and politics, in practice they included under
their personal exploits the achievements of various surrogates, such as
generals, the high priest of the Sun, and administrative officials called
the ‘Inca’s substitute’ (Incap rantin). Hence, in the words of Peter Gose,
‘… the Inka became a kind of umbrella figure or corporate persona,
who hierarchically subsumed under his identity an entire supporting
cast of lesser individuals’ (Gose 1996: 17).
My description of virtual political authority shares much in common
with Maurice Bloch’s arguments regarding the role of ritual in consti-
tuting royal authority as transcendent in relation to everyday life
(Bloch 1987: 271–4). However, I see no need to limit the constitution of
such transcendence to ritual action, unless defined in its broadest sense
as habitual practice. I also see no need to limit transcendence to specifi-
cally royal (i.e. personal) political authority. Certainly, the modern
state is constituted as a transcendent idea (e.g. ‘distinct from both ruler
and ruled’), one that transforms actual relations between people into
virtual relations between citizens and the state. Hence, transcendent
sovereigns and sovereign states are not the same thing, but they are
different manifestations of the problem highlighted by Jessop at the
beginning of this chapter.
Summation
In this chapter I have argued that what we frequently call state-

formation entails not the formation of an entity, but the configuring
of relationships around political authority made transcendent and
grounded in violence. In the remainder of this book I will follow Adam
Smith (2011) and others in calling this configuration ‘sovereignty’,
with the understanding that sovereignty entails the political effect of
transcendence, as much as it does authority and violence. It remains
to explore how and why sovereignty, understood in these terms, has
been achieved in different times and places – a task that will occupy
our attention for the remainder of this book.
2
Coercion and consent
The pack of scoundrels tumbling through the gate, emerges as the

Order of the State.
Stanley Kunitz, The System (1971)
Sovereignty is easy enough to imagine, but more difficult to imagine in

a realistic manner, one in which a master plan is not evident, in which
specific people applaud, resist or ignore specific projects, in which
necessary resources are not automatically present, and in which the
material interests of potential rulers, officials and subjects may be in
direct conflict with one another. Why should practices and strategies
cohere in this manner, often against all odds and against the material
interests of at least some of those involved?
Order, legitimacy and wealth
One innovative attempt to come to terms with such issues is John

Baines and Norman Yoffee’s (1998; 2000) comparative account of the
relationship between power and culture in early Egypt and Mesopotamia.
Baines and Yoffee contend that in both of these early civilisations it is
precisely the civilisational component that must be grasped if one is
to understand the production and reproduction of political authority.
In particular, Baines and Yoffee argue that political authority in early
states is embedded in a matrix of what they refer to as ‘high culture’.
They argue that in both Egypt and Mesopotamia the cultural activity
of an inner elite created dominant systems of meaning that hinged in
particular on the imagining of an encompassing cosmological order.
Order in this sense provided a single intellectual framework capable

of incorporating religious, social, political and economic life. As such,
it made the actions, duties and existence of kings and other elites
central to the stability and continuity of the cosmos itself, in so far
as they were called upon to maintain or renew order through ritual,
adjudication or force. This, of course, served to legitimate specific
dynasts, but it also provided definitions and criteria for legitimacy
independent of individual regimes and hence available for activation
beyond the temporal and/or spatial limits of any given regime. Over
time, the reproduction of a common cosmological order could provide
a consistent framework for reviving political authority and political
centralisation despite the regular collapse of specific regimes.
For example, in the case of Pharaonic Egypt, the reproduction of
‘high culture’ as the matrix of political authority allowed the revival
of centralised rule expressed in symbolically consistent forms after
periods of political decentralisation (i.e. after the First,, Second and
Third Intermediate periods). Notably, it even provided a literary and
visual language of authority for rulers such as the Ptolemies, who
were explicitly non-Egyptian in their ethnicity. In the case of early
Mesopotamia, a regional ‘high culture’ provided a common matrix
for political authority, both between competing city-states and within
city-states, where distinct temple and palace institutions were simul-
taneously semi-autonomous and mutually engaged in ruling and
administering the city’s population, resources and territory.
For early China, Kwang-chih Chang (1983: 9–33, 107) has argued
that a shared ideology of clan hierarchy based on divine descent
provided a mythological framework within which different clans could
legitimately compete for rulership. Evidence for early centralisation
in the Erlitou and Erligang phases (beginning of the Bronze Age in
northern China), that stands in contrast to the later segmented and
competitive political landscape of Late Shang and the remaining ‘Three
Dynasties’ period, suggests that this segmented political landscape
should perhaps be seen more historically and less structurally than
Chang implies (see Liu 2009). Regardless, Baines and Yoffee’s focus on
Coercion and consent 29
cosmological order is rather helpful in understanding the continuity

to be seen in the reproduction of political authority in Late Shang and
Zhou-period China despite the shifting fates of specific regimes and
lineages.
In this context, the amassing and utilisation of wealth was a key
component in materialising, celebrating and reproducing the cosmo-
logical order. This occurred directly through temple and funerary
ritual, as well as military campaigns. It also occurred indirectly through
the production of buildings, monuments, spectacles and objects that
overwhelmed and/or enchanted the senses. Secondarily, access to, and
knowledgeable consumption of, the products of ‘high culture’ also
served as an identity token, forging shared experiences and a cohesive
worldview amongst the ‘inner elite’ of early Egypt and Mesopotamia.
Hence, economic production, as the production of wealth, was not an
end in itself that needed justification; rather it was the means to an end
(i.e. the production and reproduction of order) whose justification lay
in the perceived legitimacy of that end.
Understanding the constitution of political authority in terms of
a wider order that is not contained or produced within the limits of
a governmental apparatus is one of the more insightful aspects of
Baines and Yoffee’s broad comparison of Egypt and Mesopotamia. This
allows us to understand historical continuity in the representation
of authority in spite of the rise and fall of specific regimes. It also
highlights that polity formation incorporates a cultural logic, and not
merely the adaptive logic of managerialism or the exploitative logic of
political economy. Hence, order and its legitimation are elaborated,
believed in and pursued beyond what is strictly necessary in order to
organise a population or mask exploitation.
However, as a number of commentators have pointed out (see
Richards and Van Buren 2000), Baines and Yoffee’s approach is severely
limited by its strong emphasis on the exclusive and exclusionary
nature of ‘high culture’. While the problem of non-elite integration
is continually raised in both of Baines and Yoffee’s papers (Baines
and Yoffee 1998: 232–3, 240, 246; 2000: 15, 16–17), they repeatedly
conclude that not only is there little evidence for the participation of
non-elites in the ‘high culture’ of Egypt and Mesopotamia, but there is
also little evidence for the materialisation of any alternative ideologies
(Baines and Yoffee 1998: 240). Apparently, in both early Egypt and
Mesopotamia the alternative to ‘high culture’ was no culture. Not
surprisingly, this position has been resisted by scholars otherwise
sympathetic to Baines and Yoffee’s approach.
Janet Richards (2000), for example, documents the development
of mass cemeteries in the First Intermediate period of Egypt and the
widespread distribution of amulets worn on the body that consistently
replicate the symbols and signs of Egyptian ‘high culture’ including
those from the supposedly exclusive domain of hieroglyphic writing.
Richards also points to an increased emphasis on, and dispersal
of, spatial (e.g. temples, processional ways) and material (e.g. stele,
statues) products inviting spectatorship of, or participation in, state
rituals and high cultural practices during the Middle Kingdom. This
is paired with the rise of literature concerned with moral order, and
the ‘private’ orientation of elites to this order (cf. Parkinson 2002). In
stories such as the ‘Tale of the eloquent peasant’, this moral order is
clearly extended to the treatment of non-elites, suggesting that such
persons presented moral and ideological problems for elites that would
be difficult to imagine in the strictly bifurcated terms used by Baines
and Yoffee (see Routledge 2003).
Norman Yoffee’s own study of law courts and conflict resolution in
early Mesopotamia (Yoffee 2000), as well as work on Old Babylonian
local councils by his student Andrea Seri (2005), points towards a
complex accommodation of, and interrelationship with, local forms
of authority on the part of central governments in early second
millennium bc Mesopotamia. Again, this would seem to contrast with
the rather one-sided picture of elite exclusivity painted by Baines and
Yoffee’s general comparative account.
More striking perhaps in its use of long-term archaeological
evidence is Rowan Flad’s (2008) recent study of the development of
oracle bone divination in early China from the late Neolithic through
the Shang and early Zhou periods. Evidence for osteo-pyromancy,

so named because the bones are burned to catalyse the cracking on
which divination is based, can be traced continuously from third
millennium bc Longshan sites through to early first millennium bc
Western Zhou sites (Flad 2008: 406–18). As is well known, during
the Late Shang period (late second millennium bc) many such oracle
bones (especially cattle scapulae and turtle plastrons) were carefully
prepared and inscribed as part of divination rituals closely linked to
Shang rulers. Divination in the Shang court was clearly an example of
what Baines and Yoffee would call ‘high culture’, a practice involving
elite specialists and closely linked to early Chinese concepts of order
and legitimacy. At the same time, these inscribed oracle bones are
preceded by a long history of less specialised, and widely dispersed,
osteo-pyromanic practices involving uninscribed bones. Furthermore,
even within Shang contexts, inscribed bones are far outnumbered by
uninscribed bones and it seems unlikely that such divinatory activities
were a state monopoly (see Campbell 2008; Jing 2008).
Shang osteo-pyromancy appears to be a case of what I have elsewhere
termed entrainment (see Routledge 2004: 185–90). By this I mean the
linking of cultural practices, with their own history and social distri-
bution, to political authority such that the former not only serves to
reproduce the latter, but is also transformed by it. This transformation
can occur both in the mechanics of the cultural practice itself and in
its indexical value as a sign of political authority. My original example
was the use of alphabetical writing in the Iron Age Levant, whose
Bronze Age invention and early Iron Age development is not effec-
tively analysed as an exclusive product of ‘high culture’ or the state. At
the same time, alphabetic writing does become linked to the function
and representation of kingship in the course of the Iron Age, and is
transformed as a result. Similarly, osteo-pyromancy has a long history
in China, and both its appeal to Shang rulers and the power implicit
in the specialised skill of royal Shang diviners, lay in the fact that
divination was embedded, and valued, in a wider cultural context. The
tension between royal co-optation and this wider cultural context for
divination is perhaps acknowledged in the fourth century bc text Kuo

Yü where it is explained that in the past everyone could communicate
between earth and heaven via gifted shamans (hsi/wu), but this broke
down when households began to perform their own religious rites and
humans and spirits freely intermingled, with calamitous results. To
restore order, free communication between heaven and earth had to
be severed and carefully managed by the royal court (see Chang 1983:
44–5).
While embedded in long-standing cultural practices, the
entrainment of osteo-pyromancy by Shang rulers resulted in develop-
ments such as the addition of inscriptions and specialised preparation
of the bone surface that presumably transformed what it meant to be a
skilled diviner. One need not presume, therefore, that Shang rulers had
to hold a monopoly on osteo-pyromancy in order to gain power from
their specialised skills in divination. Once cultural practices become
marked by their association with political authority, they can serve
as an indexical sign, calling to mind that authority in contexts where
political authority plays no direct role as either author or authoriser.
Rational choice?
In a recent, rather monumental cross-cultural survey Richard Blanton

and Lane Fargher (2008) present an argument almost directly opposed
to that of Baines and Yoffee, addressing what has been called ‘Hobbes’s
dilemma’ in the context of premodern states. ‘Hobbes’s dilemma’ is
summed up in the simple question ‘what holds a collective (such as a
state) together given the tendency for human beings to pursue their
own self-interest?’ Blanton and Fargher suggest that this question
cannot be answered in the uni-directional terms of elite domination.
They apply rational choice theory to argue that, as rational agents,
subjects in premodern states (‘tax-payers’ in their terminology) would
have actively negotiated and agitated for the provision of key ‘public
goods’ (e.g. security, adjudication, law enforcement, water supply,
transport routes, etc.), in return for the resources they surrendered to

Baines and Yoffee’s ‘inner elites’ (‘state principals’ in their terminology).
Such negotiated consent was necessary, as coercion alone would be too
expensive and ineffective a means of ensuring ‘tax-payer compliance’.
Hence, according to Blanton and Fargher, collective action, as the
negotiated trade-off of public revenues in return for public goods, is a
variable relevant to all states, modern or premodern. Using quantitative
analysis, they suggest that the degree to which a state is characterised
by collective action is determined largely by the degree to which ‘state
principals’ are dependent on extracting revenue from tax-payers rather
than from other sources (e.g. war booty, trade monopolies, royal
land-holding, etc.). Furthermore, Blanton and Fargher suggest that
states characterised by collective action would develop more extensive,
socially penetrating, bureaucracies in order to discover, address and
enforce public will and the provision of public goods.
Blanton and Fargher are certainly correct in arguing that the consent
of subjects plays a role in all political authority, and that this consent
cannot be presumed but must be continually elicited and reproduced
over time. The question is whether or not this consent is simply a
matter of balancing individual interests against the common good (or
a ruler’s ambitions against her subjects’ needs), as Blanton and Fargher
seem to suggest. Notably, Blanton and Fargher accept as given several
fundamental categories and relationships that, as we saw in Chapter 1,
become rather problematic on closer examination.
The state as an entity and the virtual relationship of ruler to
subject, with its accompanying right to both extract resources and
provide public goods, are presumed rather than examined by Blanton
and Fargher. For example, they do not ask why the state principals/
tax-payer relationship exists in the first place, nor how it is reproduced
over time.
While the diversity of interests present in any human collective
is central to ‘Hobbes’s dilemma’, Blanton and Fargher’s tax-paying
subject is an undifferentiated and homogenous rational actor. Yet,
certain of their key public goods, such as the enforcement of tax-payer
compliance and the prevention of ‘free-riding’, already imply that

tax-payer consent and the definition of a public good is not undif-
ferentiated, but rather is something that potentially divides subjects
on the basis of their interests (e.g. rich/poor; male/female; urban/
rural, etc.). Hence, consent as realised through the provision of public
goods incorporates (and to some extent is predicated on) an element of
physical or symbolic coercion (i.e. the ability to propagate a particular
understanding of the common good).
Finally, political subjects may be rational agents in terms of the
choices they make, but such choices are never infinite, hence ration-
ality is always relative to the choices that are available to any particular
agent. People do not necessarily consent to contribute corveé labour
to a state-sponsored canal project because they know it is the best
of all possible options for the management of water in their region.
Rather, choices are made relative to the immediate costs and benefits
of compliance. So, when speaking of consent, we need to consider the
conditions of possibility that any particular political subject encounters.
We need to ask how the choices available to different subjects related
to their different subject positions. Minimally, in Blanton and Fargher’s
analysis, ‘state principals’ and ‘tax-payers’ faced different choices in
premodern states and hence would have operated with different defini-
tions of rational action. The key question is how different kinds of
subjectivity were made rational within any given political configu-
ration (cf. Smith 2004).
What, then, of our original question? Baines and Yoffee have raised
a number of key points regarding the constitution of political authority.
Importantly, such authority is embedded in, and legitimised by, a larger
social (and frequently cosmological) order. Yet, Baines and Yoffee’s
exclusive focus on ‘high culture’ as the producer of this order leaves us
with a difficult dichotomy. It seems that we have only elite consent on
one hand and the bare coercion of non-elites (or worse – nothing at
all) on the other. In contrast, Blanton and Fargher have argued that, as
rational agents, the consent of subjects is essential to the constitution
of political authority in any context. Yet, as we have seen, assuming that
this consent is a simple matter of balancing the interests of individual

rational actors leaves unanswered most of the questions raised in
Chapter One regarding the virtual nature of political authority and the
intimate relationship it engenders between coercion and consent.
Thinking systematically about this intimate relationship is no easy
task. In particular, one needs relatively precise analytical tools and
concepts in order to understand how political authority can rest on both
coercion and consent, how it can serve both to dominate and to incor-
porate subaltern populations (who themselves may both accept and
resist political authority) and how it can simultaneously reflect, select
and shape cultural practices. Developing such analytical tools was a
central concern of Antonio Gramsci’s richly enigmatic political writings.
Hence, a brief foray into Gramsci’s thought will prove an essential
digression in our exploration of transcendent political authority.
Antonio Gramsci
Antonio Gramsci was an Italian intellectual and political activist who

played a leading role in the Italian Communist Party and was jailed
by Mussolini from 1926 until just before his death in 1937. Gramsci
is best known for the notebooks he kept while in prison, in which
he attempted to think through in very broad terms what would be
required to establish a proletarian state in Italy. In doing so, Gramsci
felt it was necessary to understand the general failure of working-class
revolutionary movements in the industrialised West, and the general
resilience of the ‘bourgeois’ state in the face of numerous political and
economic crises in the decade following the First World War. This
process caused him to give particular attention to the relationship
between the state, culture, leadership and power. Gramsci’s approach
to these issues charted a distinct path marked by his commitment to
the class analysis of orthodox Marxism on one hand and to historicism,
with its emphasis on concrete historical circumstances and human
volition, on the other (Morera 1990).
While this is not the place, nor am I the scholar, to present an

extended discussion of Gramsci’s thought, one can gain a basic appre-
ciation of his contribution to political theory through considering a
number of key terms to which Gramsci’s main ideas are linked. In
particular, these are ‘Hegemony’, ‘Intellectuals’, ‘Common Sense’ and
‘Historical Bloc’.
Hegemony, coercion and the state

For Gramsci, the key to understanding the state, both as something
to overcome and as something to perfect, was to consider not only
the administrative apparatus of a polity but also the social totality that
supported and reproduced that polity. As he writes:
… the State is the entire complex of practical and theoretical activ-

ities with which the ruling class not only justifies and maintains its
dominance, but manages to win the active consent of those over
whom it rules. (Gramsci 1971: 244)
For Gramsci, political society is the formal apparatus of government,

and especially the seat of coercive powers, while civil society is the appar-
ently private world outside of this formal apparatus, which none the
less is intimately linked to it. Gramsci expresses this relationship in the
simple formula: ‘state = political society + civil society, in other words
hegemony protected by the armour of coercion’ (Gramsci 1971: 263).
The organic relation between political and civil society is most
evident in constituting the consent of subjects to being ruled. For
example, Gramsci writes:
The State does have and request consent, but it also ‘educates’ this
consent, by means of the political and syndical associations: these,
however, are private organisms, left to the private initiative of the
ruling class. (Gramsci 1971: 259)
As a number of scholars have shown (Laclau and Mouffe 2001:

7–41), Gramsci borrowed the term hegemony from Russian Social
Democracy, where it referred to the ideological leadership of the

proletariat over other aligned groups (e.g. the peasantry). Yet in
Gramsci’s hands the concept was greatly expanded as a tool for
analysis. To understand Gramsci’s particular use of the term hegemony
it is important to remember that his interests were both analytical and
utopian. He wanted to understand the Italian state as an historical
formation, and he wanted to facilitate the overthrow of that state and
eliminate its necessity through the transformation of Italian society.
As a result, Gramsci used hegemony to refer quite directly to the
process by which subaltern classes acquiesced to their own political
domination by ruling classes through the use of education, cultural
activity, symbolic expression, religion, language, traditional cross-class
alliances, etc. Under these circumstances, hegemony and coercion were
intimately linked as two sides of the coin by which class dominance was
organised and perpetuated. At the same time, he also used hegemony
to refer to the process by which the subaltern classes could realise their
own potential through a ‘political-ethical’ state that organised and
articulated the classless civil society that would arise from the restruc-
turing of the relations of production. Such a ‘political-ethical’ state
would eliminate the need for the coercive powers of political society
as it would (in what could only be described as a transcript of one of
Foucault’s nightmares):
… construct within the husk of political society a complex and

well-articulated civil society, in which the individual can govern
himself without his self-government thereby entering into conflict
with political society – but rather becoming its normal continuation,
its organic complement. (Gramsci 1971: 268)
Regardless of what one thinks of Gramsci’s utopian vision, it is

clear that his view of hegemony was quite different from how it is
frequently characterised in academic circles, and reflected second-
hand in anthropological and archaeological literature (see Kurtz 1996;
Crehan 2002). For example, Gramsci’s hegemony is not a kind of false
consciousness embedded in the taken-for-granted order of tradition
(e.g. symbols, values, modes of communication); that to say an unrec-

ognised generative force lying behind cultural practices (e.g. Alonso
1994; Comaroff and Comaroff 1991: 22–32). Similarly, Gramsci’s use
of consent did not simply confuse the so-called ‘public’ and ‘hidden’
transcripts of the oppressed (contra Scott 1990: 95). Rather, for
Gramsci, hegemony entailed the organisation, direction, education
and articulation of the historical reality of a given class experience.
Hence, for the bourgeoisie, the form their hegemony took via the
modern state (apparently in either its liberal or its fascist form) was
an organic expression of their historical reality, just as the proletarian
‘political-ethical’ state would be for a future classless society. This
clearly complicates the concept of hegemony, making it much more
than the manufacture of consent. However, it also creates problems
for the analytical side of Gramsci’s task, namely understanding how
and why subaltern classes would acquiesce to a hegemony that was
not an expression of their own historical reality. Clearly, coercion is
one of the answers, but it is an insufficient answer even in Gramsci’s
own terms. This problem is never directly acknowledged or addressed
in Gramsci’s writings; however, as is often the case, he provides the
tools for constructing an alternative ‘neo-Gramscian’ view in his
own passing comments and historical and cultural analysis. This will
become more evident if we turn to his use of the terms ‘intellectuals’
and ‘common sense’.
Intellectuals and common sense

Gramsci uses the term intellectual in an unusual way, one that links it
very strongly with the constitution of hegemony. For Gramsci, intel-
lectuals were not defined by the nature of their activities, so much as
by their position in a system of social relations.
All men are intellectuals … but not all men have in society the function
of intellectuals … Thus there are historically formed specialised
categories for the exercise of the intellectual function. (Gramsci 1971:
9–10)
This ‘intellectual function’ is the leading, educating and articulating of

group hegemony. According to Gramsci, intellectuals are not simply
generated by and for the state, but arise organically within fundamental
social groups, a term he used in place of class when avoiding censorship
in prison (Gramsci 1971: 5n. 1). Intellectuals in this sense can be
engaged in a wide range of activities (e.g. technical or administrative)
not typically viewed as intellectual in and of themselves, but which
none the less provide a class with ‘… homogeneity and an awareness
of its own function not only in the economic but also in the social
and political fields’ (Gramsci 1971: 50). However, Gramsci also notes
that for dominant groups intellectuals act as ‘deputies’ in exercising ‘…
the functions of social hegemony and political government’ (Gramsci
1971: 12). These functions correspond in the first instance to ‘[t]he
“spontaneous” consent given by the great masses of the population to
the general direction imposed on social life by the dominant funda-
mental group’ and in the second to ‘[t]he apparatus of state coercive
power which “legally” enforces discipline on those groups who do not
“consent” either actively or passively’ (Gramsci 1971: 12).
The effects of coercion and the appeal of a coherent vision of one’s
own life circumstances require little explanation, the ‘“spontaneous”
consent of the great masses’ is, however, more difficult to under-
stand. Gramsci’s immediate answer is that consent is caused by the
‘… prestige (and consequent confidence) which the dominant group
enjoys because of its position and function in the world of production’
(Gramsci 1971: 12). This explanation seems inadequate, except perhaps
as a form of economic coercion. However, Gramsci himself reveals an
alternative answer when he turns his attention to the specific task of
shaping a hegemony of, and for, Italy’s subaltern classes using the ‘raw
materials’ of popular culture (termed ‘common sense’).
Gramsci was an orthodox Marxist in his belief that relations of
production were the most fundamental to the shaping of history
and society, and that capitalist relations of production in particular
contained inherent contradictions that would lead to their demise.
However, Gramsci also resisted determinism and hence argued against
the ‘fatalistic’ view that such historical changes would happen on their
own, or that historical associations and established understandings of
the world would change quickly and automatically with changes in the
relations of production. Hence, in Gramsci’s view, intellectuals (in the
form of the Communist Party) had to take an active role in shaping a
new hegemony of, and for, the subaltern classes. In Gramsci’s mind this
involved the transformation of what he calls ‘common sense’, by which
he means the unsystematised practical consciousness of the masses. For
Gramsci, the everyday culture of the subaltern classes was limited in
being unconscious, uncritical and unsystematic. It was not that people
were unaware of their circumstances, but rather that this awareness
was limited by its narrow focus and its combination with a ‘hodge-
podge’ of contradictory ideas and orientations historically inherited
and/or adopted from dominate groups (e.g. Gramsci 1971: 326–7).
Most particularly, in the case of Italy, this included Catholicism. The
task of intellectuals was not to remake ‘common sense’ but rather to
construct a critical hegemony authentic to the political and economic
realities of subaltern classes, using the raw materials of ‘common sense’
(those parts that Gramsci called ‘good sense’). As Gramsci writes:
‘… it is not a question of introducing from scratch a scientific form of
thought into everyone’s individual life, but of renovating and making
“critical” an already existing activity’ (Gramsci 1971: 330–1). Certainly,
in Gramsci’s mind, hegemonic orders could not simply be concocted
but had to connect with the shared experiences of those being led. As
he states:
It is evident that this kind of mass construction cannot just happen

‘arbitrarily’, around any ideology, simply because of the formally
constructive will of a personality or a group … (Gramsci 1971: 341)
For Gramsci, then, hegemony is realised in continuous collective

action and a common orientation. This occurs when intellectuals
are able to draw out and make coherent problems and orientations
of practical experiences found in an inchoate form in the realm of
‘common sense’.
Importantly, for our purposes, Gramsci does not limit this

method for constituting hegemony to the future Communist task of
constituting what he certainly believed to be an authentic subaltern
hegemony. For example, he states that ‘[e]very relationship of
“hegemony” is necessarily an educational relationship’, and that ‘[t]his
form of relationship occurs between intellectual and non-intellectual
sections of the population, between rulers and the ruled, between
élites and their followers, leaders and led, the vanguard and the body
of the army’ (Gramsci 1971: 350). Gramsci also acknowledges that
the ‘man-in-the mass’ can have a contradictory consciousness, one
shaped by his practical activities in the realm of production, which
would link him in common cause with other workers, and another
that is ‘superficially explicit, or verbal, which he has inherited from
the past and uncritically absorbed’ (Gramsci 1971: 333). Gramsci goes
on to note that:
… this verbal conception is not without consequences. It holds

together a specific social group, it influences moral conduct and the
direction of will, with varying efficacy but often powerfully enough to
produce a situation in which the contradictory state of consciousness
does not permit any action, any decision or choice, and produces a
condition of moral and political passivity. (Gramsci 1971: 333)
Hence, for Gramsci, all forms of hegemony involve a process of articu-

lating and making coherent symbols, values, problems and material
orientations inchoately present in ‘common sense’. This is what gives
hegemony its evocative power to elicit either active or passive consent.
The difference between what Gramsci would consider a ‘progressive’
hegemony and a ‘reactionary’ one lies in which aspects of ‘common
sense’ are brought forward and articulated as the basis of a social
order. Hence, no form of hegemony ever articulates all possible values,
problems and orientations – indeed, hegemonic projects actively select,
ignore or even repress the multiple possibilities inherent in any
lived historical context. It follows, then, that ‘common sense’ always
contains within itself the possibility of constituting alternative or
counter-hegemonies (see Laclau and Mouffe 2001: 105–14 for a similar

understanding of Gramsci).
All of this might be well and good as a commentary on Gramsci’s
writings, but does it really help us understand political authority in the
past and how this might be investigated through archaeological means?
In particular, is Gramsci’s conception of hegemony not just another
example of deterministic Marxist thinking, reducing dynamic historical
realities into the repetitive shadow play of class conflict? What does one
do in non-capitalist contexts where relations of production may be
comparatively simple, or where kinship, ethnicity and status seem to
be at least as important as class in the organisation of political power?
There is no doubt that Gramsci saw relations of production as
the most fundamental social relations, and he never broke with
the belief that such relations would ultimately form the basis for
hegemonic formations. However, as Ernesto Laclau and Chantal
Mouffe (2001) have shown at some length, Gramsci’s historicism
and his active approach to the constitution of hegemony meant that
the class character of agents in his analysis was never guaranteed in
advance, but had to be constituted. We have already quoted Gramsci
to the effect that people can possess a contradictory consciousness (e.g.
verbal vs material), with distinct implications for how they view their
own historical position and how they engage in political activity. In
Gramsci’s words, ‘[c]ritical understanding of self takes place therefore
through a struggle of political “hegemonies”, and of opposing direc-
tions …’ (Gramsci 1971: 333). Gramsci, however, goes even further
than this in his conception of the ‘historical bloc’.
Historical bloc
For Gramsci, an ‘historical bloc’ represents ‘the unity of the process of
reality’ (Gramsci 2000: 193). Here Gramsci is emphasising the unity
in any given historical moment of factors generally distinguished as
structures (forces and relations of production) or superstructures
(most everything else).
As Gramsci states:
The analysis of these propositions tends, I think, to reinforce the

conception of ‘historical bloc’ in which precisely material forces
are the content and ideologies are the form, though this distinction
between form and content has purely indicative value, since the
material forces would be inconceivable historically without form and
the ideologies would be individual fancies without the material forces.
(Gramsci 1971: 377)
In places, Gramsci portrays this unity in relatively mechanistic ways,

with superstructural elements conforming to, rather than affecting,
structural ones, as in his comparison of an historical bloc to human
anatomy, where superstructure is the skin and structural relations the
skeleton (Gramsci 2000: 197). Elsewhere, he describes an historical
bloc more realistically as an ensemble of historical relationships in
which ‘… the complex, contradictory and discordant ensemble of the
superstructures is the reflection of the ensemble of the social relations
of production’ (Gramsci 1971: 366). As a political activist, Gramsci
believed that identifying and transforming the structural relations
reflected in the ‘complex, contradictory and discordant ensemble of
the superstructures’, required ideological activity (see Gramsci 1971:
365–7, 375–7). In other words, superstructure (e.g. ideology) could
in fact shape and change structure (e.g. relations of production), not
least because in an ‘historical bloc’ these were dialectically, rather than
merely causally, linked.
The historical bloc gains further significance when Gramsci links
it to his interesting, if underdeveloped, relational theory of human
subjectivity. For Gramsci, the subject exists as a node in a network of
complex relationships. As he states:
Man is to be conceived as an historical bloc of purely individual

and subjective elements and of mass [i.e. collective] and objective
or material elements with which the individual is in an active
relationship. … the synthesis of the elements constituting individu-
ality is ‘individual’, but it cannot be realised and developed without
an activity directed outward, modifying external relations both with

nature and, in varying degrees, with other men, in the various social
circles in which one lives … (Gramsci 1971: 360)
The implications of this perspective are multiple. First, the question

of whether or not hegemony has a necessary class character is largely
misplaced with regards to Gramsci’s political thought. In so far as
particular relations of production exist as part of the nexus of social
and material relations that constitute individual subjectivity, then class
is latent in that historical bloc. There is no guarantee, however, that it is
this class experience that will be selected, brought forward and articu-
lated in constituting any particular hegemonic order.
The second implication of Gramsci’s historical bloc is that, in so far
as hegemony is part of the network of relations constituting individual
subjects, it can have a recursive relationship with ‘common sense’. In
other words, everyday life is not simply a primordial generator of social
and cultural resources to be selected and articulated in constituting
hegemony, it is also shaped historically by the existence of hegemonic
orders. Hence, hegemonic orders can be reinscribed on everyday life as
an historical experience.
Applying Gramsci
A Gramscian perspective involves analysing the formation of hegemony
as a social (and indeed cosmological) order through the selective
articulation of cultural resources embedded in everyday life, such that
specific interests are disseminated as general (and indeed essential)
interests. Because hegemony articulates values, symbols, practices and
institutions that are intimately linked to people’s lives, it often ‘rings
true’ in terms of emotions, experiences or necessities of everyday
life and hence forms a basis for voluntary consent to political and
economic domination. At the same time, because the formation of
hegemony is, by definition, partial (i.e. selective), it is never secure and
is always open to resistance or reformulation. Hence, the continued
reproduction of hegemony is intimately linked to relations of physical
and symbolic coercion. One could say, therefore, that hegemony

defines the possibilities of consent, while force defines its limits.
Finally, because both everyday culture (Gramsci’s ‘common sense’) and
hegemony are historically constituted, the experience of hegemonic
power can be reinscribed into everyday life as a cultural resource.
It follows, however, that the specific uses made of such reinscribed
hegemony are potentially open.
A Gramscian approach to political authority clearly incorporates
Baines and Yoffee’s concern for order, legitimacy and wealth in early
civilisations, while at the same time providing a means of under-
standing how such orders could relate to both elite and non-elite
experiences of everyday life. A Gramscian approach would seem,
therefore, well suited to addressing the question of how political
authority was made transcendent in given contexts, in terms of both
theory and practical application. This said, several points remain to be
clarified if we are to effectively link our discussion of Gramsci to the
issues and case studies discussed elsewhere in this book.
It should be clear that what we called political authority in Chapter
One relates to what Gramsci termed political society. It is the seat,
therefore, of coercive powers in Gramsci’s equation. While Gramsci
gave much attention to the constitution of hegemony in civil society, he
gave little attention to the constitution of political society or its coercive
powers, taking their existence for granted. Yet, as David Kertzer notes
with regards to coercion,
[T]hese sorts of power to constrain are not simply, or even essentially,

a physical power, for they presuppose a symbolic power. The power of
the government to compel people to act in certain ways is based on its
ability to mobilize people to do its bidding. (Kertzer 1996: 3)
In essence, therefore, the coercive powers of political society must

themselves be constituted as an aspect of hegemony. Here we can
begin to see how the practices and strategies of political authority,
coercive force and transcendent identification intersect and reinforce
one another under the umbrella of hegemony as a moral order.
A second clarification relates to Gramsci’s conception of intellectuals.

Gramsci was of course correct to identify the existence of historically
and culturally defined roles where the intellectual functions (selecting,
ordering, articulating, interpreting) of hegemony are concentrated (e.g.
priest, general, scribe, scientist, politician), and one might productively
revisit archaeological discussions of elites, administrators or specialists
by means of Gramsci’s discussion of intellectuals. However, limiting
the intellectual function of hegemony to these roles misses much of
the important hegemonic work done on a day-to-day basis by virtually
everyone. Indeed, for archaeologists it seems more productive to
focus on intellectual products (texts, buildings, art, material symbols,
administrative systems, etc.) that select, order and articulate cultural
resources, rather than on intellectuals. This is more than just a case of
making the most of archaeology’s inconvenient absences. Rather, such
a focus shifts our attention directly to material practices. For example,
in the case of Shang oracle bones, there is no question that intellectuals
in Gramsci’s sense are at the centre of this practice. However, so too
are the bones themselves, their preparation, processing and interpre-
tation. Treating these simply as materialisations of already existing
power (e.g. Demarrais et al. 1996) ignores the historical specificity of
the practices involved and thereby misunderstands the materiality by
which hegemony is constituted.
For the archaeologist, one last problem derives from Gramsci
himself, as it is clear that he viewed his theory of hegemony as relevant
only to modern societies. As he writes:
In the ancient and mediaeval State alike, centralisation, whether

political-territorial or social … was minimal. The state was, in a
certain sense, a mechanical bloc of social groups … within the circle
of political-military compression, which was only exercised harshly at
certain moments, the subaltern groups had a life of their own, their
own institutions etc. The modern state substitutes for the mechanical
bloc of social groups their subordination to the active hegemony of the
directive and dominant group. (Gramsci 1971: 54n. 4)
Ironically, Gramsci’s view would seem rather close to that of Baines

and Yoffee, in that hegemonic activity in ancient states is presumed to
be limited by social fragmentation (cf. Giddens 1985; see Routledge
2003 for a critique), while social cohesion rests largely on the irregular
exercise of bare force. As we have already seen, such a view of
premodern political authority fails to account for its embeddedness in
widely shared cultural practices and values. Such practices are clearly
selected, developed and at times transformed almost beyond recog-
nition; however, they still signal a degree of reciprocal interaction that
remains unaccounted for in exclusively elite-centred views. Instead
of presuming that the absence of a mass public culture means the
absence of hegemonic activity, it is more interesting and relevant to
ask how hegemonic orders are achieved in spite of, or even by means
of, ‘a mechanical bloc of social groups’ (see Routledge 2003). Indeed,
this is precisely what we intend to do in the next chapter as we look
closely at the Imerina Kingdom of nineteenth-century Madagascar in
order to understand how hegemony made possible the transcendence
of political authority and, at the same time, served as a site for resisting
and contesting that authority.
3
Hegemony in action
The kingdom of Imerina in central Madagascar
I see the numerous firenena across the land, they are all mine, I love
them all, but I will mix hairs as I please.
Andrianampoinimerina, King of Imerina (Larson 2000: 178)
In an extremely interesting article, Susan Kus and Victor Raharijaona

(2000) detail the intimate relationship between standard house plans
amongst the Merina of highland Madagascar and the representation of
kingship under Merina rulers of the nineteenth century. In particular,
the singular authority of the ruler came to be embodied in the central
pillar (andry) of the national palace, just as andry in Merina houses
could serve as a symbol and an index of the household head. This
subtle appropriation and transformation of common-sense under-
standings of the world fits rather well with the Gramscian perspective
outlined in the previous chapter. Indeed, almost every aspect of Merina
royal policy, from state ritual, through royal building programmes to
royal labour service, can be understood as having selected and trans-
formed widely shared Merina traditions in the interest of constituting
and reproducing the hegemony of Merina sovereigns. For this reason,
the kingdom of Imerina provides an excellent case study through
which to explore the question of hegemony, raised in an abstract form
in the previous chapter.
History and archaeology of Imerina
The kingdom of Imerina was located in central highland Madagascar

and is best known for its rapid expansion by conquest between
ad 1780 and 1830 (Figure 3.1). At its peak, Imerina incorporated
about two-thirds of the island of Madagascar (c. 380,000 km2) and a
population of upwards of one million people (Campbell 2005: 136–7;
Raison-Jourde 1991: 34). Information on this period of expansion
comes from a variety of sources. The most important indigenous
source is probably the transcriptions of royal oral histories made in the
1860s and known as the Tantara (Tantara ny Andriana = ‘the Histories
of Kings‘; Callett 1908). European sources include travel accounts of
merchants, slavers, and missionaries as well as military and ambassa-
dorial personnel. Archaeological surveys and limited excavations have
been carried out in parts of central Madagascar with some regularity
since the 1970s, although both geographic and chronological coverage
remains uneven. In what follows, I have relied on secondary sources
that make use of the above primary sources, with the exception of some
of the archaeological evidence.
Imerina is the name of the territory of one of Madagascar’s major
cultural groups, the Merina. The early history of both Madagascar’s
settlement and the settlement of its interior highland is uncertain and
subject to controversy (see Kent 1970). For our purposes, the period
prior to the eighteenth century is not immediately relevant except
for one key aspect of Merina oral histories. According to the Merina,
their ancestors displaced a group called the Vazimba in order to claim
Imerina. As a result, Vazimba ancestors were viewed as playing an
important role in the landscape of Imerina as sources of undomesti-
cated and antagonistic primordial power associated with water, rocks
and the underworld (see Bloch 1986: 42–3; 1987: 280; Kus and
Raharijaona 1998: 61–4) .
The economy of highland Madagascar is based on riziculture in
valleys and on terraced slopes. A key ecological factor is that large
sections of these valleys were swampy and required labour-intensive
Hegemony in action 51
Figure 3.1 Map of the Kingdom of Imerina

drainage projects via ditches and levees in order to be brought under rice
cultivation (see Berg 1981; Campbell 2005: 18–30). In Merina tradition,
kingship is closely associated with the initiation of such drainage
projects, most notably the king Andriamasinavalona (‘The Noble finely
enveloped by sanctity’), who may have reigned at the turn of the eight-
eenth century and was credited with both the unification of Imerina
and an extensive restoration and expansion of the rice cultivation zone.
Archaeological evidence in the form of survey data from north-eastern
Imerina (western Avaradrano) is more ambivalent on the scale and
timing of these irrigation projects, with intensive rice cultivation evident
from the end of the fifteenth century (Wetterstrom and Wright 2007:
286), but large-scale centralised irrigation not unambiguously attested
until the end of the eighteenth century (Schwarz 2007: 295).
According to the Tantara, Andriamasinavalona could not decide
between his children in choosing a successor and divided his kingdom
into four parts on his death – these making up the four traditional
provinces of the heartland of Imerina. This led to internecine warfare
for three generations. European accounts provide support for the idea
that the eighteenth century was a period of conflict and fragmen-
tation (see Berg 1985; Larson 2000: 49–147). The rise of plantations
in Mauritius and other Mascarene islands increased the intensity of
slaving activities in highland Madagascar. With slaving came a more
extensive introduction of muskets into warfare and the circulation of
silver coinage. Rulers eager for silver and muskets actively sold both
war captives and their own subjects into slavery. An increase in the
tendency to demand tribute in silver rather than in kind, and the
need to use silver to buy relatives out of slavery, seems to have put
pressure on local descent groups, who became more mobile, smaller
and more fluid in terms of membership and royal affiliation (Larson
2000: 156–7).
Archaeological research, at least in north-eastern Imerina, suggests
that we need to be cautious in estimating the scale and universality
of these social disruptions. Here the early eighteenth century sees a
slowing of the previous century’s very high rates of population growth,
but no population decline (Dewar 2007: 101–3). Settlement patterns

show considerable continuity, with an increase in the number of very
small sites being the primary change. An increased prominence of
fortifications at larger sites is the principal archaeological evidence
complementing the historical evidence for civil war (Wright 2007:
108). At the same time, the limited scope and chronological precision
of the available archaeological evidence also means that evidence for
social disruption in the mid-eighteenth century may be obscured by
latter developments, or better attested in regions that have not been
subject to intensive archaeological investigation.
Starting in about 1780, Andrianampoinimerina (‘The Noble desired
in the heart of Imerina’) used trade connections to build up silver
and muskets in order to attract displaced followers and usurp his
uncle as ruler of one of the four provinces. He then began a series
of aggressive military campaigns that united Imerina, and began
to expand its territory outward at the expense of its neighbours.
Andrianampoinimerina attracted displaced descent groups, promising
protection from slavers, while at the same time acquiring resources
via the sale of war captives into slavery (see Berg 1985; Larson 2000:
147–56; Raison-Jourde 1991: 40–4).
Aggressive expansion peaked under Andrianampoinimerina’s son
Radama I, whose reign from ca. 1810 to 1828, saw tens of thousands
die in wars of conquest, many of whom were conscripted Merina
soldiers who died of fever and disease some distance from home
(Larson 2000: 217–40). While taxation, especially in the form of uncut
silver coins called hasina (which also means ‘blessing’), was fiscally
and symbolically important, Merina royalty relied at least as heavily
on labour service. Under Radama I the extraction of labour service
reached extreme levels, both via conscription into largely unpaid
military service and via the extension of local obligations to a wider
cross-section of the population to compensate for the young men
absent in military service.
Unlike earlier periods, archaeological evidence from the late eight-
eenth and early nineteenth centuries shows dramatic changes very
much in keeping with expectations derived from the documentary

evidence. Long-established sites are abandoned and new settlements
founded; primary and secondary centres grow considerably in size and
are heavily fortified. For the first time, the old capital of Ambohimanga
may have had a population larger than could be fed from its available
catchment area, suggesting either a dependence on tax/tribute or the
extension of rice cultivation through large-scale irrigation schemes
(Schwarz 2007: 295). A line of fortified frontier settlements is now
discernible in the far south of Imerina’s traditional territory, and for the
first time gun flints occur as regular finds in excavations (see Wright
2007: 89–90, 108–11).
Archaeological evidence also supports the documented
tradition that, even while founding a new capital at Antananarivo,
Andrianampoinimerina reconstructed the old capital of Ambohimanga
as a ritual centre on a grand scale. He appears to have followed cosmo-
logically significant spatial orientations in laying out the gates and
fortifications (Kus 2007; Kus and Raharijaona 1998; Belrose-Huyghues
1983) and deployed monumental stonework in their construction,
most particularly in the case of the stone discs used to block gates
(Gabler 2007).
Finally, preliminary archaeological evidence from the region
of Andrantsay, south-west of Imerina, seems to show evidence for
Andrianampoinimerina’s conquests (see Crossland 2001). Here the
settlement pattern exhibits dramatic changes near the turn of the
century, with most of the oldest, best fortified sites, at the highest
elevations and with the most built tombs (both signs of higher status
residents), being abandoned. About half of the settlements with ditch
fortifications occupied in the nineteenth century appear to have been
new foundations at the time of the Merina conquest (Crossland 2001:
834).
Radama I and his successors (with the exception of Queen Ranalova
I) formed close ties with Great Britain via the London Mission Society.
Under Ranalova I, state ritual was considerably elaborated, perhaps as
a means of counter-balancing the spread of Christianity and European
influence (Bloch 1986: 18–22). Her son and successor Radama II

reversed these policies and in 1869 Protestant Christianity was adopted
as the state religion during the reign of Ranalova II. British and
French conflict over economically important colonial interests in the
Mascerenes led to a French invasion of Madagascar and a bloody war
from 1883 to 1885. During this invasion Britain remained uninvolved,
essentially ceding Madagascar to the French sphere of influence in
exchange for a free hand in other areas, such as Zanzibar (Ellis 1985:
32–3). Eventually, after widespread uprisings against the French from
1895 to 1899, the kingdom of Imerina was dissolved and replaced by a
French colonial government (see Ellis 1985).
Ritual and the ideology of kingship
While a Merina polity, in some form or another, existed long before

1780, the Tantara gives particular attention to Andrianampoinimerina
as the founder of a renewed Merina order. Stress is always placed upon
the singularity of the ruler and the unity of Imerina, something that is
symbolically represented over and over again in the deeds and words
of Andrianampoinimerina (see Kus and Raharijaona 1998: 67–8; 2000:
104–6; Raison-Jourde 1991: 48–52). This key theme is summed up in
the Malagasy proverb: ‘When there are three lords there is famine and
hunger, when there are two lords wives and children are lost as slaves,
but when there is one sovereign children are chubby’ (Veyrières and
Mèritens 1967: 36, as quoted in Kus and Raharijaona 1998: 68). While
there is certainly a rhetorical element to Andrianampoinimerina’s
prominence in the Tantara, the archaeological record clearly attests to
the transformative nature of his reign.
As Maurice Bloch has argued on numerous occasions, this renewed
Merina order revolved around royalty and was embodied and realised
via public ritual. Two of the most prominent public rituals studied
by Bloch are the ritual of the royal bath (Bloch 1987) and the
circumcision ritual (Bloch 1986). In both cases, Bloch stresses the
ways in which royal rituals incorporate and transform rituals, beliefs

and values associated with local descent groups. Merina and other
highland Madagascar groups tended to be organised into preferen-
tially endogamous, bilateral descent groups that ideally share rights
in land and residence in villages. These were known in the nineteenth
century as firenena and were ranked in social status based on genea-
logical distance from key royal figures in the distant past. Major status
divisions were between so-called ‘nobles’ or andriana (from whom
royalty could, at least notionally, arise), commoners or hova, and slaves.
One long-term dynamic of the Merina kingdom was a tension between
the desire to embed royal hegemony within the moral order of descent
groups, while at the same time to break down their autonomy as alter-
native centres of loyalty and identity.
Both the ritual of the royal bath and the circumcision ritual seem
to represent royal appropriations of common Merina rituals (Bloch
1986: 116–18; 1987; Kus and Raharijaona 2001: 117–20). At the heart
of both, for example, is the transference of blessings downwards from
elders to juniors via the use of water associated with the Vazimba
ancestors. Merina royalty were inserted into this chain of blessing by
organising and centralising these rituals such that the sovereign served
as the initiator of a national chain of blessing. In the case of the royal
bath, the washing of the king’s hair on the New Year initiated the repli-
cation of this ceremony down a line defined by status and seniority.
It was also the occasion of annual visits to family tombs, linking in
this way ancestors, descent and royalty. According to the Tantara,
while declaring that he did not change the ways of the ancestors,
Andrianampoinimerina decreed that circumcision rituals were to
occur only once every seven years, and to be initiated by the king and
carried out by descent groups in order of their status and seniority,
starting with royal children (Bloch 1986: 113–17).
In a functional sense, Merina royalty justified its rule through ritual
by placing itself in the most senior position in this chain of blessing, in
essence nationalising the ranking structure of descent groups. Rather
conveniently this included the reciprocal delivery of uncut silver coins
(hasina) upwards along the same chain, and indeed both the royal bath
and circumcision rituals served as occasions for the delivery of taxes
and tribute to the sovereign. However, the role of such rituals in the
constitution of political authority goes much further.
The symbolic content of both of these state rituals is complex and
provides many examples of the appropriation and transformation of
‘common sense’ practices, symbols and values. For our purposes, one
of the most striking is the way in which Merina royalty drew upon
the undomesticated power of the Vazimba ancestors. Regular Merina
were dependent on their ancestors to neutralise, as it were, the wild
powers inherent in water related to the Vazimba. Sovereigns, however,
did not have ancestors in the same way as regular Merina, in the sense
that they were singular, standing apart from descent groups and being
buried in individual tombs (Kus and Raharijaona 2001). At the same
time, Merina rulers could lay claim through the mythical founder
of Imerina, Ralambo, to being both the conquerors of the Vazimba
and their partial descendants, as Ralambo’s mother was said to be a
Vazimba queen. Hence, Merina royalty could partake of, and control,
an undomesticated source of power that stood outside of Merina
society as defined by ancestors and descent groups. In other words,
the political authority of Merina rulers was not constituted from crude
appropriations of popular symbols and rituals. Rather, it involved the
transformation of those symbols and rituals by representing them in
terms of a transcendent order, one that reinforced the wider hegemony
of senior over junior kinsmen, while at the same time transcending this
hegemony in the singular position of the ruler.
Bloch’s analysis of Merina royal rituals gives us something of
a structural model for the constitution of political authority in
Gramscian terms; something that we might represent along the lines
of: Appropriation → Transformation → Reinscription. This is concep-
tually useful, but it also rather static and fails to capture the historical
dynamic of political authority as something that is both asserted and
contested in relation to given hegemonic formations. Two examples
from nineteenth-century Imerina will give us some sense of the
dynamic nature of this process. The first is a single event involving

a haircut, while the second is a longer-term process with a more
clearly archaeological component, involving the development of burial
practices.
Hegemony in action I: Radama I’s haircut
Both Merina men and women wore their hair in elaborate plaits in
the eighteenth and nineteenth centuries and had almost 30 terms for
these hairstyles (Larson 2000: 240). One means by which war captives
were distinguished on enslavement was by shearing their hair. Indeed,
an eighteenth-century king of Marovatana, known for illegitimately
enslaving his own subjects, was referred to as ‘Mr. big scissors’ (Larson
2000: 243).
The singularity of the king is also compared to his hairstyle, for,
just as the various strands of the king’s hairs were woven together and
ultimately all attached to his head, so too were the various strands of
the Merina woven together into a single, beautiful whole by the king
(Larson 2000: 5–6, 178–9). However, in 1822, Radama I decided to
cut his hair short in the style of his British military advisors who had
helped him establish and train Imerina’s first standing army (Figure
3.2). Public reaction to his haircut was divided. On one hand, many
soldiers in Imerina’s newly professionalised army, as well as officials
and some mission-school students, followed Radama’s lead and had (or
were required to have) their own hair cut short as a sign of their affili-
ation with the king and his modernising regime. On the other hand,
protest broke out most particularly amongst women, who led those
stating that Radama I had become a slave of the British.
Hair as a symbol of descent group fertility was especially associated
with women, as was the privilege of plaiting men’s hair. A group of
4,000 women gathered to petition Radama to regrow his hair and sever
his links with the British government. Interestingly, in their exchange,
both Radama I and the protesting women grounded their arguments
Figure 3.2 Radama I (source: Owen 1833: 118)

in the hegemonic language of ancestral tradition. For the women,

however, this implied constraints on the centralising and aggrandising
behaviour of Radama, while for Radama his political innovations were
justified by the traditional prerogative of Merina royalty to do as they
pleased (Larson 2000: 248–9, 253). Rather strikingly, Radama is quoted
as saying that whoever had brought this petition to him should be
executed, as ‘they have forbidden me to carry out my desire. What they
have done prevents me from being king’ (as quoted in Larson 2000:
250). In fact, Radama I is said to have gathered 2,000 soldiers, requiring
them to swear that on the king’s order they would put to death the
instigators of these protests even if it were their own parents. He then
singled out four leaders from amongst the protestors and ordered
soldiers who were members of the women’s own descent groups to
bayonet them (Larson 2000: 252).
The king’s hair may have been a royal symbol, straightforwardly
ideological as a static representation of order. At the same time, it
was also a language through which the limits of political authority
were contested, and on which the coercive powers of the kingdom of
Imerina were brought to bear (Larson 2000: 253–7). Here we can see
hegemony in historically dynamic, rather than structural terms. As
Larson (2000: 256) notes, in adopting the logic of royal hegemony in
their protest Merina women were at a distinct disadvantage, yet at the
same time this was the most potent moral force available to them in
confronting the king.
Larson’s point can be emphasised by considering Peter Gose’s
analysis of royal oracles in the Inca Empire (Gose 1996). Gose notes
that the Inca tradition of mummifying and venerating dead kings in
regional shrines, where these past rulers could then provide oracles
of relevance to contemporary events, was a means of asserting and
negotiating subaltern interests within the framework of absolute royal
power. Because these oracles originated from his royal predecessors,
rather than living subjects, they could be heeded without undermining
the absolute rule of the reigning Inca. In the words of Peter Gose,
‘[o]racles made the king listen without having to listen to a rival.
Conversely, oracles gave those without power a voice, but not a voice
of their own’ (Gose 1996: 16). While the protests of Merina women
occurred in a moment of political crisis, they share with Inca oracles
the attempt to transform royal hegemony into a ‘weapon of the weak’ in
James Scott’s (1985) terms, even as its deployment served to reproduce
that hegemony as a ‘public transcript’.
Hegemony in action II: Tombs, kin and kings
What makes the story of Radama I’s haircut all the more striking is that
the moral order to which the protesting women appealed, one in which
Merina kingship preserved and protected the world of descent groups
and ancestors, is credited in its establishment to Radama I’s father,
Andrianampoinimerina. In the Tantara Andrianampoinimerina is
credited with re-establishing the fixity of descent groups in Imerina,
after the chaos of the eighteenth century. In particular, he is quoted to
say that every descent group must return to its own land and ancestors
and essentially stay put. Only the king, again using the metaphor of
hair, was free ‘to mix the hairs’ of the Merina as he pleased by moving
people from their ancestral land. Andrianampoinimerina is also
credited with formalising the rank order of the various Merina descent
groups, ascribing certain privileges on the basis of that order and
empowering fokonolona (councils of elders) to adjudicate and make
collective decisions within each descent group (Larson 2000: 180–2).
Central to the fixing of descent groups in space was the construction
of ancestral tombs (Figure 3.3). Andrianampoinimerina is credited
with telling the Merina to build megalithic stone tombs to house their
ancestors (Larson 2000: 184, 191). The construction of megalithic
tombs, and especially the movement of their monumental stone slab
walls, required considerable collective effort in the form of work teams,
which in turn emphasised the shared solidarity of tomb-building
groups (Kus and Raharijaona 1998: 58–9; Larson 2000: 188–90). As
Susan Kus and Victor Raharijaona have pointed out (1998: 68, 73–4),
this also created an experiential link between collective labour in the

service of the ancestors, and labour service to Andrianampoinimerina.
Such royal labour service focused on stone-moving projects, such
as the erection of megalithic gate-stones in settlements fortified by
the king, and in the large-scale renovation of the ancient capital,
Ambohimanga. Interestingly, one factor that made the spatial orien-
tation of Ambohimanga’s gates cosmologically significant is that these
same orientations were used in the construction of tombs and houses
(Belrose-Huyghues 1983; Kus 2007; Kus and Raharijaona 1998: 72;
2000: 106).
Tombs also provided potent cultural resources to be contested in
the reproduction of political authority. As noted above, visits to family
tombs in order to rewrap the dead and receive blessings from the
ancestors were incorporated into the annual state ritual of the royal
bath (Bloch 1987). This practice was expanded outside of the Merina
Figure 3.3 Stone slab tomb, central Madagascar (source: Sibree

1870: 247)
heartland in conjunction with the expansion of the Merina state. At the

same time, in contrast to family tombs, royal tombs were singular, not
collective, and such tombs were built above ground rather than being
semi-subterranean (Kus and Raharijaona 2001: 118). Monarchs were
buried at night rather than at sunset, as they were the ‘sun-that-is-not-
two’. The ambiguity, indeed virtual denial, of royal mortality meant that
the heir or heiress did not assume mourning clothes or participate in
the funeral, but when he or she assumed office it was said that the ‘sun
was newly risen’ (Kus and Raharijaona 2001: 118).
Archaeological surveys in north-eastern Imerina suggest
that, although stone built tombs were not an invention of
Andrianampoinimerina, there are marked changes in tomb
construction that can be dated to the late eighteenth or early
nineteenth century. Dating tombs in central Madagascar is
problematic, in that tombs in continuously occupied sites were
continuously modified. Looking only at newly founded sites with
tombs, from each successive time period, Henry Wright and his
colleagues were able to suggest that tombs constructed of very
large horizontal and vertical slabs were a development of the late
eighteenth–early nineteenth century, although these occur alongside
already established construction methods, such as the use of dressed
stone blocks (Wright 2007: 71–3). Interestingly, another development
of the late eighteenth-early nineteenth centuries in north-eastern
Imerina is a reduction in the proportion of sites containing tombs
within the settlement boundaries and a greater correlation between
the presence of interior tombs and other signs of settlement impor-
tance (Wright 2007: 109). Documented Merina traditions, all of
which post-date Andrianampoinimerina, suggest that only adriana
(noble) descent groups had tombs within their settlements (Kus
and Raharijaona 2001). Archaeological evidence might, therefore,
support the Tantara tradition that the ordering and fixing of the
relative status of descent groups was associated with the reign of
Andrianampoinimerina and became embodied in traditions and
privileges related to the nature and placement of family tombs.
The most famous aspect of Malagasy burial customs, the elaborate

secondary burial ritual involving the handling and rewrapping of
disarticulated ancestral bones at collective tombs termed famadihana
(see Bloch 1971), was not initially the dominant mode of burial in
Imerina. Although secondary burial is attested in eighteenth-century
accounts, the famadihana in its present form only emerged towards the
end of the nineteenth century, especially after the end of the Merina
state (Raison-Jourde 1991: 711–21; Larson 2001). Its development is
linked in part to a long struggle over burial between descent groups
and rulers.
One of the ironies of the apparent strengthening of descent group
identity and its attachment to specific ancestral places under the
Merina state is that Merina rulers, and particularly Radama I, moved
people away from their ancestral homelands on an unprecedented
scale via military, labour and administrative service. In this context of
military expansion, descent group identity became strongly focused
on funerary rituals. In the early nineteenth century, secondary burial
in family tombs was primarily the result of burying soldiers who had
died in Imerina’s wars of expansion (Larson 2001: 131–6). Funerary
rites were preferentially for primary burial, with elaborate expenditure
in terms of sacrificing bulls for feasts and equipping of the tomb with
wealth in silver and household goods (Larson 2001: 139–42).
Françoise Raison-Jourde (1991: 714–15) credits the decline in
primary burial rituals focused on wealth expenditure to the estab-
lishment of Christianity as the state religion under Ranalova II (reigned
1868–83), who banned the sacrifice of bulls at funerals as a pagan
custom. However, royal resistance to elaborate funeral rituals can be
traced back almost to the beginning of Imerina’s expansion. Radama I
is said to have opposed the effort expended by soldiers preparing and
returning their comrades’ bones, as well as the wealth being expended
on funerals. He declared that soldiers should be buried where they died,
encouraging them to view all of Imerina as their own country (Larson
2001: 134–5). Similarly, the expense of Radama’s extensive military
campaigning, combined with a marked decline in the availability of
silver coinage following the banning of slave exports in 1820, led to

something of a monetary crisis towards the end of Radama I’s reign.
Hence, in 1824 he absolved the Merina of the obligation to repay loans
taken out specifically for the purposes of holding funerary feasts and
depositing the requisite amounts of wealth in tombs. This had the
effect of making creditors unwilling to loan silver for the purposes of
funerary rituals, keeping precious coinage in circulation and hence
available for taxation (Larson 2001: 143–4). Radama I’s campaign
against the interment of wealth seems to stand in direct opposition
to the Tantara’s tradition that Andrianampoinimerina encouraged
the Merina to invest their wealth in collective tombs (Larson 2000:
191). Ironically, it also stands in direct opposition to Radama I’s own
funeral in 1828, which was said to include the burial of 10,300 silver
coins and imported goods worth £60,000 with the king’s body (Larson
2001: 144).
Over the course of the nineteenth century, political, economic
and religious pressures encouraged descent groups to reduce their
emphasis on displays of wealth in primary burial rituals, and to focus
efforts on the elaboration of secondary burial rituals. Importantly,
while the burial practices of kin groups developed in relation to royal
(and later colonial) attempts to control them, the outcome of these
changes was not necessarily what Merina rulers had envisioned.
It seems that the hegemonic order in which the king made the
land secure for the ancestors as exemplified in collective tombs also
served to limit kingship and strengthen the independence of descent
groups. Secondary burial in some form or another was a long-standing
tradition in highland Madagascar, but the famadihana in its ethno-
graphically famous form seems to have developed as part of long-term
struggles between Merina royalty and Merina descent groups over the
meaning and control of the central cultural resources of ancestors,
tombs and wealth.
Summation
The history of Imerina provides a concrete example of how hegemony

constitutes transcendent political authority by appropriating and trans-
forming shared cultural resources embedded in what Gramsci called
‘common sense’. Merina history also highlights the role of coercion,
both in law and in physical force, as the limits of hegemony were
reached through events like Radama I’s haircut.
At the same time, the case of Imerina has also taken us beyond
Gramsci in important ways. Gramsci’s utopian goals meant that he
gave little attention to the uses subaltern groups might make of
actually existing hegemonic orders, as he viewed these as inauthentic
to the experience of proletarians and peasants. Yet, as we saw in the
case of Merina women and Inca oracles, existing hegemonic orders
often provide the most powerful and pervasive arguments available to
subaltern groups in the immediate context of conflict with authority.
Hence, polity formation is not simply a case of domination on one side
and revolution on the other. Rather, it involves on-going debate over
what hegemonic orders do and do not include, framed by relations of
force and carried out in contexts where cultural and material resources
are not evenly distributed amongst participants. All of this involves
material practices and contexts that are at least partially discernible in
the archaeological record. It also involves much more than just politics.
It is to this ‘bigger picture’ that we now turn.
4
Beyond politics
Articulation and reproduction in Athens and
the Inca Empire
There are processes at work, therefore, that define regional spaces

within which production and consumption, supply and demand (for
commodities and labour power), production and realisation, class
struggle and accumulation, culture and lifestyle, hang together as
some kind of structured coherence within a totality of productive
forces and social relations.
David Harvey (1985: 146)
On page 172 of the sixth edition of Renfrew and Bahn’s (2012) widely
used text book Archaeology: theories, methods and practice there is
a table summarising the relationship between different social fields
(i.e. social organisation, economic organisation, settlement pattern,
religious organisation, architecture) and different levels of social
organisation (i.e. mobile hunter-gatherer groups, segmentary society,
chiefdom, state). Each cell in this table links a distinct social field with
an overarching social form by means of a trait with archaeological
implications. It is, in essence, a kind of pocket field guide to the identi-
fication of social evolution in the archaeological record. This, of course,
is exactly the sort of thing against which the critics of neo-evolutionism
have consistently railed; a view in which politics, economy and social
relations are treated as attributes of some larger entity (a ‘culture’ or
‘society’) that has taken the phylogenetic form of the state. As noted
in previous chapters, this unitary position assumes what it needs to
explain; namely how sets of seemingly disparate words and actions,

routines and spectacles, resources and beliefs constitute the virtual
powers invoked by the phrase ‘the state’ (cf. Kurtz 2006). That said,
Renfrew and Bahn’s table has at least one clear advantage; it forcibly
reminds us that political power encompasses much more than just
politics.
Here is our problem. If we begin from a cross-cultural and transh-
istorical perspective, it is difficult to identify necessary and consistent
relationships between political power and other aspects of life that
do not trivialise the lived realities of specific historical contexts.
Centralised powers do not always mobilise resources in the same
way, not least because they hold no consistent relationship with the
realm of production (not to mention distribution or consumption).
Furthermore, one cannot point to a common mode, or degree, of
integration when it comes to the relationship between domestic groups
or communities, and hierarchical polities. Indeed, even the boundary
between rulers and ruled is not acknowledged, defined or represented
consistently across time and space.
The incoherence of human society as a general category underlies
the failure of unitary views of the state as a social or institutional form.
It is why social evolutionists are forever fated to choose between reduc-
tionism and taxonomy when faced with the diversity of human social
life. However, acknowledging the incoherence of human society as a
general category presents us with a definite problem, namely how then
do we explain the apparent coherence of specific social formations?
To be more precise; when sovereignty is reproduced in any
specific historical context, it requires the assembly of a whole range
of key elements into relationships that are neither wholly determinate
nor wholly independent. Modes of production, accumulation and
consumption, modes of identification and differentiation (e.g. kinship,
gender, age and status), domestic, task and interest group composition,
not to mention technologies, objects, landscapes and taskscapes, are
all examples of elements whose interrelationship can be central to the
reproduction of sovereignty in given contexts. This does not mean that,
Beyond politics 69
for example, politics subsumes economic production or vice versa.

What it does mean is that the practices, strategies and relationships
of sovereignty must bring politics and economic production together
and take account of each, however imperfectly. This strategic process
of assembly, of articulating elements in a manner that makes sover-
eignty possible, is another way of understanding hegemony. Such acts
of assembly are never guaranteed, but must be achieved actively and
strategically in specific places under specific historical circumstances.
This makes it rather interesting and important to explore how, in
given contexts, the relevant elements ‘hang together’ to form a ‘spatio-
temporal fix’ (Jessop 2006) giving particular realisations of sovereignty
whatever stability and durability they may possess.
In this chapter I want to go beyond the more obvious issues of
demarcating and asserting political power that have concerned us thus
far in order to explore how it is that things ‘hang together’ in specific
historical contexts and how this relates to hegemony and sovereignty.
Common answers that depend on static, stable or holistic views of
society, such as homeostasis or functional interdependence, have been
extensively and effectively critiqued in the archaeological literature
(e.g. Shanks and Tilley 1988: 138–41). So too have explanations rooted
in methodological individualism, such as rational choice theory, that
posit a priori universal human subjects with fixed hierarchies of desire
(see Johnson 2006 with responses). Equally problematic are perspec-
tives that seem to valorise unconstrained agency (see Johnson 2006
with responses), as if the indeterminacy of human social life meant that
specific human lives were all equally indeterminate and unconstrained.
More interesting are approaches that actively struggle with the
problem of determinacy and contingency in given social formations.
One place to start is with Marxism’s, and especially Western Marxism’s,
long struggle with the problem of economic determinism and historical
contingency (Jay 1984). As we noted in Chapter 2, this tension was a
central concern of Antonio Gramsci in developing his concept of an
historical bloc as ‘the unity of the process of reality’ (Gramsci 2000:
193). Gramsci’s insistence that material forces and ideologies were
inseparably realised in given historical blocs partially dissolved the

orthodox Marxist (i.e. the Second International) division between a
society’s determining economic base and its determined ideological
superstructure. At the same time, Gramsci located the realisation of
an historical bloc not at the abstract level of society, but rather in the
formation of specific human subjects who exist at the intersection of
material forces, history, social relations, ideologies and individuality.
In doing so he introduced a sophisticated and penetrating means
of addressing what might be termed the contingent determinacy of
specific historical contexts (see Gramsci 1971: 352–60).
One drawback to Gramsci’s formulation, already noted in Chapter
2, is that it retains an essential core of economic determinism. For
Gramsci, an historical bloc is simultaneously the contingent outcome
of specific historical conjunctures and pre-determined as the logical
outcome of the relations and forces of production. It is rather difficult
to ‘have one’s cake and eat it too’ in this manner, hence it is useful to
consider approaches that help us extend the concept of historical bloc
beyond Gramsci.
In their ‘post-Marxist’ manifesto Hegemony and socialist strategy,
Ernesto Laclau and Chantal Mouffe (2001) explicitly sought to root
out Gramsci’s residual essentialism by means of post-structuralist
discourse theory. For example, in their terms a ‘social and political
space relatively unified through the instituting of nodal points and the
constitution of tendentially relational identities is what Gramsci called
a historical bloc’ (Laclau and Mouffe 2001: 136, emphasis in original).
Here the phrase ‘tendentially relational’ is significant because Laclau
and Mouffe recognise no necessary or fixed identities or social relation-
ships. For them, the social world is overdetermined, in the sense of
always containing multiple possibilities that prevent it from being fully
intelligible in a fixed form. Hence, there is no society in the sense of a
singular totality. Instead, there are only ongoing attempts to differen-
tiate elements from within the social world’s myriad of possibilities and
articulate them together in a discursive formation around some privi-
leged nodal point, thereby seeking to transform and fix their identity
Beyond politics 71
with regard to each other (Laclau and Mouffe 2001: 105–14). In Laclau
and Mouffe’s formulation an element can be virtually anything (a
practice, person, category, object, etc.) that is discursively recognised
as a ‘difference’ in the linguistic sense. Laclau and Mouffe consider
such discursive formations to be hegemonic formations when they are
shaped in terms of antagonistic forces, that is to say, in terms of the
subordination or exclusion of oppositional positions.
Laclau and Mouffe do not really answer the question of why things
‘hang together’ within an historical bloc; rather, they redefine what it
means to ‘hang together’. Hence, at this abstract level social, forma-
tions are entirely contingent. In part this is because Laclau and Mouffe
actively deny that causes can be external to given discursive formations,
as would be the case in the orthodox Marxist precept that relations and
forces of production shape human consciousness and social forms. In
so far as they might answer our question, it would need to be in terms
of the internal dynamics of given historical blocs; yet here one finds
their analysis impoverished in at least two ways.
First, Laclau and Mouffe deny the existence of non-discursive
practices, using Wittgenstein’s concept of language games (Laclau and
Mouffe 2001: 108) to deny that mental and material activities can be
segregated in this manner. Objects exist, but they cannot interact with
humans except through the discourse that makes them evident. Here,
Laclau and Mouffe could learn much from Actor-Network Theory, an
approach with which they otherwise share many parallels.
At about the same time that the first edition of Laclau and Mouffe’s
book was published, Actor-Network Theory (ANT) began to take
shape within Science Studies through the work of scholars such as
Bruno Latour, Michel Callon and John Law. ANT sought to develop
radically empirical descriptions of science and technology as networks
of relations between actors without presuming in advance what
constituted an actor (e.g. human v. non-human; organic v. inorganic;
organism v. institution) or privileging causes conceived as external to
the network (e.g. power, social structure, economy – all of which would
be seen as network effects, not causes, in ANT).
ANT makes little or no direct use of either Gramsci or Laclau and

Mouffe. However, like Laclau and Mouffe, ANT scholars tend to deny
society as an entity, and argue that the social exists only in the process
of assembling relational networks of actors (Latour 2005); that such
networks are contingent and heterogeneous (Law 1992; Latour 2005);
and that the agency of actors is constituted within, rather than prior to,
such networks (Law 1992).
Where a major difference lies is in the active agency that ANT
ascribes to the material world. This is not simply a case of discursive
fancy bumping up against cold hard facts. In ANT the natural world
is as indeterminate and rife with hidden possibilities as is the social
world (indeed they are the same world!). Furthermore, these emergent
possibilities only reveal themselves in relation to other human and
non-human actors, that is to say, within specific networks. However,
these acts of revelation are not merely the result of human discursive
recognition (e.g. the cluster of practices, discourses and identities that
arose with the identification and naming of AIDS). Rather, the material
world itself has a kind of agency in that the qualities and affordances of
different configurations of matter (and other non-human actors) shape
how humans relate to, and by means of, these different non-human
entities (e.g. the specific qualities and affordances of HIV have made an
obvious difference to the cluster of practices, discourses and identities
linked to AIDS).
Power, therefore, is not merely reflected or ‘materialised’ in matter.
Power is a quality of networks linking human and non-human actors
as mediators or intermediators in a chain of interactions. Bruno Latour,
for example, is fond of citing the intense, time-consuming, face-to-
face interactions that are required to build and maintain relations of
dominance in troops of baboons where, in comparison with human
relations, the material mediators constituting power are very limited.
As he states:
If sociologists had the privilege to watch more carefully baboons

repairing their constantly decaying ‘social structure’, they would have
witnessed what incredible cost has been paid when the job is to
Beyond politics 73
maintain, for instance, social dominance with no thing at all, just

social skills … It’s the power exerted through entities that don’t sleep
and associations that don’t break down that allows power to last longer
and expand further … (Latour 2005: 70, emphasis in original)
This symmetrical (to use an ANT term) appreciation of the mediatory

role of the material world is missing from Laclau and Mouffe, and yet
the agency of matter in relation to human beings is certainly a key
factor in why things ‘hang together’ in specific contexts. This is because
networks and discursive formations are not merely sets of relations,
they are also sets of material interdependencies.
For the real import of this point to become clear, however, we must
introduce a second factor missing from Laclau and Mouffe and also
from most of the ANT literature, namely the historical component of
Gramsci’s historical bloc. To quote Stuart Hall’s sympathetic critique of
Laclau and Mouffe: ‘Their problem isn’t politics but history. They have
let slip the question of the historical forces which have produced the
present, and which continue to function as constraints and determina-
tions on discursive articulation’ (in Grossberg 1986: 58). Material forces
and ideologies gain their ability to shape and constrain not through
universal relationships of causation but through historical relation-
ships of emerging material interdependence. Living in particular
environments, utilizing particular technologies, being born into the
midst of existing practices, dispositions and values, means that over
time social fields are linked to one another, the possibilities of existence
for specific subjects are constrained, and life choices gain an element
of path dependency. By selecting and articulating existing elements in
a given historical context, hegemonic projects attempt to capitalise on
these material interdependences in the interest of forming and repro-
ducing sovereignty.
In the remainder of this chapter I want to explore these points
through two rather different case studies; namely Classical Athens
and the Inca Empire. In both cases, my concern is to show how
hegemony weaves political power into the material interdependencies
of objects, identities, technologies and life-ways. For the sake of
brevity and comparability, in both cases I will emphasise relational

networks involving politics, gender, domestic groups, labour and a
single category of material culture.
Articulating power in classical Athens
Perhaps the most famous discussion of the interdependency of

politics, gender and labour is to be found in Aristotle’s The Politics.
Here, Aristotle designates the household (oikos) as the foundation of
the Greek city-state (polis) and the foundation of the household as
an ‘… association formed by men with these two, women and slaves
…’ (Aristotle The Politics I.ii.1252b). Following his own analytical
method, Aristotle breaks down the complex politics of the polis into
its simplest constituent parts, and there finds a triad relating male
citizens to women and slaves by means of the household. Aristotle’s
view that the polis could be built upwards from the relationships
of the oikos (see Nagle 2006) is interesting since these reproductive
relationships, based on dependency, labour and gender, are precisely
what is excluded from the civic politics of the polis (see pp. 81–5).
Hence, we will begin our exploration by considering the paradox
of citizenship, slavery and gender in Classical Athens under the
democracy (508–322 bc).
Polis, citizen, state?

In the vast literature on the polis as a political form, one frequent theme
is the problematic relationship between ancient Greek understandings
of the nature of the polis and modern understandings of the nature
of the state. By polis I actually mean Classical Athens (508–322 bc),
which was in fact an atypical polis in terms of its large population and
far-reaching democratic institutions. It is, however, the Greek polis
from which most of our surviving evidence originates and, it must be
said, it is an interesting case to consider in it own right.
Beyond politics 75
The polis of Athens incorporated both the city of Athens itself

and the peninsula of Attica (Figure 4.1). For much of the period
from Cleisthenes’ democratic reforms in 508 bc until the dissolution
of Athenian democracy by Antipater of Macedon in 322 bc, Athens
was at least notionally governed directly by its male citizen body.
The Assembly (ekklesia) ratified decrees proposed by the Council
(boule), and its membership was open to all adult male citizens
registered in one of the 139 municipalities (demes) into which
Athens and Attica were divided (see Hansen 1991). Normally, 6,000
or more citizens would attend and vote at Assembly meetings. The
Council acted as an executive body that drafted the agenda for the
Assembly and consisted of 500 members chosen by lot for one year
terms. Even with some councillors serving more than one term,
Mogens Herman Hansen (1991: 249) estimates that in the fourth
century bc one-third of all male citizens over eighteen years of age,
and two-thirds of all male citizens over forty years of age served at
least one term as a councillor. In other words, under the democracy
the governmental apparatus of Athens was the entire body of male
citizens.
Figure 4.1 Map of Classical Greece featuring the polis of Athens

For classic state theory, the governmental apparatus of Athens

presents a problem in that, despite its scale, centralisation and evident
agency, it is not distinguished from its citizen body. In the words of
political theorist Justin Rosenberg, the polis
… does not have any existence other than the political self-organi-
sation of the citizenry. It has no bureaucratic apparatus to which the
decision-making authority of the populace is formally alienated and
which might provide a basis of ‘independent’ interests and capacities.
It is anything but autonomous in this restricted empirical sense. And
yet it talks like a state! (Rosenberg 1994: 79)
While this ‘problem’ has been raised many times, for present purposes
it will suffice to consider recent exchanges between Moshe Berent
(2000; 2006) and several critics (Grinin 2004; van der Vliet 2005;
Anderson 2009; cf. Miyazaki 2007), regarding his claim that Athens is
best classified as a stateless society.
Berent follows Weber in understanding the state as that set of insti-
tutions holding a monopoly on the legitimate use of force in a given
territory (Berent 2000: 258–60; cf. Weber 1978: 54). In other words,
the state is specifically the sum total of the police, army, judiciary, and
legislature, etc. … distinct from the citizen body of a given territory. For
Berent it is precisely this idea of institutions distinct from the citizen
body that does not hold for Classical Athens. In particular, Berent
notes two aspects of Classical Athens that render it a stateless society.
The first is the distribution of the means of coercive force amongst the
citizenry (esp. self-armed hoplite land-owners as a volunteer army),
rather than its concentration in coercive apparatuses (i.e. the police
or a standing army), and a concomitant reliance on volunteerism and
‘self-help’ in law enforcement and civil defence. The second is the lack
of structural differentiation between ruler and ruled, in that most
offices rotated annually between citizens drawn by lot.
Much of the exchange between Berent and his critics hinges on
differing interpretations of the evidence and the initial definition of
the state. For example, both Grinin (2004: 127) and van der Vliet
Beyond politics 77
(2005: 128) argue that the troop of Scythian archers who were among
the public slaves owned by the polis of Athens essentially acted as a
police force. Berent (2006: 144), in contrast, argues that their role was
limited to keeping order in defined contexts and they therefore did not
constitute a police force in the sense of having a general responsibility
for law enforcement.
Looking past these definitional debates we do learn some interesting
things about Athens. As Virginia Hunter (1994) has shown, Berent
is right to stress the limited nature of formal coercive apparatuses in
Athens. However, what the polis did was sanction and legitimise the
use of seemingly ‘private’ coercive force by means of legal decisions and
legislative decrees (Hunter 1994: 188). Indeed, Edward Harris (2007)
goes much further in emphasising the strict limits placed on ‘self-help’
in legal matters and the need for legitimation through judicial and
legislative offices. Justifiable homicide (against robbers or adulterers
caught in flagrante) in Athens was an interesting case of a private
action that existed only through its public legitimation (Adam 2007).
For example, Makoto Miyazaki (2007) argues that an Athenian citizen
could legitimately exercise coercive force, but only if he could represent
himself as an agent of public interest. He writes:
… when Euphiletos, the speaker of Lysias 1 and cuckolded husband,

killed Eratosthenes the adulterer of his wife, he claims that he said ‘it is
not I who am going to kill you, but the nomos of the polis, which you
transgressed and had less respect for than your pleasure’. (Miyazaki
2007: 96)
Hence, the aggrieved husband portrays himself not as one exacting

personal vengeance, but as one performing as an agent of the polis,
enforcing its laws and acting in its interests. To use the terms we
introduced in Chapter 1, the polis provided the ‘principle of intelligi-
bility’ for the ‘private’ actions of citizens carried out as agents of public
interest.
The polis as a virtual authority legitimising private action is
paralleled by the role of the polis as the guarantor of private
exchange relations through the minting and assaying of coins.

For example, Athenian silver coins were tested and counterfeits
removed from circulation by Athenian public slaves with the title
Dokimastes (‘tester’). In both cases, the guarantor of the value of
a coin, or the legitimacy of a court ruling, may have been the
political community, but it was a political community that had
been hypostasized as a collective agent, capable of expressing a will,
making decisions, and maintaining an historical presence across
generations. This point is implied by Mogens Herman Hansen
(1998) in arguing that the word polis could be used to refer to a
transcendent public power in a manner not unlike current uses of
the word ‘state’. It is also argued explicitly by Gregory Anderson
(2009), who contends that in Classical Athens demos (‘people’)
could also refer to a hypostasized power with a collective will and
not just the specific individuals constituting the Assembly of the
polis at any given point in time.
One may protest that Hansen’s and Anderson’s respective interpre-
tations of polis and demos as virtual entities underemphasize the fact
that male citizens were actively conscious of constituting the polis and
of being themselves the demos. Hence, Berent’s argument that public
power and the citizen body were never distinguished in Classical
Athens. Be this as it may, one must accept that the citizen body of
Athens was constituted as a virtual public power in other ways. As
Mogens Herman Hansen writes:
The polis was a society of citizens. It was a male society from which
women were excluded; all foreigners were also excluded, and metics
and slaves, though domiciled in the polis, were not members of it,
a fact of which they were reminded every day of their lives, when
the citizens went off on their own to deal with affairs of state in the
Assembly or the Council or the courts. … Yet every day, when the
meetings to deal with affairs of state were over, citizen, metic and slave
went off to work side-by-side as artisans, traders or farmers: in the
economic sphere the stranger was a part of the society, though in the
political sphere he was not. (Hansen 1991: 62)
Beyond politics 79
Hence, the practice and ideology of citizenship was a foundation not

just for self-government, but also for the governing of others who,
outside of the political domain, constituted a significant component
of a wider Athenian society (especially with regard to production,
commerce, religion and education; see Cohen 2000). Interestingly,
in so far as Athens could be said to have a non-elected civil service,
it was largely constituted by public slaves and the female dominated
priesthood of Athena. Hence, male citizenship was constructed by
separating it from women, slaves and foreigners (metics) in two direc-
tions; first by limiting self-government to male citizens and then by
separating self-governing male citizens from what might be anachro-
nistically termed public-sector employment. This points to male citizen
identity as the site of significant hegemonic activity: something that
requires our closer attention.
Athenian citizenship was primarily acquired as a male birthright,
initially through an Athenian father, but after 451 bc only when
both parents were of Athenian birth. While wealth, status and ability
varied considerably amongst citizen males, the political domain was
defined by a notional equality between male citizens in terms of the
ability and opportunity to participate in the political process: what
Ian Morris (2000: 111, quoting Dahl 1989) calls the ‘strong principle
of equality’. This notional equality was embodied in a ‘middling
ideology’ where the male citizen was made equal with his fellow
citizens through their mutual autonomy, ideally with none rising too
high, nor falling too low in relation to others (Morris 2000: 112–13).
It is true that Athenian literature preserves what is often called an
elite view (e.g. ‘the Old Oligarch’, Plato, Aristotle) that sought to
divide or rank the citizen body according to wealth, occupation,
ability or birth. Indeed, Athens did see short-lived coups to install
an oligarchy on two occasions (411 bc and 404 bc) during the life of
the democracy. However, this elite view does not seem to have ever
formed an effective counter-hegemony to that of the middling citizen,
and in many cases operated within, rather than against, the hegemony
of Athenian citizenship.
The middling ideal of Athenian citizenship was embodied in the

hoplite farmer-soldier, a citizen with enough land to support his
household and to pay the costs of arming himself and participating
in the Athenian army (Hanson 1995: 221–89). However, the citizen
body was not limited to hoplites, including as it did extremely wealthy
landowners as well as those who were relatively poor and of low
status (e.g. thetes), such as artisans, shopkeepers and the landless.
Furthermore, although slave ownership was also an ideal, held by
Aristotle to be necessary for full participation in political life (The
Politics III.v), not all Athenian citizens owned slaves. In effect, Athenian
democracy was a fractious coalition of poor and rich citizens who
occupied different positions in relation to the means of production
(especially control of labour and land).
If we remember our discussion of Gramsci, we can see that Athenian
citizenship was a form of hegemony. Essentially, it represented as
universal the interests of a limited segment of society (native freemen)
for the purposes of political rule, and it did so by the selective articu-
lation of shared cultural resources, in this case the ‘strong principal of
equality’ amongst others. However, in contrast with Gramsci’s views,
the hegemony of Athenian citizenship did not produce a political
apparatus that was directly coterminous with class in the sense of only
investing slave-owners or land-owners with political power (cf. Nafisi
2004). This does not mean that politics in Athens transcended class
or wealth, indeed quite the opposite is true (Davies 1981; de Ste Croix
1981). However, it does highlight the fact that we cannot presume that
political power is simply another name for class or wealth, and instead
we must examine the historical dynamic of their articulation.
Free men and others

The apparent autonomy of the political domain in Athens was achieved
by demarcating it very carefully; promoting notional equality by
excluding as far as possible relations of dependency. Hence, the
sharp contrasts between male citizens, Athenian women, slaves
Beyond politics 81
and foreigners (metics) that structured citizen identity. Under the

hegemony of middling citizenship, relations of mutual and hierarchical
dependency were represented as internal to the oikos (household),
most obviously in the case of relations between husband, wife and
children, but also in the relationship between master and slave. Note,
for example, that even the largest mining ventures and craft workshops
staffed by slaves were organised and understood as extensions of the
household of the owner (Harris 2002). Less directly, resident foreigners
were also dependent on citizen households in the sense that they could
not own land, were required to pay a special tax, and were required by
law to have a citizen sponsor.
The Athenian polis played little direct role in economic production,
and limited its surplus extraction to very specific activities and
segments of society. To a large extent, this was made possible by the
tribute Athens extracted from the Delian League as its de facto empire,
as well as the duties levied on the abundant trade running through
Athens’s port at Piraeus. However, economic production and polis
revenues were also dependent on volunteerism and the ‘private’ actions
of autonomous households, especially the wealthiest ones. A rich silver
strike at the Laurion mines led to a rapid expansion of mining opera-
tions in the fifth century bc that generated considerable revenue for
Athens. However, this revenue came in the form of lease agreements
with individual citizens, who then organised a portion of the silver
ore extraction and processing as an enterprise of their personal oikos,
manned by their own slaves (Hooper 1953; 1968). Similar sources of
revenue included the leasing of public land to (usually wealthy) house-
holds, and various tariffs associated with foreign trade.
Interestingly, direct extraction of surplus did occur in two contrasting
cases. Resident foreigners (metics) were required to pay a regular
head-tax that was a simple obligation to the polis with few implications
beyond the generation of revenue. In contrast, the wealthiest citizens
were expected to use their private resources to fund specific public
projects (termed ‘liturgies’ by scholars) in the form of festivals, perfor-
mances and especially the construction and manning of warships
(Davies 1981; Gabrielsen 1994). To some extent this ‘privatised’ many

of the polis’s major expenditures and gave wealthy Athenians some
latitude in terms of displaying wealth in a socially acceptable, and at
times politically expedient, manner.
Despite the apparent autonomy of politics from production in
Classical Athens, ancient authors were very much aware that the
equality of male citizens, and hence Athenian democracy itself, was
made possible by specific relations of production. Motivated by conflict
between aristocratic landowners and the poor in the early sixth century
bc, the Athenian legislator Solon is credited with admitting the poorest
citizens into Athens’s then oligarchic assembly, abolishing debt-
bondage and reforming land tenure. While the historicity of Solon’s
specific reforms remains an open question (see Blok and Lardinois
2006), archaeological surveys do seem to register significant changes in
settlement and land use in the form of dispersed farmsteads and more
intensive farming practices by the time of Cleisthenes’ democratic
reforms in 508 bc (Bintliff 2006; Forsdyke 2006). In particular, the
increased autonomy of poorer citizens that came with the elimination
of debt-bondage and other forms of direct dependency (e.g. serfdom)
correlates with a rapid increase in Athens’s reliance on slave labour in
the realm of production (Finley 1981; Jameson 1977/78).
Athenian literature of the fifth and fourth centuries bc suggests
an aversion even to wage-labour on the part of Athenian citizens
(Cohen 2002; but cf. Wood 1988: 137–44). Poor citizens might work
as craftsmen or shopkeepers; however, this was generally organised as
self-employment, in contrast to the large workshops owned by wealthy
citizens and staffed with slaves (Harris 2002). Literary evidence suggests
that, when forced to hire themselves out, poor citizens preferred poorly
remunerated seasonal employment such as olive harvesting, as this
did not involve continuous dependence on an employer (Burford
1993: 190–1). Publicly-funded payments for service as a rower in the
Athenian navy, or for attendance at the Assembly, may have provided
another means for the poorer citizens of Athens to avoid selling their
labour directly to fellow citizens. Hence, for Classical Athens, it seems
Beyond politics 83
that democracy was made possible, at least in part, by the transfer

of economic dependency from poor citizens to non-citizens; namely
slaves and foreigners (contra Wood 1988; see Foxhall 2002; Jameson
1992).
While the political apparatus of Classical Athens may have had a
relative autonomy with regard to class, it was directly coterminous with
gender. Athenian women had a certain status as members of citizen
households destined to create further legitimate citizens (Osborne
1997), and as leading figures in interlinked rituals of the lifecycle and
the agricultural cycle (Dillon 2002). However, in terms of the political
life of the polis, women were excluded in that they were not permitted
to vote or participate in the Assembly. Furthermore, in Athens there
was a strongly gendered division of labour in social and biological
reproduction, with appropriate female labour, sexuality and authority
ideally defined and contained by the household.
Athenian literature and art frequently map gender onto a spatial
contrast between domestic and civic domains; principally the house
as the domain of women and the outdoors in general, and key
civic spaces, such as the Assembly, the Agora or the gymnasium in
particular, as the domain of men. In this literature, women operating
outside of the domestic context, or men operating within it, often serve
as exemplars of shame, humour or liminality, as in the case of forensic
speeches regarding adultery, the role of the hetairai (highly cultured
prostitutes), or Aristophanes’ comedies premised on collective political
action by Athenian women (e.g. Lysistrata or Ecclesiazusae = ‘Women
in the Assembly’) (see Davidson 2011). As Lin Foxhall (1994) notes,
this binary engendering of space in Athens was part of a masculinist
ideology, rather than a straightforward description of the lives of
Athenian women and men. This ideology stressed the formation of
autonomous male citizens as heads of households (kyrios) through
civic institutions and homoerotic relationships that drew young men
away from their natal households. In doing so, this ideology ignored
the obvious role of women as widowed mothers and new wives in
forming these same ‘autonomous’ households across generations.
In reality, this engendering of domestic and civic domains was never

a simple case of female oppression and housebound seclusion, as might
be suggested by an uncritical reading of the literary sources. Within the
oikos women both managed household resources held in common and
had independent title to the portion they brought into marriage through
their dowry (Foxhall 1989). Indeed, it was this merging of inheritances
that made marriage and household formation key strategic elements in
the perpetuation of names and property across generations. Outside of
the house women maintained extensive social networks that entailed
moving about the city on visits to the houses of other women (Taylor
2011). Women were also engaged in extra-household tasks that could
require regular trips outside of the house, as in the case of filling water
jugs at civic fountain-houses (Nevett 2011). In rural areas it is apparent
that women on less well-to-do farms would, of necessity, have partici-
pated extensively in outdoor labour associated with agriculture (Scheidel
1995; 1996). Finally, citizen women did, on occasion, engage directly in
extra-household commerce, selling goods in the Agora or taking on paid
employment, although such cases were often seen as indicative of the
poverty and ill fortune of the woman’s household (Brock 1994).
Despite, or rather because of, its ideological nature, the represen-
tation of gender in Athens was a key site of hegemonic activity. As
James Davidson notes:
Whether or not this discourse of the real can be said accurately

to reflect reality, it nevertheless represents a big, important fact in
itself and one that is readily enmeshed with other central discourses
concerned with Greek self-definition, class, politics and citizenship.
(Davidson 2011: 608)
Importantly, this ideology allowed the male head of the household

(kyrios) to be imagined as ‘… the individual empowered to cross the
boundary between household and community’ (Foxhall 1989: 31),
representing the oikos in the polis. This in turn had implications, not
only for the constitution of gender in Classical Athens, but also for the
constitution of politics.
Beyond politics 85
We can now return to Aristotle’s paradox of the polis being consti-

tuted by domestic relationships (between men, women and slaves) that
were explicitly excluded from civic politics. Drawing a sharp divide
between civic and domestic spheres and mapping this to an equally
sharp divide between masculine and feminine identities allowed Athens
to be understood as a collection of households represented by their
male heads. The autonomy of these households allowed male citizens
to participate as notional equals in civic politics, without requiring the
direct redistribution of land or wealth. As we have seen, this autonomy
was achieved in part by limiting direct economic dependency between
citizen households. This in turn was facilitated by the employment of
slave labour (and foreign metics) by wealthy households engaged in
large-scale economic activities.
Houses for citizens?

As the preceding discussion suggests, houses constituted a material
setting of particular significance to Athenian citizenship, one that has
been of particular interests to archaeologists, and one that meshes well
with our present focus on gender and labour (Figure 4.2).
The transition from Early Iron Age Greek houses, with few internal
divisions and a clear reliance on exterior workspace, to the multi-
roomed Classical house with its internal courtyard has attracted
considerable scholarly attention (Coucouzeli 2007; Morris 2000: 280–6;
Nevett 2010a: 22–42). This attention has focused somewhat narrowly
on the timing of the transition, what it ‘means’ in terms of changing
social structures (i.e. the increased importance of the oikos relative to
larger kin structures), and what it ‘represents’ in terms of gender and
citizen ideologies.
In interpreting the courtyard house, archaeologists initially looked
for evidence of ‘women’s quarters’ (gunaikon; Walker 1983), in keeping
with the long-standing, text-based view that Athenian women lived
in ‘Oriental seclusion’ from men along the lines of an Ottoman harem
(see Wagner-Hasel 2003). The existence of a rigid internal gender
Figure 4.2 Greek courtyard house, fifth century bc, Olynthus House
Avii4 (redrawn from Robinson and Graham 1938: Fig. 5)
division within Greek houses was soon questioned on textual (Cohen

1989) and archaeological (Jameson 1990) grounds. Instead, scholars
drew on more nuanced interpretations of Islamic housing to argue that
the key point of gender separation was between residents and guests,
and hence between the inside and outside of the house (Nevett 1995).
For example, figurative representations attest to female veiling outside
the house, with the veil linked metaphorically to the ‘protective’ walls
of the house in a variety of textual sources (Llewellyn-Jones 2007). As
Beyond politics 87
a key indicator of the division between resident and guest, Lisa Nevett
(2010a: 43–62) points to the development of a special room, which
she interprets as the andron (‘men’s quarter’) of literary sources, used
for symposia and to mediate the presence of male guests within the
house. Identified by painted walls, mosaic floors, demarcated space for
couches, an off-centred doorway to prevent lines of sight into other
rooms of the house, and proximity to the main entrance, the andron is
an identifiable feature of urban houses in Greece by the fifth century
bc (Nevett 2010a: 55–6; Coucouzeli 2007 and Morris 2000: 280–6 date
this development as early as the eighth century bc).
Both Nevett (2010b) and Westgate (2007) have noted that the
Classical Greek courtyard house was intentionally oriented inwards,
presenting a closed façade to the street. Its single entrance restricted
access to the andron, where interaction with male guests was mediated,
and to the internal courtyard in which domestic activities were shielded
from view and through which most rooms in the house had to commu-
nicate. This was a space suited to the control of visual and physical
interaction, especially between household members and outsiders.
From the perspective of the male head of the oikos this facilitated the
protection of household resources and the household’s reputation,
as well as the monitoring of female sexual activity within the oikos,
ensuring legitimate heirs, future marriage alliances, and unencum-
bered slave labour. In short, according to both Nevett (2010b) and
Westgate (2007), the courtyard house embodied and reinforced the
citizen ideology of autonomous, self-sufficient households that could
be represented in the political domain by their male citizen heads.
What then are we to make of this connection between Greek houses
and Greek politics?
One limitation of ideological approaches to Classical Greek houses
is that they fail to account for houses as a site of articulation between
politics, economics and identities except in terms of symbolic messages
and social aspirations. In other words, politics occurs discursively and
is reflected materially. Yet the autonomy of the Classical Athenian
household was not just a symbol of appropriate gender relations; it was
also the practical foundation of an economic unit of production and

consumption. The household basis of labour relations (i.e. slavery and
the gendered division of labour) and wealth accumulation are what
allowed the emergence of wealth differences between citizens without
the emergence of relations of dependency. These economic relations
were learned and reinforced in the bodily experience of the house
and household land-holdings. These experiences were embodied in
terms of who moved freely as a member and whose movements were
constrained as a guest, by why and in what manner household members
crossed the threshold into the public domain, and by the actualities of
creating, accumulating, managing and using household resources. In
this sense the house was a key site for constituting, and not just materi-
alising, the material interdependencies of economy, politics, gender
and identity in Classical Athens. These material interdependencies are
what was activated, reinforced and reinscribed through the practices
of citizen hegemony. In this way, politics, labour, gender and domestic
groups ‘hung together’ in a distinct way that was mediated by the
physical form of the Classical courtyard house and the daily patterns of
movement and interaction that this physical form afforded. In Latour’s
terms, the Classical courtyard house was an entity that did not sleep,
creating associations that did not break down (Latour 2005: 70). The
significance of this distinct ‘spatio-temporal fix’ (Jessop 2006) will be
clearer if we turn to the very different case of the Inca Empire.
Labour, gender and power in the Inca Empire
Ruled by a divine king (Sapa Inca), geographically and demographi-

cally immense, possessing a centralised tribute economy lacking any
standardised medium of exchange, and a social system based on
gender complementarity (distinct, but equal, male/female roles), it
would be hard to invent a polity more different from Classical Athens
than the Inca Empire. Yet, much as in the case of Athens, political
power in the Inca Empire required the articulation of distinct social
Beyond politics 89
Figure 4.3 Map of the Inca Empire

fields, such as labour, production, gender and domestic groups. These

articulations were multiple, but come together in a particularly striking
way in a network centred on labour service, chicha (a beer made from
masticated maize), and acllacona (‘chosen women’ who brewed chicha
and wove textiles for the use of the Inca Empire).
The Inca Empire (Figure 4.3) both expanded and collapsed rather
rapidly, although it was founded on a longer prehistory of state-
formation that included the consolidation of distinct ethnic populations
in the Cuzco Valley, Peru, and immediately adjacent regions (Bauer and
Convey 2002). Much of what we know regarding the specific events of
Inca imperial history depends upon accounts written by Spanish and
indigenous authors in the century following Francisco Pizarro’s kidnap
and murder of Emperor Atawallpa in ad 1532. According to these
chronicles, the Inca state entered its imperial phase of rapid expansion
following Pachacuti Inca Yupanqui’s usurpation of the throne and
miraculous defeat of the invading armies of Chanca in ad 1438. Over
the next 90 years the Inca Empire expanded along the Andes mountain
range to incorporate over 2 million km2 and perhaps between 6 and
14 million people (McEwan 2006: 93–6). On the death of the Emperor
Wayna Qhapac (Spanish: Huayna Cápac) in 1528, a war of succession
broke out between his sons Atawallpa and Wascar Inca, leaving the Inca
Empire weakened and divided on the eve of the arrival of the Spanish.
The diversity of the Andes meant that the Inca Empire incorpo-
rated a patchwork of distinct ecological zones and ethnic groups.
In integrating this diversity, Inca imperial policy encouraged stand-
ardisation, while allowing – even promoting – regional differences
in specific domains (e.g. Wernke 2006). The key to Inca policy was
something we have already seen in our discussion of hegemony and in
other case studies, namely the selective appropriation, transformation
and reinscription of shared practices and values (e.g. Jennings 2003;
Silverblatt 1987). Here, the fact that the diverse ethnic groups of the
Andes seem to have shared many practices, orientations and values
with each other, and with the Inca, was strategically important to Inca
imperialism.
Beyond politics 91
One such shared practice was the pan-Andean tradition of labour

service. Andean households were traditionally organised into extended
groups called ayllu that combined kinship, bi-lateral inheritance and
territory to form local communities. Households regularly contributed
labour to assist fellow ayllu members or to undertake collective tasks
of relevance to the community as a whole (Murra 1980: 90–2). In
general, this labour service could take one of two forms. As delayed,
symmetrical reciprocity, called ayni, shared labour implied equality,
with participants continually creating and repaying the debt of mutual
assistance over time (Flannery et al. 2009: 35–6, with references). As
hierarchical, asymmetrical exchange, called min’ka, labour was recom-
pensed immediately, usually with food and chicha, as in the case of
work done for the ayllu head (cuarac) who then rewarded the labourers
with a feast. Importantly, the min’ka arrangement implied lower status
and dependency on the part of those giving labour and receiving food,
distinguishing in this way between owning and working the land (Gose
2000: 86).
The Inca imposed a form of labour service on those they conquered
(LeVine 1987; Murra 1980: 89–186), known as mit’a (‘turn’) in reference
to the fact that each household took their ‘turn’ fulfilling the obligations
placed upon higher-level groupings to which they belonged. Work
included a wide variety of tasks, such as agricultural labour, weaving,
brewing, construction, mining and warfare. Most labour service was
carried out on specific projects by people based in their own commu-
nities. Under these circumstances, the Inca Empire cast itself in the role
of host, providing food and chicha, as well as cloth and other material
goods, to work parties during their period of service. As LeVine
(1987: 15–17) notes, labour service could also require full-time work,
sometimes away from home (e.g. mining, warfare), with communities
rotating the obligation amongst their members and the Inca adminis-
tration providing for the material needs of those in full-time service.
In contrast to this rotating labour service, the Inca Empire expro-
priated the labour of some of its subjects more continuously, thereby
associating them with the Empire in a categorical manner (Murra 1980:
153–86). This included skilled specialists (camayoc) in various activ-

ities, such as craft production; colonists (mitmaq) moved wholesale to
farm, populate or defend new territory for strategic reasons; retainers
(yanacona) who were usually non-Inca elites assigned as servants to
high-status Incas; and finally ‘chosen women’ (acllacona) who are of
particular interest in this chapter.
Acllacona (sg. aclla) or ‘chosen women’ were young girls selected
from amongst the Inca provinces for their beauty and deportment (see,
for details of the Spanish accounts of acllacona: Alberti Manzanares
1986; Costin 1998; Gose 2000; Silverblatt 1987: 81–108; and Surette
2008: 14–66). Once selected, the girls were cloistered in an acllawasi
(wasi = ‘house’), where they were taught by senior ‘chosen women’
(mamaconas, ‘esteemed mothers’) to be highly skilled weavers and
brewers of chicha, as well as knowledgeable attendants in state temples
and shrines (Figure 4.4). Depending upon the status of their families
they may have also been engaged in food preparation and agricultural
work on the lands of temples and the Emperor (Surette 2008: 46–8).
Conflicting ages for selection are given in the Spanish chronicles,
but most suggest that the girls were prepubescent, and all stress the
importance of virginity during their years of cloistering, often drawing
parallels with Christian nuns. It does seem that sexual activity, and
indeed most male contact, was strictly limited for acllacona as the Inca
Empire asserted control over their reproductive potential. Girls of the
highest status (e.g. members of the Inca royal family) were housed in
the acllawasi of Cuzco, associated with the Temple of the Sun. Known
as ‘wives of the Sun’, these acllacona wove cloth that was exclusive to
the Inca Emperor and his Queen (the Coya). Lower status girls were
housed in acllawasi in provincial centres and could be designated
‘wives of the Inca’.
In being separated from their ethnic communities and assimilated
to the governing apparatus of the Inca Empire, acllacona were elevated
to a high status during their residence in the acllawasi. At the end of
their residency acllacona seem to have had one of four fates: 1) most
commonly they were given in marriage to soldiers, nobles or other
Beyond politics 93
Figure 4.4 Acllaconas spinning wool within an acllawasi compound

overseen by a mamacona (source: Poma de Ayla 1936 [1615]: 298)
officials as a sign of the Inca’s esteem, usually, but not always, in their
native province; 2) less commonly they became secondary wives or
concubines of the Inca himself; 3) they remained in the acllawasi and
assumed the role of mamacona, training and overseeing future genera-
tions of acllacona; or 4) they served as sacrificial offerings as part of
the annual festival cycle and in response to extra-ordinary events (e.g.
disease, natural disasters, military defeat). Upon their death, sacrificed
acllacona became oracular deities (huacas), honoured and consulted in
regional shrines.
Each of these four fates served to generate hierarchy and transfer
it to the Inca Emperor through the assimilation of acllacona to the
Inca state. According to the Spanish chroniclers, the Inca Emperor
appropriated the right to allocate women in marriage throughout the
Empire by requiring men to seek official sanction in order to wed.
In this way, the Inca Emperor inserted himself into the superior role
of ‘wife-giver’, and subsumed marriage and household formation to
his Empire. Giving acllacona as brides to reward imperial service or
cement an alliance was both an extension and an amplification of the
Inca Emperor’s ‘wife-giver’ status, fusing reproductive relations with
empire building. The remaining three fates merged acllacona into the
governing apparatus of the Empire even more fully, permanently alien-
ating them from their natal communities except in the transformed
role of living or deified intermediaries (Gose 2000: 88; Silverblatt 1987:
94–100).
My interest in the acllacona stems from the link they provide
between mit’a labour service and the Inca Empire. First, as brewers,
acllacona were central to the provisioning of workers with chicha as
well as food on behalf of the Empire. How far down the social ladder
and how far out into the provinces this extended is not clear. Chicha
has long been a high-value food item in the Andes with ritual and
political significance (Hastorf and Johanessen 1993; Jennings and
Bowser 2009). This relates in part to the difficulties of growing maize
at altitude and to the cultural and religious significance attached to
intoxication. The brewing of chicha on a large-scale, as would occur in
Beyond politics 95
the case of large festivals and the feeding of work parties, is both time
and labour intensive (Jennings 2004). As Jennings (2004: 252) notes,
the labour cost of large-scale production can be reduced by using larger
brewing pots (e.g. 170 litres versus 80 litres), and indeed one finds such
large brewing pots on Late Horizon (Inca period) sites in the Andes.
However, large pots are typically associated with centralised brewing,
as their manufacture and movement stretches the labour capacities of
women in individual households (Jennings and Chatfield 2009). At the
same time, because chicha does not keep or transport well, it must be
brewed near to the site of its consumption. Hence, the Inca Empire and
its agents would have had need of large-scale brewing labour and facil-
ities spread across the Empire in order to sponsor its feasting activities.
Spanish chronicles suggest that at its height the acllawasi in Cuzco
housed 1,500 acllacona, while Pedro de Cieza de León, writing in 1553,
mentions the former existence of acllawasi at 21 different locations,
noting several as housing more than 200 women (Surette 2008: 37–8,
80–2). Pilar Alberti Manzanares (1986: 161) suggests that acllawasi
were located in 28 primary population centres of the Empire, leading
Flannery Surette (2008: 38) to estimate a total population of 6,900–
15,500 acllacona at the height of the Inca Empire (1,500–2,000 for
Cuzco and 200–500 for provincial centres). Hence, it seems reasonable
to suggest that, at least in Cuzco and major provincial centres, acllacona
were available to brew and serve chicha on behalf of the Inca Emperor
on a scale commensurate with the needs of the mit’a system. However,
Gose (2000: 87) thinks that acllawasi were even more widespread than
this, citing evidence from the Spanish chronicles for the presence of
acllawasi even in small settlements in order to suggest that acllacona
provided most of the labour needed to brew and feed those doing mit’a
service in all parts of the Empire.
Very few excavated structures have been identified as acllawasi,
the best example of which is from the site of Huánuco Pampa, a
provincial administrative centre built as a new settlement by the Inca,
and located in Huánuco Department of modern Peru. To the north of
the site’s monumental plaza, excavators identified a walled compound
built of fine-cut Inca masonry and measuring ca. 100m. x 140m. This
compound had a single gated entrance and enclosed a plaza and 50
barrack-like rectangular buildings each measuring between ca. 65m2
and 95m2 (Morris and Thompson 1985: 70–1, Fig. 8). Within this
compound excavators found spindle whorls and bone weaving imple-
ments (Morris and Thompson: 70, Pl. 42–3), as well as ‘[h]undreds if
not thousands’ of large jars believed to be for the brewing of chicha
(Morris and Thompson 1985: 70). The restricted access, large-scale,
repetitive housing, and dense artefactual evidence relating to weaving
and, most particularly, brewing, all support the impression provided
by the Spanish chronicles of acllawasi as distinct institutions capable of
housing more than two hundred acllacona.
Further archaeological evidence points to the wider impact of Inca
food provisioning. Scholars have identified a repertoire of Inca pottery
based on forms and decorative motifs that appears as a distinctly
foreign element in ceramic assemblages in Inca provinces outside of
the Cuzco region. According to Tamara Bray (2003), the most widely
distributed forms within this repertoire outside of the Cuzco region
were the so-called aríbalo, a long-necked jar thought to be for the
carrying and serving of chicha, the pedestalled cooking pot, used for
boiling food, and the shallow plate, a basic serving vessel (Figure 4.5).
Furthermore, Bray (2009: 119–21) notes that, in contrast to the Cuzco
region, in the Inca provinces medium and large aríbalos dominate over
small ones, suggesting that they were used primarily for consumption
of chicha in larger groups.
While Andean communities maintained a mixed-farming regime
throughout the period of the Inca Empire, there is some evidence that
the significance of chica led to an increase in the cultivation of maize
in areas conquered by the Inca (Hastorf 1990). Certainly, maize figured
significantly in Inca tribute demands and was one of the key staples
stored in the large-scale centralised storage facilities found throughout
the Empire (see LeVine 1992). Maize also played a significant role in the
ideology of Inca imperialism. The ploughing and the planting of maize
were ritually linked to warfare as quintessential masculine activities.
Beyond politics 97
Figure 4.5 Inca ceramic figurine of a man carrying an aríbalo of

chicha on his back (photo credit: Werner Forman Archive/Museum für
Volkerkunde, Berlin)
Indeed, the story of the original conquest of Cuzco at the dawn of Inca
history included the introduction of maize by the founding Inca couple
of Manco Capac and Mama Huaco, and was re-enacted annually by the
Inca Emperor by breaking the earth to commence the ploughing and
sowing season (Bauer 1996).
While I have focused primarily on their role as brewers, acllacona
were also renowned as weavers of very fine cloth (see Costin 1998;
Murra 1962; Surette 2008). Under the Inca, all households were
required to produce cloth as tribute. These textiles were stored in
central storage facilities across the Empire and distributed to soldiers

and corvée labourers while they were doing service for the Emperor.
However, acllacona wove much finer, more highly valued textiles using
rarer materials such as the wool of the wild vicuña and, in the case of
tunics for the Inca himself, even bat hair. Decorative patterns were
distinguished by status and by ethnicity, making fine cloth garments
quite literally markers of identity. At the same time, sumptuary laws
made the distribution and possession of this fine cloth a central
dynamic in the micro-politics of desire, favour and reward, bonding
elite subjects to the Inca Emperor and his governmental apparatus.
Brewing and weaving were not, however, merely tools of imperi-
alism made efficient and exclusive by the cloistering of acllacona. It has
to be remembered that in the Inca Empire both brewing and weaving
were women’s labour par excellence. The gendered complementarity of
work in the Andes meant that weaving and brewing were represented
as female contributions to the collective enterprise of the household
and the ayllu, just as agriculture and warfare were represented as male
contributions. Hence, in the first instance, this was labour shared with,
rather than extracted by, men (see Gose 2000; Silverblatt 1987: 3–39).
Women’s labour played a particularly important role in the feeding of
work teams and in the provisioning and organisation of feasts. Hence,
women’s labour was key to both meeting the reciprocal responsibilities
generated by ayni labour relations, and making possible the hierarchy-
forming aspects of asymmetrical min’ka labour service through the
provision of food and chica to teams of subordinate workers.
Irene Silverblatt (1987: 81–108) has argued that the Inca use of
acllacona to appropriate female labour as a tool of imperialism was
part of the transformation of a system of gender complementarity
into a system of patriarchal hierarchy linked to class dominance (via
conquest). Peter Gose (2000), however, has offered a more subtle view,
arguing that the Inca Empire used acllacona to constitute itself in terms
of both male and female modes of generating hierarchy. We have seen
already how the acllacona played a role in defining the Inca Emperor
as the ultimate ‘wife-giver’, a role in which junior males would carry
Beyond politics 99
out subordinate service to senior males in order to eventually be given

a bride. Service to the Inca Emperor took a variety of forms, including
agriculture, construction and warfare. However, all tribute to the Inca
Emperor was cast as labour service through a legal fiction whereby the
Inca claimed ownership of all of the resources of his domain (e.g. fish,
maize, wood, etc.). Hence, in collecting tribute, the Inca Emperor was
not represented as receiving goods, which would imply debt and subor-
dination. Rather, he was represented as receiving in service the labour
expended on extracting, making or raising resources that he already
owned (Gose 2000: 85–6; Murra 1980: 29). Since the Inca Emperor
sanctioned all marriages in the Empire, one means of interpreting
tribute was therefore as bride service: labour given by a junior to a
senior male in return for the right to marry (Gose 2000). At the same
time, provisioning both tributaries and direct labourers with food and
chica prepared by acllacona (as well as distributing the textiles these
same women produced) meant that the Inca Empire could thereby
also turn tribute into hierarchy-forming min’ka labour service, casting
itself in the female roles of cook, brewer and weaver by means of its
acllacona agents.
Already, prior to the formation of the Inca Empire, households,
labour, gender and chica were related through a series of material
interdependencies that had developed over many centuries (Hastorf
and Johanessen 1993). These networks were particularly evident in
local and regional politics (Gose 1993). By inserting himself into such
networks as the ultimate ‘wife-giver’ and as the ultimate min’ka host,
the Inca Emperor and his agents went a long way towards neutralising
intermediary polities and articulating households, labour, gender and
chicha directly with Inca imperial sovereignty. In Gose’s words, under
Inca imperial hegemony, household and state ‘… did not embody
fundamentally different social principles or moralities. … Rather,
they were related as microcosm to macrocosm, in which the state
appeared as a “generous” super-household of the senior generation,
with abundant land and daughters to distribute to prospective sons-in-
law’ (Gose 2000: 93).
Summation
One of the most interesting contrasts between Inca imperial hegemony

and the citizen hegemony of Classical Athens is the manner in
which the former depended on the fusion of domestic and political
domains as much as the latter depended upon their sharp distinction.
In a checklist like Renfrew and Bahn’s table this might appear as a
problematic difference to be resolved, or as a culturally variable trait
to be ignored. Yet, as we have seen, in each case these contrasting
demarcations of domestic and political domains were central to the
production and reproduction of sovereignty. In each case, relations
of material dependency were assembled and made to work in the
interests of sovereignty by means of practices, strategies and technol-
ogies that might be collectively labelled as hegemonic. We have
seen that hegemony as an act of assembly, of making things ‘hang
together’ in the interests of sovereignty, involves relational networks
of material dependency between people and material culture, as well
as modes of production, cultural practices, historical dispositions and
inherited traditions. In the next chapter we will move on to explore
more thoroughly the analysis of hegemonic practices, strategies and
technologies.
5
Spectacle and routine
If France can be guided to peace and kept tranquil by shows, shows

may in the end be as useful to them as Parliaments.
‘Government by Shows – The Paris Fêtes’, Illustrated London News,
Saturday August 21, 1852 (Issue 5745): 138–9
Filling out forms, registering land, even paying taxes, might be

considered the equivalents of sacrifice: little ritualised actions of
propitiation by which one wins the autonomy to continue with one’s
life.
David Graeber (2007: 21)
Having discussed what we mean by sovereignty (Chapter 1) and

hegemony (Chapters 2 and 3), as well as how all these things ‘hang
together’ in given historical blocs (Chapter 4), I now want to turn
our attention more directly to the question of the practices, strategies
and relationships of sovereignty. If we are interested in analysing
the practices of sovereignty in premodern contexts, we must first
address a rather particular problem. As we noted in Chapter One,
Michel Foucault defined governmentality as a specifically modern
development; but what then are the distinctly premodern practices of
governance? Looking closely, we find that on this point scholars divide.
As is often cited, Foucault himself characterised premodern society
as a ‘society of spectacle’ where ‘power was what was seen’ (Foucault
1977: 187). Similar is Jürgen Habermas’s (1989: 1–14) contrast between
the development of a ‘public sphere’ in post-eighteenth-century Europe
and what he terms the ‘representative publicity’ of medieval and
absolutist Europe. According to Habermas (1989: 27–31), the public
sphere is an imagined neutral space distinct from the private world

of personal relationships and the political world of the state. It is a
third space in which issues of public good can, in theory, be brought
forward and rationally debated without reference to the identity of the
protagonists. The development of a public sphere hinged on certain
key developments of modernity, such as the idea of the state as a
doubly impersonal institution that administered society (see Chapter
1) and the strict division of public and private domains encouraged by
industrial capitalism. For Habermas (1989: 7–12), the public sphere
contrasted with a premodern European society in which public good
was derived from ‘higher’ principles (e.g. nobility, kinship, chivalry
and, ultimately, divine authority) that were embodied publicly in the
speech and actions of a select group of elite representatives. Hence,
Habermas implies that, in the case of premodern Europe, authority
was dependent on its representation before the public rather than by
the public (McCarthy 1989: xi)). In other words, it was dependent on
spectacle.
The corollary of this focus on the visual nature of premodern
authority is Anthony Giddens’s (1985) claim that the ancient state was
limited by its inability to penetrate the daily lives of its citizens. In
James Scott’s (1998: 2) words, ‘[t]he premodern state was … partially
blind; it knew precious little about its subjects, their wealth, their
landholdings and yields, their location, their very identity.’ In other
words, the necessity of visible power derived in part from the ancient
state’s inability to monitor and manage populations in their mundane
activities and thereby shape political subjectivity in the manner of the
modern state.
This view of premodern political authority as bound up in spectacle
is reflected in the focus on display, monumentality and explicit,
visible ideology in the archaeological analysis of premodern polities.
Recognition of the importance of spectacle in premodern contexts
cross-cuts otherwise opposing views regarding what spectacle does
and why political authority exists. For example, both those who
see public spectacle as promoting social solidarity and collective
Spectacle and routine 103
identification (e.g. Inomata 2006) and those who see it as a tool of

elite manipulation (e.g. Demarrais et al. 1996) agree that the visual
and aural experience of an explicit ideology was the primary means
through which premodern political authority was constituted.
In contrast, scholars of so-called bureaucratic empires (e.g. Rome,
the Inca Empire, Mesopotamia from at least the Early Dynastic III
period, China from at least the Western Zhou Dynasty) often stress the
elaborate and far-reaching measures employed by institutional agents
within these polities in order to document, track and manage the daily
minutiae of production, taxation and labour service (e.g. D’Altroy
1992; Eisenstadt 1963; Feng 2008; Steinkeller 1987). In such polities,
temple and palace institutions seem to play a role in the management
of everyday life that sits uncomfortably with the image of the ancient
state as a thin veneer of spectacle.
Here we might contrast Foucault’s society of spectacle with a society
of routine in which political authority is manifest through its ability
to structure, organise and manage various aspects of everyday life. If
spectacle is power made visible through representation, then routine is
power made invisible through regularisation. A concern for routine is
evident in managerial (Wright 1977) and rational choice (Blanton and
Faragher 2008) approaches to the early state, where the provision and
administration of ‘public goods’, such as subsistence security or judicial
mediation, is held to both justify and necessitate the existence of the
state. However, one also finds an emphasis on the routine manifes-
tations of political authority in practice-based approaches, which
emphasise the spatial and temporal structuring of daily life for political
ends through practical means such as building projects or landscape
modification (Monroe 2010).
Ultimately, the choice between spectacle and routine is a false one,
as there is a continuum of practices between these two extremes that
are manifest to different degrees in different polities and different
historical contexts. However, it is not enough to simply say that the
practices of sovereignty in premodern polities can be placed along a
continuum defined by spectacle and routine. In this chapter I hope to
go further and show that spectacle and routine are not independent
kinds of practices but are, in the end, directly linked through the
medium of social performance.
Performance and social reproduction
The terms ‘performance’ and ‘performative’ have been widely used in

archaeological literature (e.g. Joyce 2005). Yet, despite this ubiquity, the
precise implications of a performative orientation for archaeological
interpretation remain ambiguous and contested, especially when one
turns to the analysis of political relationships. As Takeshi Inomata and
Lawrence Coban note (2006: 12–16), social theorists have employed
the concept of performance to analyse embodied human practice all
along the continuum from spectacle to routine; taking us from the
structured artifice of formal theatre to the performative utterances
of everyday speech. In sketching an ‘archaeology of performance’,
Inomata and Coban (2006: 16) prioritise scale and context in marking
off spectacles as community-oriented and community-forming perfor-
mances worthy of special attention for their potential as collective
political acts. Ian Hodder (2006: 81–3), in contrast, rejects the very
idea of distinguishing spectacle from routine, especially on the basis
of its public context, scale or purported political efficacy. For Hodder,
the structured patterns of movement, activity, decoration and burial
attested (and constrained) by houses at Neolithic Çatal Höyük indicate
a regulating of the residents’ bodies that was no less performative or
political than the public rituals of ancient kingdoms (Hodder 2006:
83). For Hodder, spectacle is simply ‘a showing and a looking’ (Hodder
2006: 82) and performance is ‘a dimension of action that bridges to
meaning and communication’ (Hodder 2006: 85).
Hodder’s argument is problematic in that spectacles are not merely
‘a showing and a looking’, but also a ‘doing’. This involves human
bodies/beings directly. Hence, the markedness of spectacles in terms of
scale, place and sensorial intensity makes a difference to their effects;
to how they are experienced and to how they reference and orient
relations between human bodies/beings (Houston 2006: 135–9; Kus
1992). The point is not to construct a typology of practices that divide
the continuum from routine to spectacle, but to recognise that when
embodied practice is formed and performed with different content,
in different contexts, at different scales and with different degrees
of markedness, it makes a difference to its effects and to its point of
reference. Even the apparently repetitious world of Neolithic Çatal
Höyük was punctuated by less regular events, such large-scale feasts
(Hodder 2005) and perhaps even off-site rituals of a collective nature
remembered in the murals and curated fauna for which the site’s
houses are duly famous (Hodder 2006: 85–7).
Hodder is correct to emphasise continuity in the performative
nature of both spectacle and routine, in the sense of habituating bodily
practices. Indeed, this continuity is the central organising principle of
Paul Connerton’s influential book How societies remember (Connerton
1989). Connerton is interested in the transmission of memory as a
shared and social phenomenon. He focuses on what he calls habit
memory as bodily skills and dispositions that are cultural, in that they
must be learned from others (e.g. walking, playing the piano, bowing,
waving good-bye), but which are performed habitually and in a manner
that is not wholly dependent on symbolic representation. Connerton
organises his text in a manner analogous to my use of spectacle and
routine, examining two extremes through the categories of commemo-
rative ceremonies and bodily practices. Connerton shows that, in the
case of both commemorative ceremonies and bodily practices, habit
memory forms an important substrate linking the formation and
transmission of collective memory in any society.
Less clear is precisely what is being remembered when bodies are
culturally constituted through habit memory. Certainly, collective
memory includes symbolic representation through the memoriali-
sation of specific events, and the repetition of narratives that convey
shared concepts of identity and of social and cosmological order.
However, Connerton includes much more than this in his attention
to what he terms ‘incorporating practices’ (Connerton 1989: 72–3).

By incorporating practices, Connerton refers to information trans-
mitted intentionally or unintentionally by the bodily activity of one
person in the presence of another, as in the case of smiling or
conversing. He contrasts this with ‘inscribing practices’, such as writing
or drawing, where information can be stored and transmitted in the
absence of the author. As Connerton (1989: 78–9) admits, there are
problems with treating inscribing and incorporating practices as two
mutually exclusive and exhaustive categories of transmission practices,
something that becomes clear when we introduce material culture into
the mix (see pp. 106–7). However, as is evident from even a moment’s
reflection on what is incorporated by the category of incorporating
practices, Connerton is concerned with social memory, not simply
as the remembrance of things past, but rather as social reproduction
itself: the memory of how to live in the world in a particular way.
The significance of this view of social memory becomes more
clear if we consider the role of material culture in social memory so
conceived. Andrew Jones (2007) has considered the place of material
culture in the constitution of social memory as an embodied practice.
Jones does this by means of C. S. Peirce’s tripartite classification of
signs, focusing in particular on material culture as an index (Peirce
1931: 531). An index is a sign that conveys meaning as a product of
its close association with what it is signifying, as in the truism ‘smoke
means fire’ (Peirce 1931: 531). Indices, in this sense, are not necessarily
fixed in nature, but can both accrue and recede over time and place by
means of material associations (hence, in the early twentieth century it
was also said in some quarters that ‘smoke means progress’).
Elsewhere (Routledge 2004: 154–5) I have discussed ‘kingly things’ in
the Iron Age Levant as indices of kingship in explicitly representational
terms, that is to say as material culture that called kingship to mind
via its associations with royalty. However, Jones’s focus on embodied
memory opens up some new possibilities in terms of understanding
material culture as indices that ‘call to body’ as much as they ‘call to
mind’. As Jones (2007: 53) points out, material culture can involve
inscribing practices in Connerton’s sense of transmitting information

in an objectified (disembodied) form, but material culture is also
physically engaged with and hence can involve incorporating practices
as well. These incorporating practices may elicit what Connerton
(1989: 22–3) terms personal and cognitive memories (i.e. explicitly
representational memories), due to the indexical associations of the
material culture with past contexts and practices. However, physical
engagement will almost certainly also elicit Connerton’s habit-memory,
in Heidegger’s sense of the ‘readiness-to-hand’ of objects (Heidegger
1962: 95–101). Engagement with material culture elicits habit memory
in how to deploy tools, how to move through space, how to comport
oneself, where to look, and when to speak. Equally, engagement with
material culture could fail to elicit such habit memory, highlighting
one’s foreignness, lack of knowledge, awkwardness, displacement, etc.
Instinctively knowing which fork to use for your salad at a formal
dinner implies much more than the immediate mechanical process of
eating.
Embodied experience is performed in different contexts with
different implications for social reproduction; implications that are
resident in both symbolic representation and bodily dispositions. If
social reproduction is the memory of how to live in the world in a
particular way, realised across different contexts with different implica-
tions, then hegemony is the attempt to link these contexts together and
thereby shape what is and is not remembered in the name of certain
interests and not others. In the cases that interest us, sovereignty is part
of what hegemony remembers: the transcendent nature of political
apparatuses, their necessity, legitimacy and authority in key domains,
and their relation to relations of force. The question is how is political
authority remembered?
Let us begin with spectacle as the site most frequently linked to the
reproduction of political authority in premodern polities. What does
spectacle do, and wherein lies its political efficacy? Clifford Geertz
(1980: 121–36) has famously argued against the idea that public
spectacle can be analysed as a kind of false consciousness; a bit of sugar
to coat the bitter pill of power. According to Geertz, Western political

theory has, since the sixteenth century, remained wedded to ‘the great
simple’ (Geertz 1980: 134) that politics is ultimately only about power
in the form of coercion, violence, domination and mastery. For Geertz
the reduction of symbolic representation to a political instrument, one
that masks, obscures or ameliorates the real workings of power, cannot
make sense of the ‘theatre state’ of nineteenth-century Bali, where
‘[p]ower served pomp, not pomp power’ (Geertz 1980: 13). According
to Geertz, what mattered in Balinese royal ritual was the constitution
and reproduction of a coherent view of reality, with the king as the
exemplary centre holding together an orderly social hierarchy, one that
arguably did not exist separate from these ritual contexts.
Geertz’s central critique is an important one. The cultural poetics
of any public performance are its central intent and any interpretation
that ignores this fact in the interest of uncovering what is going on
‘behind the mask’ of ritual will be a misinterpretation. At the same
time, Geertz simply inverts the equation that he critiques, turning a
mere mask into a mere masque. Geertz’s insistence that the symbolic
coherence of royal rituals resides at a level wholly distinct from the
political reality of daily life means that in the end he provides us with
no means of understanding either why state rituals took the particular
form that they did, or why they should have proved compelling, effica-
cious or necessary (Bloch 1987: 294–7).
As an alternative, Inomata and Coben are keen to shift our analytical
focus away from examining public performances as a ‘closed system of
their own aesthetics’ (Inomata and Coben 2006: 17) in order to under-
stand spectacle as a distinct kind of cultural performance. Spectacle, in
these terms, is a cultural performance whose political effects, such as
communal integration or mass communication, might be said to reside
to a large extent in the scale and sensorial markedness of the event.
Treating spectacle as a cross-cultural category grounded in a common
human body raises difficult questions regarding the universality of
sensorial (and somatic) experience (cf. Joyce 2005; Moore 2006). Yet
these difficulties are not so problematic if one frames the category
of spectacle historically; sensorially marked events may not have

collective political effects of necessity, but they have certainly had such
effects at many times and places in many different cultural contexts.
Acknowledging the sensorial impact of spectacle does not, however,
take us very far. It does not, for example, explain the intellectual
activity surrounding the content of spectacles, nor does it explain their
diversity or culturally specific efficacy.
Both Adam Smith and Susan Kus and Victor Raharijaona recognise
this problem directly in relation to the political efficacy of Urartian
figurative representations of ritual spectacles (Smith 2006) and Merina
royal public discourse (Kus and Raharijaona 2006). For Smith (2006:
126), ‘to attempt to understand political spectacle without an under-
standing of the visions that not only strike the eye, but stir the
imagination, would seem to miss something fundamental about the
phenomenon’. Similarly, Kus and Raharijaona conclude that if ‘we come
to understand that a sensorial ethos is implicated in any alternative way
of being in the world, then we are challenged to come to understand the
“how” of the specific materials used to create and sustain the political
…’ (Kus and Raharijaona 2006: 323). This recognition that the content
of public spectacle was meaningful and appealing to specific people in
a specific context brings us back to our discussion of hegemony.
In Chapter Two we saw how hegemony hinged on the selective
articulation and transformation of cultural resources that rang true
because of their generation in the realm of common sense (i.e.
day-to-day existence). Both Smith and Kus and Raharijaona explain
the political efficacy of performance in their case studies in a similar
manner. Smith (2006: 126–7) argues that the bountious offerings of
produce and animals carried by processions of worshippers in Urartian
figurative art were a key to the political efficacy of these images, as
abundance in the practical realm of daily production was harmoni-
ously united with consumption in the cosmological realm of deities.
Kus and Raharijaona point out how the naming of places, rhetorical
performances and movement between and within places were insepa-
rable components of royal public discourse in the early Merina state,
playing on local values that viewed speech, memory and landscape as

continuous with one another.
Hegemony resides in the constitution of a moral order and a
coherent view of reality, much as in the case of the idealised patterns
standing at the centre of Geertz’s cultural analysis. Yet, as I have argued,
this order is selectively constituted by articulating elements relevant to
some experiences while excluding those relevant to others. In other
words, one need not, indeed one cannot, choose between cultural
poetics and political efficacy as each is embedded within the other. A
coherent view of the world by definition excludes the incoherence of
social and economic asymmetry, resistance and disenchantment.
If ritual performance is aimed at constituting or reproducing a
moral order, and addressed to a hegemonic view of reality, questions
of audience and intention become less pressing than when political
efficacy is seen to reside strictly in the conveyance of a compelling
message from ruler to ruled. Ritual performance can be addressed to
a broad or restricted audience, to the performers themselves, to the
gods, to past ancestors or to future descendants. What matters is that
spectacle participates in, and thereby embeds its participants in, the
construction and reproduction of a particular moral order.
This said, we have still not addressed the specific question of political
authority. Public spectacle occurs in communities of very different
sizes with very different political structures. It is not, therefore, the
exclusive domain of centralised and transcendent political apparatuses.
Rather than spectacle having the function of producing/reproducing
political authority, it seems much more the case that political authority
must inscribe itself onto (i.e. entrain) public spectacle in order to be
reproduced.
It is interesting that in the case of both Urartu and Imerina one is
dealing not only with public events, but also with the marking and
memorialisation of these events in enduring media (e.g. stone) under
the direction and sponsorship of a political apparatus. In the case of
Urartu, Smith (2006) is dealing exclusively with representations of
spectacle. As with much royal art, these representations have a generic
quality that need not refer only to specific past events, and indeed may
not refer to any specific event. Generic banquet scenes, processions of
tribute and pious offering scenes can provide models or summations
of spectacle in which actual spectacles participate and to which actual
spectacles aspire.
Even when representing specific places and events, the memori-
alisation of spectacle can shape and transform those places and
events. Ömür Harmansah (2007), for example, provides an extended
discussion of expeditions to the ‘source of the Tigris’ at Birkleyn in
eastern Turkey by Neo-Assyrian kings, and their commemoration of
these expeditions in reliefs and inscriptions both at the cave of the
water source itself (Dibni Su) and back home in various Assyrian
cities. The source of the Tigris was already a significant place within the
imaginative landscape of Assyria, since most of the Assyrian heartland
was situated along, and dependent on, the Tigris River and its major
tributaries. Harmansah (2007: 181–4) suggests that the inscribing
practices of Assyrian kings transformed this already significant site by
incorporating it into the rhetoric of Assyrian kingship. The continual
inscription of Assyrian monuments at the source of the Tigris, and
at home, transformed it into a ‘site of memory’. In the imaginative
landscape of Assyria the source of the Tigris became a place tied to the
king himself through the actual (via expeditions), and virtual (via their
representation) spectacle of his transcendent body projected outwards
to sacred and significant places on the edges of the empire (Harmansah
2007: 195).
Inscribing practices seek to commemorate and fix spectacle in
particular ways. They also transform the places of spectacle, and
inscribe political authority on spaces in a manner that begins to
move us along the continuum of performance away from spectacle
and towards routine. As in the case of Shang oracle bones, spectacles
become entrained as indices of political authority. For example, like
so many of the royal rituals of Imerina, the erection of stones to
commemorate collective agreements and significant foundational acts
was not so much a practice invented by royalty as one appropriated,
and thereby transformed, by royalty. Indeed, monolithic stones, erected

to commemorate founding treaties or the ancestral tombs that define
villages remain focal points for communal gatherings, judicial proce-
dures and collective ritual to this day in highland Madagascar (e.g.
Graeber 2007: 69–72; Kus and Raharijaona 1998). In the nineteenth
century, under the monarchy, when agreements were reached a stone
was erected and each party paid a tax in silver (hasina) to the monarch
(Graeber 2007: 408n. 26). In this way the monarch became the
guarantor of the agreement and was associated with the stone, which
thereafter persisted as a feature of the landscape of a given village or
field.
To summarise, when considering the techniques of sovereignty
in premodern contexts, our choice is not really between societies
of spectacle and societies of routine. Certainly, the degree to which
techniques of sovereignty are encountered in the context of one’s daily
routine will vary considerably between polities and historical contexts.
Some polities track their citizens closely and others seem to exist only
in flurries of pomp and circumstance. However, in all cases we are
dealing with social performances that involve both symbolic represen-
tation and habit memory; both callings to mind and callings to body,
as it were. What matters is that the disparate contexts in which people
remember, or are forcibly reminded of the place of political authority
in their world are linked by reference to a coherent hegemonic order.
We can, therefore, define three points on the continuum from
spectacle to routine. The first is the production of objects of direct
intellectual reflection. These are the messages of utility and pleasure, of
morality and public good, or of intimidation and fear, communicated
through speeches, texts, images, performances, and built environments,
amongst other means. These explicit spectacles serve to communicate
messages of political authority. However, they also serve to regularise
and routinise that authority through habitual bodily participation
(e.g. as the audience for commemorative ceremonies), through the
creation of settings and backgrounds (e.g. plazas, monuments and
temples), and through the entrainment of media, contexts, objects
and even time (e.g. festivals and calendrical observances) via their
hegemonic associations. This routinisation of spectacle is the second
point on our continuum and provides a bridge to the third: those more
mundane techniques of sovereignty, such as the enumeration, taxation
and regulation of subjects, more commonly considered as the routine
manifestations of political authority.
What is interesting about these three points on the continuum
from spectacle to routine is how often they prove to be inseparable in
any given moment of social performance. Most particularly, spectacle
and routine do not divide between ideological and practical activities.
Praying and eating are both constituted in equal measure by elements
of symbolic representation and habit memory. The interpretive signifi-
cance of this fact will become clearer by briefly considering the
relationship between ritual and water management in Classic Maya
polities.
Royalty, routines and rituals: The place of water in

Classic Mayan polities
Mayanists have shown a strong interest in the archaeology of perfor-

mance (e.g. Houston 2006; Inomata 2006), and, honestly, who can
blame them. Classic-period Mayan centres were structured around
a core of built forms, such as plazas, stepped temple platforms, ball
courts and inscribed stele, which seem oriented to facilitating and
representing ritual events. By the Classic period (ad 250–900), the
interaction of built spaces and images was highly developed, linking
Mayan cosmology and local rulers in the context of collective festivals
and ritual enactments of varying scales and levels of exclusivity
(Sanchez 2005). Stele positioned around plazas and reliefs carved on
building facades and stairways (along with polychrome vessels and
wall paintings) depict deities, mythological figures, rulers and sacri-
ficial victims, as well as dancers and other ritual performers (Houston
2006; Inomata 2006; Schele and Miller 1986). This emphasis on
Figure 5.1 Map of the Classic Maya heartland with sites mentioned
in text marked
royal-sponsored festivals and public ritual evident in Classic Mayan

centres fits with a model of Mayan polities as weakly bounded and
radiating outwards from rulers who compete to control lesser rulers
and a dispersed rural population with shifting, or cross-cutting,
allegiances (e.g. Inomata 2004).
The built environment of most Classic Mayan centres also included
sophisticated and extensive systems aimed at harvesting (Scarborough
and Gallopin 1991), spreading (Dunning and Beach 1994), or
controlling (Davis-Salazar 2006; French et al. 2006) the flow of
water. While sub-tropical to tropical in terms of vegetation and
rainfall, precipitation in the Maya heartland is highly seasonal, with
a marked dry season from January to May. Approximately 70 per
cent of the Maya territory (i.e. Petén and the Yucatán) is a lowland
limestone shelf, creating a classic karstic environment where access
to surface water is limited but subterranean water is abundant (Veni
1990). Much of the remaining Maya territory is made up of volcanic
highlands and foothills, where watercourses are more abundant, but
deeply incised or subject to seasonal flooding. Hence, water was
available to the Maya, but only at certain times and places, and in a
manner that required knowledge and favoured active management
(Figure 5.1).
Not surprisingly, water is also one of the core elements in Mayan
cosmology, imagery and ritual (Brady and Ashmore 1999; Fash 2005).
For the Maya, water provided the link between terrestrial, celestial and
subterranean worlds, most notably in the form of mountains on one
hand and caves, springs and sink-holes (cenotes) on the other (Brady
and Ashmore 1999; Vogt and Stuart 2005). Water glyphs, and water-
related imagery such as water-lilies, were widely used on stele and
architectural reliefs (Fash 2005; Schele and Miller 1986; see Figure 5.2).
Similarly common was imagery associated with caves and mountains,
widely viewed as the abode of the rain god Chaak and sources of both
ground water and rainfall (Adams and Brady 2005; Brady 1997; Vogt
and Stuart 2005). At a number of sites, including Chichén Itzá, Copán
and Dos Pilas (Brady 1997; Brady and Ashmore 1999, Davis-Salazar
Figure 5.2 Si(j)yaj K’in Chaak II impersonating the water-lily serpent.

Machaquilá Stele 4 (source: Photo by Linda Schele. Reproduced by
permission of the Foundation for the Advancement of Mesoamerican
Studies Inc.)
2003), the public architecture seems to have been laid out to incor-
porate, or align with, a cave or spring.
Caves, springs and cenotes, all subsumed under the word ch’een
in Mayan hieroglyphs (Vogt and Stuart 2005: 157–63), are clearly
associated with central institutions of political power in Classic Mayan
polities. Most famous is probably the so-called ‘Cenote of Sacrifice’
at Chichén Itzá, where votive offerings of gold, jade, pottery censers
and imported goods, as well as some human victims, were deposited
in large numbers from the Late Classic through Post-Classic periods
(Coggins and Shane 1984). However, royal (or a least elite) ritual use
of caves can also be seen in many other contexts, such as the murals
and glyphs from Naj Tunich cave complex in the Petén district of
Guatemala (Stone 1995). At the same time, as is the case amongst
recent Mayan communities (Adams and Brady 2005; Vogt and Stuart
2005), archaeological evidence suggests that the ritual use of caves
extended beyond strictly elite practices and prior to the Classic period
(Brady and Prufer 2005).
In general, ethnographies, especially the work of Evon Vogt
(1969) in highland Chiapas, have documented the important role of
small-scale, water related ritual in the everyday life of recent Mayan
communities. Archaeological evidence suggests that such small-scale
water related rituals, especially those associated with springs and
caves, may have played a similar role in the daily life of ancient Mayan
‘commoners’ in the lowlands. Such rituals were therefore available for
appropriation and transformation by elites in the Classic period for the
hegemonic purposes of royal display (Davis-Salazar 2003; Lucero 2006;
Scarborough 1998).
A number of scholars have linked these two aspects of water, arguing
that water management and water rituals interacted to constitute and
legitimise the authority of Mayan rulers. The control of water and
the control of people is an equation that has been made frequently
by scholars over a very long period of time, most famously by Karl
Wittfogel in his book Oriental Despotism (1957). The position of
the Maya in the ‘irrigation management = despotism’ argument has
always been ambiguous, as such arguments focused on large-scale

riverine irrigation, a technology that is generally unsuited to the karstic
lowlands and incised volcanic highlands of the Mayan homeland. In
the case of the Maya, the emphasis has not been on the administration
of water canals for the purposes of irrigation, so much as on the
management of still-water reserves during the dry season.
In mapping water catchments and reservoirs at a number of Mayan
centres, most notably Tikal, Vernon Scarborough (1998; 2003: 108–15;
Scarborough and Gallopin 1991) has argued that in the Classic period
Maya rulers created large convex watersheds that captured and stored
run-off rain water (Figure 5.3). This was achieved by building urban
centres on elevated ground, using watertight masonry, and channelling
run-off water downhill from core architectural complexes into large
built, or modified natural reservoirs. Scarborough and Gallopin (1991:
659) define three kinds of reservoirs in Maya centres: 1) central
precinct reservoirs, located in the settlement epicentre adjacent to
the primary plazas and public architecture; 2) residential reservoirs,
adjacent to residential neighbourhoods; and 3) bajo-margin reser-
voirs. The last of these reservoir types were located on the outskirts of
sites adjacent to bajos, which are wetlands created by surface depres-
sions in the limestone bedrock of the lowlands (Dunning et al. 2006).
The degree to which ancient Mayan farmers modified bajos for the
purposes of intensive agriculture (e.g. raised fields) is a point of some
controversy; however, at the very least the margins of bajos formed
important agricultural land in many parts of the lowlands (Dunning
et al. 2006). Central precinct reservoirs usually had the largest storage
capacity and were most closely associated with water symbolism and
elite ritual performance. They were also connected by a combination
of canals and topography to bajo-margins in such a way as to either
recharge bajo-margin reservoirs in the dry season or to directly irrigate
bajo-margin fields (Scarborough and Gallopin 1991: 360). Hence, as
the name suggests, central precinct reservoirs played a central role in
concentrating and providing both potable water and irrigation water
during the dry season.
Spectacle and routine
Figure 5.3 Schematic cross-section of a Classic Maya convex run-off water collection system (redrawn from
Scarborough 1998: Figure 2)
119
Scarborough (1998: 136–7) notes that these water management

systems concentrated water in a quantity and quality that otherwise
would not exist and in a manner that allowed for its control by resident
elites. Yet, according to Scarborough, the dispersed nature of Mayan
settlement and land-use left rural sustaining populations with alterna-
tives to dependence on elite controlled reservoirs and hence people had
to be drawn into urban centres. To this end, Mayan rulers appropriated
and elaborated water-related rituals that were centrally important to
the world-view and social reproduction of the Maya.
For Scarborough (1998: 149), Mayan centres, in which the run-off
water from temple platforms and plazas filled reservoirs, were under-
stood as ‘water mountains’: man-made versions of the mountains
from whose caves and springs both watercourses and rain clouds were
said to emanate. For example, Scarborough points to the depiction
of mountains composed of flowing water in the Tlalocan mural at
the central Mexican site of Teotihuacan, and to the fact that altepetl,
the Nahuatl (Aztec) word for village or community, translates as
‘mountain of water’. By recreating a key site in the ideological landscape
of Mesoamerica, one that produced a stable supply of water like its
counterpart in nature, and then leading key rituals central to repro-
ducing those supplies of water over time, Mayan rulers inserted
themselves into the very centre of the universe.
Scarborough’s version of the ‘water = power’ argument is what
we might term the moderate version. He accepts, for example, that
Mayan rulers could never fully monopolise water management in local
communities (Scarborough 2003: 113). Scarborough also remains
equivocal regarding the extent to which elite power was primarily
founded on the control of water, and the extent to which dispersed
settlements were dependent on centralised reservoirs during the dry
season.
Such equivocation is not found in what might be termed the
extreme version of the ‘water = power’ argument put forward most
forcefully by Lisa Lucero (2006). Lucero’s argument largely parallels
Scarborough’s, in that she argues that Mayan rulers collected and
concentrated still-water reserves and that they drew dependents to

urban centres and legitimized their control of still-water reserves
by appropriating and elaborating key rituals. However, Lucero goes
much further in terms of specifying the nature and mechanics of the
relationship between Mayan rulers and Mayan commoners.
Lucero argues that broad categories of Mayan polities (i.e. minor
centres, secondary centres and regional centres) correlate directly with
the form and extent of centralised water management within those
polities. In essence, she argues that the degree to which Mayan elites
were able to extract surplus from Mayan commoners was directly
correlated with the degree to which Mayan commoners were dependent
on elite controlled still-water reserves during the dry season. Like
Scarborough, Lucero gives a large role to ritual performance, seeing it
as necessary to drawing people into major centres and placing Mayan
rulers at the heart of key activities for reproducing the Mayan world
(e.g. water purification and renewal). Yet, despite the attention she gives
to ritual, in the final analysis ritual performance still functions primarily
as a legitimising false consciousness in Lucero’s argument. Power, which
she defines principally as the ability to extract surplus, resides in the
practical dependence of commoners on elite-controlled water.
The limits of Lucero’s mechanistic view of the ‘water = power’
equation become evident when we begin to look beyond major lowland
sites with large-scale reservoirs such as Tikal. For Lucero, the presence
or absence of large regional centres is negatively correlated with the
availability of dry-season water sources. Local centres, such as Saturday
Creek in central Belize, do not develop into large centres because elites
could not control the dispersed year-round water sources available to
farmers (Lucero 2006: 72–3). As Kevin Johnston (2004) has pointed
out, this ignores the evidence for household wells, and hence water
self-sufficiency, found amongst small dispersed sites in the lowlands.
Lucero (2006: 37) states that such wells are found in territories where
regional centres did not develop, but this ignores Johnston’s larger
point that wells are difficult to identify at small rural sites and the ones
that are known may only be the tip of the iceberg.
Whether or not Johnston is correct, Lucero’s arguments seem to run

into trouble when we look beyond the lowlands and consider regional
centres in well-watered piedmont and riverine zones. At Copán,
in western Honduras, a large regional centre of the Classic Maya
period developed in the water-rich Copán valley. Amongst the more
prominent water management features of this site are the conduits,
drains and causeways that provide drainage and flood control for the
principal architectural group and elite residential clusters of Copán
(Davis-Salazar 2006). Also prominent are the artificially modified
lagoons that serve as stable freshwater sources and are the focal point
of a number of residential clusters in and around Copán (Davis-Salazar
2003). As Barbara Fash (2005; Fash and Davis-Salazar 2006; Davis-
Salazar 2003) has pointed out, the lagoons in particular are associated
with both water and royal, or elite, iconography such as water-lily
headdresses, tuun glyphs and half-quatrefoil frames (both symbols of
water-bearing caves). Fash and Davis-Salazar (2006) have suggested
that these water-hole groups may have formed key social units within
Copán with water-management devolved to elites within these groups,
who may have taken primary responsibility for the offerings and feasts
attested by archaeological remains within and around these lagoons
(Davis-Salazar 2003).
Davis-Salazar (2006) notes that drainage and flood control installa-
tions exhibit very different construction techniques and materials, both
between different residential clusters and between such clusters and the
principal architectural group at Copán. She credits this difference to
the various water-hole groups being responsible for the construction
and management of drainage systems in their own neighbourhood.
The rulers of Copán are said to have coordinated water management
at a valley-wide level, perhaps by overseeing interaction between these
various water-hole groups by means of a council (Fash and Davis
Salazar 2006). Rulers also took central place in the ritual cycle viewed
as essential for renewing and controlling water supplies.
In both Copán and Tikal, water management and water ritual have
been linked to each other and to a central political authority, but in
distinct ways and with reference to distinct mechanisms. One could

also consider the case of Palenque, another large regional centre located
in a water-rich area in Chiapas (the site’s Maya name, Lakamha’, means
‘wide waters’; see French et al. 2006: 145). The site has 41 known
water sources within its boundary, as well as nine watercourses that
flow through it (French 2007; French et al. 2006). It also possesses a
rich collection of iconography related to water-ritual, especially in
the ritual connection made between sweat-baths and caves at the site
(Child 2007). Like Copán, water management at Palenque is oriented
towards flood control, although the system at Palenque seems to be
more uniform across the site and, therefore, perhaps centrally planned
(Davis-Salazar 2006). However, water-management at Palenque does
not create resources to draw farmers in from outlying areas (where,
after all, water would be available all year round). Rather, at Palenque,
water-management helped to create the site itself, in that the main
plaza is raised above drainage conduits that canalise the flood-prone
watercourses, and extended over uneven ground that would not
otherwise support large-scale architecture (French 2007). In other
words, large-scale water-management is linked to political power, not
because it ensures that centralisation is the only option, but because it
creates the possibility of living in a particular way. Without centralised
water-management there would still be dispersed Maya farmers, but
there would be no Tikal, Copán or Palenque.
Lucero (2006) argues that riverine centres differ from those in
the lowland karst zone in that settlement density is higher in the
immediate vicinity of sites such as Copán or Palenque. According to
Lucero, this limits the mobility of farmers and allows rulers to control
access to land as well as to water. By making riverine centres a special
case, however, Lucero overlooks the clear symmetry between highland
and lowland regional centres. She does this in the interest of preserving
a causative link between her settlement types and the centralised
control of still-water reserves in the lowlands. But is this link worth
preserving? Making the control of still-water reserves determinative
inevitably leads us into ‘chicken or egg’ arguments regarding the
origins of water dependency vis-à-vis social inequality; it also reduces

ritual to a kind of false consciousness and it draws our attention away
from the common ‘world building’ practices of Maya rulers in both
water-rich and water-poor regions.
It would be more productive to reframe the relationship between
water management and water ritual in Mayan kingdoms in terms of our
discussions of hegemony and the techniques of sovereignty. There is no
doubt that water was a central resource (both materially and culturally)
for the Maya, and that rulership amongst the Maya was constituted
through practices that included the physical and ritual management
of water (amongst other key resources such as maize, ancestors and
blood). Supplying water and ensuring its renewal via essential rituals
was a kind of public good, one that articulated widely shared Mayan
ideas and practices regarding how the cosmos was ordered and how
one survived and prospered in this world. It provided, therefore, one
basis for consent to sovereignty in Gramsci’s terms (see Chapter 2).
At the same time, centralised rule was not the only, or even the best,
way to live in this region. Indeed, resource distribution, especially in the
lowlands, favours dispersed settlement. Hence, the physical and ritual
management of water was a public good relative to particular interests
and in the perceived absence of certain options (like seasonal mobility).
The hegemony of Mayan kings required the constant assertion that ‘the
way things are, is the way things are’. This involved strategic practices
asserting the centrality of kingship to the reproduction of the social and
cosmological order, practices that raised the cost of non-compliance,
and practices that caused kingship to fade into the background as the
setting for everyday life through the built environment, the seasonality
of ritual observance, and the provision of stored water.
Such techniques of sovereignty cannot be partitioned into the
spectacular and the routine, as each involves the possibility of both
active reflection and embodied habituation. Drawing water ingen-
iously collected from the run-off of a temple platform and standing
in a crowd to watch rulers impersonate the water-lily serpent contain
elements of both the spectacular and the routine. These elements not
only co-exist, they also co-determine one another. Water management,

in both the physical and cosmological sense, routinised the spectacle of
grand architecture and ritual observance by ordering lives to be lived
in a particular way. These were lives in which one drank water from
this reservoir, irrigated one’s fields from that channel, and ensured
the renewal of rainfall by making these offerings in that temple, or
by participating in this dance in that plaza. In other words, the basic
‘scaffolding’ of life in a Mayan centre constituted the centrality of
Mayan rulers, even as that centrality provided a meaningful expla-
nation for the order and continued existence of this particular way of
life. That this was not the only life that Mayans could imagine or live is
evident in the difficulties Mayan rulers experienced in attracting and
retaining subjects.
Summation
In this chapter, I have argued that the practices, strategies and relation-
ships that constitute a given political order involve elements of both
symbolic representation and habit memory; they are simultaneously
spectacular and routine. It remains for us to consider how this all
fits together, how sovereignty emerges from spectacular and routine
practices that draw together and create interdependent elements
within a given historical bloc under a specific logic of hegemony. Of
particular concern is something we have yet to address directly, the
violence at the heart of sovereignty, those relations of force that set
limits on what interests hegemony will articulate. It is not insignificant,
for example, that besides the life-giving substances of water and maize,
the other substance central to Mayan kingship was blood. How can
the spectacle of violence, which inherently pits the interests of some
against others, be routinised and subsumed within the social perfor-
mances of hegemony? In the next chapter we will take up this question
through an extended case study of the Royal Tombs of Ur.
6
Routine lives and spectacular deaths

The Royal Tombs of Ur
Hoe, do not start getting so mightily angry! Do not be so mightily

scornful! Is not Nisaba the Hoe’s inspector? Is not Nisaba its overseer?
The scribe will register your work, he will register your work. Hoe,
whether he enters five or 10 giĝ in your account, Hoe – or, Hoe,
whether he enters one-third or one-half mana in your account, Hoe,
like a maid-servant, always ready, you will fulfil your task.
‘The Debate between Hoe and Plough’
(ETCSL translation t.5.3.1: lines 186–93)
Between 1926 and 1934 Leonard Woolley excavated and documented

some 2,110 tombs dating to the third millennium bc and positioned
adjacent to the temple of Nanna the moon god at the site of Ur
(modern Tall al-Muqayyar) in southern Iraq (see Woolley 1934; 1955;
Figures 6.1 and 6.2). Sixteen of these tombs, all belonging to the Early
Dynastic IIIA period (ca. 2600–2450 bc), were designated as ‘Royal
Tombs’ by Woolley due to: 1) their constructed tomb chambers;
2) the abundance and wealth of their grave goods; and 3) the elaborate
nature of their interment rituals, including evidence for the ritual
killing or mass suicide of numerous attendants (Woolley 1934: 33–42).
Eighty years on, the ‘Royal Tombs of Ur’ remain one the world’s
most spectacular archaeological discoveries (e.g. Zettler and Horne
1998). In part, this is a product of the exquisitely crafted objects in
rare materials that the tombs contained and the archetypal narrative
of Woolley’s patient endeavour in excavating and conserving these
objects. However, our continued fascination with the Royal Tombs is
Figure 6.1 Map of Mesopotamia showing sites mentioned in text
also due to the incongruity, indeed the mystery, of intentional killings

regularised in large-scale funerary ceremonies. In what follows I will
use the Royal Tombs of Ur to explore how the routinised violence
of sovereignty is revealed in the exceptional violence of funerary
spectacles in Early Dynastic Ur.
Routine lives and spectacular deaths 129
Figure 6.2 General plan of the ‘Royal Cemetery of Ur’ (redrawn from
Woolley 1934, Plate 274)
Tombs, attendants and mortuary spectacles
Woolley (1934: 33–8) argued that all 16 of the Royal Tombs of Ur

shared a common burial pattern. Each tomb contained a deep vertical
shaft, with an access ramp on one side, at the bottom of which was
constructed in stone or brick either: a) a single-chambered tomb with
an open forecourt in front; or b) a multi-chambered tomb consisting of
a primary and one or more secondary chambers. In 13 of the tombs the
excavators noted the presence of more than one individual; in the other
three (PG/580, PG/1236 and PG/1631) human bones were too poorly
preserved or highly disturbed to identify individuals. In many cases the
number of bodies interred together was very high, with the six most
populous tombs containing between 22 (PG/800) and 74 (PG/1237)

bodies. In many cases it was also possible to identify a principal burial,
due to its location in the primary tomb chamber, its burial in a coffin
or on a podium, and the wealth of its associated grave goods. Most
famous in this regard is Queen Pu-abi, so identified by an inscribed
cylinder seal. Most of the bodies in these tombs, however, were located
in secondary chambers or in the open forecourts in front of the single
chamber tombs. Woolley (1934: 33) dubbed these forecourts ‘death
pits’, suggesting that the bodies therein were attendants killed for the
express purpose of inclusion in their master’s or mistress’s tomb.
Many scholars, including Woolley (1934: 41) himself, have noted
that the highly structured nature of the Royal Tomb deposits appears
to be the outcome of elaborate ritual acts (e.g. Baadsgaard et al. 2012;
Cohen 2005; Vidale 2011; Winter 2009). Bodies were both carefully
laid out and carefully dressed, at times with jewellery, personal items
or functionally specific tools, weapons and instruments executed
in valuable raw materials with a very high level of craftsmanship.
There also appear to be clear groupings and associations of bodies.
Most obvious are the groups of armed ‘soldiers’ near several of the
entrances, the bodies of ‘grooms’ found beneath the skeletons of oxen
teams still harnessed to carts or sledges, and those of ‘musicians’ found
adjacent to Ur’s famous lyres and harp. The limited number of skeletal
remains examined by Molleson and Hogson (1993; 2000; 2003: 125–7)
suggested task-specific muscular development and osteopathologies
for certain individuals, as well as a relative absence of such features
on the principal occupants of tombs, including for example Queen
Pu-abi. While uncertainty regarding the criteria for curation make
the Ur skeletons a problematic sample, these results do support the
commonsense inference that in life individuals buried in the Royal
Tombs occupied the same roles (e.g. ‘porter’, ‘cart/sledge driver’, elite)
in which they had been cast in death.
Beyond these well-known examples, there also appear to be systematic
patterns in the distribution of jewellery, especially headdresses, within
the tombs (see Gansell 2007). For example, whereas most of the crania
Figure 6.3 Tomb PG/789 showing distribution of hair adornment on

bodies (after Woolley 1934, Plate 29)
in tomb PG/789 (Figure 6.3) were associated with either silver hair
rings or no headdresses, at least eight of the eleven bodies laid out in a
row against the outside of the south-western wall of the tomb chamber
were associated with gold hair ribbons and wreaths and only one was
associated with silver rings (Woolley 1934: 65–7, Pl. 29). Similarly,
while the majority of the bodies in the centre of the ‘Great Death Pit’
(PG/1237) were dressed with gold hair ribbons and wreaths, none of
the bodies in the outermost rows (along the north-east and south-west
Figure 6.4 Tomb PG/1237 (‘The Great Death Pit’) showing distri-
bution of hair adornment on bodies (after Woolley 1934, Plate 71)
walls of the pit) were dressed in this manner, most being adorned with
gold earrings and chokers (Figure 6.4).
Overall, the jewellery suggests that the attendants were adorned and
positioned within the tomb groups in a highly structured manner that
may reflect their roles in life or in the funerary ritual (see also Vidale
2011). Alternatively, these patterns could reflect an overall aesthetic
for the burial events planned around the structured alteration of
adorned bodies (e.g. placing attendants in rows in terms of contrasting
jewellery).
Woolley suggested that the regular arrangement of the bodies and
their active association with musical instruments, weapons or teams of
oxen indicated that these attendants had died where they lay without
violent struggle after participating in an elaborate funerary procession.
This, and the common association of cups with bodies in the Royal
Cemetery, led Woolley to posit that the attendants in the Royal Tombs
had voluntarily committed suicide by drinking some form of poison
or sedative (Woolley 1934: 35–6). In Woolley’s words the ‘death pits’
represented a royal court ‘… going with their master to continue to
serve under new conditions, possibly even assuring themselves thereby
of a less nebulous and miserable existence in the afterworld than was
the lot of men dying in the ordinary way’ (Woolley 1934: 42).
That the attendants went peacefully to their deaths, as Woolley
believed, is, however, open to question. Recent analysis of CT-scans
of the crania of one male soldier (body 50 in PG/789) and one female
attendant (body 52 in PG/1237) in the University of Pennsylvania
Museum’s collection, suggests that both were killed by blunt force
trauma, probably with a narrow-bladed axe (Baadsgaard et al. 2011:
32–7). Evidence also suggests that the female may have been heated
and treated with cinnabar (mercuric sulphide) as a preservation
measure (Baadsgaard et al. 2011: 37–8).
Woolley (1934: 142) had already noted that many of the bodies
in the Ur cemetery appeared to have been ‘lightly burned’, and a
reanalysis of the curated skeletons in the Natural History Museum
(London) confirmed a prolonged heating of the bodies, much as in the
case of the University of Pennsylvania Museum sample (Molleson and

Hogson 2003: 123). Hence, it seems likely that the regular arrange-
ments of the bodies in the best preserved ‘death pits’ (e.g. PG/789,
PG/800, PG/1237) were the outcome of the post-mortem preservation
and manipulation of the attendants’ bodies, rather than peaceful, in
situ, deaths (cf. Baadsgaard et al. 2011: 39–40). In light of all of this
evidence, the Royal Tombs appear to be a kind of macabre tableau,
planned around the primary interment and realised with the bodies
of attendants killed and preserved specifically for this purpose (Vidale
2011).
Clearly, in the case of the Royal Tombs of Ur, we are dealing with
large-scale spectacles that even in the present day retain their ability
to both fascinate and disturb. In the words of Roger Moorey (1977:
40) this was ‘clearly a special rite for special people’. Yet, admitting the
uniquely spectacular nature of these burial practices simply underlines
the problem of how and why such rites occurred. The attendants’
deaths were not random acts of violence but highly ordered and
regularised events. What practices and relationships made these deaths
possible? How were these attendants constituted as ‘killable subjects’
(Agamben 1998), either in their own minds or in the minds of others
(cf. Pollock 2007a: 89–90)?
Some scholars have looked to formal rituals attested in Mesopotamian
texts in order to explain the striking burial practices of the Royal
Tombs. Early on, it was suggested that the tombs might contain
substitute kings, whose ritual killing to save the sitting king from
bad omens is sporadically attested in the second and first millennia
bc (Frankfort 1948: 264, 400n. 12). More commonly, scholars have
wondered if the tombs might represent the interments of the principal
figures of sacred marriage rituals tied to the New Year festival, whether
directly as part of the ritual (e.g. Moortgat 1949) or at the end of their
lives as a reflection of the elevated status such a role bestowed upon
them (Moorey 1977: 40).
Other scholars have connected the funerary rites of the Royal
Tombs with descriptions of death and the netherworld in Sumerian
literature (e.g. Tinney 1998). In the fragmentary Death of Gilgamesh,

for example, it appears that the hero king Gilgamesh is buried with:
… [h]is beloved wife, his beloved children, his beloved favourite and
junior wife, his beloved musician, cup-bearer and … his beloved
barber, his beloved … his beloved palace retainers and servants and
his beloved objects …’ (ETCSL translation t.1.8.1.3: Nibru fragment
lines 1–7)
On the very next line of the same tablet Gilgamesh presents valuable
gifts to Ereškigala, lady of the underworld, as well as to numerous
deities and dead notables. Presumably this allows Gilgamesh to assume
the role of ruler in the underworld, as was foretold to him on his
deathbed. In The Death of Ur-Namma (ETCSL t.2.4.1.1), the king
Ur-Namma is said to be buried with his donkeys and to present gifts
and host an opulent banquet for the important figures of the under-
world on his arrival in order to secure a worthy position.
More sociologically minded scholars explain the exceptional nature
of the Royal Tombs by arguing that they occur at a moment of
historical crisis in the transition from charismatic to hereditary (or
religious to secular) rule. Andrew Cohen (2005: 147–54) has argued
that the exceptional display of the Royal Tombs is a by-product of the
emergence of new forms of kingship based on lineage and inherited
authority, albeit still embedded in temple ritual. In this early stage the
transition between rulers following the king’s death was an unstable
and uncertain time that needed to be negotiated by means of elaborate
funerary rituals emphasising the power, legitimacy and continuity
of the ruling lineage (cf. Redman 1978: 297–8). While accepting a
similar historical context, Bruce Dickson (2006) posits the more blunt
function of intimidation and terror for these funerary rituals, calling
them ‘public transcripts of cruelty’ aimed at frightening subjects into
continued submission. In contrast, Susan Pollock (2007a; 2007b)
argues that the Royal Tombs represent the last gasp of institutional
households based on principles of charismatic leadership (such as
divine selection), with their conspicuous destruction of wealth and
labour making an extreme statement against the inherited forms of

power emerging during the Early Dynastic period.
Two sorts of problems plague both those who appeal to culturally
specific ritual and those who appeal to a general social transition. The
first is empirical. In the case of formal rituals, as Woolley (1934: 37–42)
himself already argued in his publication of the Ur Cemetery, the links
drawn to textual sources are very weak. References to the killing of
substitute kings in Assyria occur more than a millennium after the
Royal Tombs, are ambiguous, and do not entail large entourages of
attendants. Similarly, the sacred marriage rite, where the king assumes
the role of the god Dumuzi and a priestess the role of the goddess
Inanna in order to ensure the annual renewal of agricultural fertility,
did not entail ritual killing, focuses on a sacred couple rather than
single principal burials as found in the Royal Tombs and, according
to some recent scholarship, may have existed only as a literary image
rather than as an actual ritual (see Lapinkivi 2004 for literature).
Unfortunately, although The Death of Gilgamesh may provide a
literary attestation of attendant burials as seen in the Royal Tombs
(but see Marchesi 2004: 156–60), it provides no real insight into why
such practices occurred. Gift giving and banqueting on the entry into
the underworld in The Death of Gilgamesh and most especially in The
Death of Ur-Namma may provide an internal rationale for the wealth
of the grave goods found in the Royal Tombs, as well as the clear
link between the tombs and contemporary banqueting scenes (see
pp. 141–8). However, one is still left to wonder how and why this could
justify the mass killings attested in the Royal Tombs.
In terms of explaining the Royal Tombs as an ideological response to
a social transition, we should remember that the presumed trajectory
from ‘charismatic’ religious leadership to inherited and institution-
alised secular/military leadership in Mesopotamia is largely a scholarly
reconstruction. Some scholars posit an internal evolution spurred
on by the necessities of military conflict and aggressive expansion
(e.g. Frankfort 1948: 215–26; Jacobsen 1957). Others posit a growing
dominance of northern Mesopotamian traditions of inherited kingship
over southern Mesopotamian traditions of temple-based leadership

and divinely-sanctioned selection (Heimpel 1992). While either of
these may have happened, currently available evidence does not allow
us to presume that either trend did happen (Zettler 1998: 29). As is well
known, the first phases of urbanised social life in Mesopotamia, namely
the Late Uruk (ca. 3400–3100 bc) and Jemdat Nasr (ca. 3100–2800 bc)
periods, already attest most of the material trappings of centralised
rule. Besides cities, large temple precincts and our earliest written texts,
the city of Uruk also attests standardised images of the so-called ‘priest-
king’ in a variety of media (Vogel 2008: 83–134). This figure, wearing a
net skirt and a headdress oddly reminiscent of a bowler hat, is depicted
in a variety of key kingly activities that are both religious and martial
in character. It is true that the word for palace, é-gal (lit. ‘big house’),
first appears in the Early Dynastic IIIA period (ca. 2600–2450 bc),
roughly contemporary with the Royal Tombs, but the import of this
fact is unclear, especially given the limited and ambiguous nature of the
archaic texts found prior to ED IIIA.
In the ED IIIA–B periods, the three primary titles for city leaders
in Sumerian, en, ensik and lugal, are used simultaneously and at times
interchangeably, by the rulers of distinct Sumerian city-states (see
examples in Cooper 1986). More to the point, even if the earliest
Sumerian city-states were ruled by someone called an en and this
office was displaced in the Early Dynastic period by the offices of
ensik and lugal, we have no direct evidence for how individuals were
appointed to the office of en. Early Dynastic royal inscriptions show
that both ensik and lugal could be inherited, or at least held by father,
son and grandson. The en, in contrast, may have been an appointed,
charismatic leader legitimised by his religious role (e.g. Heimpel
1992), but in truth we do not know. In fact, our best evidence for
claims of divine selection, of being selected ‘from among the myriad
of people’, to use Cooper’s translation (1986: La 5.18; La 9.1), come
not from Uruk or from priestly contexts, but from the standard titular
of the Early Dynastic IIIB (ca. 2450–2350 bc) rulers of Lagash. Most
of these rulers can be shown to be the direct descendants of previous
rulers of Lagash and all of them claim the title ensik or lugal, or both.
In other words, in the fragmentary evidence available to us the only
rulers claiming to be chosen by the gods are those who seem to have
inherited the title from their fathers. The supposed evolution from
charismatic to inherited authority cannot be demonstrated directly,
but must be presumed to exist in the many gaps conveniently present
in the available evidence.
This said, even if we cannot presume a religious/charismatic to
secular/institutional transition, it remains the case that the Early
Dynastic Period, and especially ED III, is a period of significant social
and political change. For example, regional surveys indicate that both
the size of urban centres and the proportion of the population living
in such centres in southern Mesopotamia peaked during the ED III
period (Pollock 1999: 45–77). Hence cities, and presumably also their
temple institutions and military and political leaders, were playing a
larger role in the lives of more people than ever before. Furthermore,
ED IIIB inscriptions make clear that the rulers of these urban centres
were competing fiercely to control land, people and each other;
attempting to forge new, larger-scale political units via conquest and
patronage. In other words, one can make a rather strong case that
the Royal Tombs were built at a time when political institutions and
community identities were in flux.
So, one might still explain the exceptional nature of the Royal
Tombs in terms of some sort of legitimation crisis on the basis of
broader evidence for social change in the ED III period. However,
the exceptional nature of the burial rites attested in the Royal Tombs
presents a second, more significant problem if one is seeking to situate
these rites within some sort of general explanation (be it formal ritual
or social crisis). If the Royal Tombs are the outcome of some general
social or political transition in Southern Mesopotamia, then why do
we not find other Royal Tombs in, say Uruk, or Umma or Lagash?
Certainly, sampling and survival play a major role, but as several
scholars have stressed (Moorey 1977; Selz 2004), even within Ur these
burial rites constitute an exceptional case.
Elaborate and wealthy third-millennium tombs have been found

elsewhere in Mesopotamia (e.g. Eickhoff 1993; Pollock 1999: 205–17).
Indeed, tombs containing carts and oxen/donkey teams are known (if
poorly documented) from Susa in south-western Iran (Mecquenem
1943: 103–4, 122–3) and Kish in northern Babylon (Moorey 1978:
104–10). The three (semi-) published chariot tombs from Kish have
been compared to the Royal Tombs at Ur, as they may have been
constructed as a deep shaft with a ramp (Moorey 1978: 104–6) and
there are multiple bodies in several of the tombs. However, to quote
Seton Lloyd (1969: 48), the Kish tombs were ‘badly excavated, the
excavations were badly recorded and the records were correspond-
ingly badly published’ (Lloyd 1969: 48). As a result, it is very difficult
to reconstruct burial practices in these tombs, and indeed, after
examining the unpublished field records, Guillermo Algaze denied
any parallels with Ur beyond the presence of animal-drawn carts (see
Algaze 1983/4: 149–54). Where scholars do agree is that, in terms of
material wealth and overall organisation, the chariot burials at Kish are
not at all comparable to the Royal Tombs of Ur (Moorey 1978: 105–6;
Algaze 1983/4: 153–4).
Moorey resolves the problem posed by the exceptional nature of
the Royal Tombs by stating that the key may lie ‘with a cult practice
special to Ur’, since ‘[t]he Sumerian city-states each had their own
peculiar traditions and we are still far from understanding many of
them in their own terms’ (Moorey 1977: 39). Moorey is correct in the
narrow sense that these burial customs are specific to Ur and their
proximate causes may never be known given the available evidence.
At the same time, Moorey’s solution is inadequate in the larger sense
of understanding how and why the conspicuous consumption and
ritual killings attested to in these tombs came to be seen as a suitable
expression of whatever local circumstances may have prevailed in Early
Dynastic Ur.
Despite their exceptional nature, the Royal Tombs are clearly struc-
tured by, and embedded in, symbols, values, materials and ideas that
were widely distributed in Early Dynastic Sumer. We must, therefore,
140
Figure 6.5 ‘Peace’ side of the Royal Standard of Ur showing a banquet scene (source: ©The Trustees of the British
Museum)
ask how the exceptional spectacle of ritual killings and conspicuous

consumption found in the Royal Tombs might have arisen from, and
found acceptance within, the more routine workings of Sumerian
society in the mid-third millennium bc In other words, we need
to look at how this exception might prove the rule of rule in Early
Dynastic Sumer.
One rather striking feature of the Royal Tombs is the way in which
they reference elite banqueting practices in multiple ways. Banquet
scenes are a generic theme in Early Dynastic art (Schmandt-Besserat
2001). Typically, the core of each scene is a pair of seated figures
drinking from cups or from a common pot via straws and being
attended to by servants. Banquet scenes are amongst the most common
themes on cylinder seals from the Royal Cemetery of Ur, including
the seal of Pu-abi herself. Beyond this, one finds banquet scenes
elsewhere in the Royal Tombs as inlaid decoration on the sound boxes
of lyres (Woolley 1934: Pl. 116) and most famously on the ‘Standard
of Ur’ (Woolley 1934: 266–74). Here the so-called ‘Peace’ side of the
Standard includes a banquet scene, with a procession of servants
bringing bounty to a larger-than-life-size figure drinking with his
guests in the upper register (Figure 6.5).
Many scholars (Baadsgaard et al., 2012; Cohen 2005; Pollock 2003;
2007a; Selz 2004) have commented on the ways in which the bodies
and grave goods in the Royal Tombs have been laid out to suggest
participation in a banquet. Andrew Cohen (2005: 82–93, 167–220), for
example, has shown that the ceramic assemblages in the Royal Tombs
are largely limited to serving and eating vessels, with a notable paucity
of cooking and long-term storage vessels. Woolley (1934: 36) himself
commented on the frequency of drinking vessels in the Royal Tombs,
with many being proximate to the hands of individual skeletons.
Interestingly, one also finds many distinct artefacts in the Royal Tombs
that are themselves depicted in banquet scenes.
One rather notable example is the occurrence of bull-headed and
bull-shaped lyres in the Royal Tombs (Woolley 1934: 249–58, Pls
107–21), along with depictions of such lyres in banquet scenes on
Figure 6.6 Soundbox of lyre (U. 10556) from tomb PG/789 showing
a satirical banquet scene with animals and mythological creatures
(source: University of Pennsylvania Museum, Image Number 150888,
Object Number B17694)
cylinder seals (Woolley 1934: 21–2, Pl. 193,), on the ‘Standard of Ur’
(Woolley 1934: 266–74, Pl. 91) and on the sound box of one of the
lyres itself (Woolley 1934: Pl. 105; see Figure 6.6). Bull-shaped lyres
are also depicted at other mid-third millennium sites on cylinder seals
(Amiet 1980: Pl.91:1200–1201) and stone plaques (Boese 1971: Taf.17:
1), showing that this very particular object was intimately linked to
the representation of elite banqueting practices across Early Dynastic
Sumer. Through such dense, and very specific, cross-referencing we see
how deeply embedded the Royal Tombs were in the themes, imagery
and material culture of elite life in mid-third millennium Sumer.
Indeed, one could go so far as to say that the carefully placed bodies
and objects of the Royal Tombs were simply another medium for the
representation of banquet scenes.
Banqueting in early Mesopotamia was linked to two key moral
requirements, namely feeding the gods and feeding the dead (Jagersma
2007). These obligations included regular food and libation offerings,
as well as large-scale festivals and feasts. Interestingly, libation scenes
on cylinder seals (Amiet 1980: Pl. 100–3) and stone plaques (Boese
1971: Taf. 18: 1–4; 21: 4) bear some parallels to banquet scenes, with
priests and other elite mortals acting in the role of servants pouring
out liquids before seated deities holding cups. Irene Winter (2009) has
pointed to sets of pouring vessels in many of the Ur tombs that match
those of libations scenes. While Winter’s emphasis on the multiple uses
of such pouring vessels is suitably cautious, it remains likely that these
burial rites involved libations like those performed for deities.
With their analogical connection to the provisioning of the gods and
their direct connection to the provisioning of the dead, funerary feasts
would seem to be doubly marked as significant events in the ritual life
of Mesopotamia. Yet, neither this symbolic content, nor the various
social and political functions ascribed to feasting and commensal
politics in the comparative literature, seems to satisfactorily explain the
exceptional nature of the Royal Tombs.
Brian Hayden, for example, has argued that elaborate funerary
feasts are cross-culturally significant in what he terms trans-egalitarian
and early state societies (Hayden 2009). For Hayden this significance
stems from the fact that, in directing attention to the dead, funerals
are lifecycle rituals that allow morally acceptable ostentation. As such,
they provide a context for expanding and reinforcing alliances through
the gifting of food, as well as a context for signalling group wealth and
success (Hayden 2001; 2009). In particular, Hayden’s ‘political ecology’
asks that we look past the specific symbolic content of the Royal Tombs,
and focus on the supposed adaptive function of elaborate funerary
feasts as a stage for forming and affirming alliances and clientship.
In keeping with Hayden’s cross-cultural model, elite funerals in
Early Dynastic Mesopotamia do seem to have involved large disburse-
ments of food and drink to lower-ranking individuals. Two ED IIIB
(2450–2350 bc) texts from Girsu show that the reigning queen Šagšag
twice dispersed large amounts of bread and beer to lamentation
specialists (gala), ‘wives of elders’, male relatives of the deceased and
female temple servants (a total of 618 people) involved in mourning
rites for Bara-namtara, her predecessor as queen (Cohen 2005: 56–7,
157–62; Jagersma 2007: 293–4; Selz 2004: 198–9). Admittedly, it is not
clear that these disbursements were viewed as acts of hospitality or as
payment for service as mourners. For example, in his famous reforms,
Šagšag’s husband Uru-inimgina, king of Lagash, seems to stipulate a
set fee in bread and beer to be given to lamentation specialists and ‘old
women’ during rites of mourning (Cooper 1986: La 9.2). At the same
time, the wider corpus of ED IIIB texts from Girsu indicates that large
numbers of people attached to an institution headed by the Queen of
Lagash, known as the ‘house of the woman’ (é-mí) and later the ‘house
of the goddess Bau’ (é-dba-ú), shared in disbursements of food and
drink from the institution’s stores during religious festivals throughout
the year (Beld 2002: 96–196; Jagersma 2007: 303–8). So, the fact of
large feasts, in which elites provided food and drink to their clients,
dependants and subordinates, is well established for Early Dynastic
Mesopotamia.
This said, looking beyond the specific symbolic content of the Royal
Tombs to its adaptive function is not as simple as Hayden implies. While
the burial events surrounding the Royal Tombs may have involved
feasting, the tombs themselves are also representations of feasts; in
other words it is their symbolic content that tells us these are feasts in
the first place. Furthermore, if the real story of funerary feasts is always
and only one of building advantageous alliances, one might wonder if
killing attendants was really an effective recruitment strategy! There is
no doubt that clientship and other forms of dependency were a vitally
important component of banqueting in Early Dynastic Mesopotamia
(see Beld 2002). However, as we shall see in a moment, it is impos-
sible to analyse such relations without reference to the institutional
framework (especially temple and palace households) and specific
economic relationships within which this dependency was realised.
Somewhat better suited to discussing the Royal Tombs as represen-
tations of feasts might be Michael Dietler’s category of the ‘diacritical
feast’ (Dietler 2001). For Dietler, feasts in hierarchical societies
sometimes cease to be vehicles for forming and reinforcing dyadic
relations and social solidarity. Rather, they can serve to emphasize
social distinctions and provide a venue for intra-class social compe-
tition. For Dietler, diacritical feasts are marked by the emphasis on
signs of exclusivity: exotic foods, highly marked and often expensive
accoutrements and settings; as well as elaborate preparations and
practices that imply specialised knowledge. While we have only limited
knowledge regarding the foods included in the Royal Tombs (see
Ellison et al. 1978), the gold, silver, lapis lazuli and ostrich eggshell
vessels (Woolley 1934: Pls 156–7, 160–5, 170–4), the gold, silver,
copper and lapis lazuli drinking straws (Weber and Zettler 1998: 139),
the ivory, turquoise and shell inlaid musical instruments and furniture
(Woolley 1934: Pls 91–126), the elaborate jewellery (Woolley 1934: Pls
127–50) and virtually everything else found in the Royal Tombs fit well
with Dietler’s definition of a diacritical feast.
While Early Dynastic banquets were certainly represented as
marked events of an elite nature, it is less clear that they were repre-
sented in terms of competitive gifting – what Marcel Mauss (1990: 6–7)
termed agonistic exchange. Denise Schmandt-Besserat (2001) points
to the Sumerian literary composition ‘The Debate between Winter

and Summer’ (ETCSL t.5.3.3) as an example of such elite competition.
A close examination of this text, however, reveals something rather
interesting about feasting and social competition in Sumer. There are
two incidents in the story that might be classified as feasts. Initially, as
part of competing to be acknowledged as the superior season, Winter
and Summer contend by attempting to outdo each other in their
presentation of food offerings to the god Enlil in his temple. However,
once Enlil declares that Winter is superior (because Winter’s water
makes Summer’s harvest possible), Summer hosts Winter in his home
and presents him with beer, wine, a banquet of succulent food, as well
as gifts of gold, silver and lapis lazuli; pouring out ‘brotherhood and
friendship like oil’ and ‘bringing sweet words to the quarrel’ (ETCSL
translation t.5.3.3: lines 310–15). In other words, the context in which
Winter and Summer are competing is not as hosts, but as fellow
worshippers (and hence servants) currying the favour of Enlil in his
own house (i.e. his temple). This competition is not left open; rather, a
hierarchical order is established through Enlil’s declaration of Winter’s
superiority. Summer hosts Winter in his home in order to end their
competition, re-establish their relationship and acknowledge Winter’s
superior rank.
While social competition is likely to have played a role in the actual
dynamics of Early Dynastic banqueting, it is not the prime emphasis
of the visual and literary representations of banqueting in which the
Royal Tombs participates. Rather than competitive exchange between
elites, representations of feasts in Early Dynastic Sumer would seem
to celebrate, and incorporate guests into, already established hierar-
chies marked by the upward movement of abundance. Most obvious is
the presentation of offerings to deities who were the principal guests
at religious festivals held in their honour. However, there is also an
emphasis on the provisioning of the central celebrants whose drinking
from cups or straws stands at the centre of most banquet scenes.
A striking aspect of Early Dynastic banqueting scenes is their stress
on service. In most cases, servants equal or outnumber banqueters in
these scenes, and, in the more elaborate scenes with multiple registers,
the portage and presentation of consumables takes up the bulk of the
visual field. In the case of the banquet scene on the ‘Standard of Ur’,
there is a literal upward movement of service as those in the lower two
registers seem to be processing animals and goods towards the large
central figure and his banqueting companions in the upper register,
as if the entire scene culminated in their acts of consumption (see
Figure 6.5).
The thematic importance of service and servants is also highlighted
by examples of satirical banqueting scenes in which animals take the
role of humans (see Amiet 1980: Pl. 99). On the soundbox (U. 10556)
of one of the lyres from the Royal Tombs (Woolley 1934: Pl. 105) we
have a three register scene, in which animals and mythical creatures
appear as musicians and servants in a banquet context and the upper
register is occupied by a ‘hero’ figure in the ‘master of the animals’ pose
between two recumbent ‘bull-men’ (Figure 6.6). Less subtle is an Early
Dynastic IIIA cylinder seal from Ur (Figure 6.7), predating the Royal
Cemetery, on which an ibex, three equids and several small mammals
provide music, drink and food for a seated lion (Legrain 1936: Pl.
20:384). As if to make sure that the message is not missed, to the right
of the satirical banquet scene is an animal combat scene in which a
lioness is using a dagger to attack an ungulate. Anthropomorphised in
this way, servants become the prey and sustenance of those whom they
serve. At the same time, the very fact that this relationship of service
is satirised highlights its centrality to the genre of banquet scenes. In
other words, servants in the act of service may be secondary and subor-
dinate within Sumerian banquet scenes, but they are primary to those
scenes in the sense of conveying their central message.
In one sense, in emphasizing service in Sumerian banquet scenes
I am simply pointing to the second of Dietler’s two characteristics of
diacritical feasting; stressing the reproduction of class divisions rather
than intra-class competition between elite banqueters. However, we
should remember that priests and rulers take this same role of servant
when providing food offerings or libations for deities. Indeed, to quote
Figure 6.7 Early Dynastic seal from Ur featuring a satirical banquet

scene with animals (source: LeGrain 1936: Plate 20 no. 384)
Igor Diakonoff, from a representational point of view, in the third

millennium bc ‘no person is without his or her lord (human or divine),
and thus … everyone is someone’s slave’ (Diakonoff 1987: 2). In other
words, in Early Dynastic Sumer the theme of service was subsumed
into a larger representational order that emphasized upwardly oriented
relations of obligation. In this sense, the representation of service in
banquet scenes did not merely aim to reproduce those immediate class
divisions. Instead, banquet scenes encompassed service in a world
defined by obligations to superiors, including obligations to the gods,
and thereby sought to rationalise everyday class divisions in terms that
were ultimately cosmological.
Susan Pollock (2007a; 2007b) has long stressed that the key link
between all of the Early Dynastic burials in the Ur cemetery, and
most particularly the Royal Tombs, was membership in the large
institutional households that structured the political economy of Early
Dynastic Sumer. Certainly, as Gianni Marchesi (2004: 175–81) has
shown, epigraphic evidence from the Royal Tombs provides strong
support for the assertion that at least some of the principal female
burials bore the title ‘nin’, meaning ‘lady’ and designating high status
(Marchesi 2004: 186–9). This suggests that we are on safe ground in
connecting the Royal Tombs to at least one of the major institutional

households of Ur. Which one seems less significant if members of the
same families were controlling both temple and palace households in
given cites or polities, as is suggested by Girsu texts. In other words,
while Early Dynastic Sumerian polities were not temple-states, it seems
likely that the palace and major temples did form a core that consti-
tuted what might be called a governmental apparatus and an inner
elite, to which the principal burials in the Royal Tombs seem to have
belonged.
For Pollock (2007a: 100; 2007b: 216–17) the Royal Tombs represent
the symbolic death of one these great households upon the death of
one its principal figures. She suggests that middle- and upper-ranking
members of an institutional household, along with the household’s
symbolically valuable possessions, would have joined the principal
deceased figure in death through participation in an actual or symbolic
funerary banquet in the tomb itself (Pollock 2007a: 102n. 8; 2007b:
214). As noted above, Pollock (2007b: 216–18) believes that this display
of extreme loyalty to a household was a dramatic public expression of
traditional, divinely determined and uninheritable leadership, against
a rising tide of inherited, dynastic power. Pollock (2007a: 100–6)
suggests that the willingness of these courtiers to go to their own death
was an outcome of their constitution as ‘docile bodies’ in the regular
workings of large institutional households. Most particularly, Pollock
(2007a: 101–5) argues that routine participation in formal banquets,
performing specific roles with specific expectations, would help
inculcate one’s relative rank and obligations as embodied practice (i.e.
habit-memory). Pollock suggests that this routinisation of one’s identi-
fication within the institutional household would make following one’s
lord or lady into the tomb less a matter of choice than of expectation.
Essentially, Pollock is arguing that the Royal Tombs were constituted
by both spectacle and routine; that the extreme logic of the funerary
spectacle was embedded in the routine logic of regular banqueting
practices. At the broadest level this explanation is compelling. If the
ritual killings of the Royal Tombs were understood as the logical
outcome of who had died (e.g. ruler, queen, etc.) and who was to
be entombed with the deceased (e.g. maidservant, musician, groom,
etc.), then this logic had to be embedded in the routine workings of
those institutional households that defined queens and maidservants
in the first place. However, two details of Pollock’s argument seem
problematic; namely her insistence on the elite nature of the attendants
buried in the tombs and her narrow focus on banquets as the primary
context for formation of the ‘docile bodies’ of the attendants.
One problem for Pollock’s assumption that the dead of the Royal
Cemetery ‘held ritual/cultic or managerial posts’ (Pollock 2007a: 99)
within an institutional household is the evidence for heavy, task specific,
skeletal modifications suggested by Molleson and Hogson (1993; 2000;
2003: 125–7) in their study of a limited number of surviving skeletons
from the Royal Tombs. Pollock (2007b: 214n. 4) rejects their conclu-
sions as exaggerated, pointing to the sampling problems I have noted
above. However, even if the osteological evidence for extreme manual
labour on the part of the buried attendant is problematic, it is inter-
esting that the roles in which these bodies have been cast within the
Royal Tombs are to a large extent the roles of service emphasised
in Early Dynastic banquet scenes. Musicians, grooms, soldiers and
maidservants are the supporting cast of both the Royal Tombs and the
banquet scenes. If we accept recent evidence (Baadsgaard et al. 2011)
that at least some of the attendants’ deaths were violently assisted by
blows to the head, then the question of who was included in the Royal
Tombs and why becomes rather problematic. I think that Pollock is
correct in asserting that burial in the Royal Cemetery was selective and
based on one’s position within an institutional household. However, the
nature of such households was quite complex, as was the composition
of their dependent membership. Understanding how some members
of these households were constituted as ‘killable subjects’ requires us
to go beyond banquets as public expressions of those households, not
least because the step from the table to the tomb is a rather large one.
Instead, we need to examine the economic and political relationships
that defined households and dependency in the Early Dynastic period.
Early Dynastic archives from ancient Girsu (modern Telloh) and

Šuruppak (modern Fara) give us some insight into such relations of
dependency. The archive at Girsu belonged to a large institutional
household headed by the Queen of Lagash. The archives at Šuruppak
seem to have been produced by two institutional households (‘the
palace’ and ‘the house of the city’) of uncertain relation (cf. Visicato
1995 and Cripps 2007), both of which may have been headed by the
same governor. At the core of these institutional archives was the
routine (usually monthly) disbursement of rations (barley, oil and
wool) on the part of the large, land-holding households to a variety
of individuals (e.g. Gelb 1965; Maekawa 1973/4). It is also clear that
not all residents of either Girsu or Šuruppak received rations from the
institutional households in question. Giuseppe Visicato (1995: 25–6)
has estimated the number of dependants of the ‘palace’ in Šurruppak
as 2,500–3,000 based on the number of its administrative sub-divisions
and their average membership. At Girsu, actually documented recip-
ients of rations range between 428 and 697 individuals per year,
although only an estimated 30 per cent of this archive has survived (see
Prentice 2010: 20, 65 and 79). Both Tell Fara (ancient Šurruppak) and
Telloh (ancient Girsu) are tells of ca. 100 hectares in area, and hence
their populations can be estimated at perhaps 10,000–25,000 residents.
So, the direct dependants of what appear to have been the primary
institutions at Šurruppak and Girsu would have represented a sizable
minority of the population.
At the broadest level, both the Girsu and Šuruppak archives
distinguish two ways in which people were attached to these large insti-
tutional households as dependants (see Cripps 2007; Prentice 2010).
In the first category are people who have been given land allotments
for personal use by an institutional household. In return they were
required to perform some kind of service to that institution, usually
manual labour (such as canal digging and harvesting) or military
service (Cripps 2007; Maekawa 1973/4). Tablets show that at both
Girsu and Šurruppak people in this category received barley rations
for only part of the year, apparently while fulfilling their labour service
requirements. It is not clear whether all such persons fulfilled these

labour obligations themselves, or whether some may have organised
and supplied workers from their own dependants who were then fed
by the institution’s barley rations during the period of labour service
(see Prentice 2010: 74–82).
In the second category are people who received barley rations every
month, all year round, from the stores of the institutional household
in question (see Prentice 2010; Visicato 1995). These involve people
categorised by a wide variety of professional titles covering labourers,
craftspersons and priestly and domestic staff. The distinction between
those receiving monthly rations all year round and those with land
grants is not marked by profession. Indeed, members of at least
14 professions (e.g. cooks, cupbearers, leather workers, boatmen,
shepherds, smiths, etc.) are found in both categories at Girsu (Prentice
2010: 72). However, there does seem to be a distinction in rank, since
approximately half of the individuals named on ration lists for those
given land grants reappear on the other rations lists as supervisors of
those receiving monthly rations (Prentice 2010: 72). At both Šurruppak
(Visicato1995) and Girsu (Prentice 2010) it is evident that the receipt
of rations is mediated by supervisors defined according to profession,
who are responsible for mustering and directing dependant labour
in exchange for special consideration such as land grants. At Girsu
it would also appear that such supervisors were provided with the
means of production (e.g. tools, raw material) and assigned quotas by
the Queen’s household with the possibility of accumulating debts or
surplus in relation to the institution (Magid 2001: 323–4). Interestingly,
Glenn Magid (2001) has pointed to evidence from Girsu suggesting
that some craftworkers on continuous monthly rations may have
lacked their own immediate households and lived with their super-
visors in what might be termed dependent workshops.
In contrast to the situation with craft-workers, whose relations
to the palace or temple – not to mention their basic living condi-
tions – may have been mediated by their immediate supervisors, at
both Šurruppak and Girsu there are categories of people receiving
continuous monthly rations described as ‘belonging to’ or ‘registered at’

the palace (šà-dub-é-gal). Such people are usually listed by profession
and these professions are what one would expect of a large domestic
staff – cooks, cupbearers, messengers, ‘inner room’, ‘store room’ and
‘hot water’ attendants, scribes, hairdressers, cleaners, etc. They also
replicate some of the occupations depicted in banquet scenes and
represented among the attendant burials in the Royal Tombs. While
they too had supervisors, one suspects that some of these people were
direct dependants of the palace, in the sense of belonging to no other
household, or possessing no other means of subsistence.
Focusing for a moment on those who seem most dependent on
institutional households, it is evident that at least some such labourers
were unfree. There is textual evidence from Girsu suggesting that
some low-status workers were sold to the Queen of Lagash by their
overseers (Prentice 2010: 144, 147–8). One might point out that the
two sales documents in question involve workers called igi-nu-du8
(‘the blind’), a category of very low-status dependants who usually
worked under the supervision of gardeners, and indeed were here
sold by gardeners in both texts. A common explanation for this term
is that these were prisoners of war who were intentionally blinded
and enslaved (Gelb 1973). This may well be the case, however; such
transactions were not unique, with other sales documents indicating
that gala (lamentation singers) as well as a ‘foundling’ were also sold
to the Queen of Lagash, although here the sellers were their parents/
guardians, hence this may have some relation to debt bondage
(Prentice 2010: 147–8). Amongst low status textile workers, classified
on ration lists as gemé-dumu (‘women and children’), one finds some
referred to as ‘purchased’ (Prentice 2010: 55–9). This is not to imply
that all dependant workers at Girsu were interchangeable chattels of
the ‘house of the woman/goddess Ba’u’. At Girsu there is also evidence
for differential ration rates that seem to relate to worker experience
and/or skill (Prentice 2010: 91–5), while amongst female weavers
some were promoted over time from weavers to overseers (Prentice
2010: 54–5).
It seems that, within these institutional households, resource

allocation and labour obligations both incorporated class relations (e.g.
between full-time dependants and supervisors, who held land grants
and perhaps also partial control over the means of production and the
products of labour) and masked them within an overall ideology of
obligatory service by all. To quote Susan Pollock:
Ideologically, the notion that people of all social stations – and even
the gods – received rations may have contributed to a sense that
everyone was ‘in it together,’ that everyone participated in laboring
for the common good and received remuneration according to their
contribution. (Pollock 2003: 32)
Put simply, the textual evidence from Early Dynastic Sumer suggests
that a significant proportion of the population was connected in
some way to the large institutional households through networks of
dependency. However, the degree, regularity and implications of this
dependency varied considerably. Some people came into irregular
contact with temple or palace households, perhaps through occasional
public works programmes (like canal digging) or indirectly through
their relations with local elites. In essence, they lived independently
but in the wake of the economic power of temples and palaces. Others
were granted the means of production by these institutions, in the form
of land, raw materials and perhaps even the allocation of dependent
labour. Such persons appear to have been free to accumulate their
own wealth once they had met certain obligations to the granting
institution. Still others had no independent household of their own,
nor any independent means of subsistence; indeed some were officially
chattels, in the sense of being purchased by the institution. For those
on the domestic staffs of palaces and temples, this may have meant
direct dependency on the principal elites of these institutions, for
whom they performed a variety of domestic tasks. Such tasks likely
included the feeding, attending to and entertaining of elites and their
guests, much as was represented in banquet scenes and in the Royal
Tombs of Ur.
Banquet scenes were celebratory scenes, just as banquets themselves

were celebratory events. They revelled in the abundance produced
by institutional households, realised along chains of obligation and
command linking dependant workers to human rulers and ultimately
to the gods themselves. This was a hegemonic representation of ‘the
good life’ in Early Dynastic Sumer, yet it was also one that, in the
extreme case of the Royal Tombs, made imaginable the ritual killing of
hundreds of people.
The philosopher Giorgio Agamben (1998; 2005) has argued that
the arbitrary foundation of the modern state’s power is revealed when
exercised in supposed exceptions to the rule of law, such as in states
of emergency or in cases that create ambiguous legal identities (e.g.
concentration camps, detention of illegal immigrants, the ‘enemy
combatants’ of Guantanamo Bay). These ‘states of exception’ highlight
the ability of the state to exercise coercive power outside of its own legal
frameworks, and thereby they undermine the normative assumption
that state rule derives in the first instance from law rather than
force. Agamben is particularly interested in the manifestation of state
power as biopower in Foucault’s sense of the shaping, containing and
excluding of physical existence, what Agamben describes as ‘the politi-
cization of bare life’ (Agamben 1998: 6).
Like modern ‘states of exception’, the extremity of the Royal Tombs
of Ur does reveal a particular species of biopower at work in the consti-
tution of institutional households in the third millennium bc However,
here ‘bare life’ is revealed by inclusion in, rather than exclusion from,
the state. In the Royal Tombs the very existence of the attendants (their
identity, their kinship, their past and their future) was swallowed up by
the ritual drama of the funerary banquet. Within the Royal Tombs, the
attendants’ bodies have no independent referent, they are stripped to
the bare facts of their biological life, and once this has been done that
life proves easy to extinguish. This was a spectacle in the purest sense:
unusual, extreme, overwhelming, laden with messages of order, power,
aspiration and fear. Yet, I would argue that this spectacle was imagi-
nable because it had already been constituted in the routine workings
of institutional households; regularised through the organisational

structure of institutional accounting, the distribution of rations and the
management of labour. The existence of people without independent
households and without independent means of subsistence, people
whose labour could be freely expropriated as a condition of their
remaining alive, was both materially and symbolically central to the
reproduction of institutional households. ‘Killable subjects’ were not
formed only on the plane of ideology and spectacle. The spectacle of
the Royal Tombs was made possible by the material interdependence
of institutions, attached labour, production and rations, all articulated
within a hegemonic order of universal obligatory service.
We do not know for certain if the attendants buried in the Royal
Tombs originated from amongst the household’s most dependent
members. We do not know if they were loyal and honoured volunteers,
prisoners of war, destitute or merely surplus to purpose. What we do
know is that, within the funerary spectacle of the Royal Tombs of Ur,
such distinctions ceased to matter. Regardless of their origins, those
going to their deaths were subsumed into roles of service, replicating
the hegemonic order of institutional households in a ritual drama that
fused authority, violence and transcendence into sovereignty. Played
out in these funerary spectacles was an order in which each gave
service according to his or her station. For some this meant offering
libations to the gods, for others it meant giving time or barley, and for
some it meant a narrow-bladed axe in the back of the head.
7
Conclusion
The hazardous necessity of comparison
I have often found that Angels have the vanity to speak of themselves
as the only wise; this they do with a confident insolence sprouting
from systematic reasoning.
William Blake, The Marriage of Heaven and Hell
As a child I read many more comic books than I care to admit. Yet,
despite my familiarity with the genre, it came as a revelation that the
seemingly solid colours on the page were actually an effect created
by thousands of tiny dots. What I had taken to be a simple case of
colouring between the lines turned out to be something rather more
mysterious, if only for its ability to hide in plain sight. From that point
on, the seemingly solid images of comic books shifted in and out
of focus, as I alternately forgot and remembered the lesson of their
composition as clouds of coloured dots.
This experience of images that became solid or dissolved depending
upon how well I remembered the realities of their composition provides
a good analogy for the focus and purpose of this book. As noted in
Chapter 1, what we call the state is an effect, rather than an entity; its
solidity and the smoothness of its surfaces depends directly on our
forgetting the practices, strategies and technologies through which
this state-effect is constituted. A large part of this book has been about
keeping this state-effect in view, crossing our eyes, as it were, in order
to see clouds of dots instead of solid shapes.
This focus is narrower than and rather different from what is usually
discussed in literature on the archaeology of the state. Normally,
debates focus on the existence and variability of a bauplan of the

state as an institutional structure shared by distinct entities, whether
this be in terms of specialised decision making, resource allocation,
dispute resolution, the monopolisation of force, or all of the above. To
stretch my comic book analogy, these discussions are primarily about
comparing the pictures on the page, while taking for granted their
solidly coloured surfaces. In contrast, I have focused on the gestalt
experience of seeing pictures on a page composed of dots. Although
I have, in this way, raised critical question regarding the reality of the
state as a universal entity, I have also explored what I believe to be a
real phenomenon involving the intersection of authority, force and
transcendence.
With some modification (see Chapter 1), this book can, therefore,
be seen as an example of what Adam Smith (2011) has termed
‘archaeologies of sovereignty’. Here I understand sovereignty to be the
effect of the intersection of authority, force and transcendence, and
hegemony to be the condition that allows this intersection to occur.
This perspective differs from archaeologies of the state, in that it seeks
always to disaggregate and disassemble sovereignty in order to keep
visible how it is made. At the same time, it differs from both particu-
larist and deconstructivist positions in that it recognises a general
phenomenon that can be fruitfully discussed in relation to a number
of distinct cultural and historical contexts.
This last point raises rather directly the problem of comparison. To
be clear, the issue is not comparison per se, in that comparison is always
necessary if we are to have knowledge rather than endlessly discrete
sense impressions. What really matters is how, what and why we are
comparing; in other words, what do we hold to be comparable, and
what do we hope to gain from making such comparisons?
The first question is perhaps the most straightforward, namely
the question of comparative method. In this book I have engaged in
what Smith and Peregrine would term ‘intensive comparison’ (Smith
and Peregrine 2012), in the sense that I have focused on contextual
detail across a limited number of cases. Smith and Peregrine contrast
Conclusion 159
this approach with what they term ‘systematic comparison’, meaning

the statistical analysis of a limited number of attributes across a large
number of cases (Smith and Peregrine 2012). This distinction is a
long-standing one in anthropology where, for example, it appears in
Claude Lévi-Strauss’s discussion of the comparative method as it relates
to mechanical and statistical models of social structure (Lévi-Strauss
1967: 279–81). Lévi-Strauss (1967: 281) argued that, because cultural
variables are never identical across ethnographic contexts, large-scale
comparisons are not really very good comparisons at all. Hence, an
anthropologist must choose ‘either to study many cases in a super-
ficial and in the end ineffective way; or to limit oneself to a thorough
study of a small number of cases, thus proving that in the last analysis
one well done experiment is sufficient to make a demonstration’
(Lévi-Strauss 1967: 281). Yet, as Caspar Bruun Jensen (2011: 5) notes,
Lévi-Strauss’s search for the ‘golden event’ that is both a single case and
a general demonstration presents an intractable problem: how does
one identify a ‘golden event’ without already knowing what it demon-
strates? Certainly this is no less of a problem than that of comparing
multiple cases of not-quite-the-same-thing.
In part, at least, the problem lies in what one hopes to achieve by
means of comparative study. In this book, I have been much concerned
with generalising arguments. For example, I have suggested that
transcendence is paired with violence in the constitution of sovereignty
through the need to overcome constraints on the legitimate use of force
that are general and integral to human collectives. I have also suggested
that hegemony demonstrates a general pattern of appropriating, trans-
forming and reinscribing the cultural resources of ‘common sense’ as
the basis for consent in given political regimes. However, unlike Claude
Lévi-Strauss and Michael E. Smith, I have not offered these generalisa-
tions in order to reveal universal principles in themselves. For example,
I am not concerned to extract the abstract principles of political power
that structure diverse historical contexts, as if universal significance
was demonstrated by the removal of historical content. Rather, I have
developed general arguments for the express purpose of re-immersing
them in specific historical contexts. In this sense, my generalisations

are less models to be tested than tools to facilitate comparative discus-
sions without sacrificing cultural and historical detail. It is for this
reason that my use of ‘intensive comparison’ is particularly important
to my goals, as my case studies are meant to develop key concepts
through continual re-immersion into the historical and cultural details
of specific contexts. This in turn, I hope, will stimulate on-going discus-
sions across specific contexts, but also by means of those contexts, in
order to develop, modify and challenge the general concepts I have
presented in this book.
So, my goal in working comparatively is to generate dialogues across
specific contexts on the common processes of sovereignty for the
purpose of both understanding the past and acting in the present. But
can one actually do this? Does the presumption of common themes
and common analytical concepts not subordinate these different
pasts by universalizing a very particular present (twenty-first century,
Euro-American, academic)? In other words, is my analytical framework
not itself a form of hegemony? At a time when teleological universal
histories of political organisation are both popular and abundant (e.g.
Diamond 1997; Fukuyama 2011), it seems important to ask if compar-
ative study can indeed take any other form. Here we would do well to
consider the unlikely figure of anthropologist Louis Dumont.
Dumont famously (or infamously) framed the study of India
in terms of an ideology in which hierarchy (in the form of ritual
purity), as a shared value separable from power, served to encompass
opposition and thereby constitute a coherent social whole (see
Dumont 1980). The specifics of Dumont’s interpretation of India
have been subject to extensive critique, many of which have been
recently reprinted (Khare 2006). Our concern is not so much with
these details as with what Dumont saw as both the purpose and the
implications of his work. Dumont’s anthropology was unabashedly
comparative. However, for Dumont the purpose of comparison
was not to track similarities but to highlight differences, and to do
this not as an act of cross-cultural translation, but rather as a form
Conclusion 161
of ideological auto-critique (Dumont 1975). In other words, the

outcome of noting the difference between an Indian emphasis on
collectivism and hierarchy and a Western emphasis on individualism
and egalitarianism was as much to highlight the ideological nature
of Western frames of reference as it was to understand the ‘exotic’
nature of Indian thought.
As some scholars have noted (e.g. Kapferer 2011), it is in this
last regard that Dumont’s work finds parallels in the more recent
‘ontological turn’ within anthropology (e.g. Descola 2006; Viveiros
De Castro 2004). For example, Eduardo Viveiros De Castro (2004)
has argued that the classic anthropological move of rendering appar-
ently irrational indigenous understandings of the nature of existence
(ontology) rational by explaining them as coherent alternate views
of a single shared world (‘one world, many worldviews’) implicitly
subordinates these ontologies to our own. As he writes, ‘[de]scribing
this world as if it were an illusory version of our own, unifying the two
via a reduction of one to the conventions of the other, is to imagine
an overly simple form of relations between them’ (Viveiros De Castro
2004: 14). Instead, in a manner similar to Dumont, Viveiros De
Castro argues that ‘[a] good translation is one that allows the alien
concepts to deform and subvert the translator’s conceptual toolbox
…’ (Viveiros De Castro 2004: 5). Hence, the purpose of comparison
is to encounter and reflect on alterity; indeed, according to Viveiros
De Castro (2004: 11), ‘[s]ince it is only worth comparing the incom-
mensurable, comparing the commensurable is a task for accountants,
not anthropologists’.
From such a perspective, my attempts to reflect on premodern
political formations by means of concepts derived from the
critical analysis of the modern state may have the appearance of
accountancy masquerading as anthropology. Dumont in particular
might have seen in both my emphasis on the composite nature of
sovereignty and my focus on conflict, force and power, the ideology
of possessive individualism that he argued has dominated Western
social thought since the Enlightenment (Dumont 1986). My attempts
to ‘look behind the mask’ of political power are cast in terms that
are unlikely to have been meaningful to those living in the historical
contexts I purport to explain. Indeed, my generalising arguments
(about sovereignty, hegemony, violence, etc.) risk reducing the
specific conventions of these contexts to further examples of my own
conventions, rather than allowing them to ‘deform and subvert’ my
‘conceptual toolbox’.
In response, it is important to stress the place of both violence
and hegemony in my arguments. The cases analysed in this book are
defined by violence. In other words, violence is not a trait ascribed
to some other phenomenon such as the state; rather, this book is
about sovereignty as a particular species of institutional violence. My
observations apply only to the degree that this species of institutional
violence occurs in any given context.
The centrality of violence to sovereignty so defined means that
conflict and resistance are already embedded in the cases that I
have chosen to study. Hence, these contexts do not confront us as
a coherent ‘Other’; they are already divided by unequal access to
the means of violence. Here the meta-cultural status of hegemony
becomes important. Hegemony is a totalising project; it aims to
exclude or subordinate alternate views. However, hegemony is not
itself a totality. As argued in Chapter Two, hegemonic projects select
cultural resources from amongst the many possibilities generated by
everyday life and configure these resources in the interests of sover-
eignty. Hence, alternate or excluded configurations can always take
shape within a single cultural formation. In other words, the critical
analysis of sovereignty is not the exclusive domain of an external,
objectivist, unmasking of power. Internal critiques of sovereignty are
always possible, critiques that may well draw on the same cultural
world and the same ontological assumptions that form the basis of a
given hegemonic project. We saw this in Chapter Three in the case of
Imerina, where conflict over Radama I’s haircut and collective burial
rituals saw royals and non-royals configuring the same cultural values
in different ways and towards different ends.
Conclusion 163
Let us consider once again the issue of slavery in Classical Athens.

In Book I of The Politics Aristotle makes the ontological argument that
some human beings are slaves by nature. I suspect that most readers
of this book would find Aristotle’s views unacceptable; they are also
incommensurate with the concept of equality under the law that
underpins, amongst other things, the Universal Declaration of Human
Rights and the legal systems of most modern liberal democracies. In
this context, explaining Aristotle’s arguments historically or culturally
would amount to an act of anthropological ‘translation’: a means of
understanding Aristotle without conceding any truth to his beliefs
on the nature of slaves. One might find Aristotle interesting, but one
would never doubt that he was wrong.
Aristotle’s unacceptable understanding of natural slaves is precisely
the kind of alterity that Viveiros de Castro admonishes anthropologists
to encounter rather than translate; except for one thing. The way in
which Aristotle frames his argument makes it clear that his views were
not shared by all of his ancient Greek contemporaries, even if these
opposing positions are not well attested in the surviving literature (see
Cambiano 1987). Hence, one cannot encounter the alterity of the (as
in singular) ancient Greek perspective on the nature of slaves without
reducing and essentialising the diversity of views that existed at the
time of Aristotle (cf. Lloyd 2010: 208–9). There is no ‘ancient Greek
model’ to be given priority over our comparative themes and analytical
concepts. Indeed, to some extent, we are simply entering into the
debate. But this is not really the crux of the matter.
In Plato’s Republic, Socrates suggests that a rich slave-owner did not
live in fear of his slaves because, if they rose up, the whole city would
come to the aid of each private citizen. However,
… if this one man with his fifty or more slaves were lifted out of the
city by some god, together with his wife and children, and deposited
with his slaves and other property in a deserted place where no free
man could come to his assistance, how frightened would he be on his
own behalf and that of his wife and children lest they be killed by the
slaves? – Very frightened indeed! (Plato, Republic, IX.578e)
While he disagrees with Plato on many things, here Aristotle sees the
same danger in so far as he deems it necessary to consider measures
to prevent slave revolt in discussing the planning of an ideal agrarian
landscape (Aristotle, The Politics, VII.x.1330a).
Whether or not one viewed slavery as a natural state, slave resistance
was clearly thinkable: something for rich men to fear and against which
freemen were prepared to mobilise. Hence, the point is not simply
that there was a diversity of opinion on the nature of slaves in ancient
Athens, but rather that slavery itself, as a relationship of violence,
represented a fault-line within Athenian society, one that ran through
indigenous perspectives on politics, citizenship, labour or wealth.
Comparing these sorts of fault-lines in different societies is one way
in which the comparison of similarities can play a key critical role that
goes beyond the imposition of the present onto the past. On issues
such as citizenship and slavery in ancient Athens, where the terrain
was divided by relations of violence and where the means of self-
representation were unequally distributed, comparison can provide us
with terms of reference (e.g. sovereignty, hegemony, etc.) distinct from
those generated by the powerful in the past. Comparing fault-lines in
different contexts also reminds us that there are always people on either
side of these social and political divides, highlighting the need to seek
out traces of those who have been excluded or forgotten. Returning to
our comic book analogy, comparing fault-lines in different contexts
helps to make dots visible where one might otherwise have seen only
solid colours. In this way, comparison is essential to the critical study
of sovereignty.
The benefits of comparison do not exist only for the past; indeed
quite the opposite is true. Hence, the comparison of fault-lines is as
relevant to modern liberal democracies as it is to ancient Athens or
the kingdom of Imerina. For example, equality under the law may
be a founding principal to most liberal democratic states, but this
principal is also configured in particular ways involving relations of
violence that we often forget or take for granted. Equality under the
law is linked to citizenship and territory, which define where and to
Conclusion 165
whom the law applies. Citizenship and territory are in turn linked
to collective identity and national pride, but they are also linked to
mandatory birth registration, residency permits, immigration criteria
based upon wealth and education, the presentation of passports, the
policing of borders, and the arrest and forcible deportation of ‘illegal’
non-citizens. Equality under the law is also linked to policing and
law enforcement, which in turn entails (amongst other things) the
legitimate application of force against citizens (and others). As we all
know, equality under the law does not result in the actual equality of
citizens. Indeed, in most liberal democracies with a capitalist economy,
the law acts to favour some forms of equality (e.g. equality in the right
to own property and in the protection of property) against other forms
of equality (e.g. equality in the distribution of property), with police
and judicial powers deployed to ensure that this is the case. At the same
time, citizenship and territory define when the application of force is
a police matter and when it is a military matter; in other words when
force is applied by the state within the law and when it is projected
outside of the law. Furthermore, in states of emergency the law itself
can be suspended for the purposes of preserving or rescuing the law
and the institutions of democracy, when these are viewed to be under
threat from, for example, terrorism or insurrection.
These many relationships of violence may well be necessities of
living in large-scale communities; the price of liberty, as some would
have it. Yet, in comparative perspective these unfortunate necessities
begin to look somewhat different. For example, the maintenance
of order, and the continuation of a particular mode of life, were as
important to Egyptian Pharaohs and Inca Emperors as they are to the
politicians and citizens of modern liberal democracies. Indeed, as I
hope this book has shown, there are many grounds for comparison in
this regard. The difference, many will say, lies in the content of what is
being defended and the manner and limits placed on this defence; but
this is exactly my point.
The relationships of institutional violence that we perpetuate are
not simply unfortunate necessities of complex polities; they represent a
choice regarding what forms of order and what modes of life we deem
worthy of preservation and at what cost. Comparative study does not
make this choice for us, but it does make clear that a choice is being
made; it requires us to ask: ‘what exactly are we defending and why?’
It requires us to recognise the hegemonic projects within which we
are embedded and to ask whose interests are being perpetuated and to
what ends? We may be happy to find that the interests are our own and
that our consent is freely given; but even this should arise from critical
reflection and not from virtual relations that we take for granted.
To my mind, open and informed discussion of such issues holds
a greater potential to ‘subvert and deform’ our ‘conceptual toolbox’
than does an exclusive focus on the alterity of the past. In this sense, a
comparative archaeology of sovereignty has a role to play in stimulating
reflection, debate and action on how we live together in the world. We
could, and many do, pursue this role like one of William Blake’s angels,
speaking of ourselves as ‘the only wise’ and producing teleological
histories of the world as we already know it to be. The more difficult
and rewarding task is to embrace the devil in the contextual detail,
thinking comparatively while allowing ourselves to be challenged by
the diverse possibilities of human existence. This is the task that I set
myself in writing this book, one that could only ever be imperfectly
realised given the limits of time and my own personal abilities. Yet
such imperfection has its place if it opens the door to future, better
informed, discussions regarding political power and human diversity
from a long-term perspective. After a decade of war, terror and crisis,
such discussions are certainly needed. What form they take and what
action they engender will depend, at least in part, on decisions made
by people like you, the readers of this book.
Bibliography
Adam, S. (2007) Response á Edward Harris. In Symposion 2005: Vorträge zur

griechischen und hellenistischen Rechtsgeschichte. Austrian Academy of
Sciences: Vienna; 177–82.
Adams, A. and J. Brady (2005) Ethnographic notes on Maya Q’eqchi’ cave
rites: implications for archaeological interpretation. In Brady, J. and K.
Prufer (eds), In the maw of the earth monster: Mesoamerican ritual cave
use. University of Texas Press: Austin, TX; 302–27.
Abrams, P. (1988) Notes on the difficulty of studying the state (1977). Journal
of Historical Sociology 1(1): 58–89.
Agamben, G. (1998) Homo Sacer: Sovereign power and bare life. Trans. D.
Heller-Roazan. Stanford University Press: Palo Alto, CA.
—(2005) State of exception. Trans. K. Attell. University of Chicago Press:
Chicago.
Alberti Manzanares, P. (1986) Una institución exclusivamente femenina en la
época incaica: las acilacuna. Revista Española de Antropología Americana
16: 153–90.
Algaze, G. (1983–4) Private houses and graves at Ingharra, a reconsideration.
Mesopotamia 18–19: 135–91.
Alonso, A. (1994) The politics of space, time and substance: state formation,
nationalism, and ethncity. ARA 23: 379–405.
Amiet, P. (1980) La glyptique mésopotamienne archaïque. 2nd edn. Editions
du centre national de la recherche scientifique: Paris.
Anderson, G. (2009) The personality of the Greek State. Journal of Hellenic
Studies 129: 1–22.
Aristotle (1962) The Politics. Trans. J. Sinclair. Penguin Classics.
Harmondsworth.
Baadsgaard A., J. Monge, S. Cox and R. Zettler (2011) Human sacrifice and
intentional corpse preservation in the Royal Cemetery of Ur. Antiquity 85:
27–42.
Baadsgaard, A., J. Monge and R. Zettler (2012) Bludgeoned, burned, and
beautified: re-evaluating mortuary practices in the Royal Cemetery of
Ur. In A. Porter and G. Schwartz (eds), Sacred killing: The archaeology of
sacrifice in the ancient Near East. Eisenbrauns: Winona Lake, IN; 125–58.
168 Bibliography
Baines, J. and N. Yoffee (1998) Order, legitimacy and wealth in Ancient Egypt
and Mesopotamia. In Feinman, G. and J. Marcus (eds), Archaic states.
SAR Press: Santa Fe, NM: 199–260.
—(2000) Order, legitimacy, and wealth: setting the terms. In Richards, J.
and M. Van Buren (eds), Order, legitimacy, and wealth in ancient states.
Cambridge University Press: Cambridge; 13–17.
Bartelson, J. (2001) The critique of the State. Cambridge University Press:
Cambridge.
Barrett, J. and I. Ko (2009) A phenomenology of landscape: a crisis in British
landscape archaeology? Journal of Social Archaeology 9(3): 275–94.
Bauer, B. (1996) Legitimization of the state in Inca myth and ritual. AA 98(2):
327–37.
Bauer, B. and R. A. Covey (2002) Processes of state formation in the Inca
heartland (Cuzco, Peru). AA 104(3): 846–64.
Beld, S. (2002) The Queen of Lagash: Ritual economy in a Sumerian state.
Unpublished Ph.D. dissertation. University of Michigan: Ann Arbor,
MI.
Belrose-Huyghues, V. (1983) Structure et symbolique de l’espace
royal en Imerina. In F. Raison-Jourdes (ed.), Les Souverains de
Madagascar: L’histoire royale et ses resurgences contemporaines. Éditions
Karthala: Paris; 125–52.
Berent, M. (2000) Anthropology and the Classics: war, violence, and the
stateless polis. CQ 50(1): 257–89.
—(2006) The stateless polis: a reply to critics. SEH 5(1): 141–63.
Berg, G. (1981) Riziculture and the founding of monarchy in Imerina. The
Journal of African History 22: 289–308.
—(1985) The sacred musket: tactics, technology and power in eighteenth
century Madagascar. CSSH 27: 261–79.
Bintliff, J. (2006) Solon’s reforms: an archaeological perspective. In Blok, J.
and A. Lardinois (eds), Solon of Athens: New historical and philological
approaches. Mnemosyne, Bibliotheca Classica Batava Supplementum.
Brill: Leiden; 321–33.
Blanton, R. (1998) Beyond centralization: steps toward a theory of egalitarian
behavior in Archaic States. In Feinman, G., and J. Marcus (eds), Archaic
states. SAR Press: Santa Fe, NM; 135–72.
Blanton, R. and L. Fargher (2008) Collective action in the formation of
pre-modern states. Springer: New York.
Bibliography 169
Bloch, M. (1971) Placing the dead: Tombs, ancestral villages, and kinship
organization in Madagascar. Waveland Press Inc.: London.
—(1986) From blessing to violence: History and ideology in the circumcision
ritual of the Merina of Madagascar. Cambridge University Press:
Cambridge.
—(1987) The ritual of the royal bath in Madagascar: the dissolution of death,
birth and fertility into authority. In D. Cannadine (ed.), Rituals of royalty:
Power and ceremonial in traditional socieities. Cambridge University Press:
Cambridge; 271–97.
Blok, J. and A. Lardinois (eds) (2006) Solon of Athens: New historical
and philological approaches. Mnemosyne, Bibliotheca Classica Batava
Supplementum. Brill: Leiden.
Boese, J. (1971) Altmesopotamische Weihplatten: Eine sumerische
Denkmalsgattung des 3. Jahrtausends v. Chr. De Gruyter: Berlin.
Bourdieu, P. (1999) Rethinking the State: genesis and structure of the
bureaucratic field. In G. Steinmetz (ed.), State/Culture: State formation
after the cultural turn. Cornell University Press: Ithaca, NY; 53–75.
Brady, J. (1997) Settlement configuration and cosmology: the role of caves at
Dos Pilas. AA 99(3): 602–18.
Brady, J. and W. Ashmore (1999) Mountains, caves, water: ideational
landscapes of the Ancient Maya. In W. Ashmore and A. B. Knapp
(eds), Archaeologies of landscape: Contemporary perspectives. Blackwell
Publishing: Oxford; 124–45.
Brady, J. and K. Prufer (eds) (2005) In the maw of the earth monster:
Mesoamerican ritual cave use. University of Texas Press: Austin, TX.
Bray, T. (2003) Inka pottery as culinary equipment: food, feasting, and gender
in imperial state design. LAA 14(1): 3–28.
—(2009) The role of chicha in Inca state expansion: a distributional study of
Inca aríbalos. In Jennings, J. and B. Bowser (eds), Power, drink and society
in the Andes. University Press of Florida: Gainesville, FL; 108–32.
Brock, R. (1994) The labour of women in Classical Athens. CQ 44(2):
336–46.
Brück, J. (2005) Experiencing the past? The development of a
phenomenological archaeology in British prehistory. AD 12(1): 45–72.
Brumfiel, E. (1992) Distinguished lecture in archaeology: Breaking and
entering the ecosystem – gender, class and faction steal the show. AA
94(3): 551–67.
170 Bibliography
Brumfiel, E. and T. Earle (eds) (1987) Specialization, exchange and complex

societies. Cambridge University Press: Cambridge.
Bruun Jensen, C. (2011) Comparative relativism: symposium on an
impossibility. Common Knowledge 17(1): 1–12.
Burford, A. (1993) Land and labor in the Greek world. The Johns Hopkins
University Press: Baltimore.
Callet, F. (ed.) (1908) Tantara ny Adriana eto Madagascar: Documents historiques
d’après les manuscrits malgaches. 2 vols. Académie Malgache: Tananarive.
Cambiano, G. (1987) Aristotle and the anonymous opponents of slavery.
Slavery and Abolition 8(1): 22–41.
Campbell, G. (2005) An economic history of imperial Madagascar, 1750–1895.
Cambridge University Press: Cambridge.
Campbell, R. (2008) Comments on Flad (2008). CA 49(3): 420–1.
Cartledge, P., E. Cohen and L. Foxhall (eds) (2002) Money, labour and
land: Approaches to the economies of ancient Greece. Routledge: London.
Chang, K.-C. (1983) Art, myth and ritual: The path to political authority in
ancient China. Cambridge University Press: Cambridge, MA.
Chapman, R. (2003) Archaeologies of Complexity. Routledge: London.
Child, M. (2007) Ritual purification and the ancient Maya sweatbath at
Palenque. In D. Marken (ed.), Palenque: Recent investigations at the Classic
Maya center. Altamira Press: Lanham, MD; 233–62.
Clastres, P. (1977) Society against the State. Trans. R. Hurley. Urizen Books:
New York.
Coggins, C. C. and O. C. Shane (eds) (1984) Cenote of sacrifice: Maya treasures
from the Sacred Well at Chichén Itzá. University of Texas Press: Austin TX.
Cohen, A. (2005) Death rituals, ideology, and the development of early
Mesopotamian kingship. Brill: Leiden.
Cohen, D. (1989) Seclusion, separation, and the status of women in Classical
Athens. Greece and Rome 36(1): 3–15.
Cohen, E. (2000) The Athenian nation. Princeton University Press: Princeton.
—(2002) An unprofitable masculinity. In P. Cartledge, E. Cohen and L.
Foxhall (eds) (2002) Money, labour and land: Approaches to the economies
of ancient Greece. Routledge: London; 100–12.
Comaroff, J. and J. Comaroff (1991) Of revelation and revolution. Vol. 1.
University of Chicago Press: Chicago.
Connerton, P. (1989) How societies remember. Cambridge University Press:
Cambridge.
Bibliography 171
Cooper, J. (1986) Presargonic inscriptions. Sumerian and Akkadian royal

inscriptions 1. American Oriental Society: New Haven.
Costin, C. (1998) Housewives, chosen women, skilled men: cloth production
and social identity in the late Prehispanic Andes. In C. Costin and R.
Wright (eds), Craft and social identity. APAAA 8. Wiley-Blackwell:
Washington, DC; 123–41.
Coucouzeli, A. (2007) From megaron to oikos at Zagora. In R.
Westgate, N. Fischer and J. Whitley (eds), Building communities:
House, settlement and society in the Aegean and beyond. Proceedings
of a conference held at Cardiff University, 17–12 April 2001. British School
at Athens: London; 169–81.
Crehan, K. (2002) Gramsci, culture and anthropology. University of California
Press: Beskley, CA.
Cripps, E. (2007) Land tenure and social stratification in ancient Mesopotamia:
Third millennium Sumer before the Ur III dynasty. BAR International series
1676. Archaeopress: Oxford.
Crossland, Z. (2001) Time and the ancestors: landscape survey in the
Andrantsay region of Madagascar. Antiquity 75: 825–36.
Dahl, R. (1989) Democracy and its critics. Yale University Press: New Haven.
D’Altroy, T. (1992) Provincial power in the Inka Empire. Smithsonian:
Washington, DC.
Davidson, J. (2011) Bodymaps: sexing space and zoning gender in Ancient
Athens. Gender and History 23(3): 597–614.
Davies, J. (1981) Wealth and the power of wealth in Classical Athens. Arno
Press: New York.
Davis-Salazar, K. (2003) Late Classic Maya water management and
community organization at Copán, Honduras. LAA 14(3): 275–99.
—(2006) Late Classic Maya drainage and flood control at Copán, Honduras.
Ancient Mesoamerica 17: 125–38.
De Mecquenem, R. (1943) Fouilles de Suse, 1933–1939. Mémoires de la Mission
Archéologique en Iran (Mission de Susiane) 29. Paul Geuthner: Paris.
Demarrais E., L. J. Castillo and T. Earle (1996) Ideology, materialization, and
power strategies. CA 37(1): 15–31.
Dewar, R. (2007) Processual assessment of population changes in Western
Avaradrano. In H. Wright (ed.), Early state formation in central
Madagascar: An archaeological survey of western Avaradrano. Museum of
Anthropology, University of Michigan: Ann Arbor, MI; 101–3.
172 Bibliography
Diakonoff, I. (1987) Slave-labour vs. non-slave labour: the problem of

definition. In M. Powell (ed.), Labor in the ancient Near East. American
Oriental Society Series 68. American Oriental Society: New Haven, CT; 1–3.
Diamond, J. (1997) Guns, germs and steel: A short history of everybody for the
last 13,000 years. Vintage: London.
Dickson, D. B. (2006) Public transcripts expressed in theatres of cruelty: the
Royal Graves at Ur in Mesopotamia. CAJ 16(2): 123–44.
Dietler, M. (2001) Theorizing the feast. In M. Dietler and B. Hayden (eds),
Feasts: Archaeological and ethnographic perspectives on food, politics and
power. Smithsonian: Washington, DC; 87–125.
Dietler, M. and B. Hayden (eds) (2001) Feasts: Archaeological and ethnographic
perspectives on food, politics and power. Smithsonian: Washington, DC.
Dillon, M. (2002) Girls and women in Classical Greek religion. Routledge: London.
Dumont, L. (1975) On the comparative understanding of non-modern
civilizations. Daedalus 104(2): 153–72.
—(1980) Homo hierarchicus: The caste system and its implications. University
of Chicago Press: Chicago.
—(1986) Essays on individualism: Modern ideology in anthropological
perspective. University of Chicago Press: Chicago.
Dunnell, R. (1980) Evolutionary theory and archaeology. In M. Schiffer (ed.),
Advances in Archaeological Method and Theory. Volume 3. Academic
Press: New York; 38–99.
Dunning, N. and T. Beach (1994) Soil erosion, slope management, and
ancient terracing in the Maya Lowlands. LAA 5(1): 51–69.
Dunning, N., T. Beach, and S. Luzzadder-Beach (2006) Environmental
variability among bajos in the southern Maya Lowlands and its
implications for Ancient Maya civilization and archaeology. In L. Lucero
and B. Fash (eds), Precolumbian water management: Ideology, ritual and
power. University of Arizona Press: Tucson, AZ; 81–99.
Ehrenreich, R., C. Crumely and J. Levy (eds) (1995) Heterarchy and the
analysis of complex societies. APAAA 6. American Anthropological
Association: Arlington, VA.
Eickhoff, T. (1993) Grab und Beigabe: Bestattungssitten der Nekropole von
Tall Ahmad al-Hattū und anderer frühdynastischer Begräbnisstätten im
südlichen Mesopotamien und in Luristān. Profil Verlag: München.
Eisenstadt, Sh. (1963) The political systems of empires. Transaction Publishers:
New York.
Bibliography 173
Ellis, S. (1985) The rising of the Red Shawls: A revolt in Madagascar

1895–1899. Cambridge University Press: Cambridge.
Ellison, R., J. Renfrew, D. Brothwell and N. Seeley (1978) Some food offerings
from Ur, excavated by Sir Leonard Woolley, and previously unpublished.
Journal of Archaeological Science 5(2): 167–77.
Fabian, J. (1983) Time and the other: How anthropology makes its object.
Columbia University Press: New York.
Fash, B. (2005) Iconographic evidence for water management and social
organization at Copán. In E. W. Andrews and W. Fash (eds), Copán: The
history of an ancient Maya kingdom. SAR Press: Santa Fe, NM; 103–38.
Fash, B. and K. Davis-Salazar (2006) Copán water ritual and management:
imagery and sacred place. In L. Lucero and B. Fash (eds), Precolumbian
water management: Ideology, ritual and power. University of Arizona
Press: Tucson, AZ; 129–43.
Feinman, G. and J. Marcus (eds) (1998) Archaic states. SAR Press: Santa Fe, NM.
Feng, L. (2008) Bureaucracy and the state in early China: Governing the
Western Zhou. Cambridge University Press: Cambridge.
Finley, M. (1981) Was Greek civilization based on slave labour? In B. D. Shaw
and R. P. Saller (eds), Economy and society in Ancient Greece. Penguin
Books: London; 97–115.
Flad, R. (2008) Divination and power: a multiregional view of the
development of oracle bone divination in Early China. CA 49(3): 403–37.
Flannery, K. (1998) The ground plans of archaic states. In G. Feinman and
J. Marcus (eds), Archaic states. SAR Press: Santa Fe, NM; 15–57.
—(1999) Process and agency in early state formation. CAJ 9(1): 3–21.
Flannery, K., J. Marcus and R. Reynolds (2009) The flocks of the Wamani:
A study of llama herders on the punas of Ayacucho, Peru. Left Coast
Press: Walnut Creek, CA.
Forsdyke, S. (2006) Land, labor and economy in Solonian Athens: breaking
the impasse between archaeology and history. In J. Blok and A. Lardinois
(eds), Solon of Athens: New historical and philological approaches.
Mnemosyne, Bibliotheca Classica Batava Supplementum. Brill: Leiden;
334–50.
Foucault, M. (1977) Discipline and punish: The birth of the prison. Vintage:
New York.
—(2007) Security, territory and population: Lectures at the Collège de France,
1977–78. Trans. G. Burchell. Palgrave Macmillan: Houndsmills.
174 Bibliography
—(2008) The birth of biopolitics: Lectures at the Collège de France, 1978–79.

Trans. G. Burchell. Palgrave Macmillan: Houndsmills, UK.
Foxhall, L. (1989) Household, gender and property in Classical Athens. CQ
39(1): 22–44.
—(1994) Pandora unbound: a feminist critique of Foucault’s History
of Sexuality. In A. Cornwall and N. Lindisfarne (eds), Dislocating
masculinity: Comparative ethnographies. Routledge: London; 133–45.
—(2002) Access to resources in classical Greece: the egalitarianism of the
polis in practice. In P. Cartledge, E. Cohen and L. Foxhall (eds), Money,
labour and land: Approaches to the economies of ancient Greece. Routledge:
London; 209–20.
Frankfort, H. (1948) Kingship and the gods: A study of ancient Near Eastern
religion as the integration of society and nature. University of Chicago:
Chicago.
French, K. (2007) Creating space through water management at the Classic
Maya site of Palenque, Chiapas. In D. Marken (ed.), Palenque: Recent
investigations at the Classic Maya center. Altamira Press: Lanham, MD;
123–32.
French, K., D. Stuart and A. Morales (2006) Archaeological and epigraphic
evidence for water ritual and management at Palenque. In L. Lucero and
B. Fash (eds), Precolumbian water management: Ideology, ritual and power.
Tucson, AZ; 144–52.
Fukuyama, F. (2011) The origins of political order. Farrar, Strauss and Giroux:
London.
Gabler, S. (2007) Entries, gates and discs in Western Avaradrano. In
H. Wright (ed.), Early state formation in central Madagascar: An
archaeological survey of western Avaradrano. Museum of Anthropology,
University of Michigan: Ann Arbor, MI; 63–6.
Gabrielsen, V. (1994) Financing the Athenian fleet: Public taxation and social
relations. The Johns Hopkins University Press: Baltimore.
Gansell, A. (2007) Identity and adornment in the third-millennium BC
Mesopotamian ‘Royal Cemetery’ at Ur. CAJ 17(1): 29–46.
Geertz, C. (1980) Negara: The theatre state in nineteenth century Bali.
Princeton University Press: Princeton.
Gelb, I. (1965) The ancient Mesopotamian ration system. JNES 24(3):
230–43.
—(1973) Prisoners of war in early Mesopotamia. JNES 32(1–2): 70–98.
Bibliography 175
Giddens, A. (1985) The nation-state and violence. University of California

Press: Berkeley, CA.
Glatz, C., A. Kandler and J. Steele (2011) Pottery production in the
Hittite capital: Cultural selection and drift in the bowl repertoire.
In E. Cochrane and A. Gardner (eds), Evolutionary and interpretive
archaeologies: A discussion. Left Coast Press: Walnut Creek, CA;
199–226.
Goldwasser, O. (2002) Prophets, lovers and giraffes – wor[l]d classification
in ancient Egypt. Göttinger Orientforschungen IV. Reihe Ägypten 38.
Harrassowitz Verlag: Wiesbaden.
Gose, P. (1993) Segmentary state formation and the ritual control of water
under the Incas. CSSH 35(3): 480–514.
—(1996) Oracles, divine kingship and political representation in the Inka
State. Ethnohistory 43(1): 1–32.
—(2000) The State as chosen woman: brideservice and the feeding of
tributaries in the Inka Empire. AA 102(1): 84–97.
Graeber, D. (2004) Fragments of an anarchist anthropology. Prickly Paradigm
Press: Chicago.
—(2007) Lost people: Magic and the legacy of slavery in Madagascar. Indiana
University Press: Bloomington.
Gramsci, A. (1971) Selections from the prison notebooks of Antonio Gramsci.
Ed and trans. Q. Hoare, and G. Nowell Smith. International Publishers
Co.: New York.
—(2000) The Antonio Gramsci reader: Selected writings 1916–1935. D. Forgacs
(ed.). New York University: New York.
Grinin, L. (2004) Democracy and early state. SEH 3(2): 93–147.
Grossberg, L. (1986) On postmodernism and articulation: An interview with
Stuart Hall. Journal of Communication Inquiry 10: 45–60.
Habermas, J. (1989) The structural transformation of the public sphere: An
inquiry into a category of bourgeois society. Trans. T. Burger. Cambridge
University Press: Cambridge.
Hansen, M. H. (1991) The Athenian democracy in the age of Demosthenes:
Structure, principles and ideology. Trans. J. A. Cook. Basil Blackwell:
Oxford.
—(1998) Polis and city-state: An ancient concept and its modern equivalent.
Munksgaard: Copenhagen.
Hansen, T. and F. Stepputat (2006) Sovereignty revisited. ARA 35: 295–315.
176 Bibliography
Hanson, V. (1995) The other Greeks: The family farm and the agrarian roots of
Western Civilization. The Free Press: New York.
Harmansah, Ö. (2007) ‘Source of the Tigris’. Event, place and performance in
the Assyrian landscapes of the early Iron Age. AD 14(2): 179–204.
Harris, E. (2002) Workshop, marketplace and household: the nature
of technical specialisation in Classical Athens and its influence on
economy and society. In P. Cartledge, E. Cohen and L. Foxhall (eds),
Money, labour and land: Approaches to the economies of ancient Greece.
Routledge: London; 67–99.
—(2007) Who enforced the law in Classical Athens? In Symposion 2005:
Vorträge zur griechischen und hellenistischen Rechtsgeschichte. Austrian
Academy of Sciences: Vienna; 159–76.
Harvey, D. (1985) The geopolitics of capitalism. In D. Gregory and J. Urry
(eds), Social relations and spatial structures. Palgrave Macmillan: London;
128–63.
Hastorf, C. (1990) The effect of the Inka state on Sausa agricultural
production and crop consumption. American Antiquity 55(2): 262–90.
Hastorf, C. and S. Johanessen (1993) Pre-Hispanic political change and the
role of maize in the central Andes of Peru. AA 95(1): 115–38.
Hayden, B. (2001) Fabulous feasts: a prolegomenon to the importance
of feasting. In M. Dietler and B. Hayden (eds), Feasts: Archaeological
and ethnographic perspectives on food, politics and power. Smithsonian:
Washington, DC; 23–64.
—(2009) Funerals as feasts: why are they so important? CAJ 19(1): 29–52.
Heidegger, M. (1962) Being and time. Trans. J. Macquarrie and E. Robinson.
Basil Blackwell: Oxford.
Heimpel, W. (1992) Herrentum und Königtum im vor und frühgeschictlichen
Alten Orient. Zeitschrift für Assyriologie und Vorderasiatische Archäologie
82(1): 4–21.
Hobbes, T. (1985 [1651]) Leviathan. Penguin: London.
Hodder, I. (2005) Socialization and feasting at Çatalhöyük: a response to
Adams. American Antiquity 70(1): 189–91
—(2006) The spectacle of daily performance at Çatalhöyük. In T. Inomata
and L. Coben (eds), Archaeology of performance: Theatres of power,
community and politics. Altamira Press: Lanham, MD; 81–102.
Hooper, R. J. (1953) The Attic silver mines in the fourth century bc Annual of
the British School at Athens 48: 200–54.
Bibliography 177
—(1968) The Laurion mines: A reconsideration. Annual of the British School

at Athens 63: 293–326.
Houston, S. (2006) Impersonation, dance and the problem of spectacle
among the Classic Maya. In T. Inomata and L. Coben (eds), Archaeology
of performance: Theatres of power, community and politics. Altamira Press:
Lanham, MD; 135–55.
Hunter, V. (1994) Policing Athens: Social control in the Attic lawsuits, 420–320
BC Princeton University Press: Princeton.
Inomata, T. (2004) The spatial mobility of non-elite populations in Classic Maya
society and its political implications. In J. C. Lohse and F. Valdez, Jr. (eds),
Ancient Maya commoners. University of Texas Press: Austin, TX; 175–96.
—(2006) Plazas, performers, and spectators: political theaters of the Classical
Maya. CA 47(5): 805–42.
Inomata, T. and L. Coben (2006) Overture: an invitation to the archaeological
theatre. In T. Inomata and L. Coben (eds), Archaeology of performance:
Theatres of power, community and politics. Altamira Press: Lanham, MD;
11–44.
Jacobsen, T. (1957) Early political developments in Mesopotamia. Zeitschrift
für Assyriologie und Vorderasiatische Archäologie. 52(1): 91–140.
Jagersma, B. (2007) The calendar of the funerary cult in ancient Lagash.
Biblotheca Orientalis 64(3–4): 289–307.
Jameson, M. (1977/78) Agriculture and slavery in Classical Athens. The
Classical Journal 73(2): 122–45.
—(1990) Domestic space in the Greek city-state. In S. Kent (ed.), Domestic
architecture and the use of space: An interdisciplinary cross-cultural study.
—(1992) Agricultural labor in Ancient Greece. In B. Wells (ed.), Agriculture in
ancient Greece: proceedings of the seventh international symposium at the
Swedish institute at Athens, 16–17 May 1990. Paul Åström: Göteborg; 135–46.
Jay, M. (1984) Marxism and totality: The adventures of a concept from Lukács
to Habermas. University of California Press: Berkeley, CA.
Jennings, J. (2003) The fragility of imperialist ideology and the end of local
traditions, an Inca example. CAJ 13(1): 107–20.
—(2004) La chichera y el patrón: chicha and the energetics of feasting in the
prehistoric Andes. In C. Conlee, D. O. and K. Vaughn (eds), Foundations
of power in the prehispanic Andes. APAAA 14. American Anthropological
Association: Washington, DC; 241–59.
178 Bibliography
Jennings, J. and B. Bowser (eds) (2009) Power, drink and society in the Andes.
University Press of Florida: Gainesville, FL.
Jennings, J. and M. Chatfield (2009) Pots, brewers and hosts: woman’s power
and the limits of central Andean feasting. In J. Jennings and B. Bowser
(eds), Power, drink and society in the Andes. University Press of Florida:
Gainesville, FL; 200–31.
Jessop, B. (2006) Spatial fixes, temporal fixes and spatio-temporal fixes.
In N. Castree and D. Gregory (eds), David Harvey: A critical reader.
Blackwell Publishing: Oxford; 142–66.
—(2007) State power. Polity Press: Cambridge.
Jing, Zhichun (2008) Comments on Flad (2008). CA 49(3): 424–5.
Johnson, M. (2006) Discussion article: On the nature of theoretical
archaeology and archaeological theory. AD 13(2): 117–82.
Johnston, K. (2004) Lowland Maya water management practices: the
household exploitation of rural wells. Geoarchaeology 9(3): 265–92.
Jones, A. (2007) Memory and material culture. Cambridge University Press:
Cambridge.
Joyce, R. (2005) Archaeology of the body. ARA 34: 139–58.
Kapferer, B. (2011) Strathern’s new comparative anthropology: thoughts from
Hagen and Zambia. Common Knowledge 17(1): 104–10.
Keeley, L. (1996) War before civilization: The myth of the peaceful savage.
Oxford University Press: Oxford.
Kent, R. (1970) Early kingdoms in Madagascar 1500–1700. Holt, Rinehart and
Winston: New York.
Kertzer, D. (1996) Politics and symbols: The Italian Communist Party and the
fall of Communism. Yale University Press: New Haven.
Khare, R. S. (ed.) (2006) Caste, Hierarchy, and Individualism: Indian Critiques
of Louis Dumont’s Contributions. Oxford University Press: Oxford.
Kohl, P. (1984) Force, history, and the evolutionist paradigm. In M. Spriggs
(ed.), Marxist Perspectives in Archaeology. Cambridge University Press:
—(1987) State formation: useful concept or idée fixe? In T. Patterson
and C. Gailey (eds), Power relations and state formation. Sheffield
Publishing Company: Salem, WI; 27–34.
Kuper, A. (2000) If memes are the answer what is the question? In
R. Aunger (ed.), Darwinizing culture: The status of memetics as a science.
Oxford University Press: Oxford; 175–88.
Bibliography 179
Kurtz, D. (1996) Hegemony and anthropology: Gramsci, exegeses,

reinterpretations. Critique of Anthropology 16(2): 103–35.
—(2006) Political power and government: negating the anthropomorphized
state. SEH 5(2): 91–111.
Kus, S. (1992) Toward an archaeology of body and soul. In J.-C. Gardin
and C. Peebles (eds), Representations in archaeology. University of Indiana
Press: Bloomington, IN; 168–77.
—(2007) Appendix E. Ambohimanga: creating a capital and a polity in
thought, deed and dirt. In H. Wright, H. (ed.), Early state formation
in central Madagascar: An archaeological survey of western Avaradrano.
Museum of Anthropology, University of Michigan: Ann Arbor, MI; 301–12.
Kus, S. and V. Raharijaona (1998) Between earth and sky there are only a few
large boulders: sovereignty and monumentality in Central Madagascar.
JAA 17: 53–79.
—(2000) House to palace, village to state: scaling up architecture and
ideology. AA 102(1): 98–113.
—(2001) ‘To dare to wear the cloak of another before their very eyes’: state
co-optation and local re-appropriation in mortuary rituals of Central
Madagascar. In M. Chesson (ed.), Social memory, identity and death:
Intradisciplinary perspectives on mortuary rituals. APAAA 10. John Wiley
& Sons: Washington, DC; 114–31.
—(2006) Visible and vocal: sovereigns of the early Merina (Madagascar) state.
In T. Inomata and L. Coben (eds), Archaeology of performance: Theatres of
power, community and politics. Altamira Press: Lanham, MD; 303–29.
Laclau, E. and C. Mouffe (2001) Hegemony and Socialist Strategy: Towards a
Radical Democratic Politics. 2nd edn. Verso Books: London.
Lapinkivi, P. (2004) The Sumerian sacred marriage in light of comparative
evidence. State Archives of Assyria Studies 15. The Neo-Assyrian Text
Corpus Project: Helsinki.
Larson, P. (2000) History and Memory in the Age of Enslavement: Becoming
Merina in Highland Madagascar, 1770–1822. Heinemann: Portsmouth, NH.
—(2001) Austronesian mortuary ritual in history: transformation of secondary
burial (Famadihana) in highland Madagascar. Ethnohistory 48(1–2): 123–55.
Latour, B. (2005) Reassembling the Social: An Introduction to Actor-Network
Theory. Oxford University Press: Oxford.
Law, J. (1992) Notes on the theory of the actor-network: ordering, strategy
and heterogeneity. Systems Practice 5: 379–93.
180 Bibliography
Leahy, A. (1995) Ethnic diversity in ancient Egypt. In J. Sasson (ed.), Civilizations

of the Ancient Near East. Vol. I. Charles Scribner’s Sons: New York; 225–34.
Legrain, L. (1936) Ur Excavations III: Archaic seal impressions. Trustees of the
British Museum and the Museum of the University of Pennsylvania: London.
Lemke, T. (2007) ‘An indigestible meal?’ Foucault, governmentality and state
theory. Distinktion: Scandinavian Journal of Social Theory 15: 43–64.
Lenin, V. I. (1976 [1917]) The State and revolution: The Marxist teaching
on the State and the tasks of the proletariat in the revolution. Foreign
Languages Press: Peking.
Leonard, R. and T. Jones (1987) Elements of an inclusive evolutionary model
for archaeology. JAA 6(3): 199–219.
LeVine, T. (1987) Inka labor service at the regional level: the functional
reality. Ethnohistory 34(1): 14–46.
LeVine, T. (ed.) (1992) Inka storage systems. University of Oaklahoma Press:
Norman, OK.
Lévi-Straus, C. (1967) Structural Anthropology. Basic Books: Garden City, NY.
Lincoln, B. (1994) Authority: Construction and Corrosion. University of
Chicago Press: Chicago.
Liu, L. (2009) State emergence in early China. ARA 38: 217–32.
Llewellyn-Jones, L. (2007) House and veil in ancient Greece. In R. Westgate,
N. Fischer and J. Whitley (eds), Building Communities: House, Settlement
and Society in the Aegean and Beyond. Proceedings of a conference held at
Cardiff University, 17–12 April 2001. British School at Athens: London;
251–8.
Lloyd, G. E. R. (2010) History and human nature: cross-cultural universals
and cultural relativities. Interdisciplinary Science Reviews 35(3–4): 201–14.
Lloyd, S. (1969) Back to Ingharra: some further thoughts on the excavations
at East Kish. Iraq 31(1): 40–8.
Lucero, L. (2006) Water and Rutual: The Rise and Fall of Classic Mayan
Rulers. University of Texas Press: Austin, TX.
Lucero, L. and B. Fash (eds) (2006) Precolumbian Water Management:
Ideology, Ritual and Power. University of Arizona Press: Tucson, AZ.
Lull, V. and R. Micó (2011) Archaeology of the origin of the state. Oxford
University Press: Oxford.
McCarthy, T. (1989) Introduction to J. Habermas, The Structural
Transformation of the Public Sphere: An Inquiry into a Category of
Bourgeois Society. Trans. T. Burger. Ploty Press: Cambridge; xi–xiv.
Bibliography 181
McEwan, G. (2006) The Incas: New perspectives. W. W. Norton & Co.: Santa
Barbara, CA.
McIntosh, S. (1999) Pathways to complexity: an African perspective. In S.
McIntosh (ed.), Beyond Chiefdoms: Pathways to Complexity in Africa.
Maekawa, K. (1973/74) The development of the E-MI in Lagash during Early
Dynastic III. Mesopotamia 8–9: 77–144.
Magid, G. (2001) Micromanagement in the é-mi/dBa-ú: notes on the
organization of labor at Early Dynastic Lagash. In T. Abusch, P.-A.
Beaulieu, J. Huehnergard, P. Machinist, P. Steinkeller (eds), Historiography
in the cuneiform world. Proceedings of the 45th Rencontre Assyriologique
Internationale. Part 1. CDL Press: Bethesda, MD; 313–28.
Mann, M. (1986) The Sources of Social Power. Vol. 1. Cambridge University
Press: Cambridge.
Marchesi, G. (2004) Who was buried in the Royal Tombs of Ur? The
epigraphic and textual data. Orientalia 73: 153–97.
Marcus, J. (2008) The archaeological evidence for social evolution. ARA 37:
51–66.
Marinetto, M. (2007) Social theory, the state and modern society: The state in
contemporary social thought. Open University Press: Maidenhead.
Marken, D. (ed.) (2007) Palenque: Recent Investigations at the Classic Maya
Center. Altamira Press: Lanham, MD.
Mauss, M. (1990) The Gift: The form and reason for exchange in archaic
societies. Trans. W. Halls. Routledge: New York.
Meskell, L. (2002) Private life in New Kingdom Egypt. Princeton University
Press: Princeton.
Mitchell, T. (1999) Society, economy, and the state effect. In G. Steinmetz (ed.),
State/Culture: State-Formation after the Cultural Turn. Cornell University
Press: Ithaca, NY; 76–97.
Miyazaki, M. (2007) Public coercive power of the Greek Polis. On a recent
debate. Bulletin of the Institute for Mediterranean Studies, Waseda
University 5: 101–14.
Mizoguchi, K. (2009) Nodes and edges: a network approach to
hierarchisation and state formation in Japan. JAA 28(1): 14–26.
Molleson, T. and D. Hodgson (1993) A cart driver from Ur. Archaeozoologia
6: 93–106.
—(2000) The porters of Ur. Isimu 3: 101–18.
182 Bibliography
—(2003) The human remains from Woolley’s excavations at Ur. Iraq 65:
91–129.
Monroe, J. C. (2010) Power by design: architecture and politics in precolonial
Dahomey. Journal of Social Archaeology 10(3): 367–97.
Moore, J. (2006) ‘The Indians were much given to their taquis’: drumming
and generative categories in ancient Andean funerary processions. In
T. Inomata and L. Coben (eds), Archaeology of performance: Theatres of
power, community and politics. Altamira Press: Lanham, MD; 47–79.
Moorey, P. R. S. (1977) What do we know about the people buried in the
Royal Cemetery? Expedition 20/1: 24–40.
—(1978) Kish excavations 1923–1933. Claredon Press: Oxford.
Moortgat, A. (1949) Tammuz: Der Unsterblichkeitsglaube in der
altorientalischen Bildkunst. DeGruyter: Berlin.
Morera, E. (1990) Gramsci’s historicism: A realist interpretation. Routledge:
London.
Morris, C. and D. E. Thompson (1985) Huánuco Pampa: An Inca City and its
Hinterland. Thames and Hudson: London.
Morris, I. (2000) Archaeology as cultural history. Blackwell: Oxford.
Murra, J. (1962) Cloth and its functions in the Inka state. AA 64(4): 710–28.
—(1980) The economic organization of the Inka state. JAI Press Inc.:
Greenwich, CT.
Nafisi, M. (2004) Class, embeddedness and the modernity of ancient Athens.
CSSH 46(2): 378–410.
Nagle, D. B. (2006) The household as the foundation of Aristotle’s polis.
Nevett, L. (1995) Gender relations in the Classical Greek household: the
archaeological evidence. The Annual of the British School at Athens 90:
363–81.
—(2010a) Domestic space in classical antiquity. Key Themes in Ancient
History. Cambridge University Press: Cambridge.
—(2010b) Domestic culture in Classical Greece. In O. Hekster and S.
Mols (eds), Cultural Messages in the Graeco-Roman world. Acta of the
BABESCH 80th Anniversary Workshop Radboud University Nijmegen,
September 8th 2006. Peeters: Leuven; 49–56.
—(2011) Towards a female topography of the ancient Greek city: case studies
from Late Archaic and Early Classical Athens (c.520–400 bce). Gender
and History 23(3): 576–96.
Bibliography 183
O’Brien, M. J., R. L. Lyman, A. Mesoudi and T. L. VanPool (2010) Cultural

traits as units of analysis. Philosophical Transactions of the Royal Society B
365: 3797–806.
O’Connor, D. (1995) Beloved of Maat, the horizon of Re: the royal palace in
New Kingdom Egypt. In D. O’Connor and D. Silverman (eds), Ancient
Egyptian Kingship. Brill: Leiden; 263–300.
O’Connor, D. and D. Silverman (1995) Introduction to D. O’Connor and
D. Silverman (eds), Ancient Egyptian Kingship. Brill: Leiden; xvii–xxvii.
Osborne, R. (1997) Law, the democratic citizen and the representation of
women in Classical Athens. Past and Present 155: 3–33.
Owen, W. F. M. (1833) Narrative of voyages to explore the shores of Africa,
Arabia and Madagascar performed in H.M. ships Leven and Barracouta
under command of Capitan W.F.M. Owen by command of the Lord
Comminsoners of the Admiralty. Volume II. Richard Bentley: London.
Parkinson, R. (2002) Poetry and Culture in Middle Kingdom Egypt: A Dark
Side to Perfection. Equinox Publishing: London.
Pauketat, T. (2007) Chiefdoms and other archaeological delusions. Altamira
Press: Lanham MD.
Paynter, R. (1989) The archaeology of equality and inequality. ARA 18: 369–99.
Peirce, C. (1931) Prolegomena to an apology for Pragmaticism. In C.
Hartshorne and P. Weiss (eds), Collected papers of Charles Sanders Peirce.
Vol. 4. Harvard University Press: Cambridge, MA; 531
Plato (1974) The Republic. Trans. G. Grube. Hackett Publishing Co:
Indianapolis, IN.
Pluciennik, M. (2005) Social evolution. Duckworth: London.
Pollock, S. (1999) Ancient Mesopotamia: The Eden that never was.
—(2003) Feast, funerals, and fast food in early Mesopotamian states. In
T. Bray (ed.), The archaeology and politics of feasting in early states and
empires. Kluwer Academic/Plenum Publishers: New York; 17–38.
—(2007a) The Royal Cemetery of Ur: ritual, tradition, and the creation of
subjects. In M. Heinz and M. Feldman (eds), Representations of political
Power: Case Histories from Times of Change and Dissolving Order in the
Ancient Near East. Eisenbrauns:Winona Lake, IN; 89–110.
—(2007b) Death of a household. In N. Laneri (ed.), Performing death: Social
analyses of funerary traditions in the ancient Near East and Mediterranean.
Oriental Institute of the University of Chicago: Chicago; 209–22.
184 Bibliography
Poma de Ayala, F. G. 1936 [1615] El primer nueva corónica y buen gobierno.

Sigio XXI: Institut d’Ethnologie: Paris.
Prentice, R. (2010) The exchange of goods and services in Pre-Sargonic Lagash.
Ugarit-Verlag: Münster.
Prentiss, A., I. Kuijt and J. Chatters (eds) (2009) Macrevolution in human
prehistory: Evolutionary theory and processual archaeology. Springer: New
York.
Pyburn, K. A. (ed.) (2004) Ungendering civilization. Routledge: London.
Raison-Jourde, F. (1991) Bible et pouvoir à Madagascar au XIXe siècle:
Invention d’une identité chrétienne et construction de l’État (1780–1880).
Éditions Karthala: Paris.
Redman, C. (1978) The rise of civilization: From early farmers to urban society
in the ancient Near East. W. H. Freeman & Co. Ltd: San Francisco.
Renfrew, C. and P. Bahn (2012) Archaeology: Theories, methods and practice.
6th edn. Thames and Hudson: London.
Richards, J. (2000) Modified order, responsive legitimacy, redistributed
wealth: Egypt 2260–1650 bc. In J. Richards and M. Van Buren (eds),
Order, legitimacy, and wealth in ancient states. Cambridge University
Press: Cambridge; 36–45.
Richards, J. and M. Van Buren (eds) (2000) Order, legitimacy, and wealth in
ancient states. Cambridge University Press: Cambridge.
Richerson, P. and R. Boyd (2001) Institutional evolution in the Holocene: the
rise of complex societies. In W. G. Runciman (ed.), The origin of human
social institutions. Proceedings of the British Academy 110. Published for
the British Academy by Oxford University Press: Oxford; 197–204.
Robinson, D. M. and J. W. Graham (1938) Excavations at Olynthus Volume
8: The hellenic house: A study of the houses found at Olynthus with a
detailed account of those excavated in 1931 and 1934. The Johns Hopkins
University Press: Baltimore.
Rosenberg, J. (1994) The empire of civil society: A critique of the realist theory
of International Relations. Verso Books: London.
Rosenberg, M. (2009) Proximate causation, group selection and the evolution
of hierarchical human societies: systems, processes and patterns. In A.
Prentiss, I. Kuijt and J. Chatters (eds), Macrevolution in human prehistory:
Evolutionary theory and processual archaeology. Springer: New York; 23–50.
Routledge, B. (2003) The antiquity of the nation? Critical reflections from the
ancient Near East. Nations and Nationalism 9(2): 213–33.
Bibliography 185
—(2004) Moab in the Iron Age: Hegemony, polity archaeology. University of

Pennsylvania Press: Philadelphia.
Sanchez, J. (2005) Ancient Maya royal strategies: creating power and identity
through art. Ancient Mesoamerica 16: 261–75.
Ste Croix, G. de (1981) The class struggle in the ancient Greek world: From the
Archaic age to the Arab conquest. Gerald Duckworth & Co. Ltd: London.
Scarborough, V. (1998) Ecology and ritual: Water management and the Maya.
LAA 9(2): 135–59.
—(2003) The flow of power: Ancient water systems and landscapes. SAR Press:
Santa Fe, NM.
Scarborough, V. and G. Gallopin (1991) A water storage adaptation in the
Maya lowlands. Science 251: 658–62.
Schele, L. and M. E. Miller (1986) The blood of kings: Dynasty and ritual in
Maya art. Thames and Hudson: New York.
Scheidel, W. (1995) The most silent women of Greece and Rome: rural labour and
women’s life in the ancient world. (I). Greece and Rome 42: 202–17.
—(1996) The most silent women of Greece and Rome: rural labour and
women’s life in the ancient world. (II). Greece and Rome 43: 1–10.
Schmandt-Besserat, D. (2001) Feasting in the ancient Near East. In M. Dietler and
B. Hayden (eds), Feasts: Archaeological and ethnographic perspectives on food,
politics and power. Smithsonian Institute Press: Washington, DC; 391–403.
Scott, J. (1985) Weapons of the weak: Everyday forms of peasant resistance.
Yale University Press: New Haven.
—(1990) Domination and the arts of resistance. Yale University Press: New Haven.
—(1998) Seeing like a State. Yale University Press: New Haven.
Schwartz, G. (2007) Appendix D. Population and production in eighteenth-
century Madagascar: analyzing archaeological evidence of the subsistence
economy of the central highlands. In H. Wright (ed.), Early state
formation in central Madagascar: An archaeological survey of western
Avaradrano. Museum of Anthropology, University of Michigan: Ann
Arbor, MI; 289–300.
Selz, G. (2004) Early Dynastic vessels in ‘ritual’ contexts. Wiener Zeitschrift
für die Kundes des Morgenlandes 94: 185–223.
Seri, A. (2005) Local power in Old Babylonian Mesopotamia. Equinox:
London.
Shanks, M. (1999) Art and the early Greek city state: An interpretive
archaeology. Cambridge University Press: Cambridge.
186 Bibliography
Shanks, M. and C. Tilley (1988) Social theory and archaeology. University of

New Mexico Press: Albuquerque, NM.
Sibree, J. (1870) Madagascar and its people. Notes of a four year residence.
With a sketch of the history, position and prospects of missionary work
amongst the Malagasy. The Religious Tract Society: London.
Silverblatt, I. (1987) Moon, sun and witches: Gender ideologies and class in
Inca and Colonial Peru. Princeton University Press: Princeton.
Skinner, Q. (1989) The State. In T. Ball, J. Farr and R. Hanson (eds),
Political innovation and conceptual change. Cambridge University Press:
Smith, A. T. (2003) The political landscape: Constellations of authority in early
complex polities. University of California Press: Berkeley, CA.
—(2004) The end of the essential archaeological subject. AD 11(1): 1–20.
—(2006) Representational aesthetics and political subjectivity: the
spectacular in Urartian images of performance. In T. Inomata and L.
Coben (eds), Archaeology of performance: Theatres of power, community
and politics. Altamira Press: Lanham, MD; 103–34.
—(2011) Archaeologies of sovereignty. ARA 40: 415–32.
Smith, M. (2005) Networks, territories, and the cartography of ancient
states. Annals of the Association of American Geographers 95(4):
832–49.
Smith, M. E. and P. Peregrine (2012) Approaches to comparative analysis in
archaeology. In M. E. Smith (ed.), The comparative archaeology of complex
societies. Cambridge University Press: Cambridge; 4–20.
Spencer, C. and E. Redmond (2001) Multilevel selection and political
evolution in the Valley of Oaxaca, 500–100 bc JAA 20: 195–229.
Stein, G. (1998) Heterogeneity, power, and political economy: some current
research issues in the archaeology of Old World complex societies. Journal
of Archaeological Research 6(1): 1–44.
Steinkeller, P. (1987) The administrative and economic organization of the Ur
III state: the core and the periphery. In McG. Gibson and R. Biggs (eds),
The organization of power. Aspects of bureaucracy in the ancient Near East.
The Oriental Institute: Chicago; 19–41.
Stone, A. (1995) Images from the underworld: Naj Tunich and the tradition of
Maya cave painting. University of Texas Press: Austin, TX.
Surette, F. (2008) The acllacona: The Inca chosen women in archaeology and
history. Unpublished M.A. dissertation. Trent University.
Bibliography 187
Taylor, C. (2011) Women’s social networks and female friendship in the

ancient Greek city. Gender and History 23(3): 703–20.
Teeter, E. (1997) The presentation of Maat: Ritual and legitimacy in ancient
Egypt. Oriental Institute: Chicago.
Tilley, C. (1990) Michel Foucault: Towards an archaeology of archaeology.
In C. Tilley (ed.), Reading material culture. Blackwell: Oxford;
281–347.
Tilly, C. (1975) Reflections on the history of European state-making. In C.
Tilly and G. Ardant (eds), The formation of national states in Western
Europe. Princeton University Press: Princeton; 3–83.
Tinney, S. (1998) Death and burial in early Mesopotamia: the view from the
texts. In Zettler and L. Horne (eds), Treasures from the Royal Tombs of Ur.
University of Pennsylvania Press: Philadelphia; 26–8.
Turchin, P. (2011) Warfare and the evolution of social complexity: A
multilevel-selection approach. Structure and Dynamics 4(3): 1–37.
Veni, G. (1990) Maya utilization of karst groundwater resources.
Environmental Geology and Water Sciences 16: 63–6.
de Veyrières, P. and G. de Méritens (1967) Le livre de la sagesse malgache:
Proverbes, dictons, sentences, expressions figurées et curieuses. Éditions
Maritimes et d’Outre-mer: Paris.
Vidale, M. (2011) PG 1237, Royal Cemetery of Ur: patterns in death. CAJ
21(3): 427–51.
Visicato, G. (1995) The bureaucracy of Šuruppak: Administrative centres,
central offices, intermediate structures and hierarchies in the economic
documentation of Fara. Ugarit-Verlag: Münster.
Viveiros de Castro, E. (2004) Perpectival anthropology and the method of
controlled equivocation. Tipití 2(1): 1–20.
van der Vliet, E. (2005) Polis. The problem of statehood. SEH 4 (2): 120–50.
Vogel, H. (2008) Wie man Macht macht: Eine macht- und genderkritische
Untersuchung der frühesten Repräsentationen von Staatlichkeit.
Unpublished Ph.D. dissertation, Free University of Berlin.
Vogt, E. (1969) Zinacantan: A Maya community in the highlands of Chiapas.
Harvard University Press: Cambridge.
Vogt, E. and D. Stuart (2005) Some notes on ritual caves among the
ancient and modern Maya. In J. Brady and K. Prufer (eds), In the
maw of the earth monster: Mesoamerican ritual cave use. University of
Texas Press: Austin, TX; 155–85.
188 Bibliography
Wagner-Hasel, B. (2003) Women’s life in Oriental seclusion? On the history

and use of a topos. In M. Golden and P. Toohey (eds), Sex and difference in
ancient Greece and Rome. Edinburgh University Press: Edinburgh; 241–52.
Walker, S. (1983) Women and housing in classical Greece. In A. Cameron
and A. Kuhrt (eds), Images of women in Antiquity. Croom Helm:
Beckenham; 81–93.
Weber, J. and R. Zettler (1998) Metal vessels. In R. Zettler and L. Horne
(eds), Treasures from the Royal Tombs of Ur. University of Pennsylvania
Press: Philadelphia; 125–42.
Weber, M. (1978) Economy and society. 2 Vols. G. Roth and C. Wittich (eds).
University of California Press: Berkeley, CA.
Westgate, R. (2007) The Greek house and the ideology of citizenship. World
Archaeology 39(2): 229–45.
Westgate, R., N. Fischer and J. Whitley (eds) (2007) Building Communities:
House, Settlement and Society in the Aegean and Beyond. Proceedings of a
conference held at Cardiff University, 12–17 April 2001. British School at
Athens: London.
Wernke, S. (2006) The politics of community and Inka statecraft in the Colca
Valley, Peru. LAA 17(2): 177–208.
Wetterstrom, W. and H. Wright (2007) Appendix C. A contribution to the
paleo-ethnobotany of the central highlands of Madagascar. In H. Wright
(ed.), Early state formation in central Madagascar: An archaeological survey
of western Avaradrano. Museum of Anthropology, University of Michigan
Press: Ann Arbor, MI; 281–8.
Wilson, D. S. and E. Sober (1994) Reintroducing group selection to the
human behavioral sciences. Behavioral and Brain Sciences 17: 585–654.
Winter, I. (2009) Reading ritual in the archaeological record: deposition
pattern and function of two artefact types from the Royal Cemetery of Ur.
In On Art in the Ancient Near East. Collected Essays Vol. 2. Brill: Leiden;
pp. 227–70.
Witfogel, K. (1957) Oriental despotism: A comparative study of total power.
Yale University Press: New Haven.
Wood, E. M. (1988) Peasant-citizen and slave: The foundations of Athenian
democracy. Verso Books: London.
Woolley, L. (1934) Ur excavations II: The Royal Cemetery. Trustees of the British
Museum and the Museum of the University of Pennsylvania: London.
—(1955) ) Ur excavations IV: The early periods. The British Museum: London.
Bibliography 189
Wright, H. (1977) Recent research on the origin of the state. ARA 6: 379–97.
—(ed.) (2007) Early State Formation in Central Madagascar: An
Archaeological Survey of Western Avaradrano. Museum of Anthropology,
University of Michigan: Ann Arbor, MI.
Yoffee, N. (2000) Law courts and the mediation of social conflict in ancient
Mesopotamia. In J. Richards and M. Van Buren (eds), Order, legitimacy,
and wealth in ancient states. Cambridge University Press: Cambridge;
46–63.
—(2005) Myths of the archaic state: Evolution of the earliest cities, states, and
civilizations. Cambridge University Press: Cambridge.
Zettler, R. (1998) The Royal Cemetery of Ur. In R. Zettler and L. Horne (eds),
Treasures from the Royal Tombs of Ur. University of Pennsylvania Press:
Philadelphia; 21–32.
Zettler, R. and L. Horne (eds) (1998) Treasures from the Royal Tombs of Ur.
University of Pennsylvania Press: Philadelphia.
Index
Abrams, Philip 19–20 bauplan (pl. baupläne), evolutionary

aclla (pl. acllacona) 90, 92–9 9, 13, 158
acllawasi 92–6 Berent, Moshe 76–8
Actor-Network Theory 71–3 Birkleyn (Turkey) 111
adriana 56 Blake, William 157, 166
Agamben, Giorgio 155 Blanton, Richard 32–4
Algaze, Guillermo 139 Bloch, Maurice 25, 55, 57
Ambohimanga (Madagascar) 54, 62 boule (council) 75
Anderson, Gregory 78 Bourdieu, Pierre 15
Andes, Andean 90–1, 94–8 Bray, Tamara 96
Andrantsay (Madagascar) 54 Britain, British 54–5, 58
Andriamasinavalona 52
Andrianampoinimerina 49, 53–6, camayoc 92
61–5 Çatal Höyük 104–5
andron 87 caves, Mayan use of 115–17
Antananarivo (Madagascar) 54 Chaak 115
Antipater of Macedon 75 Chanca 90
Archaeology: theories, methods and Chang, Kwang-chih 28
practices (6th edn.) 67 Chapman, Robert 11
aríbalos 96–7 ch’een 117
Aristophanes 83 Chiapas (Mexico) 117, 123
Aristotle 74, 79–80, 85, 163–4 chica 90–9
Art and the early Greek state 14 Chichén Itzá 114–15 (‘Cenote of
articulation 20, 22, 38, 44, 67, 80, Sacrifice’) 117
87–90, 109 China 28–32
Assyria, Assyrian 111, 136 Erlitou 28
Athens 7, 67, 73–88, 163–4 Erligang 28
Attica 75 Longshan 31
Avaradrano (Madagascar) 52 Shang 28–32, 46, 111
ayllu 91, 98 Western Zhou 29, 31, 103
ayni 91, 98 Clastres, Pierre 18
Clausewitz, Carl von 17
Babylon 139 Cleisthenes 75, 82
Bahn, Paul 67–8 circumcision ritual (Madagascar)
Baines, John 27–30, 45, 47 55–6
banquets/feasting 141–8 Cohen, Andrew 135, 141
bajo 118 commensal politics 143–6
Bara-namtara 144 comparison, cross-cultural 158–66
Bartelson, Jens 3 Connerton, Paul 105–7
192 Index
convex watershed 118 Foucault, Michel 1, 4–5, 37, 101, 103,

Cooper, Jerrold 137 155
Copán 114–15, 122–3 governmentality 21, 101
Coya 92 ‘principle of intelligibility’ 20–2,
cuarac 91 77
Cuzco 90, 92, 95–7 sovereignty 21–2
Foxhall, Lin 83
Darwinian archaeology 10, 12–13 France 55, 101
Davis-Salazar, Karla 122
The Death of Gilgamesh 135–6 Gallopin, Gary 118
The Death of Ur-Namma 135–6 Geertz, Clifford 107–10
The Debate between Winter and gender 74, 83–8
Summer 146 Giddens, Anthony 24, 102
Delian League 81 Girsu (modern Telloh) 144, 149,
demes 75 151–3
demos 78 Gose, Peter 25, 60, 95, 98–9
Diakonoff, Igor 148 Graeber, David 101
Dibni Su 111 Gramsci, Antonio 6, 35–47, 66,
Dickson, Bruce 135 69–73, 80, 124
Dietler, Michael 145, 147 ‘common-sense’ 38–41, 44–5, 49,
Dokimastes 78 57, 66, 109, 159
Dos Pilas 114–15 hegemony, hegemonic 36–47, 49,
Dumont, Louis 160–1 60–1, 68–9, 79–80, 84, 88, 90,
Dumuzi 136 99–101, 109–12, 117, 124–5,
156, 158, 162
é-gal 137 historical bloc 42–4, 69–71, 125
Egypt 14, 23–4, 27–30 intellectuals 38–41, 46
Ecclesiazusae (‘women in the Greece, 14, 163
Assembly’) 83 Grinin, Leonid 76
ekklesia (the Assembly) 75, 78, 82–3 Guantanamo Bay 155
en 137 gunaikon 85
ensik 137–8
entrain, entrainment 31–2, 110–12 Habermas, Jürgen 101–2
Eratosthenes 77 Hall, Stuart 73
Ereškigala 135 Hansen, Mogens Herman 75, 78
Euphiletos 77 Harmansah, Omür 111
Harris, Edward 77
famadihana 64–5 Harvey, David 67
Fargher, Lane 32–4 hasina 53, 57, 112
Fash, Barbara 122 Hayden, Brian 143–4
firenena 56 Hegel, Georg W. F. 10
Flad, Rowan 30 Hegemony and Socialist Strategy 70
Flannery, Kent 2 Heidegger, Martin 107
fokonolana 61 heterarchy 10, 16
Index 193
hetairai 83 Queen of Lagash 144, 151, 153

‘high culture’ (in early states) 27–32 Latour, Bruno 72, 88
Hobbes, Thomas 10, 18, 32–3 Laurion silver mines 81
Hodder, Ian 104–5 legitimacy 3, 17–18, 28–9, 31, 45,
Hogson, Dawn 130, 150 107, 135
Honduras 122 Lemke, Thomas 5
hoplite 76, 80 de León, Pedro de Cieza 95
Horus 23 Lévi-Strauss 159
House of the Woman (é-MI)/House of libations 143, 147
the goddess Bau’ (é-dba-Ú) 144 liturgies (Athens) 81–2
houses, Ancient Greek 85–8 Lloyd, Seton 139
hova 56 Louis XIV 23
How societies remember 105 lyre(s) 130, 141–3, 147
huacas 94 Lysistrata 83
Huánuco Pampa 95–6
Hunter, Virginia 77 ma’at 23
Madagascar 6, 47, 49–66, 112
Illustrated London News 101 Mama Huaco 97
Imerina 6, 47, 49–66, 110–11 mamacona 92–4
Inanna 136 Manco Capac 97
Inca Empire 7, 60, 67, 73, 88–100, Mann, Michael 16
103 Manzananes, Pilar Alberti 95
Inca Emperor/ruler (Sapa Inca) 25, Marchesi, Gianni 148
88, 92–9, 165 Marovatana 58
Incap rantin (‘the Inca’s substitute’) Mascerene Islands 52, 55
25 Mauritius 52
incorporating practices 106–7 Maya, Mayan 7, 113–25
index/indexical 31–2, 49, 106–7, 111 Classic 113–19
Inomata, Takeshi 104, 108 Late Classic 117
inscribing practices 106–7, 111 Post-Classic 117
Iran 139 Marxism, Marxist 11, 35, 39, 42,
69–71
Jensen, Caspar Bruun 159 memory, social 105–7
Jessop, Bob 9, 25 Merina 49–66, 109
Johnston, Kevin 121 Meskell, Lynn 14
Jones, Andrew 106 Mesopotamia 27–30, 139, 143
Early Dynastic 103, 127–8, 136–9,
Kish 139 141, 143–55
Kuo Yü 32 Jemdat Nasr 137
Kus, Susan 49, 61, 109 Old Babylonian 30
kyrios 83–4 Uruk 137
metics 78, 81, 83, 85
Laclau, Ernesto 42, 70–3 min’ka 91, 98–9
Lagash 137–8 mit’a 91, 94–5
194 Index
Mitchell, Timothy 19–20 Pu-Abi, Queen (nin) 130, 141

mitmaq 92 public sphere 101–2
Miyazaki, Makoto 77
Molleson, Theya 130, 150 Radama I 53, 58–61, 64–6, 162
Moorey, P. Roger S. 134, 139 Radama II 55
Morris, Ian 79 Raharijaona, Victor 49, 61, 109
Mouffe, Chantal 42, 72–3 Raison-Jourde, Françoise 64
Ralambo 57
Nahuatl 120 Ranalova I 54
Naj Tunich 117 Ranalova II 55, 64
Nanna 127 rational choice theory 32–4, 69, 103
neo-evolutionism 3, 6, 9–11, 67 ration system (Mesopotamia, Early
networks 12, 16 Dynastic III) 151–3, 156
Nietzsche, Friedrich 4 Renfrew, Colin 67–8
Rome 103
oikos (household) 74, 81, 83–8 Rosenberg, Justin 76
order 27–31, 34, 44–5, 55–7, 60–1, routine 103–5, 111–13, 124–5, 149,
65, 101, 110–12, 155–6 155
Oriental Despotism 117 royal bath, ritual of (Madagascar)
Orwell, George 4 55–7, 62
Osiris 23
osteo-pyromancy (oracle bones) sacred marriage rite (Mesopotamia)
30–2, 46, 111 134, 136
Šagšag 144
Pachacuti Inca Yupanqui 90 Saturday Creek (Belize) 121
Palenque 123 Scarborough, Vernon 118–20
Peirce, Charles Sandars 106 Schmandt-Besserat, Denise 145
Peregrine, Peter 158 Scott, James 61, 102
performance, performative 104–13 Seri, Andrea 30
Peru 90, 95 Shanks, Michael 14
Péten 115, 117 Silverblatt, Irene 98
Pharaoh 23–3, 165 slaves, slavery (Ancient Greece) 74,
Piraeus 81 80–3, 85, 87–8, 163–4
Pizarro, Francisco 90 Smith, Adam T. 6, 14–19, 26, 109,
Plato 79, 163–4 158
polis 14, 22, 74–85 Smith, Michael E. 158–9
political authority 15–18, 22, 26–31, Solon 82
33, 45, 57, 62, 102–3, 107, sovereignty 18, 21–2, 26–7, 68–9,
110–13, 156–8 99–100, 103, 107, 124–5, 158
The Politics 74, 80, 163 ‘spatio-temporal fix’ 69, 88
Pollock, Susan 135, 148–50, 154 spectacle 101–13, 124–5, 149, 155–6
post-processual archaeology 4, 14 ‘Standard of Ur’ 140–3, 147
Private life in New Kingdom Egypt state, state-formation 2–4, 6, 9–15,
14 17, 36, 67–8, 102–3, 157–8
Index 195
Sumer, Sumerian 139, 141, 143, Urartu 18, 109–10

146–8, 154–5 Uru-inimgina 144
Surette, Flannery 95 Uruk (Warka) 137–8
Šuruppak (Fara) 151–2
Susa 139 Vazimba 50, 56–7
vicuña 98
Tantara ny Andriana (‘the Histories of violence, force 17–18, 66, 125, 128,
Kings’) 50, 52, 55–6, 61, 63, 65 155–6, 162–6
Teotihuacan 120 Visicato, Giuseppe 151
thetes 80 Viveiros De Castro, Eduardo 161–3
Tigris River 111 van der Vliet, Edward 76
Tikal 114, 118, 122–3 Vogt, Evon 117
Tilly, Charles 17
Tlalocan mural 120 Wascar Inca 90
water management (Mayan) 115–25
Umma 138 Wayna Qhapac Inca 90
Ur (Tall al-Muqayyar) 127–8, 139, Weber, Max 17–18, 76
147–9 Winter, Irene 143
cemetery 129, 133, 136, 141, 148 Wittfogel, Karl 117
Ur, Royal Tombs of 7, 127–56 Wittgenstein, Ludwig 71
CT-scans 133 Woolley, Sir Leonard 127–33, 136–41
body preservation 133–4 Wright, Henry 63
jewellery 131–3
PG/580 129 yanacona 92
PG/789 131–4 Yoffee, Norman 11–12, 27–30, 45, 47
PG/800 130, 134 Yucatán 115
PG/1236 129
PG/1237 130, 132–4 Zanzibar 55
PG/1631 129 Zawiyat er-Rahman 24

Archaeology and State Theory Subjects An

Uploaded by

Copyright:

Available Formats

Archaeology and State Theory Subjects An

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Archaeology and State Theory Subjects An

Uploaded by

Copyright:

Available Formats

Archaeology and State Theory

Series editor: Richard Hodges

Against Cultural Property John Carman

Subjects and Objects of Power

50 Bedford Square 1385 Broadway

First published 2014

© 2014 Bruce Routledge

All rights reserved. No part of this publication may be reproduced or transmitted

No responsibility for loss caused to any individual or organisation acting on

British Library Cataloguing in Publication Data

ISBN: epdf: 978-1-47250-410-4

Library of Congress Cataloging-in-Publication Data

Typeset by Fakenham Prepress Solutions, Fakenham, Norfolk NR21 8NN

List of figures vii

3.1 Map of the Kingdom of Imerina 51

APAAA Archeological Papers of the American Anthropological

ARA Annual Review of Anthropology

CAJ Cambridge Archaeological Journal

CSSH Comparative Studies in Society and History

ETCSL Electronic Text Corpus of Sumerian Literature

JAA Journal of Anthropological Archaeology

JNES Journal of Near Eastern Studies

LAA Latin American Antiquity

SEH Social Evolution and History

We cannot speak of the State-thing as if it was a being developing

Today, most of us live in large-scale political communities that

conceptually difficult. In other words, it is something worth thinking

structure of governance or a structure of coercion, a social contract

comparatively and what are the political implications of engaging (or

engage with large-scale politics in the past is in fact closer to Foucault,

book as an extended argument, rather than as a general overview. It is

gender, production and material culture will be examined in relation

hierarchy. It is now commonly acknowledged that polities named as

abstract taxonomies, it fails to explain just how the state is formed as an

under which traits that improve the reproductive fitness of a group

Even scholars with impeccable post-processual credentials, engaged in

into premodern contexts extends and naturalises asymmetrical power

came to gain varying degrees of ascendancy over all other social

authority is not that a political apparatus does, or does not, ‘gain

in Smith’s terms. Warfare occurs between communities lacking any

De-centring the state

A starting point is to note that the modern state is one example of

… presents institutionalised power to us in a form that is at once

The state idea, therefore, attributes ‘… unity, morality and independence

modern capitalist societies. For Abrams the state idea is necessary

state, especially in a book that is ultimately concerned with premodern

Admittedly, for most monarchies even this level of abstraction

The subjects of the Pharaoh were also defined by material practices

frequently sanctioned between genders and ages or within kin groups

In this chapter I have argued that what we frequently call state-

Coercion and consent

The pack of scoundrels tumbling through the gate, emerges as the

Sovereignty is easy enough to imagine, but more difficult to imagine in

Order, legitimacy and wealth

One innovative attempt to come to terms with such issues is John

Order in this sense provided a single intellectual framework capable

cosmological order is rather helpful in understanding the continuity

the Shang and early Zhou periods. Evidence for osteo-pyromancy,

divination is perhaps acknowledged in the fourth century bc text Kuo