STATES OF IMAGINATION

Ethnographic Explorations of the Postcolonial State

Edited by Thomas Blom Hansen and Finn Stepputat

11111111101111111111

Duke UniversiryPress Durham and London zoos

 CONTENTS

Preface vi¡

Introduction: States ofImagination i

Thomas Blom Hansen and Finn Stepputat

1 STATE AND GOVERNANCE

"Demonic Societies": Liberalism, Biopolitics, and Sovereignty 41

Mitchell Dean

Governing Population : The Integrated Child Development

Services Program in India 65 a^^^ II
t/s en g ian^, Voa^tL.^m ^^in C1•/' 1^
Akhil Gupta

The Battlefield and the Prize: ANC's Bid to Reform the

South African State 97
Steffen tensen

Imagining the State as a Space: Territoriality and the

Formation of the State in Ecuador 123
D aoor Duke Universiry Press Sarah A. Radcliffe
All rights reserved
Printed in the United States ofAmerica on acid-free papero
Typeset in Quadraat by G&S Typeserters, inc. II STATE AND JUSIICE
Library ofCongress Cataloging-in-Publication Data appear
on the last printed page ofthis book. The South African Truth and Recdnciliation Commissidn:
A Technique of Nation-State Pormation 149.
Lars Buur
Recon s tructing National Identity and Renegotiati ng Memory:
PREFACE
The Work ofthc TRC 182
Alerta J. Norval

Rethinking Citizenship: Reforming thc Law in Postwar Guatemala 203

Rachel Sicder

Governance and State Mythologies in Mumbai 221

Thomas Blom Hansen

iii STATE AND COMMUNITY

Before History and Prior to Politics: Time, Space, and

Territory in the Modero Peruvian Nation-State 257
David Nugent
The ideas that led to the conception of chis volume grew out of the intellectual
Urbanizing the Countryside: Armed Conflict, State Formation,
environment o£ the research program Livelihood, Identity and Organization
and the Politics of Place in Contemporary Guatemala 284
in Situations of Instabiliry, which materialized thanks to the Danish Council
Fino Stepputat:
for Development Research, the Centre for Development Research in Copen-
In the Name of che State? Schools and Tcachers in an Andean Province 313 hagen, Institute ofAnthropology at Copenhagen University, and International
Fiona Wilson Development Smdies at Roskilde University. The LIVELY program brought
together researchers from a variery of academic fields in an effort to synthe-
The Captive State: Corruption, Intelligence Agencies, and
size ongoing theoretical and empirical work on violent conflict, migration,
Ethnicity in Pakistan 345
and popular culture. Because [hese subjects converged on che culture of states
Oskar Verkaaik
yet challenged the idea of the state as something given and immobile, we
Public Secrets, Conscious Amnesia, and the Celebration of conceived the idea of an international seminar on che States of Imagination
Autonomy for Ladakh 365 in che postcolonial world and invited a number of prominent and talented
Martijn van Beek scholars for the event.
The idea of States of Imagination was enthusiastically received, and the
Bibliography 391
seminar, which took place in Copenhagen in ehe midst ofwinter 1998, elicited

Abour the Contributors 415 a collection of fascinating papers, most of which appear in this volume. We
are grateful to the participants of chis seminar for contributing to an intellec-
Index 417 tually stimulating seminar and for challenging and reflning the common
approach that now emerges from the volume. We would also like to express
our appreciation of the support and inspiration we have received from our
colleagues in the LIVELY program: Karen Fog Olwig, Bodil Folke Frederik-
sen, Preben Kaarsholm, Henrik ROnsbo, Ninna Nyberg SOrensen, and Fiona
Wilson, as well as Peter Gibbon, Poul Engberg-Pedersen, Jan Ifversen, Karuti

vi CONTENTS
 Kanyinga, Thandika Mkandawire, Afonso Moreira, Kirsten Westergaard, and
other good colleagues.
INTRODUCTION States of Imagination
Finally, we would like to thank our secretary, Annette Smedegaard Chris-
tiansen, for working hard with us throughout these years, as well as the two Thomas Blom Hansen and Fino Step,putat
anonymous reviewers for Duke University Press for their encouraging and
constructive suggestions, which have improved the final product.

... as if every man should say to every man, 1 Authorise and give up my Right of
Governing my selfe, to this Man, or to this Assembly of men, on this condition, that
thou give up thy Right to him, and Authorise aH his Actions in tike manner.-Thomas
Hobbes, Leviathan

The state has once again emerged as a central roncero in the social sciences.
It has also been rediscovered by practitioners of development and powerful
international agencies such as the World Bank (1997), which now advocates
"good governance" by lean and effective structures of government. However,
in the vocabulary of World Bank economists the state and its instimtions re-
main strangely ahistorical entities, a set of functional imperatives of regula-
tion arising from society but devoid of distinct characters and different his-
torical trajectories. In this influential train of thought the state is always the
same, a universal function of governance. In the 1970s, theories of the capi-
talist state also privileged the state's functions in reproducing labor and con-
ditions for accumulation of capital over its forms and historicity. Also, when
Evans, Ruschemeyer, and Skocpol (1985) "brought the state back in" as an
actor in its own right, their conceptualization of state revolved around certain
assumed core functions and historical tasks that every state presumably had
to perform.
The current rethinking of the state occurs ata juncture where the very no-
tion of the state as a regulator of social life and a locos of territorial sover-
eignty and cultural Iegitimacy is facing unprecedented challenges. Ethnic
mobilization, separatist movements, globalization of capital and trade, and

Vi¡¡ PREFACE
 well as Foucauldian notions of governance through knowledge-practices and
inrcnsificd niovement of people as migrants and refugees al] tend tu under-
different governmentalities, that is, che forros of nrentalité suffvsing tech-
miue che sovereignry of state power, especially in the posrcolonial world.
niques of gouvernement have informed other careful enipirical studies of gov-
The equations among stare, economy, society, and nation thar eonsritutcd thc
ernment and stateness. Among there, rhe works of Timorhy Mitchell on
dominan[ idea of statencss in che nventieth centu ry have been undermined
Fgypr (1988) and Parcha Charterjee en India (cg., 1993) have gained wide
from below by growing demands for decentralization and autonomy, and
from aboye by the nnperatives uf supranacional coordination of monetarv. currency within anth ropology.
Thc contributions in this volume all share rhis dcnaruralizing approach to
environniental, and military policies in new configurations afrer che cold war.
stare and governance in the posrcolonial world; rhey all smdy che state, poli-
At rhe same ripie, tic discourse of rights and che proliferating demands for
ties, and notions of authority empirically froni a variery of ethnographic sites;
a variery, of entitlemcnts have expanded and transformed che meanings of
and they all position themselves in the space between a Gramscian and a
citizenship. The paradox seems tu be that while che authority of che state is
Foucauldian position on power, government, and authority where much of
constantly questioned and funetionally undermined, there are growing pres-
che reconceptualization of che stare has been raking place. This is, however,
sures en states to confer full-fledged rights and entitlements en ever more
a field fraught with tension and contradictions. In Gramsci's understand-
citizens, to confer recognition and visibility on ever more institutions, move-
ing, stare power emerged from che capacities, the will, and che resources of
ments, or organizations, and a growing demand en states from che so-called
classes, or segments thereof. This "will to class power" gave birth to projects
internacional community to address development problems effectively and ro
of political-cultural hegemony and strategies of social transformation aiming
promote a "human rights culture," as che lates[ buzzword goes. This paradox
at che consolidation of class domination. Gramsci did not merely see the stare
has to do with che persistente of the imagination of che state as an embodi-
as an executive of che bourgeoisie, as older Marxist theories had held, but
ment of sovereignry condensed in che covenant, as Hobbes saw iq as che
maintained che foundational role of class power becoming realized in the
representation of the volonté generale producing cicizens as well as subjecrs; as
forro of a stare: "The historical uniry of che ruling classes is realized in the
a source of social order and stabiliry; and as an agency capable of ereating a
State and their history is essentially che history of States . . . rhe subalrern
definite and authorized nation-space materialized in boundaries, infrastruc-
classes, by definition, are not unified and cannot unite until rhey are able to
mre, monumenrs, and authoritative instimtions, This myth of che state seems
become a `State'" (Gramsci 1971: 52). Gramsci rried in other words to de-
to persist in the Pace of everyday experiences of che often profoundly violen[
naturalize che stare by pointing tu its essentially political, and therefore un-
and ineffective practices of government or outright collapse of states. It per-
sists because che state, or in timbo ali ed s verei rvernmerft, remains stable, parcial and always violent character.
This line was also taken by subsequent Marxist and post-Marxist schol-
pivotal in our very imagination of what a society is. Whether we agree on
arship inspired by Althusser and Poulantzas, wherein che stare remained
what che stare means or not, "ir" is, nonetheless, central to all that is not state:
thoroughly "socialized" and epiphenomenal, that is, an expression of social
civil society, NGOS, che notion of a nacional economy, che market, and the
relations and ideological configurations and, hence, less interesting as a phe-
sense of an internacional community.
nomenon in its own right. Also, a range of attempts in the 1g8os to creare a
This parados of inadequacy and indispensabiliry has robbed the stare of as
"stare-centered" approach to che relationship between stare and society failed
naturalness and has enabled scholars from many disciplines to smdy state-
to escape a simplistic dichotomy berween "state" and "sociery." In most of
ness asa historica l and contingent construction. Following Philip Abrams's
[hese writings che stare remained, somewhat paradoxically, a rather unex-
(1988) importan[ unpacking of che stare in rheoretical terms and Corrigan
plored but unified "social actor" along with other ubiquitous and abstract
and Sayer's (1985) work on che state in Britain, a growing body of work has
social forces whose infernal relationship, as in Marxist scholarship, deter-
begun tu chart hisrorical trajectories of stare formation in various parís of
mined che shape and functions of instimrions and directions of policies (e.g.,
che world. Much of chis work has been inspired by Gramscian notions of
class power articulated through always fragile and contested hegemonies, as Evans et al. 1985; Migdal 1988).

INTRODUCTION 3
2 INTRODUCTION
 Even in Laclau and Mouffe's (1985) influemial poststrucmralist rethinking
somewhat broader perspective on the ambiguities of the state: as both illu-
of hegemony and politics beyond social determinism, the question of the
sory as well as a set of concrete institutions; as both distan[ and impersonal
state remained submerged in a wider category of "the political," now liber-
ideas as well as localized and personified instirutions; as both violen[ and
ated from the straitl'acket of essenrialist thinking but also far removed from
destructive as well as benevolent and productive. Modern forms of state are
empirical categories. In this perspective, the state, orjust institutions, remain
in a continuous process of construction, and Chis construction takes place
entirely political, that is, alterable and floating, and only appear as relatively
through invocation of a bundle of widespread and globalized registers of
stable "nodal points" in discursive formations (Laclau and Mouffe 1985: 112-
governance and authoriry, or, as we prefer to cafl it, "inguages of stateness."
13) oras relatively routinized forms of power that have become "sedimented"
The central proposition of [his volume is rhat ne
and stable as their political origins have been effaced (Laclau 1990: 34-35).
practices must Iscern and explore [hese different inguages, their localized
Foucault found issues of legitimacy and sovereignty less relevant. Instead
mean s eenealot;ies and ^n-ía rhey appear couched in mvthholo-
he explored how modernity was marked by the emergente of a broader field
gies of power, as practical, often nonpolitical routines or as violen[ imposi-
of government of conduct-of the self, of the family, of institutions, of the
tion is requires that one study how the state tries m make itself real and
'i body, and so on. Foucault famously remarked, "We need to cut off the King's
tangible through symbols, tests, and iconography, but also that one move
head: in political theory it still needs to be done" (1g8o: 121). In Foucault's
beyond the state's own prose, categories, and perspective and study how the
view, the intensified regulation of modern societies was not a resuft of the
state appears in everyday and localized forms: in brief, to study the state, or
penerration of the state as a center of power, but the other way around: the
discourses of the state, from "the field" in the sense of localized ethno-
modern state was an ensemble of institutional forms made possible because
graphic sites, whether "inside" or "outside" of the evanescent boundary be-
of the general ' governmentalization" of socieries, that is, the specific ways
rween sociery and the stare that usually crumbles when subjected to empirical
human practices became objects of knowledge, regulation, and discipline.
scrutiny.
In Chis view, the modern state is not the source of power but the effect of a
wider range of dispersed forms of disciplinary power that allow "the state"
to appear as a strucmre that stands apart from, and aboye, society (Mitchell Languages of Stateness
1999: 89).
In a recen[ article on the character of the modern state, Pierre Bourdieu out-
As has been remarked by many of Foucault's interpreters, one finds little
lines in his inimitable style the problem of studying the state as one of escap-
interest in the state or in politics in Foucault's writings (see, e.g., Hindess
ing the "thought of the state": "To have a chance to really think a state which
1996: 96-158; Ransom 1997: 101-53). Although frequentlyinvokedinstudies
still thinks itself through those who attempt to think it, then, iris imperative
of resistance, Foucault had very little to say about resistance as such beyond
to submit to radical questioning al] the presuppositions inscribed in the re-
mere reactions to new strategies of power, a kind of ubiquitous inertia he at
aliry to be thought and in the very thought of the analysr" (1999: 55). Bour-
one point likened to chemical processes (1982: 209). Instead, his interestwas
dieu remarks wryly that given the ease with which "social problematics"-as
rather consistendy in the conditions of possibilities of politics: how certain
rhey are diagnosed from the poro[ of view of states claiming to represen[
disciplinary forros, certain styles of knowledge and governmentalities made
society and the common good-are taken over by the social sciences and thus
specific policies plausible, specific forms of rationality thinkable, and forros
given the stamp of quasi-independent thinking, social scientiscs may in fact
of political discourse possible and intelligible.'
be singularly ill-equipped to mee[ this task.
Can [hese stances, harboring such different epistemological strategies, be
Bourdieu expands Weber's classic formutation and characterizes the for-
reconciled? The answer is that they obviously cannot be reconciled com-
mation of the modern state as a process of concentration whereby "it," the x
plerely, but also that they may not need to be. Our argument is that keeping
that is the state, acquires a monopoly of physical and symbolic violente over
[hese rwo perspectives in a productive tension with one another affords a
a territory and its popularion. The state condenses four types of "capital":

4 INTRODUCTION
INTRODUCTION 5
 ing about the state as an entiry that alwayslalready consists of certain fea-
violence, economic capital (tax and regulation), informational capital (eur-
ricula, validation of larowledge, etc.), and symbolic capital (juridical dis- tures, functions, and forros of governance, ler us approach eaeh actual state

course, nomination, validation, cte.). fogether they constituye, argues Bour- as a bistorieally specific configuration of a range of languages of stateness,

dicu, capital étntique. state capital, the (nteta)authority to validate or invalidatc some practical, others symbolic and performative, that have been dissemi-
nated, translated, interpreted, and combined in widely differing ways and se-
other forros of authority, that is, to have tire last word in a territory, to have
che last judgment ([999: 67). To maintain this supremo position as Che sum- quencesacrossthe globe.
Without pretending to be exhaustivo, we single out three practical lan-
mit of society, cach institucional fieid dial sees itself as a parí of the sute
musa devine elaborare institutional rites, schemes of classifiearions, hierar- guages of governance and threc symbolic languages of authoriry as particu-

chies of competenee, achievement and honor to retain order and a distante larly relevant for an ethnography of che state. The forme[ are (i) che assertion

between itself and "sociery" as well as other parts of che state. Bourdieu's of territorial sovereignty by the monopolization of violence by permanent and

concern is not so much whether or how the state governs but rather how the visible military and police forces; (z) the gathering and control of knowledge
of che population-its sine, occupations, production, and well-being-of
specific authority of the state, its stateness and its hegemonic loeation at Che
Chis territory; and (3) the generation of resources and ensuring the reproduc-
center of society, is (re)produced through symbols and rituals.
Although Bourdieu's outline of the symbolic registers of the state does not tion and well-being of the population: in brief, development and manage-

acknowledge its own mirroring of French étatisme, it does rather usefully re- ment of the " nacional economy."
These languages of governance, always underpinned by knowledge-
mind os that the disciplinary forros of power of the state constantly are en-
gaged in a perpetuated reproduction of rhe state, its institutions, its hierar- generating techniques, have historically been disseminated, exchanged, and

chies, its own languages and forms of identities produced and sanctioned by transplanted globally, including in the non-Western world. As we know, this

its procedures. The state not only strives to be a state for its citizen-subjects, has been a highly unequal exchange of technology, flowing mainly froto the

it also strives to be a state for itself and is expected by populations, politicians, colonial powers to che colonies, late[ from che so-called developed world-
"proper" languages of stateness in its practices capitalist and socialist-to so-called underdeveloped countries. Today, NGO5
and bureaucrats to employ5 1
and international aid agencies have emerged as major transmitters of new
and symbolic gestores. A ^vnvlU vi-G:
Asw
e tr to
administrative technologies in the field of development. This technology
y understand how státes r
onto
e porary Africa and Asia are
imagined and designed we are inadvertently thrown back en che historical 1 `^or^ ^r ) transfer and exchange has involved export of a range of techniques: how to
,
development of modern forms of governance and sovereignty in Western ckW •
o-° ^^^^^'' .set up secret services and military logistics, budgetary models and taxation
" biopolitical governance such as
Europe. In che eyes of politicians , rebels, planners , and social scientists, the systems , entire packages of "high-intensity
the-art systems monitoring deforestation, participatory local develop-
history of European state formation continues tu provide powerful images of state-of-
what a proper state should be. As Crawford Young suggests , " Both colonial- ment projects empowering women, and the seructure and procedures adopted
ism and resistance to it yielded diffusion of a notion of stateness whose lin- by commissions trying to produce che truth about past regimes or arrocines.
to come The larger and more imprecise imaginarion of the state as an expression
cage lay in che European core" (1 994: r6 ) It is important, therefore ,
to grips with the historical specificities and contingencies that shaped that of effective territorial sovereignry and authoriry capable of protecting and
historical experience , as Mitchell Dean points out in his discussion of Fou- nurturing population a conomy became from the 194os on the dominant
cault's work en sovereignry and biopolities in this volume . global form of politic 1 community . Expressed in rhe programs and rhetoric
Instead of seeing state formation in the postcolonial world as a flawed of nationalist movements in the colonial and posteolonial world, authorized
the principies of the Non - Aligned Movement and count-
imitarion of a mature Western forro , we need to disaggregate and hismricize by the UN Charter ,
a legitimate represen-
how the idea of che modern state became universalized and how modern less other documents , the nation-state is (or should be)
_ The production of states as not
forms of governance have proliferated throughout che world. Instead of talk- cation of che will and interests of its citizens
j

INT RO D UCTION 7
6 INTRODUCTION
,
1i -1 ^i! u Cz d.) IAL n
 only loci of governance but centers of authoritative power has usually taken
che state or the government circulated? How do [hese genres relate to more
place through deployment of three symbolic languages of authority : (
t) the elaborate languages of political contention and the style in which state and
institutionalization of law and legal discourse as the authoritative language
governmental authority is imagined ? These are some of che questions we
of the state and the medium through which che state acquires discursive pres-
ente and authority lo authorize ; ( suggest could be asked and explored anew . As shown by several of che con-
z) the materialization of the state in series
tributions in Chis volume , stateness does not merely grow out of ofhcial, or
of permanent signs and rituals: buildings, monuments ,
letterheads, uni- "stately," strategies of government and representation. The attribution of
forms, road signs, fences ; and (3)
the nationalization of che territory and che
stateness to various forms of authority also emerges from intense and often
institutions of che state through inscription of a history and a shared com-
localized political struggles over resources , recognition , inclusion , and influ-
munity on landscapes and cultural practices. Í
ence. Whereas certain forms of state intervention may be loathed and re-
The first three are technical languages, the Foucauldian aspect, one may
say, of practical governance , discipline , sisted , other and more egalitarian forms of governance , or more benign
and productive biopolitical gover- forms of authority, may at the same time be intensely desired and asked for.
nance; the latter three are symbolic languages aiming at reproducing che
Everyday forms of state power, in other words, are always suffused with and
imaginafion of che state as that specific authoritative center of a society in
mediated by politics: contestation of authority , open defiance, as well as at-
principie capable of issuing what Bourdieu calls the "las[ judgmenc" None
tempts to divert or privatize resources.
of these languages necessarily goes together or presupposes each other; each
has distinct historical trajectories , The centrality of the state to virtually every modern notion of a society thus
meanings, and degrees of sophistication
in every case and locality . The essential thing is however means that che exploration of forms of state, stateness , and government in-
, chal a state exists
only when [hese " advert endy traverses a deeply normative ground. The state and modern gov-
lana f gove na d authority combine and co-
extst one way or che other _The decisive step in the invention of che modern ernance is not something one can be for or against as such , for che simple
reason that we cannot escape it . One can and should criticize specific forms
nation-state was exactly when che sovereign state became entrusted with ex-
of governance , undesirable institutions , and oppressive state practices, and
panding tasks of managing che social and economic well-being of its people,
to protect, reproduce , and educate in; citizens many contributors in this volume do so. Implied in such critiques are not
, to represent che nation, its
history, and its culture ( visions of the absence of government or che state as such, but rather the
s), and to reproduce boundaries and institutions en-
possibility of other, more humane and democratic forms of governance.
abling che political community lo be recognized by other states as a proper
state. While not ignoring Chal state power harbors the potencial of unprecedented
brutality in che name of a chilling dehumanized and scientistic utopia, as
Exploring che state through ethnographies thus raises the question of the
limits of government : shown by Scott ( 1998 ), we advocate a more disaggregated and less essential-
Where does che state begn and end? What is che speci-
izing smdy of che state by foregrounding che local , che emic, and che vernacu-
ficity of the state as opposed to other forms of authority and governance that
exist alongside it - in communities , lar notions of governance , state authority, and resistance to state power. In-
within enterprises , in localities, and in
stead of deploring che crisis or even collapse of postcolonial states in terms
families? Standard governmental practices in general are not considered to be
part of the political sphere . of che repercussions for regional stability (see, e.g., Zartman 1994 ), we find
They may be everyday routine government actions
it more pertinent to explore che local and historically embedded ideas of
such as censos taking, primary health care programs ,
road construction, al- normality, order, intelligible authority, and other languages of stateness. The
phabetization programs -
all routine practices undertaken by inconspicuous
government employees . constan [ recurrence of notions of stateness as a guarantee of order and or-
The fact chal these routines are considered outside
dinary life , shown by several contributors in Chis volume in various parts of
che domain of political contention and its variable languages ,
however, is che world, is thus nota barrier co critical engagement with che phenomenon
of importante. In what ways do people taik about and act en [hese forms of
government practices ? of che state, but its most fundamental condition.
Through what genres are narratives and knowledge of
The contributions in Chis book are organized according to che three dimen-

8 INTRODUCTION
INTRODUCTION 9
 graphical imagination.'l'hrough elaborare cartographies and education, the
sions of rhe state outlined aboye: first, as technologies of governance encouo- space of rhe state was domesticated as the proper place of rhe nation. This
tered in rhe guise of classifications, forros to he filled in, roles to be obeyed, was rhe sparial matrix within which local eommunities hence could be in-
epistemologies learned, and so on: second, symbolic representations of the
scribed, fixed,and ranked.
suite as a locus and arbiter oí justice and a symbol of a largor society: thid, Other important inrerventions aiming at producing "normal" srates were
the invocations of rhe state as a ser of institutions that can recognize, adju-
performed by development agencies, international donors, and rhe thrust oí
dicare, and authorize, thar is, invest its authority in and give legitimacy ro development thcory thar all supported the vicw of rhe state as an "agent of
cerrain representatives, fonns of communiry, public symbols, and also be-
modernization," an island of modernity and rationality, a part of the so-called
come loci of resistance and conrestation. In thc remainder oí this introduc- modero sector, and so on. This coneeption obviously disregarded rhe fact that
don we unpaek these aspects of state and polities and elaborare a bit on how most colonial administrations were designed tu exercise forros ofgovernance
each of rhe contributions in this volume feeds into rhem. and control of populations and territory that often were crudely extractive and
much less fine-grained and less intensive than in the European homelands.
As Mahmood Mamdani (1996a) has pointed out forcefully in the context
Colonialism , Modernity, and Governance
of Africa, colonial administration often relied heavily en indirect rule, on a
This volume grows out of an engagement with a range of postcolonial expe-
somewhat random brutality, and en local notables ro whom many details of
riences of government , authority, and notions of rights in Africa, Asia, and governance and tax collection were entrusted. As colonial administrations
Latin America . The central question is , of course , to what extent [hese expe- were turned roto the backbone of rhe new postcolonial nation-states, their
riences can be understood and interpreted through theoretical lenses that rely, excessive centralization and bifurcation in rural and urban segments, their
heavily en the historical traje r I of state formarion in Western arome As habits of summary governance at a distance, their lack of independent judi-
we suggested aboye, the more productive answer to [his question is, in out
ciaries, and rhe heavy-handed techniques deployed to control rhe majoriry of
view, lo abandon totalizing and culturalist notions of certain enduring "East-
their populations were suddenly diagnosed as developmental flaws, as lack of
ern," "African ," or "Western" forros of state and instead disaggregate and
modernity, as "weak states."
trace how various languages of stateness , not necessarily all purely Western Reflecting on rhe "flawed" states in rhe non-Western world, Samuel Hun-
in origin, have been spread , combined, and vernacularized in various parts of tington opened his controversia) book with rhe sentence "The most impon
the world. rant distinction among countries concerns not their forro of government but
At the same time, it is also pertinent to remember that the Western imagi-
their degree of government" (1968: 1). In his text Huntington acknowledged
nation of the state , however rraversed by myths and historical fiction, remains rhe capaciry of communist movements to transform states into effective ve-
rhe globally most powerful idea of political order in rhe twentieth century,
hicles of governance, and he recommended, infamously, rhat rhe Western
institutionalized in the international state system after 1945 . The most central world should realizo that states governed by so-called praetorian regimes
presupposition underlying this system is that all states in principie are, or will
were more likely to creare order, stable institutions, and economic growth
become, similar , or at least mutually intelligible , in their structures and in rhe [han democratic regimes likely to be overwhelmed and destabilized by rhe
rationalities governing their actions . Such an ahistorical understanding of overload of expectations from a wide array of interest groups. Huntington
the state was eagerly embraced by the nationalist political elites in rhe post- N,.e3 v Qtc^
was fascinated by rhe ability of strong governments ro "normalize" rhe state
colonial world , anxious to transform their states into "normal " nation-states . apparatuses of developing societies, that is, to discipline them, extend rhem,
That task entailed, among other things, that rhe state was represented effec- make rhem capable of effective penetration of evermore social and economic
tively to its citizens and eommunities and that it manifested itself effectively
relations.
on its territory. As pointed out by Sarah Radin her piece on Ecuador, Turning che state roto an autonomous actor capable of swift social reforni
many states approached Chis task through a systematic production of a geo-,

INTRODUCTION II
IO INTRODUCTION
 was, however, an agenda and a desire shared broadly across the political spec-
trum both within and outside the postcolonial world . Radical regimes in Af- the state genus occurs as a process paralleling the development of the modem
state" ( 44). This means, argues Young, that out ideas of what a colonial state
rica, Asia, and the Middle East swept to power promising rapid moderniza.-
tion and strict social discipline , often biending elements of Soviet planning, was, did , and wanted to achieve have to be historicized and seen in the context
militarization , and notions of secular modernity taken from Atatürk and Nas- of the wider development of governmental technologies and political imagi-
naries at their time . There were , for instante ,
ser. In Latin America dreams of using the state, and sometimes the military, enormous differences among
in a rapid and pervasive drive for development , modernity, and recognition the maritime Portuguese empire bent on the "revenue imperative" (52), the
were nurtured by young radicals, bureaucrats , and officers . In the burgeoning Spanish empire organized around extension of regal power and the authority
development industry similar ideas of transformation through stronger and of the Church to new territories , and the austere mercantile rationalities gov-
more effective governance were prominent , not so much aiming at national erning the early British and Dutch empires . It was, Young reminds us ,
only in
grandeur and recognition , but driven by desires to impart economic devel- the nineteenth century that the European powers began the systematic con-
opment, political stability, and techniques of poverty eradication to the post- struction of specific institutions aimed at governing the colonial populations
colonial world . Much of this developmental desire meant that profoundly and territories.

political issues of social transformation and institutional designs were de- One corollary of chis observation is that colonialism in Latin America liad
politicized and put in the hands of developmental agencies and experts who a complexion completely different from the high - noon imperialism that hit
transformed them into technocratic designs and further removed govern- the African continent in the r88os. The states of Latin America can hardly be
ance from the realm of the everyday , the vernacular, and the intelligible (Fer- regarded as postcolonial states in the lame sense as their counterparts in
guson 1990). Africa. T11e rationalities governing state practices in Latin America developed
alongside [ hose of Europa and North America although in a mimetic cela
As we know by now, many of [hese attempts to strengthen the state turned
into outright authoritarian regimes that promoted the interests of narrow tionship, as Michael Taussig ( 1997 ) has pointed out. por Taussig ,
the efforts
elites and, in effect, eroded the instimtions and authority of the state. The at creating illusions of proper states in "the European Elsewhere" remain
travestic, shot through with utopias and an (often absurd
existence of oligarchic structures of power, the organizafion of entrenched ) zeal in the face of
class interests , and the production of new, self- a colonial history that refuses to support any narrative of autochthonous au-
interested bureaucratic elites
were obviously central to this development . Another problem was that the thority (57-61 ). The desire to modernize ,
the eager embrace of cientismo and
very form of the apparatuses employed to effect this grand transformation of rational governance among national and local elites in large parts of Latin
America were fueled by the circulation of the languages of stateness men-
the postcolonial societies bore the indelible mark of colonial designs just as
the inventory of techniques employed in [hese tasks were steeped in colonial tioned aboye , that is, new techniques of control and knowledge through
which societies , communities, and salves could be improved
notions of control, policing, and summary governance of communities rather , governed, and
than citizens. appreciated.' David Nugent' s contribution to this volume gives an arresting
Colonial states, however , were never full-fledged states, Crawford Young WL n^ ¡Ilustrador) of how the desire to become modern also was a powerful impulse
(1994) reminds us . They had no sovereignty, in pohncal mobilizations in Pero in the first half of che twenáeth rento ry. Io
no autonomy, no embedding in
society and remained appendages te powerful European military and admin- 0x a different and yet related context, Tiro Mitchell ( 1988 ) has shown how nine-
istrative complexes . The incompleteness and abnormality of the colonial state teenth - century Egypt was (self- ) colonizad with
out ovar[ colonialism through
was in fact one of the central criticisms waged against imperial rule by na- N che mternalization of scientific genres of knowledge , modern methods of
tionalists , from the early " Creole nationalists" around Bolivar, to the founders administration and surveillance , and styles of cultural self- objectification

of the Indian National Congress and nationalists throughout Africa. Young through European registers by Egyptian intellectuals and administrators.
argues that " the emergente of the colonial polity as a distinctive species of But the feasibility of governmental technologies was always constrained
by the location of a state in the wider international economy and by the re-

12 INTRODUCTION

. INTRODUCTION 13
 of the state" that Ernst Cassirer (1946) saw as a malign product of fascist and
sources and revenues at irs disposal in the domestic economy. As Fernando
organicist ideology is, we would argue, absolutely crucial to the organization
Coronil reminds as in his study of the formation of rhe Venezuelan oil econ-
and the experience of coherence and order of modero societies in most parts
omy, "ILi capitalist societies, command over persons rests in the hands of the
of the world. The estire idea of political legitimaey, of the difference hetween
state, while command over resources lies in the hands of capital" (1997: 64).
naked power and^auth-ority thc idea that 'the Law" is something rhat stands
His swdy forcefully demonstrates how the modern Venezuelan state was re-
above the contingencies of everyday life and incarnates a certain collecrive
strucmred and reimagined as the country emerged as a leading oil producer
justice, the crucial diseourse of rights as something that once defined and
in [he world bur also a captive in the larger global economy, completely de-
authorized become unassailable and inalienable: all hinge on the perperuated
pendent on foreign technology and global oil prices. Not only were state poli-
introduced inm myth of the state's coherence and abiliry to stand "above society," as it were.
cies reconfigured around this abundant resource, oil was also
Since Hobbes theorized the absolutist state, European notions of political
was reconstructed as a mod-
the dorninant political imagination as Venezuela power and the state have undoubtedly been starkly reductionist. To para-
s-oil) and a timeless
em "oil nation ": the sublated uniry of a natural body (
phrase Hobbes: "Covenants without the sword are but words," and at the
collective body of the people, the nation (67-117).
basis of the state, of power, of legitimacy, we find, purely and simply, vio-
It is these actually existing forms of governance and the trajectories of in-
lence. In this view, royal pomp, state rituals, and modero ideological forma-
stitutions and representations of the state in various parts of the postcolonial
the usual nega- tion essentially serve to efface and occlude Chis foundational violente, which
world that this volume explores. Throughout we try ro avoid is the origin of a state. Clifford Geertz has called this "the great simple that
deinstitutional-
tive prefixes ( weak, disorganized , incoherent , illegitimate, remains through all sophistications . . . politics, finally, is abour mastery:
ized, etc.) that still enframe the problematics and the puzzles to be solved in Women and Horses, Power and War" (1980: 134). This, argues Geertz, has
most political science and development studies literature en the postcolonial
led to an unfortunate blindness toward the importante of symbols and ideas
state. One of the most promising avenues away from this deadlock is to dis- in their own right to statecraft and state power. Geertz retrieves the impor-
aggregate the state into the multitude of discrete operations , procedures, and
people. By tance of ibis in his study of the classical Balinese theater state, the negara, a
representations in which it appears in the everyday life of ordinary poliry whose basis of sovereignry was its status as "an exemplary center-a
treating the state as a dispersed ensemble of institutional practices and tech-
microcosm of the supernatural order." Pomp, ritual, and spectacle were not
niques of governance we can also produce multiple ethnographic sites from
devices to represent the state or occlude its violent nature; they constituted
where the state can be studied and comprehended in terms of its effects, as
the core of the state that was based en the "controlling idea that by providing
well as in terms of the processes that shape bureaucratic routines and the
a model ... a faultless image of civilized existente, the court shapes the world
designs of policies.
around it" (13). The dramas of that poliry were neither lies nor illusions,
concludes Geertz: "They were what there was" (136).
State, Violente, and Justice: Between Imaginary The negara should remind us that there are languages of stateness other
and Apparatus than those invented in Europe in the past two centuries but also that the
rationality of intent, purpose, and action often imputed to modern states-
However analytically useful it is to denaturalize the state and to get beyond
by the analyst as well as the citizen-subject-tends to occlude the important
the state's own prose and problematics of social order and development, we
mythical dimensions of the modern state. Maybe it is the very idea of state
should not forgct thar the notion of the state probably remains the most pow-
aetions as guided by an abstract, omniscient, and rational intelligence-
erful lens through which society, nation, and even the ubiquitous but elusive
ceaselessly celebrated and vilified in novels and films on spies and intelligence
notion of "the economy" is imagined. The modern state is not just a set of
agencies-that constitutes the very core of the myth of the modern state.
rationalities or institutional forros. It has also acquired vital mythological di-
The widespread idea of the state as a thing is indeed at odds with basic
mensions that give as authoriry both historical aura and weight. The "myth

I NTRODUCTION 15
14 INTRODUCTION
 tendencies in how states develop. As modero forms of governmentality pene- reaucracy and among political figures as the state's own myth of itself and is
trate and shape human life in unprecedented ways, the practices and sites of
constantly enacted through grand state spectacles, stamps, architecture, hi-
governance have also become ever more dispersed, diversified, and fraught
erarchies of rank, systems of etiquette, and procedures within the vast ex-
with interna) inconsistencies and contradictions. This has not necessarily
panse of the bureaucracy. But do these elaborate state rituals actually manage
weakened the state in terms of the capacity of policies and designs to create
to create or reproduce a state mythology coherent enough for the state to
social effects. The strength of the modero state seems, on the contrary, to
impose itself on populations with effective authority? Or are these spectacles
be its dispersion and ubiquity. The modern states of, say, Western Europe
and rituals of the state more for interna ) consumption among bureaucrats,
are today more diverse, more imprecise in their boundaries vis-á-vis other
clerks, accountants, officers-a dailg routinized reassurance of the impor-
forms of organization, more privatized or semiprivatized than ever before,
tance and power of the state that actually serves to strengthen the sense of a
more integrated in supranational structures and yet apparently stronger than
unified stateness of dispersed forms of government?
ever before. The new role of the state is, Helmut Wilke (Igqz) has argued, to
Drawing on Foucault's insights regarding the specifically modern reorga-
supervise governance by semiprivate organizations, local authorities, self-
nization of space and time into routinized, repetitive, and internalized disci-
governing bodies of all kinds, NGOS, and so on, rather than to actually govern
plines and forms of surveillance, Mitchell (1999) argues that the "appear-
directly.
ance of structures" on the basis of these micro-operations seems to be one
The neoliberal attempts to restructure and trim the apparatuses of post-
of the most fundamental features of modernity in general and the preemi-
colonial states, originally designed for "low-intensity governance," along
nent feature of the "state effect" that modem governmentality produces. The
similar unes, however, have rarely produced a similar fiexibility and enhanced
state is the "abstraction of political practices" analogous to the way capital
capacity. The predominant organization of postcolonial governance as "com-
is the abstraction appearing from labor: "We must analyze the state ... not
mand policies" have meant that IMF-prescribed delegation of powers to the as an actual structure, but as the powerful, apparently metaphysical effect
local level of the state more often than not has produced deep fragmentation,
of practices" (89). The multiple practices involved in policing and control-
lack of coordination, and an undermining of the notion of the state as a
ling territorial boundaries is that which creates the nation-state as effect,3
guarantee of social order. por subject populations or citizens, the experience
just as the technicality of the legal process is what (re)produces the notion of
of the state has, in many cases (e.g., in the postcommunist world), changed the Law.
from a frightening Kafkaesque labyrinth of impersonal power finto the ran-
The question is how these insights can be given historical substance and
dom brutality of a state parceled into smaller fiefdoms run by local bureau-
differentiality and how we can create ethnographic sites from where such
crats and police officers. In the more extreme cases of state collapse, as wit-
"state abstractions" can be studied. One obvious, if very underdeveloped type
nessed in western and central Africa in the Iggos, state administration ceases
of study is that of the bureaucracy itselfl its routines, its personnel and their
to be a factor in everyday lile, which is thrown back in an almost Hobbesian interna ] cultures, gestures, and codes, its mode of actual production of au-
state where raw military might emerge as the ultimate basis of legitimacy.
thority and effects by the drafting of documents, uses of linguistic gentes,
Even then, amid chaos and bloodshed, some warlords attempt to create iones
and so en: in brief, an anthropology of the policy process that looks at it as
of stability and to erect something resembling a state: taxation instead of
ritual and as production of meaning rather than production of effective poli-
random plunder, dispensation of 'justice" through courtlike ritual instead of
cies per se. Michael Herzfeld's (Igqz) work on the symbolic registers devel-
instant killings, territorial control, and, in some cases, appeal to subjects in
oped by Western bureaucracy has laid out valuable conceptual groundwork
the narre of a shared community or destiny.
for such studies but has not been followed by thorough ethnographic work
There is little doubt that a mythology of the coherence, knowledge, and
that could demonstrate its wider relevance. Herzfeld's own text manages to
rationalities of the (ideal) state exists, thrives, and empowers many otherwise
produce only brief but highly interesting ethnographic illustrations of the
widely discrepant practices. This myth is carefully cultivated inside the bu-
imbrications of bureaucratic categories and idioms and everyday lile, mainly

16 INTRODUCTION
INTRODUCTION 17
 cal leaders and agencies of che state. These gentes represent widcspread,
from contemporary Greece. Lars Buur's analysis in this volume of the minu-
popular, and highly interesring poinrs of access to how che state is mytholo-
n ae of dic everyday workings of che South African Truth and Reconciliatiou
gized, externalized, and abstracted from ordinary existente yer is bclieved ro
Conimission represcnts another example of how anrhropologieal sensibili-
be omnipresent.
ties applicd to the routines, riruals, and underlying assumptions of legal- Achille Mbembe's (rggz) well-known and controversia) depiction of rhe
administrative practices can yield valuable inssights. obscenity and absurdity connected with public exereise of state power in rhe
'1 he discourse of rights that has assumed such critical importante in inter-
"postcolony," in casu Cameroon, points to che importante of the state as
nacional politics and has given risa to a whole new human rights sector within
an enrió manifesting itself in spectacles. To Mbembe, che impulse of state
che development industry is also centrally linked ro rhe myth of the state. As
power iu rhe postcolonial world, due to as origin in a system of excessively
nmch as the Law as a concept depends on the state's mythical qualities, che
violent, colonial power, is organized around an equally excessive fantasy of
practices ofsolemnly encoding certain rights in constitutions, of entrenching center of
making the imperatives of che state, mmmandement , the hegemonic
and interpreting these rights in judicial practices and invoking them in polit-
sociery. But chis is impossible because "the postcolony is chaotically plural-
¡cal rhetoric also hinges en the efficiency of the imagination of the state as a
istic and ... it is in praetice impossible to creare a single, permanently srable
guarantor of (hese rights. If that imagination is ineffective, the discourse of
system out of all che signs , images and markers eurrenr in che postcolony ...
rights is inconsequential. a hollow pretence , a regime
Chis is why, roo, the postcolony is, par excellence ,
Even more important, if the legal apparatus of che state is unable to trans-
of unreality ( régíme du símulacre)" (8).
late [hese abstract rights finto actual and effective judgments that prescribe
Mbembe Cakes popular forros of ridiculing of power, jokes of a sexual na-
precise entitlements that can affect people's lives or empower them, che ap- so en, as
ture about che roen in power , their bodily functions , appetite, and
peal to a larger notion of rights makes no sense. Instead, local communities is dealt with . In itself, there
examples of how this hollowness of state power
engage in small and inconspicuous acquisitions of entidements-to land,
is nothing African or even postcolonial about such joking about rhe elevated
customs, employment, and so en. Rachel Sieder's contribution analyzes the
and yer profane representations of power.° But che specifically postcolonial
protracted efforts in Guatemala to develop new and more inclusive forros of
feature, Mbembe suggests , is che way che state is excessively fetishized in
citizenship in a sociery characterized by deep and enduring differences be-
pomp, ritual, and entertainment , and che way [hese spectacles are disarmed
tween an elite "awning" the state and che large Indian communities. In
but also domesticated through jokes and humor to an extent that there is a
Sieder's analysis it is apparent that encoding rights in legislation is not a
coexistente or conviviality berween che official and the everyday world: "In
question of handing down entidements to grateful subjects, but a complex , non che least of which
fact officialdom and che people have many referentes
negoriation of how existing entidements stipulated in customary law and lo-
is a certain conception of the aesthetics and che stylistics of power ... it must
calized setrlements can be codified and inscribed in more durable and ab-
furnish public proof of its prestige by a sumptuous presentation of status" (9).
stract rights. Thus, che case illustrates how one of che most powerful lan-
Mbembe's argument is that to be legitimare , power must be represented
guages of stateness-the codification of social relations in law-works in
within already established registers of pomp. The holders of power must
processes of state formation at che turra of che century. present themselves as firm but also generous and endowed with an excessive
Another way of studying che myth of che state is te regard it as a form of argument that African poli-
appetite . This resonates with I. F. Bayart's (1991 )
"social fantasy" eireularing among citizens and communities. This fantasy
tics remain ognized around specific discursiva registers that often predate
is produced and reproduced by numerous encounters, everyday forms of de- understood both as a practical
colonialism such as he "politics ofthehelly "
fiance and obediente, ranging from fantasies of the mighty and evil state
nnlitics of feeding populations and a symbolics _pf power around metaphors
hatching hyperrational designs (a gente popular among radical groups on the
of eating and digesting . Mbembe uses che execution of two prisoners in Cam-
left as well as en che far right) to popular gentes of conspiracy theories that
ereen and che cheering crowds gathered to witness it as an example of chis
often impute almost superhuman omniscience and omnipresente to politi-

INTRODUCTION 19

18 INTRODUCTION
 specific mode of postcolonial power that thrives on a certain measure of com-
tact with the ancestral spirits, gradually shifted their loyalty to the liberation
plicity and involvement of broader sections of the population through enjoy-
movement , which had a decisive effect en the course of the war . As indepen-
ment. Unlike the excruciating torture and execution of Damiens for regicide
dence was announced on the newly renamed Zimbabwe Broadcasting Cor-
that Foucault made famous in the opening pages of Discipline and Punish, the
poration , it was followed by a ZANU song celebrating the spirit of Grand-
execution in Cameroon was organized more like a theater celebrating the
mother Nehanda , who had been in the first Chimurenga (in 1896) and in the
splendor of the state . To Mbembe, the grotesque (and ineffective ) character
second as well ( 217-18 ). In the following months and years , this "national-
of power in the postcolony is revealed in its lack of seriousness , its indul-
ization of the mhondoros " continued , and the picture of Nehanda was always
gente, its theatrical and obscene nature, and its successful involvementofthe
placed aboye that of Mugabe at oficial functions . It appears that some of the
population in "cheap imitations of power so as to reproduce its epistemol-
spirit mediums later shifted their loyalties away from the ruling party, but
ogy" (1992: 29).
nationalization was in a sense complete when the new state authorized the
Taussig has also recently explored how the idea of the state is fetishized
association of tradicional healers to practice as doctors , and when spirit me-
through a range of magical transactions and spectacles , from spirit posses-
diums were given a special license under this association and the right to use
sions to oficial textbooks and monuments , in a Latin American country. Not
the letters "sM" (!) in advertising and oficial communication (219-20).
unlike Mbembe 's work, a sense of the absurd and surreal sufTuses the repre-
These and other works Nave opened a field that approaches the construc-
sentations of power that ceaselessly revolve around stories and images of the
"Liberator " (Bolivar), his court , his black general Paez, later elevated to the tion of the state in everyday life, although they remain within convencional
areas of anthropological research : magic, spirits , the body. Severa¡ of the con-
status of El Negro Primero, a figure that connotes a primitivity and virile
tributions in this volume attempt to approach more routinized and less dra-
power of the plains , both loathed and desired by the urban elites (Taussig
matic forms of folk theories of the state and of political authority . Fiona Wil-
1997: 94-95 ); the death of Bolivar, his second funeral en South American
son's analysis in chis volume of the narrative of a rural schoolteacher in Peru
soil as the founding moment of the state , and the rumors that his heart had
demonstrates rather strikingly how notions of the proper nation-to-be, of
been removed "to live en in every South American " ( 104). The story about
modernity, of the ideal Peruvian peasant exist in forms that in many ways are
how a commander of the M-19 guerrillas in 1974, Alvaro Fayad , stole the
separated from the actually existing state in the arca. Oskar Verkaaik also
Liberator' s sword, the ultimate fetish of the state, from a museum is particu-
illustrates the pertinence of rumors of intelligence , informers , and the sup-
larly arresting. The communist guerrillas were at first indifferent to such sym-
posed "capture " of the state by ethnic conspiracies in contemporary Pakistan.
bols, but once in their possession they, too, fetishized the "thing," wrapped
However, in many cases the myth of the state is actually sustained by the
it in multiple layers of cloth and plastic so that it literally greca and grew to
rather mundane practices of authorization and recognition carried out by the
such a size that it stuck out of the trunk of a car and had to have a small red
state: the acts of authorizing marriages and registration of deaths and births,
flag tied to it in traffic ! This state " thing" disappeared, and rumor has it that
the recognition of deputations or representatives of communities or inter-
it was given to Castro ( 1go-94). Taussig's point is close to Mbembe 's: state
ests as legitimate and reasonable and thus entitled to be consulted in policy
power is fetishized through displays and spectacles but becomes effective as
matters, the certification by the state of institutions , professions , exams,
authority only because it invades , and is appropriated by, everyday episte-
standards , and so on. Such practices reproduce the myth of the state by lit-
mologies of power, of the magical, the spiritual, and the extraordinary.
erally implanting it in people's lives, as revered documents carefully stored or
In his work en the role of spirit mediums during the guerrilla war in Zim-
proudly displayed on the walls , as stamps, permits , titles from where certain
babwe, David Lan (1985 ) also points to the crucial role of spirit mediums,
i entitlements , social status, and respect flow. The upholding of a certain im-
mhondoros , in legitimizing the struggle of the Zimbabwe African National
age of the state as a revered object of respect and authority is often vital to
Union (zANU-va) in what became known as "the second Chimurenga," the
the status , livelihood , and identiry of millions of people. Nowhere is the im-
second war of liberation . Lan shows how the spirit mediums, being in con-
portance and dependence on the mythological dimensions of the state more

20 INTRODUCTION
INTRODUCTION 21
 something that can be "conquered" or "eaptured" rhrough polities. In Kara-
pathetically represented than in Che profound disillusion one can read in the
chi and Hyderabad, rumors depict Che Pakistani state as eaptured by Punjabis
rugged faces of elderly Russians, desperately displaying their medals and dis-
and feed into the sense of displacement and loss so pivotal m Muhajir iden-
tinctions trom rhe Soviet era in rhe vain hopo of extracting a minimum of
tiry, as shown by Verkaaik. In his essay on legal inquest and policing in con-
respecr when receiving their pensions that now are reduced to mere crumbs.
temporary Mumbai, Thomas Blom liansen shows how police ofHcers and
A great many of the institutions that governed the everyday life of Soviet citi-
social workers share the conviction that the "politicization of the state" con-
zens are still in place and many of the routines are unchanged, but Che power
stituyes an obstacle to effective and rational governance. This diagnosis of
of the mvth of the state has vanished. competitive polities as Che very source of Che decay, corruprion, and weaken-
In her thoughtful piece, Alerta Norval deals with another instance of the
ing of the state is shared widely by bureauerats, development workers, social
vanishing of one rype of state, the South African apartheid state, and the
scientists, journalists, and certainly millions of ordinary people in contem-
imaginative attempt to provide the new order, the New South Africa, with a
porary India.
new authorized national history through the narrative built by the proceed-
A different way of exploring the relations between polities and the state is
ings of the Truth and Reconciliation Commission. This is a history of evil and
in the context of radical political changes when new governments elaborate
its exorcising, of forgiving and overcoming, but it is also, crucially, an attempt
strategies and policies for profound reforms of the state, in this volume rep-
tu reduce a more fundamental sense of undecidability and to conceal the im-
resented by the essays en the current South African transition. Within polit-
possibility of complete reconciliation. ical science Chis theme is often discussed under the rubric of regime or form
of regime, one of three dimensions of governance: state (the notion of the
State, Community, Hegemony, and the Art of Politics state as legal and military structures of considerable permanence), govern-
ment (the wider instirutional structures and administrative procedures), and
A number of contributions m this volume deal with the relationship between
regime (the political organization and will in power). Although this is a nec-
the state and "its other," rhe social identities, practices, and allegiances that
essary unpacking of the term state, it is clearly not radical enough to allow
states "are formed against," te use Corrigan and Sayer's expression (1985: 7).*
for any ethnographic exploration of the state and state-community relations.
The notion of community is often used to depict the other of the state,
As noted by David Nugent (1994), the bulk of the literature en state-sociery
whether local, political, religious, or other communities, that are imagined
relations argue en the basis of an implicit model that posits two abstractions,
to be located outside, but in relation te, the state. Much of the recent discus-
state and community, as two essential and bounded entities in opposition to
sion of the nature and dynamics of this relationship draws en the notion of
each other. One is seen as essentially expanding, transforming, and coercive,
hegemony, thus posing the question of how noncoercive forms of domina-
the other as essentially conservative and actively resisting imposed transfor-
tion are constituted and how communities are brought within the purview of
mations. This, however, is only one of several possible "juncmres" of state-
the state. community relations. In the present volume, Nugent himself shows how the
In particular, remembering Gramsci's preoccupation with practical polities
imagery of this kind of state-communiry opposition in the case of Chacha-
and the construction of intellectual and moral leadership, we may ask about
poyas, Peru, is an outcome of a historically specific process of transformation
the importance and dynamics of polities, understood as a distinet social field,
from the 193os to the 198os. In the 193os, in the context of an emerging
in relation to the constitution, negotiation, and change of state-centered he-
populist regime, the petit bourgeois Chachapoyanos were actively involved in
gemonies. In other words, how do political operators control or transform
producing themselves as a community of citizens, as well as che state as an
the state en behalf of specific economic and social groups?
effectively ruling apparatus in Che province. But from the late 196os, relations
In this volume, the relation between politics and the state is explored in
have deteriorated and the communiry has developed into an antimodern, anti-
two different ways. Several of the contributions deal with popular perceptions
statist "traditionalism" depicting the state as external and imposing.
of polities as something that tends to "pollute" the state, and the state as

INTRODUCTION 23
22 INTRODUCTION
 Finn Stepputat shows how village populations in postconflict Guatemala
community is imagined either as something good, pure, and authentic or as
engage in a similar, not necessarily coerced, extension of state institutions.
something dangerous, unpredictable, and ungovernable, as in the aboye case
Intertwined with struggles for communal leadership and collective recogni-
of india. Genealogically, these opposed imageries of community and state
tion, many villagers strive to develop their places into urbanlike, formal sites
can be traced to the tension between romanticism and rationalism in the Eu-
of governance with public services, offices, parks, and other elements that ropean tradition (Hansen 1997a).
symbolize the recognition of communities of citizens. In the process, "local
Insofar as the notion of community is locked in an opposition to mod-
community" is stabilized as a territorial, administrative entity, an interface
ern, racional society it calls for images of localization, boundedness, reci-
between state and population. In this sense, the labeling and institutionaliza-
procity, and so on, but also of tradition, backwardness, parochiality, and im-
don of a village-community works as a kind of enframing of segments of
mobility. Such images are often invoked in patemalist governmental or non-
the population (Mitchell 1988). The struggle for state-centered yet autono-
governmental interventions on behalf of the communities, such as those
mous urbanization feeds the local appropriations of the system of political
promoted by the indigenista movement in several Latin American countries.
representation and contributes to the creation of a space of local politics:
Here anthropologists and other urban intellectuals developed policies and
the"politics from here" as opposed to the (less legitimate) "politics from
techniques for the integration of "backward" and "humiliated" indigenous
there."
people into the nation-state. Several progressive governments have taken this
Martijn van Beek offers usa different example of state-community relations
ensemble of images, policies, and techniques en board, for example, in Mex-
in which the introduction of bureaucratic categories of inclusion evokes prac- ico after the revolution.
tices of representation and belonging that go beyond the "either resistance
An increasing number of studies Nave shown how state-centered represen-
or compliance" dichotomy. Analyzing the process leading to political au-
tations have worked to incorporate communities in a hierarchicallyorganized
tonomy for Ladakh in India, van Beek shows convincingly that ethnicization
yet homogeneous nation-state through strategies that relate certain identities
and communalization is the price for being included in the liberal democra-
to certain spaces, time sequences, substances, and so en (e.g., Urban and
cies of the contemporary world of nations. At the same time, however, the
Scherzer 1991; Coronil and Shursky 1991; Rowe and Schelling 1991; see also
British colonial fear of "communalism" and the posterior denial of commu-
Alonso 1994). Who are at che center, who are at the margins? Who belong to
nalism in India have prernised the specific forro of political representation of
the past of the nation, who belong to the future? In this way, representations
the Ladakhi. They have been recognized and granted autonomy as eight dif-
tend to naturalize some groups occupying positions in government or in the
ferent tribes. Such exclusive categories are at odds with the disorder of social
political system and other groups occupying inferior positions. However, we
practices of identification in Ladakh, but because tribalism and communal-
need more studies to scrutinize the institutional aspects of such hegemonic
ism are extremely powerful images in this context, the population engages strategies. We have to ask how the opposition and the boundaries between
in everyday dissimulations in order to practice the exclusivist categories of
state and communities have come into being, which differences and identities
political inclusion.
that have been subsumed are the main opposition to the state, how relations
Contrary to the school of subaltern studies, our usage of the notion of are organized and negotiated across the boundaries, how communities are
community is not a priori; it is not referring to a more or less autonomous
represented, and by whom. Community may well be represented differently
repository or space of resistance to domination, homogenization, and dis- by different politicians, by schoolteachers, and by other contenders for lead-
ciplinary techniques. In spite of his relational analysis of the formation of ership positions.
community, Partha Chatterjee, for example, sticks to a binary opposition,
In her study of one of the many marginal regions in contemporary Indo-
not between "state and civil society" but between "capital and community"
nesia, Anna Tsing notes that the formation of local leadership is of vital im-
(1993: 13). Like the abstractions of people or "the popular"-usually de-
portance in the incorporation of the region into Indonesia's dominant lan-
fined through negativity, as opposed to elite and elite ways of doing things- guages of stateness: "At the border between state rule and the wild stand

24 INTRODUCTION
INTRODUCTION 25
those who da reto define, dcfy and dcmand ad ministration. These are the men exist outside or independently of che state bar are vital components in domi-
whom 1 cal) 'leaders' hecause thcy are inibitious enough to ccll che govern- nant ideological formations that try to delirio the cliaracter and limirs of the
men[ that tlicy represen[ che conmiunity and their neighbors that they repre- state, how politics is supposed ro be conducted, how institutions should func-
sent che scate" (1993: 72). Tsing shows clic multiple roles pcrformed by [hese tion, and so on. Thev are, in other words, intrinsic co che svmbolic producrion
inca as brokers benveeu governmen tal agencies and che community they of stateness. As Mitchell (Iggg) has shown persuasively, che emergente of
hring roto existenec ni a clear forro, but also how [hese roles produce al] political science as a discipline in posnvar America was indeed informed bv
aurhority that is employed in setding local disputes within che communities' che larger fear of community expansion and by a quest for American global
own discursive registers and pracrices, such as marriage cases (127-53). Ole hegeniony. It also become a highly influential analytical and disciplinary rech-
paradox is that as thesc marginal arcas are subjected to more intensive gov- nique thac sought to devise a eertain universal and "normalizing" conceptual
ernance, the importante of these men tends to recede: "Local leaders invoke vocabulary within whieh "state," "political system," "society," and "econ-
the authority of the state, but often lose out if che state arrives" (iSi).5 omy" could be understood as discrete and concrete entibes available for anal-
The contributions in Chis volume share an understanding of hegemony as ysis and amenable to governmental intervention (77-So).
a process of constructing "not a shared ideology but a common material and In che same vein, but at a different level, Akhil Gupta's analysis in this
meaningful framework for living through, talking about, and acting upon volume of an extension program for anganwadis (day tare centers) in North
social orders characterized by domination" (Roseberry 1994: 361). In this India points usefully to the way in which ostensibly technocratie sehemes
sense state-community relations may be interpreted as hegemonic processes Nave profoundly political effects. Gupta shows how the program problema-
that over time develop "a common discursive framework": a shared, state- tizes gender inequality as a developmental problem, and how the official de-
authorized language of cognition, control, and contestation. An importan[ piction of increased independence of women as a possible source of devel-
feature of such a framework is rhe formation and delimitation of a distinct opment and economic gain slowly encroaches en older discourses en gender.
field of politics, including che definition of specific spaces for politics and a This practica) reconceptualization of gender may over time contribute to the
common "language of eontention" for struggle and negotiation among dif- transformation of gender relations in North Indian villages, although not nec-

ferent political actors (363). essarily in the emancipatory direction envisaged by che policymakers.
This perspective allows os to scrutinize che complex processes through Hegemony thus also works through che development of technocratie pro-
which certain phenomena become objects of political debate and eventually granas and institutions that govern by virtue of routines , interna) bureaueratic
political intervention, and others not; how ideological formations produce logics, and allotted resources without being directed by political forces in any
distinctions between che politically permissible and nonpermissible, between strict cense. In this sense, hegemony is diffuse and difficult to pin down. If
proper public conduct and improper. Transitions, for example, from conflict we further consider the complexity of governance in modern, politically plu-
to postconflict or from one political regime to another, are privileged contexts ralist, and decentralized states where politicians and bureaueratic bodies at
for che investigation of ways that reginles pursue and negotiate inclusions different administrative levels may dispute decisions, divert programs, and
and exclusions from che limelight of che political field. struggle over jurisdiccional boundaries, it becomes very difficult to perceive
In other words, the delimitation of che field of politics defines che bound- of hegemony as being closed, monolithic, and coordinated. Compared to the
ary between undisputed, naturalized, and commonsensical doxa and the het- older and more centralized apparatus inherited from che colonial powers, it
erodoxa of politically disputed alternatives-at least in the discursive gentes is undoubtedly much more difficult ro dominare and control a multilayered
that symbolieally eonstitute stateness, such as che distinetive bureaueratic "le- and intensely competitive system of democratic government that has emerged
galese." At the heart of the doxa of the political fleld are the professional in many postcolonial societies.
languages and the concepts and categories by which social and political sci- Partha Chatterjee (1998) has argued recently that convencional distinctions
ences analyze state, econoiny, and politics. These forros of knowledge do not between state and civil society are unable to capture che richness and speci-

26 INTRODUCT'ION INTRODUCTION 27
 ficity o£ che actually existing forms of political struggle one finds in post-
perpetuating logics, bereft of any unifying and encompassing rationale. For a,
colonial societies. The state and civil society belong, he argues, te the same
new political regime to be effective in implementing parís of its professed
conceptual world of orderly negotiation of interesas properly organized and
objectíves, it needs to produce a fairly coherent " state project," as Bob Jessop
conducted according to certain rules and conventions : " The institutions of
(1990) has argued . Jessop suggests that political forces that desire to trans-
modern associational life [were] set up by nationalist elites in the era of co-
form a society must lec themselves be absorbed in thoroughgoing institu-!
lonial moderniry ... [ and] embody the desire of this elite to replicate in its
bona) reforms and in a certain reinvention of che state . Governance and at-
own sociery che forms as well as substance of Western modernity" (6z).
tempts to transform social structures through administrative reform in the
Against such a yardstick of formal and educated debate and organization
main do cake place inside che technocratic confines of government depart-
most forms of politics and negotiations of power in the postcolonial world
ments. Only insofar as a thorough reform of institutions Cake place, as it
inevitably appear chaotic and lacking in purpose and formality , Chatterjee
happens at the behest of ANC in South Africa right now or as che Congress
suggests . Instead, he coins the term "political sociery" for che zone of nego-
Patty gradually brought about in India in che 199os and r96os , may we legiti-
tiations and mediations between state and population , wherein the main me-
mately talk of a political force actually dominating [ he state in any meaning-
diators are movements , political parties, informal networks, and many other
ful sense.
channels through which the developmental state and the vast majority of the
But often, such structural reforms are not carried through. Yesterday 's revo-
population interact . This distinction between civil and political sociery is per-
lutionary regimes end up focusing en political changes of symbolic value or
tinent and highly relevant . If we are to understand how issues of welfare and
en crude nationalism rather than en implementing structural reform. The
questions of rights and democracy are negotiated in che postcolonial world, case of Zimbabwe , where revolutionary rhetoric has been combined with the
we need to understand the dynamics , the tacit rules, and che historicity of its
persistence of a colonial structure of landholding and agricultura ) econorny,
many political societies.
immediately comes to mind . Without exaggeration , we might say that most
However, the rough and tumble of competitive politics at many levels does
contemporary societies remain governed by yesterday 's administrative sys-
not necessarily mean that hegemony is ineffective . On che contrary, a low leve)
tems and procedures.
of political and ideological coordination , a diffuse nature of power, a non-
Bearing this relative "ungovernability" and inercia of che state in mind, che
reflexive routinization of governance and political processes , and a common-
issue of how ruling parties rule and how we can study che ways political op-
sensical acceptance of the domination of political life by certain groups and
erators operate may be restated in a slightly more precise form. We mention
families may well be what make hegemony endure. But if this is the case we
only three of the modes of direct political intervention in che processes of
must consider the question of how, or if, political operators are able to estab- governance that authors in chis volume touch en:
lish control and alter hegemonic relations through the state apparatus.
r. Prior to any major change in policies or institutions, most governments
will appoint committees to review an area, conceptualize che problems, and
The Art of Polítia
recommend solutions in white papers or reports . Such committees are often
To understand how political forces deal with the state , how they seek to ad-
staffed by senior bureaucrats whose own entrenchment in che social world
dress and reproduce the constituencies and social interesas they consolidated
and che languages of che bureaucracy ensures that their diagnosis remains
(or created ) on their road to political office, we need to analyze more carefully,
within the dominant discourse and the proposed amendments of govern-
and with more ethnographic precision , what ruling parties do when they rule.
mental techniques remain moderare and gradual. Within well-established
Assuming political power does not mean that a new government can
sectors with complex and closely woven networks ' of entitlements , systems of
change institutional routines overnight or that social practices within che bu- rank and promotion , and so en, reforms are often extremely dif&cult to push
reaucracy can be easily modified. The state is an enormous and amorphous
through. Certain sectors of the governmental apparatus , due te their origins
mechanism that functions along a whole range of discrete and often self-
in the colonial order or a military or authoritarian legacy, have been allotted

28 INTRODUCTION
INTRODUCTION 29
 valuable texts in our understanding of rhe dynamics of broader conflicts be-
considerable autonomy, and over tinte Nave developed extremely resilient
tween social elasses and communities. David Nugent's exploration of state
forros of organization, recruitmcnt, and functioning that few political parties
fbrmation and rhe struggle over political power and designs of institutions
dare to confront. The militare, che police, clic system of prisons and correc-
in Peru in rhis cenrury is a fine example of che insights such an approach
cional institutions are often such almost self-governing instimtions within rhe
can vield.
larger network of che state. Although rhe state as a whole may be fragmented
The most widely used and most immediately effecrive method used by
and "weak," we need to cake note ofvarying degrees of "softness" and "hard- 3.
political force ro exercise power and to consolidare as popularity and support
ness" in different seetors of rhe state apparatus. Ir is in the face of this prob-
is intervention roto rhe implementation and administration of specihe poli-
lem of resilicnce and reluctante toward reform and scrutiny within the secu-
eies and regulations at che local leve). When lists of [hose entifled ro new
rity apparamses that a range of "truth-producing" technologies and strategies
agricultura) credits are drawn up, when ehildren are admitted ¡oro govern-
of reconciliation have becn employed by new and democratic regimes in Latin
ment schools and colleges, when new clerks are employed in a government
America and South Africa. Instead of an all-out confrontation with the often
department, when liquor licenses are issued, when builders are allowed to
unrepenting executioners of yesteryear, these processes have sought te by-
build on certain plots-rhe lis [ is endless - local politicians are often in-
pass che security apparatuses and instead creare a common platform for a
volved in putting pressure en local of icials. Nacional-leve) politicians are in-
broader and collective catharsis of che excesses of past regimos.
volved in similar efforts, only on a larger scale concerning the sanetioning of
2. It is tempting for political parties eager to show results tu create new
large industrial projects or large construction projecrs.
governmental institutions or programs instead of reforming or terminating
This parí of rhe political vocation involves the ability to construet a large
[hose in existence because of the web of entitlements, resources, and insti-
network of contacts, mutual favors, and economic resources that enable po-
tutional routines surrounding a given sector. New programs and discourses
litical operamrs to put pressure on a local bureaucrat (threats concerning pos-
are believed to be able to bypass and displace older and exisring srmetures by
sible transfers are common) or to win influence in local boards and commis-
virrue of the energy and hegemonic strategies pursued by a new regime. This
sions, to make friends with influential bureaucrats, and to rise in rheir own
intricate play berween older forros of governance and new forms of rationali-
political parry. From the point of view of rhe consumer of government ser-
dies seeking to hegemonize a fleld of intervention lies at che heart of Steffen
viles, chis local art of politics also requires a certain command of rhe "rules
Jensen's paper in Chis volume en rhe attempts to reproblematize che field of
of game" as well as the discursive register through which bribes and kick-
crime, policing, and correctional institutions in South Africa.
backs are talked about and constructed as reasonable within a local cultural
The result seems tu be that each new regime builds a number of new insti-
economy, as Gupta's (1995) path-breaking paper on corruption demon-
rutions or nurses particular arcas with grearer care and zeal, often reflecting
strated. In a similar, if more generalizing, vein, de Sardan (1999) has recently
che larger ideological formations and communities out of which they have
pointed to a range of cultural logics and everyday forros of reciprocity and
emerged. In intensely competitive democratic semps, che result seems to be
obligation in sub-Saharan Africa that contribute to the reproduction of what
that each political movement or parry seeks to establish and maintain zones
he tercos the "corruption complex." A very substantial part of rhe everyday
of loyalty, reproduced through flows of patronage, in various parts of rhe
forros of governance and political power is exercised chis way. Police con-
bureaucracy. It is a process that, needless to say, often brings forth the intrin-
stables are told to look rhe other way or to arrest someone particular. The
sic fragmentation of rhe state to an extent thar sometimes jeopardizes its
name of one farmer is delered from the loan scheme and che relative of a
central mythological dimensions. The long-terco result of such competing
leading family is entered instead. Municipal authorities are told to ignore the
and suceessive projecrs of domination and social reform through constan[
consrruction of unaurhorized buildings. Examples are legion.
formation, addition, and restructuring of state institutions is a morphology
This forro of power, however, cannot easily be equated with the power of
of governance, rhar is, hisrorical layers of instimtions that left traces and
certain classes or communities, nor can it necessarily be taken as proofof the
documents as they were reformed or rebuilt. Such morphologies can be in-

INTRODUCTION 3r
3o INTRODUCTION
 1
power of one particular party. Most political figures are involved in chis "po-
In the face of such obvious ambiguities of both resistance and power it
litical retailing" that has very little to do with dorninating or restructuring the
seems that imposing a universal dialectic of power and resistance on diverse
state, but merely with influencing the course of a few of the micro-operations
of the state . But the net result of these millions of everyday interventions in and complex situations may narrow rather than open the scope for interpre-
tation . In his foreword to a new edition of Ranajit Guha's path -breaking Ele-
the functioning of local institutions is, of course , that governance becomes
increasingly " porous" and fragmented at the local leve! and that the imple- mentar. Aspects of Peasant insurgency in Colonial India , james Scott argues that
Guha has avoided such a narrowing of perspective and has avoided reading
mentation of most policies are deflected, if not stunted.
the past in terms of the present : "At every toro Elementary Aspects emphasizes
Resisting Reyimes the dangers of reading the process of insurgency with a political grammar
based on mid - twentieth cenmry nation-state political forms. In the place of
The category of resistance remains a very unclear and opaque term in spite
formal organization ... Guha finds informal networks ... in the place of for-
of the enormous literature en the subject in all the disciplines of the social
mal messages and public conflict, Guha finds the world of rumor" ( 1999: xiii).
sciences. Much like the state-community dichotomy discussed aboye, the
Both Guha's ( 1999) rich interpretation of a century of peasant insurgencies
very definition and concepmalization of resistance , of defiance or insurgency
in colonial India read "against the grain" in official reports, and Scott's (1985,
is vitally dependent on the character and clarity of the regime , or state, that
is opposed . Resistance , most anthropologists , lggo) work on everyday forms of defiance , joking, and other kinds of "low-
historians , and sociologists
intensity resistance " have recorded a valuable range of acts of defiance, or
agree, is a category and a type of social practice that cannot be understood or
passive insubordination , in local and emic terms far removed from the world
presupposed outside its historical context. Yet , there is something universal
of formal politics and organized opposition. The question is, Does the uni-
and transhistorical in the way resistance is conceptualized and assumed. This
versal category of resistance or insurgency actually obtain in all these con-
was true of Marxist scholarship but also applies to contemporary work of a
poststucturalist persuasion , inspired among other things by Foucault's re- texts? Do we not tend to inscribe a somewhat heroic dimension into actions

mark "Where there is power there is resistance , and yet , or rather conse- that local actors may see as mundane, unexceptional, and maybe deeply am-
bivalent? Guha is aovare of such ambiguities in acts of defiance, the blurring
quendy, this resistance is never in a position of exteriority in relation to
power" ( 1978: 95-96). of lines between regular crime and collective resistance against authorities
This remark seems to afhrm resistance as an anthro-
pological universal , something that is alwayslalready out there (1999: 77-108 ). Insurgence and resistance remain, nonetheless , the overall
. If we cannot
leas through which he interprets the British colonial reports on dacoitry, loot-
see "it," out conceptual tools must be inadequate and insensitive to the local-
ing, killing of aristocrats-events that often were mediated through idioms
ized and emic categories that are the medium of resistance . But Foucault's
of religion and community.
notion of the imbrication of resistance in every operation of power has, along
with work by James Scott , Michel de Certeau , and others Out argument is that we need to be more sensitive to the historicity and
, made resistance
polyvalent nature of the expressions , symbols, and acts we intuitively may
into a much wider and more ambiguous category than it used to be. Lila Abu-
Lughod observes : " What one finds now is a concern with unlikely forms of register as resistance . The lives and acts of ordinary people may as well be
resistance , subversions rather than large-scale eollective insurrection, small intertwined with the lives of the powerfut in the "illicit cohabitation" that
Mbembe writes about, and revolts or resistance may well serve ends, repro-
and local resistance not tied to the overthrow of systems or even to ideolo-
gies of emancipation" (1990: 41 ). Abu-Lughod demonstrates that the use of duce structures of domination , or create new forms of power that are more
repressive and violero [han those preceding them. The works of Guha, Scott,
lingerie and dreams of romantic love among young Bedouin women indeed
and others have been crucial in wresding the question of resistance and revolt
is a form of resistance to patriarchal forms of domination , but that these
out of the clutches of a powerful teleology that saw "primitive " revolts of
practices also entail submission to other dominant ideologies , such as the
peasants or marginal people as prepolitical stages of emancipation that could
privatization of the individual and the family and market - driven consump-
emerge fully as poliricallproletarian consciousness only in the modern (and
tion (43-55).
Western) agas Out argument, however, is that we should go one step further.

32 INTRODUCTION
INTRODUCTION 33
 tionaries. As tire general ideas about what a state was, what it could do and
As wc try to make che state a Icss natural cntity we should also endeavor ro
should do expanded and diversified in ninetecnth-centmy Europe, rebels
reverse che tendency toward reading resistance ti) the state finto every mun-
and revolutionaries also developed ever more refined ideas about thc utopias
dano social act and instead listen to and record che discourses, the organiza-
"counterrepublics" rhey desired ro set up, wirh clic Paris Commmlle in 1871 as
tion. and che costes[ of tliat which from a distance appears as resistance. h he
a paradigmatic example.
result will inevitably be more perplexing and unclear but also more interest-
Twentierh-century guerrilla warfare, famously theorized by Mao Ze-dong,
ing. as Shcry Ormer (1995) argues in her critique of what she terms thc
set new standards for the sophistication of tire "parallel stare" organizad by
"edmographic refusal" lo engage resistance entpirieally.
revolutionaries in liberated zones or as "night governments" in contested
Lee os bricfly mention two of tire ambiguitics involved in resistance and
arcas. These were state structures that used some existing structures of gov-
revolts against states: first, the logic of emulation of dominant orders by reb-
ernance (taxation, territorial control, village councils, etc.) but also often
els and revolutionaries; second, the collapse of states and emergente of war-
aimed at introducing radically modern discourses, for example, of gender
lordism that has (re)appcared in the late twentieth century. As pointed out by
equality finto marginal peasant communities by che use of "state-of-the-art"
Erie Wolf (1969) in his classic work en peasant wars in che twentieth century,
techniques of organization and surveillance, new procedures of justice, and
these rebellions grew out of a complex and layered interaction between local-
more. In many cases, guerrilla movements inspired by Maoist doctrines be-
ized disgruntlement and the desire for upward mobiliry among "middle peas-
carne exceptionally effective builders of such parallel states by pursuing dra-
ants" that was given a certain interpretation, direction, and shape by an edu-
conian reform of social structures, removal of competing centers ofauthority,
cated, ideologically sophisticated leadership. The ordinary peasant had litde
such as religious institutions, and extension of control and political surveil-
idea of socialism or the catharsis of revolution but often desired a return ro
lance. During thirty years of guerrilla warfare the Eritrean People's Liberation
an idealized state of social equilibrium governed by proper moral conduct of
Eront created such a "shadow state," partly organized around che kerbeles of
patrons and officials, albeit in unequal relations (1969: 276-303).
the lineage society of the highlands and around che imperatives of war and
Guha points to how revolts inevitably cake place within social imaginaries
production, bu[ always controlled by Che disciplined cadre structure of the
structured by prevailing arrangements of power and only rarely transgress
movement. These structures became the backbone of the new independent
established notions of authority but rather tend tu reproduce [hese: "peasant
state in 1993 ( Iyob 1995).
kings were a characteristic product of rural revolt ... and an anticipation of
Here, as among che Tamil Tigers in the laffna península of Sri Lanka and
power was indexed en some occasions by the rebels designating themselves
the Sendero Luminoso in the Andean provinces in Peru , an all-pervasive logic
as a formally constituted army (fauj ), their commanders as law-enforcing per-
of militarization, a strong ideology of personal sacrifice and the ennobling
sonnel (e.g., daroga , subhadar, nazír, etc .)-all by way of simulating the fune-
death in struggle , and devotion to what was believed to be an elevated lead-
tions of a state apparatus" (1999: lo). ership at che heart of the shadow stare created organizations that were both
This logic of negative emulation and reproduction of structures of gover-
effective and terrifying in their determination ro control people, resources,
nance and, in effect, languages of stateness seem to be a recurrent feature
and territory (Degregori 1891). Peasant populations in these arcas carne lo te-
of rebels and revolutionaries in many parts of the world: from the emperor
alize that che governance practiced by [hese organizations often was harsher,
crowned by rebellious slaves in Haiti in che late eighteenth cenmry, to the
more effective, and less negotiable [han that of che old regime. In Pero, for
peasant leader of che nineteenth-cenmry Taiping revolution in China whose
instante, Chis harshness worked against the Senderistas, as did che impris-
millenarian dreams of a new utopian state led hico to regard himself as a new
onment of Sendero Luminoso's mythical leader Guzman, the father and
emperor, a new "Son of Heaven," ro countless local revolts that produced
"teacher" of the new shadow state, in 1992.
similar effects of negative emulation. Such effects were, of course, also highly
If some of [hese militant movements almost suffocated their new subject
productive in terms of giving idea, coherente, and structure to the parallel
populations in too much and roo tight governance, che opposite seems to be
systems of governance, control, and sovereignty set up by rebels and revolu-

INTRODUCTION 35
34 INTRODUCTION
 true of some of the "collapsed" states in Africa. Here, warlords and
strong- ethnographic gaze, a strict Foucauldian view of modern governance as the
men have broken the territorial sovereignty of che state, throwing large terri-
inexorable global spread and proliferation of certain discursive rationalities
tories into an almost Hobbesian predicament of apparently randomized but
and certain technologies tends lo crumble. These forms of governmentality
also ethnicized violente. Rather than shadow states, we are dealing here with
do exist and their techniques and rationales do circulate, but they only effect
shadow economies. The control of territories, people, and natural resources
practical policies or administrative practices in a rather slow and often indi-
is not bureaucratized but relies en alliances, engagement in transnational
rect way: sometimes as justiflcations for new measures or norms, sometimes
economic networks, and coercion (Richards 1996; Bayart et al. '998). Chill- simply as a form of "scientific" diagnosis, but always in competition with
ing practices of literally inscribing the sovereignty of the warlord into popu- older practices and other rationalities.
lations by maiming and disfiguration- as in Sierra Leone, Liberia, and the
We have pointed out in this introduction that the state, governance, and
Lord's Resistance Army in northern Uganda-seem to be integral lo these
the effects and subjectivities shaped by the languages of stateness of our time
more mobile and floating sovereignties.
need to be denaturalized and studied in rich ethnographic detail as an inte-
These "commercial-military complexes," free from the weight of bureau-
grated part of the cultural economy of postcolonial societies. We have also
cracies and creditor demands, are much more flexible than their developmen-
pointed out that we cannot assume that an institution, a policy paper, a dis-
talist opponents, and throughout the '99os severa) of the latter have mim-
cursive construction, a protest, or the practices of a government official are
icked the practices of che warlords (Duffield zoo'). William Reno ('998) and "the lame" all over che world. We have emphasized that the state is not a
Mark Duffield ('998, zoor) have argued that this kind of warlord society may
universal construction and that states have widely different histories, interna)
be seen as an innovative system of political authority corresponding lo the logics, and practices that need to be understood and studied. Yet there are
neoliberal world order, rather than as abnormal, and temporal, deviations many similarities, a real and effective circulation of a range of languages of
from the governmentalized state,
stateness around the world, and very real and enduring mythologies of state.
This does not mean, however, that the languages and images of stateness If the state as an actual social form is not universal, we may suggest that the
evaporate. The weight of che international system of states and their all- desire of stateness has become a truly global and universal phenomenon.
encompassing rituals of authorization reinforce the need for articulations and In that light, we should perhaps regard the rhetoric of state officials, che
complicity between the warlord systems and state institutions. Purthermore, nicely crafted white papers and policy documents, the ostensibly scientific
representatives of the nonbureaucratic warlord systems may justify their ac-
forms of governance, the grand schemes and organizational efforts of gov-
tions with reference lo previous violation of their rights and their exclusion
ernments, with al¡ their paraphernalia of vehicles, titles, and little rituals, as
from state systems. As Monique Nuijten (x998) has illustrated in her study of parts of a continuous state spectacle asserting and affirming the authority of
land, domination, and politics in rural Mexico, state subjects do not neces- the state. These spectacles only occasionally succeed in producing che specific
sarily give up claiming rights and entitlements just because the state represen-
social effects they aim at, but always reproduce che imagination of che state
tatives never fulfill their promises. In this sense, she argues, the state may as the great enframer of our lives.
best be understood as a "hope-generating machine."

The Great Enframer Notes

The study of localized political struggles, of the functioning of local institu- x. This is how Foucault put it in an interview with Paul Rabinow shortly before his death (Rabi-
tions of governance, of often disorderly and ambiguous forms of defiance or now 1984: 381-90).
insubordination, of celebration of the myth of the state and its physical rep- 2. For an incisive argument concerning che transformadon oflogics of sute formation and gov-

resentations should caution os when it comes to drawing conclusions regard- ernance froto che late Spanish empire to che early nacional sute in Central America, see
ROnsbo (1997).
ing the uniformity of how the current global languages of stateness are spo-
3. The genealogy of one of che great fetishes of the modern mobile world, che passport, has
ken, understood, and converted into policy and authority. If subjected lo an recently been explored by lohn Torpey (2000). In fascinating detail Torpey explores how trates

36 INTRODUCTION
INTRODUCTION 37
 in euro pe, .Joug ditterenr bur: iro converging fines, ni nce ninereenr h cenrury gradti Il'y es-
1 STATE AND GOVERNANCE
rablirh a monnpoiv creer virar he lean. `(le Iegitjrnam mcans of mobility."
p As demonstrated he I<amm^nvieh (193 ) in his elassie s¢tdy ot medieval English Iaw, roval
authorilc emhodied te nce Ring trua divirled inm u sublime eternal bode anda profane, honran
hodv rhar Isn vas nc, ohjecr ti Innuorrahle teles. jokcs, and Ole popular wit See I lamen ,.
in diis rolunre for fuelher di.cussion of the releeance of Kantororvie h's worle tor the

f atoe pract lcee.

ind andnatpolog iras providcd velueblu ronnribn rioos rhar aro useful Ibr nc, anal,-
sic nf dm inreractinn berrvee n local aurhnritie. and itutiturious of rhe modero eran,. se, fiar

exmnplc, Vincent 1990 . and Gledhill 1994.

ed book Primifiue Rebela.
6. This is rhe case, Eor inseance , in Hohsbawm 's(ig6g ) othenvisc celebra r

38 INTRODUCTION
 "DEMONIC SOCIETIES" Liberalism, Biopolitics,
and Sovereignty

Mitchell Dean

In a lecmre at Stanford in 1978 Michel Foucault said: "Our societies have

proved to be really demonic since they happen to combine those two games-
the city-citizen game and the shepherd - flock game-in what we call modern
states" (1988c: 71).
What could such a statement possibly mean? Was this a case of hyperbole,
an attempt to catch the ear of an American audience , designed to convince
them of his critical intent and credentials ? Was it something we were meant
to pass over quickly and move on to the more detailed analysis of different
forros of what he called "political rationality "? Or did this statement sum-
marize and encapsulate a certain dimension of what Foucault would identify
as the o goverm rit, as 1 suggest in this essay ? As is well-known, the
-IK
yrit^ literature on those "arts of governing" arises, for Foucault, with the
crisis of spiritual governmept and_lhe_Qastorate in thé hffétntfrand-siz--
teenth cénturies . Ir ingmres finto a multiplicity of problems " concerning the
`right way ' to govern children, a family, a domain, a principality " ( Foucault:
1997a: 68 ).' Foucault's theme, therefore , is of "a more general questioning of
:../. government and self-government , of guidance and self-guidance ," of all the
attempts at the calculated direction of conduct undertaken by various authori-
ties. This extends " beyond the state " to the programmatic activity of com-
munities and community organizations , unions, corporations , and associa-
tions of all kinds. II even extends to those attempts we ourselves make to
transform our own conduct with different objectives in mind so that we might
become a different kind of person . In brief, government for Foucault refers
us to the theme of the "conduct of conduct."
 -Phiss essav is in large parí a critical exposition of Foucault's general scheina something in this i nterpretation. Here, however, 1 argue that wc should also
tbr understanding these arts oí governing. It argues that this schema cannot understand this terco in a much stronger sense, as a reference to what bedev-
be properly understood without referente to concepts such as biopolitics and ils modern fbrms oí government and rule, and to that which, at leas[ in parí,
sovereignty. 1he point of ibis evposition i5 not l h ilol ogi cal, however. Ratheo can help us pose the problem of political danger and evil. 1 suggest that Fou-
it is to deepen our understanding of the resources that an analytics ofgovern- cault sought to pose ihe problem of the political danger as that which lurks
ment has ro draw en and of some of irs critical bases. Ir is also to suggest in our rationalities and techniques of government in the various attcmpts ro
some of the lirnits oí Foucaults interpremtion of political rationaliry. Fou- combine elements of the "shepherd-fiock" and "city-citizen games." Ibis
cault's 1978 statement is a usefld epigraph for this task. understanding is borne out by certain other texts that are in a not distant
Ibis statement is notwithour its problems. Foucault assumes that we know proximity to this one. The Stanford lectores, with their focas on pastoral
what he mearas by "out societies." In common with the vast majority of Eu- power, reason of state, and Polizeiwissenschaft, are based en material from Fou-
ropean and North American theorists, he was resolutely "metropolitan" and caulPs course un "Security, Territory, Population" in 1978 (1997a: 67-73).
Eurocentric. His analyses tend to assume that one can talk relatively coher- There is more than a trace of the same problem of political evil in evidence in
ently about forros of government, rule, and power that are interna] te, as he the extraordinary final lecture Foucault gave in his previous course in 1976
puts it, "what we cal] modem states," in isolation from the relations between (19976: 213-35) and the final chapter in the first volume of his History of
states and other political formations. As many have suggested, most notably Sexuality (1979b).' There, the shepherd-flock game takes the forro of the mod-
Ano Laura Stoler (1995), the study of colonial power relations and colonial ern life politics, biopolitics, and the city-citizen game is cast in the language
bodies is largely absent from Foucault's studies of discipline and sexuality. and practice of sovereignty. in both of [hese earlier texts, Foucault uses Na-
More recently, a number of writers have noted that Foucault's account of the cional Socialism as a key example of certain of the dangers perhaps more
art of government finds its limit in a relative neglect of issues of international discretely manifest in other doctrines of rule.
government, and the international forros of government that are among the This further elaboration raises other types of problems. These arise from a
constitutive conditions of sovereign states? It is thus fair to say that Fou- potentially "totalizing" reading of the aboye statement. On che one hand,
cault's analyses of the arts of government, with some significant exceptions such a statement might appear to suggest that we do not have to engage in a
and indications in his lecmres, are largely "internalist." That is, they ignore genealogy of the specific ways in which the arts of government are shaped
the international arts of government that are the condition of these relatively and imagined within particular contexts. It is of course necessary to consider
autonomous, sovereign, territorially bounded states, and che practices that not only the pervasiveness of different rationalities of rule, such as sover-
assign populations to specific states in the modern system of states. When eignty and biopolitics, but also how they are articulated in specific contexts.
Foucault discusses " our societies," we know he means ihe populations that This is the case, particularly in che colonial and postcolonial era, when such
are assigned to the territorial unities of liberal-democratic states of Europe rationalities are articulated with alternative, older, or "indigenous" forros of
and North America and perhaps certain of their antipodean derivatives. More authority and government.° On che other hand, such a statement might be
deeply, he also appears often to assume that these states unproblematically read in a manner that underplays the contingent, unruly nature of these ratio-
coincide with relatively autonomous and bounded unities called societies. nalities and che manner in which they become available to be used by different
Turning to the terco "demonic," it would perhaps be permissible to read political actors in contestation. Indeed, tu follow Hindess (1997), one might
this terco as analogous to Socrates' ' daemón," which, as Pierre Hadot puts it, suggest that che use of the term "political rationaliry" in this context is rather
was both a kind of inspiration that carne over hico in an irrational manner limited in that the identification by Foucault of political rationaliry with ratio-
and bis real "character" (1995: 164-65). The demonic would thus be che nalities of government does not leave room for the political per se. An ex-
irrational and inspirational character of modern states, even accounting for ample of [his is when a liberal rationaliry o£ government is concerned to limit
something of their dynamism and capaciry for political invention. There is or regulate che effects of the partisan action of factions on che work of gov-

42 STATES OF IMAGINATION "DEMONIO SOCIETTES" 43

 ernment. We need to consider the way ratonalities of how we govern and are
two forms of rule advances our understanding of our imagination of the po-
governed are articulated with rationalities and forces that seek in some man-
litical in two ways. First , it establishes the importance of the longue durée of
ner to affect the employment of such rationalities . Thus, we need to consider
two very broad - and intertwined - trajectories of rule. Second, it argues that
what Weber called " politically oriented action ," that which " aims to exert in-
many current problems and dangers are located not in one or the other of
fluence on the government of a political organization ; especially at the appro-
priation , redistribution or allocation of the powers of government these trajectories but in the attempt to put together elements of the rationali-
" ( r968a:
ties found along these trajectories in the government of the state. That is,
55). Thus, we need to examine action that is oriented to affecting the gov-
whether political actors take the form of an incumbent regime, a parry, or a
ernment of an organization , as much as action oriented to shaping conduct.
social movement , those who attempt to affect the government of the state are
Or, to put it in Foucault 's terms (1988a: ig ), we need to discuss the ways in
forced, in very different contexts , to try to force together aspects found along
which these rationalities and technologies of government are articulated with
these two trajectories . And, to allow for Agamben, the point of articulation is
"strategic games between liberties ," that is, how they enter into a field of
found in the different conceptions of life.
political forces.
Foucault's statement , then, locates the problem of political danger in the
My exposition aboye suggests that the demonic character of modem soci-
combination , the "tricky adjustment" between two modes of exercising rule.
eties, according to Foucault , stems from the unholy combination of two dis-
The shepherd - flock game , or what he elsewhere calls pastoral power, has its
tinct trajectories: that of sovereignty and that of biopolitics . It may be that
birth in Hebraic and early Christian religious communities . Its genealogy
such a view cannot be maintained . The effect of Giorgio Agamben 's (1998)
concerns its transformation into a centralized and largely secular exercise of
argument would appear to be that there is a much closer and longer bond
power oven populations concerned with the life and welfare of "each and
between politics and life than Foucault allows . Drawing on Aristotle, Agam-
all" with the development of the administrative state in Europe in the seven-
ben argues that the Greek understanding of politics contained two concep-
tions of life : zoe, or bare life, is distinguished from bios, or politically or mor- teenth and eighteenth centuries . The city-citizen game has its sources in
Greco-Roman antiquity and notions of the polis and res publica and con-
ally qualified Efe. The constitution of the political is made possible by a kind
cerns the treatment of individuals as autonomous and responsible political
of exclusion of bare life from political li£e,that simultaneously makes bate Lfe
actors within a self-determining political community . This mode of exercis-
a condition of politics . Life is included ir P-01-,tic-s, not y e etg enth-
ing power has been transformed by modere liberal and republican doctrines,
century emergence o politics , but at the very genesis of Western concep-
notions of direct and representative democracy , and, most crucially, by the
tions of politics , by its very exclusion.
key status of citizenship being granted to certain rnembers of the population
This is not the place to rehearse the details of Agamben ' s appreciative cri-
within che territorial surte.
tique of Foucault , but he does indicate the possibility that Foucault has un-
One way in which the attempted articulation of these elements may be
derestimated the extent to which sovereign forms of power were constituted
viewed as demonic concerns the vacillation over the status of the welfare
in relation to notions of I e. n so doing , Agamben might be read as contrib-
state. On the one hand, national governments are loath, for a variety of
uting to an analysis of how modere democratic conceptions of sovereignty
reasons, to do anything that might undermine the responsible freedom of
already contain assumptions about life in relation to politics that open them
those who can exercise active citizenship and even seek to reform social pro-
to readily and somewhat surreptitious - cvlonization by biopolitical impera-
tives. However, Agamben himself may risk a lack of attentiveness to the spe- visions so that it might transform certain groups luto active citizens. On the
other hand, governments must find a way of providing for those with needs
cific character of modere biopolitics.
whether due to human frailty and mortality or the nature of the capitalist
Despite all the possibilities of reductive readings of ethnocentrism, of an
labor market itself. The genealogy of the welfare state seems to be bedeviled
unexamined focus on nation, and of a form of totalization that threatens to
by this problem of trying to find a norm of provision that can adjust the
neglect historical singularity and the fleld of contestation , chis statement of
competing demands of a subject of needs with the free political citizen. We

44 STATES OF IMAGINATION
"DEMONIC SOCIETIES" 45
 One key domain in which [hese processes exterior to but necessary ro gov-
can note that that genealogy would also show that Chis problem of welfare
ernment are constitured is biopolitics. This terco designares the very broad
states is also a problem of the relarions and conrpetition among sovereign
terrain on and against which we can locate che-liberal critique of roo much
states, most recently reconfigured as an issue of economic globalization.
government and its advocacy of Benjamin Franklin's "frugal governmenC
Imporrant as rhis welfare state problem and irs ramifications are, 1 want to
(Foucault 1997a: 77). Biopolitics is a politics concerning che adminisrration
focus here on another aspect of chis demonio character of inodern states.
of ¡ife, particularly as if appears at che leve¡ of populations. It is "che endeavor,
"rus is che character of what I cal¡, for want of a better terco, authoritarian
begun in the eighteenth cenrury, to racionalizo probleans presented to govern-
forms of rulo` This terco encompasses those practices and rationalities im-
mental practico by che phenomena characreristic of a group uf living human
inanent to liberal government itsclf rhat are applied tú certain populations
beings constimted as a population: health, sanitarion, birrhrate, longeviry.
held to be without the attributes of responsibie freedom. More direcdy, it
cace" (73). It is concerned with matters of life and death, with birth and
refers to nonliberal and explicitly authoritarian types of rule that seek to op-
propagation, with health and illness, both physical and mental, and with che
erare through obedient rather [han free subjects or, at a minimum, endeavor
processes that sustain or retard che optimization of che ]ife of a population.
to neutralizo opposition to authority.
Biopolitics must then concern the biological, social, cultural, economic, and
Very broadly, then, this retranslation of Foucaults sentence en che demonic
geographic conditions under which humans live, procreare, become ill, main-
namre of modern states amounts to something like che following: All ver-
tain health or become healthy, and die. From chis perspective, biopolitics is
sions of what might loosely be called modern arts of government must articu-
concerned with heredity and environment, with che family, reproduction and
late a biopolitics of che population with questions of sovereignty. And it is the
genetics, with housing, living and working conditions, with what we call
combination of [hese elements of biopolitics and sovereignty that is fraught
"lifestyle," with public health issues, parteros of migration, levels of eco-
with dangers and risks. I mrn first to che triad liberalism, biopolitics, and
nomic growth, and the standard of living. It is concerned with che biosphere
sovereignty before examining nonliberal types of rule.
in which humans dwell.
Drawing en the work of Robyn Lui-Bright (1997), we might say that there
Liberalism, Biopolitics, Sovereignty is an infernal and an external side of biopolitics. There is a social form of
government concerned to govern the life and welfare of the populations that
From che end of che eighteenth century until perhaps quite recently, [here
are assigned to certain states; diere is also a kind of international biopolitics
existed a common conception of government This was true for those who
that governs che movement, transitions, settlement, and repatriation of vari-
criticized and sought to limit existing forms of government and those who
ous populations, including refugees, legal and ¡Ilegal immigrants, guest
argued for their extension, their coordination and centralization. Government
workers, tourists, and students. This international biopolitics is a condition
would be regarded as a unitary, centralized, and localized set of institutions
of the assignation of populations to states and thus of social government of
that acted in a field rhat was exterior to itself. It would no longer be purely
concerned with "che right disposition of things arranged to a convenient any form.
Biopolitics is a fundamental dimension or even rrajectory of government
end," as La Perriére had argued (cited in Foucault 1991: 93). The government
from the eighteenth century concerned with a government of and through
of things would mees the government of processes. To govern would mean
the processes of life and che evolution of ¡ife. It constitutes as its objects and
to cultivare, facilitate, and work through the diverse processes that were to be
targets such entities as the population, the species, and the race. In Foucault's
found in this domain exterior to che institutions of government. These pro-
narrative, however, che detailed administration of life by biopolitical (and, it
cesses would variously be conceived as vital, natural, organic, historical, eco-
should be added, disciplinary) practices is not coextensive with the entire field
nomic, psychological, biological, cultural, or social. They would be processes
of politics and government. There are at least rwo other dimensions of rule
that both established che paradoxical position of life as at once an autono-
that are important here: economic government, which is infernal to the field of
mous domain and a target and objective of systems of rule, as at once ex-
government conceived as the art of conducting individuals and populations,
cluded and included within che exercise of sovereign power.

"DEMONIO SOCIETIES" 47
46 STATES OP IMAGINATION
 and the theory and practices of sovereignty. Both are invoked by liberalism as The notion of sovereignty is far from a universal and, like other concepts,
a means of criticizing and halting the effects of the generalization of the should be understood in its historical variation according tu specific regimes
biopolitical norm of the optimization of lile. However, both economic ratio- of practices and forms of political rationality. Furthermore, as the state's con-
nality and sovereign power, however democratized, might also be viewed as cern for its own preservation, sovereignty might be a particular end of gov-
ultimately referring to another conception of life-that of the zoe or bate ernment. Indeed, securing the sovereignty of states is an end of the art of
life. Economic rationality presupposes a world of limited resources and rela- international government. The existence of a system of sovereign states has
tive scarcity in the face of human needs and justifies itself in terms of the as its condition a form of governmental regulation of the internacional or-
"standard of living" or even "quality of life." The discourses of sovereignty, der. Foucault suggests that internally, however, in western European societies
particularly in notions of human rights, presuppose juridical and political from the Middle Ages sovereignty is principally conceived as a transcendent
status of individuals and populations by virtue of their mere existence as hu- form of authority exercised over subjects within a definite territory. Its main
man beings.
instruments are laws, decrees, and regulations backed up by coercive sanc-
Biopolitics, then, first meets quite distinct forms of political rationalityand tions ultimately grounded in the right of death exercised by the sovereign. It
knowledge concerned with the role of commerce in civil society. These take operates through spectacle and ritual, it prohibits forms of action, it seizes
as their theoretical object the notion of the economy as a self-regulating sys- things, bodies, and ultimately life itself. The symbolic language through
tem largely coincident with the boundaries of the nation. In doing so, polit- which sovereignty operates is one of the sword, of blood, of family and alli-
ical economy presents limits to the biopolitical ami of the optimization of the ance. In certain states, from the end of the eighteenth century, we know that
lile of the population. These limits are most clearly articulated by Malthus sovereignty has been "democratized" in that we have witnessed the develop-
and his absolutely crucial discovery, in the relation between the processes that ment of mechanisms of representation by which those deemed to possess
impel the growth of population and those natural ones that provide the sub- the required attributes can participate in the choice of who should stand in
sistence for the increasing quantity of human lile, of a realm of scarcity and the place of the sovereign. One aspect of the democratization of sovereignty
necessity. The bioeconomic reality discovered and enshrined in the work of has been to create a universal language of rights by which efforts are made
the English political economists of the early nineteenth century will be used to regulate the conduct of sovereign states by various international govern-
to generate new norms of government. Such norms must be factored against mental agencies . Another is that sovereignty and the language of rights has
the optimization of the lile of the population by biopolitical means, even if proven polyvalent enough to accommodate the claims of movements for self-
they are consistent with this goal of the optimization of life. determination among indigenous and other colonized peoples.
The notion of sovereignty has its own history and its own effects: it is The other aspect of sovereignty mentioned but not emphasized by Foucault
characterized by a power of life and death that, according to Foucault, was is that the notion of a nominally separate state with territorial integrity, sub-
"in reality the right to take lile and let live" (197ga: 136). Sovereignty under- ject to noninterference by outside powers, is itself a governmental product
goes its own transformation: in the juridical theories of the seventeenth and and a consequence of the "externa!" dimension of doctrines of "reason of
eighteenth centuries, such as those of Thomas Hobbes and Samuel von Pufen- state" (Foucault 1gg1: 104). The city-citizen game not only entails relations
dorf, he finds a more limited account of the sovereign right of death as con- among putatively self-governing citizens, as Foucault tends to stress, but the
ditioned by the defense of the sovereign. The end of sovereignty is, however, formation of and relations among what aspire to be self-governing political
the continuation of sovereignty itsel5 it is caught in a kind of "self-referring communities . One of the features of the modern political world, which may
circularity" (Foucault 1gg1: gg). Thus, Foucault argues that, ifwe take Pufen- be dated from the agreements of Westphalia concluded in 1648 following the
dorf's definition of the end of sovereign authority as "public utility" and seek Thirty Years War, is that these fictive self-governing political communities
to define the content of public utility, we find little more iban that subjects have come to be represented as independent states. That is, they are political
obey laws, fulfill their expected tasks, and respect the political order. unities with definite territorial boundaries, secured by the principie of non-

48 STATES OF IMAGINATION
"DEMONIC SOCIETIES" 49
 norms thar lead to rhe ereation of a coordinated and central ized administra-
interferente of one sovereign stare in rhe interna ) affairs of another Claims
tion of life need to be weighed against rhe norms of economic processes and
to sovereignry by such communities llave thus become identical with elaims
rhe freedoms on which they depend and rhe norms derived from tire sover-
te he a state "1 he ciry - citizen game therefore concerns rhe panoply of reeh- eign subject of rights. This is why, for liberalism, rhe problem will not he a
niques by which tire members of,a population are formed or form rhemselves
rejecrion of biopolitical rqgulátiombut a wáv of man.^gingit.
moto a political community, and by which they seck ro exercise sovereignry. It
Liberal government is a particular form of articulation of rhe shepherd-
also includes rhe arts of international government by which certain popula-
Hock game and rhe ciry-citizen game. Ir assembles a pastoral powt=t lattákes
ti ons are assigned ro [hese nominally independent sovereign atares and that ni c forms f aa biopolitics of the adminisrration of 1ife and a form of sover-
regulare rhe coexistente of states with one another. eignry that deploys the law and rights to limit, tú offer guarantees, te make
The relation of rhe arts of governing and sovereignry is not the replacement
sáPe and aboye all to justify rhe erations of biopolitical programs and
of one by the other but each acting as a condition of rhe other. On rhe one
disciplinarv prac[ices. Liberalism, however, can never fully check rhe ' de-
hand, the existente of nominally independent sovereign states is a condition
monic" possibilities contained within Chis mix, as recent revelations about
of foreing open those geopolitical spaces on which rhe arts of government
rhe way in which liberal - democratie states ( like [hose in Scandinavia ) have, in
can operare. On rhe other hand, a set of supranational agreements and regu-
rhe course of rhe rwentieth cenmry, practiced forced sterilization in rhe narre
lations of populations is a necessary condition of the world inhabited bythese
of a eugenic utopia on certain of its populations . Even more pervasive has
sovereign states.
been rhe tendency within certain national states ( Australia , Canada), having
We are now in a position to locate rhe third term of out triad, liberalism.
ceased to attempt actual genocide, te commit forms of cultural genocide on
As Foucault puts it in regard to biopolitical problems: "`Liberalism' enters
indigenous populations within their borders in rhe narre of their own we11-
rhe picture here, because ir was in connection with liberalism that they began
being, such as in rhe case of rhe removal of children from their parents and
to have rhe look of a challenge. In a system anxious to have rhe respect of
families. Although rhe biopolitical imperative does not account for all that
legal subjects and to ensure rhe free enterprise of individuals, how can rhe
bedevils liberal - democratic states, it is remarkable how much of what is done
`population' phenomenon, with its specific effects and problems, be taken
of an illiberal character is done with rhe bes[ of biopolitical intentions.
roto account? On behalf of what, and according ro what rules, can it be man-
But why is this so? First , note, following Mariana Valverde ( 1996), that lib-
aged?" (1997a: 73). eralism has itself never been entirely identical with irsel£ Valverde shows how
According to Foucault, liberalism can be understood as a form of critique
rhe liberal conception of rhe juridical and political subject has a form of ethi-
of excessive government. It should be approached, however, as a critique not
cal despotism at its core , contained in notions of rhe possibiliry of improve-
only of earlier forms of government such as police and reason of state butof
ment and habit . The history of liberalism as an art of government shows
exisring and porential forms of biopolitical government. That is, liberalism
(L^^ how a range of illiberal techniques can be applied te [hose individuals and
e that rhe government of the processes of hfe
p hita) opulations who are deemed capable of improvement and of attaining self-
might take. It might criticize [hose forms , for example, in which bio ol government (from women and children to certain classes of criminals and
^
norms will be comprornised by a lack of understanding of economic norms.
paupers). Moreover , asa form of colonial governmentaliry , liberalism canjus-
It might also criticize rhe detailed regulation of tr^biological propesses of,
'r aurhoritarian ",pes of government for [hose regions deemed unimproved,
the soos, and the tendencies toward tate racisUj found in biouolitics, by z 1/ "1'[ify
ro use John Stuart Mill's judg-
✓c,^^,^lifrica, or degenerare and static , like China ,
au appeal to rheframework of right-either legal or natural - that it will f k an} advance
ment. For such nations, their muy hopeo ma mg s te p s m
codify as rhe theory and practice of demoerarie sovereignry . If liberalism
SG 1 depends on rhe chances of a good despot" (cited in Valverde 1986: 361).
emerged less as a doctrine or form of rhe minimal state [han as an ethos of
Second, and we might think fortunately, things have not quite mrned out
review {. Chis ethos needs to be situated in rhe rationalization of rhe field of rhe way Mill envisaged. The appeal to rights within rhe democratized frame-
bion lirlcaproblems . Por liberalism , ir is always necessary to suspect that
work of sovereignry has proved a resource thar has enabled some degree of
one is governing roo much. This is because rhe impcratives of biopolitical

"DEMONIC SOCIETIES" 51
50 STATES OF IMAGINATION
 success on the part of political movements in liberal democracies such as
those of other states. Does this mean that the forro of government of such
those of wornen , colonized peoples, racial and ethnic minorities ,
and people
states is assembled from elements that are radically different from the ones
with disabilities . However, although rights daims within sovereignty dis-
we have discussed here? Does this mean that state socialism and National
courses might enhance and protect the position of certain groups, others
Socialism , for example , cannot be subject to an analysis of the arts of govern-
have not been so fortunate. This is especially so of those whose political iden-
ment? The answer to both [hese questions , 1 believe, is no. The general ar-
tity is defined as without citizenship, state, or even means of subsistente,
gument of this essay is that the exercise nf_onvernment in all modern states
that is, merely by the fact that they are living. This would appear to apply to
entails the articulation of a form of pastoral power with one of sovereign
prisoners , ¡llega¡ immigrants , and even today unemployed people, al) groups
pmves`iberalism, as we have just seen, makes that articulation in a specific
defined by bare lile.
way Other types of role have a no lees distinc [ive response to the combination
Finally, although liberalism may try to make sale the biopolitical imperative
of elements of a biopolitics concerned with the detailed administration of life
of the optimization of lile, it has shown itself permanently incapable of ar-
and s rthat reserves . . the zi Df_ th to itself
resting-from eugenics to contemporary genetics-the emergente of ratio-
Consider again the contrastive tercos in which it is possible to view bio-
nalities that make the optimization of the life of some dependent on the dis-
'^^ politics and sovereignty. The final chapter in the first volume of the History of
allowin h?-rs. 1 can only suggest some general reasons for
Sexuality that contrasta sovereignty and biopolitics is titled " Right of Death
t is. Lbberalism is fundamentally concerned to govern through what it con-
and Power over Life." The initial tercos of the contrast between the two reg-
ceives as processes that are external to the sphere of government limited by
isters of government is thus between one that could employ power to put
the respect for the rights and liberties of individual subjects. Liberal rule thus
subjects to death , even if chis right to kill was conditioned by the defense of
fosters forros of knowledge of vital processes and seeks to govern through
the sovereign , and one that was concerned with the fostering of lile. Never-
their application . Moreover, to the extent that liberalism dependaon th€for-
theless, each part of the contrast can be further broken down . The right of
mation of responsible and autonomous subjects through biopolitics and dis-
death can also be understood as "the right to take life or let live"; the power
cipline, it f t rs thc f governmental practices that are the ground of
over life as the power "to foster life or disallow it." Sovereign power is a
such r ' almes . Further, and perhaps most simply ,
we might consider che
power that distinguishes between political lile (bios) and mere existence or
possibility that sovereignty and biopolitics are so heterogeneous to one an-
bare life (zoe). Bate life is included in n
other that the derivation of political norms from the democratization of the
iI`very
s exclusion Eropolitical life. In contrast , biopolitics might be thought
former cannot act as a prophylactic for the possible outcomes of the latter.
to include zoe in bios: stripped down mere existence becomes a matter of
We might also consider the alternative to this thesis , that biopolitics captures
political reality. Thus, the contrast between biopolitics and sovereignty is nor
and expands the division between political life and mere existence , already
one of a power of life versus a power of death but concerns the way the
found within sovereignty. In either case, the framework of right and law can
different forros of power treat matters of lile and death and entail different
act as a resource for (orces engaged in contestation of the effects of biopower;
conceptions of life. Thus , biopolitics reinscribes the earlier right of death and
it cannot provide a guarantee as the ef & cacy of such struggle and may even be
power over lile and places it within a new and different forro that attempts to
the means for the consolidation of those effects.
include what had earlier been sacred and taboo , bare life, in political exis-
tence. It is no longer so much the right of the sovereign to put to death his
Sovereignty and Biopolitics in Nonliberal Pule enemies but to disqualify the life-the mere existente - of those who are a
1 threat to the life of the population , to disallow those deemed "unworthy of
There are , of course , plenty of examples of the exercise of sovereignty in the
li£e," those whose bare lile is not worth living.
twentieth century that have practiced a decidedly nonliberal forro and pro-
This allows us, first, to consider what might be thought of as the dark side
grato of national government both in relation to their own populations and
of biopolitics ( Foucault 1979a : 136-37). In Foucault's account, biopolitics

52 STATES OF IMAGINATION
"DEMONIC SOCIETIES" 53
 notion and teclin iq ues of population had given risa , at rhe end of rhe nine-
does flor put an end ro the practice of war: ir provides it with new and more
teenth century, to a new linkage among population , rhe interna ¡ organizatiou
sophisticated killing machines. There machines allow killing itself to be re-
of stares, and the competition berween states . Darwinism, as an imperial so-
posed ut rhe leve l of catire populations. Wars becoine genocidal in the twen-
cial and political program, would plot thc ranking of individuals, popula-
tieth century. The same state thar cakes on rhe duty to enhance the life of rhe
tions, and nations along the common gradicnt of fltness and thus mensure
population al so exercises rhe power of death over whole populations. Atoinie
efficiency.° However, rhe series "population , evolution, and race" is not simply
weapons are rhe key weapons of chis process of thc power ro put whole pop-
a way of thinking about rhe superioriry of rhe "white races" or of justifying
ulations to death. We mighr also consider here rhe aptly named biological
colonialism, but also of thinking about how to trear rhe degenerares and rhe
and chemical weapons that seck an extermination of populations by visiting
abnormals in one's own population and preven [ rhe further degeneration of
plagues upon them or polluting rhe biosphere in which they live to rhe point
rhe race.
at which bate life is no longer sustainable. Nor does the birth of biopolitics
The second and most importan [ funetion for Foucault of chis biopolitical
put an end to the killing of one's own populations. Rather, it inrensifies that racism in the nineteenth cenmry is that "it establishes a positivo relation be-
killing-whether by an "ethnic cleansing" that visits holocausts upon whole
tween rhe right to kill and rhe assurance of life" (Stoler 1995 : 84). The life of
groups or by the mass slaughters of classes and groups conducted in rhe
rhe population , its vigor, irs health , its eapaei ries to survive , becomes neces-
narre of rhe utopia to be achieved. sarily linked to rhe elimination of inrernal and exrernal threats. This power to
There is a certain restraint in sovereign power. The right of death is only disallow life is perhaps bes[ encapsulated in the injunctions of rhe eugenic
occasionally exercised as rhe right ro kill and then often in a ritual fashion
project: identify [ hose who are degenerare , abnormal, feeble-minded, or of
that suggests a relation to the sacred. More often, sovereign power is manifest an inferior race and subject them to forced sterilization; encourage those who
in rhe refraining from rhe right to kill. The biopolitical imperative knows no
are superior, fit, and intelligent to propagare . Identify [ hose whose life is but
such restraint. Power is exercised at rhe level of populations and hence wars
mere existente and disqualify their propagation; encourage [hose who can
will be waged at that level, en behalf of everyone and their lives. This polar partake of a sovereign existence and of moral and political life. But this las[
brings os to rhe heart of Foucault's provocativa thesis about biopolitics that example does not necessarily establish a positive justification for the right to
[here is an intimare connection berween the exercise of a life-administering
kill, only rhe right to disallow life.
power and rhe commission of genocide: "lf genocide is indeed rhe dream of
If we are to begin to understand rhe type of racism engaged in by Nazism,
modero powers, chis is not because of a recent return of rhe ancient right to
however, we need to take roto account another kind of denouement berween
ki11: it is because power is situated and exercised at the Ievel of life, rhe spe- the biopolitical management of population and the exercise of sovereignry.
cies, rhe race, and rhe Iarge-scale phenomena of population" (1979a: 137). This version of sovereignry is no longer the transformed and democratized
Foucault completes chis same passage with an expression that deserves more foro founded en rhe liberty of the juridical subject, as it is for liberalism, but
notiee: "massacres become vital." a sovereignry that Cakes up and transforms a further element of sovereignry,
There is thus a kind of perversa homogeneity berween rhe power over life
its "symbolics of blood" ( Foucault 1979a: 148).
and the power to cake life charaeteristic of biopower. The emergence of
For Foucault , sovereignry is grounded in blood-as a reality and as a sym-
a biopolitical racism in che nineteenth and twentieth centuries can be ap- bol-just as one might say that sexualiry becomes the key field on which
proached as a trajectory in which Chis homogeneity always threatened to tip
biopolitical management of populations is articulated. When power is exer-
over roto a dreadful necessity. This racism can be approached as a fundamen-
cised through repression and deduction , through a law over which hangs rhe
tal mechanism of power that is inscribed in rhe biopolitical domain (Stoler
sword, when it is exercised en the scaffold by rhe torturer and the executioner,
1995: 84-85). For Foucault, rhe primary function of this form of racism is to
and when relations berween households and families were forged through
establish a division berween [hose who must live and [hose who must die,
alliance, " blood was a reality with a symbolic function." By contrast , for bio-
and to distinguish [he superior from rhe inferior, rhe fir from rhe unfit. The

"DEMONIC SOCIETIES" 55
54. STATES OF 1 MAGI NATION
 politics with its themes of health, vigor,
fitness, vitality, progeny, survival, and
yace, "power spoke of sexuality and to sexuality" (Foucault 197ga: 147). that of the fundamental role of the human sciences in the atrocities of that
For Foucault ( 1979a: 149 - regime (Peters lggg ). The late Detlev Peukert drew on studies of psychiatry
50), the novelty of National Socialism was che
way it articulated " the oneiric exaltation of blood," of fatherland under National Socialism , the history of compulsory sterilization programs,
, and of the
genetics , eugenics , medicine , social policy, and education , and his own work
triumph of che race in an immensely cynical and naive fashion ,
with the par-
on social - welfare education , to argue that "what was new about'Final Solu-
oxysms of a disciplinary and biopolitical power concerned with the detailed
tion' in world-historical tercos was the fact that ir resulted from the fatal racist
administration of che life of the population and the regulation of sexuality,
family, marriage , and education .' dynamism present within the human and social sciences" (1993: 236). Again
Nazism generalized biopower without the
limit-critique posed by the juridical subject of right, but it could not do away we witness a fundamental division of the population , on this occasion made
with sovereignty. on a particular qualitative distinction between "value " and "nonvalue" and a
Instead, it established a set of permanent interventions
finto the conduct of the individual within the population and articulated chis treatment of the Volkskórper or body of the nation that consisted in "selection"
and "eradication ." Peukert argues that twentieth-century medical and human
with the "mythical concern for blood and the triumph of the race." Thus,
the shepherd-flock game and the city-citizen game are transmuted into the sciences are confronted by what he calls a "logodicy " that tríes to resolve che
eugenic ordering of biological existente ( of mere living and subsistente) dilemma between the rationalist dream of the perfectibility of humankind and
the empirical existente of human finitude , of illness, suffering , and death.
and articulated on the themes of the purity of blood and the myth of the
fatherland. One resolution of chis dilemma is the projection of the rationalist project
away from the finite individual onto a potencial immortal body. In the German
In such an articulation of these elements of sovereign and biopolitical
forms of power , case, what Foucault called the species body of the population is mapped onto
the relation between the administration of life and the right
to kill entire populations is no longer simply one of a dreadful homogeneity. the body of che Volk or race. The biopolitical imperarive is rearticulated with a
It has become a necessary relation . The administration of life comes to te- kind of mythicized version of sovereignty. Like Foucault , Peukert argues that
quite a bloodbath . It is not simply that power, the logic of Nacional Socialism, with its concern for the nurture and improve-
and therefore war, will be
exercised at the leve¡ of an entire population . ment of the immortal Volksk ó rper, had a double significance : heroic death
It is that the act of disqualifying
the right to life of other yaces becomes necessar^for che fostering of the life on one side and eradication on the other (242).
of=ame Moreover, the elimination of other yaces is only one Pace of`t ee National Socialism is one contingent, historical trajectory of the develop-
purification of one's own race ( Foucault 1997b : ment of the biopolitical dimension of the social , medical, psychological, and
231). The other part is to
expose the latter to a universal and absolute danger, to expose it to the risk of human sciences that occurs under a particular set of historical circumstances.
death and total destruction . For Foucault , One should not underestimate the factors operative in German society, the
with the Nazi state we have an "ab-
solutely racist state, an absolutely murderous state andan absolutely suicidal historical legacy of war and revolutionary movements, the nature of German
state" ( 232), all of which are superimposed and converge on the Final Solu- polity, or the economic crises of the early twentieth century. Nevertheless,
tion. With the Final Solution , the state tries to eliminare Peukert and Foucault would both agree that the kind of state racism practiced
, through the Jews,
all the other races, for whom the Jews were the symbol and the manifestation. by the Nazis that would lead to che Final Solution was quite different from
This includes , in one of Hitler' traditional anti-Semitism insofar as it took the forro of a "biological politics,"
s last acts, the orden to destroy the bases of
bate life for the German people itself. " as che German historians call it, that drew on the full resources of the human,
Final Solution for other races, the
absolute suicide of the German race" is inscribed , aecording to Foucault, in social, and behavioral sciences.
the functioning of the modern state (232). In this regard, Peukert's retrieval of the process by which the human sci-

Foucault's analysis of the political rationality of Nacional Socialism finds ences move from a concern with "mass well-being" to acting as the instru-
ment of "mass annihilation " remains extremely interesting. In the case of
confirmation in the work of recent German historians en at least one point,
"social-welfare education," he identifies a number of phases ( 1993: 243-

56 STATES OF IMAGINATION

"DEMONIO SOCIETIES" 57
45). l;irst, there was thc formulation of the problem of thc control of youth caro of che needy roto a technology of mass an nihilation. However, given that
in che late nineteemh century wirhin a progressivist discourse in which every many, if not all, the forros- oí` knowledge and tech nologies oí govermnent
child liad a righr to physical, mental, and social Etness. This was followed by (including the coneentration camp) were che produet of politics characterized
a phase oí rourinization and a crisis of confidence exemplified by che failurc at leas[ broadly by liberal tbrms oí rule, Chis docs suggest thcre is no room
of legal schemes oí detenrion or protectioa of [hose who were "unfit" or for eomplaceney and rhat che liberal critique of biopolitics cannot offer thc
°ineducable." Flic rhird phase, coinciding with che final years of the Weimar kind ofguazantecsit claims to. Foucault is right to provoke os with che Idea
Republic has disturbing overtones for our own period. Herc diere were a that che assurance of life is connected with the death eommand and to claim
series of scandals in young people's homes and a debate about che limas of thar "che coexistente in political srruetures of large destructiva mechanisms -
edueability coupled with welfare stare retrenchment. This debate introduced and institutions oriented toward che tare of individual life is something

a new eost- benefits trade -off wiúl services allocated on che basis of immedi- puzzling and needs some investigation" (1988b: 147). Mass slaughters may

ate return , and che criterion of "value" was brought into the calculative frame- ot necessarilyor logically follow from the forros of political rationality and

work. Value at chis stage may or may not be determined on the basis of race pes of knowledge we employ, but they do not arise from a sphere that is

or genetics , but the ineducable were excluded in 1932 from reform school opposed to that rationality and knowledge. It is crucial to realize, as Peukert

educarion . Afrer 1933 [hose who opposed the racial version of determining argues in his book Inside Nazi Germany, that racism was a social policy, rhat is,
value were foreed inm silente, compulsory sterilization of che genetically un a policy that was concerned with the elimination of all those who deviated
healthy was practiced, and concentration camps for the racially inferior estab I. from an ever more detailed set of norms and che reshaping of society roto a
lished. However, even Chis program faced a crisis of confidence and the uto{ "people of German blood and Nordic cace; four-square in body and soul"
pian goals carne up against their limits and the catalogue of deviance became (q8$: ao8).
greater and more detailed. The positiva racism of youth welfare provision ( What Peukert cannot address is the rationality of what he conceives as che
now met the negative radicalization of a policy of eradication of those who, irracional component of Nazism. Although he understands che role of che hu-
in che language of the order that represents the crucial step in the Final So- man sciences in che formation of Nazi biological politics, he tends to consign
lution, are deemed "unworthy of life" (lebensunwertes Leben). The biopolitical the themes of blood, race, and Volk to an irracional sublimarion contained
government of life had arrived at the point at which it decided who was worth wirhin them rather than viewing them, as Foucault does, as rearticulated. elt=
living. With the technology o£ morder up and running, the social and human ments of sovereign power. Th's brings us to che central distinctiveness of
sciences "are engaged in a parallel process of theoretical and institutional Foucau t's comments . National Socialism is not regarded as the pinnacle of
generalization that is aimed at an all-embracing racist restructuring of social che total administration of life undertaken with che help of che human sci-
policy, of educational policy, and health and welfare policy" (Peukert 1993: ences and biopolitical technologies , as it mighr be by the Frankfurr School
245). The terin Gemeínschaftsfremde (community alien) carne to embrace fail- and their descendants. The key point for Foucault is that Nacional Socialism
ures, ne'er-do-wells, parasites, good-fiar-nothings, troublemakers, and those is regarded as a articular artieularion of specihc elements of biopolitics and
with criminal tendencies and threatened all [hese with detention, imprison- its knowledge of populations and individuals and sovereignry. Iris not simply

ment, or death. che logic of che bureaucratic application of che human sciences that is at issue
The phrase "those unworthy of life" is striking because it so clearly reso- b [ th ^r"may efiycial discourse wirhin a biopolitics of the population
nares with the biopolitical attempt to govern life. It suggests a distincúon and its linkage with themes of sovereign identiry, autono , and political
between diose who are merely living and [hose who are worthy of existente communiry. This forro of sovereignry has been drained of all its potential to
as a parí of a social or political community. We should be clear that there was claim and protect rights by che removal, following Bauman (1989: up, of all
nothing necessary in the path of National Socialism and that there were cru- counterbalancing resourceful and influential social forces.' A political dis-
cial steps of che conversion of knowledge and services concerned with che course that divides populations on che basis of race has certain fairly obvious

58 STATES OF IMAGINATION "DEMONIO SOCIETIES" 59

 political dangers . However, one that makes the welfare and life ola racialized
gades, or street or lane committees , and their subgroups (474-75). The lat-
population the basis for nacional sovereignty and political community could ter are responsible for "one hundred or so households ... to keep records
be viewed as more clearly " demonic."
of their family plans, contraceptive use and monthly cycles " ( 475). This de-
Unforrunately, this story of biopolitical racism does not end with Nazism.
tailed chain of command uses such instruments as local meetings of mar-
Foucault also insists that the possibilities of state racism are found in many
ried couples, certificates of permission to become pregnant, allowances for
versions of the articulation of biopolitics and sovereignty , including many
couples of single children , and harsh fines for those conceiving without a
varieties of socialism ( 1997b: 233 34; Stoler 1995: 96-97
- ). For Foucault, the certificate and those with more than one child.
problem with socialism is that it has a kind of state racism inscribed in its The Chinese policy thus inscribes sovereign elements (of decree , interdic-
premises and that, even if it has sometimes criticized biopower, it has not tion, punishment , and reward ) within a detailed biopolitical intervention into
reexamined the foundations and modes of functioning of racism . When so- the intimate lives of its population . It does chis not in the narre of the father-
cialism analyzes its own emergence as a result of economic transformation, land, blood, and racial purity, but in terms of the targets envisaged by the
it does not have need for an immediate recourse to these racist motifs. When plan. On one point, it is clear that Chinese policy is nonliberal in that it does
it insists en the necessity of struggle to socialist transformation , a struggle not rely on the choices , aspirations , or capacities of the individual subject.
that is against the enemies within the capitalist state, Foucault argues, it Ibis does not stop it having some similarity with early liberal policies, par-
necessarily revives the theme of racism . Moreover, when socialism takes
ticularly Malthusian - informed poor policies. In both cases, the process of
upon itself the task of managing, multiplying , and fostering lile, of limiting
economic liberalization and the recommendation of prudential procreation
chances and risks, and governing biological processes , it ends up practicing
are linked. One tries to privatize the costs of imprudent propagation onto
a form of racism that is not properly ethnic but evolutionary and biological.° individuals, families, and their offspring ; the other tries to prevent burdens
The enemies within on which this racism will be practiced are the mentally en the developmental dreams of the socialist state.
ill, the criminal , political adversaries , and-with ,
say, China' s one-child pol- The study of governmentality has yet to open up the extensive discussion
icy-imprudent parents and their potential offspring ( Sigley 1996 ). In the last
of authoritarian and nonliberal governmentality . Foucault's analysis of Na-
case, we find a form of government that combines market - based norms and
tional Socialism is a striking contribution to this problem for a number of
biopolitical interventions into the intimare lile of the population in a non- reasons. First, it shows that this case of what might be thought of as, to put
liberal manner in order to realize the objective of the quantity and quality of it mildly, a nonliberal or authoritarian form of rule is composed , like liberal
the population necessary for the socialist plan.
rule, of biopolitical and sovereign elements. It also places National Socialism,
This kind of evolutionary and eugenic racism is one that can be practiced like liberalism , within the development of a government of biopolitical pro-
against one ' s own population in the narre of optimizing its quantity and qual-
cesses. This does not mean that we should efface the differences between
ity. Thus, the Chinese government " claims that , not only is it possible to
liberal and nonliberal rule. Nor is this analysis an attempt to undermine criti-
know in detail the object to be governed , but, further, it is possible to predict
cal distinctions between such forms. What it does illustrate, however, are the
the precise outcome of any possible intervention " ( Sigley 1996 :
473). Ibis dangers inherent in biopolitical rule and in the articulation of the shepherd-
kind of rule is nonliberal in that , first, it does not use any version of the liberal
fiock and city-citizen games that Foucault held as central to modere politics.
subject to limit or try to offer guarantees for or make sale population pro- The continuities between authoritarian and liberal governmentality , together
grams, and, second, it seeks to unite market -
style economic norms with bio- with the recovery of the illiberal components of liberalism , remind os of the
political ones . Thus, national authorities assign numerical targets, provincial
dangers of not calling into question the self-understanding of liberalism as a
authorities translate these into birth quotas that are distributed among pre-
limited government acting through a knowledge of the processes of life, yet,
fectures and counties , which in turn divide the quota among communes,
at the same time, safeguarding the rights of the political and juridical subject.
townships , and so on. The policy is implemented at the microlevel of bri-
There is no necessity that means that our most general rationalities of rule

6o STATES OF IMAGINATION
"DEMONIC SOCIETIES" 61
 [his will often mean despotic provision for decir special needs with the ama
such as sovereignry and biopolitics will in el ti ctably lead to rhe truly demonio
of rendering rhem autonomous by fostering capacities of responsibility and
eventualities wc have continued to witness right lo thc beginning of rhe
self-governance. Under certain conditions, however, frustrations with sucia
nvenev-flrsr centurv. Nor, however, is there any guarantee that the appeal te
programs of improvemenr ,ay lead te fmrms of knowledge and political ra-
rights within liberal democracies and the international eommuníty of states
will guard against such eventualities, as rhe contcmporary confinement of tionality that identify certain groups as without value and beyond improve-

illegal immigrants in camps in liberal democracies attests. Elements within ment, as those who are merely living, whose existence is hut zoe. Liberal

sovereignry and biopolitics will continue to provide resources for political ra- regimes of government can thus slide froni rhe "good despot" fbr rhe ini-
provable to sovereign interventions te confine, tu contain, te coerce, and ro
tiomdiry and action in Weber's sense of the attempt to influence rhe govern-
eliminare, if only by prevention, [hose deemed without value. It is trae, per-
ment of organizations. But there can be no system of safeguards that offer us
haps, thar many of out worst nightmares tend to be realized when these ele-
a zone of comfort when we engage in political action. When we do so, Fou-
cault's position here seems to suggest, we enser a zone of uncertainty and ments of sovereignry and biopolitical rule are articulated somewhat differ-
endy from the way they are in liberal democracies today. This should offer no
danger because of the governmental resources we have at our disposal. We
reason for complacency even for those who find themselves marked as rhe
might add that rhe price of not engaging in political action is equally great, if
not greater. A condition of informed political action remains an analysis of mature subjeets within rhe boundaries contained by liberal-demoeratic con-

rhe acrors involved, the contexts of their action, the resources at hand, rhe stitutionalism, les alone those who curremly remain in need of a good despot

tactics used, and the ends sought. Though handling this relation berween within and outside [hese boundaries. It offers even less room for complacency

biopolitics and sovereignry remains tricky, we must establish an analysis of for [hose who find themselves occupying rhe position of rhe good despot.

rhe way an implementation of programs of rhe administration of tire opens

fresh arenas of contestation, negotiation, and redefinition around citizen-
ship, democracy, and rights. We must also be prepared to admit, nevertheless, Notes
that the appeal to rights might link this form of contestation to the powers it 1. Por rhe most succincr and instructive summary of his redections on che arts of government in
contests, particularly when such an appeal concerns rhe rights of [hose with- English, see Serurity , Territory , Population and The Birth ef Bio - Politics, che summaries of his 1978
out any status bus their mere existence. and 1979 courses at the Colige de France in Foucault (1997re 66-79). See also his justly

neral
The more ge araum nt ad B h epe ° 'n^^ á°rra p'Gtics ni rst famous lecture un rhe development of early modero discussion of government and his intro-
duction of the concept of govemmentaliry (1991).
combine rhe resourees gF h' I+°^°"t -ter=, t'^ .cánon,
z. 1 would draw attendon ro three important papers in Chis area by Hindess (1998), Lui-Bright
and sexuali e a°a rc Iogjc of overeienrv based en right terriWry
(1997 ), and Dillon (1995 ) that have opened up chis question ofthe relations berween interna-
death, and blood. Moreover, this biopolitics captures lile stripped naked (or
tional government anda system of sovereign states.
the zoe that was the exception of sovereign power) and makes it a matter of 3. The 1976 lectures un war, race, and rhe modero suite have been brought to our attention by
political lile (bios). It follows that, given that we continue ro live in a system Ann Laura Stoler in Roce and the Eduration of Desire (1995).
of modero states, we must face up to forms of biopolitical racism, that is, a 4. An excellent example of [his is che exemplary genealogy of rhe language of modero constitu-
tionalism provided by James Tully (1995) froni che perspective afforded by che polirics ofcul-
racism that follows not simply from discrimination, scapegoating, or insti-
rural diversity and rhe resourees of che alternative tradinons of common constirutionalism.
tutions, but from the elements by which we are compelled to rhink about and
Tully draws un tire work of che larer Witcgenstein to challenge die assumpnons of modern
imagine states and their populations and seek to govern them. This is as constitutionalism. 1 should also mention Pat O'Malley's (1998) study of the articularion of
true for liberal arts of govermnent as for nonliberal rule. The liberal arts programs of government of aboriginal people in western Australia and ferina oE indigenous
of governing through freedom mearas that liberalism always contains a divi- governance.
sien berween [hose capable or deserving of rhe responsibilities and free- 5. Por an extended discussion of notions of "authoritarian governmentaliry" asa framework for
understanding Indonesian pohtics, see Philpott (1997, especially chap. 4).
doms of mature citizenship and those who are nos. For [hose who are nos,

62 STATES OF IMAGINA'TION "DEMONIC SOCIETIES" 63

 6. On changing conceptions of population at the end of the nineteenth cenmry and their relarion
to Darwinism , social reform , and eugenics ,
see Rose, The Psycholagiml Compla ( 1985).
7. For an admirably documented account, see Pine, Nazi Family Policy GOVERNING POPULATION The Integrated Child
(1997)-
8. 1 hace not the space here to discuss the similariries and differences berween Zygmunt Bau- Development Services Program in India
man's (1989) importan[ account of the Holocaust and the present undersranding of the spe-
cific character of Nazi tacism as a rationality of extermination. Bauman's account concurs
wirh the one presented here insofar as ir
Akhil Gupta
presents the Holocaust as something that trust be
understood as endogenous m Western civilization and its processes of rationalization rather
than as an aberrant psychological, social, or political pathology. Moreover, to the extenr that
bis account stresses the collapse or nonemergence of democracy, it indicares the failure of
the democratization of sovereignty as a fundamental precondition of Nazi rule, a theme that
echoes those of Hannah Arendt's famous book, The Origins ofTotalitarianism (1958). Foucault's
brief remarks seem to add co or qualify Bauman's arcount in two ways. They first offer the
possibility of a closer specification of the kind of rationality and technology that make pos-
sible a racialized politics and poliey by demonstrating its biopolitical character. Such
a view
enables as to get a clearer undersranding ofthe role of the human sciences in such a politics.
They also suggest that Nazi politics articulares this biological politics with alternarive tradi-
tions and frameworks of sovereignty such as those of fatherland, Volk, and blood. It is not I remrned to Alipur village in western Uttar Pradesh for a short spell of field-
simply the imperative of the totalistic administration oflife that
accounts for the mentality of work in the summer of 1989. Villagers excitedly told me about all the changes
Nazi rule, but the way rhe biopolitical discourses and sovereign themes are reinscribed and
modified within one another.
that had occurred in the four years since I had last been there doing a study
on agricultural change. Among the new features of Alipur were an anganwadí,
g. Note that Foucault's point here extends one made by Hannah Arendt. ' Practically speaking,"
she states, "it will make little difference whether totalitarian movements adopt the partero of a center that provided day cace, nutrition, and inoculations to children and
Nazism or Bolshevism , supplementary nutrition and health care to pregnant women and mothers of
organize the masses in the narre of race or class, pretend to follow
the laws of life and namre or of dialectics and economics" (1958: 313). Biopolitical sute infants. I was so intrigued by Chis new development that in my second field-
racism can be justified in terms of the goal of the evolution to an ideal society or optimizing work project on the ethnography of the state I included the Integrated Child
the quality ofthe population as much as the evolution of the
race. The practice ofstate racism Development Services (ICDS), more popularly known as the anganwadi pro-
en populadons retarding social and polircal evolution does not necessarily always speak the
language of race. gram, among the development projects that 1 studied.
The ICDS program is interesting for a number of reasons. It is one of the
fastest growing development programs run by the Indian state. Launched
with only 33 projects in 1985, the ICDS program had expanded co 1,356 proj-
ects in the next ten years, and to 5,614 by 1999 (Government of India 1985:
4). The anganwadi program grew even when the Indian government started
cutting the budgets of other social welfare programs in the postliberalization
era initiated in Iggi. Allocations for ICDS in 1998-1999 went up co more than
twice their 1990-1991 levels.' Another reason for paying attention to che ICDS
program is that it was one of the first interventions that attempted to control
population growth yates by paying attention to the quality of the population.
It thus provides os with one of the first large-scale examples of the kind of
approach to population planning that, alter the Cairo Conference, has at-
tained the status of ofhcial dogma. Theoretically, the Ices program helps

64 STATES OF IMAGINATION
 sorne of her reactions, as well as rhe reactions of rhe anganwadi workers who
provide us with a nearly perfect example of thc regulation, care, and docu-
are subjected lo these visits. 'Fhe section thar follows shifts attention away
tuentatiou of rhe po pularion, especially [hose members of the population
from surveillance lo enumeration, one of rhe most important instruments of
(women and children) who are poorly represenred in otficial staristies. Such
government. 1 show what kinds of data are collected by anganwadi workers
attention ro rhe welfare of [he popularion is a fono of biopower, one of the
and what happens ro [hese data once they are passed up rhe hierarchy. Finally,
hallmarks of "govermnentaliry," a term introduced by Michel Foucault (r99r)
1 consider rhe question of resistance: how ir is te be understood within the
to analyze modes of government thar are nor necessarily parí of the state
governmentaliry literature and what forro ir takes in rhe case ot the anganwadi
apparams.
The enlargenlenr of rhe scope ofgovernmental regulation and concern rep- program.

resented by the Ices program produces new kinds of subjects and new kinds
of resistances. Among the new subjects produced are bureaucrats whose job Governmentality
is to focus on the weak and disempowered mem ers (women and chi ren)
(r99r) in which
of the weakest and most disempowered groups within rural north Indian so- The term "governmentaliry " comes from a lecture by Foucault

ciery (lower and scheduled castes); anganwadi workers, who are implemen- he drew attention r o all the processes by hich the conduct of a population
the state ; b/discourses,
ters of the reos project but also t emse ves en ' ries of state benevo- is governed : by institut' ns and agencies/including
norms, and identities aud by self- regulation , teehniques for the disciplining
lence; and, finally, rhe poor women and children who find themselves objects
and cace of the self: olitical economy as knowledge and apparatuses of se-
of state attention and discipline but also beneficiarles o supp ementary nu-
curiry as technical means operate en the population as a target ro constitute
trition, educational services, and health este that may not be available lo their
/gO4ernmentality as the dominant mode of power since the eighteenth Gen-
economically better-off rural neighbors. These new subjects are placed in
han attempt to exhaustively define governmentality, a job
structurally dependent but antagonistic positions. Bureaucrats attempt lo dis- . tury (ron). Rather [
(ir that has been well done by others , especially Mitchell Dean in Governmentality
cipline and control the anganwadi workers, who, in turn, attempt to disci-
pline and control their rural charges. Those in subordinate positions, in mm, j , 999), and by others such as Nikol , crer Mil lee and Nikolas
Shearling (1997), 1 wish
resist rhe mechanisms of surveillance that are employed en them. What do y,^c ` , _rY Rose (199o ), Pat O'Malley , Lorna Weir , and Clifford
phasize a few features of governmenta li ry.
t emphasize
such acts of resistance by the women who are the objects of state surveillance,
Foucault argues that since the eighteenth century, population became rhe
protecrion, and investment mean for an understanding of governmentality?
object of sovereign power and discipline in a new way, so that rhe growth of
And how does one tic the ethnographic analysis of everyday practices of re-
the welfare of rhe population within a given territory, [he optimization of its
sistance to a structural understanding of inequalities ofgender and class?The
capabilities and productivity, became the goal of government (r99r foo-1o1).
analysis that 1 develop in rhe rest of this essay pursues these quesrions in
The goal of "good government" became no[ simply rhe exercise of authority
greater detail.
over the people within a territory or rhe ability lo discipline and regulare them,
The section that follows attempts to lay out some of the more general fea-
but fostering their prosperity and happiness.2 Thanks lo the rise of the sci-
tures of governmentaliry, so that my use of this still somewhat unusual terco
ence of statistics, "population' became an independent realm and force in
is olear in rhe remaining parts. The section that follows contextualizes the
social lile separate from the srate and che family. As an aggregate statistic, the
ices program in the history of family planning campaigns conducted by the
population had its own intrinsic rhythms and regularities and exerted its own
Indian government and provides some dctails about the bureaucratic struc-
effects en the economy and en the nation. Population becomes rhe new aim
ture of ICDS. The thed section provides a Glose look at actual practices of one
of government: an object whose control, regulation, welfare, and conduct
rcDS office. In particular, 1 am interested in the practices of surveillance,
become rhe main goal of government.
exemplified in rhe "surprise inspection," the mosr common instrument of
Governmentality is concerned most of all with "the conduct of conduct,"
rule and regulation. 1 follow one officer on her inspection visits and record

GOVERNING POPULATION 67
66 STATES OF IMAGINATION
 that is, with che myriad ways in which human conduct is directed by calcu- --and.siee -Bnt -n naging a population involves an immersion in the details and
lated means (Dean Iggq: ro). Such a definition of government harks back to minutiae of people's lives. Here mechanisms of discipline and regulation are
the original meaning of the term, before it became hitched ro a particular important nos merely as repressive measures but as facilitators of new podes
relationship with the state: before, that is, the words "government" and of accountabiliry and enumeration. Although neither Foucaultfl r varlous
"state" started being used almost as synonyms in academic discourse. Miller commentators on governmentaliry have had much to say about this topic, any
and Rose (1990: i) point out that in advanced liberal democracies, "political discussion of discipline and regulation must entail a corresponding emphasis
power is exercised ... through a mulcitude of agencies and techniques, some en questions of resistance (O'Malley 1998). What forms of resistance do che
of which are only loosely associated with the executives and bureaucracies of new technologies of governmentaliry engender? And, if we don'[ take govern-
che formal organs of state." In fact, che state has to be seen as "a particular mentaliry to be a system that was set in place once and for all in the Enlight-
form thar government has taken, and one that does nos exhaust the enment, bus as an ever renewing and ever deepening process, then we have
ficid of
ca culations and interventions t at consutute it" (3). Like Weber, Foucault is to consider how governmentaliry is itself a conjunctural and crisis-ridden en-
mteresred m inechanrsms of government that are
found within state instim- terprise, how it engenders ¡ts own modes of resistance and makes, meets,
tions and outside them, that in fact cut across domains that we would regard molds, or is contested by new subjects.
as separare: che state, civil society, the family, down to the intimare details of
what we regard as personal life .
As an example of governmentaliry, take
Po l i
(& 1 A,n , 1
"A Silent Revolution"
ta
aamily
hamily
elalamay be e
state policies that promete or regulare an opt¡mal family size throu g h tv If one were seeking a model of governmentaliry today, it would be hard ro
incentives , advertising campaigns
^^public healthpoljpies zoning laws, and so L., come up with a better example than che ICDS program. We can see this very
on. But there are also r fálie jSoy¡ce h ntay'
Promete a particular family clearly when ICOS is located ¡n its historical context and positioned more
size in the form of ]cave
Po1'e 'es, che provrsion of msurance, and che like; explicidy within state agendas. The Ices was launched in 1975, soon after
JUa ° women's magazines and popular culture may influence how many children a 21Í che formulation of the Nacional Policy for Children. It was spurred by aware-
yn couple desire; the comments of neighbors, coworkers, and teachers might ness that India exhibited some of che world's highest rases of infant morrality,
draw attentionot th ose wh 0 violare11 pD^?
soctetal no s" y-
vi aving too many
ehildren or nos enou g h . All [ h ese are forces of governmentaliry
,
and berween
r.n
raRf ^Jl^
morbidity, and malnutrition and exrremely high rases of maternal morrality
during birth. According to che United Nations Development Program Human
"1 w the concern with che population-¡ts health, longeviry, productivity. re- Development Report for 2000, che infant morrality and under-five morrality rases
s urces-mar is so central to state policy , think tanks, and privare agencies are still 69' and ros per r,ooo, respectively, and che maternal morrality tate
al¡ the desires that inform and regulare che sexual b e h av i or an d intimara stands at 410 per 100,000.
eranons within the " privare" and " domestic " realms of
( heteronormative ) »V" The goal of che anganwadi program was te provide a set of services that
families and marriages are a series of relays that "^
transmit and translate ideas, ^\c r" consisted of supplementary nutrition for pregnant women and young chil-
pracrices, and policies from one rea l m to th
" - (i i e other. Governmentality allows dren and education, immunizations, and prevenrive medicine for poor and
os to bring under one analytical lens the entire domain, showing che opera- lower-caste children. The immunization program was operated by che Health
tion and role of state agencies within á w¡der field nf ct¡on and ¡ntervent¡on Department, which tan the Primary Health Centers (PHCS). It thus took ad-
made possible by a ranee of social actors and discourses. vantage of che presente of a large number of children and "at-risk" women
It should be clear that governmentaliry does nor narre a negative relation- in che anganwadis te inoculate children, pregnantwomen, and nursing moth-
ship of powe one characterized entirely by discipline and regulation. T e ers against the most common diseases.' After experimenting with supple-
emphasrs, rather, is on itcducr¡ve
nr0 imensron eovernm nr I'rv ' i,^ e mentary nutrition programs that produced generalty poor results (Tandon,
health Ramachandran, and Bhatnagar 1981: 382), che ICOS program was initiated to

68 STATES OF IMAGINATION
GOVERNING POPULATION 69
providc a package of well-integrared cervices that would combine nutrition. nora, could cllange people's practices (this was called the knowledge-attitude-
health, educarion, and day caro for children under six years of age and nutri- practices, or ICAP model). When 1 was growing up in India in che ig6os and
tion and health caro for pregnant women (Heaver 1989; Sharma 1986; Tan- 1970s, it was impossible ro miss che inverted red triangle tliat was a symhol
don, Ra ma cha nd rail, and Bhatnagar 1981).' of birth control. It was acto mpanied by che slogan, Hum do, hamuare do ("Us
"Che Icor prograni in any one block (a block consists of an adm inistrative two, our two"). Sometimos, there was an additional graphic displaying a man.
unit of appros i mately one hundred villages) was considcred a "project," and a women, a boy, and a girl. However, when it beeame clear that methods of
each project received funding independendy. In Mandi subdistrict (tehsil ), popularion control built on niodernization theory were ineffective, tyiar better
there were rwo icDS programs- In Mandi block, the program liad been oper- knowledge of conrraeeption and rho inculcation of "modero" actitudes fhiled
ating since 1885, whereas it liad begun in the other block in 1990-1991. The ro alter birth control practices, a seno of frustration set in among policymak-
stmeture of comniand of the ICDs bureaucracy at che district level was as ers, culminating in che draconian measures adopted during the Emergency in
follows: it was to be headed by a disrrict oxogram officer (nunl; the two child 1975-1977. By all accounts, che forced sterilization of men during the Emer-
development project officers (cDPOs), who headed the program at che le-ve, gency, especially among the poor and politically weak segments of che popu-
of che block reported M e CDPO was che head of the office and larion, only impeded subsequent government interventions in this field.'
supervised a clerical staff, which included an account clerk, another clerk who In the post-Emergency period, there was a lull in family planning cam-
did other jobs, a peon, and a driver; che COPO was responsible for overseeing paigns. However, in the next decade, a new consensos emerged among gov-
che work of the four supervisors (Mukhya Sevikas), the eighty-six anganwadi ernments and international development agencies that focused on the qual-
workers in che block, and their eighty-six helpers; the anganwadi workers íty of che population (Dasgupta 1990). The logic, since inscribed as official
were responsible for the day-to-day functioning of centers in villages, which dogma by the Cairo Conference, simply states that lower birthrates are highly
targeted poor and low-caste women and children as beneficiaries. The angan- correlated with higher status for women, accompanied by better nutrition,
wadi centers were supposed to operate every day from 9 A.M. to 1 P.M. Be- education, and health tare for them and their children (Cliquet and Thien-
cause it was not feasible for a single anganwadi worker to run a center, take pont 1955; Sen 1994). In other words, investment in che development of"hu-
care of as many as forty-five children, teach che children, cook food for them, man resources" of "human capital" was expected to pay high dividends, es-
supervise their medical care, and maintain the records, che anganwadi worker pecially when targeted to women and children. This was, in fact, che explicit
was provided with a helper. The helper's duties included doing all che odd language in which che "IGDs Experience" was summarized in a government
jobs associated with the anganwadi, including rounding up the children to brochure: "The experience of IcDs during its first decade (1975-1985) indi
arrend the center, doing the cooking when the centers were supplied with cates that it has the potencial of becoming a silent revolution, a profound
food, and cleaning the "school." In Mandi, all the helpers, anganwadi work- instrument of communiry development and human resource development"
ers, and supervisors, as well as the cDPO were women; the rest of the office (Government of India 1985: 24).
staff were men. The adverse impact of rapid growth on che welfare of the population had
Apart from humanitarian concerns with che high mortality of infants and long been recognized: Indian policymakers and politicians routinely empha-
pregnant women, other factors may have contributed to che support given to size how development efforts are slowed by rapid popularion growth. The
the anganwadi program by the Indian government and by international devel- debate then concerned what the best methods were for reducing the rafe of
opment agencies such as UNICEF. The chief factor, no doubt, was a concemn growth of che popularion. Having failed to persuade people to adgpt birth
with population control.` At the time the ICDS seheme was launched, the In- control practices by a vigorous advertising campaign, an increasingly des-
dian government already had substantial experience with an aggressive birth perate political leadership attempted to use authoritarian measures te sterilize
control campaign. It had used a modernization theory model, according to people during che Emergency under che leadership of Indira Gandhi's younger
which exposing people to information would change their actitudes, which, in son, Sanjay Gandhi. However, the resulring backlash forced Mrs. Gandhi out

70 STATES OF IMAGINATION GOVERNING POPULATION 71

 of power in the elections of 1977. Since then, che icDs program took over as
gram, supplementary nutrition to remove deficiencies, and provision for the
a technology of birth control chal sought to reduce gross birthrates by focus-
care and nutrition of expectant and nursing mothers (Baig 1979 : 340). These
ing on the quality of life for those children already born, that is, by reducing
themes were spelled out even more explicitly in a memorandum attached to the
the mortaliry and morbidity rates for infants and by reducing maternal mor-
Nacional Policy by the Indian Council for Child Welfare (iccw), which argued
tality. If governmentality draws attention to how to govern and how we are
that "che child is an investment in the future of che nation and must, therefore,
governed (Dean 1999: 2), then IcDS is an excellent example of how che gov-
be an integral part of economic planning" (Baig 1979: 334-35).
ernance of the population carne to be reformulated alter a moment of crisis.
The relationship berween IcDs and economic planning for the nation was
In fact, if one does not frame che population question narrowly, one can
seen in its relationship to other development progranis. The supplementary
appreciate how che IcDs brings forth an explicit conjunction berween the
nutrition aspect of icDs was funded from a range of sources. For instance, in
development of human resources, communities, and che nation. The relation-
Mandi block, the program had switched in che late 1g80s to a "wheat-based"
ship berween the population, political economy, and sovereignty so central to
program. This was mentioned several times by the Icos staff in Mandi, and
govemmentality is revealed quite clearly. In a country like India, children un-
at first I could not understand why they were placing so much emphasis on
der fourteen constimte a large proportion of che population (42 percent, ac-
chis fact. 1 later learned chal what was significant about the program was that
cording to che 1971 censos, and 38.6 percent at the end of che 1970s). More
the wheat allocated to it carne from the Food Corporation of india (FCi). The
than So percent of all children Uve in rural areas and have poorer access to
Fci was the body that purchased whear from farmers in che area at support
government services than their urban counterparts (Dasgupta 1990: 1302).
prices set by the government. This policy of buying all che wheat that farmers
The Government of India's Nacional Policy for Children proposed fifreen
could sell at preannounced prices was one of the cornerstones of the green
measures to achieve che goals of fulfilling children's needs. It stated:
revolution and had led to che accumulation of large surpluses in government
The nation's children are a supremely important asset. . . . Children's go-downs. The state's use of this "surplus" wheat for che ICDS thus took che
programmes should find a prominent part in our nacional plans for che results of agricultural development policies and, quite literally, fed them into
development of human resources, so chal our children grow up to be- its welfare policies. Therefore, che development of agriculmre and che devel-
come robust citizens, physically fit, mentally alert and normally healthy, opment of human resources were placed in a synergistic relationship that
endowed with che skills and motivations needed by socíety.... It shall would lead to che development of che nation. The wealth of che nation was
be che policy of che State to provide adequate services to children, both thus tied ro che welfare of its population.
before and afer birth and through che period of growth, to ensure their Because che goal of governmentality is to manage "the population," mecha-
full physical, mental and social development. (Baig 1979: 339-41; em- nisms for intervening roto che affairs of communities and individuals are
phasis added) necessary. The icDs program depended on community intervention for its
efflcacy. Anganwadi workers were recruited from the villages in which they
Such a statement reinforced directives in che Constimtion of India that pro-
served to garner local support for che program. One of che most important
vided for, among other things, free and compulsory education for al¡ children
components of local participation was chal the space for che anganwadi was
up to age fourteen. The authors of che Constitution directed the government
to be supplied by che community; thus, there was no provision for rent in che
to attempt che provisioning of educational services to alI children within ten
ICDS budget. Apart from reducing the cosí of administering che program,
years from che commencement of the Constitution (Dasgupta 1990: 1304).
such a requirement was intended to provide che community with a stake in
The point 1 wish to make here is that concerns with the needs of "nacional
the operationof che anganwadis. This created a great deal of difficulty for
development" were not incidental to programs aimed at children. In fact, the
anganwadi workers and was one of their chief sources of complaints about
National Policy for Children explicitly conceived of che icDs scheme as a re-
che program.'
sponse to its first three directives, which proposeda comprehensive health pro-
The efforts to involve che community in the program did not always work

72 STATES OF IMAGINATION
GOVERNING POPULATION 73
 The IcDs office in Mandi was located in one of the sido streets that led off
verv well. Instead of allowing nongovernmental agencies, voluntary organi-
rhe husv road that served as the major shopping center of the "modera" part
zations, and community groups to infiuence the design of the program and of town. When 1 first went looking for the Icos office, 1 walked right past i4
the methods and kinds of services it delivered. the tcns scheme, like mane this happened on more than une occasion. 1 had been rold to look tbr rhe blue
other governinent programs, imponed a top-down administrative strucmte
UNICEF jeep that servcd as the unofficial mascot for rhe mcns program. 1
with elaborate bureaucratic procedures that had a "slot" for community par- missed the office because the jeep was missing, and, unlike other government
ticipation (Dasgupta 1990:'315). Ironically, such a slotwas deemed necessary ofhces that displayed large sigas, rhere was 110 outward indication that an
ro make the program conform ro the requirements of "participatoty devel-
office existed in that building. 1t was a nondescript space, consisting of a
opment" rhat, according to development orthodoxy, was one of the lessons
small driveway barely large enough fot a vehicle and a narrow flight of stairs
learned from the high failure cate of development projects in the past. Com- to one side. One went up the stairs to a terrace, no wider than eight feet,
munity participation was essential if government was to be seca not as some-
which had a series of doors opening up to it on the left. There were three
thing externa] and imposed but as an intrinsic mode of discipline that led to
rooms. The first room housed the maro office, where the two clerks had their
regular and predictable patterns of conduct and that grew out of, and carne
desks and where the peon usually stood. The second room was used primarily
"naturally" to, communities and selves.
as a storage space. The third room, furthest from the stairs, was the office of
the dynamic and articulate CDPO of Mandi block, Asha Agarwal. She was a
Governmentality and State Surveillance thin, small-boned woman in her thirties who spoke with a confident and en-
gaging tone , sitting behind a fairly large desk in a sparsely fumished and
The concern with the size and qualiry of the population embodied in Che ICDS
decorated room. Like other officers, she had a buzzer on her desk, which she
program was exhibited in techniques of regulation, enumeration, and accounta-
pressed whenever she needed to get the attention of the peon.
bility. There were a host of procedures and rules about the day-to-day func-
tioning of the anganwadis that made little sense when viewed from a botto n-
Inspection Trips
line perspective of gains in health and nutrition. Yet such regulations were
The chief instrument of bureaucratic surveillance was the surprise inspection,
not incidental to the ICDS. As important as were the goals of the leas pro-
which funetioned to ensure that the goals of regulation, enumeration, and
gram-reducing infant mortality and maternal mortality, increasing educa-
accountabiliry were met. Regulation took the form of seeing that the angan-
tional achievements for girls, providing supplementary nutrition to decrease
wadi centers functioned at regular days and times and that the workers and
morbidity-the methods to achieve those goals were equally important. The
children were obeying instructions about how the day tare center ought to be
size of the population was being controlled, but, perhaps more important,
run, how the schooling was to be accomplished, how the facilities had to be
new subjects were being created. And the only way to ensure regulation, enu-
maintained. Enumeration was important in that one of the officer's primary
meration, and aecountability was through a process of surveillance. Although
responsibilities was to monitor the degree to which the anganwadi workers
the methods of surveillance would never reach the Benthamite ideal of the
collected data, especially information about women and children who were
panopticon, which ensured that subjects would regulate their own behavior
the targets of the ICDS program. In this sense, che object of the officer's
because they never knew when they were being observed (Foucault t979a),
surveillance was the degree to which the anganwadi worker monitored her
the goal of the program was to achieve similar results. lf anganwadi workers,
client population. As we will see, in practice, Chis model worked quite un-
children who were being served by the anganwadi, expectant mothers, and
evenly. Accountabiliry was accomplished by a series of checks that ensured
the families of the "clients" would all behave "naturally" in regular, predict-
that the data recorded by the anganwadi worker matched what could be ob-
able, patterned ways, then surveillance would have been most successful (and
served during che inspection. For example, if the worker claimed that forry
hence unnecessary). But surveillance didn't always achieve exactly what it set
children regularly attended the center and ate their meals there, she might be
out to do, and that is what makes Chis story really interesting.

GOVERNING POPULATION 75
74 STATES OF [MAGINATION
 asked to explain why only twenty students were there during a surprise in-
spection , or the claims of the worker about what she taught the students ficial vehicle, Asha managed to use inspection trips effectively and, although
might be tested by giving the students an impromptu test. she could not inspect as many centers as she would Nave been able to with a
One of the greatest challenges facing the bureaucrats was to ensure that vehicle, she still managed to monitor quite a few.
the village women who had been hired to run the anganwadis were in fact I accompanied Asha Agarwal on a couple of inspection trips. She had care-
operating them. Asha Agarwal would often impress upon me the importance fully planned out itinerary so that we would visit centers that had a record of
of inspections for ensuring the "proper operation " of anganwadi centers in good performance , but the fact that there were "
surprise visits" meant that
her block. She reinforced her point by relating the following anecdote to me. they could not serve as public relations exercises . The first trip was on a cold
When she had taken over the Mandi office, it had been without a CDPO for and overcast day in February 1992 , soon after the office had received a fresh
several months ; the government had appointed one of the previous supervi- disbursement of funds for purchasing petrol . The blue IcDs jeep, which had
sora to be a temporary CDPO. Asha pointed out that appointing a supervisor been lying idle because of insufficient funds for repairs and petrol, was
to monitor the functioning of anganwadis was doomed to failure because "in- coaxed loto lile by the driver. Because anganwadis were supposed to operate
charges," as the temporary CDPOS were termed, were at the same level in the from nine in the morning to one in the afternoon , we left the Mandi office
official hierarchy as other supervisors and, hence , lacked the authority to just before nine. Asha and 1 sat alongside the driver in front, while the super-
"pul] up" ( khainchnaa ) other supervisors and anganwadi workers. In addition, visor responsible for the areas we were visitingwas in the back. Out first stop
the office jeep had not been operating for a year because the money to repair was the petrol station . Asha informed me that the state government had re-
it had not been sanctioned by the state government ; the previous in-charge quested all offices to cut their expenditures by 20 percent . Because the office's
had used the lack of a vehicle to justify not making inspection trips. This, in annual report was due at the end of March, they had to run around even more
than usual . She wondered aloud: How did the state government expect her to
turn, had enabled the supervisora to slack off, and the clerks too felt that they
could get away with not fulfilling their responsibilities . Once supervisora cut expenditures so drastically and still get all the work done for the annual
report?
stopped going on inspection trips, anganwadi workers felt they had nothing
to fear and they ceased operating the centers . Thus, claimed Asha, because The first village we stopped was Kalanda . There were two anganwadis in
the in-charge had not made inspection trips, the "whole system " ground to Kalanda that had been operating since 1885 , when the IcDS project began in
a halt. Mandi block . 1 was told that it was a primarily Muslim village, and we en-
During Asha ' s first few months in Mandi, the rcDs jeep was still inoper- countered an impressive mosque at the entrance te the village rather than the
able, so she , too, had not conducted any inspection trips. But when it ap- temple often seen in Hindu - majority villages . The village was most unusual
for the well - maintained quality of its inner roads and the complete absence
peared that the vehicle would not be repaired in che foreseeable future, she
started taking public transportation m pay "surprise visits" to the centers. of sewage water and garbage on the streets . 1 was told that many men in the
She went to four centers and was horrified to find that none was funetioning. village were masons and had volunteered their labor to ]ay the roads and che
She felt that it was useless going to other centers because she would then end drains. Some of them had gone te work as laborers in the Middle East and
up having to give warnings to all the workers under her charge. So , at the next had come back with small fortunes, which accounted both for the relative
monthly meeting of anganwadi workers, she announced that if she found grandeur of the mosque and the neat-looking houses.
centers that were not functioning , she would take disciplinary action against The first anganwadi that we went to inspect was housed in a dark room
the workers concerned . Following that warning, she resumed going on in- that served as the storage arca for a farm family. A huge pile of lentils (arhar)
spection trips. At first , she found attendance at the centers spotty. However, occupied half the room , completely covering one wall and a good proportion

once the word spread that she had started issuing warnings and docking of the floor space. The anganwadi worker, a pleasant and energetic woman,
workers' pay, the centers started operating again. Thus , quickly sent the helper to round up additional children to add to the fourteen
even without an of-
who were already there. Asha asked the children to count numbers and to

76 STATES OF IMAGINATION

GOVERNING POPULATION 77
 That day, we visited another village with two more centers and liad a re-
recite the alphabet, which they did with practiced case. Onc child in particu-
nlarkably similar experience in that the nvo anganwadis that were well ron
lar, who was a little older than the rest, had written down numbers all the way
liad workers who liad high school degrees, whereas the anganwadi workers
m one hmtdred on bis date, and had also memorized all the poems and songs
in the other two were womeu whose qu al ifications exceeded the ni¡nimum
they had been taught. While we were at Chis center, a number of children
required for the job. Asha's explanation for this was that highly trained
carne in, looking washed and scrubbed. Asha told me that Che teaeher had
women often felt that Chis job was below their dignity. They usually prcferred
onle a high school degree. but seemed m be doing a good job with the chil-
other, better-paying jobs, but accepred tire posirion of in anganwadi worker
dren. She castigated tire angamvadi worker for flor removing the charts,
for lack of other oppormnities. Asha said thar ir was probably a mistake
which functioned as teaching aides, froni the wall where the Ientils had been
to hire people who were overqualified, because such people would never be
piled. "It is your job to look alter the charts," she told her. "When you knew
happy in a job where their skills were underutilized and in which they felt
that the crop was going to be stored there, why didn't you remove the charts
underappreciated.
beforehand?" After inspecting the attendance registers and writing a brief
During her inspection trips, Asha referred to the attendance registers, in
report in the inspection register, which noted when the inspection took place,
which anganwadi workers liad to record the number of children who carne to
how many children were there, and what the children had demonstrated, we
the center in order to evaluate the performance of an anganwadi. If she did
left that center and headed for the second one.
not find a center open or functioning properly during a surprise visit, she
The second center in Kalanda was in the porch of a house. When we
docked the anganwadi worker's pay for that day and left a note requesting an
reached there, the anganwadi worker was nowhere to be seen. There were a
explanation (spashtíkaran) for why the worker was not there. Repeated ab-
handful of very young children present, along with the helper. When asked
sences or delays in responding to the cDPo's demand for an explanation re-
where the worker was, the helped claimed that she did non know. Asha and
sulted in extended pay cuts; however, a decision to terminate employment
the supervisor attempted to coax some of the children to stand up and recite
required a great deal of documentation and careful groundwork on the part
the number table or identify objects on an alphabet chart; however, none of
of the CDPO.
them opened their mouths. It was hard to tell whether this was out of fear of
One example of repeated abstention from duty was provided by Sona Devi,
the visitors or because of the unfamiliarity with the task. We waited for a few
an older, widowed woman with three children who lived in a large village
minutes, then headed back to the jeep. As we were leaving, che anganwadi
called Hamirpur that had three anganwadi centers. Asha told me that she had
worker carne hurrying toward us. She apologized profusely and blamed her
found Sona Devi's center closed during her last three inspection visits. Asha
delay en the fact that the bus she was traveling en liad broken down. Asha
opened her inspection ledger and showed it to me as evidente: it indicated
chastised her in no uncertain terms. Even if her bus liad broken down, she
that Asha liad reached the center at 12:3o and found it closed. When she
said, that was no excuse for reaching the center at 11:15 instead of 9. The
asked why there were no children at her center, Sona Devi replied that they
worker lamented her fate, saying that it was her bad luck that the one day
liad all gone to see a play (nat) being performed in the village. But when Asha
when she started late was the day we happened to arrive. She tried to persuade
checked this story with workers at the other centers, they were unaware of
us to come back to the anganwadi for a few minutes, but Asha wanted to see
any play being performed in the village at that time- Yet, despite Sona Devi's
centers in other villages that day, and it was fast approaching closing time.
poor record, Asha had so far resisted firing her; all she had done was cut her
On out way to the jeep, Asha noted wryly how much better seemed the
pay for not performing her job.
center operated by the woman who was only "high-school pass" comparéd
Asha then proceeded to give me more examples of how difficult it was for
to the second one, despite the fact that the second anganwadi worker had a
her to fire, and therefore discipline, anganwadi workers even when she knew
master's degree. The children at the first center seemed better taught and the
they were not doing their jobs. One, Balvanti, used to manage a center in her
teaeher displayed more enthusiasm, observed Asha. She appeared surprised
natal village. When she got married and left for her husband's village, her
at this because, in the past, she had found that the better-educated "teaeher"
father requested that his younger daughter be made the anganwadi worker in
had done a really good job.

GOVERNI NG POPULATION 79
78 STATES OF IMAGINATION
 place of the older. Asha told him that she could not do that because she was
been absent from their centers. Both women responded within a day, saying
required to advertise the position, and that, furthermore, all new positions
that they could not be at the anganwadi because their children had suddenly
were reserved for scheduled caste applicants.' Asha waited for, but did not
taken iii. Asha told me that this was the excuse that she was given most fre-
receive, a letter of resignation from Balvanti. Afier her marriage, Balvanti re-
quendy. If, during a surprise visit, she found a center that was not function-
turned to her natal village for a few months before she moved permanendy to
ing, she would visit it again in a few days, usually within a fortnight. lf she
her new home.'° During the time that Balvanti was back in her parents' home,
found that the center was still not operating, she would leave a warning and
she resumed operation o£ the anganwadi. Asha added that Balvanti had been
would dock the worker's pay for yet another day. Shordy thereafter, she would
a conscientious worker and had done a very good job of running the angan-
visit the same anganwadi for a third time; if it was still not operating, she
wadi. But eventually, Balvanti had left permanendy for her husband's village.
would leave a third warning and would thereby prepare the way for suspend-
Whenever she returned to her parents' home for brief periods during the year,
ing the worker. Anganwadi workers were not government employees and,
she would reopen the center and operate it for a few days. But for most of the
hence, could be fired afier the third warning. However, Asha usually gave
year the anganwadi rernained closed. Once, when Asha had gone on an in-
them another chance. "When we go to higher officials to get rid of some-
spection tour and found the center closed, she went to Balvanti's house to
one," she explained, "they tell us, 'First make the file thicker."' In other
verify her whereabouts. She was told that Balvanti had just left for the fields
words, get more material, more paperwork, before taking any action. "The
on some urgent business. But a small child who was standing there piped up,
thicker the file," she said, "the easier it is to get a decision to fire someone."
"She hasn't gone to the fieed, she has gone to her own honre!" Thus, her
Thus, the surveillance exercised by the CDPO through her inspection trips
family member's lie was exposed. During previous inspection visits, Asha had
was not always matched by a capacity to discipline and fire workers. The
tried to persuade Balvanti's family that she would be better off resigning than
CDPO's authority was limited to withholding the pay of workers who were
getting fired. She told them, "This is a government department. By resigning,
not doing their job. To relieve a worker of her job, the CDPO had to first
she leaves with her setf-respect intact. By getting fired, she brings disrespect
assemble an unimpeachable record of the worker's misdemeanor and then
to herself." Yet, eighteen months afier that inciden[, Asha had notyet received
convince her boss that such drastic action was justified.
a resignation letter. Asha added that it was imperative that she fire Balvanti
On one of my visits tu the ICDS office on a warm day in February 1992, the
before a new consignment of food was allocated to the ICDS program be-
staff had pulled the desks and chairs onto the narrow porch to take advantage
cause, if she waited to relieve her until afier the food had been supplied, the
of the sunny weather. While I was talking to Asha, a man carne up the stairs
chances were that Balvanti, knowing that she would have to resign soon,
and headed into the office. Afier consulting with the clerk, he haaded a slip
would appropriate the food. Asha also knew that Balvanti would come back
of paper to Asha. It was an application for leave on behalf of his wife, who
to her parents' horile for Holi (the spring harvest festival) and was afraid that
was an anganwadi worker. The application requested leave for a few days
she might restar[ the anganwadi for a few days; Asha would then have to
because she was ill. Accompanying the application was an impressive stack
conduct three more inspections to fire her. Asha underlined the difficulty of
of papers, including an x-ray, which the man plopped down in front of Asha.
her task by noting that the two registered letters she had sent Balvanti were
He said that if Asha didn't believe him, she could look at the medical papers
not returned to the office, nor was there evidence that they had been delivered.
and convince herself that his wife was telling the truth. Asha categorically
Asha surmised that Balvanti's family probably knew the village postman and
refused to believe the man. She said that she had made surprise visits to that
had cajoled him to hand the letter over without their signing a receipt. She
center on two occasions and found it closed both times. What was more,
had drafted another letter terminating Balvanti's employment and was about
many villagers had come to her complaining that the center did not function.
to send it to her boss, who would have it signed and delivered officially.
She told the man that if his wife could not operare the center because she was
Asha showed me examples of letters in which she had put two anganwadi
ill, she should Nave applied for medical leave, and Asha would have been
workers on notice and had demanded a written explanation for why they had
happy to endorse such an application. Alternarively, his wife could have ap-

80 STATES OF IMAGINATION
GOVERNING POPULATION 81
 The recalcitranee of subordinates was only one instance in which che hands
plied for "casual" leave (which, However, vas limited lo rwenry days every
of coros were tied. Sona Devi, whose pay liad been docked by Asha, had
vear). But, Asha emphasized, rhe worker could nos keep che anganwadi closed
decided ro puf political pressure en Asha lo restore her stipend. One day, five
i ndefinitcly hecause she was ill and continue ro draw a salary as if rhe center
oren arrivcd and told Asha that she had no righr lo spcak tu Sona Devi in aro
were opera. Asha added aliar she had not eet received a response (spashtikaran)
"insulring tono" Asha presenred her case m diem and asked rhem what rhey
to che letters she liad lefr at the center. She demanded lo know why, if che
wanted her lo do, given che fact that Sona Devi's angamwadi was found to nos
worker was ill, tire center was nor being run by her helper: "If your wife
be functioning en there different occasions. Ignoring her quesrion, che roen
cannot make ir ro che anganwadi un certain days, why is rhe helper absent? I
said thar rhey were nos asking her to do anything, rhey just wanted ro warn
should find che helper lat che center] oven if che anganwadi worker is not
her to nos "niisbchave" wíth her workers. Asha became enraged as she re-
there." The man defended his wife, saying that she could not force the helper
countcd what happened: "First, she [Sona Devi] docs something wrong, and
to show up. But Asha did nos give up her fine of questioning. If his wife went
then she fries to put [political] pressure on me! That makes me even more
to che anganwadi regularly, why did che attendance registers nos demonstrate
angry."
that fact by listing che narres of che children who were present? "When 1 went
V ,,/i Surprise inspectious and registers were rwo devices by which regulati^
there," Asha said, "none of che registers had been filled." That charge finally
rand accountability were pursuedd through devices of enumeration. It was nos
broke che man's resistance. He then switched tactics and claimed that it was
; JV only that superior officers at " higher" levels traveled in jeeps, irwas also that
hard te entice children m come to che anganwadi when there was no food
rhey traveled to conduct inspectious, to discipline, reward, encourage, and
(poshtahaar) provided to rhem. Asha claimed that, by that logic, none of che
punish. Registers helped rhem do just that, for registers enabled rhem ro
anganwadis in Mandi should have been operating, as there was no food being
check their observations against what had been noted. For example, Asha
distributed at any of rhem. Defeated by that battery of arguments, che man
complained that workers who tan anganwadi centers in their bornes ofren
lefr. When he had gone, 1 asked Asha if she intended to fire thar particular
brought in additional children when rhey saw the dust of che jeep in che dis-
anganwadi worker. To my surprise, she said that she did nos think ir necessary
tance. Thus, by che time che campo acrually reached che center, there were
to resort to such a drastic step. "This was only my second warning to her,"
many children there even if che anganwadi had nos been operating. However,
she said. "Wc have to allow for the possibiliry that there are ofren genuine
she managed to catch che worker's "deception" in such cases by checking che
reasons why che center is nos open." Anganwadi workers sometimes carne
names of che children present against che names (if any) entered in che atten-
back en track after repeated warnings and, in that particular case, she would
dance register. The coPo's ability m swoop down en the space of che angan-
wait a little longer.
wadi worker was thus mediated by e semiotic of dliiLj^a smoke signa) deliv-
Asha proceeded to tell me about other centers that were in trouble, drawing
ered by that very device, che jeep, t at enabled her to suddenly enser che space
en che cases of Balvanti and Sona Devi referred to earlier, and emphasized che
of che worker.
difficulties she had in doing anyrhing to remedy che situation. Asha had al-
The surveillance exercised by superior officers on their subordinates was
ready been to Balvanti's village twice, and a supervisor had visited once, and
parí of che routine functioning of the Indian state. However, chis kind of
rhey had both found che anganwadi closed. In addition, when Asha went to
monitoring did nos easily translate roto control and discipline. The authoriry
inspect che center, villagers complained that it no longer functioned properly.
of che copo, as of any superior officer in che hierarchy, could be subverted,
However, oral complaints were of little use, and Asha was frustrated in her
deferred, or denied through a range of tactics. The workers who were che
efforts te persuade villagers to write down what rhey told her. "The problem,"
objects of surveillance by their superiors did nos merely conform and police
she said, "is that when you ask someone to give you a complaint in writing,
themselves as expected R rh r rh cimply regulating and normalizin
thev at once withdraw what is orherwise vociferous criticism. With govern-
per f superior ofhcerc to exercise surveillance en thelr workers some-
ment work, unless you have something in writing, you cannot build a case
times provoked disruptive reacrions that threatened che hierarchical assurnp-
and Cake any action."

GOVERNING POPULATION 83
82 STATES OF IMAGINATION
 tions of bureaucratic order. Government by state bureaucracies did not
state to justify a range of actions . My argument is that statistics function
m o se - government by anganwadi workers; there were
to gauge things like activity levels and health , monitor the actions o£ social
significant points of tension and friction in the art of government.
agents, and regulare the behavior of populations. AII these functions extend
I found a similarly contested relationship between anganwadi workers and
beyond the state and belong to the realm of government more generally.
villagers. As part o£ theirjob , workers were required to collectvast amounts of
The most time - consuming activity of anganwadi workers consisted of
data, particularly about women and children , segments of the population that
documenting and generating statistics . A plethora of registers recorded such
had not been as extensively surveyed , counted, classified, measured , injected,
things as how many children attended the center each day and who they were:
or schooled in the past . The monitoring that superior officers exercised on
their name , father's name, and caste . A nutrition register recorded how much
workers was meant partially to ensure that they were , in toro , conscientiously
food and fuel was consumed each day. A third register was used to record the
monitoring the population they were " serving."The next section deals explic-
birth dates of each child boro in the village and the parents ' names, ages, and
itly with the relationship between
anganwadi workers and " their" villagers.
castes. Similar records were kept of all deaths. The name, age, and caste of
each pregnant woman and a record of the outcome of the pregnancy were
State Mechanisms : A Numbers Game? recorded in another register. A travel log maintained a record of when and
why an anganwadi worker was missing from a center . An inspection register
Although they were not government employees , anganwadi workers were ex-
was maintained where supervisors, the CDPO, and other visitors recorded
pected to behave as such in one important regard : they had the crucial re-
their impressions about the functioning of the anganwadi.
sponsibility of generatin ria1 statistics for rhe state In a approprla
Maintaining all these records posed a daunting challenge to most angan-
image, Hacking (1982 ) has characterized the activities of the modera state as
wadi workers, particularly those who lacked the requisite cultural capital in
generating,an "avalanche of numbers ."" In the anganwadi program, record
the form of mathematical skills. Sharda Devi was a worker in the village of
keeping often appeare e I I self; it also had far-reaching effects
Bhaipur, a few miles from Mandi . Her husband was a self-taught "doctor"
in mapping, surveying, and tabulating the population and, most important,
who was also the community health worker for the village . When I met her
in potentially monitoring the lives of women and children.
during one of my visits to the center, she complained to me about the mathe-
Enumeration is a critical modality of governmentality ; it is through the col-
matics involved in the supplementary nutrition program. Different quantities
lection of statistics that the conduct of conduct can be effected . What kinds of
of food, measured in grams, had to be given to children , pregnant women,
statistics are collected , who collects them, and how they are used all affect the
and nursing mothers. Then the totals had to be added up for each category
regulation of populations , techniques of accountabilityand the formatinn off
and for all groups each day. These totals were next tallied against the amount
group identities . Foucault has pointed out the family resemblance between
of food actually left in the center. When the COPO carne on her inspection
statistics and the state : the rise of statistics is integral to the science of the
trips, one of the things she looked for was discrepancies in the registers and
state that developed in Europe at the end of the sixteenth century (1991: 96).
the actual amount of food remaining at the center. Sharda Devi 's husband
Kaviraj ( 1994) links enumeration to a specifically modern form of commu-
asked pointedly, " How do they expect a person with an eighth -grade educa-
nity identity, which he opposes to "fuzzy" communities . It is through the
tion to do all this? If I didn't help her , she would never be able to manage the
purposeful counting of peoples as members of certain kinds of groups that
books." However, everyone , including the cDPo, understood that the object
arguments can be made about representativeness , about majorities and mi-
of the exercise was not so much to detect " corruption" as to keep the record
norities, about who is falling behind and who is ahead in income data, edu-
straight so that no aspersions of corruption could reasonably be made. Thick
cational achievements , and so on. Statistics are not just collected by the state,
files and carefully totaled numbers were more important than actual action,
and they certainly are not always collected or employed in the interests of the
because the logic of bureaucratic justification demanded written evidence, a
state. Aggrieved groups can quite effectively marshal statistics against the
fact often lost on semiliterate or illiterate people in rural arcas.

84 STATES OF IMAGINATION
GOVERNING POPULATION 85
 of enumeration, classifiearion, and recording thar operated on a segment of
But rhere was more to tire functioning of anganwadis than generating
the population whose low level of literacy and lack of parricipation in the
numbers so that Cables and columns totaled up correctly. A silent revolution
formal eeonomy liad kepí it relarively insulated from Che chronotope of state
was indeed raking place through diis program, and ir was not just in the
surveillance. W hat differentiated the anganwadi worker from tire census taker
"development of human resources." Perhaps for rhe hrst time in Che history
was precisely rhe degree of familiarity with tire village that no outsider could
of rhe nation records werc being kept en births and deaths in rural areas.
ever obtain. Even when tire worker kept her distante from the social life of
Anyone who has arrempred to do a census oí an indian village knows how
Che village, its politics and divisions, she still knew a great deal more abour
difficult ir is ro record precise ages and dates of hirth, as the rechniques of
individuals and families [han any other state ofhcial could possibly Irnow.
Che modern W srern r-^n.; of the nation and `irs' pópuÍation are (mis)-
More imporranr, Che worker learned a great deal about women in rhe village,
translated roto inconmensurable modes and methods of recording the pas-
a seginent of Che population relatively insulated from Che gaze of other (malo)
sage of time as it intersects with life histories. 1 looked at several registers
state officials.
in anganwadis that recorded information about births. The mothers' ages,
However, just as there were limits to Che state's surveillance of the running
which may have ranged from fifteen and up, were all carefully recorded as
of anganwadi centers through the direct supervision of Che CDPO, there were
over the legal age of marriage: eighteen years and six months, or nineteen, or
limits to its surveillance through practices of data gathering by anganwadi
twenty-two. However, the birth of children in the village since the anganwadi
workers. The anganwadi program did not merely or mainly result in Che in-
program began in Mandi was recorded to the day, and sometimes to the hour.
corporation of women and children roto the relentless march of Che machin-
The registers contained similar, relativety aceurate information about deaths.
ery of state surveillance. 1 give rwo reasons below why such a conclusion
1n other registers were to be found data on inoculations, the weight of infants
would exaggerate claims about increasing state control and underestimate
and pregnant women, attendance at Che anganwadi, and so en. The angan-
Che spaces for resistance and ambivalente created by [hese very processes.
wadi program had resulted in a quantum leap in data on women and children,
Whether or not Che state's role in rural life was inereased by Che anganwadi
particularly with respect to fertility and infant mortality.
program, it is clear that new modes of governmentality were being introduced
It may be objected that because anganwadi workers were not trained census
by Che prograni, techniques of enumeration and data gathering, and novel
takers, the qualiry of Che data they collected was suspect. And, indeed, be-
technologies of regulation and accountability.
cause there were no mechanisms to check whether a worker had recorded all
Despite anganwadi workers' bes[ effbrts, villagers ofien resisted their ef-
Che births and deaths ¡ti a village or had accurately noted down the exact birth
forts to collect statistics . At one of Che monthly meetings of workers that 1
date of individuals, it would have been impossible to tell if a birth or death
attended, several stood up to document the difficulties they experienced in
had been recorded correctly. However, what needs to be pointed out is that,
collecting the information they were required ro enter in the registers. One
unlike census takers, anganwadi workers either lived in Che village in which
worker said that villagers refused to allow their children's weight te be mea-
they worked or went daily to Che village to operare the anganwadi. They could
sured. One day, as part of her duties, she had weighed some children; the
gather a great deal of information from the children who carne to Che angan-
next day, a litde hoy fell ill, and his sibling told the res[ of the family that the
wadi. Further, births and deaths were major public occasions in villages in
child had been weighed the previous day. Measuring children's weight and
western Uttar Pradesh and thus could be hidden only with grear diffrculty.
pronouncing them healthy was considered reason enough to attract envy, Che
Finally, anganwadi workers had no incentive to not record or to misrecord
"evil eye" so feared by people in western Uttar Pradesh.12 After that day, none
such events. Thus, although there were few mechanisms to double-check the
of the households in the village would allow their children to attend rhe an-
records kept by workers, there was little reason to be suspicious of Che figures
ganwadi. They rold Che worker, "When you don'[ feed Che children, why do
that had been entered in the registers.
you weigh them?" She could not convince them that no harm would come to
The broader point that 1 wish to make here is that, seemingly by-products
children by weighing them.
of the functioning of Che anganwadi program were series of numbers, modes

GOVERNING POPULATION 87
86 STATES OF IMAGI NATION
 Similarly, some workers reponed that when they went from door to door
to do a survey of the population , causes programs to fail, rather [han as a constitutive parí of rule, of "govern-
people often refused m cooperare with them.
"Why do you ment from below" (1998: 157). To see resistance mainly in tercos of that
come to out house to do the survey when we have to come to
you for inoculations and injections ?" which causes state programs to fail privileges che perspective of planners and
che workers were asked with impeccable /
bureaucrats and fails to acknowledge che central role played by resistance in
logic. "You should just sir at che center and do che survey there." Workers
also V -shaping discourses, instimtions, and programs of role. At its limit, O'Ma ey
described their difficulties in asking questions about al] members of a family. ^rtr
They were challenged by villagers with the words suggests, resistance can so fundamentally transform regimes of rule that it
, " When you leed only the
can creare a "significan[ source of instability" (170) for [hose regimes. Resis-
children, why do you want to cake a survey that includes everyone ? Why do
tance, in other words, might provoke a shift in strategy of modes of gover-
you want to find out who has died-are you going to leed the dead roo ?" The
nance that alter them so fundamenmlly as to be constitutive rather than acting
workers raid that they had no good response te such questions and were
merely as a source of constraint.
sometimes unable to persuade villagers to cooperare with them.
State surveillance did not mercase by virtue Resistance, of course, provokes reactions from [hose in charge of pro-
of che fact of data collection. VI V^
When analysts use che concept of che state , the imoressinn rnnvr..P.t :" ,.a., o [ grams and policies. One such fascinating mssle about che meaning of work
an co si e instimtion that has LbLV in che anganwadi concerned icDS's componen[ of schooling . Contrary to the
otentlauy greac capacities off`ó
r c n-
state's efforts to portray them as voluntary workers, most anganwadi workers
trolling, monitoring, and manipulatin¢ its vooulatior^However , wh
en one
disaggregates the state and analyzes the workings of individual bureaucracies
and programs like the ICDS, it becomes more dif&cult to conceptualize a co- ^^
that I interviewed referred to themselves as te rs consciously eliding che
difference berween themselves and schoolteachers . The st a t e, en c h e ot h er
-
G C4

hand,employed the discourse of mo[herhoo m represen [ine the efforts

ordinated, systematic instimtion that can exploit che data collected by its vari-
/ f '"^^^•^^n voh worker^ as By this logic, what workers did in che
ntaarry.
ous apparamses. In fact, [he leve] of coordination between agencies and bu-
creches was deemed an extensinn nfwhar a " o...,.1 rnnrhpr " ...,.,, u "-. Ann
reaucracies of che state implied by the term surveillance, with its connomtion
at home: che only difference was that che worker performed that funetion for K -n
of linkages between data collection and repression, suggests capabilides that
more children than would normally be found in a household. By che state's f^
che state may not possess. The Ices project at che block level diligently ac-
logic, therefore , the work done by a worker differed in scope from what she
cumulated the data collected by che anganwadi workers and passed them on
did at home but was q uaht a ti ue 1 'ent
y equlva 1 to mor enng
h . In contras [, by re-
[o che district level, from where [hey were fed finto nacional statistics on the
ferring to themselves as teachers, anganwa 1 w asized how simi- 1u
program. However, it was not clear if anyone at che district leve) was process-
lar their work was te that performed by teachers in elementary schools.
ing che data and using them for any purpose whatsoever. I was told of moun- 11 c ose
Thus an ganwa Ai ;wor ers h to emphaslze che quahtatIvely dli ferent nature
[ains of data awaiting processing, from which conclusions might be drawn
of work in che anganwadi as compared to che home.
and lessons extracted. Hacking's (1982) image of an "avalanche of numbers"
Anganwadi workers were proud of [hose students who liad either refused
is appropriate: it suggests an undirected and undisciplined flow, triggered by
to leave their centers ro go to a "Montessori" (che narre for any school that
che smallest disturbance in a mountain of numbers threatening to cascade
down without being directed to any end by any agency. charged tuition and claimed to teach English as a subject) or returned to che
anganwadi because [hey liad learned so much there..Once, toward che end of
January 1992, 1 dropped in to visir che anganwadi in Alipur. Sharmila, che
Governmentality and Resistance Brahmin woman who was che anganwadi worker, pointed te one of che girls
in her class. Before she joined che anganwadi, che girl used to walk a fair
The examples of resistance offered in che previous section raise che question
distance to a Montessori in an adjacent village. When she started attending
of che role of resistance in understandings of governmentality. O'Malley ar-
che anganwadi regularly, she discovered that her classmates knew more [han
gues that resistance has roo often been leen as a "negative externality" that
she did. She was, in fact, asked by one of che other little girls if she liad

88 STATES OF IMAGINATION
GOVERNING POPULATION 89
learned anything at all at the Montessori! Sharmila commented that, because On anorher occasion, she explained why "it was a bad idea" for a woman
che Montessori charged Rs. g a month in tuition and the anganwadi school to run an anganwadi center in her own house. She said that in such a center,
raught children for free, people in the village assumed that the edueation stu- it became impossible ro determine when a wornan was actually working and
dents reccived at the Montessori was betrer. "They don't value rhis education when she was doing her own housework. in addition, Asha felr that ir rhe
hecause it is free," Sharmila concluded. At anorher center in Kalanda, de- center was at che worker's honre, there was no way to ensure that she was not
scrihed aboye, the "star smdent" had been removed from the anganwadi and diverting the food provided for supplementary outririon of children for her
sera ro rhe village school by his parents. However, he ran away from there own family's necds. "]r is far better for che woman to go to a center some-
and carne hack ro the anganwadi because he liked being at che center. where else because then she goess with che attimde that she has ro work for
At one of rheir monthly meerings, anganwadi workers complained thar, che next three hours." Not only would it make it easier to kcep surveillance
ironically, the superior education provided at the anganwadis actually created on the worker, it would also maintain che center as a separare space imbued
problems. The workers claimed that as soon as the children learned a little with the authority of the state.
bit at che anganwadi, their parents felt that they were "too bright" to stay The problem, she said, was that when she told women not to establish
there and would transfer them to a Montessori or a government-run primary centers at their own homes, they replied, "What can we do? Get os another
school. This resulted in high turnover, as many children left che anganwadis place, and we'll move." Thus, one of Asha's biggest frustrations with the
soon after beginning their edueation. The workers added that Chis was bad program was that no money had been allocated for rent. The space for the
for the children because, in che government schools, they were packed eighry anganwadi was supposed to be donated by the village to ensure local partici-
to a class and che teachers were usually found sipping tea in the courtyard pation in the program. However, in Mandi block, where rhe program had
instead of teaching. They point out that teachers in the government schools been operating for six years, landlords, nervous of government intentions
were paid thousands of rupees for their "efforts," whereas anganwadi work- toward their properry and aware of laws that favored tenants, were systemat-
ers were compensated little for giving children in d ivi d ua l attent i en. ically taking back the spaces they had loaned to the centers.
The tension berween "voluntary worker" and "teacher" was symptomatic Women who worked in che anganwadi program were thus expected to per-
of a more general contradicrion that underlay the design of rhe anganwadi form honre-like duties in the centers, extending their "natural" roles uf caring
program. On the one hand, the anganwadi program was clearly built en the fiar children and cooking meals. However, they were also expected to main-
notion that women, as the "natural" caregivers for children, would be bese taín the boundary berween their home and their workplace, because blurring
suited to bring health and educacional interventions to young children and to those boundaries made surveillance and control impossible. In harnessing
pregnant women and nursing mothers. On Che other hand, anganwadi work- women's energies by extending their domestic roles to che public sphere for
ers were expected to be "professional" in carrying out their duties and were the development of the community and che nation, che anganwadi program
bound to an even more impressive array of bureaucratic procedures and rec- was unable to mediare the tensions created by domesticating che state, as
ord keeping than their better-paid counterparts in government service. conflicts berween control and performance became intractable. Control re-
The tensions berween anganwadi workers' status as voluntary workers or quired women to operare che centers in a workspace, a nondomestic setting;
paid professionals were manifest in the selection of sites for anganwadis. One however, che tasks they were expected to perform were considered domestic
day, as we walked to en anganwadi center, Asha took che opportuniry to return and home-like duties. Predictably, anganwadi workers resisted such a for-
to one of her favorite themes: the diffiiculties created by che fact that there was mulation by taking the opposite position en those issues, arguing for an eli-
no provision to rent space for the anganwadis. "When someone gives you a sion of work and domestic spaces en che grounds that it was hard to run a
place for free," she said, "you can't tell them that they don't have a right to center in borrowed space and insisting en a distinction berween domestic
use that space for their personal use. And then tire result is that even if the work and che teaching they did in che center. Thus, even che bes[ efforts of
children don't have a place tu sir, there is little you can say m the owner of state officials to govern the conduct of anganwadi workers, who were, afrer
the house." all, their own employees, did not bring about che results they had hoped fbr.

90 STATES OF IMAGINATION GOVEANING POPULATION gr

 Instead of the program's objectives becoming the objectives of the workers
through a process of internalization (especially pregnant women) and children , and why the program was geared
and self-discipline, recalcitrant workers so heavily to poor and lower -caste groups . These were the groups who most
redefined their activities and changed the Bite of the anganwadi to their own
home. Here needed protection , as they were " at-risk," "vulnerable," and leas [ able to help
is a case where resistance took not the form of refusal, bus largely
ó£ reinteroretatíon . themselves.
It was not that anganwadi workers necessarily changed what
Investment in human capital and benevolent protection colluded with che
they were doing; they were in a semiotic struggle about che merning of their
actions. third feature that was a necessary effect of s h a program and thar was the
vas[ mercase in che monitoring , surveillance , and regulation o£ che target
p population .'° This happened through a vanety o mee a sms. nganwadi
Conclusion
wor required to collect a staggering array of statistics . It is signifi-
Governmentality, Dean and Hindess (1998 ) can[ that most of che data collected by workers dealt with che segment of che
remind us , is equally about the
population that was leas[ represented in official records. Almost any type of
government of conduct and the conduct of government .
Examining the Icor record kepí by che state - land titles , school records, bank accounts , postal
program allows os to see precisely how che latter shapes the former
. In other savings accounts , electriciry connections , loans-recorded the name, age,
words, how does the functioning of one government program chape che
residence , and sometimes caste of che father . Typical records read: "Ravinder,
conduct of workers, clients , and bureaucrats ?
State efforts to alter, re g ula te , age 16, slo [son of) Gavinder, Village Khurd ," or "Poonam, age io, dlo Gav-
monitor, measure ,
record, and reward che con uct of olitica Isem ow-
inder, Village Khurd." Men were present in a variery of capacities^ as sons,
ered groups of lower-caste women and children
m anaQe t h e sizP anA q I ' n husbands , fathers, and even grandfathers . Significantly, women were largely
the onrn,l +: ".. nr
rh„ en ",e .:..-... 1-
1r absent or were mentioned only occasionally as wives and daughters . Because
u. u m nselt was changed as a resut t
othis interaction, as these
Qrouos imbued che state with their own agends , interpretations women were less likely to attend institutions of formal learning orto transact
, and actions.
I began thts essay by noting che c e o business at a bank or post office, they were far less visible in che ofhcial rec-
porta of a "benevolent" program i e ord. It was precisely this segment of che population , heretofore insulated
che Icos in che face of certain incontrovertible and tragic facts relating to che
Q.v from che apparamses of state surveillance , rhat che anganwadi workers' sta-
high mortality and morbidity yates for children and pregnant women in India.
tistics were systematically bringing under che gaze of che state.
The program promised to inexpensively increase the human capital of che na-
tion and, thus , promete its rapid development There was another aspect to che increased ambit of state surveillance. This
. Investing in children , che logic
went, was investing in che nation ' s future. Another , had to do with che anganwadi workers, who were themselves ambivalently
closely related payoffwas positioned as simultaneously implementers of che Icos program and bene-
its potencial to bring down fertility rates .
Better health and education and ficiaries of state benevolence . Workers we
increased rases of survival for children were expected to have long -
term ef- what they did at the centers was defined as volunteer work fiar whic ey
fects in slowing down population growth. As Suman Nayyar ,
a training ofhcer ere recompense not ' I Parnctpauon in che
in the Icos directorate ,
succinctly put it: "If a couple know that their children
won't survive , they'll have more. Education is che best contraceptive program subjected them to an intensive regime of regulation and surveil-
" ( Inter- lance . Inspection trips to monitor che performance of anganwadis were made
Press Service , January 7 , 1987)."
by severa ) leveis of officials, who thus subjected them to a level of monitoring
The ideology of humanitarian intervention ,
given an economic racionale as that exceeded even that exerted on regular employees of che state.
an investment in the future of the nation ,
was accompanied by a disco irea^F
protection . However, state surveillance was nos without its ambiguities and ambiva-
The premise of benevolent protection implicit in the Icos scheme
lences. In coping with che requirements for paperwork placed on them and
almost never drew comment from any of its planners or participants. The
che insistence on procedure followed by Indian bureaucracies , anganwadi
commitment to protection explained why the IcDS program targeted women
workers gained a great deal of competente in navigating and circumventing

92 STATES OF IMAGINATION
GOVERNING POPULATION 93
rhe procedures of rhe burcaucracy. They learned too to creare paper trails thar -, i hL"governmentaliuy demonstrates to rhe contrary that [hese are roo feamres of a
tA , , __ _ h r Anee „nr reside in rhe state. In rhe
prevented them frota being cowed by the high- handed methods of the bil
o government of conduct, rhe state is only one among a number of heteroge-
reaucraev and jts demands for 1iterare In addition , rhe requirement that
neous institutions and cannot simply be assumed ro be rhe dominant player
ersrtaleI m rrainin ^
rvol k o centers
J
- for thrre months at [he bcginning of thcir ` , C,^° J.
tenure and to rhe office in Mandi for their monthly meeting had severa ) un- ro _ Nor can
it be assumed dar rhe conduct desired by planners, policymakers.

intended consequcnces. In western Uttar l,radesh, women ahnost never made ^b a n d bu reaucrars is aetually achieved for rhe subjects of [hese policies may
well alter rhe nature of rhe programs themselves, and rhus change rhe con-
such trips on their own, that is, without 111C "protection" M no accompanying
. , Or duct of govern ment as nmch as gavera ment changes th em.
malo And yer, because they were required to do so, women ¡ti groups of rwo JJ'.
or more embarked en diese trips. In so doing, They began to traverse a public Vy

space formerly closed to them by rhe constraints of "honor" and rhreats of

gendered violente. Finally, one cannot understate rhe transformative impact
r. The exact amounts were as follows: expenditures in 1990 - 1991 were Rs. 268 crome vs. Rs.
on rhe lives of anganwadi workers of rhe discovery that they could teach,
603 crore in 1998-1999 ( approximately $151 million).
and often teach better than (relatively) highly paid reachers in government 2. What is interesting in chis discussion of governmentaliry is the extent to which the welfare
schools. As described aboye, workers ' greatest source of pride rested in rhe of populations within territorial states is discussed in isolation from rhe wider context in
favorable comparisons made by students and parents berween their centers which [hose srares existed . In particular, if one is thinking of Buropean states in the eigh-
and rhe privare schools that charged tuition. teenrh century, rhe welfare of rhe population within [hose trates was intimately linked to rhe
: their colonial
In addition to [hese features of ambivalente and ambiguiry were instantes welfare o f [ h ose or h er popu l a tions who were rhe object of dreir sute's actions
of resistance that defied rhe monitoring and surveillance functions of thc subjects.
3. For rhe purposes of comparison , rhe UND1' report puts rhe infant mortaliry rase in 1960 al
ICDS program. Suc resi tape ranged from villagers who refused to answer
165 per 1,000 live birihs , indicating that it has been more [ han halved in rhe pass forry years.
survey questions, to anganwadi workers who challenged rhe CDeo at rhe 4. The health componen for children included rhe adminisuation of large doses of Vitamin A,
monthly meetings , to workers who used rhe rules of rhe bureaucracy lo resist ¡ron and folie acid tablets , DPT and Bcc immunization shots, and rhe monitoring of mal-
being fired from their unrecognized " jobs." Resistance also found expression nuaition by weighing or measuring midarm circumference ; for pregnant women, measures

in the workers ' refusal lo see their work at the centers as an extension of rhe included ¡ron and folie acid tablets and tetanus shots.
5. In a survey conducted in twenry-seven of rhe thirry - [hree blocks in which che ices progr:nn
work of child care, cooking, nurturing , and attending to the sick that they did
began, Tandon , Ramachandran, and Bhamagar ( 1881: 380) reported that 76 percent of chil-
at honre. Workers insisted on calling themselves teachers and rhus underlined
dren under [hree years ofage in rural areas and 78 percent of those under six were mainaur-
rhe similariry of their work to that performed by primary school teachers. ished. Severe nralnurrition was fomrd in z1 percent of rural children under [hree and in
I trust it is olear that my emphasis on ambiguiry and resistance is not in- z6 percent oftribal children in rhe same age group.
tended ro efface the state 's surveillance and regulation , rhe productive prolif- 6. Concern with rhe sine of rhe population arose both from fears of rhe inadequacy of rhe food
supply as well as from its potencial to impose demands on sute senices. India's population
eration of sratistics , and rhe preoccupation with questions of rhe popula-
was 441 mullan in 1960, 884 million in 1992 , and has grown te over a billion by rhe year
tion. I have emphasized throughout that governmentaliry is never just about
2000,
control, it is most of all about a concernawith rhe population , with its size, bur
7. The overwhelming defeat of the Congress in rhe elecfions that followed has been anributed
also with its health, happiness, and productivity. It is precisely this relation- by most political observers co the coercive tactics used for birth control during rhe Enrer-
ship berween rhe state's increased capacity for rhe surveillance and control of gency. Sanjay Gandhi's followers often invoked China as an example of "successful" popu-
women's lives and its concern with saving the lives of children, particularly lation control, and democracy was held ro be rhe chief reason for India's failure. For a fas-

girls, and protecting millions of others from more acule forms of malnutri- cinating analysis of China's efforts in Chis field , see Greenhalgh (1999).
8. This issue is analyzed in greater detall lamer in [his essay.
tion and distase (Sen 1990) that becomes hard to grasp in convencional aca-
9. Scheduled castes are the lowest castes in rhe catre hierarchy.
demic discussions that pit the state against civil society. An understanding of

GOVERNING POPULATION 95
94 STATES OF IMAGINATION
 lo. This custom is known as gaunaa in western
Utrar Pradesh.
u. See also the work of Cohn (1987) and Appadurai (1993) en the Indian
state. Appadurai srates
the problem particularly well when he says "Statisdcs were generated
in amounts that far
THE BATTLEFIELD AND THE PRIZE
defeated any unified bureaucratic purpose" (316).
ANC's Bid to Reform the South African State
a. The belief in the "evil eye" is not limited tu people in western Uttar Pradesh, but is common
in large parís of the South Asian subcontinent.
13. There are various versions of this basic formula. 1 have heard the
Steffen Jensen
lame sentiment being
expressed as "Development is the best contraceprive."
14. Although I cannot develop Chis fine of argument
here, I note ehat the use of military and
missionizing metaphors in development projecrs is remarkably
similar to those employed in
colonial conquest: God, guns, and glory.

The new South African state is one in which formal expressions of democracy and
human rights should be backed up by mass involvement in policy formularion and
implementation. It is a state which should mobilize the nation's resources co expand
the wealth base in the form of a growing economy. It is a state which should conrinu-
ally strive lo improve people's quality of life. Such a state should ensure that alI citi-
zens are accorded equal opportunities within the context of correcting che historical
injustice.-African National Congress, Strategies and Tactics

When the African National Congress (ANC) assumed power after South Af-
rica's first democratic election in April 1994, one of the party's main tasks
was transformation of the state apparatus (ANC 1997). The ANC' leadership
viewed the old state apparatus as its antithesis: exclusive, without legitimacy,
and the instrument of an inherently racist white minority. The party set itself
the historical task of transforming the state into a state for all South Africans,
without vested interests and fair. This begets the question, and the main focus
of this essay, of how a political party can bring about reform of the state. I
argue that reform is happening on many different levels, often as unintended
consequences and not as deliberate as the ANC seems to think. Most of the
time, transformation is effected through very localized, not always heroic,
power struggles.
I focus on the transformation of the field of safety and security, or issues
of crime and violence in a broad sense, as a privileged point ofentry. Nowhere
was the former state more compromised than when it dealt with crime and

96 STATES OF IMAGINATION
violence (set Cawthra 1993; Shearing and Brogden 1993). In a way, the secu- wirh something altogether more important : che well-being of all South Afri-
rity apparatus carne ro incarnare che apartheid state in its most condensed cans, and being che keepers of whar is rermed rhe "national democratic revo-

and bruralizing forro. Yet, another reason for analyzing the field of safety and lution" ( 7).' This also influences ANC's understanding of the state:
security, although en a contrary note, is thar crime is pereeived as endanger- The new democratic government is faced wirh che challenge of chang-
ing South A&ica's 1ragile transition to dcmocrace (National Crime Prevention ing [the injustices of rhe past], as parí of irs strategie rask oí creating
Strategv 1996)- It is contrary because prevention oí crime is flor a heroic is- a united, non-racial, non-se x ist and democratic society. In rhe first in-
suc relating ro the purty's srruggle against apartheid. l.ike al] issues of oven- stante, this government derives irs legitimacy and legality from che demo-
doy governance, it is piccemeal and contingenr. When t'he ANC ascended to crarie processes which saw to irs birrh, However, che state machinery we
power, dio parry was quite unprepared to deal wirh [hese piccemeal issues inherited contains many feamres oí che pass . Formally, it is a state based
that were unheroic, often tainting, problems of governance. on a democratic constitution , a state which is obliged ro serve the aspi-
Before beginning my investigation of the more detailed processes of trans- rations of the majority . However, che emergente of a truly democratic
formation, 1 consider some questions pertinent to this problem . A first set of sute depends on the transformation of che old machinery , a eritieal
questions evolves around how che ANC conceives irs own role in transforma- part of the NDR [national democratic revolution ]. Such transformation
tion and whar perception it has of its antagonist, the apartheid state. A sec- should see che location of the motive forres of the revolution at the helm of the
ond set of questions evolves around the character of chis apartheid state, not state, as the classes and strata which wield real power . The challenge rhat
leas[ the security forres, the police, and how they maintained order. Finally, 1 faces [hese forces in this phase is to ensure that the elements of power
go to the murky and contradictory world of Richmond in the KwaZulu Natal they have captured are utilized rapidly to transform the state , while at
Midlands, which was and is one of the most eontested arcas of South Africa. the lame time placing it at the centre of the transformation of South Afriea's
A virtual civil war has been fought during the past two decades among the political, economie and societal relations. (ANC 1997; enrphasis added)
apartheid police, the ANC in its different forros, che conservative and ethnic-
What is clear from che quote is that the ANC believes in the state as a tool to
based Inkatha Freedom party ( IFP), and, of late, the United Democratic
be employed by a political party in the interest of the people . Here, we hear
Movement (UDM). I look at Richmond because iris exemplary of the contra-
clear reminiscences of a state controlled by a vanguard organization-much
dictory nature of South African politics, both before and after 1994 . Only alter
like a Jacobin idea of a "few good men" at che helm of the sute.
dealing with [hese issues do 1 venture into analyzing transformation in more
This Jacobin understanding of che state also informs the ANC's perception
detail. But let me begin with the ANC.
of the apartheid state, on the one hand, as designed solely to preserve a white
minority's privileges , and, en the other hand , as omnipotent , reaching from
A Historical Charge top to bottom of society. Apart from explaining that the security forces were
particularly successful in hunting down activists and infiltrating ANC struc-
In che Strategy and Tactics paper from the fiftieth congress, the ANC states the
tures, che image of an omnipotent state as che adversary of the ANC also gives
following about irs role in South African society: "The ANC is ... called upon
che parry the legitimacy to rule and reform.
to win over to irs side [hose who previously benefited from che system of apart-
heid: to persuade them to appreciate that their long-term security and comfort
are closely tied up wirh che securiry and comfort of society as a whole. In chis A Draconian State
sense therefore, the ANC is non a leader of itself, nor just of its supporters. His-
Especially within the ANC, the apartheid state is legendary for irs abiliry to act
tory has bequeathed on it the mission to lead South African society as a whole
relentlessly, without hesitation or remorse. The political will, wielding che
in che quest for a truly non-racial, non-sexis[ and democratic nation" (ANc
state, decided to move three million people under thc Group Arcas Act-and
1997: ra). Thus, che ANC is non simply a political organization but is charged

98 STATES OF IMAGINATION ANC'S RID TO REFORM 99

 it did. Ir decided to lift a vast number of white Afrikaners out of deep pov- in military-style compounds, and on duty in the special forces in the town-
erty-and it did so in just thirty years ( Norval 1996 ; Sparks Iggo
), and did so ships for months on end, white police officers also lived out a colonial police
to such an extent that , as Malan sarcastically writes, "Only the durnbestwhite
officer existence . According to more personal statements from police officers
men failed to prosper" ( 1990: 129).
This happened on the backs of nonwhite working in the riot units , this isolation distorted the worid and turned all
people in a system of racial exploitation, but it was also effected through a blacks into enemies .' Also a signature of colonial policing, vigilante groups
radical expansion of the state apparatus that made room for the employment were armed and trained to target mosdy ANC -aligned organizations like the
of Afrikaners in a gigantic afrmative action move. Thus, at the end of the UDM (Cawthra 1993: 109-26).
Ig8os, about half the Afrikaner workforce was employed in the state sector The apartheid state also employed other counterinsurgency strategies be-
(Saff IggS :
a1). A significant number ended up in the security apparatus, ei- sides mere repression and violente . Up through the Ig8os , the so-called total
ther in the police or in the military, with a considerable stake in upholding strategy ' or "Winning the Hearts and Minds" (WHAM ) strategy became in-
the system of apartheid.
creasingly important . When an area had been "calmed down," a concerted
As Cawthra writes, the police and the military " carne to regard thernselves effort was to be made to win the population of the area over to the govern-
as defenders of the 'free world' and 'Christian civilization' against the com- ment's point of view through substantial social and economic development
munist threat " ( 1993: z). This made the security apparatus
' s prime task po- (Sparks 1990: 354).' As hindsight tells us, this strategy backfired. The colo-
litical and, as Cawthra writes, "
overlaid with military imperatives " ( z), with nized people rejected the offer ofinclusion in this form ofselective economic
military-style wars waged against liberation forces, first in the frontier states development, and instead of "stabilizing " the areas , they often became more
of what are now Zimbabwe , Namibia, Angola, and Mozambique and later
violent as the security forces clamped down even harder (Norval 1996). This
inside the townships of urban South Africa. The
entire political task was ar- made activista cal) the strategy WHAM BAM. The way the strategy was imple-
ticulated through a Christian - nationalistic discourse of apartheid
. As Minister mented, with a highly arbitrary allocation of resources , only increased the
Kruger said in 1977 about the South African police, shortly after the interna- levels of violence and interna) strife. The security forces were not too unhappy
tionally condemned Soweto massacre where many, mainly children, had died: about chis, as the strategy also aimed at destabilizing antiapartheid structures,
"It is so that the all-knowing God thought fit to exercise His authority on but as we shall see below in relation to Richmond , the security forces also
earth using the service of people, parents , officials of the Government
. There- lost control to a large extent.
fore the wearers of the fine uniforms of che police are also the mandate hold- The picture that emerges is one of a police force enmeshed in political
ers of God " ( in Argus 1977 ;
qtd. in Shearing and Brogden 1993: 41). counterinsurgency toward a colonized majority of nonwhite people. This in-
Although more than 5o percent of the force was nonwhite ( Shearing and cluded a minimum of crime prevention and detection (Cawthra 1993: 5). Most
Brogden 1993 ), Afrikaner nationalism was prevalent within the police force.
efforts were aimed at inciting violence and quenching the first of an increas-
The style of policing, developing " from bush-war to counter
- insurgency" ingly hostile population.
(Cawthra 1993 ), resembled colonial policing where much of the actual vio-
There were, however, important differences . Policing in white areas was
lence was carried out by nonwhite help troops ( in South Africa called similar to policing in most other Western countries ,' with one big difference:
kítskon-
stabels or special constables ), who by their acta were alienated even further
white people were told to police their own areas as well. For that purpose, the
from their own communities . Poverty brought them to the force and alien-
government set up the Nacional Security Management System, encouraged
ation kept them there . That blacks policed blacks also went well with the
neighborhood watches, and let private security companies play a still bigger
ideology of separare development, where each population group should gov- part in policing, all of which resulted in a profound militarization of white
ern its own affairs. But it was not only the black police personnel that lived South Africa (Norval 1996: 24-47).
the lives of colonial police officers . Drawn from the most vulnerable circles
Nonwhite people were also policed for different purposes . One example of
of Afrikaner society ( Cawthra 1993 :
3), given a special mission, often isolated chis was that the policing in the "coloured" 7 areas was directed toward crimi-

100 STATES OF IMAGINATION ANC'S BID TO REFORM 101

sal offenders. This was carried out to such an extent as to turn che coloured Zulu-based Inkatha (which, with che advent of democracy, became IFF). Over

population hato what must be one of che most imprisoned populations in che che past two decades, this virtual civil war has cost thousands of people thcir
lives and repeatedly made even more luto temporary refugces as they tried to
estire world (Pinnock 1984: Midgley 1975)' What is important for our pur-
escape che violence (e.g., Goodenough iygga, i9996; Strudsholm 1997). In
pose is that although few coloureds were charged wirh political offenses, che
methods employed by che police were not ven different from [hose employed 1987, Nkabinde organizad rallies and led groups of young Inkatha supporrers

in che policing of political insurgente Thus, Fernandez (1991) has docu- against the supporters of che UDM (the interna) wing of che then exiled ANC).

mented how che police in the Western Cape employed the most gruesome This continued until 1989, when he was talked luto joining ihc ANC by the

torture (electricity, rape, beatings, animals, etc.) against suspected coloured ANC strongtnan in the KwaZulu Midlands, ilarry Gwala. Gwala cnjoycd much
influence internally in che ANC, with friends right up to the top of che part<.
crim inals.
To summarize, the police had wide-ranging powers that could be used with Nkabinde's decision to change his allegiance was probably the turning point

discretion on the parí of the individual officer. These powers were employed for the ANC in che KwaZulu Midlands. In the famous Baffle of che Forest in

in che policing of nonwhite people in the harshest and most violent way, March 1991, Inkatha, which until then had controlled most of Richmond,

regardless of whether the offense was political or criminal. Consequently, lost twenty-three men (Goodenough 199ga). As che years passed, rumors of

vast numbers of nonwhite people were imprisoned, beaten, tortured, or Nkabinde's criminal links and violent exploits, also against his own, became

killed, and many political organizations were infiltrated and often destroyed more and more frequent. The morder in 1994 of che Midland's ANC youth

(Fruth and Reconciliation Commission [TRC], vols. 2 and 3, ig98). It is partly leader, Mzwandile Mbongwa, was indicative of how Nkabinde turned his vio-

because of this that che image emerged of an omnipotent apartheid state, lence against his own instead of che now largely defeated ¡F1,. This morder

controlled by purpose and will of an Afrikaner elite. This (ANC) perception of was committed by Nkabinde's SDO (Self-Defense Unir), which by then had

che apartheid state, however, is rather oversimplified. To exemplify chis, lec become a small army accountable to no one but ehemselves.°
me investigate how South African politics were played out in Richmond, one During che 199os, ever more people were displaced from their honres by

of ehe country's most contested localities. This also provides a sinister illus- che violenee, and ever more people were killed. In 1997, Nkabinde was ex-

tration of che problems facing any attempt to transform South African sociery. pelled from che ANO after having obtained a seat in che provincial legislature.
The ANC finally moved to expel him after rumors that he had informed en che
ANO during the latter parí of the 198os and in che 199os. The problem for
Sifiso Nkabinde: The End of Politics? the ANC was that ihey had known for six years that he informed en them. In

On January 23, 1999, Sifiso Nkabinde, one of che most famous South African an interview, lacob Zuma, the deputy chairman of the ANC, explained that che

political figures, was murdered in Richmond, KwaZulu-Natal. This marked ANC only very seldom expels people but tríes te deal with them inside che

che end of a remarkable saga in South African history. In August 1999, che organization (Strudsholm 1997). It is also fairly well substantiated that che

Maíl and Guardian could report thar che local police had known about the sus- ANO knew about che violente being inflicted on their own, as in che case of

pects for four montlis prior to an acres[ ("police `Sat en' Nkabinde Docket" Mbongwa. Strudsholm suggests that the real reason che ANC failed te expel

1999: 2). The newspaper could also reveal that che local police even had dock- Nkabinde was fear of losing support in the KwaZulu Midlands if they did.

ets on che suspects but had failed to arrest them, and that ir was only when Another explanation is that his mentor, Harry Gwala, had been able to protect

there was a change of guard that rhey were arrestad. As ir turned out, ihey him until then. After Gwala's death in 1996, nobody in the ANC world exon-

were all closely linked to che ANC, one of them a bodyguard for the mayor of erare Nkabinde anymore.
After Nkabinde was expelled in 1997, he joined the ANC's new rival party,
Richmond, Andrew Ragavaloo (2).
Sifiso Nkabinde began bis political career in 1987, when he took parí in che the UDM, where he obtained a position as national secretary general.- Soon

virtual civil war of Richmond en che side of che conservative and ethnically after, his Richmond past caught up with him again, and he was charged with

102 STATES OF IMAGINATION ANC'S BID TO REFORM 103

 sixteen cases of murder. To the dismay of the ANC
, he was released from were instrumental in the creation of the self-proclaimed warlord ( Strudsholm
prison a year later due to lack of evidence .
The ANC began talking about 1999). In other words, a picture emerges in which the statements from the
"Third Force " elements in the old state (
see below) that had an interest in
fiftieth national congress , quoted aboye , have no resonante . This schism be-
fueling the violence in the Midlands. The role of the police was certainly far
tween ANC discourse and practice has not passed unnoticed , as testified to in
from admirable . The Human Rights Committee
( HRC) and the Network of
the following statement by the leader of the UDM , Bantu Holomisa, com-
Independent Monitors (
NIM) investigated the issue of police involvement in
menting en the role of the ANC in the violence in Richmond : "It should be
the Nkabinde saga. The report preves again and again that specific police
noted that while the ANC leaders are parading the Capitals of the world as
officers indeed had helped Nkabinde .
The allegations ofhis informing for the human rights activists, they en the other hand had taken the decision to
police were also proved beyond doubt ( HRC/NIM 1998 ).
Minister of Safety eliminate their political opponen [s by instructing [heir juniors to do their dirty
and Security Sydney Mufamadi and National Commissioner of the Police
work " (1998: 8). Holomisa's comments must, of course , be seco as an at-
George Fivaz closed down the Richmond police station in 1998, after yet an-
tempt to exonerate Nkabinde and UDM of guilt in the violence in Richmond,
other killing spree in Richmond following Nkabinde's release. This was not
but Holomisa makes a further charge. Contrary te its own discourse , [he ANC
the first time police officers had been removed from the station .
The same is not acting fairly and in the interests of all South Africans ; they are involved
happened in 1997, and lince 1995 all investigations into political violence
in local politics and, as such , are parí of the violence in Richmond.
have been handled by different national units ( HRCINIM 1998 ).
All Chis oc- The ideological construction of an omnipotent state resisted by a hernie
curred because there were serious indications that the police
" worked hand movement dissipates when confronted with the murky world of local Rich-
in hand with Nkabinde," as the ANC put it in their motivation for the removal
mond politics . Although the police have now arrested the suspected ANC-
of the head of the Richmond investigating officers, Johan Meeding ( quoted
aligned killers of Nkabinde, many other people had reasons for wanting Nka-
in HRCINIM 1998: 31).
binde dead: he was involved in a taxi rank dispute;" he had grievances with
But the ambivalent role of the police is not restricted to the fanning of
the provincial leadership inside his new party, the UDM, whom he had man-
violence. Before 1994, the arcas controlled by Nkabinde were for al] purposes
aged to sideline ; and he had a dispute with his own bodyguards , whom he
no-go arcas for the police, who had to obtain permission from Nkabinde to
had apparently just fired ( Strudsholm 1999 ). Everything points to the realiza-
enter Richmond. This changed when the police and Nkabinde made an agree-
tion that politics in Richmond cannot be interpreted as a simple bipolar con-
ment at a Community Policing meeting in 1994. Suddenly, the police had
flict between the ANC and different antagonists , as political and academic
"freer" access to the arca . In the words of Meeding ,
the police " went from a commentators usually do (e.g., Goodenough 199ga, igggb; HRCINIM 1998).
caseload of 400 unsolved cases to loo . The solving rate went up from
This is no[ the "end of politics," but another kind of politics , far from the
zo percent to 8o percent " ( HRCINIM 1998 :
31). The police offlcer who finally politics envisaged by the ANC leadership . It is contingen[, fragmented, and
arrested Nkabinde en the sixteen counts of murder "noted that he had the
essentially determined by the local setting.
impression, Nkabinde and bis SDU led the Richmond police by the nose" (31).
This indicates that not only did the police use Nkabinde ;
he used them as
well. No doubt the police, and before 1994 the apartheid police, helped create Postapartheid Politics Revisited
Sifiso Nkabinde for the purpose of destabilizing the different antiapartheid
Richmond may be an extreme example of the intricacies of South African
organizations . Involuntarily,
they also facilitated the development of a niche
postapartheid politics, but it is by no means unique. In other places as well,
in which Nkabinde was allowed to roam freely ,
and, as time went by, they lost ANC people were involved in local power struggles. Given the social and his-
access te Richmond and control over him. In the end ,
they, like everybody torical complexities , there is even a certain inevitability te this. Therefore,
else in Richmond , were at the merey of Sifiso Nkabinde.
rather than questioning the ANC's political legitimacy (only the South African
By disregarding his brutality and connections with the police, [he ANC
electorate can do that), the reading 1 propose is one that sees the ANC as

104 STATES OF IMAGINATION

ANC'S BID TO REFORM 105
 leadership always conflrms the support of their own people, as they did with
having to engage in a number of ofien localized and fragmented power strug-
Nkabinde sis years before he was expelled, when he became counterrevolu-
gles, emerging as the antagonist of the omnipotent apartheid state and therc-
tionary. Holomisa reflects Chis when hc says that whenever somebody re-
fore rhe transfbrming force of society per se. Ler me look at two related ways
minds che ANC "of their unfulfilled promises to the masses, they are termed
of asserting che ANC'S position as "Leaders of the Transformation."
counter revolutionary" (1998: 9). 1 lolomisá s charge that it is "a tonn of de-
One way of doing Chis has been to celebrate the antiapartheid struggle in
fense mechanism by the ANC" is less important here. What is important is
such a way as ro underline the achievements, uniry, and courage of the mem-
that the ANO does not seem m have a public language tu deal with the intense
bers in rheir struggle against the omnipotent state. The public holiday calen-
local power struggles hut must revert ro che bipolar language of che struggle
dar sinec 1994 offers an insight finto deis form of celebration. As [he ANO
against apartheid. Thus, [hose who are with the ANC are, by dehnition, in
assumed power, most Christian and all apartheid holidays were abandoned.
favor of transformation, and those who are against the ANC are opposed. The
Instead, South Africa now celebrates Human Rights Day, Freedom Day, Work-
ANO knows that Chis categorization is problematie, but publicly, they must
ers Day, Youth Day, National Women's Day, Heritage Day, Reconciliation Day,
appear undivided behind transformation.
and the date of the 1999 election. All [hese national holidays celebrate the
This means that the past very much informs che present (postapartheid
new South Africa with its respect for its citizenry-all in opposition to the
politics), both discursively and in actions: discursively because che past is con-
former regime.
standy recycled as struggle narratives, and in actions because old alliances
These holidays, as well as the national policies launched at the fiftieth
and che modus operandi of the struggle against apartheid are nos easily for-
congress, for that master, are what we might term state spectacles. They
gotten. Thus, there is a great deal of inertia in che ANC stemming from che
are nos only for the consmnption of the electorate and signs of change, bus
time when che ends (che defeat of apartheid or mere survival) justified che
are also meant for interna) consumption in che state apparatus and the
means. This is necessary to keep in mind when evaluating che ANC's attempts
ANC. Their purpose is to instill in the ANC and the bureaucracy che feeling
te fulfill its historical mission.
that they are part of a "new thing" in delivering equal services to the people
and generally working for rhe comnon good of the people (ANC 1997).
Furthermore, che state spectacles have che purpose of appeasing al¡ che lo- The State Concepmalized: The State as
calized power struggles that constantly tear at the party's image as a uni- Battlefield and Prize
fied body. 1 have attempted to problematize the ANC's implicit understanding of state-
The fiftieth congress in Mafikeng is an illustration of this. Before che con-
hood, by which che state is perceived as potentially omnipotent, be that for
gress, che newspapers abounded with stories of interna¡ strife in the party,
better or for worse. 1 have also questioned che general assumption that the
and it was with a fair amount of trepidation that the party commenced its
struggle was an affair between the ANO and che apartheid regime. Instead, a
congress. Thus, the Electronic Maíl and Guardian could ask "Can New Leader-
much more complex picrure emerged.
ship Heal che ANC?" (1997e). Afier the congress, Minister of Environmental
In line with [he introduction to chis volume, 1 propose a double perspective
Affairs Pallo Jordan exclaimed chas "the conference demonstrated maturiry of
on che state inspired by Foucault and Laclau respectively. 1 agree with a Fou-
the ANC's membership who were not side-tracked and distracted from their
cauldian interpretation (e.g., Foucault 197ga) of the state as a set of frag-
course by demagogy and posturing by anybody" ("The ANO Triumphs at
mented, noncoordinated institutional practices, contested through local-leve)
Mafikeng" 1997). On che one hand, [hese state spectacles give che image of
power struggles. But the bureaucratic knowledge and experience inseribed in
the new state as normalized in contrast to the apartheid state, which the ANO
institucional practices produces an inertia within che institutions that is very
viewed as pathological. On the other hand, the spectacles are aimed at por-
difficult to change. It is these institucional practices that che ANC must engage
traying the ANO as "a polirical will" shas had taken over the reins of govern-
with throughout che state apparatus.
ment and now determined the course of the state.
Institucional inertia has yer another twist to it in South Africa: the negoti-
The ANO has dealt with the contingent local policies in yet anotherway. The

ANC'S BID TO REFORM 107

io6 STATES OF IMAGINATION
 ated settlement. This settlement brought the ANC to power in a fairly peaceful
manner. It emerged during the period up to the election in the country ungovernable. Furthermore, as we saw in the case of Richmond,
1994, when there
were very real fears that South Africa wouid descend roto civil war. The settle- violente was a legitímate means of securing areas under their control. Only
ment was a compromise in which different positions had to be balanced, and alter 1994 did the party begin seeing crime and violente as issues of order.
the ANC leadership couid not dictare the settlement. Under the threat, en che There were several reasons for Chis change in approach. First, violence and
one hand, ofAfrikaner right-wingviolence, and on the other, of an imminent crime undermined the parry's legitimare claim to rule. Second, large seg-
bureaucratic collapse, the party had to accept that it could not terminare loyal ments of the South African population became preoccupied with crime and
employees of the former regime. As a matter of realpolitik, the negotiated violence as the most serious problem facing sociery.'° That the ANC now had
settlement basically stated that their experience from the apartheid era was ro deal with violente and crime as an issue of order (as opposed to ungov-
useful as apolitical bureaucratic expertise. South Africa had to ` let bygones ernability and revolution) and the ascension of crime and violence unto the
be bygones," as Mandela (1994) put it in his inaugurarion speech. Although political agenda meant that the party had to rely un the despised security
rhere have been serious attempts to deal with the past, most notably through apparatus, which had been predominantly involved in counterinsurgencyand,
the Truth and Reconciliation Commission" (see Lars Buur's and Aletta Nor- because of the negotiated settlement, remained largely unreformed, even ex-
val's pieces in this volume), the past is still a source of tension that informs onerated and professionalized. As an illustration of how difficult that made
the power struggles over transformation in both implicit and explicit ways. the transformation process, Minister of Justice Dullah Omar expressed deep
But the state is not only institutionalized knowledge and practices. It also mistrust to his own staff at the beginning of his term in 1994: "1 cannot trust
has, following Laclau (1985: rao-rz), a symbolic function of forming a kind anybody here" (interview with former staff member, October 1997). It was,
of moral high ground. Thus, by taking control of the state, the ANC can per- in other words, a highly complex terrain in which the party had ro effect
form the task of spearheading the NDR of society. Thomas Blom Hansen change-not only of the system but also of the levels of violente and crime.
talks of something similar when he asserts, in his contribution to rus vol- Let me now turn to an ethnographically detailed examination of the pro-
ume, that there is a constitutive split berween a profane dimension of the cesses of transformation and how the ANC attempts to effect change. 1 take
state-incoherent, brutal, and pardal-and a sublime state principie. In this my point of departure (as one among other possible examples) in the ANC's
sense, I discuss the state not only as a battleground but also as a prize." attempt to imagine and creare new, unambiguous, and normalized institu-
Thus, one must have in sight both dimensions of statehood, both as frag- tions, in caso the Secretariat for Safety and Security. This institution, the new
mented power struggles over institutionalized practices and as sublime state government's attempt to exercise some civilian, political control over the po-
principie, when analyzing the ANC's reform endeavors. But one more lice, was soon enmeshed in local political power struggles. It is [hese power
issue is
important to consider: transformation depends on how the field to be re- struggles that 1 propose to follow. To the end of the essay, 1 will show how
£ormed is perceived, problematized, and categorized. In other words, the ANC leadership seeks to exonerare the blame for present failures, com-
it mat-
ters whether it is reform of welfare policies or the security apparatus. Let me bating crime and transforming the police, by recycling the problems through
therefore briefly consider the field of safety and security as the ANC began to "struggle [Third Force] narratives."
conceive it when the party ascended te power in 1994.

The Secretariat for Safety and Security:

Violence as a Problem of Governance Eradicating the Past through Institution Building

Before 1994, the ANC never really conceived of safety and security as a prob- If new institutions are to be created in accordance with the negotiated setde-
lem. Violente and crime were primarily read as a necessary consequence of ment's principie of reconciliadon, they must not be seen as creating or rein-
the unrest in the country, and as such fit well with the ANC strategy of making forcing divisions. The creation of the Secretariat for Safety and Security is a
case in point. The official purpose of the Secretariat was to creare "civilian

108 STATES OF IMAGINATION

ANC'S BID TO REFORM 109

oversight" of and a policymaking mechanism for the police, therehy dealing mand that their professional expertise be acknowledged. In other words, thc
with the most serious problems of the past. The Secretariat's policy director Secretariat works for reconciliation and does nos reinforce tire divisions ot
asserrs: "Ir vas a response to ... [the historical problems of the police] that rhe pase Furthermore, the police, as a professional state agency, are partly
a very strong oversighr was created, controlled by severa¡ people, demoerati- exonerated of past guilt in Clic sense that the experiences obrained in the pass
cally elected through our democratic process, with a minister directing in his are useful. The police can work democratically and impartially for the benefit
capacity to analyze and ro formulase policy" (interview with Desiree Daniels, of al) South Afrieans. What is asserred here is rhat the police Nave mended

September 1997). their ways, andlor that rhe transgressions of che past were the decds of a few
In other words, the hitherto unaccountable police [orce was now to be sub- wayward individuals. Chas, the now cleansed police should be able ro work
mitted to polirical priorities. The priorities were not only about controlling in the interest of the entire South African population.
the police but also aimed at changing the focos of the police force. In line This is the official, nonconflictual version: politicians in the ANO-led gov-
with the National Crime Prevention Strategy (NCPS 1996), which outlined the ernment working together with the police in the pursuit of the common goal
new government's all-inclusive plan to combat crime in ways departing from of ridding South African society of rhe curse of crime in a democratie and
past atrocities," nonwhite South Africa should be policed in the same way as aecountable way and with a clear understanding of who is doing what. How-
white South Africa: accountable and service- and victim-oriented in direct ever, just under the nonconflictual surface, confusion and disagreement seem
comrasr to the hitherto colonial-style policing, where the focos was en coun- to reign-not least in understanding and coming to terms with the role of
terinsurgency and control. This also meant a reallocation of resources and the Secretarias "From the beginning there was an adversarial relationship be-
personnel. To bring this about, the national and provincial Secretariats were tween the police and the Secretarias With che passing of time, its role will be-
given rhree roles: policy planning, monitoring of implementation, and coor- come clearer or less needed. From the beginning the Secretariat did as much
dination. Also, the police accepted the construction with civilian polirical policy as it could and the police continued oblivious and unconscious, with-
oversight formulating policies and with the police as operational entity. In a out caring what the Secretariat was doing" (interview with Wilfried Schdrf,
status report, the police describe their relationship with the Secretariat: "The Institute of Criminology, University of Cape Town, October 1997).
Departlnent `South African Police Service' has been renamed the `Department In other words, the Secretariat had great difficulties asserting itself vis-á-
of Safety and Security,' consisting of the South African Police Service and a vis the police. At the institutional level and in terms of competencies, the
`Secretariat for Safety and Security"" ... distinct and separare of the opera- Secretariat had no direct influence over the police. Thus, the national com-
tional component. This Secretariat comprises civilian employees, thus estab- missioner reports directly to the president rather than to the minister ofsafery
lishing civilian oversight and enhancing the democratization of the Service" and security. This means that the role of the Secretariat is merely consultative,
(South African Police Service [sAps] 1997a: 31). or, as the police plan puts it, "to `direct"' the commissioner (SAPS 1997b: 28;
This is echoed in the Police Plan 1996/97, where the question is asked quotation marks in original )." Due to the ambiguities, the minister and the
whether this relationship with the Secretariat will not lead to a politicization national commissioner have both fought over the right to leadership and over
of the police (something thar has a bad ring to it in South Africa since apart- the blame for the failure to combat crime. These power struggles have been
heid). This, the plan asserrs, has two aspects: "On the one hand it is the role fought out in the press but also through secret investigations. The ANC min-
of politicians to distil the wants, demands and needs of the communities." ister's chief accusation against the police has been their "inabiliry to trans-
The politicians must, however, "take seriously the professional judgment of form," whereas the police's main problem has been what they see as "poliri-
skilled and experienced police officers in rhe determination of policies" (SAPS cal interference" in their work'° ("Fivaz Takes on Mbeki Police Squad" 1997).
[9976:28). The image emerging from these power struggles stands in stark contras[
It is evident that the police accept the civilian oversight and prioritization to the offrcial version of the new instlmtions, where politicization is no prob-
as necessary in the democratization of the police; at the same time, they de- 1em and the police are basically innocent. The power struggles are embedded

110 STATES OF IMAGINATION ANC'S BID TO REFORM III

 in past antagonisms between the ANC and the police, where the Secretariat is delivery action mechanism). It was launched in '998 and focused on ten areas
part of the ANC's attempt to democratize and control the police. This means
of concern: marine poaching, the festive season (December to February),
that forgetting the past to work for a new South Africa has some inherent
street children, safer schools, Cardboard City (a squatter area in the middle
contradictions that make such an operation very difficult.
of Cape Town), surveillance cameras in the inner city, the Safer Cities projects
However, we cannot assume that the creation of the Secretariat did not have (report back), rent-a-cop (where citizens pay for a hobby en the beat), and
any transforming effects. Wilfried Schdrf asserts that it is an ongoing process
different initiatives under the Ministry of Justice (interview with Ronel Schoe-
that does not end in insurmountable antagonisms. Thus, "as the Secretarias man, January 1999).
grew aware of how the police operates and how changes within the police are
When looking at the list of priorities, one realizes that little attention is
brought about, the Secretariat began establishing key parmerships with opin-
given to what could be termed nonwhite problems of crime.'° My fieldwork
ion changers within the police" (interview with W. Schárf). The Secretariat
in 1997,'998, and 1999 and various crime statistics (e.g., Institute for Security
also moved in another direction. Throughout 1997, a number of provincial Studies [ISS] '998) indicate that these have to do with taxi violence, train
crime prevention summits were held at which the Secretariat asserted its po- violence, gangs, vigilante groups, shebeens (informal liquor outlets), and cor-
sition as the lead department. By having a closer look at what carne out of
rupt, brutal, and inconsistent policing. None of these problems is addressed
these summits, we can learn something about both transformation and the by the province's unit for crime prevention. Furthermore, civilian oversight of
work of the Secretariat in terms of priorities and functions. The Western Cape the police has been tened down considerably from its original center-stage
provides a good illustration, as the province is run by the National Party. Thus, position. This change from the Secretariat's inception as being charged with
the inherent problems of state transformation in terms ofinstimtional inertia both safety and security, that is, making the police more safety-oriented, is
exist in a condensed form there.
also evident in the recent renaming of the Secretariat from Safety and Security
The summit in the Western Cape was held in October 1997. Out of it carne to Community Safety. How can we explain this apparent shift in approach to
a number of projects and programs aimed at crime prevention. As an em- its functions?
ployee of the Western Cape provincial administration pus it: "Originally the There are two factors we need to look at: first, the political environment in
Secretariat was to function as civilian oversight and to do support functions. which the Secretariat works; second, how resources are allocated." The Na-
This is not so much the function anymore. Now the focos is on crime preven- tional Party governs the Western Cape, and the MEC (member of the Executive
tion much more than monitoring. That means that there is no longer the Committee) of the Department of Community Safety is Mark Wiley, a former
same leve) of contradictions between the police and the Secretariat" (inter- member of the armed forces and white. According to some people Glose to
views with Ronel Schoeman, March 1998).
the department, he has been promoting issues dear to the hearts of the white
Of course, it is difficult, Schoeman goes on, to separate the two completely, constituency in spite of promises to do the opposite, and has de facto ne-
bus the emphasis has changed from civilian oversight to a more socioeco- glected or stopped most projects that were aimed at what could be termed
nomic approach, referred to as "programs" or crime prevention. Schoeman nonwhite priorities. As an NGO person said during an interview, for ex-
stresses that the approach adopted by the Western Cape Secretariat has been ample, the gang commission aimed at the rampant gangsterism, particularly
implemented as the model for the rest of the country's secretariats. Thus, the in coloured areas in Cape Town, was neglected to such an extent that it fell
secretariats have forged a space in which they can have a role beside the po- fiat without ever achieving anything (interview with Gaynor Wasser, Western
lice, without being in opposition to them. This choice is, however, neither Cape Anti-crime Forum, April 1999).
uncontested nor politically neutral, and especially not in an institution con- But the racialization of the Secretariat goes right to the core of the depart-
ceived by the ANC in a province ron by the National Party."
ment. As one informant said, "It is a very racially divided department. They
One of the most important programs in the Secretariat's crime prevention just hired two more whites in 'programs.' When the department was estab-
unit's plan to defeat crime was the so-called MADAM structure (multiagency lished they secured it with white people en all the director posts." This is a

112 STATES OF IMAGINATION ANC'S BID TO REFORM 113

 priorities, and although che administration has worked roward establishing
typical feature of Western Cape provincial administration. Most employees
a nonconflictual relationship with the police, a space vas created wirh the
are either Afrikaners or coloureds , and the maro language is Afrikaans. As rhe
Nacional Party, is still hrmly in control, little has changed since che transition inception of the secretariats in 1994 by the nacional ANC government. New
prioririess cannot eliminare this space complerely, mcaning thar che formen
ro demoeracy.21 Of course, ir is nor evident that people with a stake in thc
activists in civilian oversight have a plarform from which ro engage in power
former regime are incapable of focusing en nonwhite issues, bur there cer-
tainly secros to be a correlation benveen che two, and maybe especially so in struggles over policing priorities in the Western Cape. Here, it is importanr
to note rhar rhese power struggles among different parts of che provincial
the case of the Secrerariar and Mark Wiley. As anorher informant put ir, "Mark
Wiley refuses completely to level criticism against any man in a uniform, be administration, traversed as rhey are by (Western Cape) politics, cannot be
explicitly political but must take orher forms. Most often, these are depart-
it military or police. But also , he is afraid of working with any structure that
just smells of ANC. Although most people in the coloured arcas voted for the mental investigations 23 into lack of professionalism, bad administration, and
malpractice. These strategies of deprofessionalizing the political adversary's
National Party, most of the neighborhood watches and civics are in some way
work are necessary because the new state is not supposed to be entangled in
associated with the ANC . They don't want to strengthen these strucrures be-
cause they are afraid of losing the election [on ) une 2, 1999]." the politics of the past, but rather is to be a professional bureaucracy. How-

The way the supposedly disinterested bureaucracy is traversed by politics is ever, although the Western Cape is heavily dominated, both on the political

specific to the Western Cape, with the result being a lack of prioritizadon of and bureaucratic levels, by people from the former regime coupled with the
discourse of a depoliticized bureaucracy, the secretariats conceived by the ANc
nonwhite arcas . Furthermore , most of those coming from the former regime,
national leadership Nave opened a space that cannot easily be disregarded,
whether whites or coloureds, have very little knowledge of the townships,
where what 1 have called nonwhite issues are prevalent . Ir is thus difficult to and can be usad in the reform of the police and che state.

do anything about crime and violente in these arcas . However, we should not
confine these problems of prioritization to the Western Cape. Many of the Third Force: Recycling the Past as Political Strategy
problems exist elsewhere in South Africa , although maybe not in the con-
densed form of those in the Western Cape. Both in the section en Richmond and in the analysis of che local power strug-
However, the Secretariat for Community Safety is not only white . Although gles over transformation, 1 refer to che perception that che police are not only

underresourced and spatially marginalized at the bottom of the provincial against transformation but are actively working against it through violence.
In South Africa this is callad Third Force. The concept emerged to designate
administration, the Directorate of Civilian Oversight (employing only former
how agencies within the apartheid security apparatus fanned violence among
activists) has managed to maintain a focos on nonwhite issues. Among these
are police actions in relation to rape in Khayalitsha (a huge squatter camp different nonwhite organizations. The fight between the ter and the ANO in
Richmond is one such example. This kind of Third Force actvvity bogan in
rwenty-five kilometers outside Cape Town ), attempts to build partnership
with che neighborhood watches in nonwhite arcas in a bid to avoid their turn- che 1g8os to split the liberation (orces, and continuad en through the 1990s
in an attempt te destabilize the ANc and show che party's inability to govern.
ing into vigilante groups ( which some of them already are), mediation be-
tween police stations and nonwhite communities , and complaints over mis- There is substantial evidente that these Third Force allegations were trua. In
Richmond, there is no doubt that the police worked with Nkabinde (MRCI
handling of junior officers ( from a list of Directorate activities , April 1999).
NIM rgg8) and that the police trained and armed IFr supporters in KwaZulu
In other words , despite political , practical , and personal reluctante to police
the townships without armed vehicles and riot squads , there is scope for an- Natal (e.g., Ellis 1998; TRC 1998). However, che story from Richmond shows
that che police often lost control of their own proteges. Furthermore, the
other approach.
Although the political leadership of provincial administration generallyat- corrupt police officers in Richmond and their cooperation with Nkabinde do
not amount to Third Force in che sense described aboye. In the present situ-
tempts (whether by design or not) not to allocate many resources to nonwhite

114 STATES OF IMAGINATION ANC'S BID TO REFORM 115

 ation, one might say that the concept of Third Force has been "deterritorial-
pointing to specific individuals or specific networks che government and the
ized." It no longer signifies anything concrete but has become an "empty
signifier" '• (Laclau 1996 ) used to explain the inexplicable ANC blame the entire police force for preventing them from reaching che goal
, such as why crime
does not decrease . See, for instante , of a crime-free South Africa . In chis way, che government dispels the notion
this shortarticle in the Electronic Mail and
Guardian: that the police have reformed and that only a few " rotten apples " are left. A
group of the enemies of transformation has been created lo which all che
Safety and Security Minister Sydney Mufamadi is to investigare claims "bad" things can be referred , thereby leaving che ANC and che government
that prisoners are being used as hitmen as part of "Third Force" opera- blameless.
tions en the violente-torra Cape Flats . The announcement was made af- The question , then, is whether there is any reality to [hese allegations.
ter a meeting en Thursday berween Justice Minister Dullah Omar and One researcher, normally close to the government , in response to Minister of
depury Minister for Intelligence , Joe Nhlamhla .
Omar said he had briefed Safety and Security Mufamadi ' s blaming the increases in crime en the nexus
Mufamadi en recent allegations of Third Force activity , and in particular berween Third Force and organized crime, asserts that it "smacks of a polit-
a claim that prisoners are being released from prison at night to perpe- ical ploy, an expedient cop out" ( Jakkie Cilliers , quoted in "Blasting a Hole in
trate violente against specific targets before returning to jail . Omar said Crime Stars " 1997 ). Also, Minister of Justice Dullah Omar asserts that al-
he had told his colleagues that though these allegations are untested, though the allegations are "untested ," they do not need urgent investigation
they are serious enough to warrant urgen[ action . (" Third Force en the but "urgent action" (" Third Force on the Cape Flats" 1997 ). It does not mas-
Cape Flats" 1997)
ter, in other words, whether che allegations are true or not. This points to an
Severa¡ remarks are required here. First , allegations of Third Force seem to interesting double feature of crime in che transformation of the South African
emerge when unusual levels of spectacular violente happen ( in this case, vigi- state. Although crime is a huge problem for the government and a test that
lante and gang violence in Cape Town ). Second, and more interesting, the the government has to pass to expand its legitimacy , it can also be a resource
violence is explained by pointing to covert operations orchestrated by sorne- in che struggles over che transformation of che state. Any claim made by
one with apparent interest in creating mayhem: someone released prisoners people unspecifically accused of Third Force activities loses all legitimacy. By
so they could perpetrare violence en unspecified " specific targets" ("Third referring the blame for crime te a category outside the influence of the gov-
Force" 1997 ). In other words , ernment-to its oldest foe, the apartheid regime's security apparatus-the
an unspecified finger is pointed atan unspeci-
fied perpetrator with, it is assnmed, an interest in discrediting the govern- government and the ANC leadership exonerare themselves of guilt and place
ment and causing chaos en the Cape Flats (or in South Africa in general). By it with a group of people, who, because of che negotiated setdement, they
pointing to Third Force activities, the government and the ANC place the normally cannot blame for anything.
blame outside themselves ; they exonerare themselves of any guilt for che This kind of construction produces a strong resonance in the general popu-
crime and violence, in the case of Omar, on the Cape Flats . Furthermore, what lation as well . Third Force has become a category in which everything work-

could just as well be highly contingent events of violence are localized in one ing against the "people " is placed. 1 have often encountered statements such
locus of evil . In a resolution from an ANC crime summit as "Third Force is clearly involved in Chis." When 1 probed, people responded,
, it is thus stated that
these Third Force activities emanate from "networks of the past " ( ANC tgg6). "Just some Third Force." But Third Force is always related to che former re-
This means that although the allegations are unspecified , they clearly point gime. One example is che Muslims in Cape Town , who are generally accused
to circles around the police . This is far from los[ on che police themselves. As of "urban terrorism .-' They always caution me that "one has to think fur-
an anonymous constable tells che Maíl and Guardian : " The government treats ther here. Things are nos always as they seem . We have information that there
the police as if all of us were involved in third force activities .... They are is Third Force involved, wanting to discredit the Muslims" (interview with
useless" (" Dearh and Drudgery en che Beat " 1997 ). Muslim woman en che Cape Flats, October 1997 ). In times of high levels of
The point is that by not
violence, it seems as though che unspeakable divisions of che present are re-

116 STATES OF IMAGINATION

ANC'S BID TO REFORM 117
 bus contrary ro che ideas of che ANO leadership, it happens very gradually and
cycled through struggle narratives. Thus, Third Force seems not only to be a
in a very dispersad manner.
political srrategy bus also a way of asking questions that are normally difhcult
As we saw in relation to che Secretarias, most bardes about transformation
ro pose, such as why there is violence in the reconciled South Afriea, or why
of che state take on a highly bureaucratic and depoliticized character. How-
che police do flor (want ro) work for che benefit of all South Afrieans-and
ever, just under che surface, che enrire process is traversed by politics, and
oven more, why che ANC has nos succeeded in creadng a normalized statc
because of institucional inercia and che negoriated settlement, it is diflicult
that is universal and fair, a state that can guaranrce the safety of ira citizens.
for all parties to verbalize political disagreements. 'fhus, many of che power
struggles Cake che form of departmental investigations roto alleged hureau-
By Way of Conclusion: Transforming the cratic malpractice. However, che ANC leadership can express its discontent
South African State openly through [he Third Force signifier. The allegations of Third Force need
neither to be specified nor tested. It is simply enough, in che paranoid polit-
1 began Chis essay by asking how the ANC, as a political party, can transform
ical climate of South Africa, to allege che presence of Third Force ro taint
the South African state roto a "normalized" state. 1 argued that che rather
virtually anybody with any stake in che former regime. They will then be rele-
Jacobin representation of itself as a parry, of the state as such, of its oppo-
gated to che subversive out-group of "people against transformation."
nent, che apartheid regime, and of the conflicts of che past can only be sus-
This strategy of creating a group of people inside che state who are opposed
tained on a purely discursive level. These representations crumble when they
to transformation helps che ANC Ieadership to verbalize politically charged
come into contact with the contingencies of local South African politics. The
complaints. Furthermore, Chis exonerares che ANC of guilt for nos making
subsequent gap has to be mended through constan[ celebration of uniry, le-
South Africa roto che place it should be: normalized, sale, and for all. In other
gitimacy, and abiliry to govern in what 1 have termed state spectacles where
words, che past is being recycled as metaphors of che struggle, enabling
che party's hisrorical mission is (re)affirmed.
people in South Africa, notably che ANC, to ralk about a diffrcult present.
The ANC's actions also seem to be informed by an inabiliry to talk about
This exercise is necessary for a number of reasons. Although the ANO has
che truly fragmented namre of che South African society and state. It is as
made giant leaps in terms of transformation and securing che peaceful tran-
though the party does nos have a vocabulary shas can express all the locally
sition to democracy to an extent that one can still talk of a "South African
specific contingencies in che public sphere. Instead, the party reverts to che
miracle," many hard issues remain unsolved. The distante between what is
bipolar language of che struggle against apartheid: everybody who does nos
and what should be seems to have diminished only marginally. The gap be-
support che ANC is against transformation or counterrevolutionary, as itwere,
tween rich and poor has nos decreased; only che composition of who is poor
and everybody in favor of the ANC, as a master of definition, is in favor of
and who is rich has changed through che creation of a nonwhite middle and
transformation. Furthermore, due to che inherently fragmented nature of che
upper class. Entire arcas are being consumed by fear of crime and violence;
state, che transforming efforts of che leadership do nos filter through rhe
South Africa's marginal place in che global economy has become more ex-
system unchanged bus are appropriated by (ANC) people in different depart-
posed; AIDS is threatening che enrire makeup of sociery; public education is
ments of che state involved in power struggles informed by local logics.
still in shambles. The ANC leadership has no or very little influence over most
Through the ethnographic account of the power struggles between che
of [hese problems. They are beyond its reach in either che IMF and che inter-
police and che ANO and inside che Secretariat of Safety and Security, 1 have
national finance sector or che localized power struggles shas take place inside
attempted to illustrate how difñcult transformation is and how the reform en-
specific departments or en the ground of South African politics. Here, the
deavors ofthe ANC encountered serious obstacles. However, che secretariat(s)
leadership can work only for the discrete, and often unintended, transfor-
forged srrategic alliances locally with people in the police, and, as che case
mation outlined aboye. But che leadership is still made accountable, by its
from the Western Cape illustrates, although people with a stake in che former
own admission and by che general population. Therefore, che "people against
regime appropriated che Secretariat, its mere existente created a platform
transformation," or "counterrevolutionary forces" as an out-group will still
from which power struggles could be fought. Transformation does happen,

ANO'S BID TO REFORM ttg

I18 STATES OF IMAGINATION
 be a necessary element in the highly fragmented transition lo "a truly non- u. The taxi industry has its roots in the fact that no transpon service was provided for the
racial, non-sexist and democratie nation." boomíng nonwhite townships. Transportation was for many years a pan of the informal
economy and one arca where some people could eam a lot ufmoney. The competitionamong
different companies and taxies was so hard tltat large -scale violence unen erupted , leaving
Notes
hundreds dead.
12. Buur argues that the Tac was to a large extent an attempt to instimtionalize the horrors of
r. In the following, "the ANC" refers to the policy formulators at the top of the party. This is, as
che past and thereby creare a space in which other issues could be discussed without constant
it rurns out, an importan[ and necessary qualificarion. The parry is so divided and consists of
reference to apartheid . This is similar to when Zizek , through Lacan, talks about transferring
so many different factions that talking about one pany is absurd.
the deepest emotions to others. He uses che example of the "weeper ": " In so-called primitive
2. The South African historian Alister Sparks attempts m explain Chis cense of being a vanguard
societies we find che ... phenomenon in the Corro of weepers , women hired to cry instead
by referring to the ANC's history of exile and the importan communist influences on its
of us: so through the medium of the other , we accomplish out duty of mourning, while wc
organizacional stmcmres. This top-down approach, Sparks says, "is a concept that makes for
authoritarianism" (1990: 403). can spend time en more profitable exploits - disputing the division of the inheritance ofthe
deceased for example " ( 1989: 35 ) . By fnstitutionalizing che horrors of che past , fornler an-
3. Personal communication with police officers working in Cape Town and east oflohannesburg
in che interna¡ Stability Units. tagonista, most notably labor unions and mainly English corporate business , can get down
to business and divide che inheritance aher che now deceased apartheid regime.
4. Total strategy was devised by President Botha in che early 1g8es on the basis of an analysis
13. Of course , there are dissenting voices within che ANC grassreots , especially among [hose we
that there was a total onslaught against South Africa by communism. This onslaught had lo
could term che communirarians , who think che state is always against che "people " (see, e.g.,
be countered on all fronts by a total strategy.
5. The W HA M Nina 1995).
strategy was heavily influenced by an American counterinsurgeney theory, the oil
14. One opinion poli instimte invesrigated che issue and found that in 1994 only 6 percent
spot theory, developed by Lieutenant Colonel McCuen. Oil spotting, of course, has the con-
noracion of oiling troubled water. thought violence and crime to be a serious issue. In 1998 , che figure had ricen tu 64 percent
(Institute of Democratic Alternative in South Africa 1998 ). Another report , furthermore, con-
6. This is what made the director ofinvestigation of Western Cape Provine tell me that in terms
cluded that chis preoecupation was relatively equally distributed along racial lines. Thus,
of detection and prevention, nothing had changed from the 1g8os: the police still worked in
tse lame manner ( interview with André do Toit, October 1997). crime and violence were no longer only " white issues" (Nedeor Project 1996).
15. The National Crime Prevention strategy was passed in x996 at che very momea[ when crime
7. The cerco "coloured" is one of che most conrested conceprs in South Africa. Throughout this
and violence became the most importan[ preoccupation of South Africans . The aims of the
piece, 1 use the British spelling of the word co indicate
that the coloureds form parí of a plan were to " establish a comprehensive policy framework " that could generate "a shaced
specific racial grouping in South Africa. In this, 1 do not mean to indicate that 1 consider rhe
understanding among South Africans " on how to comba r crime. By "integrating che policy
racial categories original. They have been produced historically throughout the past cenmry.
objectives" in all sections of che state-vertically (national, provincial , local) as well as hori-
This means , again, that specific lile conditions for coloureds have emerged
. It is in that sense zontally ( in different departments )- a "common vision around crime prevenrion which .. .
that 1 speak of coloureds as a specific group. For elaboration , see tensen and Turner
( 1996). [could] be embraced by che while of society " should be creaced ( NCpa 1996 : 5). This con-
8. 1 cannot go lato any detall now, as Chis is largely uncharted academic territory , where indeed
stimted a radical deparmre from che past, when "no organized or systemaric approach to
very few studies have been made. One promising work-in-progress in this regard is Azeem
che prevention of crime ... existed ." The new plan would focos on "stemming che
Badroodien's smdy of reformatorias in the coloured community. This is also by and large the
tide ... through che applieation of more policing capacity to selected problems " (7). Al-
subject of my Ph.D. thesis, tentatively entitled "What Place in the World: OfYoung Men, State
Transformation and Violence." though che NcPS, which for a while monopolized che discourse on crime and violence, failed
to achieve its (high-reaching) goals, it nonetheless managed to change che ways in which
9. These Self-Defense Units were created in che 199os by the ANC to protect itself againstattacks
crime prevention was talked about locally in che ministries . Thus, a subtle process of tmns-
particularly by tau-aligned Self-Protection Units. Their history has been fraught with ambiva-
formation was set in modon or helped along by che largely forgotten plan.
lence and violence, enmeshed as they were in local politics. For more elaboration, see Txc
16. This is not quite so. The ministry was called che Ministry of Law and Order and consisted
(1998a). For a more critical analysis, see, e.g., tensen (19996) and Schdrf (1997).
solely of che South African police. The executive branch, che police, was renamed che South
ro. The 0DM presented itselfas a radical departure from what existed before. The party's leader,
African Police Service to indicare that it provided service to conrmunities . Safety and Security
Banco Holomisa, proposed that it was the only one not caught up in the past, the only parry
as a terco was forged to indicate that issues of securuy were made much broader than simply
that represented a new beginning. That Nkabinde not only joined up but attained the posi-
law enforcement. Apart from the symbolic intervention as strategy in che reformation of the
non of national secretary-general destroyed a lot of people ' s hope in che 0DM and created
sute through renaming , che mixup is perhaps indicativa of che confusion about reform, here
the impression of a parry that enrolled al¡ the disgrunded bad guys from other parties.
expressed in renaming procedures.

120 STATES OF IMAGINATION ANC'S BID TO REFORM 121

 i herc -, a p., la to be nrtde aboul che kind of imprecise language used. The terco 'civil rii
oversighi' i,a another ax.unple. lis intprecision could eidter mean thar Acople drafting tlie IMAGINING THE STATE ASA SPACE
policies did noi think ribouc it, or llore likely chal ir a,., impasible to agrce una epecifie
Territorial ity and che Pormation of che State in Ecuador
terne ch1t,w,11d eicher allrnv nci pohrieians direet control or bltmcly scate that they should

nor gct envolved in control.

t8. 11 are, ve n ore che tea hn oerai ¡ e pl edge co he oucside po li ti cs. 1111 s es lar fro nt un ]ti tac lor South Sarah A. Radcliffe
Afdm 1n in.¡ n. [viles, it ie che conetitutive elente11 t in ncc eelf-perceptioo of civil servancs aII

ore r che world

tg. lu che lirst general eleeciou in tg94 tlte tornar aI archeid para', the Nacional Parh, becamu
che inalorery parte in che Western Cape wirh lustacer 53 percent of die otes.fhi, was largely
j)
due lo the fact that che coloured voced en bloc for che parry. The Western Cape ts chas che
anly province whcre che parry has any signtficant support. In clic next election, in June 1999,
che Nacional Party los[ a lot of support, but due to an alliance wirh the Democratic Parte, it
managed to keep che premiership and control over the province,
1
20. lt is a bit problentatic to make Chis distinctlon, of course. One of che main reasons for che
priorides of che MADAM saategy is the tourism industcy, which is flor only Cape Town's
problem but che whole counery's, because che city is che world's access point co che whole 4 States encompáss á=^a a `a '+thigh they carry out their actions and through-
country. The murist as concept playa a very big role in che imaginations of all sectors of
out which they claim sovereignty and power (Agnew 1999). In their regimes
sociery, However, che argument here is that alrhough tourists are importan[, securing dtent
of power, states draw on space to construct their control by means of tech-
is nor making anybody in nonwhite Cape Town safer.
21. This section is based oo interviews wirh people in or close to che secretaria[, documents, and niques of power that disperse citizens and resources and order che regimes
newspaper articles. of everyday lile (Foucault 1984, 1991). According to a siate-centric pialare, che
22. There are deep historical reasons for Chis. The formar regime used che public service exten- state views itself as establishing a spatial matrix into which subjects are slot-
sively to promete che economic oplift of, first, che poor white Afrikaners, and later che col- ted and in which che nacional economy, polity, and sociery are made (Massey
oureds, through what was known as che Department of Coloured Affairs. The laccer is a et al. 1999; Agnew 1999). Yet che establishment of that spatial outreach, and
typical product of apartheid, where each group was sopposed to manage ics own affairs.
che cechniques of control chat work through-and over -the territory, are
With che amalgamation of bureaucratic institucions affer che breakdown of che former re-
of territorial
gime, che farmer whcce Provincial Adminiscration and che Coloured Affairs Deparnnentwere questions about how over time a siate responds co the challenges
amalgamated and now largely make up che Western Cape Provincial Administration. Thus, integration and infernal contests for power. Just as state formation has re-
most employees have a stake in che former regime. Also, che administration has always been cently been analyzed as an ongoing series o cultural acts and material prac-
used to further certain groups ' economic position.
tices (Corriiggan 1994; Joseph and Nugent 1994b), so loo it is posible to ex-
z3. These investigations go en from che top co che boctonc of che bureaucracy. At che time of
amine trae procesa of spatial formaban of che state chas is, how th staXP's
my fieldwork, one staff member of civilian oversight was che subject of such an investiga-
made as a space. As Stepputat (Igg9a, drawtng ora Certeau 1984) argues, we
tion, allegedly because he in toro had invescigaced a high-ranking police officer. The power
scruggles between former Minister Sydney Mufanradi and che Nacional Commissioner also are dealing nor only wirh che creation of abstrae[ space but also che practices
took ehe fiaran of their investigating each other ("Fivaz Takes on Mbeki Police Squad" 1997), trpt gh which space i s consumed and ics organizai^ *+
24. An empty signifier denotes a "signifier without signified," i.e., a signifier chal has been Drawing en recen[ work on Latin American state formation, Chis essay ex-
detached from lis original meaning and can now be redeployed in new ways wirh no scruc-
amines che spatial "forms of rule and ruling" that have been accomplished in
coral limits (Laclau 1996: chap, 3).
Ecuador since the mid-nineteenth century. ne produccive effect of state
25. For an analysis of the "ucban cerrorism" and che Muslim populatton, see tensen (i9gga).
owr is che establishme n t F tsoverel n nd made
up of varied componen[ elements of regions, cides, and natural esources
into an integra[^dwhole. In parallel wirh Joseph and Nugent's recen[ discos-

122 STATES OF IMACINATION

 sion of the cultural process of state formation, this essay addresses the ways
ingly subject to indigenous and social movement input, so too the spatial
in which "practica) and processual dimensions of `ate formation" (1994a: 1q)
vocabularies of rule of the state are transformed. Ecuador thus provides an
are inherently spatial, based on specific geographical tools and knowledges,
example of how, as "modern governmentality ... has penetrated and shaped
and on imaginative geographies and images. Corrigan (1994: xviii) outlines
human life in unprecedented ways, the practices and sites of governance have
the various means by which rule is aceomplished namely, throu h á
also become ever more dispersed, diversified, and fraught with interna) in-
gimes use of a repertoire of terms, the learning of skills and discourses from
consistencies and contradictions" (Hansen and Stepputat, this volume).
outstd e th e state, and the hierarehical nature of political subjectivities. The
This essay is organized around a historical overview of Ecuador's tech-
history o the m ependent Ecuadorian state is a story of the deployment of
nologies of spatial power, documenting the key periods in which geographic
geographical terms and the selective adoption of ideas of spatial science from
techniques and discourses provided crucial tools for the consolidation of ter-
abroad, notably Europe. In the postcolonial period,' the project of Latin
ritorial sovereignty and means of social integration.' The process of "making
American nation building has been a profoundly spatial project, in which a
the map" is then contextualized by a brief discussion of the ways citizens are
lack of physical integration has been compounded by regional conflicts over
inducted into the "spatial vocabulary" of the state. In this section on "Seeing
the nature of the state project (see Mallon 1994; on Ecuador, see Quintero and
the Map," the extent to which citizenship is constructed around a geographic
Silva 1991; Maiguashca 1983). In this context, the question of spatial control
imagination and a spatial vocabulary is examined. It is clear that "forms of
and geographic technologies of rule have been of continued, and significant,
local consciousness" (Joseph and Nugent 1994a: 22) contribute to the forms
interest to the region's states.
of state rule. A consideration of statehood in Ecuador at the current time is
In theoretical terms, the nation-building project of the state is seen to de-
incomplete without an acknowledgment of the diverse ways in which state
rive from the confirmation of its territorial and political sovereignty over
maps have been redrawn and the geographic technologies of power seized
space (produced as the modular area of the nation, within recognized and
by nonstate actors, often in engagement with transnational flows of infor-
secure international boundaries) and the monopoly of means of violence and
mation and resources. The final section, "Remaking the Map," outlines the
physical control. With regard to the construction of abstraer space-Ander-
ways in which indigenous confederations, in come cases with transnational
son's (1991) homogeneous space of the nation-the nation-state relies on
networks, have reworked the spatial configuration of the state and, indeed,
certain technologies and rules. The methods of cartography, inventory ^ nd 6'.r
G^ the nature of its power. Tracing the history of spatial strategies of states is
censos data as well as the physical g e finte cation
m tones of ot D
( b y means - not, of course, a neutral process, and raises questions about the spatial tactics
curren ort , or educatlon )
cana of Ecua that can be taken for empowerment. By critically analyzing the ways in which
dor as 10018 of the state to confirm it
^power over-and knowledge of-its space underwrites power relations, this essay hopes to contribute to a more
SI1PnW1he consolidation of these methods, an elr mshmuona ,zannn
liberatory and decentered set of sociospatial relations.
within the nation-state over the past 150 years, provides a picture of the Latin
American nation-state. As the editors point out in their introduetion, the
¡L,
"universalization" of modern forms of rule has rested contradictorily on un- Making the Map: State Technologies of Spatial Order
equal North-South relations as well as a discursive endowment of sovereign
That ofhcial discourses and practices o£ nationhood have their own spatiality
power to citizens (Hansen and Stepputat, this volume; also Anderson 1991;
is not as widely recognized as the historiciry of national imaginings (on the
Breuilly 1982). In the late twentieth century, Ecuador is experiencing another
latter, see Smith 1991; Hobsbawm 1990). Although analysts recognize the ter-
contradictory twist in this doubleness of the nation-state, as mappings of
ritorial foundations of official nations, the diverse leveis of this category and
state and citizenship are moving out of the state's hands and increasingly into
its functioning in the circulation of signs and power practices have more re-
a transnational space informed by indigenous social movements and multi-
cently been analyzed, particularly through the concept of territoriality.' Terri-
lateral development agencies. As the negotiation of terms of rule are increas-
toriality operates in three ways: through classifications of space, through a

124 STATES OF IMAGINATION

THE STATE AS A SPACE 125
serse of place, and through the enforcement of control over space (by means rhe imagination of the state requires an understanding of diese geographies.

of surveillance and legitimization). In a Foucauldian aceount, tcrritorialih The tour periods under consideration here are 186o-1875; 193os - mid 1940s:

brings in irs trajo certain productive effecrs. In this context, state territoriality 196os and 1970s; and 198os and 1g9os.
produces rhe effect of a sovercign nation-state: ' Ir [is] by means of a process Whereas rhe late rwentieth century is arguably charaeterized by the "un-
of subjeetive representation, recognition and carrographic design, however, bundling of sovereignry," postcolonial states such as Ecuador have spent rhe
that the invenrion of the contents of a 'natural' state territory (takes] place past 15o years attempting ro consolidare sovereignry , by "bundling up" ter-
and thar a legitimare discourse about national sovereignry [is] developed" ritory and power through gcographic teehniques ( see Anderson 1991: 164:

(Escolar, Quintero, and Reboratti 1994: 347). Radcliffe and Westwood 1996 ). While professional gcographers engaged in
The state's "will to arder" operares across a number of sociospatial rela- mapping and spatial practices, the state utilization of geography also encom-
and ad-
tions engaging with rhe productive social dif erences of gender, sexuality, passed a range of groups , inchuding cultural workers, the military ,
class, race-ethnicity, and location.° Surveillance by the state-its all-seeing ministrative functionaries (Dijkink 1996).
and all-knowing overview of, and discursive claims over, its territory-is thus
analyzed as a power effect, che outcome of particular dispositions of maps, Making the Geographic Tools of
personnel, networks, and discourses. State territoriality is thus not a given Statehood Scientíftc, 186o-1875_
but rather something to be worked for, and the tools and behaviors through Although geography as a discipline was only slowly being professionalized in

which it is made are constandy being rethought and remade. In this light, Europe in the nineteenth century, rhe opportunities for the application of

Ecuador is a highly contested space, with various groups generating poly- the newly emerging practice of' scientific" geography did not go unnoticed

phonic mappings and images of the national space. At the same time, the in the slowly consolidating nation-state of Ecuador. Independent since 1821,

nation-state over the past r5o years has extended and elaborated the means Ecuadorian statehood began to be centralized in midcentury under the dicta-

of knowing and mapping its land. Foucauldian approaches have of course torship of Gabriel García Moreno, himself a lawyer and consummate politi-

been criticized for failing tu acknowledge different actors' contestations of cian. In line with other states in Latín America, the territorial question was an

power and whose tactics offer alternative avenues for sociospatial organiza- urgent one, and states turned in large measure to rhe positivist ideologies of

tion. It is only in the more recent period that these spatial tactics can be more observation and experiment for an answer (Hale 1996: 148).
In effect, during the mid-nineteenth century, the Ecuadorian state faced
readily uncovered.
By means of classifications, serse of place, and control, territoriality op- the task of retaining the country as a geopolitical entity in the face of expan-

erates to establish and extend state power by means of physical and discursive sionist claims by Peru and Colombia and as a coherent,set of insritutions that

control over (mapped) areas and the production of "grounded" subjectivities. brought together ofren fractious regional elites. When Peru and Colombia

National subjectivities gain a distinct "sense of place" vis-á-vis their interna- threatened to divide the newly independent state of Ecuador between them,

tional bordees and the internal landscapes (see Smith 1991). Territorial dis- the atomized regional elites unified around García Moreno, whose power

position and orders rest on quotidian work and social reproduction by various rested on a combination of Church backing,1 the army, and landowner fami-

institutions of the nation-state, but these practices and technologies have not lies. The state's need for spatial integration and secure boundaries was para-

existed since the foundation of independent nation-states in Latin America mount; as Quintero and Silva note, "The national question in 186o-1865 was

(nor indeed elsewhere). Rather, in Ecuador there have been four key periods the territorial question" (1994: 114). As geographic work before this date had

in which the state has either introduced new methods of spatial order or has largely comprised gentleman-naturalist assistance for French geodesic mis-

reconfigured the utilization of these spatial technologies to bring them into sions (see Radcliffe 1996), García Moreno turned to the emerging scientific

line with the state project. 1 argue that the spatial imagination of the state is, community in Europe. Establishing the Escuela Nacional Politécnica (Na-

first, always in a process of readjustment, and second, that our analysis of tional Polyteehnic School, the first non-church-established universiry) entailed

126 STATES OF IMAGINATION THE STATE AS A SPACE 127

 the recruitment of Europeans, largely Germans and Jesuits, to the new state that permitted in theory some degree of regional decentralization, although
institution. Geographers, botanists, and naturalists as well as mathemati- chis was not acted on (124).
cians were employed, and physics and chemistry were taught systematically Social aspects of nacional integration were also being developed at chis
for the first time.' stage, although they reflected the fragile sense of unity in che state, which
Theodore Wolf, a German geologist, was central to the establishment of comprised a shifting set of coalitions among landowning and new commer-
geographic techniques in the state in Ecuador. Arriving in 1870, he gen- cial elites en the coast and in the Highlands (Quintero and Silva 1994). The
erated a discourse (including Darwinism) and practice of scientific skills in currency, sucre, was established only in 1884, although the Banco del Ecuador
the geographic and geologic sciences that was to be used by the state in its had been founded in 1868 for the issuance of notes (53, 161). A national an-
territorial consolidation. Together with colleagues, such as Wilhelm Reiss, them was performed for the first time en ro August 1866, whose words were
his work in mapping the country was invaluable, establishing centralized written by the poet and politician Juan Leon Mera (García González 1992).'
knowledge about the layout and makeup of che country. Reports were regu- Primary education was encouraged, although Chis remained firmly in the
larly sent te García Moreno. The appointment of Padre Menton to the direc- hands of the Catholic Church. Although these means of establishing a "com-
torship of the new Astronomical Observatory in Quito consolidated chis move munity" of Ecuadorians were adopted, they remained extremely remote from
(Terán 1983). By 1870, the government established the post of state geologist, che bulk of the population, who were excluded from nationhood by racism,
held by Wolf, who was sent to study the mines of Loja. The inventories of povertg and lack of civil rights.
state resources were complemented by an emerging role for geographers in Overall, the Ecuadorian state by the mid-nineteenth century was just be-
public works administration (a role that was to become highly significant in ginning to organize the abstraer space of statehood. The means at its disposal
the mid-twentieth century). Wolf, for example, was made director of a drink- were relatively rudimentary, but have continued to be used up co che eve of
ing water project in Guayaquil after 1879. the twenty-first century, namely, che consolidation of geographic knowledge
The introduction of speciatists in che new science of geography was soon and skills in the arras of a few professionals, and the mapping of territory.
put to good use in the making of new spatial connections across the highly These spatial techniques were conceived by García Moreno and later presi-
fragmented territory. During his regime, García Moreno oversaw the unprece- dents as offering a regulatory power, wielded by the elites in their defensive
dented building of bridges, roads, anda railway line. Eleven bridges were con- protection of Ecuador as a geopolitical entity. Notably in chis geopolitical
structed between 1869 and 1875, and roads were built in fourprovinces, includ- power game, the state cartography did not provide a visual vocabulary for the
ing forty-four kilometers of the Cuenca-Quito road, although their coverage citizenship of the country, although it appears that García Moreno, the main
remained extremely patchy (Quintero and Silva 1994:161); che rail link between initiator of chis process, personally possessed a strong geographic imagina-
Quito and Guayaquil, for example, was not completed until xgo8. tion of the state. A ough che nacional currency and the national anthem
Cartography was a prime new tool of the state, which had previously been circulated as signs of a nacional community, they were meaningTésss fei the
reliant en eighteenth-century maps by Pedro Vicente Maldonado, who pre- bylk oi tlie, population. By che end of chis period, Ecuador as a 5oxereign
pared them while taking part in the French geodesic mission of 1763, andan -,.bnation-state was considerably more established, due in par t co che deploy-
(arguably inaccurate) midcentury update by Villavicencio (Terán 1983). Using ttfigües ófkñbwledgé ptoductibñ and to che consolidation
spa la 1&t
the new maps and political power, García Moreno changed che electoral ge- of political ties in new geographies.
ography of the coünfry in i86o, soon after becoming president. Rather [han
continue che system of theee equal representative blocs (Quito, Guayaquil, The 19305 tú mid-1940s: Modernization of State Geographies
and Cuenca), he established the province (provincia) as the basis of adminis- The next major period of state usage of spatial techniques carne in che 19305,
tration and political power, a move strongly opposed by Guayaquil (Quintero with certain elen)ents joining the state repertoire of "methods of statehood"
and Silva 1994: 123). Spatial politics were further reshaped in the Constitution in the decadés before and after. The use of mapping was much more exten-

128 STATES OF IMAGINATION THE STATE AS A SPACE 129

sive by this tinte, in borh rhe global North and South, and the application of appointed to sir on the National Cultural Council in rhe mid-1970s, as social
cartography to statehood increasingly was bound up in a move to creare a integration superseded spatial integration in state agendas.
sense of national identity among citizens. Around rhe world, newly indepen- Although by 1940 Ecuador expected that irs territorial sovercignty was rela-

denr stares in rhe twentieth century found geography a "necessary tool for tively secure and was consequendy establishing resource inventories for de-

clarifving and fostering their national identity" (Hoosen 1994: 4). Ecuador's velopment. the war with Peru in 1941 abrupdy changed that situation. W ith

bventieth-century state similarly professionalized geography in line with its die extensive territorial loss in the Amazon basta, both Ecuador's discourses

srte-building praetices, producing gcographic knowledges and inventories. of national territory and the practices of surveillance required ro secure rhe

Morcover, rhese skills and invenrories were associated with the military, remainder of rhe country changed. in rhe wake of ehe conflict, the Servicio

whose central role in nation building was taken as self-evident. What we see Geográfico Militar had rhe knowledge and personnel to place the concrete

by this time, roo, is a growing awareness of the social dimensions to nation: boundary markers along rhe new (disputed) frontier. However, rhe process of

as homogeneous space was established, more emphasis was placed on rhe representing this new border and explaining the abrupt change of shape of

social groupings within that space. national maps became an overriding concern that continued until ehe con-

In the 19205, the Ecuadorian army was appointed by the state to creare a flict's resolution in 1998 (see Radcliffe 1998, for a brief history). After Chis
national topographic map (Cortés 1960: 27). In pressing their case for the date, the 1GM produced maps showing the "dismembering of the territory of

job, the army's representatives argued that the knowledge of "patriotic fron- Ecuador," developing a vocabulary through which the relations of spatial rule

tiers" and an inventory of the country's natural wealth (geology, hydrology, could be encoded in forceful, emotional terms. Maps indicated two rypes of

forestry, agriculture, minerals) would be in the state's interest. Moreover, Ger- territorial loss; first, the territories "ceded" to Brazil (in 1777 and 1904) and
man technical skills sharing and exchanges were important in the wake of ro Colombia (in 1816, clarified in 1922); second, the "territories in Peru's

the German mission for this purpose in 1925 (N. Gómez, interview r994). power" (under the 1830 protocol, and the 1942 event; see map in Quintero
Consequently in 1927, a technical commission for the national map was and Silva 1994: 451-52). After rhe Rio Protocol of 1942, Ecuador continued
formed, followed in 1928 by rhe inauguration of training programs in topog- to dispute Peru's claim ro territory by marking one segment of the (disputed)
raphy and cartography and the placing of the Geographical Military Service border with the words "zone in which the Rio Protocol is inapplicable" (zona

(Servicio Geográfico Militar) under the army high command. The Servicio en qué el protocolo de Rio de Janeiro es inejucutable). The geopolitical interesas of

latee became the Instituto Geográfico Militar (IGM, Geographical Militaryln- the state were clear in this cartography (see Hepple 1992); moreover, the map
stirute) as the association among state mapping, geographic knowledges, and continually kept the issue alive for Ecuadorians, whose education tests con-
the military drew closer during the cold war (see below). From the start, the standy repeated rhe "history of rhe borders" in which rhe conflict with Peru
Servicio was tied to similar organizations across the Americas, as it had links featured heavily. In talking of the relationships among myth, history, and
with rhe Instituto Pan-Americano de Geografía e Historia (Pan-American In- identity, Friedman argues thatthe temporal continuity of identity is established

stimte of Geography and History) and the Inter-American Geodesic Service, (or reestablished) by means of spatial díscontinuity (1992: 194). In Ecuador's

both of which gained increasing significance in training and resource distri- national repertoire, the post-1942 visual and spatial discourses highlighted
bution during the cold war. a strongly spatial discontinuity (real and highly symbolic) with neighboring
At rhe end of the 19305 and into the 19405, geography gained a falrther, countries, especially Peru.
civilian boost through the extension of teaching of the discipline to secondary In summary, the period of the 193os and 19405 was one in which rhe pro-

school teachers. Francisco Terán, first as a teacher and then as a lecturer in duction of abstract space for and by the state continued, although increas-

rhe training college, became a key figure in the establishment of a civilian ingly under military command, in particular after the conflict with Peru. In-

group of geographers. His book, Geografía del Ecuador, was first published in ternational exchanges of knowledge were built up under die U.S. policy of
1948 and subsequently went through twenty more editions. Terán was later overseeing its "backyard," a process that increased in rhe rg6os and 1970s.

130 STATES OF IMAGINATION THE STATE AS A SPACE 131

 Yet the state did not deal with the production of territoriality as exclusively a Explaining the role of the military mapmakers, the law described a need
concern for its military: geographic knowledges and nationalist interpreta- "to control cartography and geographical publications, which circulate or are
tion of maps slowly carne into the public arena through schools and the dis- exhibited, so that they correspond to geographical reality, and that che inter-
tribution of the "logo map" showing Ecuador's new configuration.
nacional borders are correct and recognized by the Ecuadorian state." More-
over, the authority of the maps with che Rio Protocol rested on che judicial
ig6os and 199os: Attempting a Social Integration
system, under which che sale of maps without the Rio Protocol line was con-
of State Territorg
sidered traitorous and carried a sentence of up to sixteen years in prison.
This period saw increased military control of the techniques of geographic Under law, maps had to include the Ecuadorian territory from before Ig41, a
knowledge under the cold war logic of anticommunism. These decades were law that was enforced through regular inspections by the authorities of book-
also characterized by the application of spatial principies to the "problern" shops and other outlets.
of development, and particularly to the perceived need to intégrate poor Ec- Cartographies produced by the modernist IGM were oriented toward both
uadorians into the modernizing nation. In the urgent agenda for social and the national security of the geopolitical region of Ecuador and the consump-
spatial integration, spatial knowledges and imageries were increasingly dis- tion by citizens of "nationalist" imagined geographies. With regard to na-
played for public consumption, although their production was increasingly tional security, the Institute for Advanced National Studies (Instituto de Altos
centralized.
Estudios Nacionales, IAEN) coordinated with IGM geographers for territorial
In die context of cold war geopolitics and the nexus of military power with and aerial photographic information. Formed in May 1972 under a military
"development," it is not surprising to see that the map and cartographic tech- government, the IAEN was explicidy designed to elaborate a national security
niques come further under the surveillance of the (military arm of the) state. doctrine (NSD), as well as "to prepare the ruling classes (cadres) of the na-
Since che 195os the United States had been organizing fifteen Latin Ameri- tion" (Quintero and Silva 1991: 223). As elsewhere in cold war Latin America,
can states in its Geodesic Survey, also providing grants to students to attend the NSD became a state-centered rnission to block external threats to national
courses in the latest techniques in Panama. Even in 1960, the 1GM was argu- sovereignry and expand the state into the foil frontiers (Hepple 1992), justi-
ing in state national security debates for a more significant role for mapping, fying and legitimizing surveillance and military expenditure. The elaboration
and suggested that "the fortificatory project has to be better when the map of che NSD created a discourse in which che military emerged as homologous
on which it is designed is exact" (Cortés Ig6o: 23). The Law of Nacional Car- with the patria, in which interests of che military epitomized the nacional in-
tography of 1978 was a response to these issues. The law explicidy stated that terest, and where the armed forres could speak on behalf of the country.
cartography served the demands of the national development plan and na- Following from the idea of an essential, natural "national soul" of ecuatoriani-
tional security. Map making was to be centralized in one state insdtution, a dad, che IAEN viewed itself as coterminous with the very territory it was re-
task given to the IGM, which was funded by the Ministry of National Defense. sponsible for mapping. Despite contradictory shifts in emphasis toward so-
The armed forces were given overall responsibility for the security of maps cial development, the armed forres continued to expound their cartographics
and aerial photographs, and, together with the Foreign Ministry, the IGM took of state power in a state-centered notion of sovereignty, with the military
over the designation of internacional borders. With a large staff and modern regarding itself as the sole competent agent for oversecing the nation's secu-
offices, the Ic;M gained from its association with a centrist state ethos of rity and development. 1n the course of che 197os and IgSos, when rising eth-
modernist development, objective science, and the tutoring of citizens. Leg- nic group demands and weak unification were perceived as blocks to com-
islaron endorsed its role as the holder of a monopoly on geographic tech- plete nacional expression, the military had bound itself so closely to state
niques in the state, and other institutions were established to extend its work territoriality that it remained central (Quintero and Silva 1991: 231).'
in new directions (Radcliffe Tgg6). Under che Law of Cartography too, the Visual referents to the security of national borders were widely dissemi-
IGM took on a role in the publication and prepararon of informaton and nated, along with accompanying nationalist stories, during this period. The
educacional texts.
first six Republican constitutions had not mentioned the Amazon region, but

132 STATES OF IMAGINATION THE STATE AS A SPACE 133

rhe eastern regios, the Oriente, was increasingly representad in state dis- Also regionalism of countries. For example, here we have regionalism around

course as a cure arca in che nation s geographees. "Ecuador was, is and will Guayaquil and Quito" (interview, April 1994). The extension of education

be an Amazonian country" became a slogan on governmcnt-headed note- around che country also spread che nationalist ideas abour geography to a
paper in the 1970s (Whitten 1981). Oil-based development permitted exren- wider population. A geographic traiuing college fiar teachers was established
sion of stare sovereignry over the territory for the first time in rhe modero era in x973 by che govern menc, with a rep resentative of the IG,w on the board
(Quintero and Silva 1991: 165). The new focus was echocd in che geography- (interview, April 1994). The training college Centro Panamericana para Estu-
hiscory curriculum, where special emphasis was placed on the "Eeuadorian dios e Investigación Geographica (CEOElce) had an explicitly nationalist idea
discovery" of the Amazon River hy une of Pizarro's officers. The myth of prior of its role. As une director explained, "Geography is a very interesting sei-
Ecuadorian claim to che Amazon lowlands east of che Andes of course reit- ence that heles ro strengthen rhe spirit of Ecuadorian nationhood" (interview,
erated an anti-Peruvian sentimenc. In addition to a geopolitical concern with April 1994). State co-optation of intellectuals was generally high at chis tinte
che basin, policy discourses highlighting che promise of furure national de- (Quintero and Silva 1994: 299), although in che case of geographers, che en-
velopment emerged, not least due te che new petrol economy of [hese years dorsement of state policies appears to have been particularly enthusiastic.
(Restrepo 1993: 154). The ubiquiry and banality (Billig '995) of che country's Social integration was encouraged through che above-mentioned dissemi-
map and landscapes (in weather maps, en sehool walls and barracks) was nation of imagined (Amazon) geographees, as well as through extensive so-
established in chis decade. The Oriente represented an arca of security weak- cial programs. From 1963, the Civil Action programs deployed military con-
ness as well as furure development prospects. tingents in development projects, a pattern that was expanded in che later Alas
Although che 'g6os and 1970s were undoubtedly shaped by che geopolitical para la salud (health) and Alas para la cultura (culture) programs. Kibbutz-type
concerns of che post-World War II period, there was nevertheless in Ecua- setciements were also established in sensitive border iones to incorporate
dor an extensive effort by military and civilian regimes alike to overcome che young peasants into'military and development agendas. This Conscripción
profound developinent problems facing the country. Florencia Mallon (1999) Agraria Militar Ecuatoriana (CAME, Ecuadorian Military Agrarian Conscrip-
sees Chis period of Latin America's history as mne of "narional liberation" in tion ) began in 1966, and within ten years had three centers in che Sierra, one
which states address nacional questions in a more expansive (and at times on the coas[, and one in the Oriente. Where possible, civilian populations
progressive) agenda. With oil revenues to hand, Ecuador engaged in a num- were also involved, such as through festivals, old peoples homes, and links
ber of measures intended in parí to create an inclusionary nation-state. Al- with peasant leaders (Quintero and Silva 1994: 221-23). A more enduring
though it had security overtones, a range of policies contributed direcdy te institution was che establishment of a "nacional" football team by che military
strengthening civil rights and inducting a sense of narional place among citi- in 1972, a team that still exists coday although it became associated with
zens. Suffrage was extended to illiterates in che 1978 general eleccion, thereby Quito, thereby undermining its original intention.
including many rural and female voters for che first time, a third of whom In summary, a brief overview of Ecuador's use and deployment of geo-
had previously been excluded from che vote. The physical integration of che graphic skills and spatial knowledges has demonstrated che varying yet per-
country was quickened through che trebling of road mileage between 1959 sistent agenda of mapping by che state. Different instirutions have been
and 1978 (Quintero and Silva 1991: 238). North American geographers were charged with Chis task, and their relationships with the state have been at
commissioned by che state agency for development, che Junta de Planifica- times contradictory and diffuse. Nevertheless, che spatial imagination of che
ción, to carry out research into che persisten[ problems of regionalism (inter- state has arguably been central in a wide-and significant-range of aspects

view, April 1994)- of the state project. In che establishment of state territorialiry, the state has

Geography in planning was perceived as a tool to overcome uneven devel- used geography and geographers co map resources, to guarantee geopolitical
opment As one geographer active at that time said, geography "can help security, to overcome localisms, to control internacional borders, to inculcare
enorniously to overcome many problems, principally che localist problems. a leve of country, and to provide a basis for development planning. Although

134 SÍATES OF IMAGI NATION THE STATE AS A SPACE 135

 the geographic imagination of the state was initially a controlling imperas geopolitical imagination picked up by citizens. In this way, local or regional
locked into territorial defense, that agenda has increasingly taken second
horizons were brought into line with the national territory. Thus, in rural
place to the socioeconomic development concerns of the state.
Andean communities, around 8o percent of mestizo and indigenous groups
recognized the national flag and the shield. Additionally, over 75 percent
Seeing the Map recognized the map of the country (although zo percent did not) in the
mid-Iggos, reflecting high rates of primary education and frequent migration
As noted aboye, while in the process of nation building, the Ecuadorian state (Radcliffe and Westwood 1996).
has attempted to instigate afflliations and structures of £eeling of nationalism In summary, the geographic techniques deployed by the state operated not
in its citizens. As has become clear in recent research, these structures of only to establish territorial sovereignty vis-á-vis other states, but to construct
feeling are profoundly geographic, based on a popular sense of geopolitics,
internal integration among the populations inside its bordees.
knowledge of a "national" landscape (Smith 1991), and recognition of the
national map (Anderson 1991), as well as rituals of commemoration (Gillis
1994). Geographic imaginations and tools can thereby provide the basis for Remaking the Map
the "collective internalization of a territorial identity" (Escolar et al. 1994: Despite attempts to inculcate visual cartographies and geographical stories
352; see Breuilly 1993). As argued by Gupta and Ferguson, "States play a about the national territory in citizens' minds, these attempts are not hege-
crucial role in the popular politics of place-making and in the creation of monic, or ruling people's geographical imaginations. Alternative geographies
naturalized links between places and peoples" (1992: 12). Through its dis- have erupted in Ecuador during the past twenry years in ways that contest the
courses around Amazonian development promise and the visual referent of spatial order established by the state. Relations of citizenship and space have
the national (truncated) map, the state attempted to generate a communiry of been reworked in processes that illustrate the ways in which "emergent popu-
cosubjects united by a primary affiliation to the nation-state (see Crain Iggo). lar cultures and processes of state formation" interact (Joseph and Nugent
One visual technique for the establishment of state power in Ecuador is 1994a: 3). Challenging state-centered notions of a unified discrete space, dis-
the truncated map, showing the territory lost to Peru in the 1941 conflict. tinct voices and practices of "popular geographical identities" (Radcliffe and
Whether in school textbooks, on murals of public buildings, or adorning the Westwood 1996) create new spaces through which to express norions of com-
side of barracks, the map of Ecuador is widespread across the territory. This munity, citizenship, and identity (Painter and Philo 1995). Popular geogra-
visual referent to nationhood echoes Benedict Anderson's (1991) suggestion phies may attempt to "redraw the map" of state power-figuratively or lit-
that national maps become "logos" for national identity in an era of print erally-by generating new spatial order (and mapping that) as well as by
capitalism and easy reproducibility of images. Marked with the Rio Protocol voicing distinctive afflliations to territory.
line and mínima] features internally, the Togo map resonates with a memory The use of cartographic maps by Ecuador's indigenous federations to lay
of truncation; more than a guide to place, the map funetions as a com- claim to land illustrates this process and the hybrid nature of popular ge-
memorative frame for spatial identity (Radcliffe J995).° Schools, geographic ographies of identities (Lavie and Swedenburg 1996). To recuperate land
instimtes, and the media created maps of the national space while implicit from colonizers or to gain land title, indigenous organizations such as the
spatial metaphors and place-bound images circulated (Radcliffe and West- Confederación de Nacionalidades indígenas de la Amazonía Ecuatoriana
wood rogó).
(CONFENIAE, Confederation of the Indigenous Nationalities of the Ecuador-
Another mearas through which subjects construct their sense of national ian Amazon) and the Organización de Pueblos Indígenas de Pastaza (oPIP,
place is the school curriculum, which includes a section on the "history of Pastaza Indigenous Peoples Organization) prepare their own maps showing
the borders" (historia de límites). Referring to the long history of territorial the extent of their land claims, thereby denying validity to the IGM maps.
disputes with neighboring countries, the history of the borders reiterates a Originally trained by a German development consultant based in the Ama-

136 STATES OF IMAGINATION THE STATE AS A SPACE 137

 che indigenous movement has transformed che political culture of che coun-
aun lowlands, indigenous cartographic teams gained the neeessary ski lls and
try. Formally, che institutions of state rule now include indigenous leaders,
political motivations to engage in cap making. "Che confederation eartogra-
such as che Indian representatives elected roto nacional Congress and che
phers also designated territory on che ground with marks on village Crees in
leaders appoinred co Rey poses in surte administration (see below; for a sum-
consultad on with local populations. Map making theceby entered into the
dynamics of mernory, space, and identity by reminding younger villagers of mary of these changes, see Santana Igg5; Almeida 1991: Alban Gómez et al.
1993; Zamose 1994). In che next seetion, 1 examine che resulrant dynamics for
their village limirs (interview with (hlt', 1994).
Che confederations' map making and land tiding projects generate meta- che territoriality of che Ecuadorian state at che curo of che nventy-first cenmry.

phorieal and material means ro contest stace projects that deny indigenous
identity and colonize lands. These alternative nonsmte cartographies have at Globalizing che Map: New Configurations of (Nacional) Sociery and Space
their core a spatial project different from che state's. By providing visual proof
of alternatives co land colonization by settler farmers and to state develop- As oudined aboye, che spatial organization of che Ecuadorian state has been
characterized by che "bundling up" of sovereigncy through che use of geo-
ment projects thac see in che Amazon basin an " empry space" promising un-
graphic forms of knowledge and che institution of spatial vocabularies of
limited opportunities for nacional development , these maps provide a distinct
palimpsest co that organized by che state. Mapmakers in the indigenous con- citizenship . In an era of growing global interconnections of production, im-

federations are key actors in the realization of indigenous projects. They pro- ages, and populations , spatial closure can no longer be constimted around
duce maps of spaces that are to be read both as representations of material neat state boundaries (Agnew and Corbridge 1995). As a consequence of

territories and as mecaphorie and symbolic referents to a sociality constimted political-eeonomic changes and social movements, Ecuador's territorialiry
has been profoundly reworked, giving rise to what are, as yet, preliminary
by, and constitutive of, a particular geography (see Pratt 1992).
indications of what could be consolidated into new forms of spatial rule and
In che interplay between che indigenous and che state, che rimals of rule
have changed, in that the state has acknowledged che confederations' maps social relations.

and conceded to part of their demands, although without transforming ics Rescrucmring of che world economy along neoliberal Cines has dragged

primary concern in geopolitics. Whereas original concessions were of indi- Ecuador in ics wake, albeit laten than other countries in che region. The open-
ing up of che economy co global markets (connected particularly with oil,
vidual land parcels to families, later grants of land from che state were made
m legally registered communities . Additionally, che state cartographers began timber, shrimp, and banana production ) has entailed che removal of state
powers from certain sectors and che reinforcement of state intervention in
to work with che oplr topographic group, although Chis did not make che
indigenous and official projects of creating an indigenous space che same. As others (government actitudes to privatization remain ambivalent). Foreign di-

expressed by one CONFENIAE leader, "The granting of [land] titles isn't che rect investment doubled co U.S.$q.l million by 1993, when che state Consejo

solution to che problem of indigenous tecritories" (Uquillas 1993: 185). Nacional de Modernización (GONAM) "modernization" agency began ics re-
mit of further liberalizing of che economy." Key shapers of the spatialities
In summary, although che preparation and deployment of maps of national
of state power are che World Bank and IMF, promoting neoliberalism with
cerrimry were once che exclusive preserve of che military arm of che Ecuador-
ian state, that situation has now changed. The engagement of indigenous a human face.

popular mappings with state forms o£ rule resulted in che tiding of indige- As Ecuador was encouraged co enter global markets, a measure of lo-

nous land and a greater social inclusion of indigenous populations as full cal grassroots democratization was promoted by multilateral agencies. The
World Bank iniciated programs for che disbursement of funds primarily via
citizens ."' In a broad social movement , indigenous organizations since che
1g8os have contested che centralized and racist hierarchies of rule embedded nonstate groups comprising nongovernmental organizations , state agencies,

in che Ecuadorian state. Through marches, demonstrations, and coalition and grassroots organizations , with goals of capaciry building and instim-

building both within and outside che country with a variery of social actors, tional screngthening. Learning from che social funds experience to alleviate

138 STATES OF IMAGINATION THE STATE AS A SPACE 139

 the impacts of neoliberal restructuring, the programs were initially concerned
for National Development and in regional and local plans" (CONPLADEIN
with administrative "good governance" but became more engaged in shaping
civil society-state relations. Through its Indigenous Program work, for ex-
1997: 6).
The development project that grew out of chis process became che Proyecto
ample, the World Bank promoted intercultural and bilingual education proj-
de Desarrollo de los Pueblos Indígenas y Negros del Ecuador (PRODEPINE,
ects, participatory democratization and multicultural legislation and praetice
Development Project for the Indian and Black Groups of Ecuador), working
throughout Latín America, including Ecuador from the 199os.
with $go million from the International Fund for Agricultura) Development,
Moreover, internationally networked advocacy and nongovernmental
the World Bank, the Ecuadorian government, and indigenous organizations
groups have supplied information, support, and resources as diverse as train-
(van Nieuwkoop and Uquillas 2000). Drawing on the experiences of grass-
ing, airfares, policy interpretations, and computers to civil society in the
roots organizations, "base groups," and regional NGOS and federations, the
Andean countries, including indigenous organizations. As international re-
PRODEPINE agenda was designed precisely to reconfigure che spaces of de-
sources and information have flowed in, indigenous organizations have ne-
cision making, power, and territory in Ecuador. The PRODEPINE spatial for-
gotiated a changed position vis-á-vis a (reduced) state (Serrano ¡993). As a
mation was based on a decentralized alternative to existing parish and pro-
cold war scenario of suppression of interna) dissent gives way to an inclu-
vincial administration, with regional offices and local development projects
sionary" discourse of parricipatory democracy and rights, indigenous orga-
formulated by indigenous communities themselves. The aim was to target
nizations have come finto positions that grant them a eapaciry to rework the
poverty-alleviation measures and social training projects across che three
formation of the state, at least partially. This is particularly true in Ecuador,
major regions, focusing on 288 parishes. The parishes were chosen on four
where, in comparison with, for example, Bolivia, the state has been less cen-
criteria: concentrations of black and indigenous groups; high levels of pov-
tralized and homogeneous (Andolina 199q). Moreover, indigenous actors ne-
erty; high levels of social deprivation; and degree of grassroots organization
gotiate with che state in a space defined by its networks among activista in
(CONPLADEIN 1997). The policy was thus formulated on a combination of
both North and South and different Latín American countries. Involved in
needs-based criteria but also, crucially, on the notion that the populations to
transnacional networks with other countries' indigenous groups, NGOS, and
be involved in development were to be active agents in the process (having "a
solidariry activists, Ecuadorian indigenous populations are exchanging infor-
degree of grassroots organization") and not passive recipiente.
mation, strategies, and proposals with nonstate (and, in some cases, anti-
The new spatiality offered by these arrangements forms an explicit part of
state) actors (compare Slater 1997).
the diSCOUrse Of PRODEPINE. According to PRODEPINE documents, "Net-
One brief example can illustrate the process. Following from the develop-
works allow the gaining of space, presence, information, greater refiection
mentalist and increasingly inclusive agenda, the Ecuadorian state in the mid-
and capacity of consultation." A nerwork extending beyond the national bor-
Iggos founded an institution dedicated to the problems of indigenous and
ders linking diverse social actors is an integral part of the geographic imagi-
black Ecuadorian affairs. This state agency was empowered to deal with in-
nation of the movement, in which international assessment and experiences
digenous and black Ecuadorian issues in ways that belied therr past as back-
are seen as particularly beneficial (cONPLADEIN 1997: 38). The geographic
waters of state administration, deprived of funds and política¡ clout. Con-
imagination informed by indigenous and other popular social movements in
sejo de Planificacion Nacional para Indígenas y Negros (CONPLADEIN), later
the rg8os and 19gos is demonstrated in other statements. The project ana-
Consejo de Desarrollo de las Naciones y Pueblos del Ecuador (CODENPE,
lyzed che relationship between "local" scales and institutions at higher scales
Council for che Development of the Nations and Peoples of Ecuador), stated
in terms of local, national, and global relationships and the balance to be
its objective as "To put in place a process leading to the constitucional recog-
sought among them. Premising their discussion on a linkage between global
nition of Ecuador as a plurinational, plurilingual and pluricultural society, in-
and local development, the project argues that "no local project is valid unless
corporating Chis redefinition in global society, in national objectives and in the
it has possibilities for regional or micro-regional growth," while there is a
political, juridical and administrative strucmre of che stare, [and] in the Plan
need to systematize any knowledges "a escala" (in relation to seale; 39).

140 STATES OF IMAGINATION

THE STATE AS A SPACE 141
 processes of state formation. Rather, che practices of gaining geographic
Despite extensive criticisms of PRODEPIN E, in part from inrerest groups
knowledge and then represenring ir in maps, documents. and policies havc
in che state who resent such decentralization and multicultural policies, such
been integral to tire historical transformation of Ecuador, and will continue
remapping of the geographies of resource distribution and decision making
to be so hito the next cenmry. Geographers llave been employed by thc starc
hold out the prospect of territorial rcorganization in Ecuador. On the one
to carry out this task, but they are not tire only actors involved; engincers,
hand, power has been explicitly decentralized to racial groups marginalized
politicians, planners, the atraed forces, and others havc all had varying
by previous administrative and political geographies; en che other hand, these
roles. Most recently, indigenous movements prepared their own maps ot lo-
same groups are becoming inereasingly engaged in a transnational flow of
cal communities. The formulations of state spatial power are not created
information, resources, and contacts through which to define their sparial
and fixed once and for all, but rest en quotidian work-including ideologi-
and development objectives. Negotiations over the appropriateness of differ-
cal, representational, and material work-carried out by diverse social sub-
ent types of development intervention are ongoing as a result. Indigenous
jects, themselves reproduced within che inventories and power effects of state
opposition to neoliberal reforms of water and land resources, combined
with environmentalist support for their opposition to oil extraction in che territoriality.
In this analysis, the processual and dynamic nature of state geographies
Amazon arca , illustrate these processes . Spatial organization , according to
and territorialities has been emphasized. During much of che Republican pe-
PRODEPINE and indigenous practice, is organized in tercos of networks, in
riod, territorialiry was about che military-state classification of space, about
which Hows between global and local actors work around, as well as strate-
gically alongside, che state. In such reformation of state geographies, there is a (fragile) national sense of space, and about military surveillance over the
ground and its population. Nevertheless, even within this repertoire of dis-
less of a centralized homogeneous state space, but the possibility of a more
course and action, che institutions and actors engaged in a centralized, hier-
heterogeneous and empowering space. In summary, while the indigenous de-
archical sense of state power changed and developed over time. Nonmilitary
velopment agendas are currently being discussed heatedly, it has been sug-
contributions to geographic knowledge increased over the raid-twentieth
gested that che extent of change to state forms of rule under neoliberal reform
century as che educational curriculum, rituals of national celebration, and
and multilateral agency agendas are significant. As state territorial sover-
increased population migration all transformed patterns of spatial integra-
eignty over political economy is unbundled, so the processes of state forma-
tion. Classifrcations of space began to include indigenous confederation map-
tion are increasingly informed by "the emergence of forms of [indigenous]
pings and land titles. A sense of place was indelibly marked with the 1942
local consciousness" (Joseph and Nugent 1994a: 22), themselves constructed
Rio Protocol line, yet other senses of local identities enriched che geographic
in a transnacional field of knowledge creation.
imaginations of citizens (Radcliffe and Westwood Ig96).
The lack of closure around such state geographies refiects the very spatial
Conclusion nature of its task, that is, the unevenness of state power, as well as che exis-
tence of alternative geographies that make "other" spaces and identities. The
The relations among state, space, and power in Ecuador have shown a shift-
quotidian practice of geographic knowledge-making procedures attempts to
ing pattern of spatial techniques , geographic imaginations , and types of
centralize the arrangement and regularization of state space . Simultaneously,
power invested in different geographic tests and imagen. Throughout che
discourses and repertoires of visual imagery set about creating meaning
Republican period, che imagination of the Ecuadorian state has been a
around these spaces as places. In other words, space and place get called up
geographic imagination, in which the territoriality of the country has been
in che practical, discursive, and representational aspects of power, the dimen-
mapped and managed in various ways. From the centralizing and integra-
sions that underpin state power. Within che processes of state constructions,
tive, almost recuperative, project of President García Moreno in the mid-
it is thus not only the externally oriented question of geopolitical sovereignty
nineteenth century to the localizations of development policy in the indige-
that conditions state practice, but also, crucially, che internally oriented justi-
nous social movement, it is clear that space is not a passive backdrop to the

THE STATE AS A SPACE 143

142 STATES OF IMAGINATION
 fications and explanations of territory that matter. Such internal and externa) 8. International talks around securing the Ecuador-Peru border dispute-with the mediation
elements are intimately interconnected in tercos of the personnel involved in of che United States, Chile, Argentina , and Brazil - resultad in a resolution that undercut
defining the state repertoire and in terms of the audiences for their reception traditional notions of territorial sovereignty. The Ecuadorians agreed to drop their claims to

and the practices that underpin them. Moreover, with the fast-changing trans- sovereign access lo che Maranon River in favor ofa "free navigation accord" with Peru (The
Guardian, z9 January 1998).
nationalization of economies, states, and indigenous politics, the boundaries
9. Recen resoluuon of che Ecuador-Peru confict over clic bordee raises interesting questions
of the map are rapidly transformed into new, unprecedented fields of action
about che form the lago could now take.
as yet not fully explored by social actors nor mapped by social analysts. In 1o. However, internal security concerns did nor completely disappear oven then. In che aftermath
conclusion, nation-states are status of geographic imaginations, subject to con- of the 1990 Indigenous Uprising (Levantamiento Indígena), rural highland arcas were mili-
stant redrawings and new cartographies created in che power-invested spaces tarized, especially in those arcas, such as che Central Andes, where indigenous support (br
of territoriality. che uprising was strongesr. State presence and surveillance of Indian communities increased
through roadblocks, new "development" projects (viewed by snany rural dwellers i, a mean
lo enter villages and defuse politieal organization), replacement oE rural civilian reachers
Notes with inilitary personnel, and military searches oEprivate houses.
u. Over the period 1982-1988, Ecuador underwent eight adjustment and restruct:: ns„ pro-
1 am gra te di¡ to che organizers of the conference on States of Imagination atwhich this paper grams, although it retained a trade surplus by 1993. In 1986-19g1, exports to the Andcan
was first presented and commented on so wisely. Thanks are also dueto Fino Steptimatand countries quadrupled.
an anonymous reader for further commenrs en later drafts. The research froni which chis 12. The extent of"inclusion" is, of course, subject to negotiation, non leastwirh regará to gender
chapter is drawn was generously funded by the Economic and Social Research Council in issues, which are often remoce from debates on "indigenous" politice.
projects on "Remaking che Nation" (1993-1995, No. R000/231432s) and currently, under
che Transnadonal Communities Programme, in the project "Transnational Indigenous Com-
munities in Ecuador and Bolivia" (1999-zoo1, No. L214(2512023). Al! trauslations are mine
uniese otherwise noted.

r. The postcolonial period in Latin Anierica is generally defined as che 1780s to 19gos. Formal
independence For che Spanish colonies vas gained in che r8zos and 18305.
2. For an alternative chronology of regionalism and rerritory in Ecuador, see J. Maiguashca
(1983), who identifies 1830-1925, 1925-1945, and 1945-1972 as significant periods.
3. 1 use che terco territoriality in a critica) cense that atteropts to question the ways in which
geopolitics and space are utilized in power gamos.
4. In Chis regard, recent work in the discipline of geography has highlighted the role of sur-
veillance, representation, narrative, racialization, (hetero) sexualizarion , aud gendering as
key practicas through which power constitutes its social and spatial effects (see Massey et al.
1999).
5. García Moreno signed a concordar with che Pope, grauting che Catholic Church status of
sale religion in che nadan and control over all religious life, educatioo, and che readership
of books (Quintero and Silva 1994)-
6. In this, Ecuador was nor alone, although it was possibly an early advocate of the trend. By
che 18705 and i88os, various republics liad created new higher education instimrions F.
provide leadership informed by modero science. The emphasis in [hese institurions was en
encyclopedic knowledge ofsubjects, en scientific and practical concerns, and en secular¡ Sin
and sute control (Hale 1996: 149).
7. Juan Lean Mera was also author of a book entided Catechism of Geography of the Republir ef
Ecuador (Ierán 1983:183).

144 STATES OF IMAGINATION

THE STATE AS A SPACE 145
dDi i sní QNV 3LVLS II
THE SOUTH AFRICAN TRUTH AND RECONCILIATION
COMMIS SION A Technique of Nation- State Formation

Lars Buur

The South African Truth and Reconciliation Commission (SATRC ) is one of

the most prominent features of the South African transition . The processes
initiated by the sATRC have had significant influence in transforming the idea
and expectations of the new South Africa nation and state. The work of the
commission is therefore an interesting site for understanding how an idea
about a transformed nation-state is produced.
The SATRC's general task, according to the Promotion of National Unity
and Reconciliation Act No. 34, 1995, was to "[establish ] as complete a picmre
as possible of the causes , nature and extent of the gross violations of human
rights ... including the antecedente , circumstances , factors and context of
such violations [ and decide ] whether such violations were the result of delib-
érate planning on the part of the State or a former state or any of their or-
gans" (Act 1995: 8o5 - 6, 8og). One could say that the aim of establishing "as
complete a picture as possible" of atrocities committed in the past was to
create a shared cense of the past in the midst of an extreme diversiry of ex-
periences . The sATRC took it upon itself to construct a common national
history based en the "truth" about the past human rights violations com-
mitted by the past regime and its former opponents . This truth was, idealty,
m lead to a shared understanding of the extreme diversity of experiences in
South Africa.
The biggest dilemma of this imperative was the question of how the new
democracy would manage to deal both with individuals belonging to lib-
eration movements and with those from state institutions , given that both
 strong, onstage presence in the public eye that served to mark rhe trans-
groups were responsible fbr disappearances, death squads, psychological and
formation of rhe state and rhe nation. The new stare soughr ro present itself
physical torture, and other violations of human rights. This problem was in-
as a rrurh-seeking, imparcial judge, dispensing polirical and social justice, in
herendv relared to questions of how a serse of justice could be established
contras[ ro rhe exclusive apartheid state.
that neither granted hianket anmesty no¡ prosecuted every person who had
Second, 1 explore how a space was created in which rhe imbiguous con-
incited or committed human rights violations.
iiicts of rhe past and presenr could be dealt with. 1 suggest that rhe SATRC
was conceptualized as a body disconnected from rhe state and society through
The Negotiated Settlement a ránge of technicaf legislative devices Chid. 1 examine m derail the
invisible bac stage, everyday praerices of rhe SATRC: how rhe materia] tbr rhe
Que might say that this process was necessary, but also that it was compli-
new nacional history was produced and cleaned of ambiguities. Here 1 sug-
cated by the faet that nobody could elaim he or she had come out of the past
gest thar the everyday practicas were characterized by being highly compli-
conflict a clear and unconditional winner. Unlike the Second World War, there
cated translations and negotiations, which were informed by both racional
was no official winner of the war against apartheid, and hence it was nor
ideas about how modern bureaucracy should funetion and more localized
possible ro establish a war tribunal like the paradigmatic Nuremberg courts-
concerns of a polirical nature.
martial. In South Africa, the era of apartheid was brought to an end through
Finally, 1 discuss rhe effectiveness of the separation berween the visible and
a negoriated political compromise including nearly all recognized political
the invisible nature of rhe sATRC. As a consequence of rhe perspective cho-
players in rhe country.' In spite of, or perhaps because of this settlement,'
sen, Chis essay ro a large extent leaves out a range of highly relevan[ questions
tensions relared to the work of the SATRC have continued to emerge. These
relared to accountability, the need for justice in a transitional society, moral-
tensions can be diagnosed as a conflict berween implicit and explicit under-
iry, and discussions relared ro what has been described as the South African
srandings by different people, groups, and political parties of the nature of
"miracle." Instead, I concentrare on rhe nitty-gritry of the backstage perfor-
the negotiated settlement.
mance of bureaucratic micropolitics.
In brief, the explicit ofliicial expression is that there were no winners of the
past conflict. Therefore, as SATRC Deputy Chair Alex Boraine formulated it,
rhe work of the sATRC is "an honest assessment and diagnosis of the sickness The Nacional Spectacle and the Other
within out society in an attempt to give people, both perpetrators and victims, of the Rainbow Nation
an opportuniry to face the past and to star[ afresh" (1996a: 2). 1 suggest that
Truth commissions, as pointed out by several authors, have proven far more
/rhe implicit understanding is that there was indeed a winner, and everybody
e ffeective iban court proceedings and tribunals in furnishing a dramatic me-
knows it was th^e AN,c, which is still celebrating the changa from institution-
dium for communicating rhe new offirial y and law thar a transitional
alized separare ethnic development to nonracial democracy through sports
sociery might want to implement (R. Wilson 1996; Hayner 1994).' The SATRC
events, public holidays, and the SATRC. However, due to the nature of the
has proveo paraca ar y effective in Chis regard. One reason for Chis was its
negotiated settlement, Chis celebration cannot be made overtly in rhe work of
quasi-legal status, which helped prevent rhe interruption of storytelling ses-
the SATRC, test it be criticized for nor looking toward rhe firmare, and test it
sions by lawyers drawing attention to legal technicalities (R. Wilson 1996:
alienare supporters of the former regime.'
16).' This gave the process unprecedenred media appeal, nationally as well as
This essay examines rhe relation berween the `bosta
internationally (Theissen 1999). On the other hand, it also gave the SATRC
speetacle of the SATRC process, and rhe "backstage," invisible, incide of the
an enormous ideological responsibiliry, something tu which rhe usual court-
bureaucratic i chinero nf rrurh roduction.' First, 1 examine the public rep-
room in South Africa was flor quite as accustomed (R. Wilson 1996).'
resentations emerging from rhe SATRC process. 1 suggest that it was rhe es-
Through [hese public spectacles, rhe SATRC provided a controlled environ-
tablishment of rhe SATRC and the public representations emerging from its

NATION-STATE FORMATION 151

150 STATES OF IMAGINA'TION
 ment, which enabled talking about the past in new ways. That is, quite apart (1994), Boraine and Levy (1995), and Boraine (t996b). These writings assert
from its overt objective of creating an understanding of the "nature, causes that something must be removed from society, the sick must be expiated to
and extent of gross violations of human rights" (Act 1995) in South Africa, let the rainbow nation become healthy. An examination of two such media
and documenting violations and abuses in an official state setting, this pro- personifications of sickness, security police officers Gideon Nieuwoudt and
cess allowed a particular picture of the new South Africa to emerge. Eugene de Kock, is illustrative. I suggest that each looks exactly like the Afri-
In the human rights violation hearings, the victims of abuses were allowed kaner incarnate-the "good" and the "bad" sides.
to come forward and publicly reenact a carefully selected catalogue of abuses.
In the amnesty hearings, che perpetrators of human rights violations could Personification of Violente and Evil
be granted amnesty according to certain criteria, one of which was telling the Today, Nieuwoudt is serving a life sentence for a range of brutal killings in
"whole" truth. These were not just "media stunts". displaying "the tissue the Eastern Cape during the apartheid era. He has been denied amnesty by
war," as they were popularly known (a reference to the constant dishing out the Amnesty Committee of the SATRC for not "disclosing the whole truth"
of paper hankies to victims in public hearings). Besides referring to the about how he and others rortured and killed human rights activists and cadres
starkly contrasting displays of weeping victims telling painful stories and am- of Umkhonto We Sizwe° and who gave him the orders to commit human
nesty applicants showing litde emotion while confessing to gruesome killings rights violations. Internationally, the most prominent case has been the kill-
and torture, the picture that emerged was one that demonstrated how the ing of black consciousness leader Steven Bantu Biko. During these hearings,
balance of power had changed from the perpetrators to the victims-from one had only to listen to and look at Nieuwoudt to understand why he is one
one class and racial group to another.e This display is closely related to the of the most haced persons in South Africa. Speaking in Afrikaans at hearings,
way in which the new South African nation-state, the "rainbow nation," is often with an aggressive and arrogant attitude toward the victims of his brutal
imagined. actions, every word had to be dragged out of him, and he was never willing
to reveal anything or to implicare anybody.'° Although in some quarters this
"Otheríng" attitude would create respect, in the context of the SATRC process, Nieuwoudt
To create a new nacional history, a "radical other" was needed as the consti- becomes the incarnation of the "culture of mendaciry": the "bad" and evil
tutive outside of the new South Africa. In S6itth Africa, the identification of side of Afrikaner culture.
the other of the new nation-state has been the task of the sATRc. More pre- De Kock carne across in much the same manner at the beginning of the
cisely, a certain construction of the other has started to emerge from the work sATRC process, but he has since shifted attitude. Also serving a lile sentence
of the commission. This construction has been a positive identification of in prison (212 years), he is currently speaking freely at amnesty hearings,
what the "new" nation-state is, compared to the "oid" apartheid state. How- making one disclosure alter another, implicating former apartheid cabinet
ever, because of the negotiated setdement, this process could not be carried ministers and superiora. His bodily attimde is relaxed and his "naked deter-
out through clear reference to the present, because current conflicts were ex- mination to tell the truth" has been seen as an accommodating gesture to-
tremely difficult to deal with. Instead, present conflicts were constantly re- ward the victims of his deeds ("A Word, a Nudge" 1999: 24)."
arriculated in the form of the past. The other, in chis sense, became the past De Kock's shift makes it difficult for people to come to grips with him
as opposed to the present focus on transformation. today, and to deal with their own conflicting impulses to condemn him for
The psychoanalytic vocabulary of trauma combined with the legal language his evil deeds at the same time as praise him because he has "further opened
of human rights provided legitimate languages for articulating this paradox. the doors" to the dark side of South Africa's past ("A Word, a Nudge" 1999:
In particular, a whole range of illness metaphors has been used to diagnose 24). However, should de Kock be granted amnestyfor over one hundred cases
affairs of the past, which have a bearing on the society of today. The preva- of killings, torture, and inciting killings for which he has applied, many
lence of these metaphors is illustrated amply in Boraine, Levy, and Scheffer CIL South Africans would, today, accept it. This is so mainly because he has come

152 STATES OF IMAGINATION NATION-STATE FORMATION 153

to personily rhe "culture of truth telling " that characterizes rhe new South 1lalf an hour later 1 me[ die samc secretary together with a logistics
Africa. As two brothers of rhe same culture, rhe one stands out as an evil of officer in the door to thc hcarings room un the tenrh fioor. The hearing
rhe past, whereas rhe othcr is pointing toward the fumre " with a lightning had just started and rhey were whispering, discussing de Kock.
fas[ twinkle in bis eyes [who has yeti ro be caught out in a lit [promising to The secretary said, "W hat a sharne he is so small now. Betore, he was
tell] more ata later stage " ( z4)." Let me give an example of how ambiguities rail and looked really strong. That was before they put hico in jail."
like rhe ones described aboye play themselves out inside the sxrac. The logistics officer. who had been active during rhe struggle in rhe
ANC and had lost severa) of her friends while orhers had been rortured,
An 1 lonest Man was furious. "Kak, man! [Shid] How can ynu defend him? He is a killer
During rhe "Guguletu Seven" amnesty hearing in Cape Town on 17-19 No- with blood on his lrands. He is evil. 1 could go up and kill him for whar
vember 1997, de Kock was subpoenaed to give evidente. He liad supplied the he has done. He is a futláng Boer." She turned away from us in disgust
sATRC with information about Vlakplaas" involvement in the killing of the and sat down in the last row of chairs, right at the back of the room,
seven activista in 1986. His testimonies challenged some of the white secur- facing de Kock. Her last words were, "Don't listen to her [the secretary].
iry officers' versions of the incident, because they denied the killings were He [de Kock] is a fucking Boer."
preplanned. The secretary tried to explain: "He did awful things, but it is not fair
My first meeting with de Kock happened on che first day of rhe hearing. He to lock him up alone. He has changed now. He is the only one speaking,
walked around in his green military uniform, apparendy alone. Passing him, telling the truth. He is not like de Klerk and P. W. [Botha]. He is telling
I did not at first recognize him, but 1 remember noticing that I had seen the the truth, so they should treat him with more dignity and not lock him
face before without being able to fix it as de Kock's. A few minutes later 1 up like a criminal. 1 am not saying it is right whar he did, but he is telling
passed a group of staff members of the SATRC regional office in Cape Town. the truth and that is good." (Based on fieldwork notes 1997)
They were talking about the person I had passed a few minutes earlier:
This story can be interpreted in many different ways. One could, for ex-
"Lars, don't you recognize him?" ample, blame the secretary for being ncive, but that would be to miss the
1 said: "Who?" point. Apart from telling something important about the highly complicated
"The person down there," nodding with their heads toward the other process of the sATRC and how particular elassifications change over time, 1
end of the corridor. would like m highlight the particular configuration of individual Afrikaner
1 said: "Oh yes, I have seen him before, but who is he actually?" securiry personnel and its intímate relation to the overall classification of
One of them, a secretary for one of the commissioners, said: "That's the Afrikaners, the "Boer." One way of identifying the malaise of the South
de Kock. He is here because he is going to testify in the Guguletu Seven African society is to "villainize" somebody. This somebody, 1 suggest, is the
amnesty hearing. He is challenging the local policemen's stories. He is Afrikaner, a name that became shorthand for rhe old apartheid state and its
going to make a Pool out of them, challenging their lies." wrongdoings-the past of the present.
1 said: "Oh, is it really him? Are you sure? He looks very different from
the person I have seen in the newspapers and in [the documentary tele- Chains of Equivalentel"
vision program] Prime Evil." One of the most striking features of the public amnesty hearings has been
The secretary responded: "Yes, it is him. He looks different because the focus on individual Afrikaners as representations of the old apartheid
he is in prison now, he has losr so much weight. Before, he was bigger state. In particular, the peculiar configuration of individual Afrikaner securiry
and taller. His time in prison has really taken its rol) on him. What a personnel, who were mainly hostile to the sATRC process and seldom free-
shame, he is much smaller now." speaking, was linked to distinet state instimtions as human rights violators.17

154 STATES OF IMAGINATION NATION-STATE FORMATION 155

 In the work emerging from the sArRC, che specific entiry of "Afrikaner individual Afrikaners are broadly identified as the source of ovil, as ovil itself.
human rights violator" was presented adjacent to another closely related en- how can reconciliation be achieved?
tiry: the strikingly silent and supposedly ignorant group of ordinary Afrika- If a personification of ovil is necessary, this would suggest that the pos-
ners. The Afrikaner as human rights violator and che ignorantAfrikaner form sibility of reconciliation with other Afrikaners and Africans requires a simul-
a single entity: the Afrikaner of the old apartheid state. The behavior of chis taneous celebration of che good elements of the Afrikaner, good elements
entiry is made "real" by linking patterns of action to a specific "culture": the that can be incorporated in the new rainbow nation's culture of truth telling.
old apartheid state's "culture of mendaciry." Deeds and actions of the Afri- This, however, would require that the evil be personified in certain people,
kaner securiry personnel became che operational definition of Afrikaner and the whose wickedness is then explained by referente to the Afrikaner past, like
old apartheid regime. In chis way, a paradigmatic image of the Afrikaner as Nieuwoudt. Hence, the behavior of individuals would be justified, to some
an evil belonging te the past, of what the new nation-state is not, and of what extent, as being beyond their personal control, as deformations of essentially
che sATRC process was implemented to identify and leave behind is created. good Afrikaners. Then, individuals could be "cured," to draw on the medical
This process of classification has its dialectical eounterpart in a similar metaphor, and che healthy Afrikaner could bloom together with the res[ of
chain of equivalente, consisting of a particular combination of individual vic- the present rainbow nation, just as de Kock does.
tims of apartheid, [he black majority, and the new South African state. This During che SATRC process, "Do not villainize the Afrikaners" was con-
combin
ation p arily includes women, most often mothers, who voluntarily stantly stressed. This statement, however, has the precise effect of vilif'ing
go te the SATRc. They are presented as representativ aud— Afrikaners, because the "group," however disorganized and heterogeneous it
reconciliation-seeking black popo ation an image through which the SATRC may be, is by implication pointed out as different from the rest, something to
has been able to represent the new South Africa. Instead of the culture of cake notice of. On tire one hand, by representing the Afrikaner in such cate-
mendaciry, embodied in the oid apartheid state, the new South African stare gorical terms, the problem is no longer in individual human beings, but in an
is characterized by a "culture of truth telling." Rather than being the ultimare entire culture of old Afrikaners opposed ro che transformation of the new
victims of apartheid, chis group of wornen stands out as a paradigmatic ex- South Africa. On che other hand, this classifieation process also creares a new
ample of the new national subject, the subjeet te be served by che new state. category, the new nation-state, an object of identification and service for the
The Afrikaner is constituted as the past against which the new state is seenl óX a4'7 Afrikaners.
Specifically, the democratic state formulares itself in opposition to human
rights violations, behavior that infecred the "pass" In this sense, the formu- "Afrikaners Are Africans"
lation of a culture of mendaciry also identifies that which is now presumably In what is now known as che "Afrikaners are Africans" speech, South African
left behind. Therefore, when human rights violations happen in che present, President Thabo Mbeki spoke to "the Afrikanerbond" in Pretoria on 27 July
rhose actions are deemed "left over" from the past , most often assi n o 1999." In the speech Mbeki opened the door to the biggest "group" of Afri-
"THird Force" elements of the past, or as yet uneducated and rroncfnrn, d kaners who had not yet volunteered thcir services to the new South Africa.'"
rou s formerly serving the apartheid state. He asked whether South Africans "are te remain tied to the divisions and
The power of chis particular configuration of the past and the present m confiicts of che past, or whether they now can be fused roto one South African
hued with two distinct cultures is more intriguing when one realizes that journey leading into a common futuro" (Mbeki Igg9). Apart from the lan-
nrost amnesry applicants coming fonvard to the sATRC are not white securiry guage used, which reminds os of the language of the SATRC, his speech is
personnel; many of those were indemnified before che SATRC started. -Che interesting because we find here a transformation from "truth telling" and a
majority of applicants for amnesry are black, alleged members of che differ- focos en the past tu "the emergente of new ways of doing things, and find-
enr liberation movements (TRC zopo). The oppositional play ourlined aboye, ing new, creative ways of expressing ourselves and our plans for the future"
which strucmres the past and present, raises one question: To the extent that (Mbeki 1999). It is no longer the past and truth telling, but a question of the

156 STATES OF IMAGINATION NATION-STATE FORMATION 157

 where cite nacional hcad office of sxTRC was situated. Ir is Chis phenonieaott
futuro, of becoming involvecf in dovelopnient arad die uplift ot rho black ma-
1 turra to now, exploring firsr che technmeal consrrucrion of die s,crirc as sepa-
jorin. 1lis answer is a clear ves: dic Afiikaners can become parí oí che new
ated froni che state and larer rhe actual work going ora inside che pubiic te-
Soudi .Afinca hecause "Afrikaners are Alricuts." It is onh a question ofdraw-
cess points o1 dic s vrnc. Based on dic aboi e ,tssessmer.r of rhe Suund Atrlcan
iug ora dic good side or tlte :Atrikmner licritage:
transiriou, 1 suggesr that che futtction of die sATine has bcen ro institutiomtl-
1 liave bcen rold tor a long rime thcre has becn in interesting way of ize "' rhe conflicr oven che pass atad rhe highly ambivalent »,resent.More pre-
descrihing a loyal Afrikaner. Iris ahvavs raid: 1 Iy is'n goeie Afrikaner'
cisely, tire displacemcnr and separation of diese issucs froni orare and socien
11 le fS a good Afiikanerl. hhís \v as ofren uscd to describe soco cope w1](, have beeu produced rhrough a range of both technic l and legislani enevices.
should be elected to a school comntittec, parlianient or any other lead-
ership position. To some extent iris cine same in che ANC when sorneone
is referred to as a "good comrade." A Constitutiva Separation
But should a "good comrade" and a "goeie Afrikaner" work merely The daily work of cine sATRc, including che processing of data about human
for their respective interesas, or should they not be good eomrades and rights violations wirh che aun of aceumulaning a comprehensive official his-
goeie Afrikaners for che entire South Africa? (Mbeki 1999) rorical database, was governed by a scientific, positivistie ideal` The parame-
A month later, rhe reply carne from che Afrikanerbond, who pledged to ters of chis ideal are oudined in che mandare of che SATRC» According to rhe
work closely wirh che ANC government in bringing economic and social Aet of Nacional Uniry and Reconciliaron, che SATRC was required to "fune-
development tú che country ("Afrikanerbond forges ties wirh ANC" 1999). tion without polirical or other bias" and in an objective and even-handed
Thabo Mbeki's speech displaces che question froni truth relling te actual ev- manner (Act 1995: 843). Ir was therefore obliged to be so independent body,
eryday dedication in che forro of hard work, a language rhat, given their Cal- separated from rhe polirical environment that consrimted it and that granted
vinistic Christian background, is familiar to Afrikaners. The point is that Chis it time extensions (which happened several times) and funding. The Aet states
displacement could not have happened without che work of che SATRC, with that ro fulfill its obligations-documenring and laying [he foundarion for a
its "public grilling" of che "bad" sides of Afrikanerdom. The subde shifr from morally just sociery-che commission shall be "independent from any parry,
truth relling to work erhic nceded a resonating background, and Chis was government, administration, or any onher funetionary or body directly or in-
provided by rhe SATRC. direcdy representing che interesas of any such entiry" (843)." This eonsrim-
Rimalized public represenrations emerging from che pubiic sATRC pro- tive separation was produced by a suspension of temporality and a suspension
cess(es) are effective performances of che new nation-state. But che visible of place to creare new time-spaces at the margins of che polirical and social
character of che spectacles of [he sATRc as 1 have described and analyzed it domains of society.
aboye is one thing. Another side of chis story is che invisible daily work going The temporal suspension can be leen in a range of factors characterizing
en behind closed doors. When 1 visited South Africa and che SATRC for the che commission. First, rhe legally preordained end date locales rhe commis-
first time in September 1996 1 was surprised when highly esteemed scientists sion outside the "normal" remporaliry of che nation-srare. Second, chis was
writing papers and participadng in rhe public debate about che commission further enforced by vesting the commission wirh special invesrigative pow-
did not even know where the headquarters of che SATRC were simated (based ers because che sATRc could subpoena people without obtaining clearance
en fieldwork notes tg96). froni other authorities .^, The special investigative powers enhanced che ex-
Even though che SATRC was che single niost public and debated institution cepdonal position of che SATRC, indicadng that we are dealing wirh an insti-
in South Africa during 1996 and 1997, wirh a radio channel sending directly tution where time is compressed and where niany tasks have to be done in a
from public hearings and a weekly internationally awarded television pro- hurry, wirh no time for following che "normal" schedules and procedures oí
gram broadcast en nacional relevision every Sunday, very few people knew che state.

158 STATES OF IMAGINATION NATION-STATE FORMATION 159

 The suspension of place refers to the fact that most official truth commis- Constitution of South Africa, with rhe invocation of nation building, peaceful
sions do nor rk in state buildings, but rent their own premises and tease coexistence, and a new nation-state built on rhe rule of law, including the
their own equipment. In Sout 1 A rica, the SATRC set up four privately owned invocation of human rights as the measurement of "universal justice."
offices furnished with equipment leased or ourchased with f gn a r
There are many references to the judicial discourse and set of practices in
money. This way of ideally disconnecting the SATRC from the state's localizad the work of the SATRC. Besides the obvious referentes to amnesty legislation,
socia nvmes created an idea about a new time-space at the margin of the rhe judicial power vested in the power of subpoena and in the revocation and
nation-state, placed outside the nitty-gritty of everyday political struggles. framing of the work of rhe SATRC within international law and in particu-
The symbolic importante of this form of teehnical disconnectedness cannot lar rhe Geneva Convention of Human Rights, the judicial discourse and set
be underestimated. When rhe SATRC closed down some of the regional of- of practices was apparent in rhe public hearings process. Every hearing, for
fices before the release of the final report in October 5998, it was decided m example, starts by claiming "This is not a Court of Law," but the way rhe
transfer the amnesry section of the SATRC to the new premises of Magistrate hearings room is set up-with commissioners and committee members, ap-
1
gurt in Durban, KwaZulu Natal. Among staff members in the Durban office plicants, victims, and lawyers placed as if it were a courtroom-makes it
this raised discussions aboutwhether they could maintain their independence difficult to distinguish from a tour[. This image is supported by rhe language
in the work and whether people outside the commission would perceive the used, oath taking, and expected behavior, such as participants asking ques-
SATRC as a prolongarion of rhe justice system. In other words, they were tions through a lawyer.
afraid that by tila transference they would not only compromise the work of Through this double invocation of technieal separation and legal discourse,
the SATRC, but also creare, in the public mind, a sense of the sATRC's being rhe SATRC became a means of accommodating conflicr. It was a way of regu-
a" normal" tour[, judging people (interviews, Durban, 1998). lating and neutralizing turren[ conflicts related ro past responsibility and
guilt, so that [hese conflicts could be contained and dealt with within one
The Legal Discourse
segregated domain. In Chis way, the past and irs conflicts were eneoded, or-
The sense of the SATRC as disconnected and legitimare cannot be explained dered, and organized, so that past experiences could be encapsulated in the
with reference to its technical separation from sociery and state alone. The rules and legislation of rhe Constitution en which rhe SATRC was based.
quasi-scientific working procedures are only one side of rhe story. Minimally, Though rhe ideal was not always maintained, despite the best intentions of
rhe legitimacy of the SATRC is based en a convention rhat tests en the com- rhe SATRC staff, nonetheless it allowed [hese conflicts to be controlled and
mon interest in there being an institution to deal with questions about the their consequences made relatively predictable. A consequence of this pro-
past and the dilemma of guilt and responsibility in a nonrerributive manner.2s cess is rhar rhe past is determined by legislative provisions and by rhe particu-
Mary Douglas has suggested that for an institution ro turn finto a legitimare lar manner in which legislation is interprered.
social institution, it needs "a parallel cognitive convention te sustain it," so This point is often forgotten in rhe comparative study of truth commis-
that conventions and practices are naturalized in social classification (1966: sions. Nearly all descriptions and analyses of official truth commissions rest
46). por an institution to be legitimate, "every kind of institution needs a on rhe implicit proposition that norms, action, and representations are basi-
formula that found its rightness in reason and in nature" (45). In other cally the lame (see, e.g., Hayner 1994, 1996, 1997). This means thar what a
words, there needs to be in "analogy by which rhe formal structure of a cru-
commission should do, what it actually did, and the manner in which people
cial set of social relations is found in rhe physical world, or in the supernatu- in charge of commissions represent and reflect en rhe wod< done all belong
ral world, or in eterniry, anywhere, so long as it is not seco as a socially con- ro the same leve) of reality. The result of this is that descriptions and analyses
trived arrangemenr" (48). 1 contend that this analogy is the encapsulation of of rhe work of these commissions end up being far roo coherent, and further,
historical truth in a legal discourse and set of practices, thar is, in rhe law. In that importan[ information is excluded about how mandares (and their inter-
rhe case of rhe SATRC, legitimacy is based on rhe founding principies of tire na) tensions) are handled by staff members while commissions are working.

160 STATES OF IMA GI NATION NATION-STATE FORMATION 161

 In tire process of truth production, Itisrorical events hact to be recou-
The Invisible Inside
Structed. Some events becana e classified as vtttav, and others did nos. 1his
The intensitti of chis public process has also crearen a certain kind of blind- requircd a series of iuterprerations and judgmcnts by every-da; workers ¡ti che
ucss around tito work of the SvTRC. 1 he rirualized public representations sATRC. Even though cite sArao car ried )ti¡ ntanp of lis t sks in che punhic oye,
onle describe oue aspect of irs work. The pm ate process of selecting victims the everyday work ofthe cona nissíon has, unti1 recentiv, largeie escaped seru-
who tvould t ell stories publicly was basicalhv a strategic process where judg- tinv."'l his has been so partiy hecause cite dalle work of the s,va RC eras to a
ments were made bv SA-rkc staff members to identin, tire good and the bad
largo exeent inaccesible The evetydav practico of object ngandr^ o^essmg
protagonists of thc pa,[. Uf cite 22,000 statements tito 5ATRc reccncd from
pasr experieuces from tire apartheid penad in tito catan: teas done colchan.
vi tinas of human rights violations , ipprosunately ro percent appeared in any public attention. in otiter words, the commission s evetydav work fúne-
public hearings ; che remaining go percent were processed behind the seenes. tions were retracted (invisible) from society, in contrast to the theatrical naturc
This processing behind closed doors veas about finding [he truth about tire of everyday bureaucratic public routines (see, e.g., Herzfeld tygz). This raises
past bur more soecifically ir was abou[ dividing things up into relevan[ and
questions, such as what charac[erizes che truth that constitutes che funda-
irrelevant truth and abou[ removing cite debele of nacional hisior that w
tion of the new nation-state? What makes tire truth produced by che SATRC a
no[ deemed to fall within tire human of tire SATRC . legitimate and authoritative truth in nacional and international arenas? This
This seleetive , creative, and publicly unobserved production of specific lcinds is whar 1 answer in rho foliowing section by examining the ways in which the
of truth claims is explored in the foliowing sections. categories of GHRv were negotiated in che everyday bureaucratic practice of

che SATRG.
individualized Truth
The new South Africa describes itself as a rainbow nation consisring of many
The Eueryday Practices ofTruth Production
different and highly diverse fragments (peopie, histories, and ethnicities). To
The following exampie illustra[es che everyday practices that were applied in
create a shared sense of the past in che midstof this diversity, the commission
che Data Processor Unit (DPU) of che sATRC to objectify reliable information
needs unambiguous information that must then be classified according to a
abou[ past human rights atrocities. The task of the DPU was to analyze the
human rights-based "trichotomy": victim, witness, and perpetrator-a hu-
statements received by che sATxc from alleged victims of GHRVS and to cias-
man rights vioiation-based classification schema.
sify and enser rhe information in a database. The DPO was only one of several
It is importan[ to loop in mind that in the landscape of a nego[iated settle-
units working together en chis project. Its work was the second step in an
ment, tire road che SATRC has chosen is, in essence, a liberal way of dealing
ongoing process of verification of che received statements. At che end of chis
with rhe past (Mamdani r996b: 4-5; 1997: 22) because the sATRc has chosen
process statements were either dismissed as being "out of mandare" or clas-
to interprer irs mandare from an individualized perspective.z" This has been
sified according ro victims, witnesses, or perpetrators of che GURV(s) who
done in ovo ways. First, che commission primarily documented che abuses
committed by che apartheid state institutions and irs foremost opponents in appeared in the statement.
Therefore, rhe incoming statements, on which all the units depend, wete
individual tercos. Second, it tried to capture as many stories as possible from
central ro che work of che DPU. 'Pite cooperation among thc different units
individual victims of gross human rights violations (Gxavs), a process of
consisted of differenr layers of "corroboration" of che information eaptured. de
"rransparent" data collection that produced material for writing up a new
This was done through a database, applying che well-known software pro-
common nacional history. Not everything enters the nacional seript. The sto-
grato Oracle, which classified and analyzed incorning empirical material in
ries forwarded to che sATae needed to contain certain information thatsuited
such a way that differenr unitslpersons could eommunicate and change the
olear-cut definitions of victims, perpetrators, and abuses to fin che national
seript. In this way, a cerrain grammar was used for che production of the captured information.
The methodology is, in essence, "positivistic." Each stop in che ongoing
"correct" seript.

NATION-STATE FORMATION 163

r62 STATES OF IMAGI NATION
data processing permitted a systematic cross-correlation and confirmation of registeis all statements from the Western Cape, en the telephone and
the raw data captured by the statement takers and refined by the data proces- ask her what to do.
sors.'° The aim of the cross-correlation was to produce a neutral and objective During the discussion it is decided that Sheila should write a "throng
process of fact finding, as required in the mandate of the sATRC. The complex summary," which would explain why the statement is not a GHRV state-
negotiations that chis positivistic methodology implied can be illustrated by ment, or, in other words, why the statement is out of mandate. The final
the following sequence. decision would then he taken by commissioner Mary Burton later on.
Mary (or another human rights violation committee member) is the only
The Data Processors person who can Cake che final decision on whether a statement is out of
Sheila has received a statement, which at first seems problem-free.'° She rypes mandate or not.
a reference number into the correct rubric en the computer screen and the Sheila then goes to the computer and begins writing the summary.
deponent, who is also che victim, is thus automatically given a reference num- One sentence stands out: "The bar was predominantly for whites." When
ber. She then goes through the statement te find other victims. Every time a finished, she decides to create the necessary ídentities en pages 2 and 3
name is mentioned she writes the new person's name down en a piece of in che database, because che bar was for whites and it could have been a
paper and makes the first classification: who is victim, who is perperrator, who confiicr with a political motive. She then consults the room again-this
is witness, and so forth. She then begins the creation of their "identities" on time discussing with several data processors-and it is decided that she
the screen by giving them a victim, witness, or perpetrator number, so that should go and talle to some of che more experienced data processors,
they will have their own page in the database. Extracts from my field notes who work on another floor with the Investigative Unit on corroboration
illustrate this process: of statements.

A problem emerges when die screen (the data prograni) on page 2 asks We talk to the investigative data processor and show her a printout of

for more details about the incident. Sheila rereads the statement and Sheila's work. She tells Sheila that she has done good work and prom-

begins doubting whether she is dealing with a GHRV incident or a job ises ro talk to Mary Burton after the corroboration has been done by the
conflict berween different workers in a bar. Investigative Unit. Sheila tells me she feels relieved, because it is "no

Sheila thinks out load: "7t seems more like an event than an GHRV. longer my problem, and because 1 have done nothing wrong."

How can a bar fight over jobs be classified as a GHRV?" She answers 1 then ask her about the sentence she wrote that guided che rest of her

herself: "In some cases it seems so, Lars, because boycotts " were used classificatory work: "The bar was predominantly for whites." The narra-
tive of che statement did not contain this information; the statement
as a weapon by the UDF [United Democratic Frontl, so everything de-
pends en the political context. What do you think, Lars?" had not said anything about which kind of bar we were dealing with. In
"I don't know." Sheila's explanarion, she refers to her own knowledge about bars dur-
ing the time of apartheid. She explains to me that "some bars were for
Sheila Iooks around the room and goes to Julia, the leader of the unir,
who is sitting at the back of the room. They quickly go through the whites only and others were for coloreds and blacks."

starement together and consult the ten pages of GHRV classifications. Eniployees in bureaucratic institutions like the 5ATRC try to fulfill [he ideal
Alter a long discussion and several consultations with the GHRV classi- of the constitutive separation between politics and science, well Imowing
fications, they decide not to create "personal details"-the witness, vic- that it is impossible. They strive hard to follow the rules, even while simulta-
tim. and perpetrator identities-because it is not a GHRV statement. neously acknowledging that it is impossible to classify and thereby postu-
But before Sheila has left Julia's desk, Julia changes her mind. She is late that the nature of apartheid's heritage consists of discrete and distinct
not satisfied; still doubting, she decides to call Shaida, the person who entities that can easily be ordered. In part, this commitment owes to an

164 STATES OF IMAGINATLON NATION-STATE FORMATION r65

uwareness of rhe intplicarions of not fulfilling rhe ideal if ir comes to public from rhe quesrion of dealing with a statement out of mandare, ro the qucsrion
lanowledgethe fcar of "messing up," as it is popularly expressed among of how ro socure its acceprance in thc classification system, so that identities
employees of the SATRe. And pardy, rhe conunitmcnr owes to a belief in the of vicrims, perpetrarors, and witnesses can be captured in rhe database.
correemess of this tvay of working: rhe bu tea notatie ardering of rhe world. Anorher dererminanr ni the process is ihe specific histore of rhe "normar
Rut jusr as important is comntission members' awareness thar they bayo a tive classification scheme" applied by rhe sATRC." This contextual hisrory,
mission rhat is biggcr than rheuiselves. I ❑ this serse, the ideal is a guiding which rcfcrs ro unspokcn practíccs, is also a crucial facrorin thc classification
framewurk a uecessary ficriou tb rheir mericulous work, which informs process. Sincc the data processors bogan rheir work in iggh, rhe normative
acrions and decisions oven when it cannot be preserved in rcality classification scheme developed from two ro ten pago,. Nearly every week,
new domains of violence were included or removed or rhe inrerprerations oí
Classifiration the categories changed. But information about rhe changos was nor always
The aboye sequence describes the negotiations involved in rhe classification passed on to staff, who had to implement rhe changos in rheir daily work.
of information contained in a statement. Each statement usually involves sev- When a chango did filler down in the interna ) hierarchy, ir generally took one
eral elassificarions, with their own contextual story. Contextual stories con- to two weeks to reach staff members. '4 The conss equences were, first , that rhe
tain important information if one does nor want to simply dismiss the work data processors were always insecure about whether they were doing rheir
of rhe employees as "bad" or "unprofessional," and if ore wanrs to under- work correctly, and second , that they were torced to constandy rewrite rhe
stand the metamorphic process this statement undergoes before its infor- work they had already eomplered to make it match rhe latest revisions ro rhe
mation becomes en (arti)facr. classification scheme. Thus, bureaucracy is a process rather than a rigid set
The first classification originares with Sheila, bar is later transformed by of rules and rcgulations.
communication with other persons from rhe unir (Julia, other people in rhe The dilemma for rhe people worlcing in rhe DPU was that they had to make
data processing room and on other floors). At first, the information con- decisions chal had to be aligned with rhe rules and classification schema. This
tained in rhe statement does not describe a human rights violation; this is required judgments and inrerprerations , which reshaped rhe past in particular
verified by Julia after a rather long negotiation. It does nor fit the definition directions and which are part of rhe reinvention of the pass." This raises a
of a GHRV, which is "the Icilling, abduction, torture or severe ill-treatinent of question : Is bending rhe past, as we have seen in rhe aboye example, done
any person, or any attempt, conspiracy, ineitement, instigation, conimand or deliberately in bad faith ? 1 would say no. 1 suggest that instead a certain "will
procurement to commit an act referred ro" among the four basic classifica- to order " seems to be at stake in this process of bureaucratic classification.
rions (Act 1995: 1). This leads os to che crearion of identities in rhe forro of eategories of vic-
When Sheila classifies the event as a GHRV after all, Julia recognizes rhe tims, perpetrators , and witnesses and rhe specific context of this classifica-
classification as a GI-[RV. She then immediately sends the decision to another tion system.
person for further verification. When I later asked Sheila how she knew that
it was a GHRV, she responded by saying rhat she could nor know with cer- The Wíll to Order
tainty, but thar her own knowledge about bars in South Africa during rhe Week after week, those working for rhe sATRO are confronted with painful
years of apartheid made her cure, as a colored person, thar this was rhe case. stories of people whose lives have been broken down during the years of
In this categorizarion, rhe arti(fact) rhat we are dealing with-"a bar predomi- apartheid-and there is flor much they can do for them.'° Among ordinary
nantlyfor whites"-is an active participant in the negotiations benveen her and staff members, frusrrations started to surface in rhe beginning of 1997, when
Julia. In this sense, personal hisrory and local knowledge-all rhe unconsid- the victim-driven hearings were finalized and rhe amnesty process rook off.
cred factors that organizo daily life in South Africa-mingle with the classi- In eontrast with the victim-driven hearings, considerable resources were used
fication schema. on amnesty applicants in rhe forro of legal representation to make rhe process
From the moment rhe caregorization is complered, rhe discourse changes fair according to international human rights standards. A sense of betrayal

166 STATES OF IMAGINATION NATION-STATE FORMATION 167

 was common among staff members, who felt they had often exploited the proof provided by a witness within this positivistic methodology ranked lower
victims by putting pressure en them to recount experiences publicly, without in the hierarchy than other kinds of evidence . However, this changed in the
giving them anything back.17 last months of the work of the sATRC, when it was realized that they had to
However, ífpeople are classified as victims they will at least come into con- dismiss thousands of statements as out of mandate due to the lack of objec-
sideration later, when reparations are decided on. After the victim, witness, tive material verifying the victim statements. Then the witness category be-
and perpetrator identities have been defined, the data processors produce the carne important. The sATRC even went so far, in cases where there was neither
list of dependents (their identities). This list guides the distribution of repa- objective proof nor witnesses, to say that if people signed a solemn declara-
ration to victims (TRC Igg8b)." Therefore, when a person is turned away, this tion the SATRC would accept the statements as true and entitle them to repa-
may have enormous consequences not only for the victims themselves, but ration (based on fieldwork notes and interviews with staff members, 1997-
for their families as well. Without the victim classification they will miss out 1998).
on the chance of a whole range of future entitlements, including school bur- Before 1 continue the analysis, 1 will give another example of how localized
saries, medical treatment, clean penal certificates, and jobs, al] of which in- knowledge and the will to a certain order, when applied together with specific
form the prospect of a different future. bureaucratic competencies, not only influenced the classification process, but
There is, therefore, what I call the "will to a certain order""at stake among also made it a powerful tool. The example is taken from the regional Investi-
the employees of the SATRC and commissioners: a will to classify people as gative Unit (iu) of Cape Town and is partly based en my own observations
victims. This happens in a conflictual balance among a range of strategies, and partly en the retrospective reflections of the investigator whose work 1
such as the imperative of being objective, the feeling of betraying victims, and observed.
the will to give victims a chance to access che new state welfare systems. In
this sense, classifications are settled and fiad their stabiliry without anybody The Legal Mínd
necessarily formulating this preference openly. The tu was tasked with analyzing and corroborating statements to find out
On the orher hand, for every victim identified there has to be a perpetrator, whether a statement fe¡¡ within the mandate of the sATRC. This was the last
a person whose life can be ruined if wrongly (or rightly) classified as such. step in the chain of data processing before the GHRV cases were submitted
The classification of perpetrators is notjust coincidental; it springs from the to final decision making by committee members and commissioners. On sev-
logic of how GHRVS are identified and counted, When dealing with a GHRV eral occasions 1 shared an office with members of the tu in Cape Town. The
there has to be an agent. As the key words guiding the database indicate: following event took place in 1997, just before the investigative process was
"Who did What to Whom?" Often, nobody is mentioned in the statements, to be finalized.
but still the database requires a perpetrator, a request that comes up on the The leader of the unit carne into the office with a bundle of cases. The cases
screen automatically. In these cases, an "unknown" South African policeman, were handed over to the investigator with the words "Take a look at these
South African Nacional Defense Force member, SDU person, or other is writ- cases with your legal mind. 1 an1 not satisfied with the decisions taken. See
ten into the database, helping to establish the overall patterns of GHRVS. what you can do." This particular investigator was one of [bree persons in che
The importance of the identity ofwitnesses and the implications for people unit who were trained in law. The cases aff belonged to statements received
classified as such are less clear. The role of the witness category was down- from the rural arcas of the Southern and Northern Cape, arcas where the
played in the work of the sArRC from the beginning of the process. Excep- sATRC had used very few resources and therefore liad few cases document-
tions were the cases where there was no "objective" material for verification, iug GHRVS.
such as medical records, death certificates, or mortuary, court, or peison files. Over the following days the investigator looked through che cases and was
In such cases, the sATRC had to rely en the affidavits captured from witnesses. struggling with them. He told me: "All the cases elearly speak about GHRVS
The identity of witnesses was in chis sense initially less important because the but none of them have any objective material to support their claims." The

168 STATF.S OF IMAGINATION NATION-STATE FORMATION 169

 whipped, but diere was no documenta,y proof, filie was nor arrested. She
investigator in charge of the original investigations, a foreigner, liad classified
had been tear-gassed and conlplai ned of bad hcaldl and rhat type of
rhem as out oí niandate, which, according to my ot]ice fellow, vas "rcason-
thing. Well, ncere was one witness who saw [liar she w,is ^1•hippcd. You
ahle enough," but it evas ouly becuuse rho forcign investigator "did not un-
obviously had ro corroboratc chis wirncss, hur no: much, come our ot it.
derstand the historo ofviolations in South Atrica." lle then wrotc up chocases
fhen slic showcd me rhe mares ir hcr back and, invasn t 11Ice electrodes
and hended them hack lo the unir leader.
placed on your stoniach or on your fingernails- bur ir was issault. Onc
\Vhcn 1 lato intcrviewed clic investigator aboi t die oven[. he told che fol-
could say liar there reas a political motive, hecause filie +'as a political
lowing storc: candidato, bur uo direet link. lf you leave tt Ijke that ir's a bit roo cague,
The worst part of actually doing a case is that you are sitting and reducing flor so vague, but it is notbolstered enough as tu actually justifj a reasen,
everything that you've experienced on paper. 1 think that was ulrimately classify che case as a violation [within che context of che SATRC]. Sur it
che most difficult part of my work. One person is showing os exactly you contexrualize it, if you go into her psyche, if you go roto the times
where he was shot in che eye, the family members have to go and stand when it happened, if you use the graphic insight detail you have of these
in welfare queucs, the husband and father is now an invalid, and nobody people-the trauma that these people went through-highlight their
can support the family. As an investigator you have been there, you have position, how it affected thein and you bolstered che case with this in-
seco it. Every time you deal with [hese people you're moved. formation, rhen maybc.
The fact that they are hclpless is very difhcult. There is nothing you Ifyou go out of your way basically to have a person classified and there
can do and rhar sense of helplessness both makes you powerful and are no gray arcas, then either you are a victim or you are nor. If you detail
Ieaves you feeling shitty. You know exactly what can be done, you know a case like that, then a report becomes a tool with which yen elassify
that this fucking matter falls outside the mandate, hieres nothing that people. You make the recommendation; it is a GHRv, and ten ro one your
che Truth Commission can do for this person. That's why you're walking recommendation will be sent to the human rights violation comniittec
around with your list of thc Human Rights Commission, Land Commis- [who cake the final decision]. (Based un fieldwork notes from 1997 and
sion, lawyers that can take deis matter further, the rype of struetures that interview material from 1997-1998)
these people did nor have access ro [before]. But access is nor everything,
To let "a report become a tool" and "bend the rules," as it is described
access does not help them either. You are their Iast chance.
aboye, takes particular skill and knowledge about how burcaucratic systems
You can'r get angry at rhe person [for wasting your time], but once you
function and what is required to let a case pass. It requires the abiliry to bal-
leave, you just get so pissed off and you think there's nobody to blame
ance subtly between too much and roo little: to find the middle ground be-
but rhe system. And rhen you obviously relive [the stories from the field]
tween eonfiicting demands. In retrospect, it can easily be misinterpreted as
when you sit with a case and have to take a decision. This is why it is
deliberate fraud, but in my view this would be a misinterpretation. What 1
always important to have a background, a South Afrícan background, so
want to highlight is che capacity for sett]enlent thar 1 have called a certain will
that you can acmally identify with what was happening. The fact that we
to order, which gave the past a direction and a clear, unambiguous contenr.
could identify with people's plights obviously influenced the report writ-
ing as well.
You have ro judge whether a case is political or not, and you obviously Metamorphosis and die Creation of Bureaucratie (Arti)Facts
don'[ want to mislead the commission because 1 am here to do a service
We are now in a position to answer the question: What characterizes the facts
to the Truth Commission and tu the periplo [of South Africa]. But I'm
produced by che sATRC? The facts on which the new nation-state are founded
not going to lie. There are some times when you can actually bend the
are (arti)facts. These (arti)facts-the database, the identities in the forro of
rules.
victims, witnesses, and perpetrators, and the memory of the process-are
Let me give an example. There was deis woman who had been

NATION-STATE FORMATION 171

170 STATES OF IMAGINATION
 constructed through complex processes of interaction and negotiation. We
lyze the relation between the visible-the ritualized access points-and the
have wimessed a metamorphic process, where answers are wrested from the
invisible bureaucratic routines.
narratives of victims captured in statements. It is a metamorphosis in the
The constitutive separation between politics and scientific procedures can-
sense that the initial narratives were characterized by a high degree of ambi-
not be maintained if one looks behind the on-stage dintensions emerging
guiry (and instabiliry), which, througb the classification process (ofwhich the
from various kinds of official public representations (whether media, aca-
DFO and the lu are only a small part), were finally transformed into a set of
demic, final report, or ritual representations) and public transparency. Nev-
clear and unambiguous identities.
ertheless, the separation is a necessary fiction. An echo of Max Weber can
As we have seen, this stability did notcome easily. Complicated translations
almost be heard here. He pointed out: "The march of burcaucracy has de-
and negotiations took place to "find the truth about the past." Bureaucratically
stroyed structures of domination which liad no rational charaeter.... Every
eonstrueted truth necessitates, in the discourse of rights, unequivocal cate-
bureaucracy seeks to inerease the superiority of the professionally informed
gories so that entitlements in the form of reparation can be distributed. How-
by keeping their knowledge and intentions secret. Bureaucratic administra-
ever, when viewed from the perspective of everyday bureaucratic practices, the
tion always tends to be an administration of 'secret sessions'; in so far as it
bureaucracy, which was ideally perceived te be filtered for the "disturbing
can, it hides its knowledge and actions froni criticism" (1968a: 244).
noise" of political contention and personal histories, turned out ro be some-
What Weber so rightly brings to out attention is the double nature of bu-
thing far more complicated.
reaucratic instimtions. On the one hand, bureaucratic institutions accentuatc
It is in this landscape ofwill m order, uncertainry, decisions, controversies,
their rational side-here, in the form of transparent criteria for defining who
personal stories, political preferences, an Act, a database that demands infor-
are victims and perpetrators. The flip side of the transparency of bureaucratic
mation, normative classifications, and high stress levels that the foundation
institutions like the SATRC is the public erasure of irrational actions and de-
of the new nation-state is created. The point is relatively simple. When the
cisions, mainly by controlling the circulation of knowledge."' This is done by
complex translations and negotiations among different role-players are com-
hiding what is going en inside the sATRc, by making the irrational actions
plete, all that is left on the computer screen, in the investigative reports, or in
invisible, through a whole range of naturalized and logical strategies such as
the final report of the sATRC are the categories: victim, witness, and perpe-
"disconnecting" the work of the sATRC from other state institutions, security
trator. The negotiations described here vanish and we have the hard social
checks, media departments, and other kinds of ritualized "access points,"
facts, which in the future will circulate in new networks of statistics, history
and partly by emphasizing openness and transparency. One could call it para-
books, policymaking, crime prevention, political election arguments, and the
national history. doxical that it is the maintaining of the second separation, this partly invisible
and pardy transparent )gnus face of the sATRc, that makes the truth produced
by the SATRc legitimate and authoritative. Sur as Michael Taussig (1997) has
The Visible and the Invisible
illustrated, the production and reproduction of the idea about the modero
The public debate on the sATRC discusses whether the commission fulfills
state is based en more magic than we normally think. The importance of this
the imperative of writing up a new history in a neutral and even-handed man-
observation points to the significance of looking at and analyzing modero
ner, disconnected from the political and social (orces at play in South Africa.
bureaucratic state instimtions as something other than highly rational bodies.
In general, it is astonishing how much attention each step of the seventeen
(in the end, there were only fifteen) commissioners or each press release at-
tracts in the media. The SATRC has been hailed as the most transparent offi- Final Words
cial commission so far. Through a whole range of "access points,"°" such
1 have attempted to open up che ready-made facts about the past that consti-
as public hearings, public processes of eonsultation, and media coverage on
tute the foundation of the new nation-state of South Africa. 1 have analyzed
an unprecedented scale, including both radio and television broadcasts, the
how facts are manufactured and have reconstructed the double separation
SATRC has created a sense of transparency around its activities. Here, 1 ana-
between science and politics on the one hand, and between what is visible

172 STATES OF IMAGINATION

NATION-STATE FORMATION 173
and invisible ora che other. 1 did this by moving bchind tire rirualized access has seldom been qucstioned. Taking these invisible and at times hidden pro-

points provided by the SAT RC, toward a place where tire truths 011 wllicll tire eesses luto account is not che same as bcing a cultural sptciccrderber (spoil-

ucrs° cation-state is bascd are constructed. Altbough tire first constímtive sport), as certain sociological and anthropological perspectivos could sug-

divisiou-benveen science and politicsis unable to accouut for dre nego- gcst," but co calce scriously che power ot modero bureaucratic: institutions
tiations. link,, and infiuences benveen dre s.vFRC and politics, it is neverthe- with decir insistence on all-embracing coherence, knowledge, and truth. It is

less important to bear in mirad tire svords of Bruno Latour: "lt would be a also co aclcnowl edge that, even though nccy are impossible to mainrain in

misrake to denv che edectiveness ot this separation- (1993: r;). 1',7 separating predice, as indicated bv Aletta Narval in her essay in chis volunte, they can

science and politics and kceping che interface benveen che nao domains invis- still be meaningtiil practices to strive coi. One majar analytical consequenee

ible, in extremely efficient, successful, and often "trusted" kind of institu- has lo be Calcen note of in this analysis. Tire information revealed through che

tion is created. The SArRC has indeed been effective in gathering no impor- access points of the snrxc is not interesting as such, as Truth with a capital

tant, officially sancrioned body of knowledge about the apartheid past. It is T. Instead, the information Inade public in che hearing process, the media,

cqually important lo bear in mirad when we are talking about state-governed and the Final Report is important because it is here that the representation of

bureaucratic institutions that the separations between science and politics, tire new nation-state is displayed, both the "heroic" inclusions and che ex-
or what is visible and invisible, do not have a rigid or totalitarian character. cluded "other" of che nation-state.
We do not encounter a state that has an all-embracing character, nor do we
fail to fbllow the burcaucratie procedures stated in che law. What we do en- Reoisiting the National Srript
counter is the ideal of che state as all-enibracing, and Chis is something quite At first glance, hard social facts nearly always present themselves as innocent.
It would therefore be reasonable lo aslc liow hard social facts, in the forro of
different.
The robustncss of che work of die ssATRc-che representation that is under the GHRV elassifications presented aboye, interact with the national script

construction in South Africa co legitimize the new nailon-state-relies on the and with che construction of thc past (i.e., Afrikaner culture) as the Other of

work of "experts" working for the commission and che visions and imagos the new nation-state.
thcy present in their work. The veraciry of chis representation has been on For che new national script lo be accepted so that South Africa beconies
trial since the release of che Final Report in October 1998: legally, historically, internationally recognized as a modero national state, it has to pass through

sociologically, and so en. 1 am sure that to a largo extent it will pass this the legitimate media of oficial truth and reconciliation commissions. In

scrutiny-and with good reason, because it is a meticulous and gigantic other words, truth and reconciliation commissions have become one of che
work, che real effecrs of which we cannot yet begin lo assess. However, be- means, together with international tribunals, whereby new nation-states can
cause most assessments of off sial commission work will be of a normative be accepted as bona fide nation-states. Apartheid was classified by the United
kind, many of the processes dealt with in chis essay will escape scrutiny. For Nations as a crime against humanity and che apartheid regime placed out-

exacdy this reason 1 think there is a lesson to be learned from che sATRC, the side che international community of democratie nation-states, with all tire

general features of which eould have wider analytical resonance. attached implicadons of economic and cultural sanctions and so on. Thus,
One reason for pursuing this analytical perspective is that a reading of the through che work of the sxTRC and its internationally recognized celebration

growing literature ora official truth commissions shows that it seldom, ifever, of a "culture of truth-telling," South Africa beconies liberated from its prob-

dcals with the interface benveen tire scientific-bureaucratic radonale and con- lematic past and finds its proper humanitarian self. Formerly shamefully mis-

crete bureaucratic practices-between what conimissions claim to do and used and appropriated by the omnipotent apartheid regime, it calces its due

what they actually do. In other words, there is a gap in che literatura when it place among other nation-states.
comes to the ways facts are constructed. Common to the growing literature The work of che sATRC is intended to mark a definitiva break with tire past

ora official commissions is che faet that thc relationship between che practical to define the new nation -state in contrast m the former apartheid regime. By
techniques applied to objectifying pass abuses and tire resulting claim to truth looking carefully anche everyday work of che SACRO and relating its invisible

174 STATES OF IMAGI NATION NATION-STATE FORMATION 175

 character to its public performance, it is my hope that it has been possible lo use oí ihe courrroom for graphic demonstrations of its policy and For scapegoating oppo-
understand some of the quasi-magical steps that make the Olear separation nents oí its law. However, rhe new democratic government has amply demonstrated íes in-
between new and old possible. What is hidden behind rhe access points is the ' ability to reinstale che role oí law, taking finto account fraud trials such as rhe 1997-1998
case against former activist Allan Boesak, as well as rhe trial of former apartheid minister
material not deemed valuable for the national script. The work of the sATRc
General Magnus Matan and ten other high-ranking former officiais for che massive misap-
is in Chis sense in a double maneuver of both displaying and removing: mag-
propriarion oí surte funds to train and equipan illegal army during rhe late 1g8os (see, e.g.,
nification and effacement in rhe same movement. In other words, rhe CAIRO
R. Wilson 1995a: 42).
is simultaneously about both remembering and forgetting. 8. Due te rhe continued control of che economic held by an affucnt hile elite sud muid-
national corporations and che poverty most black victinrs of apartheid are still subjecr lo, the
change is ultimarely symbolic. This change is concerned with codifying rhe popular menrory
Notes
oí apartheid suffering-which has been charactenized by being Huid, unfixed, ind frag-
mentary-ro creare an official history, or what is ofren referred to as "global aurh. " See
1- The groupings rhar excluded themsclves were rhe radical left and che militan[ right , both oí
R. Wilson (1996: 17) and Buur (1999) for an elaborare analysis of how codification tooli place
which opposed che negotiated settlement die former because rhe reforme negotiated in che
in rhe SATRc.
settlement did not go far enough, and rhe latter because che changes wenr loo far,
9. Umkhonro We Sizwe is rhe name ofamc's underground army, popularly called MC. which
z. There are severa! definirions of rhe South African "negotiated settlement." The coneept of
means''che spear oí rhe nation."
"settlemenr'' as 1 use it here refers ro "rhe striking oí a compromise, steering towards a
ro. This section is based en field notes from two of rhe Bikm amnesty hearings and rranscnprs
subtle balance between opportunities and dangers, between conflicting demands-notradi-
froni rhe amnesry hearings; see saTRC website: lvuvw.ttuthorg.za.
cal solutions" (Bauman 1995: 127). Ihe fact that apartheid was ended as a result oí a nego-
u. De Kock was found guilry before rhe s.ATRC irnncsw process began and .vas jailed For aea
tiated settemenr meara that rhe negotiating parties agreed to maintain che surte apparatus
years, whereas Nieuwoudt could fase a criminal trial should rhe saTRC deny bina amaesí.
intact and ro gradually redistribute jobs through affirmative action to formerly disadvantaged
1z. When, for example, rhe deeds of English-speaking perpetrators such as Craig Williamson,
social groups, mainly black people.
which were no lees horrendous Iban Nieuwoudt's or de Kock's, were publicized by a CAIRO
3. During interviews, staff members and commissioners oFthe saTRC srressed rhe importante
hearing, he was seldom presented as anything other Iban a James Bond-type superspy (see,
of che work oí rhe commission in relation te wridng a new history legitimizing rhe transition
e.g., "Ex-Spy Williamson Clashes with Pik," 27 December 1995).
to democracy while at rhe same time stressing rhe importante oí doing so in an even-handed
13. The data discussed in Chis essay were collected in rhe SATRC in 1996, 1997, 1998 and 1999
manner (based en fieldwork interviews, 1896, 1997, 1998, 19990-
during field research conducted under rhe inspires oí che Ph.D. project Te Establish a Truth:
4 . 1 suggest chal once one declines tú accept rhe limited depth oí reified public representarions,
not only is one forced to cake finco account `backstage" acrs, but 'onstagé' dimensions, roo, Victims, Perpetrators, Experts dad che Work of che South African'1'ruth and Remncíliatien Coinmission.
Department of Ethnography and Social Anrhropology, Aarhus Universiry, Denmark, which
change character. Often seemingiy incomprehensible, invisible and insignificant onstage di-
was subenitted in zopo. 1 am indebted to Dr. Torben Vestergaard, Department of Ethnog-
mensions become visible and Cake en new meaning. The usage of rhe ternas onstage and
raphy and Social Anthropology, Aarhus Universiry; Dr. Marrijn van Beek, Department oí
hackseage is an analytical distinetion berween official, public represetieations and aces and
Ethnography and Social Anthropology, Aarhus Universirv; Dr. Kristoffcr Brix Bertclsen, The
ihe invisible everyday practices sud acrs, which often are conducted by rhe same actors. The
Danish Cultural instante, Copenhagen: Professor Andre du Te¡t. University of Cape Town;
distinetion does not imply rhar backsrage practices are more authentic or right ehan onstage
and Ph.D. student Steffen )ensen at rhe Centre (br Development Research, Copenhagen. for
represenrations and acrs: often rhe tensions and coneradictions berween them chal observers
their critica) commenrs en carlier versions of chis [ext.
enceunter do nor arise for Che persone moving between che different domains.
14. Seven young ANC activaste were killed in a police anibush in che black rownship oí Guguletu,
5 Osiel is oí rhe opposire opinion, stressing rhar mal is a more effective public speetacle,
twenty kilomerers outside Cape Town on 3 March 11,86. The police claimed then rhar che
because ir can " stinulatc public discussion in ways that fosrer rhe liberal virrues oí rolerance,
young men set up an ambush for rhe police, during which they killed them in self-defense
moderarion, asid civic respecr' (1997: 2).
(see rRC 1998a, 3: 451-53).
6. The amnesty process, en rhe contrary, turned out ro be far more legalistie [han expected. lo
15. Vlakplaas is a farm in rhe Orange Free State from where rhe c-section of die Securily Pollee
avoid rhe rvpe of critican. incurred by tus Nuremberg trials for not allowing aceused Nazi
operated and planned covert operarions against enenties oí rhe apartheid regime. De Koclc
leaders proper legal representarlo,: che new sean, through die Legal Aid Board, has spent a
was rhe lasr leader of rhe una.
stupendous amount of money on legal representation For amnesrv applicanrs.
16. This coneept is developed by Ernesto Laclau and Chamal Mouffe (1985: ra7-34)-
7. One has te be careful making suela a statement, because rhe apartieid regime nade frequent
17. I do not claam that Chis has besn a deliberare cilla rt oí rhe saTRC My point la rhar ir is a

176 STATES OF IMAGINATION

NATION-STATE FORMATION 177
 ;h
Ihc negoriared scrulement, ore could sao risas nc, Tac re presents rhe atare on a la
plcnuc cumrgiug Prona .i comhinariou r cevcral pro. cs.cs &.r Miel, clic sm5u:_ duo ro it,
(Ir , srlll a zntc ivsriutrion hura orar, insr.tufiun .tenicing ouóidc the sra
public prohlo. procided rhe spacc. tl:.:. t .c .nrhv ..I
z} Ore of che t hacen oBlc'ul 11111111 .1- 11,n, :.na1}", 1 t I lavner a h c r '. . .t i " .
S Tiro AIril.m bond +ivf nn rl llcl 11 3rcc rb l Ir 'uan clu'a cr^anvation "' 1 " .i,, C m u r. 1,.1
oili a r truth ....m. ' i ... id n hu :iris da rert .o th J,
f. üu, mlistalltalit lr o j,m I uf h t taizanu leso mphinnrv.
luyo int ol.ti n- t li'in.u Iighe mplen a tJ., 1 la i e 99:
lut . toda iawh Atn t a.. f r.. n t cc 1w 1Il1,1 ueu s t o t aservatt. e, rl tisti.m
a. Un n ht debate nc,thrl, csn is e moretni ice mstitu 'r tght ct tic ..,_m:
prntessionalx . ad imellccw I . oppae I n, u.tns Lvma. n . T s' Afrik.merbond seas cstah-
of puhl-c Ir, ali ag ano ' grilli g" inher nt ha sise ! .t.nge process.
lishc lor rhe promotnn "f n i e . u : e r poliGetl, aul...rtl. c td cconomic interesa, ,iris rloa•
ab. tA.Hlmdasi tr99nh Isl 2 [t hhcra u u lizo hui s..o et
!¡al],, sorbe Nmonal faca, 1!"r lo' aire q. ah id gocen san[ herwecn 5948 aire 1994. Sin- uuh's si0 cc caos r I_ '...u-, Ir cr..
!hcs ceo: v: ith tc[ onc, u. H.0111,11
:qq; tira ti uerairv hn,. L .red u¡tan c r.. menPOar' 11-1 .comen Shc:ancTbon' -o;gco 1,, of spc ti c d a, a s_ s n. n-1
disco sosa aipnuc l1 c „ .5 insuurvo a hmd ir ", 1
1 ics ovada ANC" x9991. mendoos ¡la particular "torced removals;' nce pass Iawsand che souoma legar' r .
Mbeki mcn-
rq Many Afrilcmers have dedicared themsd. eo tos rhe ne . South Al,¡,,, and Thabo apartheid. Instead of making it posssible m acrivdy' deal widt lit, deprived li.l g co,aditi,n<
specch severa) pro fess io os from wlacre "good" Afrlcaners have emerged , includ-
cioos ir¡ lis of30 mtlli0n people and co deal powerfully itli tiro causes, che indtvidualtzed hunsan dghre
ing pol i ocian, j oo toa lis ts, rriters , and business people. perspectwe of rhe SATRC reaches only 22,000 victints. Ir [herefore works asan adeclogical
if chal-
zo. Institutionalizatioti refiera Itere ro a "legitimized s ocial grouping ," assuming rbat cherefore rhe referente m
smokescreen justifying the social and economie order of today -
lenged. it is able to res[ its claims to legitimacy on something more tiran jusr a purely tnstru-
"banalicies ofevil."
meotal oe era ctical a rrangem e nt , for exanr píe, by hein g a bis to res[ their clac ms to legiti-
27 . ncc firsr real esposare oí rhe fragili[y of rhe everyday worie surfaced during tire Heidelberg
enough m
macy en their fit wirh trae namre of rhe univcrse" (Doaglas tg86 ; 46). Here it as Here, one of rhe co m miss ione rs.
'lavern massacre amncsry hearing in Cape Tosen in 1997 .
sao that 1 h ereby e xclo de tire idea of a u t n si¡ m tion a, u rerely an i ns au ni ental oe provis tonal
Head of che Investigadte Unit Dumisa Ntzebeza , seas implteared by a gardener, who aIleged
practica) arrangenaent. chic he had sean Ntzebeza drivireg tire car aansporfing che weapons used ira tire qttaak, He
papales
zi- 1 do not clains cha[ everyone woddng for tire 5ATRC espouces rhe ideal ofsepa rati ng Iater re[racted bis tesrimony, but tire impl ication of aire leading bario mcrnber of tiro sATRt
aod scaence . tnstead , 1 follow Schmadr, who describes idealization as being negatively defined
taroughr une of Ihc mosr serious cace rocas to tire surface among tire everyday staffmembees
polincs and science, The
(,9g3 : 6). The point is that no one disputes rho value of separattng , for or agatnst Ntzebcza,
of che sATae , who divided amo two groups : bisele and whrce
separatioo is [herefore sor a unit'ing ideal ni tire tradiraonal cense, but something chal is each case oí Gestiv. ¡sur
28. The intentions were from rhe beginning tu properly ' investigate '
held in common by being indisputable . lit Schmidt ' s analysts oí architecture and ats values favor ofa corroborarion saaregy, whea
be che and of 1997 tisis merhodology seas dropped in
and ideals , he defines indispumbiliry : " Wlmt cae find as che not yet experienced disagree-
ir became apparenc thac títere were 001 enough e esources avaalable for such a resouree-
ment-che disagreemen[ is unarticulated and peed under certain circumstances non be ar-
intensave srraregy. lt has tos be men[ioned that onc reason fosa [has shaft was rhe resoume-
cae imntediately dtscover rhan all disagree as
ciculated. But fthe values have tú be articulated , which bogan in April 1986 , nraleing
intensive ' hearings-driveti ' public work of rhe 5ATRC ,
m what they were not dtsagreeing on-and rhey won't ,ver agree thar nccy do, in fact, das-
tt nearly impossible unir [he Invescigative Unit co den¡ wirh cases tirar were not going m be
agro," (7) .
presenred in public A backlog oí thousands uf cases ( around gis percent oí rhe cases dad
az. Alchough ir is true rhat che everyday work oí processing information in che sATRC seas , which had ro be
non cocer Chis public process ) accumulared ira rhe offices of rhe sarRe
opelled out in no appendix to rhe Fanal Report, as being positivista , at as not autamatically ( basad en fieldwork material from 1996, 1997,
dealtwith to fulfill che mandare of rhe SATRC
even though chis
given rhat Chis scienrifie positivistic ideal is aemally mandated by rhe Act , ntins oí tire
1998 and interviews wirh investigarors atad leaders of rhe differenr invesrigative u
is the elaim tirar rhe commission malees in nce Final Report . In facr, che Act ttself doses not
SATRC).
refer ro che positivistic sciencific ideal. Instead, itrdividuals vithin rhe commission nade , Tisis elassifica-
29. Tiras seas done te identify rhe data , which can be categorized as ' falsa' data
choices wheia interprecing general eoncepts such as "obleetiviry ," " evos- handed," and "in- rhe normative classifications , and die
non ts comacinually negotiated among rhe tracemeni ,
dependence " an chis way. The problem is rhan by stressing cite positivistic ideal guiding rhe the statement is experamented wirh and used ni
1 stress tirar títere data analyst or processor . In Chis process ,
worie of che si-rac, one eould expect rhac it guided al¡ facets of rhe SATAC.
the eoer9day work, arad differenr ways , fuscas che normative classifications are negotiated among rhe different role-
are majo: censioos among tire process ( especially ihc vieron hearings ), , investigarors , human rights or
dtfferencand distinta stages players: coro tn tss aoners , smtemenrs rakers, daca processo rs
che product (rhe Report), and rhat rhe process irself caen chrough
and so en.'fo understand rhe eonplex process (ti
ganizatians , viciarais and perperrators ,
depending on .vhac paa of nce cA' rac process one analyzes . In días essay 1 [herefore lomee
establishang rhe faces A011[ G H Rvs a[ as impormnc tos renrember that in che mostrucrao a and
my coneerr wtch rhe bureaueratic logic as a core pare of chis eomplex and avecen process,
(for further oegotiatiola of whar "frise ' data are, trae smtemenr as ove II as rhe processors and analysts
bus I do no[ daten that rhe positivistic viethodoloo seas hegemonic all along
and tire conipurer software are a ctive particapan i s ira tire dama.
elaboration , ere Buur 1999).
30. The namcs of che everyday workers have beso changad.
a;. Becros e a mate roscimtiora cantror be used be the new governmerar m investigare uselfdoe co

NATION-STATB FORMATION 179

178 STATES 01 IMAGINATION
 31. (me coufd hace expected che use of 'labor strikes" to be more appropriate here [han the use
of 'boycon." Boveort as a strategy during the campaign of che United Democratic victim can receive is 120,000 Rands (approximately U.S.$zo,ooo) ovec a period of six years,
Front was and the lowesi amount is 80,ooo Rands (approximately U.S.Ss3,Soa). Whether the govern-
"a campaign not to buy or use certain producís or services " and was an inherent parí of tire
ment will implement the sATRC recommendation regarding final reparations remains an
language in che Dru room, where most of the data processors were sitting.
issue in South Africa, much ro the anger of people classified by che SATRC as victims of
32. The intention behind using Chis spelling of (ard)facts for artifact without brackers is that I
GHRV5, who feel becraycd by the SATRC promising relief without following through.
wani to direct atrention m the double nature of che concept. iris boca a human construccion
39. The concept is calcen from lars-I lenrik Schmidt (1993), who relies en the work of Michel
as well as a "fact" thac connotes something more salid.
Foucault:
33. The "normarive classification scheme" was tire emic name among data processors in the
40. The concept of"access points" is developed by Giddems and refers to "points ofconnection
Cape Town office for what was known as the "controlled vocabulary, of violations" by more
between ¡ay individuals or collectives and nce representativas of abstraer systems. They are
computer-fiterate and human rights data base -trained staff.
places of vul nerabiliry for abstraer sysrems, but also unen nns at which trust can be main-
34. At management and commissioner leveis ofthe SATRC che "classification scheme"or"con-
tained or built no" (1992: 88).
erolled vocabulary" was actually reduced, not augmented, from over nvo hundred violations
41. By emphasizing tire question of control, 1 can be said to differ froni Weber, who. as the quoce
to fewer chan fifty, of which only twenrv-five were used in any meaningful way to classify
indicares , saw bureaucraric instimtions as "keeping dreir knowledge and intention secrer'
people un the database (based on Internet discussions with ene of tire creators ofthe data-
(t968a: 244)-
base, Patr,ck Ball, 1999). One explanation coufd be that a range ofsubcategories was devel-
42. Most promincntly Pierre Bourdieu who, inspirad by Durkheim, states that a practica¡ bel ief,
oped for each of the mitin categories in tire "conttolled vocabulary." According to Ball, the
in Chis case, that che eonscirutive scparation between scienee and politics is maintained, is
head cara processor in the Johannesburg office organized che effort and directed its appli-
"an inherent pan of belonging" (1990: 67). Processes of belonging are, in Bonrdieu's un-
eation among data processors all over the country, bur, as mentioned aboye, the diffusion
derstanding, double-edged. On tire onc hand, they confirm the mken-for-granted under-
of the changes was rather slow. This apparent conrtadiction tells as own story about how
standing people hace of dteir own life conditions, and on the othcr hand, Ihey create "mis-
differently hierarchically placed people within the same institution can interpret and work
recognition," a kind of amnesia rowactl that which tiray confronr the raken-for-granted
with the "same" heudstic rool, cal¡ it something different, and interpret its development
differently. reiation to clic world. "Genesis implies amnesia of genesis," he polemically writes (1977:
19). The relation Bourdieu points al: is die relarion benveen what une can speak about and
3s. Because itis not possible co define every instante as a GHRV, there are, as mcntioncd, many
the unspeakable, and the role of the anthropologist, who is commined co do more thau just
checks and balances in place within che srructure: persons, normative classifications, meco-
reproduce the self-understanding of groups of people, their taken-for-granted reladon m
ings, and so on designed to pick up "mistakes."
the worid. By speaking about what people Cake for granted and objectifying ¡t, the anthro-
36. Right from che beginning of the SATRc ¡t was planned to grant what were called "urgen[
pologist can easily end up being sean as a cultural spieleerderber-telling people that ches'
rehef reparations" ro victims ofGlaRvs. Severa) countries , including Denmark and Swirzer-
misunderstand their own lile conditions. The question is whether Bourdieu in his rheory of
land, granted up te $1 million each to the Presidents Fund for Chis purpose, but che process
practiee does not put roo much emphasis en belonging and che reproduetion of a cense of
nacer really materializad. The firsc payment of 2000 Rands (approximately U.S. $350) was
belonging, thereby ending up wich an understtnding of individuals and groups ets well-
mcher symbolically paid just a few days before the ofhcial closure of the work ofthe sATRc
in July 1998. integraced entities . The question is whether people, and groups of people, can both accept
the loss of innocence that objecrification imphies--for example, by acknowledging tirar the
37. Among staff members a phrase ofren repeated when they aired [hese senrimencs
was: "Tire
constimtive separation berween politics and scienee cannot be mar ntained-and at clic sume
circus comes ro [own." This phrase refers to the intervention process, when the SATRc
time still be eommitted co it without Chis adjustment neeessarily having fatal consequences
wouid set up a hearing in a specific locality in a period of a few weeks. Stacement takers,
for their belief in, for example, che work of the 5ATRC. Thoinas Blom Hansen s contribution
logistics officers, invesrigators, researchers, the media, and finally the commissioners in
co chis volume seems co impir so.
"their big Rmws or Mercedeses" would arrive for the hearing. 'The day afer tire hearing "che
circos" would leave mwn and che local communitywould be "forgotten" (based en fieldwork
notes, 1997).
A workshop was held in Cape Town for staff members in 1997 in which severa¡ commis-
sioners and commiccee members parricipated. Tire aim of tire workshop was to creare an
invernal forum where tire frustrations arising after tire shift from victim-driven hearings co
amnesty hearings coufd be dealt with openly.
38. ¡tis proposed hy tire Repararion and Rehabilitation Committee that che maximum antounta

180 STATES OF IMAGINATION

NATION-STATE FORMATION ISI
 \A

-rmrseLn£ Chis social fintase, and rhe rae is a pardeipant 'a rhe srruggle for
instiruting a hegemonic conception of rhat feno3sy or, a, 1 prefer to cal) ir,
RECONSIRUC1INC NATIONAI. IDENTITY
social imaginare. DI iwíng on rhe dtsrinetion benveen mvth and imagmanL
AND RLNECOTIXIIN(i M EMORY The Work of rhe TRC developed in rhe wo k oi Lidio, 1 irgo, that Sourh \trica tices a su I ale
benveen a nndtiphciry oí myrhs of nationhood.' That is. die posrrransi6on
Alerta J. Norval períod has been marlcec 1 a series of srruggles -round eoneeptions ot com-
muniry, nationhood, erhuiciry, and so forrh, cach oí which contribuyes ro and
conrests rhe institurion of an overarehing conception oí idenrin. New'n
rypically emerge during and as a result ol periods of deep dislocarion and
may be regarded as attempts to suture the hssures that have opened up as a
result oí those dislocationsl . There is no doubt rhat during rhe ig8os and
199os South Africa faced an organic crisis, parí of which entailed a putting
roto question of rhe apartheid imagina)uring rhe final years of the crisis
rhe discourse of nationhood articulated by rhe African National Congress
(ANC), rhar of nonracialism , dominated rhe polirical landscape as the alter-
in Chis essay 1 attempt to disaggregate severa) important dimensions of rhe
native to apartheid. However, even Chis apparent dominante was contested
manner in which rhe memory of rhe pass is being negotiated and recon-
by altemarive discourses such as thar of "sclf-derermination" increasingly
struered in and through rhe work of rhe South African Truth and Reconcilia-
shared between rhe Inkatha Freedom Parry and rhe far-right Afrikaner Free-
tion Commission (Tac), established by an Aet of Parliament to investigate
dom Front ." In rhe years following the first democratie eleetions rhe unques-
and expose gross violations of human rights' rhat took place under rhe apart-
tioned acceptance of rhe discourse of nonracialism has come inereasingly
heid regime, covering rhe period from March ig6o to May ro, 1994.' 1 argue
under pressure . Without rhe inrmediate presente of "an enemy," rhe apart-
that te do justice to rhe complexities of this process one has to explore rhe
heid state, it has become more difficult to hold rogether "rhe people," for rhe
relation benveen memory and identity and, more specifically, memory and
uniry of rhe people depend, at least in parí, en its opposition te rhe forces of
national identity. In short, I wish to bring to the fore rhe fact rhat rhe institu-
oppression . Having to assume the mande of government in a period of tran-
tion of a new national imaginary that has to articulare a relation ro apartheid
sition characterized by its search for reconciliar ion among different sectors of
history is no simple master. The issue is complicated even further if we want
rhe sociery made rhe task of reconstructing a widely acceptable conception of
that process te facilitare an opening up onto a postapartheid, posrnational
nationhood even more difficult. It is in Chis terrain thar rhe work of the TRC
sociery. I intend to offer a certain reading of the logic of apartheid and rhe
has te be placed if it is to be understood properly
negoriarion of its memory in rhe work of rhe TRC that may point to such an
Cribes of rhe Tac have suggesred that it is nothing bus an insrrumenr of
opening, to a future that is no longer dominated by apartheid.' However, 1
rhe state, and of rhe ANC in particular.' This criticism is based upon a par-
argue rhat chis is realizable only en condition that certain of rhe inherent
ticularly primitive conception of rhe state and ofren ignores its legal and
limitations of rhe work of rhe TRC and, by implieation, that of rhe discourse
moral autonomy, as well as efforts ro establish and maintain its polirical im-
of nonracialism, are overcome. partialiry, yet there may well be a grain of truth in rhe accusation. However,
Chis is nos to be found where cribes of rhe TRC usually locate it: in an inability

Reinventing rhe Myth of Nationhood to take a critica) distante from rhe ANC, leading to an unquestioned accep-
tance of rheir "version" of rhe past" Rather, it is to be found in rhe conception
As Hansen and Stepputat point out in their introd uction to this volume, w
of nationhood emerging froni rhe pronouncements and publications of rhe
myth of rhe state may be considered a form of "social fanrasy" cireulating
TRC. To substantiare Chis claim, ir is necessary briefiy to highlight some of
among citizens and communiities. Conceptions of nationhood form part and

NATIONAL IDENTITY AND MEMORY 183

IX`
( L_Y`
 rhe maro contours of rhe conception of nonracialism that has informed ANC
shape the character of new regimes. It is on these issues that 1 wish to focos
political discourse historically.', In contrast to the identitarian conception of in this essay.
nationhood structuring the apartheid imaginary, nonracialism aims at open-
ing a space for difference of a linguistic, cultural, and religious nature with-
out, however, conceding ground en the basis of race. In so doing, it acknowl- Why a Truth Commission?
edges rhe historical legacy of apartheid but seeks tu overcome it through an
It is now two years since the TRC brought out its final report." Its tasé, in
emphasis en political equaliry regardless of race, creed, and religion, stress-
the words of Vice-Chairperson of rhe Commission Alex Boraine, was to con-
ing the need for the Tour "historie groups"-Africans, whites, coloureds, and
tribute to che healing process in South Africa through "an honest assessnlent
Indians-to build a common sociery.'°
and diagnosis of rhe sickness within our sociery in an attempt to give people,
To rhe extent that the Tac participates in the building and reconstruction both perpetrators and vierims, an opportuniry to face the past and its con-
of a "common sociery" ir is necessary to place its activities in the context of sequences and to start afresh. The Truth and Reconciliation Commission
a posttransitional democratic government facing social demands for an offi- is an opportuniry tu make a contribution to deal finally with the past without
cial recognition of rhe truth about human rights violations committed by the dwelling in it and to help ro creare the conditions for a truly new South
previous regime, and for dealing with those guilty of ordering and commit- Africa." "
ting [hose violations." The las[ two decades of this century have been marked
In atrempring to address questions central te rhe process of transition,
by a growing interest in these issues and how they affect transitions. Iris not namely, how emerging democracies come ro ternas with past violations of
possible to give a detailed account of the different contexts in which these human rights, how new democratic governments deal wirh leaders and indi-
issues acose over the past ten years. Suffice it to say that historically such viduals who were responsible for disappearances, death squads, and psycho-
processes have taken many different forros, ranging from special national logical and physical torture, and how it deals wirh the fact that some of its
tribunals (Argentina), te international tribunals (Nuremberg and Hague tri- perpetrarors eontinue to play importan[ roles in public life, South Africa has
bunals), to individuals (Chile, Paraguay, and Uruguay) and nongovernmental decided that rhe way forward is te be through rhe work of this commission,
groups (Honduras, Uruguay, Paraguay) taking their cases to national or inter- rather [han through Nuremberg-style trials." After extensive discussion, rhe
national courrs."
idea of such trials was rejected en several grounds. The first was a concern
There are, equally, many ways in which [hese processes may be approached with rhe difficulry of proving guilt in rhe context of a criminal legal jusrice
analytically. One could, for instance, focus en rhe political eonditions that system. As rhe Nuremberg trials showed, in cases of political trials where
permitted or inhibited the realization of practices of truth telling under suc- large numbers of people acted as members of political organizations, it is
cessor regimes. Such a focos would demand a detailed examination of the very difficult to determine individual legal responsibiliry. A second reason was
repressive, constitutional, and political legacies of specific transitional con- rhe difficulty in gaining evidente of such acts. II was generally agreed that
rexts, as well as of how democratizing political eonditions shape rhe ability there was very little likelihood of new evidence coming to light or of wir-
of governments to deal wirh issues of truth and justice." Alternatively, or in nesses being prepared to testify in rhe context of criminal trials. Most impor-
addition te Chis, one may focus en the manner in which collective identities, tan[, however, is the consideration that rhe granting of amnesry was a central
national self-images, and political cultures are negotiated, reworked, and re- par[ of rhe very process that made rhe negotiated rransition possible. As Des-
constructed in the course of investigations roto past abuses of human rights. mond Tutu pointed out, "Many of [hose now calling for justice through crimi-
To paraphrase Perelli, rhe resignification of the past does not serve only to nal trials supported rhe negotiated settlement at Kempron Park, and seem to
explain rhe presea[; it is, indeed, a struggle for control over the future." Thus, forget that amnesry was a crucial ingredient of rhe compromise which re-
reconstrucring collective memory and instituring new foundational myths do versed rho country's inevitable deseen[ roto a bloodbath."`
more than "deal with the past"; they act as legitimizing moments for and Instead, it was agreed rhar rhe Tac should proeeed through attempting te

184 STATES OF IMAGINATION NATIONAL IDENTITY AND MEMORY 18s

unswer, on rhe une hand, the dcmands of those who lost family arad friends membcred (or tbrgotten), by whom, and for what end." ' Indeed, arthis nmo-
to know rhe truth surrou nding those circumstalees and, on dic other, to pro- ment, whcn it is apparent rhar borla identin, and nteutory are politicat and
vide .m nppornmiq• for pcrpen atuts of abuses lo malle full diselosures of social constructs, and whcn vve can no longer assign to eidhet the status oí
rimcir crimes in return llar mnnesti'. It is onle in and through full diselosure a "natural object cve "must ralee responsíbility mor rhcir uses and abuses,
(truca telling) rhat j u sti ce as ac1< novel edgmen t can be att a roed; Chis, in tu rn. tccogn izi ng rha[ every asscrtion of identitY hnvoives a choice thar affects not
operas up rhe possibility oí reconciliation. 1 atgue rhat public recognition jusr ourselves but othen.
and achnowledgnm ent of injusrices taus constitnr, the hasis for nce attainm el Ir 1 listory avails us a ntultiplicny of examples oí ncu gravite' of chut responsi-
of jusfice." l hat Ís why dre role of publie ucl<nowl edgtucnt oí memories or bilin-lacre, une only necds ro think of rime histor} miel historiral
the pasr in nce reconstruction of rhe present and rhe future is absolutely cru- en the Holocaust-and of widely diverging wavs in which memore and iden-
cial to the whole proeess. With the creation of this commission, South Africa, tiry interact. In situations of"new beginnings, " identities are consrructed and
in the words of a commentator, "has decided to say no to amnesia and yes ti held together as much by forgetting as by remembering. Such new begin-
remembrance; m say no to full-scale prosecutions and yes to forgiveness."22 nings sometimes require the cradication of the past and engagement in what
This sentiment veas already clearly expressed in the new Constitution. Its last Benedict Anderson has callad "colleetive amnesia."" This is most evident
section, dealing with "National Uniry and Reconciliation," includes the fol- in the eonstruction of new states in postrevoiutionary situations: revolution
lowing statement: "The adoption of this Constitution lays the secure foun- has tu inaugurate an absoluta beginning that necessitates the introduction of
dation for rhe people of South Africa to transcend the divisions and strife of "new time" and a radical break with the past. Construeting a new Japan and
the past, which generated gross violations of human rights, the transgression two new Germanys after the Second World War also involved forgettingrather
of humanitarian principies in violent conflicts and a legaey of hatred, fear, than remembering." It is, moreover, of interest to note that even in rhe case
guilt and revenge. These can now be addressed on the basis thar there is a of the eonstruction of the new state of Israel, the first few years focused more
need for understanding but not for vengeance, a need fór reparation but not on the present than the past. The concept of the Holocaust only carne into
for retaliation." existente in the t95os, after the new state was flrmly established and Jews
But, one may ask, what precisely is the tole of memory in this process, and could reflect on the "pasmess of the European past." Ir was only when the
how is it to be negotiated so that it avoids two excesses-that of roo much memory of those terrible events could no longer be taken for granted that
dwelling en the past and of roo little disclosure-bota of which will make there was a powerful reason to commemorate." Even the need ro commemo-
reconciliation well nigh impossible? Indeed, one may want to reflect some- rate then has a history. Gillis argues that commemoration historically has
what further on rhe very possibility and nature of the °reconciliation" that is taken roughly three forms: prenational (before the late eighteenth century),
to be achieved, en its precise relation to remembrance, and on rhe relation national (from the American and French Kevolutions to the r96os), and post-
between memory and identiry in general. nationai forms. The early, prenational history of memory shows a sharp di-
vergence between popular and elite memory: whereas the elite classes (die
aristoeracy, the church, and the monarchical state) had a need for institution-
Memory and Identity alized memory, ordinary people relied on living memory. National memories,
Let me start with the latter. The notion of identiry depends en the idea of in eontrast, tend tu focus en the eonstruction of uniry and condnuiry so as to
memory, and vice versa. Any individual or group identiry, that is, a sense of cover over the fragility of new nations.3°
sameness ovar time and space, is sustained by remembering, and what is As our concern is with modern, national memory, it is necessary lo reflect
remembered is defined by rhe assumed identiry.' Memories are constandy further en sonie of its niost central charaeteristics. National memory is aboye
revised to suit our turren[ identiry, and this memory worl< is aiways embedded all archivab it relies on rhe immediacy of rhe recording and en tire visibiliry
in "complex class, gender and power relations that determine what is re- of rhe image." Ir is also, howevcr, acutely aware of the efforts of caca group

186 STATES OF IMAGINATION NATIONAL IDENTI'TY AND MEMORY 18]

 to make its version the basis of national identity. It is thus aware of conflicting insist en "apologies for everything that has occurred in the past," and to "let
accounts of the past. And it is out of this awareness that a different relation
bygones be bygones." Indeed, it was argued that one should not "dwell on
to the past emerges in what Gillis calls the postnational era, a relation that the real or imagined injustices of the past," butwork toward building a future
bears a strong resemblance to the struggles at the time of the Reformation "without mistrust, prejudice, hate and domination."'°
between older Catholic practices of locating the sacred only in certain times
These themes are echoed throughout the National Parry submission to the
and places, and rhe antiritualistic, iconoclastic Protestants, who demanded
TRC, which exhibits an exemplary form of "nationalist" history, with al] its
that the sacred be brought finto everyday life itself. The new iconoclasm at-
monuments and archives, its exclusions and denials." National identiry, in
tempts to desacralize the nation-state, to democratíze memory, and to retrace a Chis reading, is natural, given, and puye, constituted with reference only to
multiplícity of pasts better suited to the complexities of a postnational era."
rhe characteristics of the "nation" itself. In denying the fundamentally rela-
This account provides us with valuable insighrs into questioning the rela- tional nature of identiry, this account attempts to efface rhe difference at the
tion between memore and identiry as it is being played out in South Africa
heart of every identity, and in so doing affirms an essentialist, identitarian,
today, in rhe movement toward a postapartheid sociery. In South Africa, rhe homogeneous, and nonpluralistic coneeption of nationhood.'° Young's de-
struggle over rhe meaning of the past is by no means over, its character by no scription of nationalist monuments perfectly encapsulares the account of Af-
means setrled. Although there is a general agreement on the ovil that apart- rikaner history presented there. The submission plots "the story of ennobling
heid represented, that agreement, some argue, has been reached without a events, of triumphs over barbarism, and recalls the marryrdom of [hose who
thoroughgoing engagement with che past." One only has to think here of the gave their lives in the struggle for national existente."'°
refusal to engage with rhe work of the TRC evident in the submission of rhe
Much of the submission, then, is taken up by setting out the historical
chief of the South African National Defence Force (SANDP), General G. L. context within which "the conflicts of the pass" and "unconventional actions
Meiring, to rhe commission. Instead of focusing en rhe period under inves- and reactions" should be considered, according te its authors. The emphasis
tigation, rhe submission dealt with the transition period only. The few scant en the plural case here is noteworthy: rhe conflicts involved many forces,
referentes to the past, moreover, were couched in terms of "acts perpetrated and rhe commission is constantly reminded that "no single side in the con-
in rhe context of a war situation." Indeed, this was a eharaeteristic that flict has a monopoly of virtue or should bear responsibility for ali the abuses
marked rhe submissions ofall the main parties. It is ro these submissions, to that occurred."°" Indeed, in judging abuses, de Klerk argues thar distinctions
the wider discourses informing them, and te the manner in which they relate should be drawn among [hose carrying out orders, those carrying out orders
ro and commemorate rhe past that I mrn now.'°
"overzealously," and [hose committing nlalpractices and serious violations of
human rights. But, aboye all, rhe context of [hese actions-working within a
Identitarian Constructions of the Past state of emergency-had to be considered so as "m explain the historical
context in which they occurred."°' One is struck by rhe contrast between chis
"Wc are all rhe children of out times and rhe product of the cultural and narrative and accounts of rhe Holocaust that, time and again, con up against
politieal circumstanees finto which we were born and with which we grew rhe sheet inexplicability of its ovil. While, in rhe National Party's case, there
up," wrote F. W De K1erk.'°
is an acknowledgment of rhe suffering of all [hose involved in the conflict,
We know that during the period-effansition in South Africa, many differ- the whole submission serves to provide a context that could explain, and so
ent articulations have been given to rhe memory of rhe past. These range from jusrify, rhe actions perpetrared.4' Indeed, Chis is a feature of all the submis-
National Parry appeals to put rhe past behind us, to nostalgic demands for a sions by political parties, organizations, and state institutions.
certain resurrection of apartheid, be that in the forro of an AfrikanerVolkstaat Now, in stark contrast to rhe certainties and olear-cut justifications that
or an independent KwaZulu. National Parry discourse, for instante, continu- have informed rhe submission of the National Party and of rhe other main
ously admonished the people not to be obsessed with past grievances, not to political organizations and official institutions, the whole of rhe transition

188 STATES OF IMAGINATION

NATIONAL IDENTITY AND MEMORY 189
 petrarors of ahuses (the TRC's investigative unir sought to check [hese as care-
process irself was marked indelibly by a series of ambiguities. Because Che
fully as possible bcforehand; ir warned persons ro be named and invired rhem
end oí apartheid carne ahora flor as a resulr of a revol utona ry break oí a
ro respond ro allegations) and sought ro bring to public artenrion borh well-
complete discontintlity [bar divided tire pasr fronl tle present and furo re. it
known evems and eecryday injustices perperrated agatust persons htrherto
had tle characrer of ara impute rrausitiou-'Chis raises precise problenls for
unnamed. la chis fi rus ora tiro cvcrydayness of injustice, and nce reoceupa-
rito negot iariou of memory. trurh- and [he ins[imtion ofa just and doro ocnrtie
[ion of memory sitos by ordinary citizens, Che real significance of Che public
society. It both imposes restricrions and opens possibiliries for [hose pro-
hearings and Che scarch far justice becomes visible. As Une CO tntnentator ora
ceses of ncgoti ario s. In Che tiirsr nsrance, as de Brito notes, becrose repres-
a similar process elsewhere has par ic It does uot bring ncu dead back ro lile.
sors are flor defeated, but even giren a degree of political legirimacy by rheir
bur it brings rhem out from silente; for rheir families, it mearas Che end to au
"voluntary withdrawal" from power, successor regimes musravoid a backlash
agonizing, endless search.47
rhat may endanger Che stabiliry of Che transition.°' Second, negoriated demo-
The end of those searches, oí course, depends on Che knowledge obrained
cratic transitions ser frameworks within which truth and justice have lo be
during the whole process. As Che daughter of an aerivist argued: "I want to
pursued: democratic pluralism ensures that voices of victims are heard at Che
forgive, bur 1 don'[ know whom lo forgive." Steve Biko's mother, shorrly be-
same time as it ensures a continued voice for violators. Third, under these
fore her death las[ year, said much Che same: "Yes, 1 would forgive my son's
circumstances, there usually are state instimtions Cha[ survive; institutional-
killers. 1 am a Christian, and we Christians do forgive. But first 1 must know
ized crimes have ro be dealt with, and there must simultaneously be a coneern
what ro forgive, which means 1 musr be told fully what happened and why."'"
for not entirely destroying state instimtions. Fourth, mechanisms and insti-
The hearings consequenty have offered an opportuniry not only for survivors,
tutions dealing with abuses must be seen lo comply wirh judicial due process
bur also for perpetrators of violente te come fonvard and give full starements
so as ro strengthcn the insrimtionalization of democratic procedures, even if
of rheir participation in events. In this manner, reeonciliation was sought
that mearas rhar some of [hose who are guilry of abuses may not be brought
between Che parties parriciparing in and affecred by Che events.`9 Nor only
to book asa resulr of a lack of evidence.'° In all of these cases, Che new
high-ranking officers were brought to book, bur low-ranking police officers
regimes are engaged in ambiguous procesases of negotiation that are a far ery
and ordinary citizens were given tire chance m partake in whar 1 would argue
from demands for olear-cut ethieal stances and decisions. Itwould, therefore,
is a public memorial exercise that differs from Che "standard" nationalist uses
be strange if [hese ambiguities did not also enter into Che processes in and
of memory and monuments.`°
through which past memories and new beginnings are negotiared.
Thus, Che TRC steered South Africa away from a culture of violente. But it
also had several effecrs that arguably no other forro of engagement with Che
Toward a Different Remembrance past could have had. It has subverted Che ability of national leaders of all
persuasions to grasp and represent history in their own image; it has under-
ir is in Chis context that Che memory work of Che TRC has to be understood.
minéd borh Che possibilities, mentioned earlier, rhat may malee reconciliarion
The importan[ investigative work of rhis commission has brought ro light Che
impossible: and ir helped to avoid roo much pasrness as well as a covering
existente of state structures for identifying targets for elimination, firsthand
over of Che past. In contras[ to Che usual construcrions of memories of a
claims of ministerial-level approval of bombings and killings-all [hose ac-
nation's past, this exercise has brought m ti lit no singular past and has com-
tivirics tira[ Che Narional Party submission euphemistically called "unconven-
met xz fied narionhood The inabiliry to construcc sueh a sin-
tional actions"-as well as evenes known, until now, only ro Che communities
gular narrarive is admirted to in Che Final Report, which srates tira[ in Che pro-
immediately affected by rhem. Indeed, 1 would argue tira[ one of Che most
motion of national uniry and reconciliation, reeonciliation cannot be imposed
importan[ effects of che memory work of the'rac is Che way it has offered an
on a diverso sociery atrempting ro consolida te a fragile democracy. A healthy
occasion for survivors to gain recognition of rheir plight in full public view."'
democracy requires only respect for eommon human digniry and shared citi-
Iu [hese public hearings, submissions were allowed lo contain names of per-

NATIONAL 1DEN1I'1'Y AND MENIORY 191

190 STATES OF IMAGI NATION
 zenship, as well as tnechanisms for the peaceful handling of unavoidable remembering apartheid as a logic of closure, the work of the TRC may open
conflicts. It does not require everyone to agree
the space for difference that will not immediately be subsumed and trans-
The memory work, therefore, performs a multitude of complicated fune- formed into a logic of othering." This remembrance may be able to encirde,
tions. Through it, all discourses of nationhood and thc idea of sacrificing to mark the space of difference as such constitutive of any already constituted
one's life for national existence becorne problematized, if only as a result of identity.
the absence of a singular narrative. Through it, the past is recalled so that it
becomes possible to )cave the past behind. This memory work is to be sup-
plemented with a public archive where all the materials obtained by the cotn- l eyond the TRC?
mission will be lodged. Chis contains the seeds f , relution te past a..a to "The South African Truth Commission is only one of the structures through
memory th t m a , lead South Africa toa postnational conception of identity, a
which we should hope to dismantle the old regime of truth to replace it with
conception of identity that takes its chaca
new and multiple narratives. We must remain aware of the dangers of re-
thaf w ich was exemplary in the identitarian conception of identity that in- placing apartheid's false utopian historicism with out own new orthodoxies.
formed apartheid
As we construct new historical narratives, it will be in the curreney of tetero-
tapias, multiple idealisms, rather than the single-mindedness of utopia....
On the Possibility of a Postnational Identity Out state-sponsored Commission has no monopoly en processes of histori-
cal rectification.""
The possibiliry of a postnational (postapartheid) identiry depends en the Like all narrative structures, the plurality of discourses that, woven to-
abiliry to go beyond apartheid insofar as apartheid functions as a signifier of gether, forro the discursive strucmre of the TRC display certain blind spots.
closure. If apartheid signifies the denial of difference at the heart of identiry, These blind spots have acted as conditions of both possibiliry and impos-
a remembrance of apartheid would consist in a remembrance of (the effects sibility: they mark the exclusions en which the TRC was set up asan exercise,
of) closure as such. A pluralistie, postapartheid social order will, conse- and they ultimately put into question its parameters. A number of contem-
quently, be a forro of social division in which the constitutive nature of differ- porary writings have set out to make visible such blind spots in the discourses
ence is thought. It is this constituting function of difference, the holding- of and around the TRC. As a consequence, they upen up arcas ofinvestigation
against-an-other, that 1 would argüe becomes visible in the memory work of beyond the remit, and perhaps beyond the conceptual structure of the TRC.
the TRC. This can further be clarified by analogy to the structure of memory. These blind spots may cake a variery of forms, ranging from institutional
Remembrance (teehnically, primary remembrance or retention) serves as die tnechanisms and procedures in the actual operation of the TRC to exclusions
not-now that is constitutive of the possibility of the presence of the Now " built into its very remit. To begin with, although much reseacch still needs to
Remembrance rhus in essence points to the incompleteness of the present. If be done, a number ofwriters have begun to question the institutional mecha-
a postnational orden is to be characterized by such a remembrance, it is strue- nisms and procedures deployed by the TRC. I focos on only two sueh arcas
turally determined as incomplete. However, this remembering is not to be of here, as they impact more directly than other arcas en the main concerns
just anything. It is to be a remembering of the logie ofclosure. Thus, we have under discussion: that of the relation between che TRC and NGOS, and that
here a double signification: a remembering as such, which already reminds of the interna) procedures deployed to make decisions within substructures
os of the incompleteness of out present; and a remembrance of something, of the TRC concerning, for instance, the allocation of "victim status" to ap-
of a discourse of closure. These two moments reciprocally reinforce one an- plicants., In their study of the relation between the TRC and NGOS, Merwe,
other and serve to show that which cannot be made present. Its marking, Dewhirst, and Hamber point out that the failure of NGOS to "effectively
paradoxically, can consist only in lceeping open the space, the interval be- mobilize around the principies and strategic concerns" raised by the TRc
tween the present and the past, the now and that which preceded the now. In could be ascribed, in part, to the structure and process of the TRC 5 They list

192 STATES OF IMAGINATION NATIONAL IDENTITY AND MEMORY r93

 neeessary biological, historical, and cultural background to legirimately spcak
fnur factors that wcre of sigui1 icance : tire 1 nc s lramework ot opcration, its
abourbiack cxpcrience' undcr apartheid" will be dealt wirh. ibis is notan
political agenda, political tension within rh organization that affected the
¡die question, but ore [bar goes to tire hcari of tito new nacional imaginar}
relarionship henveen thc Trr aud rhe Neus, and its interna) managentenr
under construction iu amtemporary South Allica. lndced, there ls a bias elan
structures . For our purposes, thc tirst tliree are of particular interest. lis more narrowh, defined than "biack expcriencc' that iniorms chis imaginar'
trantework of opcration carne in tor criticisin tronl NGOS operaring outside
in dhc making. One only has lo look at rhe construcrion of dio histor} of
tire human rights sector, arguíng thar the ene was based on an overly legal-
resistance in die writings of come oE the prominenr government cominero
istie te utework and tirar ir emphasized iuvestigution and rights -based mecha-
tors on rhe TRC. Astral. Asnal, uud Roberts's account of resistance ro aparr-
nisms and procedures at rhe expense of grassroots confiiet resolurion and
heid, ir tire context of their discussion of tire work of the Lrunh Comntis-
social support work. As a result , the TRC failed to engage wirh the local com-
sien, crucially limits irself to an account of ANC resistance. Ignoring non onle
plexities of particular communities where ir held hearings; it underestimated
the role of political organizations such as rhe Pan Africanist Congress and rhe
confliets arising from economic injustices ; and its emphasis on gross viola-
Black Consciousness movement, ir also negates rhe crucial role of rhe United
tions of human rights rather than on the more common violations that "made
Demoeratic Front (u DF) in rhe final phase of interna) resistance against
up rhe day- to-day experiences of most black people" was seco as potentially
apartheid. This is no accident, reflecting as it does a very particular history of
marginalizing . The organizations also felt that the TRC'S work did not fit in
resistance froto the perspective of the ANC in exile. Coupled with the margin-
wirh the nationa l political agendas toward which the work of die TRC was
alization of UDF and other internal resistance leaders who led democratic
geared. This was also at rhe root of the internal tensions within the TRC,
opposition inside rhe country during the crucial years of the rg8os and early
arising as a result of tire need no represent different political persuasions in
tggos since rhe return of che exiled leadership, this marginalization appears
its composition . Although the TRC, as argued earlier, made great efforts to
all the more sinister. Add to that rhe carlier criricisms of ANC government
establish its impartiality, it was reluctant to engage roo closely with NGOS for
policies of "affirmative action" by minority groups such as coloured South
fear of being aceused of having roo Glose allianees with political groupings.
Africans,'I and tire practical embodiment of rhe project of nonracialissm be-
The NGOS also raised ronceros about the simplistic nature of the categori-
gins to look all the more flawed. Indeed, onc has to question whether what
zations into which the TRC divided applicants . The victimlperpetrator dichor-
appears at first to be limitations of implementation are not indicators of
omy did not necessarily refiect tire experience of the person in the street, just
deeper underlying problems with the vision of nonracialism."
as it probably oversimplified rhe categorization of major political actors. A
Nevertheless, ir could be argued that the fact that the ANC's version of
disquiet wirh Chis categorization and its construction and implementation is
nonracialism is flawed does not necessarily contaminare rhat propounded by
also expressed by Lars Buur , based en fieldwork conducted in the Data Pro-
rhe TRC. The question to be addressed here is whether the practices and view-
cessing Unit. Analyzing and problematizing rhe detailed discussions that lead
points of che TRG are marked by traces of the closure of identitarian poli-
lo persons being attributed "victim" status , Buur draws attention to the con-
tics-similar to that of rhe official ideology of nonracialism-even while
tingent decisions , all of which would contribute to the reconstruction of a
they are straining to institute a certain pluralism. Indeed, ir would be surpris-
"new," officially sanctioned history of the pass.` " The TRC has also been ex-
ing if they were not. These traces could be found, most clearly, in rhe TRO's
posed to a series of more wide-ranging criticisms . It is to these that 1 turn in
reptesentation of rhe old regime and the entity called the "apartheid state."
conclusion.
It is commonplace in che literature on apartheid to question rhe portrayal of
Addressing the very nature of rhe TRC remir, Steve Robins argues that tire
both rhe project of apartheid and rhe character of rhe state in monolithic
TRC has privileged a "modern temporal frame" by taking rhe whole of irs
tercos." Yet, the TRC consistently treats ' the snate" as a homogeneous entity .
investigation to be delimited by rhe beginning and end of tire apartheid era."
In the words of rhe Final Report, "The South African state in tire period froto
One of rhe consequences of such a framing is [has it raises the question of
the late 1970s tu fthcl early tg9es became involved in aetivities of a criminal
how rhe histories and collective memories of those "deemed not to have tire

NATIONAL IDENI'ITY AND MEMORY 195

194 STATES OF IMAGINATION
 namre when, amongst other things, it knowingly planned, undertook, con- its starting point.°° Moreover, this return to Africa, to Krotod as mother of die
doned and covered up the commission of unlawful acts.1` To question the nation, also necessitates amnesia about how and why she was forgotten. In
portrayal of the stake as homogeneous entity is uot to problematize the main contrast m che writings en Krotod during the apartheid era, contemporary
findings of the TRC. It is, however, te draw attention to the problems that appropriations offer Afrikaans-speaking South Africans a way te identity with
such a portrayal may hold for the future. The inabiliry to recognize that dif-
Africa and with a hybrid, instead of a "pure race," identity. However, Coetzee
ferent stake sectors pursued sometimes markedly different strategies, some warns against a roo easy appropriation that risks "forgetting che conflict and
arguing from greater repression and others proposing greater liberalization, destruction" at the roots of this hybrid identity. Most problematic of this ap-
is to ignore che differences in reforms that may be necessary ro construct a propriation is "the sense of completeness" of closure that it brings.
truly posrapartheid state.
The "Mistas[" exhibition is an example of the impossibiliry of maintaining
These limitations, however, deal only with exclusions operating "within the such final closure. "Mistase" functioned, not unlike che Truth Commission,
present," so to speak4 It is now necessary to look more closely at the system- as confessional space, but it had che added advantage of forcing whites "to
atic exclusion of colonia] violence in che TRC's reconstruction of the pass.- confront European colonial violence and genocide."'° Ir rendered the previ-
The character and complexiry of colonial violence and the manner in which it ously forgotten history of colonialism visible, and in so doing called forth
impacts en contemporary South Africa are made visible in a number of con- widely divergen[ responses from different communities. This was the case
temporary publications. Several new signifiers have emerged as paradigmatic especially for the design of one of the galleries of che exhibition, constructed
of what is at stake here. They include che figures of Saart)'e Baartman and so that viewers could not avoid walking en images of the aboriginal KhoiSan
Krotod, as well as the "Mistas[" exhibition.°° While each signifies its own people. Angry KhoiSan responses challenged the use of these images and of
unique singulariry, together they act as signs of a new vision of South African
fragmented body parís as a reenaetment, not a critique, of colonial violenee.
identity, one that aims ro put finto quesion any homogeneous narrative of While the "Miseast" exhibition forced whites te confront colonial violence
nationhood. The struggle for che meaning of Robben Island is a case in point. and challenged che exclusionary emphasis on contemporary history in che
Harriet Deacon traces che shifts in the symbolization of the island from a
TRC, it was also used by KhoiSan activists te construct "totalizing ethnic-
repository of "al] that was considered negative in sociery" to acdng as "a nationalist narratives that draw upon collective memories of suffering.'-
focus for renembering apartheid and as a spearhead for nacional revival."°' KhoiSan descendants used che exhibition to advance their turren[ claims to
Emphasizing the importante of this new symbol in che construction of a new land." In so doing, they invoked discourses of cultural continuiry and purity
national identity around che observante of human rights, Deacon also draws to precolonial KhoiSan ancestors, such as Saartje Baartman and Krotod.
attention to the negative possibilities of presenting too unitary an account of
its significance. She argues that, if it is to be a living monument for the new
South Africa, ir should also permit diversity "in its opening outwards ro other Conclusion
accounrs of our past and our future."°a
The meaning and significance of figures such as Krotod, and che divergent if
In similar vein, Carli Coetzee provides a nuanced discussion of the ways in not clashing appropriations to which they Nave been subjeet, attest to the
which Krotod is remembered today. Through an exploration of the figure of complexiry of che processes of che reclaiming of histories. It also throws finto
Krotod (a Khoikhoi woman in Cape Town in the seventeenth century) she sharp refief che fact that contestation of che narratives organizing and fram-
addresses both che conditions and implicarions of the fact that Krotoii, once ing che TRC does not automatically carry any "subversive" or democratizing
written out of history, is now reclaimed as onse me, rhe mother of Afrikaners connotations. Onitary discourses may be challenged only to be reinvoked ata
seeking to reestablish connecdons with Africa. Coetzee discurses the dangers more particularized level. The apparent opposition between universalism and
and advances in such a "remrn" te Krotod. Taking her as che founding mother partieularism, between an emphasis en, for instance, "che nation" and "dif-
of the Afrikaner people is te reinforce borh the urban, Cape Town-based ferent (national) groups," may thus be misleading. Neither extreme univer-
version of South A&ica's early history, and ro perpetuate che myth of 1652 as salist nor deeply particularist views of identity, of necessity, will escape the

rg6 STATES OF IMAGINATION

NATIONAL IDENTITY AND MEMORY 197
identitary logia oí apartheid. lo do so, ara enaphxsis ora the noncompletion ''bl:mket' amnesq'-the'nu: in rhe
'virar ti,, g o ,mrnmc n tt ala[) to n
and interpenerradon of both must be mahttained. This potentjally could be
y. I ir a discussion or nona, '11,,a, cariocs ronntihutiom m 1
achieveel by pluralizjng diseoti rses stich as that oí thu TRC. 1 lowever, as 1 ha ve
111''ntn3 , rhr i itrio: ( i:'r't,15S 55(tg8p-
argued, Chis poreutial cvill be realized onlc ora conditiou thar more worle is 1'nr d 11-siun of dtc'nacional tl miran" lis iha a>n:a
done lo avoid rhe closures tirata more honutgenizíng versfun of nonraciallsm lordan Irgyhi -

nmy bring about. u. 'Fhi. t.dscs the question or tvhcther puuishment t tlss:avo ne'1 11- 1 y y and ptrLL e ro uthe:
tur:n- nt dcaling wth perpemuors ,t i nlustiee,.l cito .... .,lar rhe .tbzrn huui:,h-
:nent:a:rdmi..ihle onlc whvn 1111, has ',en ara,) l.td^nrnvl :erra: o(tntl d :er.
a nauou:d consen>us ^xisr. ror non) asarnrior.'te9 9 :ol.
Notes
ta. Ibid.: 3.
1 lvou Id like m than k l' somas 61om 1 lansen and ¡!in ir Stepputat for dicir hclpful osmio enrs 13. This is die focos of De Brito's importan[ comparadve smda of [hese processes in Urugaa)

ora ara earlier version of chis essay. Sople uf die arguments were flrsr developed and published and Chile. A similar focos informs Kaye's study of tmth co mmiss lo ras in El Salvador and

elsewh ere 1 thank Blackwell'r for permi s sion tu draw en a previous ly pu blis hcd article (Nor- Honduras; see Kaye (1997: 693-716).

val rg98a). 14. Perelli (1993:154, quoted in De Briro 1997: 10) .

1. The Promotion of Narional Uniry and Reconciliation Bill of 1995 defines gross violations el i s. The commission consisted oí seventeen full-time eommissioners and had rhree separate

human rights as ihe killing, abducdon, torrare, or severe ill rrcatinent of any person be committees: a Human Righrs Violadoras Committee, which conducted public hearings for

someone aedng with a politcal obj eetive. Ir ineludes rhe pla uning of such aers and attem pts vi cti msls urvivors; a Reparadon and Rehabil itation Committee, which worked un policies and

to conrmit rhem. rccommendations arising from [hose hearings; and ara Amnesro Committee, which heard
a. The TRC was constituted by the Promotion of Nadonal Uniry and Reconciliation Bill, 1995, applications for amnesry. Che final reporr of dte Tse veas published in diftérent forros. 1

which combines rhe req u'remen ls of the interim Con stimtion veith those of human rights draw ora nhe co version of 1111 Tac wehsite (NOVember 1998).

nortes. The signiflcanee oí this is alear when it is eompared tu alrernative processes through 16. Sce Boraine (1996a). The TRC's view of their mandate is discussed in full in volume i, chap-

which trutli eommissione histotically havc been consdmted. In orosr iostanees, commis- ter 4 of rhe report. See'rac wcbsite, flle:pl D1Ihnallmhap4.htm.
sions of this sor[ are appointcd by a presiden[ ir prime minister of the country concerned, 17. Omar (1995: 2-8); Tutu (19g6a: 38-43).
and [hey have to work out their own procedures, objectives, and methodologies. The bene- r8. Tutu (r996b).
fit of a commission appointed by au Act ofl'arliament is tirar a democradcally elected group 19. There are no general or "blanket" amnesry provisions in South Africa. Amnesry has tu be

of people participated in the debate and the finalizing of rhe objectives of rhe commission applied for en ara individual basis. Applicants nmst complete a prescribed form, detailing

(Boraine [9966). For a detailed discussion oí the background un :be idea of rhe TRC in informadon perraining ro speciflc human rights violations; such disclosure should be full

South Africa, see Walt and Walr (1996: 1-211, as well as che infortnation pack published by and complete. A public hearing follows, where offenses fall roto the caregory of "gross vio-
lations of human rights." If not, amnesry, decisions may be taken in chanlbers. Several cri-
tire TAC ( n.d.).
3. Por an elabomdon of the idea of a postapartheid identiry, seo Norval (1990: 5S5-57: 1996: teria bate to be fulfilled for anmesty no be granted. These inelude tlie fact that a pardcular
act must be shown m have taken place as parí of a wider political event or lit tire service of a
x75-305).
4 1 draw liere un rhe disdncrion as developed by Laclau (rayo: 6o-68). political organizar ion, Actions for personal gain and based en ill will are excluded. Ir is also

5. For en in-depth discussion of this period of South African politics as well as of tire various importan[ m note that Use law does not require that applicants should express remorse. T hey

diniensions ofrhe crisis, see Norval (1996: chapa. 5, 6). can come tu rhe commission saying, for instance, "that rhey fought a noble srruggle for

6. Fui- ir more detailed discussion of these discourses, see Narval (1996: chap. 6; 1gg8b: 93- liberation, butthat because [hey opened themselves to prosecubon ir civil actions as a result,

rro). rhey are asking for amnesry" (¡bid.).

7. 'This criticism was voiced in many quarrers, not al[ of wlrich 'vete "conservative, In n ieir zo. "Truth" and "full disclosure" should clearly non be understood in metaphysical terms. The
study uf the reladorr benvecn rhe Tac and NGOS, Menve, Dewhirst, and Hamber point out commission can never attain "The Tmrh" lnstead, Krog suggeste tltat une should under'
that sume NCOS were reluctant te worlc tiro closely with rhe Toa, because rhe Tac was seca stand rhe idea of truth here "ras rhe widest possible compiladon oí people's perceptions,
as engaged in a national political agenda, of which rhe Inkatha Freedom Parry (1FP) and Pan stodes, monis and experiences' G995: 118).
Afriean Conference (vac) in particular were suspicious (1999: 68-69). ar. The TRC's power of subpocna aids it in working against the possibiliry of a "eonspiracy of
8. Although rhere is sume evidence drat rhe Tse has experienced dif culdes distancing itself silente."
from the ANC-parricularly visible in its aeannenr of rhe nrnv Infanroos A NC appl icatioir for 22. See Boraine (1996).

t98 STATES OF IMAGINATION NATIONAL IDENTITY AND MEMORY 199

 23. See Gillis (1994: 4).
41. The question of the historical context in which past abuses occurred raises important and
24. Ibid. 3. por a discussion ofthe need to introduce a gendered dimension finto rhe work ofthe
difficult practica¡ , echical, and philosophical issues. It is not possible to treat [hese issues
Tac, see Goldblatt and Meinrjes ( 1006).
here. However , it should be noted that alchough sucia contextualizations may be used to
25. This is now an accepted tener of ntost social and political thcory infarmed by post-
attempt to explain past abuses , they have not generally bcen invoked m shirk responsibility
serucmrallsm. It is important no note, however , chal rhe claim that identity is socia uy pro-
for such ahuses . Like other organizations , rhe National Parry has gone to grcat lengths to
duced does not also translate finto a claint that it is, consequenely,
easy te chango or challenge sketch rhe context and general aiens of apartheid policies. It did, however, recognize that
socia identities. Processes of identity formation are t, ically subjecr
to deep sedimentation, apartheid "led re hardship , suffering and humiliarion-to institutionalized discrimination
and identities could preve eztremely recalcitran[.
on the basis of rice and erhnicity," and [liar responsibility for actions should be taken b7
26. See Gillis (1994:5). This whole process [ hus is based on a conception
of identiry as ftmda- che Cabinet, rhe State Securiry Council, individual ministers , and commanding officers for
mentally relational. identities are not aehieved as self-lame and puro, in isolation from orher
all decisions taken and actions authorized. See "Sccond Submission by Narional Part" (23,
identities, but are always corstituted in relarion to an other or a series of others.
26-27).
27. See Anderson ( 1991: 8, quoted in Gillis 1904:8).
42. 'Ihe TRC takes issue wirh [his view for ohvious rcasons , arguing mar ' explaining is not
28. Ghlis (1994:12).
excusing, understanding is not forgiving ' (TRC website, par. 45, filel))Dlfinall5chap7.hmU.
29. [bid.
43. De Brito (1997: 6-7).
30. Women, nacional mmorities , and young people were gencrally adntitted ro nacional memo-
44. Even though rhe work of the TRC does not Cake rhe form of normal criminal tribunals. it
ries at un even slower pace c han to nacional representative and educational instimtions.
gives great consideration to bcing leen working within rhe constraints of due process and
(¡bid.: ro).
rhe tale of law. This is eztremely important , because rhe ramos of both victims and perpe-
31. Fierre Nora (quoted in ¡bid.: 15).
rramrs of human righrs violations are made public . An elaborare system has been pul ¡rito
3z. Gillis (1994: 18-zo).
place ro ensure cha[ perperrarors re be named are alerted ahead of rime and are invited ro
33. Although (here is general agreement that rhe work of rhe TRC
is rhe bese possible way to make representations m hearings . To this end, the TRC has an Investigative Unit that inves-
achieve justice and reconcil¡ atimi. ie continues lo be challenged frota a variety of
positions. tigares allegatiohs and colleces any relevan [ evidence.
The National Parry continues to warii against rhe possihility of che TRC'
s being tutned ¡rito 45. Ibis objective es clearly specified in de remit of rhe work of die TRC as the need ro ''restore
a w:tch -hunt ; the Freedom Front vraíces similar objections .
The I FI? veted against the act and the human and civil dignity of victims by granting theta an oppormniry to relate their ewn
has recently resorred te publishing es objections to rhe TRC in tire
national press. These accounts of rhe violations of which they are che victims . See Promotion of National Uniry
parties were afraid that the TRC may " go roo far "; other parries ,
including che CAO and and Reconciliation Act: 5ection 3 (e) (1995)
Azanian People's Organization (AZAOO), argued that ir would not "go lar enough."
46. The TRC has, however, been criticized for disregarding "everyday" abuses as a result of thcir
34. Ir is important te note thet there ate deep and important differences between submissions
focos on gross violations of liman rights . 1 retum to rhis issue late[.
by ofñcial parties and instimtions, and [hose by individuals , These differences
are also re- 47. Paraphrasing of extrae r from the Chilean Commission ' s Report, quoted in Roraine (19966).
fiecced in the degree to which positivo effects were produced as a result of submissions.
48. Quoted in Woods (1 997). The five aten who adm i tted to partic iparing in che killing of Bike
General ly speaking , 1 argue that in rhe case of official submissions ,
the exercise was nierked in 1977 are Col. H. Snyman , Lt. Col. G. Nieuwaudt, Warrant Officers R. Marx and J. Beneke,
by a largo degree of cynicism, whereas individual submissions ,
appearances , and confes- and Capo D. Sicbert.
sions were disdnguished by their thoroughgoing and soul-searching roture.
49 In the ceurse of its work rhe TRC has widened its investigations co include analysis of the
35. See de Klerk (1997).
role of che media , die legal and medical profecsions . and the business communiry in up-
36. De Klerk (1990: 68), por an in-depth analysis of National Parry discourse
duríng Che early holding rhe insdmtions of apartheid. It has also arranged for hearings on thc question of
ig90s, sea Altbeker and Steinberg (1998:49-71),
conscription.
37. "Submission ro the 'Iruth and Reconeiliation Commission by Mr. F. W. de
Klerk , Leader of 50 Por a more derailed discussion of [hese possibilities , see Norval ( igg8a: 259-60).
the Nadonal Parir," and " Second Submission of rhe National Parry to [he Truth
and Recon- si . See TRC website , par. 45, filep/ Dlfinallichap5 . htm, par. zo.
ciliation Commission ," 1997, www.truth. org.zalspeciallparty2lnpz.htm.
52 Krog, ¡ti her discussion of die work of tire TRC, railes en important question conccrning thc
38. 1 nave analyzed rhe character of apartheid discourse , wherc 1 develop rhe argument that, in
trajectory of apartheid and irs relation to die faca rhat British abuses against women and
add ilion to its historical specificity, ir is exemplary ola generalized identitarian conception
children in eoncentration camps were rever offcially icknoevledged or eondemned by rhe
of identity (Norva11996).
Rritish: "Wasn ' t rhe mere Eact that the abuses of rhe wat were never exposed perhaps not a
39. 1. E. Young (1992: 270).
key factor in rhe charac [er that formulated aparthcid ' s laws7 Vías rhe Roer veneratimi of
40. "Submission te rhe Truth and Reconciliation Commission by Mr. F. W. de Klerk, Leader of
Emily Hobhouse no[ a symptont of che desperare need for semeone ` from rhe orher sido' to
the National Parry."
recognize the wrongs thar liad been done? " She argues chat the faca that [hese abuses were

200 STATES OF IMAGINATION NATIONAL IDENTITY AND MEMORY 201

 nu1 puhhdp rc¢ognlzcd conrrlhutnd direcdv lo liiC puosihi p rhat thc,var" beeune n tnlldore
suppm'Gng rho noüon ot Vt lcanen:le t 1hrcaca , 1 guR:p" 1i995:
RETHINKING CITIZENSIIIP Reformingthe Law
1 In Iheorerio,d ri,n.etu, of lhix arguor ni dm,cs sirongly on DInt e,a iuconstruclive retd- in Posnvar Guatemala
,gnfHl.>crl'sergunu,. ,u..inrcrua ti,t-.... tsnese Ce Uunda 1197 r,1.

54 11 1 1' issue i3 in(nrizcd m (onuo!I}'s 119911 exeul l rnt.md} ol lhe p:uodoses ot do Rachel Sieder
„- Asnral.,Axmal. and Roberts U9pnil.
h 1 i nv ala 1 aras licrors within tha n soaor and C _ trcggies Io red(line ils role in the
1 pohneal ronrert. Seo 11erwu. Urt,h11,1..111,1 llmnbu 9996-

5Bletnp998).
S . L
Q. Rodios 0998: 120-40)-
do. Ibid: 137. Rubias asks, for instance, whether "new public histories" will marginahze the
experiences of eolouteds "un the grounds that they did not suffer under apartheid as much

as black South Afficans."

61. 1 deal with the question of ethniciry in general, and in eontemporary South Afinca note
specifically, in Norval (1999a: 1-z2).
62. Indieations ofthis are present in Pallo )otdari s (1998) discussion ofthe national qucstion in This essay addresses the relationships among the state, law, and citizenship
posb1gg4 South Africa. "Chis is evident, in particular, in bis treatment of the question of in posteonflicr Guatemala, examining the ways in which struggles to redefine
ethniciry. Ethniciry, for hico, is nothiug but an artifaet imposed by the apartheid regime and the state and citizenship were fought out on the legal terrain. By definition,
ao atrempt tu conrinue to forge competition with "fellow Blacks" over scarce resources.rhis
processes of political transition following prolonged periods of authoritarian
hardly is che sott of rhetotic :hanould malee one confidcnt in rhe dcptii ofrhe pluralisin of
rule or armed conflict involve transformations of che state. Under authori-
the f,rm of non racialisni espoused here.
63. W ith regard to che former, see Posel (199x); and Norval (1996: chapa. 2, 3). For a discussion
tarian regimes or in situations of internal armed conflict, violations of human
of divisions within the tate dudng the 1970s and 1g8os, see Sarakinsky (1998: 54); and rights are a eommonplaee, either because of deliberate policies toward certain
Swilling and Phillips (1998: 134-48). sectors of the population or because of the weakness of the state and the
64. TRC website CD: par 77, file:plD finall5chap6.hnn. inability of government co protect and enforce basic rights. Any transition to
65. Some of the argumenta in this section were fitu developed in a review of literature un dte
more democratic forros of governance therefore implies the (re)construction
Tac. See Norval (lgygb: 499-519).
of the state as a guarantor of democratic rights and obligations. Through
66. Sce, for instance, Robins (1998: 120-40); Coerce (1998: 112-19); and Davidson (1998:
143-60).
such processes citizenship is reframed as various legal and institucional mea-
67. Deacron (1998:161-79;178). sures are implemented to benefir and protect (hose groups and individuals
68. Ibid.: 179. who were marginalized or victimized under the previous regime, that is, those
69. Coetzee (1998: 114-15). who were effectively excluded from citizenship either de jure oc de facto.
70. Robins (ígg8:13o).
Citizenship is ofren conceived of as a fixed and nonnegotiable set of rights
71. Ibid.: t3t.
and obligations, such as (hose embodied in a written constimtion. However,
72. Davidson (1998:159).
it is in fact bes[ undecstood as a dynamic process rather than a statie jurid-
ical construet. Both in terms of its legal attributes and its social eontent, citi-
zenship is contested and constantly renegotiated and reinterpreted. Such
contestation is parccularly acute during periods of political transition. In
considering chis phenomenon in postwar Guatemala, 1 adopt the perspectiva
proposed by Ano Misehe, who has argued for a move beyond formalistic con-

202 S'FATES OF IMAGINATION

 ceptions of citizenship "towards a view of citizenship asa historically contin-
rights are currently being played out through the law across the continent,
gent, interactive vehicle of articulation, conflict and dialogue" (1995: 157-
particularly on the question of whether the increasingly internationally sanc-
58). Mische's analytical framework is particularly useful for thinking about
tioned rights of indigenous communities should be recognized within na-
the changing nature of citizenship during transition processes, as it takes
tional legal systems.' Evidently, the ways in which the state and state-society
roto account the multiple appeals to citizenship from different and often con-
relations develop and change depend en struggles and interactions between
flicting social sectors and recognizes "the potential dynamism of such ap-
hegemonic and counterhegemonic understandings and practices of "rights,-
peals in reshaping relationships between state, societal and economic actors"
1'obligations," and "justice" as constituted by different individuals and groups
(157-58)
within particular material and historical contexts. An analysis of the ways
During periods of political transition a range of actors, including domestic
people resort to and use the law-and the idea of law-in periods of rapid
elites, international donors, and intergovernmental institutions, and political
political transformation can therefore tell os much about the changing nature
and social movements artempt to advance different and often competing vi-
of the state and struggles to redefine citizenship in a particular context.
sions of the state, governance, and citizenship. The state itself can usefully
In Guatemala attempts to redefine the state and citizenship centered on
be analyzed as a series of instimtions and cites where conflicts over powerare
the peace negotiations between the government and the combined guerrilla
constantly negotiated from aboye and below. One of the primary sites of en-
forces of the Unidad Revolucionaria Nacional Guatemalteca (URNG), which
gagement where such different imaginarias and political projects are con-
were successfully concluded in December rgg6. The peace settlement set out
tested from che top down and the bottom up is the law. This is because the
a blueprint to overhaul the Guatemalan state, for decades dominared by an
law es central to claiming rights and enforcing obligations. Nationally and,
anticommunist counterinsurgency logia, and to ensure respect for human
increasingly. internationally accepted standards are often forged through le-
rights and the democratic role of law. The three decades prior to the peace
gal processes in a dynamic of "naming, claiming and blaming" (Felstiner,
settlement were characterized by gross and systematic violations; over two
Abel, and Sarat rg8o). In effecr, the legal system is converted into a contestad
hundred thousand people perished during this period, including ovar titty
site of meaning over state accountability and citizens' rights as the dominant
thousand disappeared. The vast majority of those killed, sume 83 percent,
ideas and values that underpin the law provide the framework for advancing
were indigenous noncombatants murdered by the military and paramilitary
alternarive understandings (Starr and Collier rg89). As Boaventura de Sonsa
forces (Comisión de Esclaricimiento Histórico [GEHI 1999). Throughout the
Santos (1995) has observad, law has become an increasingly globalized or
armed conflict, the judicial system singularly failed to defend even the most
transnational phenomenon. Through the law and legal encounters, differ-
elementary of citizens' rights or punish those responsible for gross viola-
cnt ideas about the appropriate balance between rights and obligations filter
tions. Throughout the peace negotiations reform of state institutions and
among the internacional, nacional, and local arenas. Indeed, Santos claims
practices to ensure protection for basic human rights was therefore sean as
that "the main driving force behind the transformation of the state and its
central to wider efforts at democratization. The peace settlement also mapped
legaliry is the intensification of transnational practices and global interac-
out a radical agenda, which aimed to include Guatemala's 6o percent indige-
tions" (rig). Certainly, claims by groups and individuals for greater political
nous population, historieally subject m discrimination, socioeconomie expío i-
inclusion, equaliry of treatment, recognition of minority rights, or punish-
tation, and political marginalization, in a new nation-buildingproject. In spite
ment of [hose responsible for gross human rights violations are increasingly
of a deep-rooted legacy of racism, by the late 199os internationalized "rights
framed through appeals m international law. In recentyears in Latin America
thinking" had become parí of the dominanr idiom of political reform in Gua-
social movements and individuals have used international human rights law
temala. This was evident in the agreement signed by the negotiating parties in
ro try to secure truth and justice in cases of gross violations of human rights,
March 1995 en the Rigtts and Identity of Indigenous Peoples, which framed
particularly when local courts fail to enforce accountabiliry (as, for example,
the specific demands of Guatemalan indigenous groups within a wider evolv-
in the Pinochet case). In addition, indigenous peoples' claims for special
ing international discourse and legislation on indigenous peoples' rights. The

204 STATES OF IMAGINATION

POSTWAR GUATEMALA 205
 nlan rights and obligations rogether wich che spccial rights of indigenous
agreemenr proponed such mensures as a conunicmenc to hilingual education.
communities or ethnolinguistic groups. Such nmlriculmral rule oí law con-
religious tolerante, and respecr tor indigenous authoricies. It also endorsed
strucrion. thcy argued, formed part of che rvider challenge ot building clic
rhe principie of spccial rights and positive discrimination widt rhe aim of
busis for a multieth nic and nmlcicultural cirizenship in Guatemala. Hoy+'ecel.
achieving greater social juscice The overall thrusc of che agreemenr aimed
their opponents, represented in rhe govennnent che privare sector, che me-
to encompass cultural diversiry wichin che unitary fiamework of a retbrmcd
dia, and che legal profession, frequently raised clic specter ot "ball<anization,"
G uacentalan state This, it was argued, would give hisrorically disadvanraged
arguing rhat granting spccial rights ro indigenous people would ultinrarcl}
groups a greater stuke ni rhe national policy and articulare difieren[ tonns of
lead ro echnic separucism and clic brcakup of rhe couucrf'1 hec also claimed
participacion iu a developing project oí multlcultural cirizenship.' Indeed, die
that recognition of indigenous customary law would result in 'ene las tor
suceessful incorporation of rhe indigenous population ¡oro a democratized
indigenous people and another law for non-1ndians," contradicti ng che con-
body politic was viewed by many as essential to securing a lasting peace.
stitucional principies of legal equality and universal citzeuship. In effect, rhe
In che period subsequent to che signing of che peace settlement the law
barde over multiculturalism and che rule of law in Guatemala illustrated wider
became a key battieground where artempts to redefine cirizenship were played
politieal conflicts over tire namre of tire state, governance, and cirizenship.
out. Efforts co construec che role of law had to address a legacy of militariza-
The following sections situare current debates about cirizenship in Guate-
tion, an ineffectual and corrupc judieiary, entrenehed impunity, and che com-
mala in historical concext and analyze transformations in che rule of law that
plex challenge of encompassing cultural differences. The 1995 Indigenous
have oceurred since che signing of che peace settlement. 1 conclude by consid-
Righcs Agreement specifically recognized che legitimacy of indigenous "cus-
ering che implicacions of recent developments for che state and citizenship in
tomary law" (derecho consuetudinario) and "tradicional" eommuniry authorities
and committed che government to incorporare both inro che official srace le- Guatemala.
gal system where they did not conflict wich nacional and internacional legal
norms of human rights.' This implied che official acceprance of legal plural- Rethinking Cicizenship
¡sin and che right of indigenous communities co resolve conflicts in accor-
Following che end of direct military rule in 1984, a constitution was promul-
dance wich their own "customary pracriees." ` In effect, it envisaged provid-
gated that enshrined universal principies. However, a history of exclusion,
ing indigenous people wich a higher degree of legal autonomy chan they had
authoritarianism, and racism and, more reeently, extreme state-sponsored
enjoyed for oven a cenrury. in the wake of che peace agreemenr policymakers
violente meant that many Guatemalans continued to view tire state as an op-
began to confront tire complex challenge of integrating internacional legal
pressive force rather than a guarantor of their rights as citizens.' In addicion,
instruments,° national lave, and indigenous customary law. A number of
until tire r99os, tire cold war and tire acutely exclusive namre of national poli-
studies on indigenous legal norms and practices were carried out by nacional
cies had engendered an organized opposition that sought to destroy che state
rsearch eenters, such as che Latin American Faeulcy for Social Science
rather than to influence or reform it' Beyond highly restricted interelite com-
(FLACSO) and tire Assoeiation for Socio-Economie Research (ASIES), a gov-
petition, che state and che law were not perceived by most people as a viable
ernment-sponsored think cank. Research and discussion foca en customary
terrain wichin which co struggle for social change.
law involving Mayan intellecmals, policymakers, and academies were also
The eontemporary difficulties involved in construeting a democratic ciri-
sponsored by che United Nations observer mission in Guatemala Misión de
zenship in postwar Guatemala can be traced back to che constitution of che
Las Naciones Unidos en Guatemala (MINUGUA), set up in 1994 tu monitor
republic itself. The consolidation of che apparatus of tire state during tire late
human rights and verify tire implementation of che peaee agreements in sicu.
nineteenth and twentieth centuries was nor followed by tire suceessful con-
The debate was heaced and often acutely polarized. Indigenous activists and
struction of an imagined nation-state peopled by an active citizenry. The na-
their supporters in incernational NGOS and tire United Nations argued in fa-
tional project of tire country's dominant civil and military elite during che
vor of developing a politico-legal system thar eould encompass universal hu-

POSTWAR GUATEMALA 207

206 STATES OF IMAGINATION
 nineteenth century did not include either the indigenous majority or most
absence of an effective state presence in many parís of che national territory.
poor non-Indians as citizens wirh equal rights, but rather treated them as
This was particularly the case in che highland departments of greatest in-
subject populations to be controlled, disciplined, and "civilized." Following
digenous concentration, such as Huehuetenango, Alta Verapaz, and Quiché,
independence, the conservatives proposed a new status for indigenous
where economic elites relied en coercion and clientelist networks to secure
people, creating a separate, subordinare legal jurisdiction for them, which in their access te labor and land.
effect recreated che colonial republica de indios. As Guatemalan historian Arturo
After 1963 che armed forces began to develop a counterinsurgency project
Taraeena (1995) has observed, although such ethnojuridical exclusion af-
to defeat the URNG, which reached irs zenith during che early Ig8os. The
forded indigenous people a degree of protection, it also meant they did not
relationship between state and subjects changed dramatically as che army
actively participare in che process of nation-state construction. This did not
consolidated a highly militarized and autboritarian forro of governmental
mean an escape from subjugation; che economic exploitation of the indige-
power throughout che rural highlands (Schirmer 1998; R. Wilson 1995b; Ko-
nous and poor non-Indian majority was-and remains-central to the con-
brak 1997). Extreme and violenr repression was combined wirh institucional
stirution of the Guatemalan nation-state. The idea of separate legal regimes
mechanisms to incorporare and control che rural, predominantly indigenous
for Indians and non-Indians was opposed by nineteenth-century liberals, yet population, including forced resetrlemenr in some arcas, obligatory partici-
che triumph of che Liberal Revolution of justo Rufino Barrios in 1871 did not
pation in paramilitary civil defense patrols, and village-based military com-
result in che exrension of futl citizenship status te all Guatemalans. Within
missioners who reported regularly to the army. Existing forros of commu-
the context of che late nineteenth-cenmry agro-expon boom, universalist
nal authority and organization were destroyed as thousands of communiry
ideals of equaliry before the law and positivist doctrine constimted che ideo-
leaders were murdered or disappeared by che military. Long-standing inter-
logical apparams for che exploitation, expropriation, and assimilation of the
and intracommunity conflicts over land were often plaved out through the
indigenous population. Communal lands were increasingly encroached upon,
lechal prisco of che counterinsurgency, exacerbating local divisions. Despite
and indigenous men remained legally subject te forced labor requirements
che transition to elected civilian government in 1985. che militarization of che
until 1944 (McCreery 1994).°
countryside and gross violations of human rights persisted throughout che
During che 1944-1954 period of reformist government, political parties, next decade. Only when che peace negotiations berween che government of
universal male suffrage, and a program of agrarian reform were introduced
Alvaro Arzú Yrigoyen and che URNG recommenced in earnest in 1994 was che
throughout rural Guatemala, providing many younger, literate, and Spanish-
prospect of transforming che militarized, authoritarian state finto a guarantor
speaking Mayan men wirh access te municipal political office (Handy 1994).
of che rights and obligations of all Guatemalans genuinely raised. The nego-
Although illiterate indigenous women were still denied che vote, these changes
tiations themselves remained a restricted affair between che belligerents, but
srimulated the development of an active civil sociery and began to lay the basis
social movements did have some limired input to che discussions rhrough che
for the development of a citizenry that included che majoriry of che popula-
Civil Society Assembly (ASC), a consultative body set up in 1995 through
tion. However, che anticommunist counterrevolution that followed che CIA-
che framework negotiating agreement. The asseinbly was able to forward dis-
backed overthrow of President Jacobo Arbenz in 1954 effectively halted che
cussion documents ro the negotiating Cable, and wirh che participation of che
eonsolidation of local political power and reasserted che dominante of cen-
umbrella Coordination of che Organizadons of che Mayan People of Guate-
tral government In subsequent decades municipal and regional government
mala (COPMAGUA) had an important role in determining rhe nature of che
throughout much of che country remained weak, and che majority of che
1995 agreement on indigenous rights and identiry. Civil sociery participation
population was politieally marginalized and subject to ever greater economic
in che peace agreements was limired,'° but che contrast to 1984, when civic
exploitation. This was in parí che product of a highly coercive state, but also,
opposition barely existed and che nature of che political rransition was dic-
paradoxically, of its weakness, which facilitated other forms of domination
tated almost entirely by che armed forces, was striking nonetheless.
and exploitation. Indeed, highly centralized government coexisted wirh che
The principal challenge to dominant liberal, universalist conceptions of

208 STATES OF 1MAGINATION

POSTWAR GUATEMALA 209
citizenship iu Guatemala has come t[om an emerge11 t pan-Maya11 ism. During
ideas of legality were used by opposition groups ro challenge and conresr

ds 199os chis diverse social and political movement questioned lile UNIsnill', oficial violations of civil aud political righrs.

uarion-state, drawing inentiou to irs discriminarory aud exclu sionary tinture

However, time proposal conrained in die 1996 pelee agrecutenrs ro incor-

(bastos an(¡ Camus 199 ; Coja 1996: Fischcr and Erown 1996; Kay Warren
poratc índigcnous "customary law" inro tlte norns and instin.rions of snnr

1998; Nelson 1999). Activisrs have drawn inspirarion and suppmt front time
law represented an even more radical departure. Here, che terrain of time law

transnationalized indigenous peoples movemenr in r11e Americas, but the

was being used to contesr fundamental assump tio us about universal cirizen-

growing sircngth of time indigenous movement also refiects che fict that ¡u- ship and time nature and uuity of time Guaremalan Arate ln eftect, [he proposal

digenous righrs huye increasingly occupied rime agenda of nongovertt mental

can he sean as part of a wider artempt by indigenous acti%ists ro secure greater
autonomy fbr indigenous communities and peoples. by recognizing rime rlght
and intergovernmental organizations in time international arena, not least time
United Nations, which afrer 1994 provided Mayan organizations in Guatemala
of indigenous com monines to decide en and apply their own procedures for
conflict resolution, the state would no longer mainrain a monopoly en time
with a strategic ally in the form of its observer mission. During the 199os
demands for indigenous rights and idealized projections of "Mayan values" production and regularion of law. This constituted a fundamental challenge

carne ro constitute a new discourse artieulated by indigenous leaders and their to time legal centralist model that had existed since Independence, based on
the indivisibility of state, government, and law. Yet the proposal can also be
supporters . This began ro affect social relations in Guatemala and helped lo
frame much of time nacional debate around judicial reform , democratic con-
read as an attempt by certain political elites in che state to renew their flag-
ging legitimacy. Santos has pointed lo "che way in which the state organizes
solidation , and citizenship during time peace process ( Sieder and Witchell,
its own decentering" (1995: 118); the commitment conrained in the peace
forthcoming).
settlement lo build a multiculmral rule of law by incorporating previously
excluded, "nonstate" legal phenomena inro srate law can also be understood
A Multicultural Rule of Law? in Chis light. Concessions lo multiculmralism would, itwas hoped, secure the

The ideas of the "rule of law" and "equaliry in che law" have long been used support of che indigenous population for a new pací of governance and also

as a legitimizing discourse of state power. However, they have also provided convince international donors that che Guaremalan political elite was serious

many groups and individuals with a means lo challenge and contest those con- about reform.
trolling the apparatus of che state by demanding that certain righrs be guar- In common with other postcolonial nation-states, Guatemala has long

anteed and legally prescribed obligations respected. During the late 198os and been characterized by a situation of legal pluralism where different systems

1990s, NGOS and popular organizations in Guatemala increasingly carne tu of obligation have coexisted within che same soeiopolitical space, albeit in a

use the law as a site to struggle for citizenship, campaigning for an end to highly unequal relationship. This stems parfly from the spatial and juridical

impuniry and respect for constimtional and human righrs. por example, hu- separation of indigenous peoples in the colonial pueblos de Indios or their post-

man righrs groups such as the Mutual Support Group (Grupo de Apoyo Mu- independence equivalents, but also from the uneven coverage of the state over

tuo, GAM) and che Nacional Coordination of Guaremalan Widows (Coordi- national territory, which means thar che nacional legal system remains largely

nadora Nacional de Viudas de Guatemala) denranded an end lo human rights inaccessible for most of the population, who are obliged to develop alterna-

abuses, the observance of due process guarantees such as habeas corpus, and tive means to regulare social relations. However, the posited division berween

time convienen and punishment of those guilty of violations. The indigenous "customary law" and "state law" is in fact a legal fiction. As much recenr

organization Consejo litnico Runujel Junam (cER)) organized after 1988 spe- scholarship en legal pluralism has shown, more often than not what are iden-

cifically to lobby state authorities lo respect che voluntary nature of the civil tified as local customary practicas are in fact colonial or postcolonial impo-

patrols prescribed in the 1985 constitution. Struggles over state accountability sitions (A. Griffiths 1997; Moore 1886; Starr and Collier 1989). State law and

and citizens' rights were fought out through rhe legal system as dominant customary law are lociced together in a historically asymmetric but mutually

210 STATES OF IMAGINATION POSTWAR GUATEMALA 211

 constirutive relationship in which state law may be understood as an attempt
patrols demobilized weeks alter the final peace settlement was signed. Com-
by che political center to dominare, govern, and adminístrate indigenous
bined with the fact that rural inhabitants became aware of the content of che
peoples, and local practices seen as a means of adaptation, accommodation,
peace agreements and che increased nacional and international currency of
and, occasionally, resistance. In one sense, as Laura Nader (19co) argues in
"Mayan rights," this mean[ many came to question the legitimacy of central-
her study of customary law in che Zapotec region of Oaxaca, Mexico, custom-
ized forros of governance and law. Many Mayan NGOS actively encouraged
ary law can be understood as a counterhegemonic strategy used by indige-
such reflection, such as che legal rights service Defensoría Maya, whieh enm-
nous communities to protect their limited and conditional autonomy from
phasized che importance of che "recovery" of indigenous customary law, or
che central state. In Guatemala, central government has never been wholly
Mayan law, as they preferred to call it. Through chis process, unofficial forros
successful in imposing its authoriry. Instimtions such as che colonial alcaldías
of dispute resolution gained a renewed legitimacy in many indigenous com-
indígenas or che contemporary alcaldías auxiliares and even, according to some
munities. In effect, che state was disaggregated and an increasing number of
recen[ accounts, che counterinsurgenc civil defense patrols instigated by che
people carne to imagine che law as something thar could be construcred from
arme were adapted in practice by subject populations to become institutions
che botrom up, notjust dispensed from che top down.
used to defnnd a qualified local autonomy (Sto¡¡ 1993; Kobrak 1997). How-
Practices varied from place to place, bus local dispute resolution was gen-
ever, it is also evident that state law is nos something somehow external to
erally characterized by extended dialogue between die parties in conflict to try
indigenous communities; no hermetically sealed spheres of "nacional state
to active ata mutually acceptable solution- The auxiliary mayor, local religious
law" and "indigenous customary law" exist, and indigenous people certainly
authorities, or other communiry leaders mediated discussion. Misdemeanors
make recourse to all means at their disposal to setde disputes, both official
were analyzed in terms of motive and effect, che guilty parry usually expected
and unoffcial." Nevertheless, in che highly contested debates surrounding
to acknowledge his or her wrongdoing, and che victim given an importan[
recognition of indigenous customary law that occurred alter che peace settle-
role in proceedings-in marked contras[ to proceedings in che courts. The
ment, advocates repeatedly emphasized che differences berween the rwo "sys-
sancrions applied were predominantly restitutive and reconciliatory in nature,
tems" of law, contrasting what they claimed was che eonciliamry and non-
such as public apologies, replacement of goods damaged or stolen, or com-
coercive nature of customary law with the punitive sanctions imponed by che
munity service (Sieder 1996; Esqui[ and García 1999). This again contrasted
nacional judicial system, widely condemned as discriminatory, costly, and in-
with che state legal system, where fines or long prison sentences were che
effective. This revealed che extent to which che moral force of "customary
most commonly applied sanctions. Indigenous activists advocated che"recov-
law" rested precisely on its imagined "nonstateness," even while its very na-
ery" of these practices and "tradicional beliefs" as parí of an overall strategy
ture was partly defined and shaped by nacional law." lis formal recognition
to strengthen Mayan identiry and improve access to justice for indigenous
would have means that something framed largely in opposition to the stare
people. The active reconstruction of a Mayan past involved collective imagin-
would in fact have become parí of che apparatus of the judicial system. Such
ings of foundational myths. traditions, and shared histories, or "imagined
a proposal was, in turn, a response to che increasing rransnationalization of
communities." In this sense, the essentialist discourse of Mayan activists,
the Guatemalan legal space, reflecred in che influence of international think-
which emphasized che harmonious nature of pre-Hispanic Mayan society,
ing on special rights and multieultural eitizenship on che peace settlement.
is best understood as a strategic resourcc, what William Roseberry has de-
Alter mid-1994, che presence of che United Nations verification mission
scribed as a "language of communiry and contention," that is, a "social and
in Guatemala ensured that the discourse of indigenous and human rights
discursive construction and imagination" (1996: 83) that subaltern groups use
reached even che most remoce rural arcas, even if in some places it was ini-
as a counterhegemonic mechanism to contest domination."
tially received with suspicion or openly rejected by [hose who maintained
However, ethnographic research demonsrrated that, in practice, indige-
close links with the army. During Chis period the direct presence of the mili-
nous people throughout Guatemala continued to engage in che strategic use
tary began tu recede. Military commissioners were abolished in 1995 and civil
of law, resorting alternatively to nonofficial institutions and practices, state

212 STATES OF tMAGINATION

POSTWAR GUATEMALA 213
 and the discourses of disputing," and that "paradoxically, larv `genders' in-
law, and, increasi ngh, m international legal foca to resolve disputes and pur-
dividuals in ways rhat define their positions both in socicty and in legal con-
sue grievances. Local eonflicts, divisions, and faetionalism frequently ex-
rexts, while also affording spacc fmr contesting tllose positions" (1y98-5, 20).
presecd themselves through Chis snategic resort to differenr dispute resolu-
l'hrough recourse ro the national courts and aso to Che international arena.
don mechanisms, and individuals and groups took their disputes to thc
niany indigenous women liave gained an increased con scio u sic ss of them-
courts or other institutions, such as MINUGUA, when they perceived it tú be
selves as individuals wirh rights and obligations. In Che process, rhey llave, at
to their advanrage ro do so." The experiences of conflict, refilge, and tesis-
leas[ potent ially, reimagined themselves as cirizens oí Che nation-stars.
tance during thc armed cottfiicr had transtbrmad understandings ot law, jus-
In general, howevcr. expecrations oí the national judicial svsrem were para-
tice, and rights for inany people. Returned relitgees, for example, liad ap-
doxically boli extremely low and high at the sane tinte. A growing nunlber
propriared and internalized universal discourses and frameworks of human
of individuals and groups expected Che state w guarantee an ever increasing
rights as a result of thcir interaction wirh NGOS and Che United Nations High
range of rights, yet their everyday expecrations of die ofhcial justice system
Commission for Refugees (UNHCR). This highly politicized group lobbied
remained minimal. Inefficieney, widespread corruption, and impuniry contin-
to secure their relativa autonomy from the Guatemalan stare, securing ofb-
uad te prevent Che full exercise of rights or Che enforcement of obligations,
cial commimients through Che peace negotiations in 1994 that Che military
generating a simation for most of the population thar approximated what
would nos enser returnee settlements without prior consultarion." Returnees
Guillermo O'Donnell has termed "low-inrensiry citizenship" (1996: 166). In
also refused to have civil patrols, local police, or even auxiliary mayors within
addition, Che marked risa in common crime, such as robbery and kidnapping,
their resertlements, relying instead en Che organizacional strucmres consoli-
that occurred in Che wake of the peace senlement focused public atrention on
dated while in exile in Mexico. The access rhey had gained to a "transna-
Che question of "law and order." Among some sectors Chis lcindled nostalgia
cional political space" (Pritnhard r995: 24) liad strengthened their perceprion
for Che time of military rute during Che carry tg8os, when Che mitirarized state
of themselves as a population with rights, and they were more willing te ap-
dispensed a highly punitive and summary justice throughout Che country. Al-
peal te both national law and international human rights law to defend their
though a rango of judicial reforni initiatives were supporred by international
inrerests. organizations such as Che World Bank and Che Inter-American Developrnent
Another importanr ehange that occurred in "rights understandings" re-
Bank, tangible results were slow to materializa. At times, reform efforts were
lated to gender. During the armed conflict, many indigenous women became
counterproductive; for example, changas introduced to Che Criminal Proce-
widows and were obliged ro occupy roles in Che family traditionally reserved
dures Cede in 1994 aimed at guaranteeing the rights of detainees, such as the
for men. Some later became importan[ local and national leaders, for ex-
right of habeas corpus, were accused by many commentators of favoring al-
ample, Rosalina Tuyuc, who became a congressional depury in 1995 for the
leged delinquents. Most dranlatic of all was Che steady mercase in mob lynch-
left-wing opposition party, the New Guatemalan Democratic Front (FDNG).
ings of suspected eriminals in rural and urban eommmnities that occurred
The protagonism of Rigoberta Menchú, awarded Che Nobel peace prize in
throughout Che country alter 1996. As state-orchestrated political violence re-
1992, also symbolized Che growing political impmtance of Mayan women and
ceded, many instantes were reported of residents detaining, beating, and, in
their projection onro a global stage. In Che walce of Che armed conflict, many
Che most extreme cases, murdering and burning [hose suspected of carrying
women increasingly challenged their traditionally subordinare role in local
out erimes such as robbery and rape.- These lynchings were attributed to a
affairs and became less willing to submir to the will of Che men in their com-
variery of factors, including Che failure of state forces to guarantee law and
munities (or customary law). Cases of neglecr, abandonment, and domestic
order and Che legacy left by Che violence of Che military counterinsurgency,
violence were taken to Che stare authorities, where they often received a more
when a brutal and summary form of `justice" was the order of Che day.
symparhetic hearing [han before local authorities. Recent experience in Gua-
Debate continued around Che implemenration of Che commitments on legal
temala bears out Che observation made by Susan Hirsch that "gender is made,
pluralism set out in Che peace agreements. In effect, Che inrernationalized
remada, and transformed in fundamental ways through legal institutions

POSTWAR GUATEMALA 215

214 STATES OF IMAGINATION
 discourse of indigenous rights and identiry contained in the £995 agreement
ernment or che United Nations observer mission from campaigning in favor
framed customary law as a localized phenomenon, intrinsically bound up
of a yes vote . In che days prior to the referendum the powerful privare-sector
with a particular vision of Mayan identiry that derived essentially from rural,
associarion Comité Coordinador de Cámeras Agricolas, Comerciales , Indus-
agricultural communities." Did recognition, then, mean creating autono-
triales, y Financieras (cACIF) rook out full - page advertisements in che na-
mous jurisdietions in Mayan villages? Was indigenous customary law to be
cional press warning of dice consequences if che package was approved. Nu-
incorporated finto state law only at Chis level? And what would the relationship
merous newspaper editorials discussed the dangers of giving political rights
be berween local indigenous authorities and the courts? The inherent ten-
to "illiterate lndians" and predicted " reverse discrimination" against non-
sions berween ethnic integration and autonomy apparent in the entire peace
Indians in The fumre. Although indigenous activists in die cides and the high-
settiement were played out through chis one issue. Some Mayan activists and
lands supported che reforms , tl£is proved insufficient to mobilize a skeptical
even some inc personnel argued in favor of separare territorial jurisdietions,
and largely uninformed electorate, and che package was defeated . The pos-
wherc "Mayan law" could be applied to che indigenous population.'a Other
sibiliry that customary law would be formally recognized as parí of che state
advocates of recognition (myself included) argued that the establishment of
legal system in che short term looked increasingly unlikely. Nonetheless, in
"separare but equal" jurisdictions would marginalize indigenous groups even
che wake of che referendum defeat indigenous activists vowed to continue
further from the nacional poliry and instead proposed the strengthening of
working to strengthen " Mayan law" in communities throughout the country
local conlict resolution mechanisms together with the greater incorporation
as an alternative to dispute resolution through che courts.
of che general procedural principies of customary law, such as conciliation
and mediation, finto a unitary judicial system (see Sieder £998; Esquit and
García r999). In 1998 opposition parties in Congress proposed a reform of Conclusions
che 1985 Constimtion to recognize indigenous communiry authorities as ju-
State formation is a continuous process negotiated both from aboye and be-
dicial bodies. The congressional formula, endorsed by most indigenous civic
low. 1 have sought to indicate che ways in which che peace process and [he
organizations, was that these authorities' resolutions and dispositions be
prospect of incorporating indigenous customary law finto che legal system
mandatory for those individuals who voluntarily opted to accept their juris-
changed che possibilities for imagining che Guatemalan state and citizenship.
diction as long as they were not in contravention of their constitucional and
The multiculturalist perspectives embodied in che 1995 agreement on indige-
human rights. in May 1999 che issue carne under renewed scrutiny when a
nous rights and identiry eonstituted a reframing of citizenship, raising che
nacional referendum took place to approve a package of fifty constitucional
possibiliry that che construction of a nacional citizenry more inclusive of [he
reforms previously approved by Congress designed to institutionalize the com-
indigenous majority could cake place. However, at che same time as these
mitments set out in che 1996 peace agreements. The proponed reforms aimed
claims were taken up by sectors of che Mayan movement, resistance from
to recognize che Guatemalan nation-state as "multiethnic, pluricultural, and
powerfid groups to formally recognizing indigenous rights illustrated che
multilingual." They also advocared che recognition of customary law as parí
highly contested nature of chis process. In addition, while certain sectors of
of [he judicial system according to che formulation previously agreed en by
che population increasingly called on [he government to guarantee certain
che parties in Congress. However, in che end, che majority of Guatemalans
rights and obligations, expectaúons of the state in general remained very low.
were indifferent to [he significance of che referendum, and the vote was
I have indicated here how Chis is a consequence of historical traditions of
marred by an abstention rafe of 82 percent.'° Many of those who did tum out
state-subject relations in Guatemala and of che current inabiliry of che stare
to cast their ballot, mainly in the capital ciry, were alarmed by the strident
to meet even che most basic of its cirizens' demands. Although che peace
warnings from far-right and conservative sectors that granting special rights
process opened che possibiliry of constructing an active citizenry of a democ-
to indigenous people would lead to ethnic division and conflict. Opponents
ratized nation-state, progress toward chis goal is far from guaranteed.
of the reforms had previously obtained legal rulings that prevented che gov-
However, [he contested process of developing a new system of law, rights,

216 STATES OF IMAGI NATION

POSTWAR GUATEMALA 217
arad justicc in the postwar period has profoundly itfcctccl undcrstandings ot gu:deduouion,and pohco decislous ovice developmenc priueilies. r t'=Cor,s o1á indigmlous
people ngit e are etrr ntI bcing draft d b , t h e 1 ur d Ecn-on md h org: u adora m
lile state, citizenship, arad relations betwccn rhe state and civil soeiery. In one
Amedc.ut Smte._ On irte melional t arad -m= den t, nbl s. se cll tger ;1996] and
serse, the proposal to founally recogiiize indigenouse ustomary law (and, bp
I'In a[([ave E.
implication, the auronomy oí local indigenous communities) represcnred a Fue CSmtple. lile agrecnient prupn.ed spu-il1, m irnnuce u'tiot 1,11 11-2,, 111,1'. .n thv
challenge ro rhe legal sovcrcignty of lile Guatemalan state. Indeed, Chis vas hrreicr prnalizrdon of sexual crimen connnilted aC nnst indigenous womcn.
oue oí rhe principal rcasons opponents rallicd against it. 1-1owever, 11 haave } libe setrleinent as t schole proponed in ie.creasc ü o:a, aumao 1,v thrtIltgh thc decenrrai-
iGIlimtotpobtimadmimstradccstrim'es and ilt aíenohenin ci wnt;cipal ❑ oa:ntcx
argied
t here rhat the- incorpora tion of customarv law foto official legaliry can
indigenous CPainu ter greater rcgioi al .unooo11n n spccidc ctl > ir aivtic c. nn.t.n'.ti. s
al so be under'tt oda a new t rm oí governmentaliro a wav in which the_
tuca net secured in tire negotindnm Etlúrts hysonme indigcuous mt sts lo sec e propor-
state maps out new territorios and commumrles extendmg ras control to
tional rcpresentarion in cottgrees on the basic ofethnolinguisdc idenun vele similarly un-
arcas previously bevond its reach. Yet such an exerelse cannot be understood sucoessful . For a discussion of dre indigenous movement' s position ora auronomy during ti,,
merely as an extension of powcr from the top down ; it also signals the man- peace negotiations , sec Coja (1997); for ara earlier position ofche Mayar movemcrtt ora die
ner in which the state becomes increasingly porous as the boundaries be- auronomy question . ser Consejo de Organizaciones Mayas de Guatemala ( 1991).

tween state and society change. If indigenous customary law had been for- 4. Customary law vas not explicirly debed in the peace agreements daemselves. Ir can be
understood as the uncodified eoneepts , beloefs, and norms thar, wirhin a given cornm uno ty,
mally recognized as parí of the state judiciary, chis would nave represented
define prejudicial aetions or crines: the processes by which drene shotdd be resolved; and
the incorporation of unofficial discourses and practices into the realm of a
lile sancrioas or resolutivas deelded and applicd.
new governmentaliry . The rejection of the constitutional reformo package in 5. Co legal plural isin, set 1. Gdffiths (1986); Iaoolcer (1975); M ' (1988, 1992. 1993); Moore
May 1999 demonstrated that the idea of a In ulticultural nation-state is notyet (tg86); and Starrand Collier (1989).
socially and culturally embedded in Guatemalan civil and política¡ soeiery. 6. Recogn ilion of indigenous peoples' right to ase tradicional legal practices 15 dearly set out

Nonetheless , irrespective of official recognition , nacional and international in Articles 8 and 12 of the ao Conventi(in 169 on indigenous and tribal loco ples, which vas
hnally ratified by che government of Gua te mala in Merch 1996 following a protracted polit-
developments in recent years have greatly increased the legitimaey of local
ical battle. Acceprance oí the Convention vas lude condicional ora its su boca inati ora ro che
forms of conflict resolution for many indigenous people. in addition , the sin-
i985 Constitution.
gular inefficiency of the Guatemalan legal system means that many people
7, Gross violations of human rights continued during che late og8os and 199os , despite che
will continue to resort tu "extralegal " practices for dispute resolution . Yet, at transition ro elected civilian government in 1985.
the same tinte , the state will also continue to be a central focos of human 8. The ideology of armed revolution that posired ara altemative , socialist niodel oí the state
rights struggles . The law will therefore remain an important site for the on- constituted a vital part oí chis equation.
9. After the onaoduction of new vagrancy legislador ora 1934 during che regime of Jorge
going contestation of the imaginaries and boundaries of the nation - state.
Ubico, poor ladino as well as indigenous tren became increasingly subject to forced labor
requirements.
ro, Por critical analyses, ser Palencia (1997) and Floliday and Sranley (forthco ming).
Notes
tt. Neither is nadonal or costomary law isolated floral che claanging internacional legal orden
París of tilos aracle derive Prona a previnos paper , '' Rethinking Citizenship : Legal Pluralismo Santos ( 1987 ) notes chal the legal eontext le now characrerized by interlegality and a mixing
and Instimtional Reformo in Guatemala,' published in Cotizenahir Studies 3, no . 1 (1999). 1 am of cultural cides , while global diseourses are loeally vernacularized and eonsrandy acqui re
greatly indebted ro Carlos Flores, James Dunkerley , Maxine Molvneux , len ny Pearcc, Finn new nteanings.
Stepputat , a nd John Waranabe for their feedback en earlier versions . Rescarch in Guatemala 12. Customary proeed ores were coneil iatory ¡ti na ro re in part because the da real of pu ni tive sanc-
vas finaneed by ara Ecunomie and Social Rescarch Couneil postdoctoral award, rhe British toons imposed by che state tribunals existed as ara altcrnarive if conciliation failed. Por a
Academy, and rhe Central Reseach Fund ofthe Universiry oí London. detailed study of rhe reladonslop becween etate lave atad cusromary legal 'alee hani sm s, see
The 1989 International Labor Organ ization ( ILo) Convention is currently Che only stamtory Esquiramad García ( 1999).
inmrnational instrument un rhe rights oí indigenous peoples. Ti establislies indigenous 13. Roseberry ntaintains :''As such comtrunities ore imagined , symbols al distinctiveness and
rights tu naroral tesource use tradicional lands , custoniary law, tradicional .1uthorities , bilin- a urh e nticity are selected amad ap propriared , wothin a social field marked by o n equaliry, 1icr-

218 STATES OF IMAGINATION POSTWAR GUATEMALA 219

 anche and contention. Languages of ethniciry, religion, and nationalism draw upon imagen
of primordial associations and identificarions, but they Cake their specific and practica) forms
as inguages of contention and opposition" (1996:83). GOVERNANCE AND STATE MYTHOLOGIES IN MUMBAI
14. As Anne Gritfiths has observed, "The administratlve and theorefical separadon of legal sys-
tems does flor extend tu people's uses of che law in arranging their own lives" (1997: 2). Thomas Blom Hansen
g. This comrnimlent was tragically broken in September 1995, when an army platoon entered
che refugee resetdement of Xamán. Alta Verapaz, and shot dead eleven refugees.
r6. A 1998 report by the UN verification mission (MINUGUA) en 119 lynchings carried out be-
tween March 1996 and March 1998 found that most had oceurred in regions of the country
worst hit by counterinsurgency violence during che ag8os, even though [hese arcas had conl-
parafively low leveis ofcriminality (MINUGUA 1998).
17. See Nelson (1999) for an incisive critique of chis view ofMayan identiry. Nv
a44,, ^ Gr✓/ t . f ^
i8. Many Mayan acrivists rejecred che terne "customary law," interpreung it as a colonial impo- Ef CK

sition that framed indigenous legal practices in terma of "custom." They preferred instead
the terco "Mayan law."
n ✓ crrw ,^e .wti^f ^i,
ig. Por more en che popular referendum, sea Arnson (5999). ^n n,,.^ ^4 rr 0 ^ vr^v{
y^..44 dh riC i..cnpo iy„ cc! M"^ ^l

In ale thos ^^jJ^d%

o ks thar need no particular and excepcional eforts, no special courage 41

or endurance, we find no magic and no nrythology. But a highly developed uragic asad -e^
connected with it a mythology always occurs if a pursuit is dangerous and irs issues
uncertain.-Ernst Cassirer, The Myth of che Stute

Shattered Myths

A passage in The Moors Last Sigh by Salman Rushdie captures che gap between
the dominant self-imagen of the nationatist elite and the cultural practices of
the popular worlds in Mumbai: Every year during che Ganapati celebrations,
when the elephant-headed god Ganesh is celebrated in huge public proces-
sions, the modernist painter Aurora Zogoiby dances in her white dress on top
of her house at Malabar Hill, displaying her rebellious sophistication as well
as her contempt for what she regards as a primitive Hindu mass festival un-
folding at the popular beach deep below her. One of the central themes in
Rushdie's novel, however, is that beneath this gap in representation, Mum-
bai's official face and che life of its aflluent elite are intimately interwoven
with the city's popular worlds as well as its most murky sides: massive cor-
ruption, organized crime, and communal politics, personified by the charac-
ter Raman Fielding, a gangster and populist politician.
This constitutive split in che life and imaginings of the city was highlighted
by two rounds of bloody riots between Hindus and Mustims in Mumbai in
December 1992 and January 1993 leaving more than one thousand persons

220 STATES OF IMAGINATION

 arson in the ciry-organizad and eneouraged be Shiv Sena-che police ac-
deatl, manr more +vouuded and more chal 15o,ooo people displaced. 1 he
tively assisie11 Hiudus. proceded theni tioni Muslim eounteratrackS, or simple
riotc dealt a major blo+v co che imagen . favored by che middle elass , ofNlumbai T
tamcd a blind ave en clic atroci[ies and pluoder he I lindo miiitants. The
as che epitome of Indiau modernity and che cite of a pragmatic , enterprising
presente oí leading e, ongress nlinisters in .t1unlhai duting che : iots and decir
eapicalist ethos . The terocite , che scale, and che political character of che riors
deployment of che newly formad Rapid Action bocee did nor curó clic sus-
made it elear dtar sectarian violence did nor just emanare fron1 residuos of
rained arcaeks on Muslim,.
irracional heliefs among the ordinary masses, as eonventional wisdom liad a
These eircuinsrances gane risa co a ranga of rumors und eouspicacy theo-
but r ' - • Ice of modernir md crtoitatlism Dur-
ries. To most of my intornlauts ¡ti che Muslim ueighhorhoods m cu u:,i
ing che riors and thcir atiermath . [hese denied but immensely poweihd force
Mumbai, oidor imaginings of che upper echelons of scate and che Concess
und dynamies aequ¡red an unpceeedenred visibilíry. Prommenc citizens de-
Parre as cites of justice and protection -ave way to a radical sense oí isolation
plored Chis and bemoaned che dcmise of public order , tolerante , and cos-
and betrayal. In March 1993 a group of people, mainly Muslims affiliated with
mopolitanism in Mumbai ( e.g., Padgaonkar 1993 ). Scholars and activists
gangster organizations in Mumbai, organizad a series of bomb blasts, killing
saw the proliferation of erime and sectarian polities as an effect of the "lum-
hundreds of civilians on a single day, wrecking bus terminals as well as che
penization" of the ciry as irs industrial economy and its larga worlcing class
stock exchange in che city. A few weeks later, consignments of arms and
llave given way to real estate speculation and small service industries (e.g.,
explosivas werc recovered at severa) places along che Maharashtrian coas[.
Lele 1995).
Allegations of the involvement of che secret and independenc Pakiscani in-
My basie proposition in this essay , however, is thar [hese concerns can be
telligence unir Inter-Service Intelligenee (isi) in assisting local undcnvorld
read as symptoms of a wider anxiety regarding public order . What was ac
nerworks in Mumbai were immediately raised by che police as well as Hindu
stake during and after the riots in Mumbai was the very "myth of che state,"
nationalist and more moderare public figures.'
che imagination of che state as a distan [ but persistent guarantee of a certain
R,egardless of che factual complexities surrounding che bomb blasts, it
social order, a measure ofjustiee and protection from violence . The shattering
soon became a well-established popular truth rhat che blasts constituted a
of chis myth obviously appeared in different guises to different groups and
Muslim answer to militant Hindus, a message of "Don't mess with us" sent
communities . To che Muslinis in Mambai ( approximarely 17 percent of the
by the mythological Dawood Ibrahim, now elevated to che status of a stern
ciry's population ), who boro che brunt of police brutaliry and ethnic rage of
godfather figure. A Muslim female teacher from Nagpada echoed what 1
militant Hindus in both rounds of violence , che riors marked che culmination
found to be a widespread sentiment when in 1993 she told me, "We all felt
of a long process of political marginalization and everyday harassment by che
horrible during [hose Tour months [December 1992 to March 1993]-all over
ciry police and Hindu extremis [ political forees . For more [han a decade Mus-
you would hear These derogatory remarks about Muslims, you felt che hostiliry
lims had been the targets of a relentless stigmatization by militant Hindu
all over, in che trains, in shops, in my school. 1 retal) riding en a train when
nationalist organizations such as the Shiv Sena (Shivaji's Army). Aided by che
a group of Hindu women spotted me and starced talking quietly. One said,
police, che dominant Marathi-, Hindi -, and English - language newspapers in
`We Hindu women should also do something. Look at that Muslim woman
Mumbai had consolidated older images of che ciry ' s Muslim arcas as dens of
there-one should throw her off che train.' . . . Al¡ Chis scopped after che
dmg peddling , smugghng, and violence, peopled by clannish , fanatic, and
bomb blasts-nor because they aecepted us, but because they feared us."
hosrile Muslims paying allegiance ro dreaded Muslim mafia dons like che
Among the substantial seetions of che Hindu middle classes and che slum
Dubai-basad gangster king , Dawood Ibrahim.
dwellers who supported militant Hindu nationalism, che shattering oí thc
The anti-Muslim bias of che police force in tire city became more obvious
myth of che state appeared in an altogether different forro. Hete it was a
chan ever during che riors, when police offrcers issued orders of "sheet to
triumphalist sense of "teaching the Muslims a lesson," of overruling and
kiil" at Muslim demonstrators , while generally milder forros of riot control
defying the srate, of eelebrating an ethnic majoritarian justice opposed te
werc administered on Hindu crowds . During one weck of riors , killings, and

STATE MYTF1oLOGI1,S 1N MUMBAI 223

222 STATES OG IMAGINATION
 what Hindu nationalist leaders had decried as the state's "pampering" and
otherWhereas che state indeed is made manifest in an amorphous
protection of minorities. During [hese heated months in Mumbai it was as if
terrain of small arenas and local institutions, governance is not necessarily
earlier, more restrained and guarded modes of naming and talking abour
C'T,Q legitimare. The production of legitimacy, that is. che naturalization of power,
Muslims gave way to che most radical xenophobic fears and fantasies that
requires constant enactment of che state as a symbolic center ofthe
ciety, so
circulated from rickshaw drivers to respectable family docrors.
source of governance, the arbiter of conflicts, che Bite of authorization-as
The supreme dictatorial leader of Shiv Sena, Bal Thackeray, has made a
delegation of power as well as che right to "write society"-in law, constitu-
name for himself through colorful and provoking rhetoric, his radical abuse
tion, rules, certiflcates, and so on.
of Muslims and political opponenes, and his outspoken contempt for theju-
tt,r,» My argument is that chis public and performative dimension ofgovernance
dieiary and populist artacks on everything associated with sarkar (the statel
and p ' the si nificance of state spectacles, political rhetoric,
government; see Hansen 1996). Since the riots, the Shiv Sena has consoli-
C>Y o and processes of political identification as well as m che pertinence of "t Ie
dated its position year by year. In che elecdons in 1995 that brought Shiv Sena
Law " public le g alProcesses and so on
to power the state of Mah2rashtra, the parry c the polis in ^ J^
levels of corro Ernst Kantorowicz (1959) has shown with great subtlery how a legal-
the metropolis. In spite of massive ve evidence of increasing leves
p- pf° olitical theory of che "king's m o bodies" developed in medieval Britain.
tion , p cont empto en
or l egal processesf ` A
and democratic --^--Q
procedures, and
ere, political authority was const ucted as a dual structure: on rhe one hand,
Thackeray's celebration of bis own "remoce control" of the government, che
1 the notion of a sublime, infallible, and eternal body of che king (the law); on
parry's popularity was not affected for several years. Amid mounting evidence at ó i th e o th er, th e pro fane, h uman, an d fa ltibl e b ody o f che king (che giver m
of corruption at the highest leve) che parry once more swept che polis at che U `,^^
-• ru r`, aws). Kantorowicz quotes Blackstone writing en che sublime dimension of
municipal elections in 1997 in Mumbai and several other cides in che state ` -
U,+ royal authority: "[The King] ... is non only incapable of doing wrong, but
(see Hansen 1998). hN ^y/'.
uuug: u. uuu m
). 11 Ino fo ll y or wea k ness " (4).
Governance and the Imaginad State As Lefort points out, che efficiency of [his construction flows precisely from
\t jLthe separation and unity of these two bodies, fwwAáw-w cnmbination of che
The state is a narre given to various practices and institutions ofgovemment,
\iprbfane and che sublime in the eves of che subiecis: "It is the image of che
not only as an analytical concept but also as a locos of authority, howevcr 3 1 ^;¡
., natural body, che image of a God made flesh, che image of his marriage, his
elusive, invoked by and reproduced by an endless range of interventions- JQ p^\
` paterniry, his liaisons , his festivals , his amusements and his feasts , but also
from validating documents and checking motor vehicles to prohibiting cer-
the images of his weaknesses or even his cruelties , in short, all che images of
tain substances and encouraging certain forms of behavior that serve the
his humaniry, that people their imaginary, that assure them that che people
public health or interesa. The state, in other words, is an organizing concept
and che king are conjoined" (Lefort 1998: 245).
thr which people in Mumbai as well as in tth d rn societies imag
This union of che two bodies was laten reconfigured as che nation, the
me che cohe ts order and its institutions butalso its
peopie, or che leader took che place of che sublime-abstraer body and made
hidden secrets, ts sources of violence and evi
governance of che empirical and profane people possible in the narre of this
This idea of che state as the Bite of sovereignry and a symbolic center of
higher principie (Lefort 1g88: 254). Lefort argues that with che advent of
political will and power aboye parcial interesas, and thus embodying a univer-
democracy chis mythical and "original" source of power becomes radically
sality of justice, knowledge, reason, and authority that exceeds any single
empry, as it can be only temporarily occupied by representatives ofthe people,
representation (a parry, an ethnic group, a class) is rooted in Western history.
or the nation (17). What can be occupied is exactly that which is more per-
According to [his idea, che state musa govern but it musa also reproduce in
manent and enduring: che central institutions and symbols of che state.'
imaginary dimension that separares che actions of he state from rhnac ofany
1 suggest seeing che imagination of che state as marked by a deep and
e-
A. 1 t K4 CypN

224 STATES OF IMAGINATION ] _ / (/

STATE MYTHOLOGIES IN MUMBAL 225

t ,1', £ lti 4 ¿a - , , 74 4L
 politicians and film heroes ro sports temas, cultural evenrs, brand narres, and
constitutiva splir. On lile one hand irs "profane" dimensilis .. rhe incoher-
srvles of consumption sha red aeross die Icugth of India as vc l as acoss casta
enec, brutality. partialiry, and the banalirv of lile techuical sides oí gover- _
and class. 'I'his national, or ar leasr narionwidc. culmte has svsrematicallc
anote, as well as [he rough-and-mmble of negoriation, conipronise, and na-
been prontoted by die state, whose crucial role in producing a 11,111011,11 imagi-
ked self-interust displaved in local pn hits. On die other hand, lile "sublime"
uary hardly can be overestimared, as IChilnani (1997: 19) has pointed out.
qualities imputcd lo a more distant state: the opaciry of rhe secrets_and
jffhe political ficld abounds with religiou imIgere, and politicians invoke
knowl dge of [he huila' e cI ^In ilcsctc ny hrddeti res(2 u rces, d . o c,
precolonial aristoeranc splendor in dress lile, peec h an d re tbr to lile wisdom
mtd innnense poseer, und the illusions of hi>her lorms ot f 1 uliuíor stirr_
of religious tests when nccy arrcmpt ro represent die authorin of lile nation-
bel_eved ro-prtveil there. Thc rcpermire ofpublic performatives of rhe star ;dr
state. However, edueation, command of English, and competente in science
from stamps to military parades and imposing architecture-serves to con- v
and administration constitute equally, ir not more, powerful registeis of au-
solidare chis imagination of rhe state as an elcvated entiry. To paraphrase ,ve
thority and sublime qualitie The bureaucrat, rhe planner, and rhe scientist,
Durkhcim, rhe celebrations of rhe rationaliry and power ofthe state represen[
the member of rhe Indian Administrative Service-rhe heavily myrhologized
attempts tu make a society worship itself and its own social order.
"steel trame" of rhe state oecupy crucial positions in contemporary polio-
If we keep Chis dualiry of rhe state in mind we may see rhe full significante
¡cal imaginaries, not least of rhe large middle class. The bureaucrat vas until
ofwhy olitical e s anda encies of the state feltthe need to launch various
VJ recently rhe heno of modern India and until rhe 1970s was depicted in Hind
orar
initiatives in Mumbai to create mechanisms for reconet ranon or at eas re-
ja films as a man of characrer and insight.
habitarion,een Muslims and Hindus afrer the r'mis. I nigua Chat these
Ontil quite recently f is "modero nationalist aristoeracy"-lineages and
Initiatives Nave been une e o
familias of high-ranking bureaucrats, setentists. and politicians-were re-
rejganizing techda ,es nf ower
_ ate, but also by reconfigunng rhe legitr
^ ferred to with awe and respect. The mark of [hese ideal narional citizens rhat
t tac y and auth . in at retrievin a myth of rhe state with-
manned rhe bureaucraey was exactly their combination of moral integrity,
out which a democra tic state cannot govern, not ceca if this state is ea
commitment to rhe larger abstract nation, and deep technieal insights. The
by a government that nurmres rhe most antidemocratic form of majoritari-
authority of edueation, especially English edueation, remains crucial also
anism, as in rhe case of Shiv Sena.
Before proceeding ro the complexities of contemporary Mumbai, leí me among ordinary people, where it often generares more respect than wealth.
Many ordinary people in India still attribute considerable authority and sub-
briefly consider whether and how such a line of reasoning derivad from me-
dteval western pobncal thought can be made relevan[ to contemporary In dia. lime qualities to institutions such as courts, ro judges, to senior bureaucrats,
Vi and so on. This testifies to how effectively the modera nationalist elite in
Are there any "sublime" dimensions of rhe state and political authority in
India throughout chis century has made education, science, rhe rule of law,
modero India? Should one insread adopt a longue durée perspectiva and in-
and the role of rhe public sector mato coro signifiers of the modera nation.
quire moto how older registeis of kingship and rhe reladons between Brah-
Complementing older registeis of public conducís, rhis nationalist regisrer
mins and khatriyas (ruling or land-owning castas) are played ourtoday?°There
has evolved loto a complex web of public languages and political imaginarias
is no doubt rhat notions of honor, patronage, and rhe appropriate behavior
that shape discourses en the srate and boost rumors and stories of trans-
of rhe landowning aristoeracy arad dominan[ castas in various parís of India
gressions of rulos, corruprion, and abuse of governmental aurhorirv. It isthc
have shaped the construction of politics rhere in profound ways.s These cul-
cense of violation of rhe idealizad sublime qualities imputad to rhe state ti,
tural repertoires seem, however, insufficient to capture rhe meanings evoked
makes suc stories may we e that ordinary Indinas are
by rhe rerm sadcar in contemporary India.
less in awe of rhe state than a few decades ago, but rhe state is still regarded
Inspire of dceply segmented and compering notions of power, leadership,
as indispensable in tercos of public order and of recognizing communities,
and legitimacy in posrcolonial india, its public culmre has produced a larga
leaders, or claims as legitimare and authentic. Bureaucrats, judges, and offi-
reservoir of shared symbols, languages, and referentes-from war heroes ro

226 STATES OF IMAGINATION STATE MYr IIOLOGIES IN MOMBAI 227

 cers are called on every day to authenticate, inaugurate and uthorize in Referente to come forward and file affidavits before the Commission" (Sri-
brief, to act as transient incarnations of authori and s mbols of the state. krishna 1998:2) and also called on the police and the government to submit
These manifestations of the authority of the state may well be a structural their versions of the events in the ciry: 2,126 aidavits were filed, 2 from the
effect of governance, as TIm Mitchell (1999:9) contends, but they are also government, 549 by the police, and 1,575 by various individuals and organi-
imporran
t pub¡ e performances that peed to be studied in their own right.° zations. Afrer severa¡ attemprs at obstruction and delay from the state govern-
1 will illustrate the split representation of the Indian state as both sub ime ment, the report of the commission finally appeared in August igg8, five and
and profane, first, by analyzing the initiatives in the fieed of the judiciary, a half years afrer the riots rook place. It was lefr to the state government to
especially the construction and impact of the Srikrishna Commission in- decide whether the massive evidence gathered by the commission should be
quiring raro die riots and the bomb blasts in 1993 and the proceedings of made available te the advocate-general ro allow the state to institute criminal
the special court mandated under the Terrorist and Disruptive Activities Act prosecution.
(TADA) investigating the bomb blasts in March 1993 .
Second, 1 discuss how Public inquiries finto serious conflagrations, major policy failures, and dis-
the government maintains public order in central Mumbai and how the asters, such as famines and revolts, have been carried out in india since the
so-called mohalla ( neighborhood ) committees were set up by the police and
1870s. The inquests of the colonial state were normally carried out by civil
NGOS in these arcas in 1994 . 1 look at the rrajectory of these initiatives and servants who interviewed police officers, victims, witnesses, and others to
how they impacted en local perceptions of state and authoriry in the context
17
establish the factual circumstances and to apportion responsibility. In the
of the strained everyday encounters between police and Muslims in the city.
independent postcolonial state, these techniques of governmental self-
diagnostics eontinued but acquired new moral and potitical dimensions as

The Srikrishna Commission: Catharsis they became intensely occupied with rooting out harmful practiees and with
and the Polirics of Truth reforming sociery through reform of the state. As Visvanathan notes in bis
discussion of corruption inquiries in the r95os and ig6os, "Public adminis-
A few weeks afrer the riors in January 1993, the government of Maharashtra tration was (now) the honre science of the modera state. The state was not
decided to set up an Inquiry Commission headed by the high court judge, seca as something gargantean, a huge organism or a giant machine, but as
Justice Srikrishna. The massive evidente of open involvement of Shiv Sena something tentative" (1998: 15).
and other political parties in the violente, and of abuse of authoriry en the With the Shah Commission in 1977 probing loto the excesses of the Emer-
parí of the police, made it clear that to reestablish its authoriry in the eyes of gency imposed by Indira Gandhi, a new and more openly "cathartie" mode
the minorities, human rights organizations, and the broader public, the gov- of inquiry was created. The need was felt to clarify the extent of the damage
ernment had to demonstrate its eommitment to justiee. The commission be- caused by Gandhi's dictatorship as well as to redeem the apparatus of the
gan its work in June 1993 with the mandare to establish "the circumstances, state. The Shah Commission was more like a hearing, with submissions from
events and immediate causes of the incidents which occurred in the Bombay victims of excessive use of state powers, as well as responsible bureaucrats
Police Commissionerate arca in December 1992 on or afrer the 6th December and politicians. Everybody appearing before the commission was allowed to
r992, and again in January 1993, en or afrer the 6th January 1993." The com- be represented by legal counsel, and the commission sought in its style and
mission was further mandated to identify "individuals, or groups of individu- proceedings ro be as close as possible ro rhose of a courtroom.
als or any other organization" responsible for the riots, as well as to assess The Srikrishna Commission emerged more like a public tribunal: it was
the effectiveness of the Bombay Police in handling the situation (Srikrishna decided to make the proceedings public and to cal¡ on interested organiza-
1998: 1).' tions and parties to be represented through legal counsel along with the com-
The commission was supposed to work like a public hearing and initially mission's own official advocate. A number of organizations and parties were
callad "all persons having knowledge about facts touching upon the Terms of represented before the commission. The Shiv Sena, the Bharatiya Janata Party

228 STATES OF rMAGINATION STATE MYTHOLOGIES IN MUMBAJ 229

(0)1'; hoth of which ' ere ruling the stars during most of die tenure of the was brought in to refortn rhe Bombay Police. Satura liad heen higlily success-
commission), ncu A11-India Milli Council (a coordination group foca uuntbcr fui in prevcnting the massive riors in Bonihay from spreadiug to 1hane..:11
ot Muslim organizati one], ncC Jamivat-F-Ulcnw (Council ol Islamic Scltolars), industrial district north of die city, tltat liad seco dre worsr rioting in rhe
nce Comutunist Panty of India (unid 1995) aud a variety of human rightn country in ncc tg8os, and as a Sikh he was tcidely assumcd tu be neutral in
asocia[ Pons such as rhe L-awyeís' Collecrive and the Committee for the Pro- tic Hindu-Muslim conflicr. 'l his move and rhe wording of die mandare oftira
recibo of 1laman Righta, finco 1993 t0 199' have all heen repies ented at the conunission made it clcar m mune police ntficers that die tire Igoverniniii
hearings. The commission could order anp public servaut ro resrift before iL was reade ro put mosr of die responsibihrv on uhe police force. Ilundreds o?
whcreas ordinary citizens ippeared onli voluntaríle. Thc same colulttary prin- oflcers who liad heen in comnrand at police starions got base writing Ieogth^
ciple applied ro [hose elecred ro puhlic office, bur given rhe public namre oí affidavirs explaining their actons and perceptions during rhe riors.
the hearings there was pressure on public figures m actually appear and lee The two handfuls of such afhdavits 1 managed ro ger access ro all ran along
themselves be cross-examined. similar unes. They began with a long, often rosy account oí rhe careen oí rhe
The hearings began as systemaric inquines roto rhe events unfolding in the officer, his specifie distinetions, often pointing to long experience in handling
critieal period in rhe twenty-six affected police districts in the cite; laten, the "mobs" or crowds. One officer referred to his long experience at police sta-
focus was on background factors and the role of specifie public agencies, tions in Bombay's mil¡ district, where on many oceasions he liad dealt with
especially rhe police force. Based en independent investigation and affidavits demonstramrs and "violent mobs." Another wenr through an almost ten-
from citizens and police officers, rhe commission and rhe counselors repre- page account of all the demanding posrings he had been through-ranging
senting rhe interested parties called and cross-examined witnesses ofvarious from Dharavi, che biggest slum in Asia, to Agripada, described as one oí' the
sorts and loyalties. After rhe cross-examination of each witness, lustice Sri- most"notorious rrouble spots in Bombay"-and rendered a virtual catalogue
krishna drew his eonciusions coneerning the chata of events or cause-effect of situarions he had dealt with in [hese locations: mass meetings, Hindu-
relationship dealt with, and [hese brief summaries and conclusions, along Muslim violence afer ericker matches between India and Pakistan, religious
with a massive amount of written material, formed rhe bulle of rhe material processions turned into "rampaging mobs," and more.
summarized in the final report. Armed with these credentials, the officers then turned to descriptions of
Thc material rhe commission dealt with was intensely politicized, just as rhe areas they were poseed in during rhe riors. One officer described the
the forro and public nature of the hearings-intensely covered by rhe press- Muslim-dominated Dongri area in central Bombay as a dangerous and unpre-
often made them rhetorical platforms for rhe counselors of rhe different par- dictable place: "This localiry has a long history of communal riors [that have]
ties. The commission openly and self-eonsciously tried to extract, or ap- been occurring hete at frequent intervals siece 1893.... Though by and large
proximate, a factual truth from a series of political interpretations. The basic rhe residente are peaee-loving citizens, incidente of antinational characrer
raslc of establishing positively who did what, when, and how was often ex- committed by a few mischief mongers tend to cause sudden escalation of
ceedingly difficult and had often to rely en inferences rather than clear-cut tens ion. The arca has earned such a notorious reputation that rhe police ma-
evidence. chinery has to be alece round rhe year. However, it is not always possible to
Lee me offer a few glimpses into how material was presented before the predict how and to what extent a situation can deteriorare."
commission." The accounts then gave vivid and detailed descriptions of rhe events and
rhe actions of the officers, rhe orders issued, rhe number of rounds fired,
and so on. From [hese aceounts a picture of chaos, drama, and eonfusion
Reconstructing "Truth" alter the Fact
emerged: of a city exploding in what the officers described as random and
As rhe decision to ser up a commission of inquiry was raken by rhe state unintelligible violence; of looting, arson, and violente breaking out behind
governmcnt in February x993, Bombay's commissioner of police during the rhe police as they dealt with one situation, reports arrived oí police vehiclcs
riots, S. Bapar, was transferred and A. S. Satura, a highly respected officer, being stoned and artacked by rioters in new arcas and so forrh. These were

230 STATISS OF IMAGINATION STATE MYTHOLOGIES IN MUMBA1

 stories of a force not properly prepared for such situations and of fear of riots. Judging from my conversations with policemen, this was notan attempt
mobs that in che accounts from che Muslim parís of che city seemed intensely to exonerate themselves or their actions. None of the policemen I talked to
hostile to the police force.
denied he had shot dead or wounded rioters. To them, the shooting was in
In the cross-examinations most officers defended their actions and asserted self-defense, che only way to respond to the mobs they dreaded so much.
che need for public order. One maintained that when Hindus were ringing the The sense of being unjustly targeted while doing your job, of being hated by
temple bells on 6 December to celebrate che demolition of the mosque in che local people, of detesting [hose same local people, and of being stabbed
Ayodhya (che so-called ghanta naad [ringing of temple bells] promoted by Shiv in the back by political leaders was a common experience. In sume ways
Sena), "It was treated as religious activity exempted under che han order of it appeared to me to be the strongest bonding, a sort of negative esprit de
che Bombay Police Act." A senior inspector from che Nagpada police station corps.'
admitted that although he was aware that Shiv Sainiks (Shiv Sena activists) The bulk of the evidence presented to the commission was, however, affi-
made highly provocative speeches in conneetion with the ghanta naad, "Itdid davits submitted by victims and bereaved families-mostly Muslim-social
not then occur to me to take any action." workers, local organizations, journalists, and many others who did not share
An inspector was asked by che counsel for che Milli Council to justify his che police's perspective. There were harrowing accounts of the brutality and
calling Muslims "aggressors." He replied, "Muslims were aggressors because rage of groups of men attacking Muslims in streets and honres, attacks often
they came out in large numbers [and] they did resort m violence. The poliee led and organized by local Shiv Sainiks; and there were frightening accounts
had to cake effective action and che Muslims who were en the streets had to of the anti-Muslim bias of che police in their "cleanup" operations. A young
bear che brunt of che police action." According to [his officer, che questions man who worked in a Muslim-owned bakery recounted before the Commis-
of why events took place, who was shot, and so on were not relevan[ te polic- sion how "commandos in light uniforms and bulletproof vests entered the
ing. Their job was to restore public order, and those who gol in che way might building. I peeped out from my hiding and saw Samshad standing, two com-
gel shot.
mandos pointing their guns at him. He folded his hands and sat down near
The issue ofwhether che police used excessive force againstespeciallyMus- che commandos and pleaded that he was a Bhaiya (from Uttar Pradesh). One
lim rioters was discussed at length at the proceedings. A police officer de- of the commandos kept saying that Samshad was a Pakistani, the other said
fended che immediate use of "extreme force" (i.e., shooting directlyat rioters that he was Kashmiri.... 1 concealed myself again, then 1 heard firing and it
m kill): "1 do not think that it is always necessary to use graded force when became quiet. 1 saw Samshad writhing in pain, blood flowing out. Fíe said his
dealing with a situation of violence. II` extreme force is resorted te, at the very prayers for a minute or two, and then he was quiet."
Hrst stage, to pul down firmly che riots, I would consider it justified." It was, however, testimonies from leading police officers that elicited most
Affidavits from ordinary constables told a less orderly story, full of horror interest from journalists and others interested in che proceedings.[° V. Desh-
of advancing mobs: "The mob did not respond to our teargas ... en che mukh, former leader of the Special Branch (sB) in Bombay, was che first high-
contrary, they indulged in heavy stone throwing ... then someone fired a gun ranking police officer to appear before the commission. A well-spoken man
at us." An elderly constable related how he was attacked with a sword and with intellectual inclinations, Deshmukh appeared very humble that day in
how his colleagues withdrew finto che police post: "1 was then lefr alone in mid-February 1997 as he stood in his uniform in che dock in the spacious
che hands of che mob, 1 was terribly frightened ... someone attacked my face room in che Bombay High Court.
and neck with a sword." Deshmukh was more frank in his admission of che failures of SB [han
Most of che accounts by constables seemed tu reflect both fear and in- anyone had anticipated. He explained che lacé of intelligence work by the fact
comprehension, as if rioters were a sor[ of natural calamiry, displaying an chal he, as well as many others, "was led to believe that the governmentwould
aggression that che policemen did not seem to connect te their own status protect che mosque [in Ayodhya, demolished in December 1992 by Hindu
as police. Instead, many of them depicted themselves as che victims of rhe nationalists]." He said that he was well aware of Shiv Sena's capaciry for vio-

232 STATES
OF IMAGINATION STATE MYTHOLOGIES IN MUMBAI 233
 Deshmuldl's reacrion hete revealed dlat he subscribed ro the widespread no-
lenco that tito party "had incired hatred against the minorito communiry,'
tion of couunun.al violeuce as simple retributiva jusnce. wherc a killing on
dl id that "ntaha aarlti were srarted 1)y Shiv Sena in late Deeembcí 1992
cach sido caneels cach and malees the two cides ovel Thc grave tailure ot
with rho purpose of Iorcing thc minoriw conununity to gire up decir nano-
dre seo to monitor Shiv Sena eould, iu this view, be counrerbalanecd by ir>
[poner] on the strects. i le liso stated rhat sn l:ncw ale the central persons
equally serious leniency mwurd Muslim connnunal organizations. Une non-
of communal organizations in tbe ciry, bur chose ro do nothing , es he and
action malees another nonaerion plausible and permissible. Shiv Senas coun-
orher police officers were rcassured that nothing nould happen in Ayodhya:
sel pursued this lino in a subsequenr arrenlpt to cxonerato l lindu policenien
"1 had issessed that nrresting acrivists and leaders uf the Shiv Sena would
by clainling that particular Muslim pollee otilecrs had acred in a p ,rriai _ud
resulr in filrther couununal violeuce.... LI1e Iension Ihat developed in arcas
anti-I lindo manner during the riots. No conclusive evidente was produced
under Shiv Sena influence fbllowing a rumor that Bal Thackeray [Shiv Sena
and du allegations were dismissed by Srikrishna (Mid-Doy, April 5, 1997).
leader] would be arrested was nota factor rhat affected my assessment. It was
The pereeption that Muslim anger or "aggression" jusrified police brutaliry
based on what had happened on previous oecasions."
and later Hindu "retaliation" liso informed the following week's testimony
In bis cross - examination of Deshmukh , Shiv Senas counsel, Balakrishna
of Shreekant Bapar, who was commissioner of police in Bombay during the
Joshi, followed a course he had pursued throughout the hearings . Instead of
riots. Bapat was widely believed te be sympathetic to Shiv Sena and Bjr, and
challenging the evidente of Shiv Sena 's involvement (which would have been
had submitted a 175-page affidavit. Over a week he was cross-examined by
a fucile exercise ), Joshi focused almosr exclusively en alleged aggressions
five counselors, among Ibero Mr. Hudlikar, representing the police force.
and atraclcs en the police and Hindus by Muslims . During the riots, rumors
Hudlikar was generous in his questioning and gave Rapar ample time ro ex-
were rife thar Muslims collecred arras in mosques and that loudspeakers on
pand on whar he had stated in writing. Rapar was particularly adamant in his
mosques were used lo incite attacks on Hindus. But Deshmukh stated that sB
rejection of the charge of having an anti-Muslim bias: "According to me the
never recovered any weapon from mosques.
larger number of minoriry communiry casualties during December 1992 can
Questioned by che commission ' s counsel, Vijay Pradhan, Deshmukh ad-
be explained en the basis of che much greater aggression of the minoriry
mitted that he had not recommended any preventive arrests in early January
communiry mobs." Shortly after, Mr. Muchalla, the soft-spoken counsel for
although it was common knowledge that Shiv Sena was inciting its followers
the Milli Council, probed furrher finto Chis, and Bapar said angrily, "It is not
in ever growing maha aartis.
trae thar action against the minoriry communiry in Deceniber 1992 was
Pradhan: Were you aware of a closed door meeting for the entice Shiv wholly unjustified." Muchalla then confronted Rapar with the pollees own
Sena leadership on December 29, 1992? (Alleged to be che time when statistics, which showed that also in January, when Hindus led by Shiv Sainiks
Shiv Sena's subsequent attack en Muslims was planned.) were Ihe undispured aggressors, most of the victims of police firings were
Deshmukh: Yes Sir, we were aware of that meeting being held. Muslims. Bapar replied, "I eannot comment en rular."
Pradhan: Did you get information abour whar happened at that Then the eommission's own counsel mole over and asked Rapar why his
meeting? affidavir never mentioned the by rhen well-established role in the riots played
Deshmuldt : Wc were informed that the agenda eoncerned collecrion of by the Shiv Sena. Rapar replied, "If there is reference to Shiv Sena, it should
funds for riot victims. be there. if there is no reference, there is none." He conrinued: "The police
Judge Srikrishna : Would such an agenda in your opinion need closed are eoncerned with offense, not with polirical afñliarion."
doors? At this point, Justice Srikrishna lost patience and asked Rapar why the
Deshmukh : No. The ss received reports on December lo that masjids issues of involvement of organizations in the riots were omitted. Visibly
[mosques] maybe were used to instigare violenee, but issued no instruc- disturbed, Rapar assured that che police before che riots "had taken action
tions ni Chis regard . . . nor did it aer against nvo Muslims who gave against organizations known to be violent."
provocative speeches in November 1992.

STATE MYTHOLOGIES IN MUMBAI 235

234 STATES OF IMAGI NATION
 Srikrishna: If this was done, why is there no reference to organizations
in your af&davit? organizations indulged in illegal acts. Some of their members have done so"
(Mid-Day, April t2, 1997).
Bapat: We had no material at hand at this point in time indicating that
any organization was involved in the riots. Like other high-ranking police officers, Satura appeared to be rather con-
cerned with protecring the police force. In his deposition, he depicted the po-
Srikrishna: Were you not aware that Shiv Sena leaders claimed that their
lice as the protectors of society by conveniently blaming the riots on Dawood
volunteers had demolished the Babri Masjid? (A news ítem splashed
Ibrahim and other criminal networks and "land grabbers." These criminals
across front pages in most of the country a few days after the demo-
lition.) felt threatened after the crackdown on illegal constructions by the munici-
pality and rhe police force in the preceding year, and "they bit back by exploit-
Bapat: No, 1 was not aware of such statements being made.
ing the feelings of the people after the demolition of the Babri Masjid," Satura
Here the judge sighed, leaned back, and said, "Thank you, sir." stated (Mid-Day, April 16, 1997).
Yet, the proceedings of the commission inadvertenrly began to expose the
The Politics of Balancing Guilt complex links among political parties, the state, and die legal system. This
became particularly evident as the main perpetrator of violence, Shiv Sena,
The evasiveness and obvious irresponsibility of leading policemen appear- assumed office in the government of Maharashtra in 1995. The principal arca
ing before the commission, as well as the apparent lack of comprehension of contention between the government and the commission concerned the
among ordinary policemen of the social world they were supposed to police release of documents and files related to the inquiry. The commission had to
testified to how ill-equipped the police were to perform basic tasks of main- seek permission from the advocate-general for release or declassification of
taining public order. The interpretations of communal riots in terms of ap- documents. In some cases, Chis was refused en the basis of "interestof state"
portioning and "balancing" collective guilt and responsibilities among face- or other compelling reasons, but generally the state could not in principie
less and abstract communities thar emerged from the commission hearings claim privilege and refuse to hand over the required documents (according to
clearly reflected a dominan[ and widespread modaliry of politico-moral dis- the Public Inquiry Act). Jusriee Srikrishna could here use powers equivalen[
course. But ir did not produce material suited for prosecution of actual and to those of a High Courtjudge and demand the release of certain documents.
concrete crimes committed by individuals. As the former commissioner of After the coming to power of the Hindu nationalisr coalition government
police in Mumbai A. S. Satura stated emphatically a few days before he was in Maharashtra in March 1995, rwelve of more [han twenry cases pending
appointed, "Our penal code and our idea of justice revolver around the idea against Thackeray for violence-inciting rheroric were wirhdrawn or classified
that individuals commit crimes and are punished, whereas political parties as (¡.e., made a securiry question and hence neither object for public prosecution
a whole do politics. There might be individuals within these parties who com- nor available for scrutiny by the commission). After a protraeted legal tussle,
mit crimes, even leaders, but they must be punished as individuals.... What the state government agreed to hand over Tour cases to the commission."The
can we do to an organization? Ban it? That is difficult to do more permanently scrutiny revealed that the cases had been pending for a very long time without
in a democracy" (interview, February 18, 1993).
any agency daring to [alce action and that Sena leaders repeatedly had rried to
When appearing before the commission in April 1997, Satura reiterared threaten rhe government to withdraw [he cases. Police Commissioner Tyagi
Chis point of view, and declined to name particular organizations respon- stated that if rhe government decided to withdraw the cases "in the interest
sible for violence. When Justice Srikrishna pointed out that a range of orga- of communal peace and social justice," the police force world hace no objec-
nizations such as Shiv Sena, Vishwa Hindu Parishad, and various Muslim tions (Times of India, April 4 1997). It should be added thatafrer bis retiremenr
organizations were usted in the government of Maharashtra's "Guidelines to from the police force in 1997, Tyagi was nominated as candidate for Shiv Sena
Handle Communal Riots" issued in 1986, Satura said, "It is true that they in the 1998 general elections in India but failed ro win a seat.
were active in social activities, but it did not come te my notice that they as As Shiv Sena and irs ally BJe carne to power in Maharashtra in 1995 the
new government began a process of obstructing the course of the investiga-

236 STATES OF IMAGINATION

STATE MYTHOLOGIES IN MUMBAI 237
t,on. ?,s lile tmtiog evidenee puinred to Shiv Sena', crucial role in bode rounds designad lo combar Sikh militants in i>uujab). As mame as 1Sq persolis vele
ut violence as tvell as tu die mas} links between the parte and rhe pol ice l(), ce, accused of complicity in rhe conspiracv. Most of these derainees were sub-
a was decided on )situar' 23, tg96. to dissolve lhe Sríloishna commission. jected ro rhe most Itumiliaring and brutal Ibrms ut interrugation and onh
Alrhough in a striet serse tire state liad legitimare power to Iissulve tic roer few o( rhem were granted baila Fven rhose who were accused uf pla}i«g minor
mission and Shiv Sena liad sworn to do so during tira elecrion emnpaign in roles in conneerion with dte arras consiguntenrs on dm coast ufMaharashna
'995, thete were few precede nts, pone of tvldch involved such blatant and in February 1993 were kept imprisoned fbr more than evo ^ears befo re rhe'
nansparear bid faith as in this case. were released becausu of laek of evideuee..A brand ncw high-sean iry.. alas
The issue of Iegitimate audturip was ti srake hese. Could die stare actually courr building was erected in Mmnbai, sud tire police displaced unusual dili-
sustain a credible aura uf impartialiry if it exercised ¡ir such an unmitigated gence in producing and gathering a massive amount of evidenee in rhe cases.
manner a clear ethniclmajoritarian forro of justice? In tire ensuing debate Before the Central Burcau of Invesrigations in Delhi took over rhe case, tire
intellecmals, activists, and political figures argued that the commission had Bombay Poliee had charged all the accused with une of rhe most serious of-
to be reinstated to support tire process of consolidating intercommuniry Tenses in the Indian Penal Code, "the waging of war against the state." These
peace and harmony, whereas legal questions of justice and prosecution of rhe charges were withdrawn only after a much-delayed intervention of rhe arror-
perpetrators of violence placed a very minor role. Under pressure from the ney general (Visvanathan 1998: 127-28). Most of rhe evidenee gathered by tire
central government it was decided to restore the commission but wirh an Bombay Poliee that indicted this larga number of people was either irrelevant
expanded mandare. It was now decided that the commission should include or of low quality. By contrasr, eorrupt officials in police and cusroms depart-
die bomb blasts in March 1993 "to give a clearer and more comprebensive ments who had made possible tire imporr of advanced explosives into Mum-
picture of the parteros of violence and civil disturbances in the ciry," as it was bai had non even been questioned.'°
argued by tire advocate-general (Indion Express, May 29, 1996)." The inclusion The TADA court proceeded ata brisk pace. It was closed lo rhe public be-
of tire bomb blasts was a rather transparent attempt te deflect the course of cause of rhe allegedly sensitiva character of tire evidenee presented there.`
the investigation but nonetheless enjoyed considerable support in tbe public Prosecution targeted individuals and rhe Shiv Sena government srated explic-
debates following the nieve. This reflected, once more, that the dominant, irly that it wanted to speed up the process against what ir called "Muslim
and official, interpretation of the riots as irrational excesses spontaneously gangsters" responsible for the blasts. Some of rhe accused were sentenced to
committed equally by faceless Hindu and Muslim communities (neither as ten and twenty years of imprisonment. A number of rhe key accused, in one
organizations nor as individuals) in extraordinary situations enjoyed wide- way or the other supposedly connected with Dawood Ibrahim, are still in hid-
spread currency. The government's decision actually authorized the formula ing. Although the case is nor yet fully concluded, it appears unlikely, given the
of "balanced and equally apportioned guilt": that every morder by Hindus strong political pressure on the TADA court, that those sentenced can expect
eould be morally neutralized by demonstrating a corresponding atrocity com- a new trial or that the government will concede that many nave been convicted
mitted by Muslims. The logic of "retributive justice" that Shiv Sena's counsel on questionable grounds.
and the Shiv Sena-led government had been at pains to establish to exonerate
itsclf seemed slowly to impose itself on the commission's work.
Diagnosties, Prescriptions, and State Spectacles:
There was a glaring contrast between the slow, contected, and in many
The Report of the Srikrishna Commission
ways aeademic character of the work of the Srikrishna Commission, and tbe
simultaneous prosecution of those accused of responsibility for the bomb The status and aurhoriry of a report of a commission of inquiry is always
blasts in March 1993. In rhe weeks following the blasts hundreds of people, precarious. It is not just an expert's reporr or a piece of research commis-
mainly Muslims, were rounded up and detained under the stringent Terrorist sioned by rhe government. Whether the government agrees with rhe conclu-
and Disruptive Activities Act (TARA, in antiterrorist law passed in the 198os sions or nor, ir is inevirably an authoritative statement en the matter under

238 STATES OF IMAGINATION STATE MYTHOLOCIES IN MUMBAI 239

 scruriny. However, the government is not compelled by law to implement or who sought to extract their particular and interested, and intensely politi-
even accept the findings and recommendations of the commission. More cized, truth-claims from the stream of witnesses passing through, while che
clearly than in any other commission 1 know of in India, the complicated and judge, positioned three feetabove the res[, concluded each cross-examination
contested trajectory of che Srikrishna Commission had inade this inquiryinto with a summing up supposed ro extraer the reasonable and the facrually plau-
a spectacular clash of different notions of the state: on one side, the decenr sible-the negotiated truth-from the maze of interpretations before him.
high court judge, supported by human rights activists and larga sections of The final report from Justice Srikrishna adopts a medicalized language. As
the press, defending the idea of che state as impartial, aboye society, and a medica( doctor, Srikrishna diagnoses and prescribes a possible cure for
committed to a universal form of justice; en the other sida, the administrativa what he tercos che "communal malady": "Communal riots, the base of this
machinery and the police employing their armory of techniques ro delay, ob- country, are like incurable epileptic seizures, whose symptoms, though dor-
struct, or influence the inquiry ro protect their own "secrets" and cover up mant over a period of time, manifest themselves again and again. Measures
their failings; and the political (orces governing the state, commirted to a of various kinds suggesred from time to time dealt with symptoms and acted
majoritarian notion of "retributiva justice," bending and threatening the ad- as palliatives without effecting a permanent cure of the malaise." The judge
ministration ro serve their ends. Although public in its form, this debate en argues further that until a complete change in social outlook and the leve) of
che authoriry of state power remained in significant ways interna) te the state, education, "communal riots must be treated, perhaps, as an incurable dis-
in terms of its mandato, resources, methodology, and authoriry. It was not ease whose prognosis calls for suitable measures to contain its evil effects"
a citizen's tribunal or an independent investigation putting the state as such (1998: 4).
en trial."' In keeping with chis diagnosis che judge is brief in bis examination of the
1 argue cha[ both the Srikrishna Commission and the TARA tour[ should causes behind the riots, admitting that they grow out of complex dynamics
be seen as state spectacles, earefully crafted public manifestations of che state of demography, class, and political discourses. Bombay had che unfortunare
as a producer of imparcial and universal justice. At the same time, [he very combination of an increasingly impoverished and isolated Muslim commu-
simulcaneiry of che two proceedings, one indicting Hindu organizations and niry and a set of very aggressive Hindu organizations (Srikrishna 1998: 25-
the police, rhe other "Muslim goondas [criminals]," represented in itself the 29). Srikrishna is emphatic in his statemenr that "[no] known Muslim indi-
discourse of "retributiva justice." Both of these proceedings were marked viduals or organizations were responsible for the riots," and equally emphatic
by [he crucial dualiry inherent in the representaflon of the state. The Sri- in pointing te che responsibiliry of Shiv Sena, not as an organization per se,
krishna Commission revealed numerous examples of the profane sides of but as "che attitudes of Shiv Sena leaders [as reflecred in] its doctrine of `re-
state power: the deplorable quality of policing and police intelligence; the taliation' were responsible for che vigilantism of Shiv Sainiks" (30).
partial, biased, and brutal conduct of the police force; and series of blatant Srikrishna regards communal riors as incurable, and he notes in a more
attempts en the part of che government ro obstruct the proceedings and to poetic vein that because "che beast in man keeps straining at che leash to jump
prevent powerful political figures such as Bal Thackeray from being prose- out" (1998: 63), effective measures are of paramount importante. The most
cuted. At the same time, the very existence of che commission, the tenaciry interesting part of the report is che judge's diagnosis of the structure and
and integrity of Justice Srikrishna, and the public exposure of misconduct, shorteomings of the police force in Mumbai and che string of recommenda-
corruption, and liaisons between politicians and the police force also pro- tions he makes to remedy these (31-62). Srikrishna paints a gloomy picture
vided a site for a certain process of public catharsis. of a complacen[, biased, disorganized force where even che most everyday
In Chis capaciry che commission has become a symbol of the resilience of routines of filing cases, physical training, and discipline are ineoherent if no[
a hi her rm of usuta, and chas a sigo of che perma- absent. Returning te che diagnostic moda, he sums up, "Despite knowledge
nence of thelubllme dimensinn nf thp sratp, This dualiry was inscribed in the of che fact that che force had been infecred by communal virus, no effective
choreography of the proceedings: before [he bench a string of counselors curative sreps were taken over a large period of time as a result of which,

240 STATES OF IMAGI NATION STATE MYTHOLOGIES IN MUMOAI 241

eonununal violenee became eltronic and as virulent symptonis showed up terrible criare-cuco Muslints admir thar. AVe should not be soft on nccnl... .
Beside , rhere are so many things we are rever told. 1,he govern nt cnr ha, a
during rhe rvo ron periods" (35).
¡he estire tono of che report mainrains chis neasured distante to che in- lot of inlormation it cunnot diselosc°
1erested parties and gives critieal and considered suuuuaries of che events ¡he deep divide betwecn che social worlds of Lindos and Musiinls cien

uufolding around each police scacion. 1'he judge examines che depositions todav traverses che city seenis ro heme al1deced non orle' rhe serse of public

by leading police offieers and political leaders. former ministers. Keading justice in rather profound 'vaya but also che public interese in diese mvo pro-

through these pagos one gecs an ever elearer pieture of che dtcory of the scale ceedings. Many educated people from all annnumicies llave calcen a Leen in-

to which Srikrishna subscribes. The tope is not legalistic but moral. He is terest in che Srikrishna Commission. A string ot independent repoets tnd

highly critieal of nonadherenee ro rules and regulations in the police force documencation of suf erings and abuses Nave emerged from NGOs and civil

but oven more critical of che ostensible lacé of committnent and lack of rights activists. Many ordinary Hindus, according to my impressions from

a moral oudook he detects in leading police officers and in many political discussions and interviews over che past Tour years, seem nonetheless to ap-
preve of che formula of balanced apportioning of guilt and responsibility for
figures.
Srikrishna's view is clearly that a sense of dury toward the nation and an che riots. II is a conveoient nonlegalistic framework that enables che ordinary

ethical view of life must be the basis of che representatives of the state. The citizen to bracket these events as events without actors and asan unfortunate

formula seems tú be "che higher the tank, the deeper the commitment," aberration from che normal order of things. 'lo most ordinary people 1 met,

a formula thac resonares with the dominant discourse of che postcolonial che commission of inquiry appeared as a somewhat inconsequential siga of
nation-state in India. The idea of che state as a moral entiry, once again enun- "che state," a manifestation of authoriry that simply was expected to restore
ciated in an official report, remains exactly ¡es most unattainable and, diere- che public order cha[ liad been upset by che riots. Sur che rh tnrir nfrhe trae

fore, most precious and sublime dimension. asa ratoraL ^t"ty as well as che legal intricacies of che proceedi s, were

In the case of che TADA court, che representation of che state was slightly mainly direered at the educated, lirerate middle creas, which always was che
different. Here, the rhetoric of secrecy and the practices of classifying even primary consntuency and coneern of te Qosic Im ial atete.
che most banal piece of evidente in the supposed interest of rhe state contrib- Among ordinary Muslims in che parís of central Mumbai 1 got to know, the

uted to creare a sense of urgeney that sonrething larger threatening che nation commission figured less prominently, althongh Srikrishna was praised as a
was at stake. The Mumbai police eagerly projected che enormous material it "secular person," in che cense of being highly educated, imparcial, and criti-
had gatbered to generare che same illusion of che effectiveness and ubiquity cal of Shiv Sena. The Muslim social world in Mumbai is non only spatially
of rhe state's knowledge and capaciry for taking en public enemies. The pro- separated from chal of Hindus, but is also demarcated by che existente of a
fane dimensions were equally obviotts in che brutaliry and partialiry of che local Urdu public sphere consisting of newspapers, journals, and local cable
police investigarions and in che harshness of their treatment of che detainees. TV. '1'o che average Hindu conversant in Marathi and Hindi, Chis world ap-
However, che serious character of che crime, che alleged connections of che pears elosed and even threatening and che Urdu press is routinely accused of

"Muslim gangsters" to Pakistan, and che secrecy surrounding che case meant spreading anti-Hindu propaganda. Such aeeusacions were also presented to

that these obvious abuses and human rights violations never generated rhe che Srikrishna Commission but were never substantiated. Rather than being

kind of public concern and debate [han Nave surrounded che Srikrishna pro- vehicles of sectarian ideology, che Urdu press seems, however, to be strongly

ceedings, especially in che English-language press catering to an educated introverted and preoceupied by issues interna) to the Muslim communiry. It

middle-class audience. In a conversation in 1997, a liberal Hindu business- was unsurprising, therefore, that che TADA case was auributed more impor-

man expressed quite succincdy to me how che scale of knowledge and vio- tance and concern than che proceedings of che Srikrishna Coinmission. Mus-

lence at che disposal of che state acquire sublime dimensions: "Set, many of lim organizations documenced che harsh neatment of che predominantly

che accused in Chis case are well-known criminals. They Nave commitced a Muslim detainces under TASA, which became a symbol of che inherently anci-

242 STATES OF IMAGINATION STATE MST IIOLOGIES IN MUMIIAI 243

 Muslim bias of the state. Calls for opposition to the case, and to dismantle
wouldn't object to os having in image of Virgin Mary in one corner of the
the TASA laws altogether, have for some years been high en the agenda of
compound."
local organizations and of severa) Urdu newspapers, and are a point often
Talking to the policemen 1 gota clear sense that the "sociology of the hoo-
used in campaigns by local Muslim politicians. As we shall ser, the TASA
ligan" that carlier had informed police work in Bombay (Chandavarkar 1998:
proceedings resonare with a long tradition of enmiry between Muslims and
180-233) now in the Muslim neighborhoods liad evolved roto a cerrain sim-
the police in the popular neighborhoods in central Mumbai.
plified "sociology of Muslims." T sed on n
admixture of stereorvpes, rumors, and circulating stories mane of them
State Speetacles and Politics in the Mohalla modifiedersions of [hose circulating in the neighborhoods Consider the
following example based en the stereorype that Muslims divorce their wives
The position of the Muslim communiry in Mumbai since Independence has
all the time. An officer explained, "One reason for the high level of crime is
been marked by an increasing leve) of socioeeonomic marginalization (Han-
all [hese young boys who grow up without a proper father-chis is because
sen 1997b). Moreover, Muslim areas have more police stations and chowlcis
[here are so many divorces and the men just leave their families behind."
(police posts on strategic street corners built of stone and, as in rhe colonial
However, anyone with a rudimentary knowledge of Muslim Personal Law
period, equipped ro be barricaded and comed roto a bunkerlike structure)
will know that aceording to this legal complex, children belong to the father
[han other arcas in the ciry. The Muslim arcas are clearly treated as seeurity
and even peor fathers would never abandon their sons (though maybe their
problems, as dens of crime and mafia activiry, in other words, arcas where
daughters), who are crucial to their social standing and future life prospects.
the "police machinery must be alert around the year," as police officers put it
The following off-the-record statement by a young and inexperienced officer
before the Srikrishna Commission. In spite of al] assurances of the change
testifies to the enormous gap benveen the social worlds of the predominantly
of mind of the police afrer the 1992-1993 riots, the infrastrucmre of rhe po-
Hindu police and that of the Muslims in [hese arcas: "In the beginningwhen
lice in [hese areas expanded considerably afrer thc riots. The practice of mak-
1 carne here, I was nervous when we went en patrol, especially at night. This
ing "preventiv;gr s" of what the police termed "notorious characters" or
hostiliry was somerhing 1 never experienced before ... but then afrer some
-
just "rowdy yo men" prior to festivals, elections, and so en was wide-
time 1 started to look them right in the eyes and pretend that I was indif-
spread and widely accepted as legitimate." It is no exaggeration to say that
ferent to them. 1 also learned more of the dirry language they use around
for young Muslims in [hese areas, the police force was an ever present and
liere ... that helped a lot. Now 1 get answers te my questions and 1 feel more
dreaded representation of a hostile state.
respected."
The police force in Mumbai is overwhelmingly Hindu and recruited from
Generally, postings at [hese stations were considered strenuous, full of
the same social environments and caste communities wherein Shiv Senas
hard work and dangers but also considerable rewards and bribes from the
masculine Hindu chauvinism has emerged. In the predominantly Muslim
brisk flesh trace and drug economy in parís of central Mumbai. As a rule,
arcas in Mumbai where 1 worked, 1 frequently visited two police stations and
officers are rarely posted more [han twelve to eighteen months in one police
met officers and constables [here as well as in the chowkis. In each of the
station, whereas constables rypically serve nvo to three years or more. The
compounds of what was supposed to be the strong arm of the secular state,
police depends vitally, therefore, en its network of informers in the neigh-
one found two or [bree small temples devoted to Ganesh or Hanuman (the
borhood, nerworks created and maintained through fiows of hafta (literally,
monkey god associated with courage and fighting spirit). These had been
"week," i.e., regular bribes or payoffs) and other economic transactions.
financed by donations from officers and constables and had been constructed
The depositions before the commission also revealed some of the many
within the past decade. "Sorne of us have questioned whether it is appro-
ties berween the police force and local political organizations, particularly the
priate," an officer from a Christian background told me, "but my superior
Shiv Sena. Officers posted at local stations generally tried to protect the police
simply replied that if there were more Christian officers and constables, he
force by keeping the top political level content and slightly misinformed. It

244 STATES OF IMAGINATION

STATE MYTHOLOGIES IN MUMBAI 245
 tiared rhe forination of utohalla committees at et'ery police station in rhe so-
secmed ol paramouut importante te) prevenr am disfigurarion ot rhe intricare
ealled problem ,creas. whicb almost e.^dusiveh happeued ro he arcas wirh a
w e b of halt a atid racit undersraudi ngs among local operarors, huilders, and
suhstantial Muslim populariou. Thc initiatives had nntnc arliniries with simi-
local stronglile n polinrians ('Lid ti s), on which the daily worltiug of die police
lar reehniques of governance enlployed oven rhe pasr eentune: bodies ofcoc-
depcuded. If nc¡ s web were brokeo up t 111o ugh mass ice arrests, ti] e police
cerned and "respectable" cirizens from all conanlunities in a ucighborhood
would llave lo work nmch harder, bereft of dae local nenvorks of intbrmcrs
viere called on to assumc res ponsib il i ry. ro calm rempers, and ro assist the
and "helpers."
police in raking preventivo aetion. As imporrant, rhe committees aimecl ar
[hese netoorks were Huid_ in ti] ti di ni en si o ual, interwoven with orbe r uet-
''recrcating confidente in out instimtions and in out democracy among, rhe
worles, cut across lines of Gaste and religion, and operared on rhe basis ot
Muslims in Chis city,'' as une of the driving torees behind nhe initiative ex-
rumors and gossip. Tú claim that someone was a police informer or paid by
pressed it." When rhe mohalla committees were set up in 1994 they nlainly
rhe police was a trivial and well-Irnown part of tacit warfare among networks.
recruited members from rhe Muslim middle classes. Many of the niembers
The status of "informer" was never positivo knowledge, and the claims by
were )clown in the locality as respected figures and often viere involved in
policemen that someone "works with us" were ofren as strategic and (luid as
voluntary work and used ro being in Glose contact with institutions of the
the rumors floating around in rhe biraderi networks,'" 1 got to meet only the
state. As a young progressive advocare known for her controversia) supporr
"publie" friends of rhe police, thar is, those who trade openlywith policemen,
to divorced Muslim women and active in rhe committee in Nagpada stated,
drinlc with them, 'fix" various things for rhe officers, and walk in and out of
`°Ihe fear of the intentions of the police was rhe biggest problem, and then
rhe police station. Many of these helpers were "marginal men," at rhe mar-
tire fear of attending meetings inside rhe police compound irself. Only edu-
gins of rhe powerful nenvorlcs, small traders, often with criminal records or
cated people who knew they enjoyed smne respect among eonstables and
from low-status families.='
officers were willing to do rhat in 1994. You can imagine how rhe atanosphere
In 1994, mohalla (neighborhood) committees were set up throughout the
was at that time.""
arcas affeeted during the riots to promete reconciliation in the city as well as
Among police offcers it was broadly assumed that rhe Muslims, especially
to facilitare the future governance of rhe urban territories in central Mumbai.
the peor and rhe uneducated badmash (hooligan), constituted rhe main prob-
Afrer the Bliiwandi-Thane riots in 1984 that also affected large parts of Bom-
lem. In rhe police analysis, riots started when such people were incited and
bay, peace committees were formed in a number of mixed neighborhoods in
manipulated by local political leaders and their imanas (priests). The police
Bombay, such as Mahim, Bandra, and Byculla."' However, over the years they
saw rhe committees as a way to "depoliticize" and contain communalism,
evolved loto platforms for local politicians who saw the committees as en
that is, to reduce it to occasional ourbursts of irracional social behavior and
opporrunity for forging ties with rhe police as well as strengthening their
to reduce rhe element of "political manipulation" [llar standard common
posirion in rhe neighborhood. None of [hese committees played any role in
sense among police officers (and many social scientists) in India hold to be
preventing violente in 1992-1993, and all were dissolved quickly after.
rhe main reason behind riots. Members of political parties and politicians
Police Comm issioner Sahani, known for his incl ination toward what one
elected in rhe arca were not admitred roto rhe committees. Based en their
officer disapprovingly described to me as "intellectual policing," launched a
carlier experience, rhe police wanted to remain firmly in control of [hese com-
new reconciliation scheme after a series of unprecedented initiatives: police
mittees, to kcep polities away "in order to curb rhe divisive effects of partisan
officers were told to cooperare closely with social activists, to attend long
interesas," as a high-ranking officer put it. The objective was more ambitious
sessions where riot victims gave detailed and moving accounts of their loss
Chis time around, for, as he said, "We want to creare a new leadership among
of children and spouses at thc hands of the police; police officers were made
Muslims." The police wanted to bypass and exclude rhe established brol<ers
to sir through week-long courses on Islam and Muslim culture conducted by
and dadas (strongmcn) in rhe mohalla and instead create a representation of
people sueh as Asghar Al¡ Engineer, a well-known activist and vocal critic of
Muslims by "civilizad" cirizens, that is, Muslims who, according to rhe stan-
the police force.
dard assumprion so central tú governarice in India for a century, by virtue uf
Thc action plan that was implemented in 1994 entailed thar the police ini-

STATE MYTHOLOGIES IN MUMBAI 247

246 STATES OF IMAGINATION
 their education had abandoned primitive beliefs and had become amenable commenced in lanuary 1997, a large number of those recruited by the police
to reason and persuasion. Through these representatives the entire commu- as nonpolitical activists tried to convert their newfound visibility and public
nity could be addressed and governed, it was assumed.. standing into a bid for a political career . As a consequence , a large number of
In their initial phase the meetings were tense and serious, committee mem- committee members had, strictly speaking , to be excluded because of their
bers told me. A retired judge who served en a mohalla committee related, "pollution" by the political world. This was of little consequence , however, as
"In the beginning, all the top officers from the station were present at the many political figures already had begun to attend the meetings after the en-
meetings. On Fridays when the streets were full of people assembling for the largement of the committees . On severa ( occasions prominent political fig-
Friday namaaz, we would all come out with the officers and stand around ures not only attended meetings , but even began to preside over functions
the crowd, very alert, watching passers-hy and making sure that no one organized by the mohalla committees . Very few of the politically involved ac-
made any provoking moves. There were tense moments, but 1 think we were tually left the mohalla committees , and Che still more infrequent meetings of
successful."
the committees at the police stations began to resemble public functions,
As the political attention faded in 1995-1996, the committees were subtly often Iavishly hosted by the helpers and friends of the police.
transformed. On the initiative of the new and flamboyant police commis- Another consequence of the expansion of the committees was that certain
sioner, Tyagi, the committees were enlarged from a maximum of fifty mem` police stations began to assume new functions of brokerage and "fixing" of
bers per police station to as many as two hundred. Many of these were the local problems , paralleling those of the local politicians they had sought to
helpers used by the police to gather information in the neighborhoods. They marginalize . An officer at the Agripada police station told me enthusiastically
d es i re d some recognition and standing in the community and the police re-
Ú' hbout bis newfound role as "fixer": "Now many people cometo us with their
warded their loyalty by conferring en them official status as community rep- usual problems-sewage, water, telephone connections, school admissions.
resentatives . This expansion and inclusion of larger groups people loto the For us it is very easy to solve-we just make a few phone calls. When I present
committees had severa( effects. First, the social prestige ofsitting on the com- myself to diese lazy bureaucrats at the municipal corporation, things start to
mittees was immediately reduced as people like Y. were included in a mohalla happen ... [nodding toward a line of people waiting in the compound]. So,
committee. Y. was a rag merchant who had become rich on buying from and as you can see with your own eyes, people have gained confidence in us. They
exploiting the rag pickers in the crea. He was a big, hefty man known for bis can see that we actually solve their problems."
violent temper and the long whip he carried when ordering his many workers The mohalla committees and the néwlssertive friendliness of the police
around. He was feared but not respected. Another new face was A., a man have indeed reduced the level of tension in [hese arcas, but they have removed
known as a supplier of all kinds of goods to the police. He walked in and out neither the mechanisms producing communal enmity nor the organizational
of the station, was always excessively servile to commanding officers, and was structures that perpetrate this violence. It seems as if the mohalla eommit-
always joking with the constables. He had a small office in the building op- ces Uju,c vided the police with a set of new techniques for keeping or-
posite Che police compound, with nothing but a table, a telephone, and a der in "trouble spots" through a networ o un erwor oneratoxs liaisons
chair. From there he could fix anything, he boasted: "Just tell me, you are my with political figures, direct mtervention in the distribution of public services,
friend, 1 will get it for you." S. was another new member. He was a lecturer at and more.
a local college and liad developed political ambitions. He had bcen politically
active in the Muslim League, later in Congress. "Now, 1 am a social worlcer. 1
Whither the State?
work only for the community," he said. S. always moved around with a couple
of whar he claimed to be "students," muscular young men who answered his The many continuities in terms of governing the badmash in Mumbai over
cell phone, brought him cold drinks, and drove bis car. the past century demonstrate how the dominant governmentaliry of the post-
As the campaign for the elections for the municipal eorporation in Mumbai colonial state in India continues to represent itself as a locos of a higher ratio-

248 STATES OF IMAGINATION STATE MYTFIOLOGIES IN MUMBAI 249

 ,aliry outside ami aboye rhe complexities and irrationalities of the lives of rhe represen[. The prevalence of such political imaginarios did malee Shn° Sena's
uusses. odarheSrikishnnComenssion; mtlthcmohallacommitteeswerein- style of governance possible te sustain. On rhe one hand, rhe p trty cele-
_ . In both brared its conquest of rhe stare governmenr on behalf of rhe majority of
n V 1 LJtses,
bli - t leA, ion
s u mr of a are Etintess reisonablenese tol-
the
m ns ordinanv Hindus and enacted rhis in gran(¡ populist spectacles and equalk
a LI ' trance, and justito-viere represented ro th ' re ^ ° audien, e z4^- grand proa) ses of employment or tree housing tu the sluan dwellels On rhe
^rp ( educated middle clase. otlter hand, U-om the top leadership to the local shnkha pra_ ukh (leader of i
As relativo peace llegan to prevail in central Mumbai, middle-class soci- local branch) there was ar rhe same time an indulgente in corrupr piactices,
ety--rhe high-ranking officers, rhe educated, and rhe Ictivists-withdrev< real estate speculatiou, and more necia criminal actiwq unprecedcnted ecen
fiom rhe mohalla commnrecs and more everyday and profane forros of gov- \, ) 1 in a ciry like Mumbai. Ibis enabled rhe party te extend complex nerworlcs olí
ernance and nerworking were reconstructe . lmilarly, rhe public spectacle o patronage, dependence, and alliances ar all levels in rhe city and the state.
rhe Srikrishna Commission has come te an end; only rhe report remains in • Moreover, Bal Thackeray' s constant assertion of his autonomy and his con-
1
libraries and in limited circulation among intellectuals and political activists. tempt for the judiciary and r h e S r ik r i s h na C omm i ss i o n remain central te his
Ironically, it fell en rhose most strongly indicted by the report, the Shiv Sena n populariry and his long-standing efforts te construct hiniself and his move-
and the Mumbai police, te decide whether and in what forro the recommen- ment as a site of authority that openly defies and challenges rhe authority of
dations of the conimission regarding reforms of the police should be imple- the stare. This obviously militares against the desire for social respectabiliry
mented. The political world in Mumbai appeared, in other words, te conduct that is equally strong among Shiv Sena's supporters and undoubtedly had a
"business as usual." In rhe elections for the State Legislative Assembly in negative inipact en Shiv Senas electoral formnes in rhe 1999 eleetions. But
September 1999, Shiv Sena los[ some of its former electoral strength also in the older myth of tbe state as the center of society appears nonetheless feebler
Mumbai, and a more centrrist coalition formed a new government. However, fiare. It is in icative o
Shiv Sena remained a force te reckon with in the ciry and the srate. As the many Muslims no longer look te rhe government but to Dawood Ibrahim and
new government in July 2000 declared it would prosecute Bal Thackeray for rhe associated myths of rhe power of rhe big-time dadas, or rhe mavericle
his role in the 1992-1993 riots, Shiv Sena leaders threatened that Mumbai lower-caste Samajwadi Party, for protection (see Hansen 2000).
would buen, and rhe police assessed, as so many times before, that arresting 1 suggested aboye that rhe attribution of sublime quahnes te rhe state is
Thackeray would not be in the interest of public order and communal har- linked te its capacity for violence. As older myths of the state and its mo-
mony. Shiv Sena remains today the most powerful organization in the ciry. nopoly en violence seem to crumble in Mumbai, competing myths of au-
These evenrs beg the question of whether the "srate spectacles" analyzed rhoriry and tales of fear cluster around the actual dispensers of violence in
aboye had been able ro resuscitate rhe myth of rhe state. The answer cannot rhe city: rhe Shiv Sena, rhe underworld, and the police. The most profound
be precise, but let me offer a few reflections by way of conclusion. paradox of this process of segmentation of authority, however, is that each
Thc hold of Shiv Sena en Mumbai and its five-year tenure in rhe state gov- segment remains dependent en the continued exisrence o tate-as a
ernment point to a transformation of the aura of the state as a site of neu- poolofresources as ' e itimateviolence, oras anordertobedefied
traliry and a certain predictability based en impersonal rules and technical and oooosed. Even while effective governanee crumbles and fragments, rhe
expertise that both rhe colonial and postcolonial stare painstakingly sought myth nf rhe unity and cnherence of rhe state must be kept alive.
te construct and maintain, not only in rerms of practices of government but
also in the popular political iniaginaries. Listening te the mazo of rumors and
tales of conspiracy that constitute an important parí of rhe popular debate en
Ñotes
government and state, one gets a sense of a widespread understanding of I anr indehted to Véronique Bénel, Jonathan Spencer, Jonathan Parry, Finn Sreppucat, Oskat
poliries as a gane and control of government institutions and eheir resources Vedcaaik, Chris Fuller, Vivek Narayanan, Daniel 1lenvitz, and orhers who provided usetul

as rhe prizes ro be won by parties and the communities they are believed te critiques and comntente en earlier drafts ur [his essay. An earlier and longer version of [his

250 STATES OF IMAGI NATION STATE MYTHOLOGIES IN MUMBAI 251

 essay appears in The Eueryday State and Society in Modera India, ed. Véronique Bénei and C. J.
own conduct, appeared lo llave loar moral authoriry over the citizens and appeared to evoke
Fullee (Delhi: Social Science Presa , zopo) 31-67. Unlese
otherwise nored, oil nanslations no feas ... The criminal clemente ,ere emboldened ro huri a crerde bomb ar tica Police
are mine.
Comnrissioner and ¡rack constables lo death without fear. The police developed a psycho-
1 happencd ro be present in the cay during tire note and on severa) ocasiona
observad po- logical fear about atracks en rhem" ( 1998: 34).
liremen literally protecting arsonists and mrning dneir back
on rampaging mobs. Similar 1o. 1 was formnare ro be able co anead most ufrhese hearings, and tire quores and observadoras
incidente all over che city were reporred widely by joumalisrs of che English- language presa,
¡ir che following are ah from my personal notes and non che official transcript.
especially Times of india.
ir. Masa prayer un sneets and foorparhs around temples invented and organized by Shiv Sena
2. The official report of che Srikrishna Commission stared that "
a granel conspiracy was and other radical Hindu organizations in Chis period lo mobilize Hindus to atrios agaierst
hamhed at che insrance of rhe notorious srnuggler Dawood Ibrahine
Kaskar, operating flora Muslime.
Dubai, to recriar and train young Muslime to vent their anger and wreak revenge by explod-
rz. The legal intricacies of chis mtpreesdented more are oudined by R. Padmanabhan in Prontline
ing bomba near vital installations and aleo in Hindu dominated arcas so as to engineer a
(April 18, 1997 ). An imporranr part of chis procesa was that rice High Court in Mumbai acm-
fresh bous of communal note " ( Srikrishna 1998 :
60). Justice Srikrishna fans to mention tire ally rejected the governmenr ' s request for withdrawal and furthermore sentenced Thackeray
vital role played by corrupt cusroms and police authorities in che entire operation. The
m a week of " simple imprisonment en che grounds of contempt of tour[" Thackeray ovas
compliciry of high-ranking ofHcials in ara operarion has recently peen docunrented by Shiv
torced to appear in che magistrales court in che suhurb of Bandra on February r7, where he
Visvanathan in Sethi and Visvanathan ( 1998: 118-28).
was released co ball. Meanwhile , che courc was surrounded by thousands of angry Shiv Sai-
3. Are artes rrng representati en at tire leve 1 of popular sulture oE
che nrythical qualities of chis niks shouting slogares and demanding rhe inmcediare rclease of tire Serrapati , che conrmander
"originar}" source of power appeared
some years ago in a TV comedy . The protagonist, ara of che army, as he is known among rhe rank and file (Mid-Day, February t8, 190-
Afeitan woman who has recendy moved ro tire Unired Srates, visits Capitol Hill and stops et
13, The ternes of referente of che commission were expanded to include ara ineesdgation ecf"rhe
che glass montre conraining che original American Constitution opened ro irs famous first
circuoestances and che immediate cause of che incidente comoeooly known as che serial
page , "We, che people ..."
She reads through che page, rerurns home , overevhelnred by Chis bomb-blasts ofrice rzrh March 1993, which occurred in che Bombay Police Commissionemte
revelation , and decides to cliange hcr life. Just
as Che domain of che sacred is fraught with arca" asid further wherher [hese were linked by commoo causes or "a common design" m
ambivalentes, bliss as well es horror, che sublime
dimension of che arate is dcperedent ora lis che riots investigated by che commission (Srikrishna 1998: 58).
darle ardes. The fascination of spies and secret cervices , from James
Bond ro Nileita m Men in 14. For an overview of che proeeedings of tire TADA court, see the arcade ")volee for Whom" in
Block m menfion some recent filnes, is premised ora che awe and fear of che brutality and
Humanscape ( Decembeug95).
ruthlessness of che arate. Tice atare appears sublime because of
ics inordinate power ro con- 15. The seerecy of che T ADA proeeedings has produced che interesting parados that che Sri-
done. redeein , and even purife violente, as René Girard (1977) has poinred
out in che context krishna Commission has encounrered many diffculties in fulfilling ics revised and expanded
ofrhe pertinente of violente and sacrifica to religioos rimals.
mandare because ir has beca impossible te gel access ro rhe many elassified documenrs used
4. The fireramre ora kingship in precolonial and colonial India is vas[
. Landmarks are Dumont in che TADA court.
(igSo), Dirks (1987); see aleo Cohn (1983) and Price ( 5996).
56. Independent human righns group organized che Indias People ' s Human Rights Connnissioo,
S. See a good discussion of conrinuities and breaks in forme of
arare ora rhe subconeinent in which shortly alter thc riots set up an inquiry headed by ovo retired judges of Ihe Bombay
Kaviraj (1997a).
High Court. Their report, 'Ihe Peoples Verdict. was published in Augusc 1993 and coneluded
6. Akhil Copra k (1995) analysis of corruption as a social practice around which ordinary Indi-
rhat rhe malo responsihiliry for rhe mota should be laid un Shiv Sana and a parcial and incom-
ans deliberare dre namre of che srare and ics ideal separador
from "rhe social" is exemplary petent police force.
as a way te grasp che everyday mneareings and significante
ofsarkar. e7. See Dieareshwar and Srivatsan' s (1996 ) exeellent piece on police pracriees eoncerning arreos
'Che contneission was aleo asked r
o recommend sreps ro be taken te i mprove che performance of young "rowdies ," thcir subsequenr classification as "rowdy-sheeters," and more.
of che police force and ro recommend admioiscrative measures that could reduce che
likeli- 18. In the predomieranrly Muslim neighborhoods I worked in, most families come from noah
hood of such incidents repeating thenrselves .
In other words , che task was daunring and India, and che srrucmres of che biraderi , rhe clanlike and durable relations of trust berreen
deeply conrroversial, as it was bound ro colude with
entrenched political and burcaurraric familias mosdy ( bit non always ) of che sanee caute, remain of paramouot importante when
intereses in che government and in che police force.
it comes ro mutual help, gctting jobs , martiage. recroiring new labor from che villages, lend-
8. This draws on my own presente at hearings in November 1996 and February
-March 1997, ing money, and so ora. However , in che urban economy, relations and economie networls cut
as well as wrinen affidavits from a range ofpoliee officers and civilians obrained from court
across [hese structures , and biraderi seems ofren to fimerion as a las[ resort in dures ofcrisis
officials.
or serious decision,
e. Summing up en che low morale of che police, Justice Srikrishna wrires: "Che police, by rheir
sg. A large nurnber of rice residente in [hese arcas are Muslim weavers of north Indias descent.

252 STATES OF IMAGINAT ION STATE MYTHOLOGIES IN MUMBAI 253

 n:usr]v y,Lp Ihe jdahu annmcni1,, : Io', t,au, anc. lotccr-cnsre coi] mm..im dtat high-status
,HUSISn:.s did nnl mc hni .e aS pmpu bluslims. The pdahas in rhis ccntun pegan lo chito
III STA'l8 AND COMIMUNffY
racogw Gun bv callin rhan, vc, a: :ir. che Acabe .cerd titr hciper rctcrriug :O dtosc scho
hhpecl th I roph.c: I-Oto 8' , .rla:an,r [h _ontc:np: t0: lul th u¡ansari. ,Iill pr

♦. '11i : h:gh-S tatus Muslim in 1hc nc] gl borhnod eid Ibour d:c police inforlile r:: Su 111( oj
thuse peopic eall rhemsclves ansaris. óur wIie an :hec hclpinge The police! Thesc meo are
j :dalias, ,, knaw ho:v Ihev ara.'.
J ¡le 1984 1 i0:s in Rhis m(1i. Thauc nP Ilombay me diser:sscd and dorunten red by tlm Conr
otittc ter clic Prorecri m elUemoa uie Kights e, 1 uva m I ,, he:sI u Bhi ^vad: Kials ( últil,c:

orne, 1984).
al. Inc similar initiabves in conjunction with rhe riots la Rembay 11 1898, sce Krishnasiral!
(1966 : 20- 45): for the 19zq rictus, s ec Chandavadcar (1998: r68-76).
_z. Sushoba Bharve, activist and self -professed Gandhian social worker (interview, Wodi, Feb-
ruary 19, 1997).
z). S. Rharve conhrmed chis view when che said, ' Kespecmble people are non very interested in
wodeing evith che police . The tout M11 always come firnvard, but we did not want that. So ve
workcd really hard to frnd good people with construenve views. lt was very diicult,"

254 SIATES 0 f 1 M AG 1 NA'1ION

BEFORE HISTORY AND PRIOR TO POLITICS
Time, Space, and Territory in the Modero Peruvian Narion-State

David Nugent

P/J L C, a°t--
Thi e say argues that local people's relation to the nation-state and moder-
niry has distinct spatial and temporal dimensions that shifr through time ac-
cording to complex changes in che organization and orchestration of power.
It also identifies three "states of imagination" for the Chachapoyas region of
northern Pero. Each "imagined state" is shown to have its own orientation
toward che nation and moderniry and to be characterized by its own distinct
"time and space effects. 1 also compare the three states of imagination to
assess che impact of each on the territorial integrity of che nation-state. At
issue are the conditions under which local populations imagine themselves
to be a part of or "outside" che historical imaginary of the modero nation-
state.
We begin with an analysis of che present, a period when nation-state and
modernity are regarded as alien to and contaminating of a local community
depicted discursively as remote, pristine, and premodern. We then consider
che historical process out of which the present emerged. Particularly note-
worthy is the fact that in decades past nation-state and modernity were re-
garded in very different terms than they are at present: in the rgzos as distant
but emancipatory forces thatwould liberate people from their oppressors; from
1930 tu 197o as integral to and constitutive of a local citizenry that repre-
sented itself as fully modem and nacional. In other words, we trace the his-
rorical process by which moderniry and che nation-state "sour" in the local
imagination to produce a curious historical inversion, one in which tradition
evolves out of moderniry_ We begin with the traditional present.'
 rever missed. As everyone Entre, they had meetings wirh each other ou a
Modero ay under Siego
regular basis, where they discos sed whatevcr matrers dice needed to discuss
to nwintain their control ()"Cr rhe tocan. Ore of rhe most imporrant rhings
In Qhe carie efrernoon of hile Iq, 19X3.1 arrived at the small food storc (bo-
thcy did ar [hese meerings, Sra. Yolanda raid, w,ts d.^ prices for rhe goods dice
dega ) in Chachapoyas where 1 always had my lunch . Upon my arrival sis
sold. Bur they also discussed such maners as hov, ro Influcncc- the police
months carlier 1 had asked Sra. Yolanda , who operated rhe bodega , ro cook
and pol iticians and how to deal wirh people cabo were a threar ro [hora. No
for inc and had been casing ar hcr store over sirca ¡'he arrangcment 1 had
onc liad actualIN over witnessed ore of diese meetings, shc admitted, bur it
wirh Sri. Yolanda proved to be unusually rewarding . 1laving lived iu Chacha-
was Cmnmou knowledge rhat they mor late at night in rhe Camara de Con-
poyas lar mosr of hcr tifo, she liad at hcr disposal a wealth of informarion
ercio (the Chamber of Commeree), a room in rhe municipality. Ir reas here.
about rhe town and its history. She reas also an extremcly good-natured, upen,
she said, in rhe dead of nighr, that they made their plans to take advantage of
and friendly person, because of which cae quickly becaue good fr iends. In-
people and settled on the excessive anrounts they would charge rhe public for
deed, Sra. Yolanda 's quick wit , wry sense of humor, and gregarious namre
meant that 1 always looked forward to rural times and to the discussions we their goods.
As a result of this behavior, Sra. Yolanda explained, when bodega operators
had about Chachapoyas.
or market vendors went to a merchant's warehouse ro get supplies of oil,
When I arrived for lunch on this particular day, Sra. Yolanda had just re-
noodles, and so on-as she had done rhat very day-they had to pay far more
turned from rhe largo store and food supply warehouse of Sr. Lazo, one of
[han should have been the case. And after paying so much for [hese goods,
the town's wealrhy merchants . 2 Lazo reas one of perhaps ten other prominent
she and rhe others were forced to sell them for almost norhing. There were a
merchants , or comerciantes , each of whom made his living by procuring
great many humble people like herselfwho had small stores or who sold food
foodstuffs and other basic necessities in bulk in distant industrial centers,
in rhe market, and they were all forced to compete wirh each other to survive.
transporring them to Chachapoyas , and selling them in smaller quantities
Everyone had to keep their prices as loca as possible if ehey wanted to attract
wholesale (and retail ) in the town . These raen provisioned all of rhe rown's
eustomers. Even so, because of rhe greed of rhe merchants, the eost of living
one hundred or so bodega operators (including Sra. Yolanda ) and all of the
kept going up. No one had any choice about rhe master, she said, because
women who had food stalls in the town market wirh small amounts of cook-
[here were only a few comerciantes like Sr. Lazo, and everyone was forced to
ing oil, salt, noodles, rice , soda crackers , toilet paper, beer, and other articles.
The population of rhe town carne either to bodega operators or to rhe town go to one of them.
Sra. Yolanda said that she simply couldn't understand what made rhe co-
market to purchase food, making it possible for women like Sra. Yolanda to
merciantes behave as they did. What she found especially baffling, she said,
make a modest living by reselling [hese iteras retail to rhe general populace.
was that they seemed totally indifferent to rhe plight of rheir neighbors. It
Sra. Yolanda was very angry when she remrned from her visir to Sr. Lazo.
was obvious, she said, that people were really suffering. Many women didn'r
When 1 asked what was wrong , she expressed deep frustration about her pre-
have enough to feed their children. Al! across town people had to cut down
dicament . With uncharacteristic bitterness and resentment she explained that
on really necessary rhings, lile powdered millo, eggs, cooking oil, and mear.
rhe comerciantes were making it inipossible for her, or any of rhe other bo-
They were eating more of foods lile yuca (which she said was fitting but nos
dega operators , to get along . Beyond that , they were doing real harm to every-
nutritious) and less of rice (which she said was far more nutritious) to make
one in Chachapoyas . The problem , she explained , was the merchants ' greed,
rhe linde money ehey had go as far as possible. Even people on government
their hunger for profits, and the pleasure they cook in taking advantage of
salaries were struggling. The government did nothing to control inflation and
people like her, who were poor and vulnerable.
wouldn't mercase salaries, so state employees found that their salaries pur-
Although rhe merchants rarely lefi their honres to go anywhere other than
chased less and less.' Retired government workers, who had fixed pensions,
their stores , had no friends aside from other merchants , and did nothing bur
she explained, were becoming desperate. Their monrhly payments were based
work and work, Sra. Yolanda said, [here was one "social occasion" that they

THE MODERN PERUVIAN NATION-STATE 259

258 STATES OF IMAG]NATION
 on what they had been earning at retirement, often decades ago, and hadn't ate as a group-probably settling on this strategy in one of their secret night-
gone up in years despite inflation. With rhe pittance they received from the time meetings. Despite rhe boycott, rhe woman liad tried to hang on. First,
government, she said, these old people could cover only a small part of their she had complained to rhe subprefect's office about rhe municipality's refusal
living costs. Many didn't have enough to eat. to enforce its own regulations. Later, she had gone to rhe prefect when rhe
According to Sra. Yolanda, however, what made matters worst of all was subprefect proved unresponsive. Ultimately, however, no one in rhe govern-
rhe fact that rhe merchants didn't really need to be taking advantage of people ment was willing to help her. As a result, the comerciantes succeeded in driv-
in rhe way they were. After all, she said, they were already millionaires. All ing rhe woman out of business. In the end, she had left Chachapoyas.
you had to do to see Chis was to walk by their fancy honres, see their shinv This story was common knowledge, Sra. Yolanda explained, and no one
new cars parked in their garages, or visir them at work. "Go by one of their was willing ro risk suffering rhe lame late as had Chis unfortunate woman.° 1
srores," she told me, ` look inside at all the merchandise they have, and see ir` then asked why rhe bodega operators did not organize ro present a united
they look like people who are struggling. Then, tell me why they are doing front to rhe comerciantes. She responded by explaining that every effort to do
Chis to us."
so had been mee by either indifference or repression on rhe parí of rhe au-
Sra. Yolanda expressed real bitrerness about rhe simation she and the odr thorieies.' With great bitrerness, she explained that rhe government had never
ers were in because of rhe merchants, a situation she saw herself as powerless taken any interest in rhe region or its people, that it did nothing to protect
to change. 1 knew that rhe municipaliry set official prices for many of rhe
them from rhe comerciantes. To rhe contrary: the government acmally helped
items Sra. Yolanda purchased wholesale from rhe comerciantes, and thus it rhe merchants prey en people. By way of example, she explained that every so
was clear that rhe price-fixing activities of rhe merchants were ¡Ilegal. Sra. often a newcomer would arrive with a truckload of goods that he would sell
Yolanda said as much herself. So 1 asked her why the bodega operators didn't at prices lower than those offered by rhe comerciantes. Seeing their monopoly
bring Chis problem to the attention of rhe local authorieies. In response, Sra. threatened, rhe merchants would arrange with rhe police so that rhe new-
Yolanda said, "How can we trust rhe municipaliry? It is te that very place that comer was falsely accused of and arrested for some serious crime. The indi-
rhe comerciantes go to make their plans to rob as!" vidual in question was then given a choice: either he could stay in Chacha-
Sra. Yolanda went en to explain why rhe government could not be trusted. poyas and Pace a jail sentence, or he could ¡cave town and never come back.
She told me rhe story of a bodega operator, a friend of hers, who had pro- Severa¡ of these falsely accused individuals, she said, had appealed te higher
tested te rhe authorieies about rhe comerciantes. Several years earlier, Chis political authorieies te free them from rhe rhreat of imprisonment so that they
woman had made a formal complaint to the municipaliry, but fomtd that rhe could operare a business freely in Chachapoyas. Their appeals, however, had
authorieies were unwilling to do anything to help her. Much te her dismay, fallen on deaf ears. As a result, all of these would-be food merchants had
however, Chis woman discovered that municipal representatives were quite been torced out of Chachapoyas, and die position of rhe comerciantes re-
willing to do her harm. Shordy after she made her complaint, she went ro mained unchallenged.
rhe warehouse of one of rhe merchants to gel fresh supplies for her store and It did no good to euro to rhe authorieies for help, Sra. Yolanda said, because
to her surprise was told that they were out of everything she needed-even they already knew what was going on. Indeed, they were directly involved in
though there were large sacks of these same items in plann sight and orher helping rhe comerciantes Cake advantage of people. This was undoubeedly
women were purchasing these goods before her very eyes. Thereafrer, she was because rhe merchants used the profits they made from local people to bribe
toid rhe same thing at all of rhe supply warehouses in rown. From that day the police, rhe politicians, and anyone else whose cooperarion rhe comercian-
en she found that pone of rhe merchants would sell her any of their goods tes required. In Chis regard, Sra. Yolanda said, the influence of rhe merchants
at any price.
extended far beyond Chachapoyas and included the industrial centers where
It was obvious to this woman that rhe municipaliry had "informed" rhe the merchants purchased their wares, renewed their business licenses, and so
merchants about her complaint, and that rhe merchants had chosen to retali- on. "The comerciantes could never do what they do," she said, "without rhe

26o STATES OF IMAGINATION TIIE MODERN PERUVIAN NATION-STATE 261

help of the government ." Ir vas local peoplc libe herself and other Chacha- pelled te generare a theory to account for Che forasteros' inexplicable behav-
poyanos who were lefi to suffer Che consequences. ior, a nccory with time and space effects oí its own. The forasteros were said
Sra. Volanda ' s angra and articulare ourburst was far from Che only expres- to be descended from one of che ten lost Cribes of Israel, who at sume time in
sion uf this kind rhat 1 board in Chachapoyas . lo rhe contrary_ lanlents just rhe remoto post crossed Che Atlantic Ocean, sailed up rhe Aruazon P iver, aud
like hers dominated public discourse . This discou sc . however, was flor lim- settled in the rown of Celendin (honre of cae merchant forasteros: see Nugent
ired to critical commentary on the disastrous stare of affairs Ches afflicting 1996: 258-61). In othcr words, the forasteros were thought ro belong ro a
Chachapoyas . Other themes werc equally importanr , in particular the har- rase differenr froni Che naturales. The modernity° they entbodied and rhe ses-
monioos , caholesome , and consensual way oí tire rhat was said to have pre- tero of srate power on which they relicd were regarded as baing as foreign n1
dated the contemporary era and the corrupting influence of the (orees of mo- ^ the people and place as the forasteros themselves.
dernity and nation-state in bringing about Che demise of traditional life.° /,Q &-' Iris peculiar indeed that Chachapoyanos represent themselves as naturales,
Implicit in Che "discourse of corruption" that dominated the public sphere a racially distinet, premodern folk community whose harmonious and con-
-stare.
during this era is a distinct "state of imagination"- a particular way of view - ( ^^
,,,4 4\ sensual way of life has been put at risk by modernity and the nation
ing the institutions and cultural values of Che nation-state, and of positioning From the 19205 through the 19706 they depicted themselves as quite the op-
Che regios and its people in relation to the nation-state in spatial and tempo- posite-as members of el pueblo (Che propio ), a racially unmarked nacional
ral terms.' "Alien" is an apt terco to characterize how the modem nation-state communiry that enthusiastically embraced che institutions and values of the
was imagined during this era. lndeed, in public discourse Chachapoyas was modern nation-state.
represented as a spatially remoto, socially pristine, and temporally premodern Having briefly discussed Che "traditional present," we now examine thc
communiry only recently subject to the dangerous and corrosiva influences of "state of imagination" that gave birth ro this racially unmarked nacional com-
the merchants, modernity, and rhe nation-state-all of which were closely munity, and in the process uncover somerhing of the region' s "modern pasr.
linked in public discourse.' Although [hese forces were said to have made
their presente felt in chis remoto region for only a short time, they were rep-
Phase 1. 1885 to 1930: The Compromised Nation-State
resented as fundamentally alien to the region and its people and as already
having done much to undermine what had traditionally been a simple, har- At the turn of the twentieth century the emerging Peruvian nation-state was
monious, consensual way of life. rent by contradictions. Although founded as an independent state in the
in Che local imagination, it was the comerciantes who most clearly em- r82os en liberal principies of demoeracy, citizenship, private property, and
bodied the corrupting influences of Che outside world, so much so that pub- individual rights and protections, the central government was not remotely
lic discourse depicted people like Sr. Lazo as both dangerous and foreign. able ro make good en these arrangements even one hundred years later. Al-
Indeed, local people's terco for themselves was naturales (natives), people though such principies were uniformly invoked in all political ritual and dis-
namrally and unproblematically of the place of Chachapoyas. Their term for course, many parís of Che country were organized aeeording to principies
Sr. Lazo and others libe hico, however, was -forasteros (outsiders; strangers), diarnetrically opposed to [hese precepts of "popular" sovereignty.
people who were "out of place" in Chachapoyas, who did not "belong." Natu- Chachapoyas was one such region. Dominating the social and political
cales emphasized a fundamental moral schism that separated them from the .J i landscape of Chachapoyas vas a group of white, aristocratic families of pu-
forasteros and Che nation-state, w hi c h were eonstrue d as en ti re l y di s ti nc t an d tatively Spanish descent who saw it as their birthright to rule over Che region's
AA^^ ,
separata from Che communiry of naturales, as belonging ata great "distance" 1 multitudinous mestizo and Indian peasant population. These elite families
from local life. Indeed, the state was noted primarily for its misuse of or its rejected any and all assertions of equaliry between themselves and Che subal-
indifference to Che local population. The forasteros were sean as so funda- tern groups over which they ruled. Rather, the elite regarded and referred to
mentally alien to all things truly of Chachapoyas that the naturales felt coro- themselves as a separare aristocratic caste, the casto española, thatwas naturally
A

262 STATES OP 1 MAGt N ATION THE MODERN PERUVIAN NATION-STACE 263

 entitled to power and privilege dueto its racial purity and cultural superiority.
lege of the elite and to the violent struggles for power that occurred as differ-
Even social interaction with subaltern groups was kept to a minimuni and
ent factions of the elite, each with its own extensive clientele of mestizo arti-
was strucrured in such a way as to require public deference and subservience
sans and "Indian" peasants, battled for regional control. On the other hand,
from che subaltern.
state institutions were designed from che center to support the public sphere
Peculiarities of che Peruvian state during chis period mean[ that the central
imagined in nacional discourse, one consisting of a mass of formally identi-
regime in Lima was torced to ally with such aristocratic families if it was to
cal (albeit male) citizens, each jurally indistinguishable from che next, all of
maintain even a semblante of control in the outlying sections of its territory.
whom were united behind che national cause of promoting "progress" and
For the central government lacked the ability to control the nacional territory
"advancement." This public sphere, and che institutions meant to support it,
directly during this period. Those who controlled che central state apparatus
also offered an explicit critique of a social order in which local aristocratic
were thus compelled to seek out clients among rural aristocratic families who
elites and their clients used unlicensed violence on an endemic scale to defend
would rule in the narre of the central regime.
their privileges, powers, and prerogatives.
In other words, although the independent republic of Peru had been born
To address the threat posed by che institutional and discursive presente of
of armed struggle against absolutist, colonial Spain, and although che central
che nation-state, che local ruling elite faction repeatedly and publicly demon-
government of Peru embraced precepts that had emerged out of che Enlight-
strated its ability to ridicule nacional injunctions about "proper" behavior-
enment, che actual operation of che state apparams depended crucially on
to violare che constitutional protections of the opposing elite-led faction, to
maintaining social strucrures of the ancien regime.
do great harm to their persons and property, and to do so with complete
Although elements of the landed class in Chachapoyas enjoyed che back-
impunity. In other words, che ruling faction went to great lengths to com-
ing of che central government, this class was far from unified. Rather, mem-
promise state institutions and nacional discourse, using both to further the
bers of the elite saw themselves as having che inherited right to rule that no
aims of social groupings (their own privileged faction), statuses (superordi-
one, neither che central government nor other members of che elite, could
nate racial eategories), and forms of interaction (violente and domination)
legitimately deny them. Furthermore, to exercise [hese rights che elite did
that were "unthinkable" (or deemed "medieval") within che discourse of the
not hesitare to use violence. There was a sense in which elite men regarded
nation-state.
themselves as having no true peers, and certainly no masters. Rather, each
ti was notjust state institutions and nacional discourse, however, thatwere
believed he had che right to use power in defense of prerogatives that were
compromised by [hese kinds of actions. The nation-state was compromised
legitimately his because of che elite station in life that he enjoyed by right of
in a second way as well. At issue here were the boundaries of che state appa-
birth. No one-nor even che state-had the right to interfere with [hese
ratus. To protect itself from the ruling elite faction, the opposing elite faction
privileges.
y/ - rganized itself finto a set of positions that replicared the state apparatus: a
As a result of chis state of affairs, in che opening decades of the century che
11ff11^^ shadow state" (Nugent 1999).'° Furthermore, the opposing faction claimed
operation of che state apparatus and the imagined communiry of che nation
that Its rig t ul posmon as t e local representative of the nation-stare had
were "compromised" in two senses. First, che region's landed elite took con-
been usurped by the ruling faction through violence and deception, and thus
scious and systematic steps to publicly contradice what the "stare stated"
that che ruling faction had no legitimate claim to che position it occupied. As
(Corrigan and Sayer 1g85): ro use state powers and public institutions to sub-
a result, at any given point in rime, ruling and opposing factions both sought
vert che very forms of egalitarian personhood and homogeneous public imag-
to occupv che same political soarP "nd hnrh rlaimed to be "the legally con-
ined in national discourse. The elite were compelled to do so for two reasons.
stituted authormes." Neither of [hese factions, however, could hold onto che
On che one hand, nacional discourse emphasized equaliry before che law, citi-
state apparatus more [han five to ten years before being deposed by the
zenship, and constitucional rights as che only legitimate basis for naconal life,
other. Furthermore, in that the rise of one faction inevitably meant perse-
principies in blatant contradiction to che racial hierarchy and inherited privi-
cution, emiseration, decline, and even demise for the orhers, each faction

264 STATES OF IMAGINAIION

THE MODERN VERUVIAN NATION-STATE 265
 Phase II. The 193os to rhe 197os: The Liberarcd Narion-Statc
onditions
these
tbughtviciously lo protect and promote irs own interesas. In [hese c
it was verv difhcult tor rhe opinase to view rhe seare as a discrete entity aboye
in 1930 the aristocratic order carne undone, as popular classes coalesced
and b odcry' ,r rhe n a [ion a tn ceno widta general will or interesa.
roto a social movemenr rhat mobilized around prineiples o citizcnship and
Rather, rhe regional elite subsumed rhe stars apparatus and used ir to violare
equality, idcals long invoked ni political ritual and discourse but thar stood in
and ridicule rhe precepts of state and narionh ood..
such glaring contradiction to ev_eryyday, life. Seizing on this radical disjuncture
In orher words, rhe atremprs of warring elite-led tactions ro legitimare aris-
bervwecn ritual language and social lile, die movemenr elaborared a telling
tocratic sovereignnv, when popular sovereignryy, represented rhe role terms in
critique of rhe existing state of affairs. It did so by affinning rhe legitimacy of
which political lile eould be represented, "compromised" rhe nation-state.
the egalitarian principies involced by rhe aristocrarie famílies and by poinring
Elite efforts to compromise the nation-state, howevei, were internally contra-
c^Ep ehe elite's consistent betrayal of [hese principies.
dictory. The more cines "misused" stare institutions to undermine the egali-
The emergence of this movemenr in Chachapoyas cannot be undersmod
tarian forms of personhood and public imagined in national discourse, the
outside of changes occurring in rhe national and international arenas. The
more powerfully they generated the image of a state form that was an inver-
late nineteenth-century crisis in North Adantic capitalism resulted in a flood
sien of their own compromised creation. Indeed, the very fact that rhe elite
of new laboring people, social doctrines, political ideologies, and investment
devoted such unceasing attention to compromising the nation-state sug-
capital pouring into Peru, which permanendy altered Peru's class structure,
gested the existente of somethin2 of which the elite were deathly afraid,
balance of political power, and state apparatus. Lima, rhe national capital,
something so powerful and so dangerous to elite oower that it
saw rhe emergence of a working-class movement and a forro of populist,
compromised and negated at every opportunity. The "thing" in question was
mass politices that challenged aristocratic control for rhe first time in the
a nation-state liberated from the abuses of che local elite, a nation-state that
country's history (Burga and Flores Galindo r979; Stein 1980; Sulmont 1977).
truly did representjustice, equality, and rhe common good.
Outside of Lima the influx offoreign capital undermined production and ex-
There were important spatial and temporal dimensions to these counter-
change relations within and between regions, resulting in movements of se-
imaginings. The ongoing public demonstrations in which rhe elite showed
cession, civil war, and escalating leveis of violence (Collins 1988; Gitlitz 1979;
their abiliry to hold rhe threat of rhe liberated nation-state at bay-that is, the
IQarén 1973; Mallon 1983; G. Smith 1989; Taylor 1986). In rhe absence of any
success of rhe elite in compromising the operation of the state apparatus
state presente in the countryside, rural elites were left to fight among them-
regionally-created the illusion (and illusion it was) that a liberated nation-
selves for control over rhe new economic opportunities o t e era.
state existed in a different spatial domain, lying beyond the reach of regional
In rhe openmg esa es o e ten ury t e central government also sought
power holders. Indeed, b c liberated nation-state had been distanced
to form a solidary national community along three axes: (r) the compression
from local affairs it w ossible or the popular cla
of space, by-w,°,,, of moccive investments in road construction to form a
liberated nation-state ot on y as a "thing" that could arrive from atar, but
more integrated national economy; (2) amassive expansion in rhe state bu-
ajo as a thing whose arriva was eing t warted by the elite. And because
reaucracy (including the military and rhe police) and rhe public sector; and
the liberated nation-state was associated with moderniry and the future and
e(3)
use
rh state control over'ptl te urs to promote eitizenship in-
rhe elite self-conseiously harkened back lo tradition and the past, the spatial
dividual ri hts, and eaua.lig hefez rh 1 A u rhe only legSiutare asrs of
plane became a temporal one as well. Not only was rhe liberated nation-state
national life (Burga and Flores Galindo 1979). In other words, just as the
being held at bay geographically by the elite, so too was the advance of mo-
central government was attempting to form a more cohesive national com-
dernity and progress being thwarted by backwardness and tradition.
munity based on rhe rhetoric of popular sovereignty, it was confronted wirh
These mixed messages were implicit in the orchestration of regional polit-
chaotic conditions that violated [hese very principies,
ical life. We now turn ro the social developments that made the messages
In Chachapoyas, nonelite groups responded by embracing rhe very prin-
explicit.

THE MODERN PERUVIAN NATION-STATE 267

,66 STATFS OF IMAGINATION
 ciples invoked in national Jiscourse as providing the only legitimate basis of
nacional life: a clettfsed state apparatus that could help make a reality of the from che broader national context. All such means of compressing time and

imagined nation-state. Two overlapping political movements, both intent on seace were viewed with a kind of millenarian awe, as capable of ushering

"democratizing" che aristocratic social orden, emerged among popular classes in an era of unlimired progress and prosperity (see, e.g., Amazonas, año i,
15 October 1926).
in Chachapoyas: the Partido Laboral Independiente Amazonense (Indepen-
dent Labor Party of Amazonas) and the Popular American Revolutionary Al- Finally, Amazonas argued that profound individual transformations would

liance (APRA). Both seized on state-endorsed notions of equality and the be required of those who wished to become one with the nation-state. Chacha-

common good and used them to imagine a democratized social order that did poyanos would have to forgo the violent and arbitrary displays of power and

away with elite privilege. In addirion to extensivo political organizing among domination that had characterized local life for so long to transform them-

subaltern groups, one of the movements published a local newspaper, Ama- selves into a "natural aristocracy" of individuals worthy of region and nation
alike.
zonas, which articulated this new image. Amazonas asserted that the nation and
the people did have a common interest, that they could build an effective The characteristics of these worthy individuals were thought to differ sig-
nificantly by gender. The remade man imagined publicly in the pages ofAme-
fürure-by rising up against the aristocratic families. The papen asserted that
Chachapoyas' elites were not a noble caste, but rather were brutal and rapa- zonas svas the very antithesis of the aggressive, dominating, violent male

cious, that they had divided, victimized, and abused the local population for of the aristocratic order. Rather, he was a peaceful, autonomous, rational,
centuries . Nor, the papen argue d , were t h e region ' s mestizos and indios un- hard-working individual, whose behavior was characterized by moderation,
(, ^~
couth and semicivilized, as the elite asserted. Rather, they were the local em- restraint, discipline, and respect for himself and others-an individual who

bodiment of el pueblo Peruano (the Peruvian people), the region's only hope for posed no threat to anyone. The actions of such a man did not need to be
sa l vat i on . El pueblo was deeply committed to democracy, equality, and constantly monitored or controlled by any externa¡ body because he policed
jus-
tice, but the state had neither extended to el pueblo its constitutional rights himself according to generally accepted principies of fair play, truth, and

nor curbed the excesses of the elite. Once rid of the elite el pueblo would be ethics. His motive in doing so was not to gain material rewards or riches.
Rather, being this kind of person was depicted as its own reward, knowing
free to become a true part of the Peruvian nation. In the meantime, however,
that one lived up to the ethical ideals to which all should aspire to live just
Chachapoyas was best understood as a remoto backwater where prenational
sentiments and archaic forms of behavior lingered that had long lince dis- lives, m be responsible members of community and nation (see, e.g., Ama-
zonas, año 1, 15 October ig26).
appeared from the rest of the civilized world (see, e.g., Amazonas, año i,
i5 Ocrober 1926; año 2, March-April 1828; año 4, May 1929). The news ayer imagined an entire community of such AutoDQMQuS, hard-
working, dignifled ethical male individuIs. These men were tu obey the
According to Amazonas, integrating the region into the nation-state would
require a series of major transformations in local life. First, it would mean same set of moral principies regardiess of race or ancestry, were to identify

establishing uncompromised state institutions that were free of elite control, themselves with the lame "common good" and "general interest." They were
to live in peace and simplicity, desire not what belonged to others, nor do
that could safeguard individual rights and protections, and that could guar-
antee equality before che law. Indeed, according to the paper, chis was the anyone harm. They were to adhere to general principies of integrity and truth,
nesty and humility. And they were to free themselves of any and all feelings
most daunting problem facing the region and its populace. Becoming a part
of fear, admiration, or envv toward the powerful. In other words, they were
of the nation-state would also mean, however, integrating the region into ,k
to become disciplined, principled individuals who were neither abusivo of nor
che nacional economy and culture. With regard to this problem the papen
obsequious to others, who possessed the "inner strength" to be fully self-
put enormous emphasis on the extension pf communicafnp, traAWorta-
reliant, independent, and autonomous. Furthermore, they were to evaluate
tion and educatin. Amazonas was particularly fascinated with roads and any
the worth of all, regardless of social position, in terms of the same ethical
means of breaking down the physical boundaries that separated the region
standards of behavior (see, e.g., Amazonas, año i, i February 1927).

268 STATES OF IMAGINATION

THE MODERN PERUVIAN NATION-STATE 269
 The remade women depicted in clic papo diffcred profoundly from the women and men of which regiou and nation were so much in nced, Arnazonm

national, banal-worl:ing, principiad individual into which alean were lo mold worked with exisring social distinctions hut gave diem a transformad signifi-

rhentsclves. The ideal svonr<rn svas one whase moral purity, natural sinrplicigl cante. Specifieally, Amazonas usad the disdain of elite groups fbr manual labor

and eniotive empathy made her worrhy oí inclusion in the national commu- as che basis for identifying the "superior man.- but inverted the terms oi

nity. Thc most importan[ obligation of a woman, asserted Amazonas, vas te moral legitimacy associared with manual labor, making it finto a virtue rather

devore herself to the domestic sphere, where sha was entrusted with the heavy than a vice. On chis basis, the papen attempted ro arase distinctions bettveen

responsibility of providing the proper environment for turning boys finto the Indio and mestizo and to malee common cause amkng all rho se who labored.

proper kind of responsible, disciplined men. Indeed, mothers were depicted For, aeeording to the newspaper, ir was the regiori s laborers who exemplified

as local embodiments of the national communiry in Amazonas, for it is they the superior man, who could malee sociery over in a new forro (see, e.g.,

who inculcare in rheir sons the kinds of values and orientations that will make Amazonas, año i, r March 1927).
them "useful elements to communiry and nation" (see, e.g., Amazonas, año i, This appeal for a "liberated" nation-state, in which el pueblo could reap all
the benefits of modero state and nationhood, attracted large numbers of the
i February 1927; April 1927).
Given this faith in the key role played by women and the honre in the pro- laboring population. In August 1930, at a moment of crisis in the national

cess of nation building, it is not surprising that every issue of Amazonas political order, [hese marginalized middle secrors risked their lives in an

included special seetionsa "Woman's Column" and a "Children's Col- armed ` revolution" that overthrew the region's arisrocratic power holders.

umn"-devoted speeifically lo these copies. The "Woman's Column" went Elderly individuals who participated in the uprising liken their assault on the

fimo considerable detail about how women were to order and maintain rheir prefecture, decisiva in bringing them ro power, to the storming of the Bastille
domestic space (focusing on cleanliness and hygiene) if they wished ir to be in revolutionary Franca. The local newspaper svas renamed Redemptíon. In it,

truly "proper," if ir was to be the kind of environment most conducive lo the calendar was set at "year one": "year one of the redemption" was 1930,

turning rheir sons finto men suitable for membership in the modern nation- year two was 1931, and so on.
state. The colunut was particularly helpful in alerting mothers to the exis- Within a few years of the 1930 "revolution" the elite-based political order,

tence of hidden dangers lurking within and around tire honre that threatened with its congeries of material, behavioral, and symbolic markers oí privilege

the integriry and order of the all-importan[ domestic sphere. In Chis way the and distinetion, disinregrated. With the breakdown of the arisrocratic order,

paper proved tbat maintaining the honre was anything bur a simple task, and "rhe people" became aetively involved in the region's most vital social aetivi-

that Amazonas liad an important parí lo play in educating mothers as lo how ties, assumed control ovar its key political positions, and reorganizad its most

it could best be done (see, e.g., Amazonas, año i, April 1927; año 4, May 1929). basic forros of religious practice." Furthermore, under the protection of lib-

It was the region's laboring poor, el pueblo Chachapoyano, who were re- erated state institutions , el pueblo elaborated a range of new sociocultural
garded as having the most potencial for becoming the remada men and forms. These "horizontal associations" (Wolf 1966) strengthened "the

women of which the new social order was to be composed. Even [hese re- people" as a group, and also celebrared the principies of equality and justice

made individuals, however, would nave lo cake one additional step if they around which the middle sectors liad mobilized as a political communiry. 2

were to overcome the divisive political affiliations of the past. They would The ascent of el pueblo finto these multiple arenas o£ social, political, and

llave lo joro forces with all other hard-working, disciplined, and committed religious rife enabled people who liad recently risked rheir lives to bring about
individuals. Only by aeting with the strengrh that carne from uniry eould even a rransformed social world to broadcast the legitimacy of the principies for

these remade individuals be suco that the elite did not subvert the process which they liad foughr-and to do so in myriad domains of social life that

of national inregration and regional regeneration (see, e.g., Amazonas, año r, until then liad remained the exclusive preserve of the regional elite. The com-
bined effect of these changas was rwofold. On rhe one hand, they resulted in
i March 1927).
In identifyi ng who was most likely to transform themselves iota the remade the emergenee of a new kind of publie culture in Chachapoyas, one in which

270 STATES OP IMAGINATION THE MODERN PERUVIAN NATION-STATE 271

 the constant assertion and celebration of equaliry replaced the hose of mate-
rial, behavioral, and symbolic markers of hierarchy previously acted out en a
1 As part of che same process that "liberated" el pueblo and the state, the
central government also established the conditions in which national com-
daily basis in the reproduction of aristocratic sovereignty. On the other hand, modity circuits could penetrate the region free from the encumbrances of
these changes created a situation in which the principies of popular sover- ..a elite politics. For with the dismantling of elite factions, in which the person-
eignty took en a transformed significante in public lile. For the first time in . nel and the property of opposing factions had not been sale from the preda-
the region's history these principies carne to define the standards to which tions of the ruling faction, the state underwrote the safety of individuals, the
individuais who oecupied positions of public trust could be heid accountable. security of their possessions, and their ability to move through the regional
This marked a radical change from the status of these principies during the space unmolested. In 1960 the first major highway was completed that linked
period of elite rule, when they had acted as nothing more than empty phrases Chachapoyas to the coast. From this point onward the goods and labor of the
uttered mechanically in political ritual and discourse by elite groups who pub- n /4({Chachapoyas region were "free" to enter markets of more nacional scope, and
licly flaunted their ability to disregard the most fundamental aspects of these foodstufts and manufacmred goods of all kinds from outside the region could
principies. likewise participate in what had formerly been the rather limited, regional
There was also an important material dimension to the ascent of el pueblo. markets of Chachapoyas (Nugent 1988: chap. 6). As the region became pro-
Beginning in 1930, successive regimes in Lima provided crucial material sup- gressively more integrated into nacional commodiry circuirs, local peone and
port te Chachapoyas's middle sectors. The government bureaucracy expanded products) were increasingly drawn into a nacional, exchange-based economy
in size and scope in the form of a growing number of separate ministries, that provided them with most of the material necessities of life.
offices, and financial institutions, and in the process absorbed growing num- By the Ig6os, then, the local populace had succeeded in realizing most
bers of the local populace (Nugent 1988: chaps. 4, 5). As early as 1950 em- aspects of the vision that had first inspired them to political aation in the
ployment in this bureaucracy had absorbed as much as 40 percent of the 192oy^Integration with the nation-state had, in the words of Amazonas, liber-
working, male population of the town (up from 13 percent in 1930; see 69- gated the region from the "ciaustrophobic insularity," the "boorish provin-
70). And because the real wages of those employed by the state roce steadily cialism," and the "remoce backwardness" that characterized the period of
throughout che 1940s, 195os, rg6os, and the first half of che 19706 (69-70), aristocratic ruley State institutions had been liberated from elite influence,
the growth of the bureaucratic sector was able te support the expansion of local lile had been democratized and modernized in virtually all ways, and the
artisanal production of many kinds (Nugent 1995) as well as a whole array of rights and responsibilities of citizenship were realities for most townspeople.
ancillary "service industry" activities (Nugent 1988). Furthermore, el pueblo dominated che region's major social activities, con-
Changes in economic, social, political, and cultural lile following che revo- trolled its key political positions, and organized its central forms of religious
lution of 1930 thus led to a significant expansion in the size of che town's practice.
middle sectors, the very social group of which el pueblo was composed. At
the sane time government investment in communication, transportation,
Phase III. 1g8o to the Presenta The Alien Nation-State
education, health care, the judiciary, and tire police force-all of which ex-
panded steadily beginning in the 19305-acted to strengthen che autonomy From the 1930s until che early 197os the institutions of the nation-state did
and security of the growing middle sectors of regional society and to make much to support the citizenship-based public sphere imagined in national
superfluous the group-based protections and coercions that had character- discourse che very public sphere that el puebl0-had he pe ring loto being
ized the earlier period of elite rule. In this way the central government re- as a result of the movements of democratization of the Igzos and r93os. By
placed the elite as the source of patronage and protection for the local com- the ig8os, however, the local populace carne to regard the modero nation-
munity, and in the process established direct dependencies with an emergent, state as a "foreign body": an entity wholly separate from, alíen te, and dan-
nacional citizenry that it helped to creare and whose rights it protected. - gerous for a way of lile depicted as simple, natural, and harmonious. So for-

272 STATES OF IMAGINATION TI-SE MODERN PERUVIAN NATION-STATE 273

 tralized lbrm of governance in which tl uur own appointecs (pretect,) viere
cipo wcre the nailon-state in,¡ ntodernity, perce i vcd to b e h y r hi s ti me th a t / —o 1,1
in given virtual monopoly control ,,el the state political Ipparatus fbr thc set-
local people sought to distante themselves fiara everything assoeiated with
tions of die narional territory (departntenrs) cntrusred te rhena. 1,1ciecrs, aud
nce nerion-sine including the ve e public splicre tirar rhev liad helped brin,
tliosc nccv appointed lo Icnscr adnahtistiarb c posta. >vct given free ein te
finto bcing Fvirh thc asistente of thc nailon-ata tc_ and the discoulac of popu-
govern according te) their own prcrogatives, without "1ntertercncr" ítont
lar sovercignty that liad foruterly acted as the idcological expression of decir
elecied polídcians (Congress liad been dissolved), poliical partes (parties
political mobflization. In orher words, by nce 19Sos local people viere no
liad been oudawed), or tito host oí orher formal and intormal mech.misms
longer willing to depict ncc public sphere as lile contingent ach ievement oí
thar liad evolved during previous decades he which people liad mude thcir
an emergen[ social group (el pueblo) wirh an explicirly modern orientation
wodcing directle in concert wirh the nation-state. Rather, by this tinte the influcnce felr in local affairs.
The military government's imposition of autocratic role in Chachapoyas es-
representation of borh the public sphere and the local populace itself liad
tablished eonditions in which the demoeratized public sphere that el pueblo
undergone a discursive inversion. Botli viere imagined as prepolitical and
liad helped bring into being was inereasingly undermined. In alliance wirh
ahisrorical, as having timeless features that viere inherent in the people and
ncc Guardia Civil (tito narional police) and a dependent judiciary, prefects and
place. So alien did the nation-state and moderniry appear to local people by
their appointecs were free to coerce wealth out of the local population and ro
the 1g8os that they reimagined their own history in such a way as to excise
persecute and harass those who did not aequiesce ro their demands, wherher
both from their past. Indced, in Chis regard, the naturalizing dimension of
eontemporary public discourse is very effective; one would never know of el forwealth, labor, subservience, sexual favors, or orher.
The military government tlaus introduced a forra of rule that was autocratic
pueblo's previous "romance" wirh forces now regarded as dangerous and for-
eign, would never suspect that che nation-state and modernity werc once re- and centralized, borh nationally and locally. it iniposed its will arbitrarily on
the populace and subverted democratic process, even as it proclaimed the
garded as powerful forces of eniancipation. The very fact that Chachapoyas's
importante of popular rule. This contradiction between the discursive repre-
"modern" history is a hidden history thus says much about how alien the
nation-statc has come to be in the local imagination. We now mr toa con- sentation of the nation and the actions of the state apparatus in some respecrs

sideration of the forces thar led tú the emergente of this most recen[ "state resembled the contradictions of pre-1930 aristocratic rule. There wcre impor-
tan[ differences, however, between the two periods. As of the i97os itwas no
of imagination," that of the alien nation-state
In 1968 severe economic problems occurring at the national level, related longer the case that the national communiry could be conceived of as being
compromised at the regional level by a backward-looking local elite, as had
to Peru's dependent position in the global capitalist economy, accompanied
by widespread peasant land invasions and social unrest led to a military coup, bcen true in the 19205. In the 1970s che national communiry was being com-

which was fbllowed by twelve years of military rule. The discourse of the promised at the national level-and by a state apparatus that claimed ro he

military regime can be characterized as "radical populist." "Che military cri- oriented squarely toward the futuro.
Faced wirh [hese eircumstanees, local people could no longer look to a
tiqued not only an unjust international economie o d fhat exploited Per
and robbed it of its resources, but also the counrry's highly unequal class liberated nation-state as an alternativo te a locally compromised state of af-
fairs, as they had in the 192os. Rather, in tenis of spatial imaginings, the
srruemre, in which an entrenched landed elite held power over the masses of
local populace was eompelled to "distante" irself from the broader, national
peasants and workers. The military government promised to address [hese
problems, tú involve the masses in the political process, and to bring to en context. Local people viere eompelled ro look inward, te themselves, for an

end the marginalization of the country's majoriry (Booth and Sorj 1983; Low- alternative m a nationally compromised state of affairs. The danger was no
longer interna], as it liad been in the 1920s, but rather external, embodied in
enthall 1975).
. A though the do t re ml nary regime s resse emocracv, p the state itself, and made manifest in the srate's behavior nationally.
Although the military state was compromised in one sense (it violarcd the
participation , and role by the peopl^ the actual srrucmre ofpower estab
a.,r_tlMilirary
e rulers set up a highly ceo- principies of popular sovereizuty- , tich le tt' te rule was based), in an-
lished by the regime s v as

274 STATES OF IMAGINATION THE MODERN PERUVIAN NATION-STATE 275

 t+ir`dc^1 ^
other it was quite whole and intact. Unlike the compromised nation-state of
institutions of the nation-state continued to provide material and political
the 192os, in the 19706 the boundaries of the state apparatus were anything
support, howe,,ver, Che subgroup maintained itself quite well.
but compromised or unclear. Local people knew exactly what "the state" was,
T1ua0 invnlyed in the mercantile sector, on th- 1 i I
and it was not they! In the 1970s it was notan element of the local population
accumulate surplus. and in considerable uantities. Some of these merchants
(an aristocratic elite) that could be blamed for compromising a state of affairs
were based in Chachapoyas, others just to the west in the town of Celendin.
(popular sovereignty) deemed legitimate and normal beyond the confines of
Merchants from both areas had long been involved in regional and inter-
the loeality (as was the case in the ig2os)_Instead, it was a group of"foreign"
individuals (prefects, judges, police), imposed on the populace from beyond regional commerce, but always on terms set by the local elite (see Nugent
1996, 1997). Merchants from both areas had also been centrally involved in
the locality by che military state, who were compromising the nacional com-
munity regionally and nationally. the movements of democratization and thus in che 1930s were able to harvest
the fruir of their own political efforts. After 1930, however, the Celendino
These conditions continued into the 198os. Although military rule ended
merchants (who in the rg8os and rg9os carne to be referred to as che "Jews"
in tg8o, and thus government institutions lost their overtly dictatorial char-
of Celendín) were able to draw on their deeper involvement in and greater
acter, in the context of economic disintegration and the brutal war with Sen-
familiarity with more all-embracing national circuits of exchange to gradually
dero Luminoso, political parties ruled from the center in an inereasingly
eliminare their Chachapoyas-based competitors
autocratic manner. Even so, they proved themselves incapable of managing
As che Celendino merchants increasingly monopolized interregional trade,
national affairs and eventually delegitimized themselves dueto revelations of
they began te impon a wide range of manufactured goods (and later mass-
massive fraud and graft. All across the country there was mass disaffection
produced food staples) into Chachapoyas on a scale never before possible.
from institutionalized political parties and processes, as people looked for
Seeing little possibility for profit by exporting agrarian producís or handicraft
alternatives to a system that had obviously gone bad. In che second half of che
production from the region, however, the merchants never beeame involved
rg8os, as the Peruvian economy disintegrated in the context of triple- and
in the ownership of land or the exploitation of local labor. Rather, they re-
even quadruple-digit inflation, and as the government wrestled unsuccess-
mained a true merchanr class, one whose interests were restricted exclusively
fully with a huge foreign debt and the censure of the "internacional economic
to the realm of exchange. The national, exchange-based economy on which
community," state institutions of virtually all kinds ceased to operate and
most Chachapoyanos became increasingly dependent for their livelihoods af-
even collapsed." This was especially true in outlying sections of the national
ter 1930 was thus brought to them by this small group of merchants who
space like Chachapoyas. As the citizenship-based public sphere once main-
were focused on commercial affairs to an unusual degree. They and their de-
tained by the state apparatus continued to weaken, Chachapoyanos perceived
scendants thus became the mediators berween the Chachapoyas region and
themselves to be a small island in a vast sea of insecurity. It was as if the
che national economy.
nacional community itself was disintegratiog.'°
As che town's middle class differentiated finto accumulating (merchant) and
Compounding local people's sense of "external rhreat," heightening their
nonaccumulating (bureaueraticlartisanal) sectors, those involved in each were
scnse of betrayal and isolation-bur adding a temporal dimension to their
thus confronted with a distinct set of lile possibilities and limitations. As this
sense of spatial remoteness-wcre changes in che exchange economy. The
occurred, the formerly unitary vision of social justice that had provided the
disintegration of the elite ordei, and th finte ration of the Chachapoyas re-
cultural basis for che entire middle class's struggle with the elite order in the
gion finto the nacional economy, had profound and unanticipated effects on
r92os underwent a similar differentiarion.`Those involved in each middle-
regional social relations. From the 1930s onward el pueblo differentiated into
dass sector selectively appropriated [hose dlmenslons o t e ongtna mora
two distinct subgroups, one mercantile and the other bureaueraticlartisanal.
v^ r i,^r correcnot
^ tyI eu new I e etrcumstances and pushe
The bureaueraticlartisanal sector can be characterized as "nonaccumulating";
the remaining ones from view.
people within it were unable te accumulate surplus. To the extent that the
For [hose in the nonaccumulating sectors economic livelihoods were se-

276 STATES OF IMAGINATION

THE MODERN PERUVIAN NATION-STATE 277
 diese circumstances it appcared to local people thar che merchanrs were will-
cure but also modest and more or icss hxed, regardless of individual effmt. ing to put [hcir hungcr for profir before oven che survival of [he local com-
Despite Reir vision of prosperity coming fiom hará work and individual muniry before che abiliry of their ncighhors to fecd [hcir children. That rhey
nierit their actual lile possibiliries in Chis regard wcre limited. Those in Chis could do so was a funcrion of [hcir control ovcr che disrribution and sale of
sector therefore expressed their newfbund strength by seizing control of Che food staples. Thar rhey would do so was something local peo pie had no easy
political, social, and cultural arenas fiom which rhey had becn exeluded by
way of u ndcrstand i ng_
che elite, by " democratizing" diese practices and opening them up ro "die As che local populace sought some way lo comprehend che inhuman be-
citize )iv as a whole. 1nvolvemenr in [:hese demoaatized puhIle acrivities be- havior of che merchanrs al[ che cultural differences and social dfstanees thar
came an important indicator ot this sectors new solidarity and social aseen- had eharaeterized Chachapoyano-merchant interactions ovcr die years carne
dancy, of their niumph over ihc elite families. mgerher in che local imagination to form a logical, ordered "esplanation,"
Because of che possibilities for aecumulation available to them, on the one that attributed te tire merchants an inherently different honrar] nature.
other hand, the small handful of merchants who monopolized long-distance Nothing short of chis, it seemed, could explain rheir indifference to che suf-
exchange were able to realize precisely [hose dimensions of el pueblo's origi- fering of others, their willingness to take advantage of che misfortune of oth-
nal vision of social justice that were closed off to che nonaecumulating sec-
ers for prívate gain.
tors: growing prosperiry through hard work and individual effort. And rhey In [hese circumstances the superior attitudes and modern pretensions of
emphasized precisely those dimensions of el pueblo's original vision thatthe rhose who mediated berween the local and nacional communiry were cast in
nonaecumulating sectors "forgot": hard work, savings, thrift, sobriety, indi- a new light. Local people felt compelled to resi-r rh I 1 'r rve hehavinr
vidual morir, and so on. yl j f these for -r d ro distinguish themselves as clearly as possible from_
- To distance themselves from che social obligations and entanglemenrs that nu
everything associated with Che merchan[ rea ten-no t onl y aecumul auon ,
posed a threat te cominercial accumulation, che merchants began to shun "greed," and impersonaliry, but also che corrupt political system that was so
che very social and cultural practices so recendy democratized by their non- directly involved in unleashing these forces on che local population. In so
accumulating peers, [hose ehat expressed che new collective identiry of che disrancing themselves, local people seized en che very "backward customs"
nonaecumulating sectors. The merchants went so far as to ridicule tírese ex- that che merchants had used to ridicule them, bur in che process were able to
plieitly social practices as "backward customs" and to identify themselves invert che terms of moral legitimacy normally associated with "things back-
with "modern," national culture. The merchants arrribute their own pros- ward" and "things modern." For in light of the highly negative attributes that
periry to their "modern" work habits and orientations. And they explain their carne te be associated with che merchant realm and beyond, the sociality,
nonaecumulating neighbors' inabiliry to prosper by pointing to [hcir "baek- interactiveness, and also the innocence and naiveté of simpler folk who em-
ward customs" and their "laziness." bodied a simpler way of life could easily hold che moral hieh grrn,nd The
Because merchants have come to oceupy tic key role of eommercial media- modern world did indeed threaten this simpler way of life, these more pri-
tor between region and srate, local people's most direct, lived experience wnh mary and enduring identificarions (of family, communiry, and friend), as thc
nacional integration has come to e em o ie y t e ir experiences wir t e merchánts themselves had taught them. For Chis very reason che modern
merchant sector. And because oí che way that relations of exc an e ame lo world had to be contained, controlled, and carefully monitored."
be struetured in che reeton-as clearly differentiated from che social and ma-
terial life of [hose in che nonaccumuláting seetors-[hese people's in[erac-
dons wwith the m e vr ti t characterized by impersonaliry and even Conclusion
hostiliry. The for oi al sis underscores the fact that states of imagination nave
With che disintegration of che national economy that began in the late distinct `time and s ace e ' istinet ways of conceiving of and organiz-
1970s, itself heavily eonditioned by the crisis in international eapitalism, ing temporal and spatial realins. In closing, 1 would like to compare che time
mueh of che local populace faced seriously deteriorating living eonditions. In

THE MODERN PERUVIAN NATION-STATE 279

278 STATES OF IMAGINATION
 and space effects of the three states of imagination discussed aboye to assess poyanos' response verged on ehe "postnational." They attempted to withdraw
the impact of each on the territorial integriry of the nation-state. Two points froto the broader national context, if only in cultural terms-to erect spatial
are at issue: the conditions under which local populations imagine them- and temporal boundaries between themselves and a modern nation-state per-
selves to be a part o£ or "outside" the historical imaginary of the modern ceived as "alien." The fact that the local populace looked to themselves as a
nation-state; and the centrality of "moderniry" in understanding contempo- I moral alternative t £_a nationally romnmmised sstate of affairs and in the pro-
rary social processes.
lt yy nlo cesssought to inscribe their identiry toco space, reflects their effort to "cor-
The compromised nation-state of che pre-r93o period was in essence pre- A cq,s rect" the failure of the nation-state. Chachapoyanos attempted to divest the
t sitori_L nnnitla lw nren ^rion I Th A (/ L .,.-..
tea, . at s, central rulers m the nauonal fiv 'state not only of moral authoriry (t iossi^ se l s) but also of terri-
capital had little if any direct control over everyday life in regions like Chacha- tonal integrity (to regula e movements and activities of these persons).
poyas, had virtually no abiliry to impose a truly national orden of social cías- i It is important to recognize, however, that Chachapoyanos' spatial and
sification. Instead, local elite rulers oversaw a social order that classified ✓k✓>' 4.e
temporal retreat froto [he modern nation-state, though in one cense explicitly
people according to nonnational criteria; insisted on their "right" to control ^('c'^a+antimodern, is in another seo se anything but that. It is true that Chachapoy-
the movements of goods and people within the regions they
monitored; and anos naturalize their connection to a distinet and autonomous place that is
refused te relinquish these "rights" to the central regime.pymptomatic of the said to be a moral alternative to, as well as to he beyond the reaches of, the
preterritori -'
, prenanonal character of the compromised nation-state was the nation-state.. It is also the case that Chachapoyanos essentialize dleir mem-
following
hen leaders in the national capital truly threatened the ability of bership in a community that is said to be premodern , unchanging , and time-
local elit s to monitor local affairs (or when nacional leaders were especially less
, one that represents a moral alternative to the corrosive and alienatin g
weak), it was not uncommon for local elites to seek to secede, to establish forces of modernity. It is revealing, however, that the equality, commonaliry,
their own territorially distinct, independent polities (as occurred in Chacha- and consensuality, said to embody the essence of this remote, premodern
poyas in the 188os and in the Ig2os, foz example).'° Neither national nor lo- communiry only emerged (discursively) with the breakdown of the elite order,
cal leaders approached the territory of the nation as a foregone conclusion. which divided the population into mutually antagonistic and warring groups
Rather. the issue of territory was a "problem" to which there were many pos- whose members would acknowledge having nothing in common with each
s ibl e " so l uti ons. "
•^\I . other. In other wocd
As a result of the "cultural revolution" of the Ig2os, however, and the emer-^ ral communit nfCha
est of the citizenshi
gence of che liberated nation- saec , t th ese con d
rnons changed. Inc nation- ^ based public sphere brought into being by el pueblo during the modern era
state became truly national and territorial for the first time, as the local popu- of the liberated nation-state. What che naturales nf ('haohanrn , oc rail rr dition
lace vested in central rulers the moral authoriry and the political right to is moderniry . Thus, even as ir appears to recede ,
moderniry continues to de-
classify social persons ( as citizens ) and te regulate their activities and move- fine rhe a
l ra _neters of contemporary social processes.
ments within state boundaries Symptomatic of the newly territorial nature of
che nation-state was the following. In political struggle, the territory of the
nation did indeed become a foregone conclusion, as did the integrity of the Notes
state apparatus. Political warfare carne to be waged only over who would con-
This essay was prepared for States of Imagination, a conference hosted by the Centre for
trol the state apparatus and the given territory of the nation. Efforts to recon- Development Research, Cope nhagen, Denmark. February 1S-15, 1998. 1 would l ike to ex-
figure territorial boundaries or to redefine the nacional polity (through seces- press my gratitude to conference organizers Finn Stepputat and Thomas Blom Hansen for
sion) ceased. inviting me to participate, and te conference participaras for their valuable questions and

By 1980 nacional and global economic crises undermined the state's abiliry commeuts. In addition, I would like to thank Catherine Besteman, Toni Biolsi, Constantine
Hriskos, and Mary Beth Milis for their insightful comments on an earlier version of thc
to maintain and order a viable national economy. As the state lost its territo-
essay. The researcli on which Chis essay is based was generously funded by che Henry L.
rial integrity in eeonomie tercos it lost the abiliry to sustain el pueblo. Chacha-
and Grace Doherry Charitable Foundation, Sigina Xl, che Scientific Research Socicty, thc

280 STATES OF IMAGINATION

THE MODERN PERUVIAN NATION-STATE 281
 MacArtln: fovndrrion r U>ctonl 1ell hu Progum rhe I'tdbuglit IIays Ore Dounral of th u t . ... patean sauna drat had fnrmcrly hect : tognizrt'. in d: mwn (Sm lu.,n dr lo-
. che Colh) College Soei:.1 Caballeros and 6t;i luan de los Ind:osl hy a single o,rona (che A_ra-: 'a Anuntal relchrntcd
f rogram Ahruad, che N mor .I 'rae r for ene 1: mmniries
bc all e l s rtfoom . : mil es of hn l.e, eund t-t :99",
Se;enct 1A r¢ Cnnuulre tlrc (.016 (cll gc In.crdr iphnary Sntdic s Cnnt Program. thr
e suli, al t", ..... a Ihe l n : A .emhle
rn. r.m ',, ni 1in> :e, : AdoLrv cl..m<ICe ..: nrr (]'c:.1 n:ndarien hrr Ara:hn:-
u(.A n vonaa. a un l : d .id uxlq rt u ,e thiis s: 'ang hora ls s : thc
.logieal he,caíeh. 1 1aa9 allt' aele........ I,c che n ,r,', suppor1 ol all [ti,,, insitw iun.^
ar[isans; f''-1 roeatitt ter'dh a oc tiro:n; yi spot[ [.leen eiuhs for }
Unleos arherv ur r d dl nut.lnloi ar mne.
ahIte. whieh esv:hliehed un arena iu 'h`nd:e naw thm-s otdei n¢-i pinte nh di ag-
.hc dixautsc i )f e t'r_prz e e neeruu [ t u ,,, ntodern iry, and narion-seae dtsctuscd
iued h e u Iv in th n srspape An e e u:d : : r c p d ue 131 n h h^ o:l
in uta }ag that tolic n d thara l r t rhecnIn out f h ch dais 'ti', taren
.ayo u r n I51 s d clubs. 161 r eh t r hui or nos ra 'r te caeh- r r1, 11, se..
•ntcrged ptrtains m th 9805. Theta : esfau-s oslice anda c:meactitur toevard
Nugunie)9a chap 8). Ir:- impottrne.e enpha ieetln bistre: rlr'e.phe vt r :'h P,'
Thc ove uvltel ndng ensplt eis, howcvcr. vt'as
pase and present expresscd during th: time-
coorestrd dotnain. Che movemms te of d enso imane: ora soughr co foro[ rooper li e sr -:nr
in 19go, rise rise of a new governnsental alud poloieal-
as deseribed in che tcxi- Reginning
zations citar could establish communip' control over rhe public sphere Thc central govern-
economic reo me in the nacional capital resuleed in a eransformation in the relationship he-
ment, on che other hand , soughrtu block rise @srman'so of rhese oigan 1aadooe and to mabe
cween region and narion - state . A diseussion of this more recent period is beyond the scopc
the public sphere a highly individualized domain. The government did nos hesitare m use
of rise present es say.
violence m pursue ice ends. A more detailed analysis of chis struggle m define che nature of
z. Lazo is nos dite actual neme oí rise person in quesrion.
che public sphere is che subject of a work in progress ( see Nugent n.d.).
3. This was a period of rising prices and rapidly Lalling real wages for most people in
13. As mcntioned above ( see n. t2 ), since the 1930s tuero has bcen ara ongoing struggle bctween
Chachapoyas.
local social clemente and rhe central government tu define die nature of tire public sphere.
q. Severa ) women madcet vendors mld inc virmally identica1 stories during chis period.
Most of the local population has favored group controle over economic and political pro-
Yolanda che police used tear gas te break un a
S. Severa ) years alter my diseussion with Sra .
ceses, whlle che government has favored a highly individualized pubGe sphere . The imple-
local protest march led by market women and small bodega operators who were protesting
mentation of structural adjusnnent - rype policies beginning ni che mid-t g7os has setrled mi,
wlsolesale price fixing by the merchante ( see Nugent 1996).
). The conse-
matter definitively in favor of che government (and rise merchant forasteros
6. Considerations of space preclude anything bus rhe mosr cursory examination of [hese as-
quences hace been devastating for most Chachapoyanos.
peets of public discourse.
14. This perception that Chachapoyas was endangered by trae broader national context was ag-
7. Discourses of eorruption have long bcen of interese to social seienris ts ( see Douglas 1966).
gravaced by the emergence of revolutionary ntovements libe Sendero Luminoso and the
In recent decades , etimas have been made tu address the explicitly political dimension co
Movimiento Fevolucio nario Tupac Amaro ( MaTA), and by che governnentS brutal response
these discourses ( see Chatterjec 1986, 1993 ; Gupta 1995).
8. Modemity was associated with rhe merchant realm in the local imagination in large part co [hese movensente,
[ of Pero in x985. Im-
because the merchants went tu great lengths to afliate themsclves with what they charac- t5. Alan García , leader of tlte APRA political parry , was elected presiden
mediarely afer the election sume people in Chachapoyas were hopeful that García ' s election
see below).
terized as " modern" work habits and cultural orientations (
would recule in positiva changes for eheir region . The policies of che García government,
9. A diseussion of why it ir s pecifically' 7ews" who hace cotice to symbolize what is regarded
however, did nothing tu alter che commereial monopoly enjoyed by rhe larga-scale merchante
as rhe alíen nature of commerce is beyond che seope of chis essay.
to. Tisis alternative stmcture of govername included individuals whose posicions mirrored [hose from Celendín , and therefore did nothing to improve che living conditions ofthe malority of

, and thus repli- che local population . Nor did García ' s government succeed in preventing the deterioration
of che departmeneal prefect, provincial subprefeces, and disrriet governors
of state institutione in use context of me war veith Shining Path and che censure of che
poste of che executive branch of government. Ir included as well individuals and
cated the if anyrhing tu
of election, who repli- intemational economic communiry . As a result, sise García regime did Infle
corporate bodies ( mayore and municipal conncils ), chosen by menos
complete with councilors for eaeh of che arcas change Chachapoyanos ' view of the naron-stare as alíen.
cated the armero re of municipal government (
16. Individuale now in rheie seventies and eighties inherired a view e f the movement of rhe e88os
included senators and congresional deputies, chucen by
distinguished hereiN. Ir even
as une whose inrent was to Form ara independcnt pofity. The documentation from die t88os
mearas ofelection, who naveled to Lima and presentad ehemselves to Congress as che legiti-
concerning tisis event (much of which has disappeared) simply records the fact chat che
mara represcnmtives for rice Department of Amazonas (see Nugent 1989).
juntas de pro- pre fect was forced to Hee Chachapoyas in che Pace of a regional upticing. After a month oe
n. Key in tisis regard were: ( p el pueblo's donination of municipal councils ,
so, during n+ ¡en he narshaled in ilicary Forces in rhe neighboring depnrtinentofCaja marca,
desocupadas ( which nade federal money availablc to localities for infrastructured improve-
el pueblo's ability to elect the prefect rcturned to Chachapoyas wieh a Iarge contingent of sol diere - canon , and so ora.
ment), newspaper publicatiun , and school board membership ; (a)
From this vantage point he was ablc to regain control of che regio n.
"eheir owñ' to Congress , initiating what people refer to jokingly as che era of "cholo polir-
el puebio s democracization of religioue practico, in particular rhe substitution
cians": and (3)

THE MODERN PERUVIAN NATION-STATE 283

282 STATES OF IMAGI NATION
 URBANIZING THE COUNTRYSIDE State formation has been theorized in a number of different ways. V ery
crudely, I distinguish among thre different approaches that have emerged
Armed Conflict, State Formation, and the Politics
from non-political science disciplines. One is the "penetration" approach,
of Place in Contemporary Guatemala
which emphasizes the infrastructural and administrative aspects of state for-
mation. They depict the process as an often violent penetration of the territory
Finn Stepputat
by army, bureaucracy, and capital. The penetration establishes networks and
nodes that (in principie) facilitare the centralized control of flows of goods,
people, capital, information, and images (e.g., Lefebvre 199a; Giddens 1985;
M. Mann 1988). Social and physical landscapes are organized, leveled, and
straightened out so as to permit the state to "see" what is going on within its
territorial boundaries (Scott 1998), thus constructing an "abstrae[ space" (Le-
febvre iggl). Here we could also include Marx's idea of the "urbanization of
the countryside." Marx focused en the commodification of space, that is, the
creation of a (nacional) market for land, but he also linked this process to
An incident experienced by a former colleague of mine from the UN buman
the effects of "enclosures" of privare properry, the concept of vagrancy, and
rights mission in Guatemala brought honre to me the point which is at che
the disciplining of the labor force (see also Giddens 1985).2
center of my argument in chis essay. While sitting outdoors in a village in
A different approach focuses on the cultural aspects of state formation,
the Cuchumatán Mountains, my colleague was chatting with a villager about
that is, the production of new subject identities through the categorization,
the armed conflict, peace negotiations, and human rights. "Rights?" the man
regulation, and routinization of everyday life and encounters between citizens-
said. "Oh yes, I've heard about rights. Over there they've gor rights, fve heard
subjects and state institutons. The central question addressed through these
them once, there," and he nodded across the deep valley in the direction of
theories is how the state is rendered natural, taken for granted, or even legiti-
the town, la Concepción, the administrative center of the arca since colonial
mate to the subject-citizens, and how the state monopolizes the power to
times. "But 1 am not cure whether we could bring them hete [traerlos hasta
acá]."' categoriza people and to make categories stick (Corrigan and Sayer 1985).
Thus, while the state is associated with the generation and sanetioning of
The universaliry of the rights of the citizen wherever he or she may roam
certain, more or less namralized orders, "the secret of state power is the way
is one of che hallmarks of the modern narion-state. Against this background,
itworks within os" (200).
the aboye depiction of rights as being place-bound stands out as being some-
There are many different answers to the question of how this "internaliza-
what particular. It points toward a broader perception of the modero state as
tion" of state power comes about, but the answers tend to adhere to a "dif-
located in cities and towns, while the population in die res[ of the nacional
fusion" model of power, in which people are reduced to objecrs of central-
territory is represented as living under a different set of conditions. The spa-
ized power: the only option left for [hose who are subjected to state power
tially differentiated representation of state and citizenship poses che question
is to reproduce or resist che power (Latour 1986). Foucault's analysis of the
of how the stare is territorialized and how different segments of the popula-
disciplinary techniques employed in the garrison, the school, the clinic, the
tion relate to the state asan idea and a set of institutions (Abrams 1988). To
church, and the family is a good case in point. However, bis laten work on
approach [hese questions, che presenr essay explores processes of state for-
government gives more scope for the active involvement and even compliciry
mation during thirty-five years of armed conflict in Guatemala, a confict that
of subjects in processes of state formarion, in particular in relation to lib-
formally ended in igg6 when the government and the guerrilla organizarion
eral and developmental states (Foucault lggr). Techniques of government-
Unidad Revolucionaria Nacional Guatemalteca (uRNG) signed the accords
counting, taxing, mapping, reporting, naming-are not necessarily monopo-
for a "firm and lasting peace."
lized by state institutions. They may circulate in different regional, nacional,

URBANIZING THE COUNTRYSIDE 285

 thcy scrutinize, which leads to certain limitations in their aualytical perspec-
or transnational networks and be employcd lar government in sites outsidc
rive. Whereas rhe flrst tends to ferishize rhe state as an actor, it shares s ith
the tate institutions. 'Chis is particularly so under the neoliberal regimes
tire second approach a rendency to regard state formation as a top-down
where citizens and groups of citizens (including "indigenous people") are
from-rhe-cantar-ounvurd prucess I suggest that he con bining the three ap-
encouraged ro devclop auronomy and sclfgovernment" Eut still, state ínsti-
proaches, rve muy he able to go beyond unir of their lrmitations. In the lol-
tutions seek to monopoliza the power to authorize and coordinare the use of
lowing 1 explore how processes of state tormation hace articulated with lo-
[hese techniques, as discussed in the inttoduction to chis volume.
calized stmggles and emergenr projects of change in rhe nertll uf Guatemala,
Although tire idea ot government or governancc opens a theuretical space
ar dte bordee widr Mexico. Pased on severa) field studies from 1994 to 1gg6.
fiar subject-citizens as agents of state formation, ir is hard te) get bcyond the
the essay focuses on tire period of tire armed conflict in Guatemala tiran die
"¡ron cage" image of modern power, to actually grant them an effectiviry as
early rg6os to 1996, arguing that rhe Mayan villages in this region and in this
subjects of power, in particular when they do not directly challenge or resist
period have been increasingly urbanizad in rhe serse dlat they have been
state power (van Krieken Ig96). In this regard, it is important to analyze how
turned inm "sites of governance." Different actors, including the villagers
techniques of government and images of state articulare with techniques and
rhemselves, have eontributed to chis development, which also has placed rhe
ideas of sclf-reflexivity and emancipation. Embcdded in processes ofmodern-
ization, this larter set of techniques, praetiees, and ideas is not directly linked villages within an extending fleld of politics in Guatemala.

to state formation, but they may produce "projects of change" that eontribute
m che formation of state-eentered orders (Asad 1993). State Formation and Armed Conflict in Guatemala
The third approach focuses dleoretically en ageney and projects of change
The armed conflict in Guatemala may be sean as three to tour decades of
in processes of state formation. This approach has developed through studies
rapid change, occasional outbursts of armed resistance, and rhe steady devel-
of stare formation in localized and historical contexts, of what we may cal)
opment of a counterinsurgency state, as Carol Smith (lggo) has named it.
state formation through rhe "politics of place." These processes have buen
The key moment of this conflict is the quasi-revolution and the subsequent
explored by anthropologists who hace discovered the state as part of their
counterinsurgency campaign in 1981-1983. In 1981, a coalition of guerrilla
localized fields of research. Peter Sahlins (1989) and David Nugent (1997) are
movements had achieved momentary support from a substancial number of
but rwo examples of studies that show how people at the margins of existing
Mayans in rhe Guatemalan countryside who baeked a counterhegemonic proj-
jurisdictions have contributed to the extension, territorialization, and central-
cct of "the poor"-peasants, workers, and students-against "the rich"-
ization of rhe state by producing or supportingstate orderin specific localities.
One underlying assumption is that the territorial nation-state, at least dur- basically the landed elite, "their" army, and "their" state. The military govern-
ment reacted by establishing a disproportionately fierce counterinsurgency
ing the twentieth cenrury, has been contingent on the containment and fixa-
program, which brought about numerous selected killings of community
tion of the population in visible and governable places, and that these places
leaders and organizers, massacres of endre village populations, and the or-
have been formed, in part, in relation to a centralized state. 1 suggest that the
ganization of rhe rural population in armed "civil patrols for self-defense,"
approach of the politics of place will give us an idea of rhe ways in which
dle PACS, with Glose to une million members in rhe Guatemalan countryside.
projects of change have emerged in specific struggles over resources, entitle-
Discussing how tire conflict may be interpreted in a historical perspective
ment, and political control in the making of localities. Such struggles over
of Guatemalan state formation, Smith (1990) presents the view that political
rhe definition of "rhe context of localities" (Appadurai 1996) are to some
life in Guatemala is constituted by a struggle between "Indians and the State,"
extent strucmred by existing cleavages of, for exaniple, class, region, eth-
a struggle that has formed both the state and the Indian communities. fhus,
nicity, and gender, and the relations between centralizing elites and those
the army's counterinsurgency campaign was carried out by "a weak but des-
deemed "locals" by the central elites.
potic state that attempted tu eradicate rhe bases for rhe autonomous Indian
Rather [han being different approaches as such, the ways of studying state
communiry once and for all" (21). Smith explains rhe auronomy of rhe Indian
formation are distinguished by the kinds of processes and subject niatter

UREANIZING THE COUNTRYSIDE 287

286 STATES OF IMAGINATION
 community-` by way of this historical opposition in which rhe Indian com-
"order of the city" to the totaliry of the territory in the form of an "infinite
munities emerged as corporate units engaged in political struggles with the
network" (Foucault 1984) akin to Lefebvres's (rggr)"abstract space" through
state since the conquest. A basic premise for the apparent unity of the Indian
which society could be seen and ordered.
communities is the establishment of territorial administrative control along
The statelsociety duality was enforced through the use of different tech-
previous territorial divisions through the church and the missions, rather
nologies of government, in particular in the late nineteenth century. The lib-
than through an alliance with an indigenous elite engaged in the indirect rule
eral land reforms, which privatized and individualized land, togetherwith the
of the conquered (21).
conceptualization of vagrancy, the laves of forced labor for those who did not
In spite of its weak "infrastructural power" over territory and popula-
possess (enough) land to be considered productive, and the regulation and
tion, the colonial state managed to establish a hegemony through the church
surveillance of the towns, became instrumental in the territorialization of the
(Smith 19go; Van Oss 1986). The postcolonial state did not. The infrastruc-
state and the construction of state, society, economy, and politics as separare
tural power of the postcolonial state, that is, the capacity of the state to "ac-
spheres.
tually penetrate civil society and to implement logistically political decisions
Together, the labor and land reforms produced a dualized society: on one
throughout the realm" (M. Mano 1988: 5), developed considerably as Guate-
side, property owners with political rights; on the other side, "docile bodies"
mala was inserted into the expanding world market from the late eighteenth
who were subjected to forced labor on private properties in the developing
to the early twentieth cenrury. The nineteenth-century liberal onslaught on
plantation economy. In other words, society was divided between citizens
the Catholic Church provided means for the formation of a national army,
and subjects. The citizens were the "white" landowners and the local "non-
which took over the Church's role as representative of the government in the
Indian" (ladino) elites in the towns who mediated relations between the (In-
countryside. Revenues from coífee exports made it possible to extend the
dian) subjects and the state. The dual society also had effects in terms of
modern infrastructure to larger parts of the territory, not least the telegraph,
identity. As McCreery (1994) has shows, the well-known hierarchical and di-
which was essential for the military control. In northern Huehuetenango,
chotomous identiry construction of ladino-Indio (as superior-inferior) was a
schools, town halls, and water supplies reached the towns-the capitals of
particular product of the epoch of the liberal reforms and the positioning of
the municipalities-beo ,een r8go and 1g,o (Recinos 1913).
ladinos as intermediaries between Indian laborers and white landowners in
However, according to Smith (lggo), the state never achieved legitimacy
the coffee export economy.
within the Indian communities. Only through coercive means could the state
Returning to Smith's position, there may be a problem of interpretation as
control the communities, and attempts to impose new institutions on them
well. She insists that the state, given the autonomy of the Indian communi-
were not successful. Smith concludes: "The fact that all such attempts to
ties, could impose institutions only by force, whereby the state became even
assimilate or incorporare the Indians into the nation in the postcolonial pe-
less legitimate. However, she ignores that "state" may be perceived in differ-
riod have been forced rather than urged on them, has led to further loss of
ent ways in different contexts and moments, and that people may hold seem-
state legitimacy in Indian communities" (17). Below, 1 return to this discus-
ingly contradictory ideas of the state as "sublime" and "profane" at the sane
sion in relation to the army's civil patrols, the EACS.
time (see Hansen, this volume). To many Mayans, tata Presidente (grand-
Smith's position points to two problems. Epistemologically, she (with many
father President) has been an ally in the struggle against exploitation and
others) tends to project the modero concepts of state and civil society into the
abuse by landowners and local representatives of the government. Also,
past as if established by the conquest (Ronsbo 1997: 59). Thereby, she disre-
Smith's position tends to ignore that in many cases Mayans have striven with-
gards how this duality developed in the political thinking in the region from
out success to become fidl-fledged members of the national society.
the late eighteenth century, when the state emerged as an entity to secure
Thus, we may conclude that the exclusion of "Indian subjects" coincides
wealth through the free acts of individual subjects, and society emerged asan
with the construction of state and society as distinct entibes, and that this
independent process to be disciplined and framed through politics (65). In
exclusion should be seen against the background of the relative inclusion and
the same period, political elites conceived of the project of extending the
protection of Indians in the colonial forros of government (Van Oss 1986).

288 STATES OF IMAGINATION

URBANIZING THE COUNTRYSI DE 289
 central srate lince tire liberal reforms, che space heyond die towns has been
Che test of che story. from che liberal la lid reforms to che rggos, may be inter-
controlled only to a limited extent, despite tire tac, [liar tire codihcadon of a
preted as a story of how che Mayaus and others hace negotiated che condi-
level of administration below rhe town cook place in clic nineteenth cemurv.
tions of dieir exclusion and their posible reinclusion. And it is che story ol-
In an act regarding municipalicies from 1879, tlrc Barrios goverrimenr ga'e
how che projects of individualiziiig land property and of establishing an "in-
che orden dhat deput mayors in fincas (privare properties, haciendas) and
finite network" chat facilitares che nsrimtional gane oí` che state bate bcen
hamlecs must be appoinred by rhe mun icipalit}' on che basis of suggestions
carried out. Still, in che wake of che armed conflict, development agencies
from "owncrs and possessors." Chas, redecting che dual socicry of che liberal
lomear die continuing imporrance of communal forros of land tenure and thc
reforms, che acc refers to rwo groups of population: che white or ladino own-
ambiguities of documcncs aud existing land registers that cause permanent
ers of land at che fincas, where their tenants (mozos-colonos) were living and
communal and intercommunal conflicts over land."
working beyond rhe reach of the state, and rhe Indian "possessors" (notown-
The strucmres for state formation were in place with che liberal reforms of
ers) of communal land in che hamlecs (quoted in Ochoa Garcia 1993: 34)
che late nineteenth century. But it was still a project. The effects of che land
Although the number of villages increased slowly over che decades, the
refornis in terms of individualization of communal land have bcen gradual
presente and control of che central state apparatus remained episodic and
(Watanabe 1992; McCreery 1994). The capaciry for surveillance (Dandeker
dependent en che goodwill of the depury mayor co give information. In most
iggo) has been limited by the central state's dependency en alliances with
outlying villages, oecasional visits by land surveyors or che treasury police
local elites andlor the subaltern opposition te local elites. As Watanabe (1997)
were the only direct contaet with state instimtions before the armed conflict
mentions, the province of Huehuetenango employed three policemen in the
of che 1g80s. Usually, direcc encounters with scace institutions took place in
late nineteenth century; ocherwise, °public" order depended on ladino mili-
the towns or che cides, and che compulsory obligations of villagers to partici-
cias, privately employed armed men at che privare estates, and locally ap-
pare in work turnouts in the public works, co work as voluntary policemen,
poinred volunteers taking turns during one year of service, a system thac still,
and co supply firewood to teachers and authorities were directed at service in
in che rggos, funccions in the highland villages (and in some towns).
The dictatorship of General Ubico from 1930 tu 1944 represented a new che town rather [han in che villages.
This, however, has changed since che 1970x, and in particular during che
move toward the strengthening and centralization of the state apparatus.
armed conflict. There has been an increasing contradiction between che ex-
Ubico engaged heavily in road construction, and che better part of the roads
clusive definition of nation-state and citizenship and che gradual acquisition
connecting che towns in Huehuetenango were built due to che vialidad, che
from the 1940s en. Also, their
law of forced labor in road construction. During che Ubico regime, the rep- of land on che parí of Mayans, in particular
resentatives of the central state in che cowns, che intendentes, were appoinred engagement in export crop production, in che commercial sector, and in che
textile industry has eontributed to increasing contradictions between their
among nonresidents and sent to the towns by che president. This opened up
the possibility of new negotiations and alliances between che central state and position as producers of wealth for che nation and che narrow de facto defi-
nidon of citizenship. These contradictions fed into che armed conflict as che
the Mayan elite of che mwns against the local ladino elite. Among older May-
attempts in che 1g6os and 1970s to acquire policical influence at local levels
ans, the image of Ubico is not as negative as the one construeted by histori-
through formal, democratic procedures were blocked by che elites.
ans. He is seen as muy stricto (very strict) and a just man who did not differ-
enciace between Indians and ladinos (see Carmack 1995: 218); his laws (the
vagrancy laws and che vialidad) applied to both groups (K. B. Warren 1978: Agents of Village-ization
149). This image of order, justice, and state authoriry was no doubt some-
In Chis section I consider some of che contemporary agents who have sought
thing that the army tried co resuscitate during the comtterinsurgency
to put villages and hamlecs onto che political map of modero Guatemala.
campaign.
Although cheir goals have been different, they have all engaged in tire noble
Whereas towns increasingly have come under che institucional gaze of the

URBANIZING THE COUNTRYSIDE 291

200 STATES OF IMAGINATION
 art of forging or resisting change by means of engaging people in an abstract institucional network that provided an abstract authorization of new modes
community with a common cause, whether modern Catholicism, armed revo- of being human, the Ac activists and communal lay priests employed and
lution, or national security and development. Before these, other projects ar- spread new "techniques of che self" (Foucault 1984) to develop a new set of
ticulared cleavages and identities that contributed to the production of vil- ethical practices around the notion of human dignity and equaliry.
lages, most notably the liberal land reforms (Davis 1997) and the political and Although Catholic Action was formed as an anticommunist movement that
agrarian reforms of the populist interlude 1944-1954. Regional variations are supported che military governments of the 1950s, it turned to a more critica)
huge, both in terms of the timing of the process and in terms of the agents position as parts of the clergy approached che program of liberation theology.
who have been active in changing the conditions and positions of villages and The movement became very involved in politics of place, articulating genera-
hamlets. Since the 1g6os, however, the combination of the growth of the rural tional as well as ethnic struggles over a communal hegemony embedded in
population, local conflicts, and che intervention of national actors has turned che liberal construction of state and citizenship, and in che spatial hierarchy
villages and hamlets roto increasingly visible and discrete loci of development of city-town-village. In Nentón, the North American Maryknoll mission en-
and "governance."'
couraged young leaders from the villages to present petitions for land and
The following sections focos on the development in the municipality of other resources at the offices of the central administration in the capital, thus
Nentón, a poor area on the northern side of the Cuchumarán Mountains as cireumventing existing patron-client relationships. In other regions, the activ-
they slope down roto Mexico. The area is dominated by cattle farms and ista of modern Catholicism engaged in issues of communal socioeconomic
corn-producing communities from where the inhabitants routinely migrate development, in particular by forming cooperarives or creating village mar-
to Mexico for agricultural wage labor. The capital, Nentón, was connected to kets that bypassed the control of che ladinos. Only the latter was rhe case in
rhe national nerwork of roads in Ig8o, which was extended into che hinter- Nentón.
land by che army's eonstruetion company, and under armed protection during
che rg8os. Half of che population was displaced at some point in the early The Guerrillas
rg8os, and che last armed encounter took place in 1995.
In the 1970s, the new guerrilla movements srarted m work in villages in cer-
tain areas in their attempt to organize che population against che govenrment.
The Catholíc Church
Moving from the periphery toward the center, as prescribed in their strategy,
One of che first agents to move beyond the towns was che Catholic renovation rhe guerrillas sought to gather military srrength and establish their "rear
movement Acción Católica (AC) and related foreign missions that founded parry"-their social and logistical basis-in areas with weak presence of the
centers and built chapels for che indoctrinization and "awakening" of the state, and of che army in particular (Payeras 1991: 18). In the case of Nentón,
population at village level. This process, baptized as "village-ization" (aldei- recognized as a ladino rown (although many old people still speak Mayan
zartón) by Ricardo Falla (1978: Si8), may be seen as a decentering tactic in languages), che guerrilla strategy produced a spatialized image of che resur-
rhe context of the struggle against che syncretie, popular Catholicism that rection as taking place "in the villages" and threatening che tocan.
beeame "tradition" (costumbre) as che reform movement tnanaged to hege- The guerrillas entered and to a large extent used che modernizing struc-
monize the practices and beliefs of modern Catholicism, known as "the reli- tures established previously by che Church, for example, by taking over the
gion." Struggling against che existing hegemony of che elders who were category of che "responsible," che leader of [he village group of activists, and
locked in parran-elient relationships with powerful town ladinos, che reform by establishing corresponding local clandestina eommittees (ec[s) that or-
movement gathered young potencial leaders at week-long seminars in che de- ganized the collection and transport o£ food, metal, money, and young men
parrmental capitals or other central towns and sent them back with new vi- for che "fighters in che wilderness." According to narrativas recorded in
sions and new capabilities in the arcas o£health, education, organization, and Nentón, the guerrillas tore down che characteristic barred wooden doors to
pastoral practices. Apart from involving converted villagers in a translocal the village jails: "We used to have a jail with bars but they tore it down: [The

292 STATES OF tMAGINATION URBANIZING THE COUNTRYSIDE 293

 attempt to vro rhe hoarts and minds of the population by recognizing social
guerrilla saidj 'He who commits a crime will meet Chis one, and he pointed
and econonric inequaliry, and promises ca substantial changos in rural living
to ltis rifle. Now rhere is no jail, 011 arras''' (feld notes 1996). f hey also
discouraged ano arteutpt of the iAluyuns to parchase land. although this liad conditions. Organizad civil defense parrols and "local development com

been a ntajor aspiration and srrugglc in cvhich many' Mayans had been en- Iees- were the nteans tor establishing >ccurity and dev elopmern -cspec-

gaged during the late ig7os: ¡he land belougs to overybody, rhey said. tively, at the subtown level. As a decree from 1983 saos: "It is convenient ro
establish a system of coordination at the narional, municipal and local level
Also, they por pressuro 011 the vil lago mayors (auxilliares) to make them stop
to make the actious of the staré s msI1rution5-aimed at real'zing a progrim
collaboratíng wirh the tocan mayor. In sum, thcy attacked those struemres and
symbols ruar conneeted rhe villagers ro ehe state apparams. At the sane tinte, of reconsrruetion and developuent- reach the smallest connnunities in rhe

the guerrillas soughr to construcr an altemative state by installing "people's rural arcas" (Linares 1988: 235).
This turn became an importan[ element of the army's identiry as the only
tribunals" and themselves as surveyors of order and justice; by wearing uni-
forms, the symbol of aimed men's loyalty ro a state (see Dandeker 1990); by truly nationalist and effective sector of ehe state, differem from rhe corrupt

registering births and deaths; by installing a comprehensive intelligence ser- police, the corrupt politicians, and the lazy state employees in offices, and
different as well from the unpatriotic oligarchy. A civil affairs officer in Hue-
vice; and by imposing taxes, drafring young men, and engaging adult men
huetenango explained to me, "We are Guatemalans, cae belong to the people.
in modern bodily exercise (drilis). As noted by Wagner (1994), disciplinary
The majority of os are indigenous-the general is indigenous, and 1 walked
techniques are floran exclusive properry of modennization offensives "from
barefoot as a kid. Look! 1 have the hair of an indigenous! ... If there is a
aboye"; [hey are as often appropriated by agents of modernization offensives
conflict in the countryside, it's our dury to enser as mediators to make peaee.
"from below."
The National Police are nor well structured for Chis, [hey don'r reach the vil-
Unlike the "real" state, the guerrilla state was a porrable, partly invisible
state with a minimum of physical infrastrueture and an unlocatable center lages. They stay in the cides. The soldier, on the other hand, leaves his garri-

somewhere in the wilderness, from where authorization of representatives, son. He only stays there fiar three or four days after having been on patrol. He

rituals, actions, and ideas emanated. This state was legitimized as a redis- provides security" (field notes 1994, 1995). Thus, the army leaves the bounded

tributive state in favor of the peor, a stare that represented popular sover- "power containers" (Giddens 1985)- the cides and the garrisons-and ac-

cignry. But as the conflict intensifled, the practices of rhe guerrillas became companies the real (indigenous) people in their villages and hamlets and,

inereasingly authoritarian. In Nentón, at least, they never engaged in rhe beyond that, in the wilderness (la montaña).
The army represented itself as an egalitarian, nationalist force of change
fields of education or health. Meetings, tribunals, organization, and the daily
and development. In Nentón, people tell stories about the strict colonel who
drills were the only rechniques with a bearing on the formation of new sub-
jectivities in which the guerrillas engaged. In these other fields they rested on broughr order to the town, organized the population, and saw ro it that every-

the involvemenr of associated agents of modernization, such as progressive body had equal shares of the relief provisions. No favoritism was accepted.
The teachers from the town were sent with the army to the villages, where the
teachers, missionaries, peasant unions, and the cooperativa niovement. But
in Nentón most of [hese left the region when fighring got fierre in 1981. population was ordered to (re)build schools. Although the presente of the
arny has been disastrous to many villages, it should non be neglected that
the very fact that urban authorities, landowners, developers, and others make
]"he Army
themselves present in the villages may legitimize existing power differentials
The next village- izing actor to appear in the villages was the army, which,
in the eyes of many rural dwellers. It takes some suffering to overcome rhe
faced with massive subversion , liad to reconceptualize its military strategy
and organization . Following the experience of the French army in the Alger- hardships of travel in the countryside.
Apart from organizing state penetration, the army employed a repertoire
ian war, " the Guaternalan army mimicked several aspects of the guerrillas'
of different techniques (known as "psy-ops"), which aimed at transforming
strategy, including the establishment of a presence at the village level, the

294 'SIATES OF IMAGINATION URBANIZING THE COUNTRYSIDE 295

 che Indian subjects. As a civil affairs officer in Cobán explained: "Our obli- bctween che villagers and the army, which allowed che former a good measure
gation lis] to make them see che situation. Indians are very susceptible, they of autonomy, provided that everybody in the village agreed. The communal
are easy to ply, just like clay" (R. Wilson 199D 47). Some of che techniques authoriry reigned over individual members, and dissidence was always dan-
did not differ from [hose employed by the Church and che guerrillas, such gerous for the individual as well as for the community.
as confession, ritualized conversion, education, physical training, and drills. The violente had an explieidy disciplining effect: "Before [when they
From che villagers' point of view, some of these interventions are seen as worked with che guerrillas], people thought of themselves as being very
civilizing, as expressed, for example, by a village leader in Trinidad: "Nowa- brave, but suddenly they became very humble," a village representative ex-
days we are somewhat refinedltanned (curtido) by all che advice we have had plained co me. But the patrols may also be interpreted as an inclusion of che
from che Church and from che army" (field notes 1995). The notions of "ad- villagers in the surte, as a kind of recognition. After all, the state armed the
vice" and "ideas" are uniformly used to conceptualize a specific (pastoral) civil population, which is an excepcional and potentially counterproductive
relationship between the modernizing agents and che Mayans. In Chis respect, measure. McCreery (1994) mentions that the Indians in most cases were ex-
che receivers do not distinguish among different agents. cluded from the nineteenth-century ladino milicias in the highland, but dur-
In terms of surveillance, the civil patrols extended considerably the army's ing che armed conflict of the 1g8os, ladinos left the zones of conflict in large
capacity. As 1 have argued elsewhere (Stepputat rgggb), the effect of the mas- numbers. Beyond the towns, only Mayans, grosso modo, were present.
sacres was co produce displacement and a spatial dichotomization of the rural Although the regular, physical presence of rhe army alongside che villag-
population between che network of villages, which became organized and ers-no matter how faz from tire roads-often cook the form of threatening
armed, and che "wilderness," a space beyond control in which the people intrusions and measures of control, che patrollers also received training, "ad-
encountered were defined as subversives. Anybody who was not from your vice," and direct linkages to a range of civil state institurions that were all
village was "unknown" and should be detained. Thus, the villages were con- under army control. Although the notion of citizenship would be misleading
strucced as discrete spaces, isolated from one another. Even known persons in chis eontexr, as che patrollers did not have any individual, constitucional
were unknown: a ]ay priest from a village in Nentón told me that he had once righrs, we may perhaps coin the subject formation as "corporate patriotship."
been detained in another village en his way to Nentbn, "with pass and ID card This formed the basis of che hegemonic "army-peasant parinership" in which
and everything." He knew che head of che patrol in Chaquial because they had che patrols may be seco as forming both parí of che counterinsurgency appa-
attended the same courses for ]ay priests in Huehuetenango and asked: "But ratus and che communal hegemony of che villages (Kobrak 1997: 17), which
why, Juán? You know me. 'Yes,' he said, `but we don't Imow what you think.'" enabled che patrollers to embody che state at che frontier." They produced
Inside the villages, surveillance worked as well. Because of the danger of surveillance among themselves and were collectively rewarded with symbolic
being accused of being subversive by fellow villagers, everybody tended to act inclusion in che nation-state, with certain liberties chac gave them material
in unison: "What one did, everybody did. Nothingwas done in secret [a escon- benefits as well.
didos]," as a patroller explained (field notes xgg6). This village was in prin-
cipie turned into a disciplinary space, a human panoptieon where the villagers Deuelopment Agencies
themselves undertook che disciplining. One could argue that small face-to- Development agencies have operated in Guatemala since che 195os, and since
face communities always have an element of strong (normalizing) mutual che early 1970s, they have increasingly targeted che rural Mayan arcas. De-
surveillance, but che new element of che armed villages was the installation pending en che accessibility, che potencial for production, and che actions of
of an exterior authoriry among the patrollers themselves. When people re- other agents of village-ization, there are huge regional differences in che de-
turned to their village from exile or displacement, theywere taught che lesson: gree co which communities have received atrention from development agen-
"Here you have ro follow che custom [el costumbre]," that is, Here you have to cies. Thus, a growing guerrilla presence in particular arcas in che late 1970s
patrol like everybody else. The patrol could also work as a "buffer institution" elicited development action from che government and USAid, and alter 1984

296 STATES OF IMAGINATION URBANIZING THE COUNTRYSIDE 297

 gerous to get mixed up wirh. Now, rhe employees "come here to serve. 1 he
rhe counterinsurgency program (e .g., "beans and rifles") was accompanied
previously excluded subjects in rhe villages Nave become entitled ro public
by come, although limired, dcvclopmenr minan ves.
selvices, which in general tercos are defined as "developmenr." If rhe villagers
Before rhe logos. Guatemala was non a signiheant recciver of aid as con,-
embark on rhe narional developmem projecr. they are included on a collectivc
pared to other Centrad American countries. However, as thc peace process has
basis as "community," while rheir codificd representativas become "civil so-
gained momentum, rhe armed condice has dwindled. and refilgees and dis-
cien,.- lf we take a look at rhe practices of development agencies. ir appears
placed peoplc hace reserrled in rhe tormer arcas of conflict, governmental and
thar die "community" is given priori[} in relation to villagers as individ ual s.
nongovernmental agencies hace focused rheir actions in [hese arcas. In Nen-
First of all, according ro rhe umvritten criteria oí severa] NGOS in rhe carca,
tón, a weil-known focus of armed conflict, a formar patrol commander and
mos[ agencies "do non work with privare fincas, but only wirh comunidades,"a
lay priest (and supporter of rhe present neoliberal government) made the fol-
notion that in chis eontexr refers to villages or hamlets." Second, agencies
lowing observation: "Before, rhe governments were only a saying [de dicho].
organize the communities for collective appropriation of rhe development
We learned who won the elections but nothing else. The governments nevar
projects through publicly registered committees and representativas. This is
gave rheir parí to the people. It was not like rhe Mexican government that
non only a technique for rhe formalization of rhe interface berween the rural
gave people their roads, their water, and their electricity.' Now it's different,
population and rhe development agencies; it is also a technique for rhe unifi-
the [public] workers come hace to serve. Before, they were only scolding os;
cation of a (temporarily) divided eommunal subject: "The most importan[
they cama to kill people, it was the slavery. Now a person has value, there is
thing is lo strengthen rhe unity of rhe community, ro creare a communal
respect. Now they take os finto account" (field notes 1996).
spirit," which is based en "tradition, rhe cultura of cooperation, and eom-
In this account, the exclusive predatory state has been substituted by a
munal work." In the postconflict context, rhe development discourse presents
more benign, apparently inclusive onc, for which the provision of security
rhe divisions of community caused by war, displacement, and mixed resettle-
and welfare lo the citizens is its reason ro be. The employees of the state have
ments as the most importan[ problem to salve. The villages of Nentón are
always acted wirh contempt roward che Maya; they have acted as superiors to
considered to be "the worst in tercos of eommunal cohesion," and the model
them. In the vernacular, the exclusive governmentaliry is conceptualized as
village of Chacaj, designed and reconstructed by rhe army, is "su artificial
lack of "respect" or laek of "rccognition" (no nos reconoce or no nos toma en
cuenta). Phenomenologically, chis governmentaliry is embodied in the occa- community.""
Development projects rarget entire communities or formalizad groups
sional face-to-face encounters in which rhe Mayans have had to carry rhe state
within them. An interesting counterexample is the case of rhe scheme of in-
employees on rheir backs across rhe moumain ranges; have had to bring them
dividual credirs for "rhe productiva reintegration of repatriares" (FOAELAP).
food and firewood; have had to cake off rheir hat, bow their head, and look
Each household was offered a credit of U.S.$t,5oo (r,ooo daily wages in ag-
to the ground when ralking to ladinos in general and state employees in par-
ticular; have had lo put up wirh arbirrary taxation and humiliating violence, riculture). After two years, fine rates of reimbursement in three villages in

although they have had some space for negotiation with superior (central) Nentón were so peor that new repatriates returning lo these villages were
denied credirs until rheir fellow villagers had paid rheir due reimbursements.
authorities (sea McCreery 1994).
As an angry and disappointed repatriare said, "They only offered us other
Now, however, the villagers note a changa in the actitudes of state employ-
projects such as latrines, vegetables, and reforestation. Iris non just. Everyone
ces and modern state substitutes, such as NGO workers and employees of
[should llave] his own account, everyone [should Nave] his own life." In this
internacional development agencies. In rhe villages of Nentón, 1 metfew May-
case, rhe recurrence ro rhe "community spirit" veas sean asan imposition.
ans who inade the distinction benveen governmental and nongovernmental.
According to rhe representativa of rhe E.U.'s integrated rural development
To the villagers, rhe agencies and rheir people were all "employees," licencia-
project in Huehuetenango, the (re)construction of rhe community spirit con-
dos, and ,instimtions," which is the indigenous terco to designate externa]
stituted rhe first step in the creation of a "municipal spirit" and, eventually,
entibes wirh resources to be solicited. Instimtions are explicitly contrasted ro
a spirit of nation. In the mid-rg9os, the new line of thinking among trans-
organizations, which are "something among ourselves" and potentially dan-

298 STATES OF IMAGINATION URBANIZING THE COUNTAYSIDE 290

 national NGOS and intergovernmental agencies involved in the arca was to
bership not only by birth but also by participating in communal labor obli-
mercase sustainabiliry by developing townshipwide associations of commu-
gations in the public works and by accepting appointments for communal
nal commirtees and representatives that managed communal credit schemes,
health promorers , dispensaries, barefoot vets, and so en. The institutions en- of lces (cargos; literally, burdens) in the struggle for, and organization of, com-
munal development projects. This is indicated by a conflict between the vil-
visioned self-sustaining region- and nationwide networks of communal or-
ganizacions . As comed by Michael Woost ( 1994 ) in the case of Sr¡ Lanka, lage of Aguacate and the settlement of returned refugees in Chaculá, in Nen-
tón. In Chis case, returned descendants from Aguacate who held inherited
the agencies wanted to develop a "nation of villages." But unlike Woost's
description , in which governmental agencies produce (ideal) villages as the deeds to ideal parts of the communal land were denied access to land on the
grounds that they, unlike the resident adolescent men, had failed to contrib-
basis of the nation-state , postconflict rural Guatemala is characterized by a
ute to the communal development projects. As suggested by Cancian (1965),
multiplicit,v of governmentally and nongovernmentally induced networks of
villages. one reason behind the downfall of the political-religious hierarchies (of the
costumbre) is that they eould not aecommodate and give status to the in-

Forming "Sites of Governance" creasing number of young men in the communities. The committees, repre-
V ery briefly, here 1 attempt to characterize the composite effects of the ac- sentations, and promoterships required by development projects serve this
function, as 1 have shown in the case of the Guatemalan refugee settlements
tions of different agents of village-ization, including the villagers them-
in Mexico (Stepputat 1992). Thus, it seems that the fiows induced by devel-
selvas, as the stabilization of the village-communiry as a "site of governance."
opment projects and public works push toward a sharper and less inclusive
Heuristically, we may describe the stabilization effects as spatial, social and
symbolic. definition of membership, and make it meaningful to reinvent the hierar-
chies of rotating offces of the costumbre. The infernal struggle to become rep-
The spatial stabilization consists in the unambiguous delimitation, mea-
resentative of a group or a communiry is also intensified, and apparently a
surement, and registration of landed property or possession in relation to
"community career" is an asset for further ascendance in the greater sociery,
authorized communities with a well-defined group of members (vecinos). The
as, for example, an employee in a development agency, a politician, or an
proliferation of land conflicts indicates that the boundaries between indi-
entrepreneur.
vidual as well as communal properties are poorly defined and registered, and
The symbolic stabilization takes the form of a remolding of the settlements
in many cases double residence strategies, migration, and shifting alliances
in the image of the town (and the city). As the infrastructure of the public
make it difficult for the agencies involved to establish an unambiguous body
services is installed, the ideas of cantar, public space;- and public order ma-
of knowledge about membership of communities involved in conflicts. In a
terialize, if at all possible, around a central park or plaza, el parque. Some-
simation where entitlement to land is negotiated rather than defined through
times, the central park is even provided with a fountainlike construction when
property relations, the agencies involved take "historical and social factors"
the water supply is installed. This urbanization is not necessarily imposed, as
into account. Thus, belonging, identity, and conflict trajectories of displace-
in the case of some model villages (e.g., Chacaj in Nentón). The design ex-
ment and suffering become resources for the negotiation of entitlement.
presses the striving of the village elite to become full-blown members of na-
Furthermore, the issue of unambiguous definirion of land has importance
tional society. When talking with representatives, the image of a checklist of
beyond the land itself. In practice, only setfled eommunities that are not en-
symbolically loaded public works emerges: "The development committee has
gaged in conflicts are eligible for the development projects, in particular for
struggled for five years now. We've achieved the new school, the clinic, and
the development infrastructure (schools, clinics, roads, water, etc.). As a rep-
the new `depury mayor building' [auxíliatura] with an office and a hall for
resentative of a group of presumably displaced people declared alter the ritual
meetings. The only thing missing now is the [covered] market.""
measurement and circumscription of their new "urban zona": "Today we are
Often, the checklist also includes a concrete basketball field in the central
happy, because now we can work with the instimtions."
park, not to mention electrieirylpublic lighting and the pavemenfof emerging
The social stabilization consists in the tendency toward a definition of mem-
streets. These latter features mark the present in the town of Nentón, but they

300 STATES OF IMAGI NATION

URBANIZING THE COUNTRYSIDE 301
belong te the (vision oí tire) fume in the villages. A significan[ absence from about.... Is your country a socialisr country?" Obviously, he was talking

che checklist is che Church. dm kev symbol of early colonial governmenrality about nacional and inrernational politics--and 111 general, the refugees' cu-
riosity and willingness te Icarn ubout politics and named pohncal systems
in ncc Indian towns_ Religion has beconte 1 conlested issue with thc prolif-
arenad the world were impressive - bur his depietion ot politics as a specihc
crarion of 11011-Catholic missions and churehes, in particular during tlie
field of knowledge and practice that "they" had entered and were esploring
aruted conflict, and it is no longer ro be raleen for granted that dte Church is
more el less collectively secros to offer a fruirful approach to the analesis o
centering setdeittents,
Boa owiug from Karen Fog Ohvig's (1997) notion of "cultural cites," we political fornls and dynamics in relation ro localized processes of state for-

suggest approaching Chis process of stabilization as a production of "cites oí mation and conflict.

governance.- " Olwig defines cultural cites as "cultural institutions which As we have seco, che process of suite formation in Guatemala has been

have developed in the interrelationship between global and local ties" (17), contested, far from linear, and not always centered in the institutions of the

such as the localized but ever changing family land and house thar are re- governing elite in the capital. The process has articulated various visions and

imagined and recreared by othenvise migrating people in che Caribbean. A versions of how individuals and communities should relate to che state, which
form the state should Cake, which elements of modernization state institu-
"site of governance," then, is a spatial, social, and cultural matrix that or-
ders tire interface between citizens-subjects and transloeal institutions. This tions should further and which they should curb. Hence, political struggles
have developed. In tire Foucauldian sense, the techniques and rationalities
is where subject-citizens must position rhemselves m claim rights or entitle-
thar produce che state as an effeet do so by simultaneously producing econ-
ment. It is also one of the sites where images of state and citizens are pro-
duced and consumed. In eomparison with Gupta's (1995) notion of tire "nec- orny and sociery as fields to be governed and framed through politics. As an

essarily localized instantiations" of tire transloeal state, sites of governance increasing number of subjects, places, and everyday practices are brought

do not necessarily comprise representativesloffices of state institutions. A within the purview of state institutions, the points for potencial disagreement
and contestation multiply. The central questions are, however: Who are in-
certain order and governabitiry is sufhacient.
Furthennore, we may notice that che communities as sites of governance cluded and formed as political subjects? Which kind of processes and prob-

are mutually exclusive in the sense that citizen-subjects in principie must be- lems are defined as political? What means of power do those excluded from

long to one community only, the community through which they can claim the political process command, and what is their relation to politics?
During most of the history of the Guatemalan state, politics as well as state
their entitlementlrights. Many of tire problems encountered and generated by
institutions have been che turf of che ladinos. As mentioned aboye, che liberal
die intervention of che development agencies in postconflict dynamics are
related to the murky business of defining unambiguous belonging. In che land reforms and other techniques invotved in che formation of che coffee
export state excluded the Indians from the emerging political domain, and
cases of confliet that 1 have followed in Nentón, che agencies were notable to
generare precise and reliable knowledge of who people "really" were and to those regarded and defined as Indians have been able to enser tire political
field only by crossing certain boundaries, that is, by leaving their communi-
which community they belonged. Communities emerged and disappeared as
ties and "redressing"-literally and symbolically-as ladinos. These ethnic
alliances and conditions changed over time; people moved in and out of cate-
boundaries and the ambivalence they have engendered have left their mark
gories (displaced, local, repatriares) and communities.
en the forro of political struggles in general and en che struggle of che "In-
dians" for reinclusion and recognition as more [han cheap sources of labor
State Formation and politics in particular.

lit 1988, a forlner lay priest and cooperative lcader from northern Huehueten- On one side of che divide, che nacional elite has been ambivalent in its

ango who had sought refuge in Mexico told me about che experience he had dealings with che Mayans. In spite of a liberal project of extending che order
of che ciry to che totality of che territory, of individualizing and privatizing
gained through conflict and exile. He depieted "politics" as one important
element of chis experience: "Little by little we have learned what politics are landholding in che name and spirir of che modera nation-state, che nacional

302 STATES OF IMAGI NATION URBANIZING THE COUNTRYSIDE 303

 elite has continuously contributed to che reproduction of a system in which has been vital for che population ro be able to distinguish berween instimtions
the Mayans existed only as corporate subjects with communal landholding, if and organizations, berween necessities and politics.
any. When Mayans finally began to engage in politics, they were excluded by At che other sida of the nacional divide, Mayans have had an equally am-
all mean if they were not strongly controlled through clientelist nerworks of bivalent relationship to their inclusion in che nation-srare through politics
the elite. In che late 1970s, when the army systematically began to persecute and to che ciry, che paradigmatic place of politics and che nation-state. Mayans
communiry leaders, organizers, and other organic intellectuals, "politics" have identifled [hese with che ideas and ways of che ladinos, who were posi-
was a central element of the complex of labeis used for identifying those who tioned as intermediaries berween che state and che Indians in che countryside
transgressed che boundaries. Thus, their abilities ro read and write, to speak of che postcolonial state. Since then, politics and che state have been che turf
Spanish, and hence, tu engage in politics were che criteria stated for perse- of ladinos. To engage in politics has been delegitimizing for most Mayans,
cuting them. One refugee remembered how an army oficer had interrogated who saw politics as opposed to sineerity, honesty, and humiliry, che attributes
him and other cooperativa leaders, asking them, "'Do you read and write?' of che real Christian, that is, che ideal Indian as involved by che colonial mis-
('No!') 'Of course you do. We don'r want you te be experienced. You have a sions. In che countryside, che politiqueros (a derogative term for [hose who.
lot of experience, you speak Spanish, you know how te write, you have poli- engage in politics and negotiate with che representativas of che central state)
tics in your heads.' The colonel was very angry: 'It's your fault that we have to are powerful but godless ladinos, che heirs of che Jews (che Spaniards) who
come hora.'"
killed (che Indian) Christ (sea Bricker 1981). To give but one example, a rep-
As che power struggles left behind the subtleties of decades of a "war of resentative of a village where che inhabitants negotiated che conditions of
posirion" and moved finto the terrains of the "war of maneuver," it became return from Mexico with a group of refugees characterized his opponents in
outright dangerous to be associated with "politics" if you were peor andior che following way: "They are very politicianlike, they are not sincere. We like
Maya. When the refugees returned from exile, che army warned the villagers honesty, sinceriry, chis is che most importan[ thing; but they are not straight
in the areas of conflicc: "Be careful. They bring their organization, they bring [reales]. They want to cake advantage of us."
politics." As an army officer told me in 1994, the refugees liad been "ideolo- Owing to [his combination of che powerlculture constructions of the
gized" afrer having spent many years abroad. In general, che army identifled nineteenth-cenmry nation-state and che conditions and stigmas introduced
politics with foreigners or foreign influence that had polluted che Guatemalan by counterinsurgency, politics was still, in che 199os, considered a dubious
countryside, and in particular with "organization," the hallmark of the popu- affair in che villages. With che policy of decentralization and che distribution
lar movements of che 1970s that defined organización as che universal means of lo percent of the state budget to the municipalities, however, che inten-
of forging political change.
sity and interesas in municipal elections have increased to a degree that by
The dangers of being associated with "organization" and "politics" have far outdoes nacional politics. The nacional political parties are involved in
lefa their marks en política¡ life in che former arcas of conflicc. When the municipal elections for practical, not ideological, matters. They are mean
nacional labor union, Unsitragua, in 1996 began to operare in che villages of of playing out che competition for leadership within communities and the
northern Huehuetenango, afrer international and nongovernmental organi- scramble for access to public resources and symbols among villages. Thus,
zations had paved che way for more tradicional política] organizations, a com- che stabilization of village-communities through public works is very much a
muniry leader consulted me about che identiry of [his entiry: "Do you know if parí of [his political fleld.
this union is an institution oran organization? We are preparing a group of
che union. Let's see if ir brings something good. It seems to have some foun-
The Polities of Place
dation. [The village of] Yuxquén solicited grains, and soon afrer, eight sacks
of maize and beans arrived. They have offlces they are not working secredy. Although municipal elections, associations, and translocal identities are gain-
Tris not some politics but [about] necessities [no es ninguna política sino necesi- ing importante, che village-communiry has been at che eenrer of política] ac-
dades]." Under che polarizad and dangerous conditions of political conflicc it tion during che 199os. With che governmentalization of che Guatemalan state

304 STATES OF IMAGINATION

URBANIZING THE COUNTRYSIDE 305
 vas rrying to make a ¡¡si of children for a vaccination campaign. Likewise,
in rhe former arcas of conflicr , village-communities are i ncrcasingly engaging
during the iggos, runiors identified certain institutions (such as rhe U; N mis-
¡ti relations with state or otherresourceful institutions withour defining rhese
sion MINUGUA) with "che sigo of che beast666," or warned villagers
relations as political. ' 1 don t Uvanr lo make politics , 1 just want lo improvc
against human vampiros, dtoleros, in rhe shape of unknown ibreiguers.
this place' is a vete common expression of indivi duals who engage as com-
lo rhe degree thar politics has become a weaningful held of action for
muniry organizers and in termediaries henveen rhe community and rhe mayor,
villagers, the reversal of erhnic and spatial hierarchics rends to legiti ni ize po-
development fdnds, NGOS, credit schemes, ministries, international agen-
litical cngagement in and cbr the villages. In particular, ovo fearures have
cies, and other institutions . They become registercel a nd authorized as repre-
marked a break with prior clienrelistic relations: clic civic Co nuairtees and rhe
sentatives , promoters , and committees ; rhey receive credentials and ID cards
villages-againso-the-town alliances. The civic committees are presented as ex-
from rhe institutions ; and rhey often use their con nections with one institu-
ponents of a legitimare "politics from here" that contrasts rhe usual —poli-
tion to further the interests of their community (and themselves ) in relation
tics from [here," that is, che politics of rhe nacional political parties thar are
lo other institutions or communities.
dominated by ladinos and based in che capital. In northern Huehuetenango,
Apart from this kind of " new clientelism " ( Palmer 1998), communities en-
for example, civic committees llave been promoted from rhe town of Jacalte-
gage in a number of other ways in communication with state institutions.
nango in an attempt to creare a regional, pluriethnic identity based on a village-
One of [hese is (threats of) violence . Within a postconflict conjuncture, when
centered, sustainable, and culmrally sensitive development in las Huistas,
che eradication of political violence becomes a marker of transition , violence
comprising five municipalicies at che Mexican frontier.
is an efHcient means of gerting che attencion of governmental and ínter
Within munieipalities, che town/village divide was often decisive for rhe
governmental institutions ." Occupation of land or government offices, kid-
ourcome of municipal elections during che r99os. Given che limired funds of
nappings , roadblocks , and threats of communal violence may be interpreted
rhe mayor, he has to make a choice becween, for example, paving more strects
as ways of achieving access co resources or to negotiations over rights and
in che cown or supporting che construction of schools and clinies in some of
enddement. In che case of Guatemala , many such actions are coached in a
che villages. In che (numerous) cases in which ladinos have been in control of
language of rights, equality, and order. One incident of threatening behavior
che town and che public services while che Maya-speaking population in che
in Nentón , for example , was legitimized as an attempt lo halt what che per-
hinterland have worked in che construcrion of public works in the town, the
petrators considered a disorderly resecclement of former refugees : " They are
town-village contradiction has been framed in che idiom of ethnicity. In rg92,
about co build their houses on chis land , without due demarcation , without
in Nentón, che first Chuj mayor was elected by en antitown coalition of vil-
landmarks, without documenta .... We need co see a map, we have to have
lages that protested againsr rhe previous "all-for-rhe-town" policy of public
licenciados and engineers measure che land." The action and its framing is, of
works. In che municipalicy, Chuj-speaking villagers held: "Now che mayor is
course, simational, but these are nevertheless calls for the kind of order srates
natural Idee us, he suppores che indigenous people. He fulfills Chis promises]
provide.
while che Spanish let us down." Anche time of chis interview (1995), che ex-
In orher cases , Chis kind of order is rejected , either in a political language-
plicit Mayan identity had non yet reached this region to any significan[ de-
exemplihed by che former refugees who have returned from Mexico with an
gree," but che ethnically informed political engagement was evident.
explicitly antistatist ideology (Stepputat ig99b )- or by way of roadblocks,
threats lo government officials, rumors , or conspiracies of silence that are
used co sea¡ off communities from che gaze and control of state or other State Formation and rhe Urbanization of che Countryside
institutions . Ideas of local sovereignty, backed, for example, by narratives of
Postconflict Guatemala, much like posrapartheid South Africa, has been
die heroic struggle of che civil parrol againsr che guerrillas while rhe armywas
turned into a laboratory for social and political science. Competing theoretical
absent, may be invoked co legitimize resistance to institucional intervention
and methodological approaches are brought together in che same arena in an
in what is considered lo be communal affairs. In one case in Nentón, an NGO
attempt co unravel rhe meaning of che conflict. What happened? How and
worker was denied access lo a village and accused of heing Ami-Christ as he

URBANIZING THE COUNTRYSIDE 307

306 STATES OF IMAGINATION
 why did it happen? And what are the effects of the conflict for the future of
with the army, villagers in the 199os engaged in the definition of the condi-
Guatemala in general and the rural arcas in particular? As Arturo Arias re-
tions and forms of services that the state should provide and the negotiatiml
marked, a review of books en the conflict reveals that the discussion therein
of limits to state control and surveillance.
of the transformation of Mayan subjectivities during the conflict tells us more
We may suggest two elements behind the apparent change in the villagers'
about the positioning and theories of the writers than about how the Mayans
position vis-á-vis state and nonstate institutions in the arca. On the one hand,
position themselves (Arias in Hale 1997: 825). The present essay, informed by
Che increased engagement of state institutions in the villages has multiplied
theoretical debates over the intersections of power, space, and culture, runs
the points of contact and possible contestation and negotiation. On the other
the same risk. However, the theoretical debates do not Cake place in a vac-
hand, if we regard the resistance toward the state asan exception from a more
uum. In the late r99os the population of Guatemala passed through a mo-
consistent, longue durée effort on behalf of the Mayans to become recog-
ment of "intense need to remember" (Hale 1997), and among scholars who
nized and included as full-fledged members of the nation, the current em-
have taken part in this moment of remembrance and interpretation of the
brace of the state appears as a continuation of a process that was initiated by
armed conflict, there is an explicit cense that "the state" in Guatemala is a
the exclusions of the liberal reforms in the nineteenth century and inter-
seriously underexplored phenomenon. There is an acure need lo understand
rupted by the armed conflict. In chis perspective, the armed conflict and the
the state.
struggle against the state may be explained by the frustrated attempts to be-
The present contribution argues that processes of state formation may
come full members of the nation in previous decades. As people in Santa
fruitfully be explored from the vantage point of the politics of place, of the
Eulalia told Shelton Davis around 1970: "Yes, we are naturales, but we are also
ways in which, in specific locations, populations become recognizable and Guatemalatecos."
governable and states imaginable and effective. This approach helps os get
The exploration also shows that nonstate actors have been instrumental in
beyond the state-centric analysis by addressing the question of how a range
bringing about preconditions for the villagers' creative engagement with state
of actors contributes to the formation of state as they articulate national,
institutions. The modernizing movements of the Catholic Church and the
transnational, and regional flows in processes of place making.
guerrillas, who gave explicit attention to the village level, provided organiza-
The empirical exploration of state formation in a former arca of conflict in
tional capacities, languages, and forms of reflection that in the presentjune-
Guatemala suggests that three important changes have taken place during the
ture of state formation have enabled the villagers lo align, lo some degree at
1980s and 199os:
least, with "the institutions." We are talking about the modern forms of gov-
r. The reach of state instimtions has been expanded territorially from the
ernance embedded in committees, meetings, negotiations, representations,
towns lo the level of villages and hamlets, not least through the presence of
planning, projects, and other practices that have not been developed through
the army.
the curriculum of the national education system. Through their engagement
2. State practices have become increasingly governmentalized since the
in the modernizing movements, villagers llave also developed a modern lan-
moment the army succeeded in imposing control over the armed insurgence:
guage for the critical assessment of the kind of order installed by state
"Now the employees come to serve us," as villagers noted in the mid-199os, institutions.
thus observing the change in governmental rationality.
Although in principie this language permits them to enter formalized po-
3. Social spaces, which are defined or authorized by state institutions, con-
litical fields, "politics" is almost exclusively focused on issues related to the
stitute sites of negotiation and struggle £or segments of the population who
"urbanization" of the villages: provision of public services (education, health,
were marginalized or excluded from effective citizenship before the armed
transport), symbols of urban lile and order (fountains, parks, plaques, monu-
conflict. In contrast lo the strategies of the guerrillas and their followers, who
ments, sports facilities, markets), access lo financial institutions, andan em-
basically sought te destroy or mrn away the existing state, and to the civil
bryonic industrialization.`" The scramble for attention from the "institutions"
patrollers in the r98os who tried lo defend themselves by allying themselves articulates identifications and cleavages between Maya and ladino, town and

308 STATES OF IMAGINATION

URBANIZING THE COUNTAYSIDE 309
village, benvecn villages, benveen competing leaders, and between individu- Notes

als and village-eommunities. fhe prime ohjecrive of the villagers is not nec-
1 apprcci,ue rhe eonstrnctivc commenn from "I houms Blom Maneen. sudipa I(at ira 1. 11, udk
essarily to order lile ni Che loculities according Lo the order of rhe state, yer R,amho, iadrel Siedar CaroLA. S nirh. NSona lóense . m Fiona P'll
rhe ensuíng politics oí place tends ro lave rhe effecr oí stabilizing rhe villages onccnvise mted, all Iranslatiou, are mine.
as sites of governance. 'f he process, however, is marked by rhe continuing i. Antltropologie t lolt n I'inclla, MINI Guau. personal comncnicatiott, 1996.

wealaresses of stare insriturions in rural Guatemala. rhe ambivalente of rhe Giddens and Fouca,lit ideas on tire inaritutianal aspeet uf errar lorm uion (i.u.. tbe
decclopntcnt of Ihe admini.rr.!ti e .tpparatus:tt I he' r.lpacin fei +ucuill:_ne 1 ¡Pace
village population tira-h-cú rhe order oí the srare, and, nor leasr. the ambiva-
I ryyól h-i eing on Aa e:, cn^ph t taus b)te' th, tl tnotioe.: ,^^ m rhe t ür.t-
lence of rhe whjte elite and the urban ladinos toward rhe Mayans. As Rachel
tiou of space and time (latid and labor) preeonditiou the emcrgence of a nr sahlatic.tt
Sieder notes in her contribution to this volume, the urban population in
na tional i e nt.
Guatemala rejecred proposals for ncc devolution of certain judicial powers to Unlike rhe South Miau studies tradition, "commuuiry" in tire I atin American smdies tradi-
village-level authorities.'° tion is usually undersrood as organized at rhe leve¡ of une colonial "Indian towns' or, altor
lf we are to judge from rhe experience in the former arcas of conflict in rhe independence, rhe municipaliry.
most isolated parts of Guatemala, rhe village-level population have inercas- 4. For example, rhe E.U. dcvelopment project in Huehuetenango, ALAI9r, could inveet only
ro percent of che funda designated for irrigation because of irregular la-nd tentare during Use
ingly become members of rhe nation-statc. However, for many practical pur-
mid-rggos.
poses, such as access Lo credit, social infrastructure, and development proj- 5. Only in rhe tesos did tire population of die highlatide surpass the estimated number of
ects, they are being incorporated as a kind of corporate citizen, as members preconquest inhabitants of [hese arcas. Tire population growth and local condices have
of a territorial commnniry that has been authorized as a site of governance. sparked a process of duplicarion of authorized serdements.
'fhus, their recognirion as citizens depends era rheir neighbors. As stated by 6. See'frinquier (1964) for en analysis ofthe experience by a French officer who parricipated in
che counterinsurgency in Algeria. The aechitecr of rhe Guatemalan countcrinsurgcncy pro-
a villager who was giving one of bis days of work to rhe construction of rae
gram, Benedicto Lucas García, was erained in the French army and parricipated in rhe Alger
village school: "Some people don'[ understand that their children need rhe
¡a la war.
school, or they think they'11 have it for free. They dodt understand that we 7. Aceording ro Gustavo Porras, rhe arme even nrimicked the narre of rhe village-level guerrilla
have to give out labor to become recognized as persons." organization , cci., Comité de Coordinación Local, Iater rus Comité de Desarollo Local (per-
Striving to become recognized as persons, to bring rae rights across rhe sonal communication , November 1994).
valley from rhe town to rhe village, seems to be a recurring theme that may 8. See ¿¡se Alonso ( 1995 ) and Stepputat (zoou) for a similar interpreration of frontier dynamics.
give us an idea of rae current state of imagination. We get the image of vil- 9. During Use tg8os rhe Mexican government gave priority te social investment at rhe frontier
te stabilize die houndary between Use revolutionary, peor Central America and Mexico.
lagers striving to make rhe state present, to become included. This image
m. Interview with employecs in rhe Centre Canadien D'etude er de Cooperation hrternationale
is difficult to accommodate within a penetration approach Lo rhe state; it
(cEei), Septeiuber 1994. The' distinction was blurred, though, because some privare estates
presupposes decades of nonrecognition and exclusion produced by violenr had been abandoned by rhe ownerladministrator during rhe confliet, leaving
ir te the tenants

forms of state penetration of rhe territory. But rae discourse of recognirion to work rhe land on their own. The E.U., uN nca, and others did carry out conununal proj-
and rights also presupposes rhe appropriarion of ideas of human dignity and ects he re, sucia as schools and water projects, but only after negoriating with che owner, who

equaliry as predicated by rae Catholic reform movement, as well as rhe idea afrerwzrds rold rhe tenants , "This school is mine, in¡,, water tank is mine," etc.
u- Interview with tire representativo of Use G.U. ALA/9t project, April 1995.
of rhe state as rhe ultimare authorizing entiry. Thus, the patrerns, processes,
12. Public space is undersrood as a historically specific eonstroction of conrmon space, as dis-
and subject matters of state formation change over time and space. Processes cussed, for example, by Sudipta Kaviraj (1998). It is a space thar is connected to the abstraer
of state formation are always working ni and through polities of place. But androriry of che arate with inereasingly codified notions ofappropriate behavioripublic order.
whereas state formation at some junctures is driven from singular centers, In rhe villages of Nentóa, for example, tire idea that drunlmnness is unsnitabie in public
other junetures, such as rhe turren[ one, are characterized by decenrered and space is emerging as an índex of eivic conduct and respectabiliry: "If 1 drink, 1 drink at honre,
diffuse dynamics. These changos are also reflected in out theories of state tranquilamente , wlthout making trouble lbulla].'' Tire misiona, rhe Pmmstant in particular,
llave been promoting thcse codes oí public behavior.
formation.

URBANIZING THE COUNTRYSIDE pI1

310 STATES OF IMAGINATION
 13. Interview with che committee secretary in Aguacate, November 17, 1996.
14. The concept is being codeveloped by che author and Henrik Ronsbo.
IN THE NAME OF THE STATE? Schools and Teachers
15. The eíficiency of violence as a means of access to resources is conditioned
by the public
attention ro che giveri crea and che presence oí resourcefid institutions
wish a peace agenda.
in an Andean Province
16. Referring ro biblical notions of che devil.
17. Villagers referred te themselves as naturales, a colonial category, índigenaus, a posicolonial Fiona Wilson
category related to che 1944-1954 reforni period, oras Mateanos or Sebaseianos
, che highland
towns te where che Chuj-speaking inhabitants belonged ¡ti che colonial period. The linguistic
tetm Chuj was used in che lame way as Mateano.
18. Struggles over access to land is another field of primary importance.
1g. The proposal fornled parí ofa major package of proposals hrz changes in che constitution-
many of them coming out ca` che peace aceord-which was rejected in che referendum in
Iggq, mainly due m che negative votes in che capital.

I want to explore a case study from Andean Peru m reflecr on the ways in
which imagen and practices of che stare are forwarded and communicated,
countered and opposed benveen central government and provincial periph-
ery. By envisaging state policy and praetice as involving specific kinds of spa-
tial flows, 1 wish lo discuss the changing forros taken by state intervention in
the provinces, and pay particular attention tu confrontations wish regard te
education and che extent to which schoolteachers, che largest group in pro-
vincial sociery employed by che state, have been bearers o£ che designs of che
stare in different historical periods.
For che bureaucracy of che modern stare, che sehool has become an cm-
blem that demarcates che territory effectively governed by the state, an insti-
tution that relays ideas about state, nation, and citizen. Nearing che frontiers
of che modern state, che school has taken over functions of the military out-
post. It is che place where symbols of nation are kepí and regularly displayed.
Here, children are taught to become citizeus by saluting the flag, standing al
attention at hearing che nacional anthem, marehing on civic oeeasions, and
learning about nacional history, geography, and ceremonial events. This as-
pect of che school has long been appreciated in Pero: in che ig6os Vásquez
noted that it was "la unica agencia estatal de la cultura nacional" (che only
state agency of nacional culture; t965: 133). The school has become a focal
point through which che state enters local public culture in a more benign
guise than, say, che police. The mission of che school is to produce cirizens
not only by educating che young but also by drawing parents finto multiple

312 STATES OF IMAGINATION

social activi tic s, includirrg observante of nationaleo inmemorations. Notsur- tice and local reaction by focusing on interpretatioas and actions in response

prisingly, Bien, schools and insti in res of higher education have also liad sym- m a key evenr in the school and national calendar, dic celebration of Fiestas

bolic aud practica) significante for amistare and revolutionary groups. Patrias, Peru's national Independencc Day.
in rhe provinces, schoolteachers wen c ar rhe o(¡ ter fringeso f the state; they,
can be envisaged as the fingers olí dic state s long arras reaching down to the
Transnt i ssion orIra nslatio n?
people, enrbodying and negotiating tire blurred meeting polis bettveen state
Extending State Power over Space
and sotierro. liowever, teachers are more [han state employees. Thcy are also
local inrellecnmis, recognized as having die aurhority and responsibility to Much thcorerical work oil die state, as Allen (1999) has pointed out, resrs on
defend and proniote their communiry, tocan, and province. This means teach- the idea of rhe state as ni impersonal organism with inherent authority whose
ers may tale up political causes that bring them finto sharp eonflict with the attributes can be transmitred over space. Ir is a straighrfonvard matter. From
state and its agents. Teachers localize and translate different kinds of cultural Chis it follows that the outcomes of state control in national territory are rela-
fiow that open up new worlds and ways of imagining for the local communiry. tively easy to discern. Strong states are considered lo possess rhe power lo
They are placed in mediating positions, often intensely ambiguous ones, en impose state role ata distante. This necessitates direct force (or the threat
two counts: they act between state and local society and they also link town of it) to extend sovereignry over space as well as a more insidious fostering
and countryside. Rural eommmnities have at times pereeived teachers as du- of structures of feeling to spread ideas of national community and national
bious emissaries, a double-edged blessing that could be beneficial but also imagination. Subject-eitizens enrolled in a nationalist project become in-
potentially harmful and cosdy.' volved in multiple forros of allegiance to rhe state. The space of state territory,
In Peru up lo the 197011, most reachers were drawn from thc dominan[ in the modera state is seen as eonstituting a planning frame in which the
urban-mestizo class. When pus in charge of the new alien, intrusive school in state bureaucracy can count, measure, and map and can instimte policies to
the rural communiry, they tended to act as místis, as mestizos with a strong educate, improve, and develop the national population. Weak states, in con-
belief in their own racial and cultural superiority, constantly mindful of the trast, are defined as unable to extend sovereignry over space, drum up nation-
unbridgeable gulf between themselves and their Indian charges. According alist sentiments, or bring progress or development to the population.
m one reading of Peruvian history, schooling and teaching were channels There are many reasons why models of state power that stress the domi-
through which a deeply racist social order was reproduced. However, expan- nating, coercive properties of states have been emphasized in official histo-
sion of higher education and the greater access of social groups subordinated ries; for one, they record the state's triumphalist view of its past. But models
in the Andean race-class hierarchy tu the teaching profession could destabi- that define state power as located in and contained by the apparatus of the
lize and totally transform Chis relationship. Educational institutions became central state overlook rhe interplay between central state and provincial politi-
sites of new struggles against prejudice and injustice waged in the llame of cal/cultural narrarives and histories and dismiss the notion a priori that state
nationalism and a new moral order, and against rhe broken promises and formation is a process that also takes place "from below." Here the smte's
betrayal of white governments in rhe distan[ capital, Lima. rheroric as lo the orderliness of its administrative structure, clear-cut hier-
This essay explores the conflictive relations between the Peruvian state and archic division of territory, and top-down lines of command paper over the
an Andean province at different periods during the twentieth century as seen complex, mutating arrangements and relationships that weld province and
through the leas of education. The diseussion is framed by reeent debates on state together. The model is encapsulated in a particular ideal-rype rypol-
the namre of state power and how states extend their power over Space. The ogy that, as in rhe case of Peru, produces particular actitudes among ruling
Peruvian state's changing spatial practices and the reactions these provoked groups that totally misread and misrepresent subterranean counterdiscourses
in Tarma, a province in the department of Junín in the central highlands, is at work in the provinces.
then examined. In the final section, 1 explore contemporary state spatial prac- How and through whar discourses and techniques, channels and networks,

314 STATES OF IMAGINATION AN ANDEAN PROVINCE 315

 intermediaries and brokers are images and understandings of state, nation,
glect gave way to enforced, structural reform, heralding a new, greatly inten-
and citizen relayed? Until recently, few attempts have been made theoretically
sified phase of state intervention in the provinces.
to inquire into the dynamic processes through which states achieve a pres-
But the turbulence provoked by structural reform and influx of state agen-
ence in space. To move away from reified, instrumentalist notions of the
cies tended to sharpen opposition to the state. This found expression in the
state, one needs to think through the connections between state power and
growing appeal of Marxism-Maoism in the 197os and eventually to the emer-
territorialiry and explore a different set of metaphors. How-and through
gence of a hard-line, ultra-left Maoist party, Sendero Luminoso (full titie, the
whom-does the state flow over space so that it can reach its outer provinces, Communist Party of Peru, the Shining Path), which promoted a strategy of
filling national territory with practices and symbolic cites? How does the state
violence and terror to bring down the state. Subversion in the Andes was
take on-and how is it given-the authority to coerce, persuade, and seduce not taken seriously by the government in Lima until it had a revolution en
spatially dispersed populations so that they act as subject-citizens and accept
its hands. In the words of Sendero ideologues, the party applied a Maoist
as legitimate forms of governance and relations of ruling imposed from the
strategy to separate the nation from the state, relentlessly attacking the state's
outside? As Gupta (1995) usefully suggests, there are two processes at work.
spinal column to remove its presence by liquidating its agents. It was hoped
One is the production of the state as a translocal presence, a shared imagi- that the convulsiona produced in provincial society would eventually weaken
nation of the state spreading like a color wash across the map and distin-
the head, that is, the government, which would need only a little push for it
guishing one national homogenizing space from others beyond its borders.
to drop off of its own accord. Senderos onslaught provoked a third phase in
The other process involves the implanting of institutions and agents that con-
state intervention: a greatly expanded military presence in the Andean prov-
stitute networks through which messages and directives can dow between inces to stamp out terrorism and the survival of a eentralist, autocratic gov-
central state and province. Color washes and fiows peed to be conceived as ernment that has made very little concession to the demand for a greater
potentially involving two-way exchange, though these may be blocked or in- decentralization of political power.
effective under reigning constellations of power.
By foeusing en the analycic space between state and province, one discards
Taking a historical view, 1 suggest that in Peru one can discern three periods the assumption that states can reckon on ready-made agents or representa-
distinguished by the forro and intensity of state intervention in the Andean tives. They have to be produced. One importan[ site is the school, another the
provinces. In each of these, the state has been represented, imagined, and military. In the business of ruling at a distance, states need to adopt, co-opt,
contested in different ways. Starting in the late nineteenth century, some sev- work through, and rely en hierarchies of intermediaries, emissaries, officials,
enty years after the end of colonial rule, the Republican state began to extend employees, and underlings. Some are socialized and disciplined to act more
authoriry and sovereignty over the territory claimed. It did so partly by dispos- or less en behalf of the state in military and bureaucratic orders that are
sessing provinces of their governing capacíties. This marked a first phase in stretched out over nacional space (often with particular concentrations on the
the spatial extension of state power. But in practice the state did little in the frontiers). But a great many others are not enrolled by the state so direcdy or
way of implementing postcolonial ideals or spearheading policies to bring formally, such as schoolteaehers, minor ofheials, and técnicos. These actors
modernity and progress to the Andean provinces. Postcolonial Pera gave birth are capable of changing face, of moving across and blurring the notional
to an "aristocratic state" whose seat of government was on the coast, not divide between state and sociery, of being tricksters who challenge in myriad
to a "developmentalist state" concerned with che well-being of in; citizens. ways the extent and nature of state power. What happens to these relations
Up te the late rg6os, Peru was considered "one of South America's weakest en the borders is integral ro processes of state formation.
states" (see Lowenthal 1975: 7). When social unrest was deepening in the One needs to envisage those carrying out the work of the state as possess-
Andean highlands and a broad consensos had emerged among professionals ing some degree of latitude and choice, a capacity to translate-as opposed
and intellecruals that the country needed a strong state, one capable of plan- to transmit-ideas about the state and the state's directives, practices, and
ning and executing structural reform, the military took power in 1968 and
messages, or "tokens" to use Latour's (1986) terminology. The translation
established the Revolutionary Government of the Armed Forces. Years of ne- model as proposed by Latour differs in several important respects from the

316 STATES OF IMAGINATION

AN ANDEAN PROVINCE 317
 movements and partics. leachers may boye becn through a nuniher el- so-
amunonly held diffusion model of state power. In the latter, messages issued
by powerbil srates are assumed to be transmitted largcly intact through ner- eializing experiences ¡ti nccir lives thar have instilled dilfcrenr mental framc-
works. ideologies. ideas of personhood Chau i nreract in a nmlripliciq of sv.iys.
works ofagents. 1 hough losing [bree duo re the fríction of space, rhe eontenr
The extension o( Cite stars over pace as borne b1- schoolrrtchers dmretole
of Cite messages has nor becn airered significantiy. AVhat counts nurst ni tite
involves the state in ncrworks and interacrions that it cannot hopo to control.
diffusion model is "the inicial force of those who hace power' ' (267). Follow-
and rhat modity, and may oven reverse, die srare's messages and dcniands.
ing a Fouctuldian lino of argunient, Latour rejecrs Chis static concept ora me
If wc lool< at Cite translation model of power iu relation ro Peru's .Maoist
grouuds thar power is uot something thut can be held or possessed. Stic-
cessfid transmission over space docs not reflect the power of Che inicial im- opposition, especially the cold, clausrrophobic ideology of Sendero ni ira

pulse: "T he spread in time and space of anything-claims, orders, artifacts. attempt to produce a mitror, mimic state, we can seo that messages pro-

goods-is in the hands of people; each of these people may act in many nounced by leaders, tire Ayacucho cupola, have flor been opera to much muta-

different ways, letting rhe token drop, or modifying it, or defleeting it, or tion or transformation. Indeed, enormous efforts were expended by Sendero
ideologues to implant a dogmatic, unchanging, untranslatable doctrine, as
betraying it, or adding to it, or appropriating it. The faithful transmission of,
firm as a religious faith. This can be seco as one reason for the parry's failure
for instante, an order by a large number of people is a rarity in such a model
ro keep its early support. But in everyday contacts, there had to be some flexi-
and if it occurs, it requires explanation" (267). Latour goes further. Because
Che token is in everyone's hands in turra, everyone shapes it according to bis bility and openness if militants were to hope to win recruirs and pursue Che
strategy of separating the nation from Che state. Whcther Sendero leaders
or her different projects: "Instead of the transmission of the same token-
liked it or nor, militaras were taking part in processes of translation, the out-
simply deflected or slowed down by friction-you get ... the continuous
comes and effects of which they were unable to predict-though they might
transformation of the token" (268).
This is a logical and appealing reformulation of the concept of power, but try to control them through expulsion or assassination.
The perspective adopted here, therefore, is that neither the Peruvian state
to what extent is a poststrucruralist interpretation relevant te a discussion of
nor its quasi-religious Maoist opponent could impose or carry out policies
relations existing between state and province in Peru? The concept of trans-
lation builds on Che notion of flow in which messages are both propelled and and designs in an unmediated way. They could not transmit commands or

altered as they pass through networks of actors, themselves mobile and chang- messages over space; instead, they have had to accept and allow for some
level of translation. The scope of Chis translation, what it entailed and its
ing. In Peru, teachers are generally physically mobile, and as local intellectuals
and cultural brokers they capture, channel, and localize new ideas and prac- consequences for provincial soeiery, is che subject under discussion here. 1

tices. But given that rhey are employed and controlled (at least in theory) by begin the story by looking at Che context of postcolonialiry and events taking

officials in local branches of the Ministry of Education, teachers have not had place in the early twentieth cenmry, when state encroachment started in car-

unlimited scope to transform "tokens" that Che state directs through the edu- nest. This period is not only of historical interest for certain ideas as to the

cation system. So where and in what do powers of translation he? purpose of education; its control and the proper relations between teacher
and community were worked out then and have continued to trame percep-
For state bureaucracies, not only has the school been considered a fixed
site of state presente, but constant attempts have been made to also "fix" the tions to the present day.
teacher. As rrainees in teacher training colleges, reachers are exposed to a
particular rationality, vision of modernity, and concept of pedagogy that are State Territorializing and Provincial Reactions
markedly diftérent from ideas and cosmovisions held in the rural community.
It would not count as education othenvisc. But the state is not the only, nor Producing a National Education
neeessarily the most effective, institution that has engaged in producing new in Che decades after Independence from Spanish role in Che r82os, liberalism

rationalities among the literate population. In Peru, proselytizing, catechizing as a political philosophy shaped Che mutual recognirion and intelligibiliry of

Protestant churches have done so; so roo have forceful, left-wing political messages exchanged benveen atare and province. Liberalism had been intro-

318 STATES OF IMAGINATION AN ANDEAN PROVINCF 319

 duced at the very end of the colonial period and had prompted the Spanish
ther te suggest that neither the local nor the continental expression of nation-
Crown to grant Indians the same civil rights as Creoles (Spanish setders who,
alism corresponded to che spatial scale required for the formation of Latin
dueto the raint of being boro in the colonies, were barred from holding office
American states after the 1820s- The spatialiry of che independent state, a
in che colonial administration and church). But liberalism proved unsetding,
former administrative unit of che Spanish empire, was flor particularly mean-
for ir fired Creole patriots to wage revolution to liberare America from Span-
ingful to Creole elites, who identified more with che spatially circumscribed
ish rule. Following Independence, liberal ideas guided che new nation build-
provinces that resembled city-states run by white or multiethnic ruling groups
ers and underpinned Republican constitutions and legislation. However, as
with powers of government over che surrounding indigenous population. Nor
Hale comments, Chis European political philosophy was applied "in countries
was the spatialiry of che new states meaningful to indigenous populations.
which were highly stratified socially and racially, and economically underde-
The old ethnic polities had been broken up under colonial rule and Indians
veloped, and in which the tradition of centralized state authoriry tan deep"
had suffered both spatial enclosure and deterritorialization, pushed around
(1996: 368).
by colonial administrators ro provide tribute and workers and goods for ehe
In Peru, at the start of the Republic, the liberators declared that all Indians mining economy.
or naturals would be known as Peruvians. Like all other citizens, they would
The effective scale of government inherited from colonial times was the
be recognized as possessing individual rights to land, that is, to the land
province. Provincial elites, euphoric at the end of colonial rule, were expand-
granted to Indian communities by the Spanish Crown. They were also liber-
ing and diversifying through che incorporation of waves of immigrants of
ated from the obligation of providing tribute in cash, labor, and goods on che
largely European origin who looked for advenrure and fortune in the New
grounds of dreir race. But in practice, it proved far from easy to carry out the
World. Tarma, a prosperous Andean province relatively close to Lima, liad
emancipatory legislation; tribute was in effect retained and Indian commu-
long provisioned the mining centers in the central highlands with workers,
nities were under great threat as a result of the change from corporate to
pack animals, and foodstuffs and was aetively promoting che recolonization
individual ownership. By che end of the nineteenth century, Iiberalism liad
of che tropical lowlands on its eastern margins and benefiting from che vastly
been transformed from an ideology of opposition to the old colonial order to
expanded production of cane alcohol, aguardiente (F. Wilson 1982). Duringthe
an ideology that coupled progress with social control and new "scientific"
nineteenth century, the province attracted ripples of foreign immigrants who
ideas of racism and positivism. In che process, the inicial emancipatoryvision came directly from Europe or via Argentina or Chile as well as from China.
of citizenship rights was dropped. Seen from the perspective of the nacional
These men (and a small number ofwomen) preferred to settle in che flourish-
elite, the mass of che indigenous population of che highlands was immersed
ing Andean town, located nor roo far from che still unruly Lima. Some hoped
in a backward subsistence economy, not reached by flows of moderniry, lack- to make their forrunes in mining or from properties in che recolonized low-
ing all vestiges of civic culnlre, and roo humble, ignorant, and uncivilized to lands; others offered professional services or were master craftsmen much in
be capable of exercising rights of citizenship.' The Constimtion adopted at
demand by properry-owning families resident in che town. Others v4ere mer-
che roto of the century included literacy as an essential artribute of citizen- chants and traders, muleteers and transporters, che largest concerns setting
ship, and Chis move effecrively barred che majoriry of che indigenous popu-
up import-export houses from the Tarma base. Newcomers, one could argue,
lation. National and provincial elites alike then carne to see education as a
would tend to have less-ingrained prejudices as to the place of an Indian peas-
nacional crusade, as che route through which ignorant Indians would, even-
antry in postcolonial Peru. Indeed, for a short period in the mid-nineteenth
tually, be civilized and integrated inro ehe new nation.
century, family histories retold in hushed voices reveal that a number of Eu-
In his discussion of imagined communities, Anderson has pul forward the ropean men married women from indigenous families and gained access to
argument that che nationalism espoused by che Creole pioneers of Spanish resources.
America was characterized by a "well-known doubleness ... irs alternating
In Tarma, as in other Andean provinces, property owners and professionals
grand stretch and particularistic localism" (rggr: 62). One can go a step fur-
(though rarely the Italian or Chinese merchants) were forming a provincial

320 STATES OF IMAGINATION AN ANDEAN PROVINCE 321

 needed tú be bridged. ibis hcightened the iniportance given to education and
elite eager to govern and even more coger to encoge in local poi iticking. There
generated a vicw rhat only a nacional education project under state control
was an unusual enthusiasm in Tarma tor transforming thc cabildo (council) of
colonial tintes roto a modern governing instimtion. This vas signaled by the could integrare the Andean provinces. Let us look at rhe cmerging conflicr

remarkable decision in the ntid-i86os ro keep copies of the mayor's corre- benvecn province and state tiom rhe perspectivo of1aruta.

spondence and Che minutes (actas) of provincial council debates, bind rhem In 1870, the state had declared rhar free priman education was to be made

finto Ieather volumes, and deposit rhem in rhe municipal archive. With a de- available in the capital of every distrier in the country ancl in 1873 handed over
responsihiliry for education to rhe provincial councils. 1 he Lima minisrry in
gree of aurononio that included the right ro colleet Laxes levied on movement
charge of education (along with justice, culrure, and welfarei relayed limitcd
and rhe consumption of luxurv goods as well as commandeer Indian labor.
provincial elites could actively govern and take sreps to improve infrastructure state subsidies ro rhe provincial councils ro pay for schools and teachers, bur
after Peru's banlcruptcy following rhe War of rhe Paeific, the provinces had ro
and establish schools.
fiad their own revenues to make up for the loss of stare subsidies. In Tarma,
The tradition of centralism, though, bit hard. As the correspondence of rhe
rhe provincial council recommended that teachers of rhe seventeen primary
Tarma mayors shows, rhe provincial elite insisted on imagining the state as
a superior juridical body and constantly asked for advice and guidance. Al- schools in rhe province try to collect small sums from parenrs, but rhe men-

though at rhe start, rhe embryonic central governmenr headquartered in Lima sure was unpopular, especially in the rural ateas.' Circulars sent by the Min-

possessed neither rhe capaciry nor the imagination to intervene directly in istry urged rhe opening of more schools and that rhe physical condition of
rhe boys should be improved: "Norhing eontribures so much to physical edu-
whatwas going en in rhe provinces, the acriviry of rhe provincial councils was
cation as military exercises."' This indicares rhe connection being made be-
partly responsible for forcing presidents to finance rhe expansion of rhe cen-
tral state administration. In tercos of polirical ideas, there had been much rween schooling and military reeruitment.

common ground between modernizing provincial and nacional elites. Sharing Tarma's municipal archive reveals that during rhe 18gos, 40 percenr to

class and roce identifications, a horizontal affcniry existed among rhe clusters So percenr of total municipal income (raised largely from consumpcion Laxes
on aguardiente) was devoted to education. The Tarma mayors wrote fre-
of whirelmestizo elites dispersed over space, but there was notyet much gen-
eral acceptanee thar a vertical, hierarchic relationship bound province to state. quendy to the Lima Ministry te try ro secure funding from rhe state, bur rhe

By rhe 1870s, after a messy period of rule by caudillos (military strongmen), Ministry replied that rhe province was wealthy enough to manage without

che central state was stabilizing and growing stronger (Goorenberg 1989, a state subsidy, given that Tarma's municipal income was known to have

1993; Klarén 1986, 2000). Stare revenues expanded gready dueto levies raised doubled since rhe war. The mayor continued ro complain that other, pressing

en rhe windfall profits earned during rhe guano bonanza (when nitrare-rich expenditures had to be posrponed due to rhe borden of education and threar-

bird shit was exported as fertilizer to Europe). These revenues financed an ened to close schools to redirect funds te public works.' But members of rhe

expanded state bureaucracy, public works, and wild schemes to throw railway provincial council were deeply divided in their views as to rhe proportion
of funds that ought to be devored to education compared te public works.
lines across die Andes with rhe aim of linking Pacific and Atlantie Oceans.
But Peru suffered ignominious defeat in rhe War of rhe Pacific (1879-1884) Despite rhe wrangling, rhe mayor made rhe claim rhat forty schools were
functioning in rhe province in 1892; the vas[ majoriry were to be found in
when Chilean troops rampaged through rhe country. The coastal nitrare fields
were los[ to Chile, and the state was bankrupted and Torced to cake costly rural hamlets where teachers from Tarma town attempted to teach Quechua-

loans from Britain. The shock of rhe defeat was salutary. A general view speaking children to read and write in Spanish.' This claim, however, sounds

emerged among intellecmals debating rhe causes of the catastrophe and pos- like an exaggeration.
sibilities for futura reconstruction that rhe country was plagued by racial, cul- A radical group of intellectuals and educationalists in Tarma, under rhe

tural, and geographic fragmentation. The gap between coast and highlands, intellecmal leadership of Adolfo Vienrich, were inspired by the challenge of
bringing citizenship within reach of rhe indigenous popularion 7 They devel-
between modern Peru and backward regions still steeped in the colonial past

322 STATES OF IMAGI NATION AN ANDEAN PROVINCE 323

 oped attractive teaching materials that drew on local natural history and resorted to the threat of fining the provincial councils who did not comply
myths and legends for use in rural schools, produced the first newspaper in with the request for information. Adolfo Vienrich, who liad just been elecred
Peru in Quechua as well as Spanish, and opened a night school for the edu- mayor, wrote at length explaining the difficulties of collecting accurate statis-
carion of artisans and their children. Not only was the Quechua language tics from far-flung districts whose authorities were themselves harely literate
valued, but Indian cultural traditions in the present were respected as well as and who kept no records." In Tarma, as elsewhere, difficulties of data collec-
Inca achievements in the past. The radical group (supported by artisans and tion probably meant there was substantial underreporting. The result of the
traders) had won the municipal elections on two occasions (in 1897 and 1903) 1902 census shocked the Lima Ministry, for it revealed that despite the na-
and installed Vienrich as mayor. Thus, attempts had been made at an early tional crusade, only 29 percent of children between the ages of six and four-
date in Tarma to define and implement an alternative form of education and teen years were receiving instruction (Contreras 1996: 11). The Lima Ministry
pedagogy and to translate the content of education so as to inspire a less decided to intervene.
racist, less unjust vision of postcolonial sociery. There had been deep criti- In 1904, the Directorate of Primary Education sent out a circular to inform
cism in the provinces of the state's view of education; in Vienrich's words, the provincial councils that the state would now cake over the responsibility
"There does not exist in the Republic, even in embryo, a form of national of guaranteeing the provision of primary education throughout the Republic.
teaching that is truly Peruvian in spirit and direction."a
To do this, better and more regular statistical information was imperative as
Attempts by the central state to intervene in provincial matters and extend to the numbers of schools in operation, their condition, and their location.
its authoriry were not considerad by Tarma's citizens as efficient or beneficial. Provincial authorities were told they could reduce the salaries of teachers who
How, then, in practice, did the central state manage to subdue the proudly failed to send data to the district and provincial authorities." Mayor Leon
independent provinces? Here Tarma's municipal archive provides a vivid, day- wrote unceasingly to Lima excusing the delays and reiteraring his predica-
by-day account of the erosion of provincial autonomy.
ment. In 1905 the government passed a resolution imposing fines on the may-
During the IS9os, the Education Ministry wrote in an increasingly peremp- ors of the provinces and districts who did not send data on time. Members of
tory tone demanding that provincial councils furnish statistical information Tarma's provincial council were both incensed and bemused. This treatment
on the exact numbers of schools, teachers, and pupils. At the start of the at the hands of minions of the state was thoroughly disconcerting to the men
decade, demands for information were sporadic, but in 1886 Tarma received who tan provincial affairs. The Tarma mayors no longer responded to Lima's
a visir from an official inspector sent by the Ministry of the interior to report demands with long letters, written in a familiar, man-to-man style. They
on local government in the central Andean departments. In his assignment now wrote more cautiously. After receiving a particularly sharp telegram that
he would also review what support was being given to education and condi- threatened to remove Mayor Leon from office unless the statistics were dis-
tions in the schools. The inspector was not well received in Tarma. He failed patched immediately, Leon wrote: "There must be some mistake."" The or-
to gain an audience with the mayor, and when trying to collect a list of the ders and threats from the state bureaucracy ridiculed local democracy, under-
children enrolled in provincial schools was fobbed off with the excuse that mined the dignity and respect of the provincial authorities, and gave them
the municipal authoriry had no representatives who could collect Chis infor- very little room for negotiation, let alone translation, of state demands. Pe-
mation in the rural areas.° Tarma paid the price for its high-handed treatment remptory demands created an image of the state as being all-powerful. The
of the inspector, for he published "most unfavorable" comments in the Lima provinces' age-old tactic of noncompliance no longer sufficed, for there were
press that, according to the mayor, were full of "grave inaccuracies."1" serious shocks in store.
In line with the crusading spirit now found in Lima, a massive effort was The state's efforts to make government rational and effrcient were increas-
made to collect national educational statistics for the year 1901-1902, not- ingly depicted in the Lima press as blocked and aborted due to slow, cumber-
withstanding the noncooperative provincial authorities. The Directorate of some, anachronistic, noncompliant provincial government. The image im-
Primary Education in Lima sent out a stream of official letters, and finally planted was of a backward, lethargic, rural elite that ought not to remain

324 STATES OF IMAGINATION AN ANDEAN PROVINCE 325

in charge. Thc wap was hcing cleared for rlre state m crush locally elected body language of clic bent, dcspised, humble, servile Indian finto che upright
government and remove irs capucities and responsibilities. Finally, in igo6, body of che cirizen, a citizca who in dhe I.arin American contexrwas expected

che control, planning, and linancing of primarv education was taken out oí to bc malo and provc bis patriotism by bearmg arras.
die hands of die provincial councils and shitted to a state ministry. 1bus, die Thc statc's erosion of provincial powcrs of go^'crnmenr and ubsorprion of

launching of a narional educarion project was acconipanicd by tlte bypassing, education liad come ar a time when co unrercu rrents vete surtacing 111 che

discrediting, and disempo vcring of provincial government, and a new sub- Andean region rhar rejected the state s crudo civilizing mission. In "arma, this

servient relarion sh ip berween province and state was boro. incipicnr indigenista lino of thought liad been pionecred in che pedagogic and
Perú s nacional education project in che carly tsvenrieth cenmry depicted che political work os Vienrich and his group from the Radical Partv. Vienrich had

Indian as ignoran[, defective, uncivilized. To integrate che Indian, Contreras been en exeeptional example of a local inrellectual who in the early igoos liad
notes: 'Armies of teachers and personael wcre sent off to the countryside, ac- adopted a soeialist perspective to analyze postcolonial sociery and to guide
companied by packs of books, pencils and maps, while poor bus diligentpeas- his political acrions as mayor. llis critique of che state ( and as representative,
." But al-
ant communities built classrooms and school yards to receive them. The task the prefect) cosí him his lile and also bis repumrion as a soeialist
involved not only teaching reading, writing and simple arithmetic, but aboye though much of his political work was expediently " forgotten," sume of his
all, to transmit the `nacional language,' which was Spanish, ... and nacional ideas lived on among later generations of local intellectuals . Opposition to
history and geography, and to inculcare nutricional habits, as well as notions of the integrationist educational project of the state carne to che fore briefly in
hygiene and of `urban living' that would improve the physical condition of the che Iggos and with greater force in che 1970s.
indigenous yace" (r996: g). Thc missionary view of education was fostered by
the local as well as rhe nacional press. In agio and igrr, almost every issue of Structural Reform and the New Philusophy of Education
Tarma's newspaper, E! Imparcial, carried an article on the imperative peed to From che first phase of state expansion in the igoos up to che igbos, che form
civilize che Indian through education: "For the future prosperity of the nation, of state intervention in the Andean provinces changed little." But the Revolu-

education is essential for the indigenous race still living in semi-savagery" tionary Government of the Armed Forces under Juan Velasco Alvarado, aided
(io January igio). "Che sigas were hopeful: "We have seen that those from their by professionals and intellectuals from che "left," was inrent on imposing a
race who know how to read and write have formed families according to civi- series of structural re£orms in che 1970s to modernize che country. Of particu-
lized custom" (7 November igio). Numbers of schools in the province rose lar relevante for the Andean provinces was a sweeping agrarian reform rhar
slowly, but opposition was voiced by the properry owners. Local hacendados aimed to dispossess che hacendado class and transform their large properties
(fámilies resident on the hacienda) had no interesr in extending education to loto cooperatives, along with a series of measures to reorganize education.
rheirgente and chased the schoolteachers away from their properties. But the military government got off to a bad stars. Two events in the late ig6os
Schooling was closely bound up with an authoritarian form of pedagogy, were later singled out as symptomatic of che underlying violent clash between
military ethos, and militarized regulation of the body through drilling, an authoritarian state and citizens with their own interests and agendas. One
marching, and sirting still. There also developed a preoccupation with hy- was che overturning of a locally directed agrarian reform in che department
giene and eleanliness, with checking the bodies of Indian children. One can of Andahuaylas and its enforced return to state control (Sanchez rg8i). The
question rhe extent to which these ideas reflected the power of the state to other was che violen[ suppression of protests by parents in che department of
transmit and impose a cornmon regulatory framework, for these ideas also Ayacucho who rejected che government's plan to levy charges on schoolchil-
emanated from the dominant class in Andean sociery. But one also needs to dren having to resit exams (Degregori 1990). In both cases, the military was

question che extent to which the practices revealed purely the mierophysies of sent to restore order, deaths oceurred, and the confrontations generated sup-
disciplinary power, for in a postcolonial situation (one never addressed by port for revolutionary parties, hardening che view that the brutal imposition
Foucault), they could also be seen as carrying liberatory potentials. It should of state power should be opposed with violente if need be. Leaders of both
be remembered that an important objective was to transform the body and movenients were later to join Sendero.

326 STATES OF IMAGINATION AN ANDEAN PROVINCE 327

 With respeta to education, the military government inherited a heavy finan-
tional territory a set of technical reforms had been mapped out and was being
cial commitment from the previous government (under Belaunde) that had
strenuously implemented with the aim of creating a homogeneous economy
promised to introduce free education at the secondary level and to mercase
and sociery. But the military government did nor succeed in selling a new
teachers' salaries. To chis, the military government added a political commit-
popular image of the state nor in replacing the old racist social order and
ment. Education was to be a cornerstone of the revolution. Velasco, in well-
discourse with visions, models, and possibilities that corresponded to popu-
publicized speeches, criticized the state educarion system and, inspired by the
lar needs or demands. In Tarma, the new state functionaries were often given
teachings of Brazilian Paulo Freire, stated that his government "aspires to
disparaging names, such as yanagringos, pseudo- or cholo gringos, who as-
creare an educational system that satisfies the necessities of the whole nation,
sumed authority and were protected by the stare but who were judged incom-
that reaches the great mass of peasants, up to now exploited and deliberately
petent impostors. Jokes about the new state agents abounded, told on jour-
maintained in ignorance, that creares a new consciousness among all Peruvi-
neys by an increasingly mobile population. The state was being experienced
ans of out basic problems and that attempts to forge a new kind of man within
as dominating and authoritarian and was being reimagined through laughter.
a new moraliry thar emphasizes solidarity, labor, authentic liberry, social jus- But it was black humor.
tice, and the responsibilities and rights of every Peruvian man and woman." `°
The contradictory processes of structural reform and the complexiry of
An Education Reform Commission (en which soma 120 specialists labored
the outcomes had non been foreseen by members of state planning agencies,
for a year) delivered a long report with many recommendations for compre-
many of whom were Peruvian and foreign academics of Marxist persuasion
hensive reform. Despite the widespread criticism Chis report provoked, it
laudably working to end feudalism and dependency and launch Andean soci-
formed the basis of the new Law on Education (Decree Law 19326) of 1972.
ery en a modernizing track. But the reforms could flor be transmitted down
The new Peruvian education was to build on three ideological pillars-hu-
through the state's administrativa hierarchy in a direct or systematic way.
manism, nationalism, and democracy-tied together by the central idea of
Instead, certain groups, such as the ubiquitous ingenieros, were leen in [he
concientízación (awareness raising) that would lead to liberation and participa-
provinces to be the main beneficiarias, for they were in a position to Cake hold
tion in che historie process of removing the old structures of dependence and
of and translate particular aspects of stare policy in the light of their own
domination (Drysdale and Myers 1975). But many schoolteachers remained interesas. Now, instead of blaming the state for its neglect, there was plentiful
unimpressed. They were openly critical of the government for reneging on
evidence at hand to accuse the state of bringing chaos and misfortune. A new
promises made by the Belaunde government for salary increases.
translocal discourse of state corruption was gaining ground while che local-
Through the reforms, secondary and higher education was expanded in
ization of state institutions sharpened perceptions of inherent contradictions
the highlands and a new nacional policy for bilingual education inaugurated.
between the state planners and the people's desires, demands, and place-
Por the first time the teaching profession was put within reach of youngsters
making practices. Agrarian reform had non given peasants control of their
from poor rural, indigenous, and cholo backgrounds, for teaching was the
land, nor could rapid training programs compensare for their lack of prepa-
only profession where trainees could work (and earn a salary) while smdying
ration. Instead, the reform had substimted state-employed administrators, few
for their qualifications in vacation times. Now young men and women from
of whom knew anything about agriculture in the Andes, for the old patrones
the subordinare groups in the Andean race-class hierarchy could hopo for
and had failed to make sufRcient resources available to manage the transfor-
social mobility through education. Mediating and keeping a watchful oye
mation. The state's cooperativa adventure started to crumble, and poverty and
en the turbulence provoked by the changes was the state-sponsored, nation- disillusion carne in the wake.
wide organization for "popular participarion" witose well-chosen acronym,
BINAMOS, meant "without masters.""
Violent Responses to State Presente
The agrarian reform led to che implanting of new state offices in the prov-
State-led structural reform inevitably failed to mear popular aspirations and
inces and brought greatly increased numbers of técnicos and ingenieros. This
expeetations. When the central state discovered it could non control outcomes
represented an era of more intensa stare presente and intervention. On na-
and feared ungovernabiliry, it put a stop to rhe top-down process of change.'"

328 STATES OF IMAGINATION

AN ANDEAN PROVINCF, 329
 among younger teachers and secondary school pupils, to eonsider a politics
Vet, in thc posrrefbrm phase. the expansion of schooling and higher educa-
of violente as legiti orate.
rion and rhe new language of concientizació n and empowernienr continuad
The reform and rapid expansion of educatiou had drawan on poorly pie-
ro poli6cize perceptions of racism, social exclusivo, and patterns of privilege.
pared teachers widhour much access to teaching mnaterials order 111.111 cheap
-1 he state' s expansion ofeducation was linked w ¡ th thc developmcnt of a new
editions of Marxist manuals (available in marke[places rhroughour the higli-
language of dissena This was a strident, belligerent forro of political speech,
lands) and crudo political pamphlets (s(3ma of which carne tiom the state's
behavior, and hody language [hat crystallized ¡ n centcrs of higher education.
own organiza tion, siNAMOs). Students who poured inro rhe provincial uni-
As Portocarrero and Oliarre 0989) note, rhe new political discourse of thc
versiries, reacher taining colleges, and instimres of higher learning were pre-
schoolteachers resred on the "critical idea." 1 his Portocarrero defines as "the
sented with an education glued to Marxist principies (Degregori 1991). yet, al-
vehement opposition to a haced order ... a discourse which is presented as a
though it was critical of the state in terms of its underlying philosophy, some
jolt ro the conscience , as a revelation about something the dominant classes
of the premises of the state's civilizing mission were retained under the guise
had previously hidden" ( 1998: 124). The critical discourse provided a simple
of producing the "new socialist man." The "good" teacher tried to interest
set of schema and methods of analysis that could transpose painful lived re-
pupils in politics and current events, but the preoccupation with hygiene and
alities lato standardized conceptual and categorical forros.
eleanliness, and with disciplining and regulating the body, continued.
The state ' s efforts to incorporare schoolteachers inro its new national edu-
From out of an alphabet soup of Maoist factions, rhe political and ideo-
eation initiative and use them to convey lis messages were far from success-
logical lead was caprured by force in the 1970s by a group of intellectuals
ful. The schoolteachers were no transmission belt, and by the 1970s increas-
and teachers from the University of Huamanga, Ayacucho. Sendero Luminoso
ing numbers became classroom Maoists (Angell 1982 ). The Ministry of
proclaimed a hard-line, highly simplified, and thus more accessible version
Education fought to cake greater control over what was taught in schools
of Maoism tirar was presented as "scientific truth" (Degregori 1991: 239). The
through a reform of the curriculum , but the educational content could be
dogmatic ideology and visions of destruction and bloody revolurion taught
communicated and translated in a variety of ways. The government was
the imagination of some of the young who looked for a way to confront the
caught out in its own rhetoric . People had been pressed to become aovare and
racism, injustice, and eeonomie and social hardships they had known from
to criticize the old order as a prelude ro giving their support to structural
childhood. The school became Sendero's most importanc gathering point and
reform. But they were no[ supposed to critique [he state or the inroads rhe
target. Paradoxically, Sendero declared the start of its armed insurrection in
military government had made inro local political processes and aspirations.
1g8o, the year Peru returned to constitucional rule afrer a decade of military
At an organizational level, rhe critical idea was brought into play in the
government. The transition had marked an unprecedented opening up of
mounting political opposition to military government . The teachers ' union,
ehannels of legal political action, and most Marxist parties opted to particí-
Sindicato Unico de Trabajadores de la Educación Peruana ( SUTEP; Union of
pate in the national election (Manrique r995b). Not so Sendero. The Parry
Peruvian Education Workers ) founded in 1972 had a membership of over one
remained clandestine and pledged itself to bring down the bourgeois state by
hundred thousand by the end of rhe decade and was rhe largest union in the
wiping out its agents in the provinces and sweeping away revisionists by liq-
country. But declared ¡Ilegal by the military government , it was not permitted
uidating al¡ instantes of popular aetion not under Sendero control. Sendero
to act as a channel through which grievances and demands could be brought
strategy involved militarizing sociery in preparation for a long war. But rhe
before or negotiated with rhe state . The union denounced the government as
kind of future state envisaged by rhe Parry was only vaguely indicated. Under
a military dictatorship and organized srrikes and popular actions to influ-
rhe New Power or New Democracy, people were told they would live in great
ence public opinion finto taking an antigovernment stand ( Hinojosa 1gg8).
harmony, free from all conrradiction, in which divisions of class, cace, gen-
In response to mounting national strike action in die late 1970s , the govern-
der, and generation would be erased, rhe project kepí on track through obe-
ment turned to repression. Union leaders were murdered , jailed (some four-
dience to an absolute philosopher-leader.-"
teen, in the case of Tarma province ), hounded, and denied employment in
In Tarma, most teachers liad been exponed to rhe critica] discourse of
state schools . The confrontation gencrated a greater susceptibiliry , especially

330 STATES OF IMAGINATION AN ANDEAN PROVINCE 33T

 Marxist-Maoist politics in centers of higher education during che 1970s and
campesinas) to fight both che terrucos (terrorists) and che murderous spirirs, che
1980s. Some had attended Sendero-inspired escuelas populares (clandestine
pishtacos, who now roamed at large on che community borders.
night schools) offering a political education that ranged from reading Mao-
The military mode of state power had to change to mark che passage from
ist texts ro dynamiting police stations. Sendero militants worked secretively
war to peace. The transition, taking place during che years 1992-1998, de-
and left characteristically cryptic codes and messages in che landscape (as
manded that che military stand up as a moral force, exact retribution by pun-
in the sudden appearance of crude slogans painted in red en walls, or the
ishing those accused of terrorism, but also reprieve [hose who offered infor-
communist flag, or unexpecred attacks) to suggest they were everywhere and
mation (through che Law of Repentance) and safeguard an orderly return to
nowhere. In che schools they tried to turn state nationalism and militarism
a state of government. The military defeat of Sendero was not accompanied
inside out by substituting the communist flag, hymns to Sendero's leader
by an attempt to minare a process of reconeiliation or national healing. In-
(Abimael Guzman under bis nom de guerre, President Gonzalo), and a new
stead, che Ministry of che Presidency that mol‹ charge of donor funds sought
calendar of heroic events. Most schoolteachers now found themselves caught
to reestablish state presence by instigating a vas[ program of public works in
in che middle, having ro translate berween the opposing demands of state and
which building schools ¡ay at che center. Now the smallest, most remoce com-
Sendero. This involved a dangerous mis of symbolic practices. They might
munities are being provided with school buildings of standard design and
hoist the red flag in public while singing che national anthem in secret, or
color whose inauguration may bring che presiden[ en a lightning visir. The
vice versa. AH che while, they had to persuade local officials of the Ministry of
school building has become che reminder in che landscape of a return to state
Education that schools were functioning as normal, for no one knew for sure
government But most schools still stand as an ernpry shell, without books,
which side these officials were on. After Sendero had dominated extensive
furniture, or equipment, and staffed by an increasingly despairing body of
areas in the Andes and numbers of deaths, physical destruction, and terror
teachers to whom che Fujimori government consistently refused to pay a liv-
were mounting, che state gave full power to che military ro crush che subver-
ing wage. One suspects that the teaching profession is being punished for its
sien and bring che "regions of emergency" back to the state government.
oppositional politics and sympathy for revolutionary action that has lasted
Sendero had gained effective control of the more sparsely populated re-
nearly thirty years (F. Wilson 2000).
gions at a distante from 'forma town and transformed them roto support
bases for the armed groups. The parry had also won some popular sympathy
by organizing che redistribution of land and livestock, symbolically showing Celebrating Fiestas Patrias
it was capable of putting into effect what the state's agrarian reform had failed
In Pero, the national myth of origin celebrated in Fiestas Patrias on July 28
to accomplish. The entry of Sendero, especially loto che northern perimeter
carnes a fraccured meaning. On che onc hand, che eviction of Spain and
of che province, provoked population displacenlent, che conflict Lending to
achievement of independence are seen as resulting from a popular struggle
enhance parterns of movement already underway. Ironically, Muruhuay, an
for liberation . On che other, che state is celebrated as extending its power and
importan[ place of religious pilgrimage in the central Andes, become a place
sovereignty through two lines of transmission : che military and che school.
of test and recreation for Sendero militants. When fighting the subversion,
In Fiestas Patrias , as Gose ( 1994 ) argues, an ideological image of che nation
the military carne to employ such brutal tactics that they became more feared
is built up through a militarized presentation of che local school system,
than Sendero, and Indian-looking people were singled out for exceptional
where la patria (che Mother- Fatherland) is treated as "a school writ large."
mistreatment. But people caught up in che conftict carne ro make a distinction
From Chis metaphorical representation , education is seen to provide "che fun-
benveen che hated military units sent up from che coasr and che units of se-
damental medium of social relations within che Fatherland " and in so doing
rranos (men from the highlands) who more easily gained people's trust. Rural
demarcares the reach of state territoriality . This ideology, though, is contested
communities were more likely ro throw in their lot with che state when they
and undermined during che celebration (70-71).
could ask paisanos for arms and help in setting up peasant milicias (rondas
Fiestas Patrias has been celebrated since che early nineteenth century with

332 STATES OF (MAGINATION

AN ANDEAN PROVINCE 333
 the terrorists. For the narrator. Chis post gave him tire chance to return to thc
parados of sschoolchildren accompanied hy military hands playing martini
community bordering his mother's borne. Belonging tu a family from thc
nmsic; thcy filo past local dignitaries, sing the national anthem, and salero
regios, he claimed to he both paisano (fellow countryman) and edueated our-
tire national flag. In later years, Independence Day has been nmrked by a pa-
sider. Ar one leve], bis story weaves together ideas abour statc and antisrate_
rado of Peru's armed forces in Lima, a celebration of military might subse-
citizen and subversive, teacher and community. At another, the story is used
quently brought m die notice of increasing numbers of Peruvians through
to justify, borh m hirnself and the listener. how he assesses and maneuvers in
television. ]tesas hardly surprising thar Sendero prohibited Chis celebration in
postconflict society, where thc danger of mistaken identity abounds. Extraer,
the Andean regions they dominated. Nor esas ir surprising, given irs conno-
from tire narrative are followed by, my commentary.
tations of liberation, that in postconflict times, towns and communities have
wanted to register tire end of civil war by returning to their celebratory cal- There are some experiences that are well worth talking about, and re-
endar, including the parados of schoolchildren at Fiestas Patrias. cording, that can serve to promote the identity of the peasant, the iden-
In this final section, 1 want to suggest that the celebration of Fiestas Patrias tity of indigenous people about their reality, about their pueblo. 1 should
in recent years can provide a window through which to observe how the un- like to talk about last year, about the month of July, when, as you know,
easy relations between state and province are now being played out. The pa- Peru celebrates Fiestas Patrias on July 28 and when ese in Tarma celebrare
rades of schoolchildren traditionally held en July 28 mark a contested site of the feast of out patron saint, Santa Ana, en July 26.
public spectacle, the time-space slot where both state and locality seek rep- With [hese opening remarks, the presentation of the national holiday is
resentation and where, in postwar times, new rules of the game are imposed, downplayed by being linked to an event in the provincial ceremonial calendar.
played out, and opposed. I look at events relating to Fiestas Patrias from two This suggests that the national will be viewed and framed by the regional,
perspectives: first, by commenting en a story told by a rural schoolteacher rather than considered a manifestation of state power.
about his role in reestablishing the parades in a community of returnees;
At the time of Sendero's occupation, many of the customs of the citi-
second, by presenting my observations of the effects produced by the govern-
zen were prohibited. Itwas prohibited thar children take parí in parades.
ment's decision in 1999 to reorganize the celebration to give greater promi-
It was prohibited that in indigenous peasant society, they could sing the
nence to the armed forces.
national anthem of Peru, or have a military escort raise the Peruvian flag.
As educators and teachers, though ese are not in agreement with the
Fiestas Patrias and Popular Nationalism
political system of the government in office, nevertheless we believe in a
The narrative 1 discuss describes with particular clarity the teller's careful
state, in a nation, in an identity, and that the teacher has to also under-
formulation of an antistate stance. It belongs to a gente of mobile popular
stand that there are ancestral customs in the pueblos, which cannot be
culture, of convivial storytelling in bars and en joumeys.'° It articulates a dis-
changed from one doy to the next. On the contrary, one must respect
course that few teachers date to express openly in the tense postconflict situ-
their customs. One of [hese is that the people are accustomed to orga-
ation where agents of the state's intelligence service are suspected of having
nize parades to celebrare Fiestas Patrias. The mothers and fathers feel
taken over from the military in keeping a watchful eye en the activities of
that they are somebody in their community when they seo their children,
schoolteachers. The ideas expressed can be seen as belonging to an opposi-
the pupils, participate in a civie-patriotic ceremony, euen though there has
tional, subterranean discourse that flows among teachers. Through stories
come no legal order from the state that obliges them to hold a parade [emphasis
such as this one, patriotic events can be vested with oppositional meanings
and subject to endless improvisation. addedj.
In 1996, the local office of the Ministry of Education offered a contract to The narrator suggests that the peasantry had embraced and made their own
the narrator to reopen the primary school in a small hamlet in the northern the customs and rituals concerning parriotism and nationalism, practices
highland fringes of the province. The population had fied en account of Sen- brought to the countryside by earlier generations of sehoolteachers. These
dero and later military occupation and liad been branded as accomplices of

AN ANDEAN PROVINGE 335

334 STATES OF IMAGI NATION
 rimals have become integral to the claims of the indigenous peasantry to
identiry and cirizenship. Sendero is blamed for failing to understand the con- The narrator positions hiniself as an active sponsor of popular nationalism

text and meaning of [hese "customs" or that this popular form of nationalism in the community, where he expresses support for the wishes of the mothers
indeed expressed a separation of die nation from the state. The narrator be- in particular. Not only does the parade involve the active participation of chil-
lieves that teachers guiding the communities representa superior, transcen- dren, but the social status of parents is reflected in how clean and well-
dental nationalism. It is the teacher rather than the government who is the turned-out the children appear. The choice of which children are allowed to

true patriot and works toward the realization of a better state, one that listens march and carry fiags, pennants, batons, or swords is a social issue, for not
to and acts on the needs of the people. all in the community have the resources allowing them to present their chil-
dren in a suitable way. The choice of who marches is partly up to the teacher,
For the people who want to carry out this kind of event, they see it as
and it is his duty to drill and instruct the children selected. In the community
a kind of distraction, for this is a pueblo that suffers from the lack of
celebrations, soccer tournaments have now taken the place of an older tradi-
many indispensable services. For example, it does not have public light- tion, corridas de taro (bull runs). The substitution is indicative of modernizing
ing, or electricity; it does not have a road that is passable by vehicles; it
tendencies and changing social relations. But these activities of the men he
does not have drinking water; it does not have a sewage system; it does
outside the sphere of the teacher, as they have nothing to do with the school.
not have a medical post. In reality, the peasantry of this community has
Instead, the narrator concentrates his energies on planning a parade for the
very little of anything.
children. The final comment encapsulates a familiar trick of magnification: a
The transcendental nationalism of the teacher and strength of popular na- tiny event, in this case a parade of schoolchildren, is linked with a vast pan-
tionalism is ¡uxtaposed with the failure ofgovernment tu fulfill its obligations orama, an overmrning of the social order whereby poor peasant children can
of bringing modernity to the countryside. But in this part of the narrative, the hope to enter the professions and serve their community.
teller is clearly interweaving an older political discourse that is now out of In the end, the parade was not realized, for the plans were interrupted by
step with the change in government policy. The Fujimori government won the appearance of unknown, threatening strangers who dressed as peasants
popular support with its emphasis on obras publicas (public works), although bur who did not have the faces of peasants. The narrator records that they
"government by public works" is providing only shoddy, empty symbols of had asked: "How is it possible that you are going to hold a parade for Fiestas
modernity. Patrias when in Peru there are enormous economic and social problems? How
is it possible that the flag flying will not be the flag of the people?" The Sen-
In the community, the only entertainment for the men is sports,
dero militants threatened to kill the teacher if he did not "take care."
sporting events, but the mothers do not participate for it is mainly soc-
cer.... But the mothers really like parades, they like to see their children 1 was not going to "cake care," as 1 believe that man is borra one doy and
cake part, parade, dress up and take part in the customary dances of the then has tu die. lf we are going to live, then it must be with dignity. It
region. It is the mothers who enjoy the parades most. 1 understand this seems to me that if we cannot live with dignity, then lile is worthless.
reality. Therefore atan assembly we held, we agreed we would celebrate And whether one dies today or tomorrow is immaterial. The srvles and
the 7uly holidays with a parade, to celebrate not only the anniversary of forms are of no interese to me. Thus, when one is able to worlc honestly
independent Peru but since the community belongs to the province of on behalf of the people, trying to stimulate the development of the
Tarma, also Tarma's anniversary. In this way the people would have a pueblo, one's death would be received as a great message for the chil-
little distraction, entertainment, a little happiness, and feel hopo that dren, who in futura would see that one has to live with dignity and that
one day their sons would become professionals and do something to nobody should push them around.
serve their communities.
The beginning of a new narrativa theme is indicated by a shift in tono and
style. Its purpose is to elaborate on the image of the heroic teacher, prepared

336 STATES OF IMAGINATION

AN ANDEAN PROVINCE 337

 muniry [han could earlier generations of mestizos, ever conscious of their
ro sacrifica hintself for che cake of che people. in chis interlude, we find an-
social and cultural superioricy. Although ideas and practicas associated wirh
other example of che trick of magnihcatiou: cite teacher's marryrdom in che
school may appear relarively unchanged, rhey acquire new meanings and sig-
seraiee ot rhe people scen as a great loessage for rhe chíldren- Thc lul plication nificance. In chis case, ve are no longer witnessing enforeed, srare-directeel
uf chis passage is ciar teachers in tira postconflicr phase peed ro redefinc their integration. Rather, educarion is recognized as che roure out of desprecio (dis-
political identity chrough a renewed co mmitnlent co che people and chas che dain). In Chis, che ideas of che narrator come close to [hose expressed nearly
Sendero militants srill lit large hace no contact wirh che popular base. a century earlier bv Adolfo Vienrieh, for whom che school cas "demoeracp in
Sadly, teachers in die peasanc communities are caught in the middle, action."fhrough che nar rarom's eyes, onc lees d i sloca tion; che parado, liags,
benveen on che one side a state thar pays miserable wages and issues crappings of nationalism have become abstracred from che stare. Fiestas Pa-
laws thar teachers must follow, and on rhe other side, groups rhat still trias has been translated into local, oppositional nationalism. Ihis reversal,
survive and wish ro impose new laws in che style and custonr of the however, must also be sean as an integral parí of stace formation "from be-
reality and ideology they stand for. The teacher is in che center. For che low," in that iniages of state and citizen are being opened up for scrutiny and
state, teachers are members of che other side, and for che others, lec us reformulation.
cal] them Sendero, reachers are state informers. So when che Peruvian
milirary is on its way, che teacher is threatened, and when che other Fiestas Patrias and Locality
group comes, he is also threatened. The Peruvian state has nor been unaware of the dislocation, nor of che poten-
cial offered by Fiestas Patrias to rally opposition in che provinces. In 1999, che
The narraror's presentation of che heroic teacher is juxtaposed by a descrip-
forma[ for its celebration was changed by government decree in a move thar
tion of che fraught position of all teachers caught benveen Sendero aud che
aimed to give greater prominence to che armed forces and police. In Tarma
milirary. While living in che communiry, che narrator had been indefatigable.
town, as in all provincial and deparemental capitals, parades of schoolehil-
He had kepí in contact wirh oeher teachers and sought out [hose he thought
dren had to star[ earlier [han July 28, in Tarma's case, ten days earlier, on
were bebind che threar to his lile. On two occasions, he had visited arrepenti-
July 18. The reason for shifting che date was to allow che state co orchestrate
dos ([hose accused of terrorism but released through che Law of Repentance
a nationwide marching competition in which milirary judges would adjudi-
afcer providing information co che milirary). He had "hard dialogues" wirh
care which schools produced che bes[-turned-out, bes[-marching children.
them, noting chal although rhey were of greater physical stature, he pos-
The winning schools at che provincial level would then compete in parades
sessed greater "character." But che day before che parade, more strangers ar-
held in che department capitals, and [hese winners would have che honor of
rived and caused alarm in che communiry. The narrator resolved to lead his
going to Lima to cake parí in che march on che day of Fiestas Patrias itself. in
colleague to safety. Recrear was nor a heroic aet, but was explained by che
Tarma, responses co che changes decreed in Lima were critica]. Some inter-
narrator's concern for che young teacher, and che stork, of che nighttime walk
preted che staggering of che parades and increased military presence as a sign
and their reception by teachers in a district capital eonscitutes a healthy anti-
of nervousness en che parí of che government and fear of retaliatory action
dote to earlier references to sacrifice.
by remnants of Sendero. Others were annoyed at che government's move ro
Although che maro message of che narrative lies in che claim being mide
take over what rhey considered a loeally organized event in che hands of local
of a separation between the nationalism sponsored by che state and che na-
schoolteachers. But on che day, many people carne along to cheer.
tionalism espoused by che people, there is much room for improvisation and
On July 18, educacional inscimtions from Tarma province-from primary
border play. Depending on audience and circumstance, che narrative can be
and secondary schools to teacher and nursing training colleges and branches
more or less explicit as to che narraror's own identification wirh che commu-
of university deparements-participated in a parade that lasted over five
niry as well as wirh che oppositional politics of Sendero. Teachers like che
hours. Musie was provided (as usual) by bands of che police and che boys'
cholo narrator can develop a quite different leve) of empathy wirh che com-

AN ANDEAN PROVINGE 339

338 STATES OF IMAGINATION

 secondary school. High stands had been erected in the central square where
the bonds of community by celebrating the work of the local schools and
spectators could sit, and a podium was built, covered by an awning, for mcm-
their teachers. Through a loudspeaker, che commentator repeated che com-
bers of the municipal authority, visiting dignitaries, police ehiefs, and mili-
munity's aspirations for the future and pledged that the youngsters of the
tary guests. In front of the podium were three tables where gold-braided mili-
district would make a patriotie contribution to Peru. The hopes of the com-
tary judges sat, and keeping the parade route clear were armed soldiers.
munity, said the commentator, rested on these children, who would one day
Holding the event together was a commentator, who shouted into the public
enter the professions and return to serve their community, thus giving voice
address system introducing the schools and reminding the crowd of the civic-
to the same trick of magnification employed in the narrative discussed aboye.
patriotic nature of the event and of Tarma's support for President Fujimori.
Schoolchildren came up to the podiuni to recite poems and speeches about
The parade was broadcast live by the local television station.
the peed te fight poverty and the exclusion suffered by most of Peru's citizens.
The parade over, at supposedly its crowning moment, the commentator
But no mention was made of Fujimori or of the government in Lima.
read out the names of the winning schools selected by the military judges.
Fiestas Patrias has become a contested political site. Taking place in 1999
Immediately the festive mood evaporated. The crowd began to boo and
was an attempt to reinsert a stronger military presence and substituta a na-
whistle, people muttered among themselves that the wrong decision had been
tional agenda for the popular celebration of community. Although the state
reached. This reaction brought consternation on the podium, reevaluation by
had managed to affect the staging of che event in departmental and provincial
the military judges, and a welcome silencing of the commentator. Alter con-
eapitals with mixed success, it had been unable to reach down to districts or
siderable delay, when many spectators had drifted away, a new decision was
communities. The greater involvement of the armed forces in Fiestas Patrias
announced. Declared as winners were the lending boys and girls secondary
can be seen as an instante of the way the Peruvian government has tried to
schools in the town, a decision that accorded with the opinion of the dimin-
reintroduce a military presence, but it is uncertain to what extent this will set
ished group of spectators. The event was significant. Local ideas as to the
a new pattern for years to come. The celebration of Fiestas Patrias in che year
hierarchic ranking of provincial schools could not be overturned, even by
2000 was overshadowed by reactions to President Alberto Fujimori's assump-
high-ranking military judges. On this occasion, ihe desire of the crowd won
tion on that day of bis much-disputed third term of office.
out. The state's effort to stage-manage the public spectacle of Fiestas Patrias
had mer with limited success. The allure of a national competition was far
less importan[ [han the local parade. But because chis had been held ten days Conclusion
before national Independence Day, there was not much happening in Tarma
This essay started by raising questions about che techniques and mechanisms
town on the day itself.
through which states make their presente felt in the provinces. A conceptual
Rural districts and communities in che province were much less affected by
framework was suggested through which relations at the meeting point be-
the change in dates or practice or by the greater military presence decreed by
tween state and province could be analyzed in terms of the flows and spatiali-
the government. District schools had sent few children to march in Tarma en
ties of state power and the possibilities for mediators to translate rather than
July 18 and were far more occupied with organizing their festivities forJuly z8.
transmit orders and directivas of the state. By way of conclusion, we should
In the district capital one hour away from Tarma town where 1 spent the day,
look again at the kind of state rhat has been described and che kind of politi-
weeks of preparation had gone into che parade that gathered children and
cal reactions it has induced. An inicial focos on flow and translation has been
teachers from all the district's schools. Children were well-turned-out, wear-
a useful way of conceiving of state-province relations; this must now be set
ing clean school uniforms or fancy dress. Ar the podium were members of
in context.
ihe municipal authoriry, all schoolteachers; representatives of the district liv-
In che periods under review, che state has appeared in che Andean provinces
ing in Lima, who were presented with a gift of live sheep; and a single uni-
in various guises. We haya seen how the increasingly impersonal and authori-
formed policeman. The festivities in the district were enjoyed as reinforcing
tarian bureaucratic apparatus of the central state rook away ihe capaciry for

340 STATES OF IMAGINATION

AN ANDEAN PROVINCE 341
 If we lonk agaiu ar rhis history, andar the suture of rhe political alrernatives
governnacnt tiom tlte provinces in the carie nventieth centuiy, and in the pro-
that emerged in tlie Andean provinces, rhen the period whcn ntost wns it
eess squashed the frtgile local democracy at rhe provincial level as well as
state and most was lost, 1 believe, can he dated to the turn olí rhe eenmrv. Thc
ideas generated locally about educatiou. Pollowing a long period of srare
undcnnining of local governmenr cut shor, an expcrilacar in po_rcoloni al
neglect, a reform ist militare governnacnt in rhe 19705 srrovc ro apply much-
democracy and instinnion building al rhe provincial level..absorption oh rhe
needed structural reform. Eur aithough wcll-iiirentioned, it did so by fiar and
provinces be an ineffecrive srate led ro the worsr of all possible wodds. This
toree and Chis served lo hiel resistance that turned to Marxism and Maoisni
loes not mean that 1 harbor a sentimental artachinent to lurma s provincial
lor irs oppositionul polirical models and straregics. In response m subvcrsion
conocil or the work of rhe radical indigenistas. The viewpoinr reelecta mp
in che Andes, clic serte firsr unleashcd violenr nulirary rxpressioo and then
awareness of che ver ✓ e and dynamism, the contesrs and condicis, tire acople
tried ro alter rhe image of che military to fit better with rimes of peace. During
evidence of an undersranding of localiry, which is m be fbund ni the letters
che 19905, an aorocrarie presidency centralized state power and took direct
of the Tarma mayors as matters of tire day were reviewed and commented on
charge of processes of recuperation within a framework of neoliberal eco-
at length during che r86os to Lgoos. With the expansion of che scope and
nomic order. In sum, one can see that although there have been swings be-
power of the central state, this local debate was stifled, and therewas no other
twecn direct intervention and neglect by the state, few openings have ap-
recognized political channel where ir could be redirected.
peared for alliance building or negotiation with the provinces.
In response ro a generally inrransigent and peremptory state, reactions in
the Andean provinces have taken an increasingly eonfrontational and violenr Notes
forro. The silencing of the earty expression of oppositional politics pushed
chis essay owes a grear deal tu colleagues ni rhe research program Livelihood: Identity and
resentment and hopes for polirical change underground, che subterranean
Organizarion in Siruarions of Instabiliry. In particular, 1 want m thank Fluir Steppurat for
political discourse surfacing now and again. With turmoil and instability
his inspiring comnients on a very preliminary draft. Unless orherwise noted, all translations
brought about by top-down structural reform with all its unintended con- are orine.
sequences, a new violent phase of confrontational politics made headway. i. Ser, for example, discussions of the uneasy relationship benveen reachcr and communirv in

The spread of a belligerent Marxist-Maoist discourse in centers of education Primov ( x980 ); Ansion (1989): Montero (1990); Contreras (x996).
a. See Mallon (1995) for a discussion of the emergence of chis discourse oí eitizenship in che
paved the way for che attempr by a small, highly disciplined group of mili-
central highlands, especially in tire conrext of rhe War of thc eacific and lis aftcrmarh.
tante to bring down the state- Once again, after their military defeat, a period
3. latina Municipal Archive (TMA) Actas, 1886, Ocrober 29. The number of sehools (seventecn)
of silencing has ensued. Political opposition, however, remains latent, just
exceeded rhe number of districts (ten) at rhe time.
under the surface, finding expression in the contests over meanings, as ex- 4. TMA Letters, ¡886, Mayor Piedra te inspector of Schools, Tarma.
emplified in the question of whose liberation is being celebrated in Fiestas 5. TMA Actas , x891, December a¡.

Patrias. 6. TMA Letters, 1892, April 13, Mayor Carranza to Director General de Gobierno, Lin a.

The schoolteachers caught in che middle have led a double Efe, moving 7. Politically, they belonged m rhe Radical Parre. In his smdent days in Lima, Vienrich had been
a disciple of che iconoclastic liberal , Manuel Gonzalez Prada.
between state and opposition. Dealing with Chis doubleness has generated
8. Quoted in rhe introducrion ro Vienrich's compilation, Fabulas Quechuas (1906; reprinted xg61)-
nor an ironic distancing but a propensity to adopr a politics of violence.
9. TMA Letters, x886, March 14, Mayor Philipps to Subprefeet, Tarma.
Emerging from Chis history is an overall vicw that the scope for translation by xo. TMA Letrers, x896, Sepmmber 8, Mayor Canrella ro che Official Inspector, Lima.
tcachers as mediating agents between state and province has been quite lim- u. TMA Letrers 1903, )uly 21, Mayor Vienrich to Prefect of )unín.
ited. IL is inconeeivable m consider social and political action taking place 12. TMA Letters 1904, April 30, Mayor Leon m district mayo,,.
without some element of translation; nevertheless, in view of the intransi- 13. TMA Letters 1905, April 4, Mayor Lcon to rhe Directorate of Primary Educados, Lima.
14. Vienrich, along with ano rher membcr of rhe Radical Parre, were hounded m death in i go8:
gence of both che state and its revolutionary opposition, there has not been
posthumousI)', he wou fa me as a ''folklorisr."
much room for mediation.

AN ANDEAN PROVINCL 343

342 STATES OF IMAGINATION
 15. There was a short turbulent period in che 194os when che state changed characier. An indi-
genista challenge to che state ' s educadon project in rhe 1940s was associated with critiques
launched by novelisi and educator José María Arguedas and by Luis Valcarel and José Antonio THE CAPTIVE STATE Corruption, lntelligence
Encinas. Por a brief period [hese roen were able ro influence state policy under rhe presidency Agencies , and Ethniciry in Pakistan
of Luis Bustamante (1945-1948).
16, From a speech given by Velasco in 1971, quoted in Drysdale and Myers (1975),
Oskar Verkaaik
17. SINAMOS is Sistema Nacional de Apoyo a la Movilización Social (National System for che
Suppor[ of Social Mobilization).
A. This was marked by a change of presiden[, when in 1975 General Velasco stepped down and
General Morales Bermudez took over.
19. Importan( discussions of Sendero's political thinking and agenda include tire following
works. Gorriti (1990, 1999) provides tire niost comprehensive chronological accounr of Sen-
dero's strategy and actions, and Poole and Renique (1992) also give a useful overview; Arce
(1989) reproduces versions of taxis by leader Abimael Guzman for foreign audiences; Har-
ding (1987) examines che work of Sendero intellecrual Diaz Martinez
; Tapia (1997 ) discusses
Sendero strategy from Guzman's notes caprured by the milirary; Taylor (1998) explores thcir
militar strategy ; Ansion et al. (1993 ) look at Sendero and education ; and Manrique
(1995a)
covers the later story Erom Guzman's arrest to the division of rhe Parry. Insightful commen- The terco PIA stands for Pakistan International Airlines, bur not according lo
taries on Sendero ' s polióes, rheir appeal, and represenration are found in Degregori (1991);
many Muhajirs, a term widely used in Pakistan for [hose whose families mi-
Ponle and Renique (1991); Starn (r991); Palmer (1994); Portocarrero (1998); Stern (1998); sud
grated from India in the years alter independence in 1947. Most live in the
Bcasley-Murray (1999),

20. 1 first heard che narrative while en a journey in 1997 and agreed with thc teller that ve could
two big cides of Karachi and Hyderabad in che southern province of Sindh.
later record it, though it lose wit and spontaneity in ti re retelling under more forma] cir- Since the ig8os, they massively support a political parry called the Muhajir
cumstances, Qaumi Movement (MQM; Muhajir National/Ethnic Movement) that has given
chape to a collective and political Muhajir identity. In 1996 and 1997 1 did
fieldwork in one Muhajir neighborhood in Hyderabad, an area of shoemak-
ers, blacksmiths, shopkeepers, streetvendors, bangle makers, rickshawdriv-
ers, and their families. In this area , most people say that che official meaning
of PIA is merely a promising veil of nacional solidarity that conceals a rather
disgusting face of ethnic exclusivism. The real but hidden meaning of PIA,
they say, is Punjabi International Airlines.
The airline offers its white-collar employees che safety of employment,
good educacional facilities, excellent medical care, housing in quiet, clean,
and rafe enclaves separated from the dusty and often dangerous city arcas
that surround them, possibilities for cheap foreign travel, and high social
status. Apart from che army, PIA may be the most attractive state-owned em-
ployer in Pakistan. But many of my Muhajir acquaintances say that they would
never stand a chance if they applied for a position in PIA. That, they insist,
has nothing lo do with merit. On the contrary, they maintain that Muhajirs
are weIl-known to be the most educated and patriotic people of Pakistan. Yet
PTA has been captured by Punjabis, and Punjabis will never allow a stranger
to enter their stronghold.

344 STATES OF IMAGINATION

 This is typical, rhey say, for che way Punjahis control al[ state institutions. about the stare. 1 argue thar the modero nation-stare is collecrively imagined
Thev do this rhrough family ties. A typical Punjabi extended family consists as a space of legitimare power, relativel}' unconneered ro and untouched by
of brothers, eousins, and ancles, who among dtemselves divide the strategi- real power siruggies thar are going ou wirhiu society, and thar this is done by
calle inapormnr posts. One js senr finto che arme, another develops a nenvork depicting che stare as aun expressio n. or caen the guardian, of trae Isianüc
umong bureauctats. a third rakes care oí rhe landed properry, a fourth doca principies of loyalry and authority. Srate funcrionarics, govermnent leadcrs,
the dirn work in rhe Punjabi Muslim Leaguc, a fifth worlos himself np finto and others associated with che process of narion building and srate tbrmation
rhe rank and file of rhe rival Pakistan Peopie's Parte, a sixth may become a in Pakistan have promoted an imago of dar stare as being opposed tu, and
judge, a sevcnth starts a family branch in thc United States, an eighth joins aunad to reform, prevuiling "un-Islamic" tendencies within seciete, suela as
P¡A, and so on. If one branch suffers, che others make up for the losses. As a ethnic or kinship loyalry and "fcudalism." At present, socieral groups and
whole, the Punjabi family prospers. emancipatory movements appropriare and use chis discourse for their own
When my informants talk like Chis, it is not always olear what they mean by purposes. Continuing lo reify the ideal of che srare as che potencial promorer
"the Punjabi family." Do they allude lo several families who happen lo be of a just society, they also acense che elites associated with state power of
Punjabi? Or do they talk about all Pakistani Punjabis who live as one extended potluting chis ideal by using ir as a veil to conceal the very particularistic loy-
family? Tire question appears irrelevanr lo them. What matters is that Pun- alties rhe stare is supposed to reform. This has added ro an increasing crisis
jabis arrange things according lo che wrong cense of loyalry. The Pakistani of stare authority, as, for instance, indicated by extremely low turnours during
stare, afrer all, is a creation for all South Asian Muslims. Its fruits belong tu nacional elections, a growing sense of displacement among eertain societal
all Pakistani citizens. Muhajirs roo nave the right lo a share of the profits groups, and a growing polarization within Pakistani society, manifested in,
of PtA. Any true Muslim would recognize thar. Yer Punjabis, like Western- for instance, the risa of ethnic and sectarian movements. 1 end with some
ers-sorry, my friend, don't Cake ir personally-care only about monee. They remados on why che widely held assumption that che risa of social unrest and
would sell Pakistan, a stare created in che name of Islam, to American capi- sociallpolitical violente is caused by a collapse of state authority is roo sim-
talists. Such an unparriotic attirude is in their blood. Did they not form the plistic, as it focuses solely on changes rhat occur within stare insritutions and
backbone of che British colonial army? A typical Punjabi would sell even his ignores possibly more importan[ transformations in society ar large.
own mother if it would benefit his interests. Small wonder, then, that Paki-
stan is in such a sorry stare today, captured as iris by unpatriotic hypocrites
Muslim Politics and che Reification of tire State
who only use Islam for their own purposes.
1 have lo admit that I have not quoted anyone in particular in Chis introduc- To a large extent, che contention of Eickelman and Piscatori (igg6) that Islam
tion. It is instead a loose compilation of many casual conversations. Perbaps eonstitutes the moral-symbolic language of Muslim politics is also true for
nono of my informants would endorse this view entirely-certainly not when Pakistan. Eickelman and Piscatori argete that most political debating, bargain-
presented to them by a foreigner or un paper. Yet the text eontains a basic ing, and cajoling in Muslim soeieties eontain explicit references or more ini-
idea that 1 llave heard often in Hyderabad, namely, that the Pakistani stare is plicit allusions to Islamic notions of loyalry, authority, respeet, justice, soli-
made in che name of Islam and belongs lo the whole of the Pakistani narion, dariry, and so on. Of course, despite elaims to che comrary, islam leaves much
bu[ is captured by one ethnic group. This essay is meant to examine more room for disagreement over such issues. In modero societies, islam may serve
fully this interrelatedness among islam, ethniciy, and images of rhe stare in many goals and objectives, as no person or instimtion enjoys a monopoly on
popular discourse among Muhajir residenrs of 1 lyderabad and Karachi. che interpretation of Islamic naditions. However, as is olear, for instante,
To do Chis, 1 relate Dale Eickelman and James Piseatori's (1996) remarks on in the case of women's movements, it is practically impossible, or at leas[
[he character of Muslim politics lo an approach for studying the stare as pro- unwise, to reject islam altogether as a common ground for social-pol¡rical
poscd by Philip Abrams (1988) and put to practice by Akhil Gupta (1995) in discourse.
che case of South Asia. 'che aim is to show how people in Pakistan can talk Apart froni and in addition to Islam, the state constitutes another legiti-

346 STATES OF IMAGI NATIGN Inc cAprrve STATE 347

 mizing principie in che name of which people claim to say or do things. it is
symbolism of state rituals and processions, state monuments, military uni-
probably true that for a long time states were built and maintained in the
forms, and bureaucratic new speech.
name of something else, such as che monarch, che emperor, or a religion; all
Building on [hese insights, Akhil Gupta ('995) has shown shas popular
che same, che idea that che state itself constitutes an autonomous principie of
images of the state hinge on che same separation of che state from sociery. In
authority seems to be at leas[ several cenmries old. This idea is based on what
an article en popular debates about state corruption in a northern Indian vil-
Philip Abrams ('g88) calls che reification of the state luto un almost sacred
lage, Gupta argues shas che very theme of corruption enables che villagers to
entity in che name of which order must be restored or people can be asked to
maintain a division berween themselves and local state representatives. State
make sacrifices-an idea of order shas to a large extent helps shape people's
functionaries enser local power struggles, bus they also deny Chis by claiming
perception of social reality. Because Abrams himself formulates what he
ro act in che name of an imparcial and translocal state. Not surprisingly, che
has in mirad more accurately [han I can, it is bes[ to quote him somewhat
villagers evaluare Chis discrepancy as hypocrisy or corruption. The villagers
extensively:
thus subscribe to che ideal of an imparcial and translocal stare and then con-
The state is nos che reality that stands behind che mask of political prac- demn che activities of local state functionaries on che basis of rhese high
tice. It is itself che mask, which prevenís our seeing political practice as expectations. The ideal of an imparcial state is maintained; in contras[, che
it is. . . . There is a state-system . . . a palpable nexos of practice and state as it is experienced locally is talked about as a pool of moral decay shas
institucional structure centered in government and more or less exten- pervades che state bus has little effeet on sociery. The separation between state
sive, unified and dominan[ in any given sociery.... There is, roo, a state- and soeiety is thus maintained ro confront che power abuse of local srare
idea, projected, purveyed and variously believed in in different societies functionaries with che moral superiority the villagers ascribe to themselves.
at different times.... The state comes roto being as a structuration The idea of a state that exists relatively independendy of sociery may thus be-
within political practice; it starts its life asan implicit eonsrruet; it is then come che basis for rebellion, resismnce, or other forms of subversive action.
reified ... and acquires an overt symbolic identity progressively divorced In che case Gupta (1995) describes, che [heme of corruption is che most
from practice as an illusory account of practice. The ideological funetion importan[ argument through which che villagers consrruct che image of a
is extended to a point where conservatives and radicals alike believe that polluted state or a besieged ideal of impartialiry and justice. Although corrup-
their practice is nos direcred at each other bus at che state; che world of tion serves a similar goal in Pakistan, 1 also think shas Muhajirs in Hyderabad
illusion prevails. The rask of che sociologist is to demystify; and in Chis have more strategies at their disposal ro make a similar argument about a
context that means attending to che senses in which che state does nos besieged and conquered state. Ethnicity, or che argument that other erhnic
exist rather [han to those in which it does. (8z) groups have capmred che state, is one of them. Apart from corruption and
ethnicity, 1 also pay attention ro a discourse revolving around secret iotelli-
It is of some importante to see shas che reification of che state can be
gence agencies. Seen as che hidden, true power players operating behind che
reached only by separating che state from sociery, and shas Chis separation is
bureaucraric fayade of the democratic state, intelligence agencies are said to
a discursive or imaginary practice. Coneretely, state functionaries are seldom,
corrupt che original ideals of che Pakistan Movement. 1 furthermore discuss
if ever, nos parí of sociery. They may belong to a particular ethnic community
che presente of rumors and che production of a quasi-secret atmosphere in
oran income group, they are men or women, and so ora; yes they derive their
which to disrribure such rumors that help construcc che image of che captive
power from che negation of their social identities and che claim that they, as
state.
stare functionaries, stand outside or aboye soeiety, act impartially, and are not
In this essay, then, I am most concerned with che myth of che state rather
direcred by any prejudice or self-interest. It is therefore importan[ to study
[han with che doings of che state. Rather [han focusing en che production of
how Chis idea of impartialiry and authority is being constructed and repro-
che state myth by state officials themselves, an issue someone like Timothy
duced by people associated with state power. This includes che study of che
Mitchell ('99') has paid attention ro, 1 limir myself mainly ro rhe popular

348 STATES OF IMAGINATION

THE CAPTIVE STATE 349
appropriation of rhis myth as ir is linked to die rhemes of corruprion, eth- In Pakistan today, words for differenr rypes of landlords, such as jagirdar.
nicin°, and [he works of intelligence agencies, al of which are "hot issues" ni zamindar, and especially ioadera, are widely used as insults in publie conversa-
popular debates in Pakistan toda. To put these popular lmages of rhe statc tinn. In specches. politicians accuse each other ot wishíng to rule Pakistan
joro pers pective, howevec 1 hrst say a few things about state formartun ín as "their privarejagir" (landed propern1). In iv series, the wadera persona is
Pakistan and how rhe idea of stare audioriry has been linked to a modernist almost by definition nce evil genios. This rather recent type of landlord who
I nterpretation of Islam_ canuor flaunt an aristocratic background, as he is usually froto a taroily who
used to produce the villagc headman, who worked a, the representarive ol
the more traditional jagirdar or zamiudar, is rypically portraycd as driving
Islam and Nationalism in Pakistan around his vast Lands in a four-wheel-drive Pajero jeep with tinred windows,
State formarion in Pakistan, as nation building, is closely related to modernist exploiting his bonded laborers (haris), raping rheir daughters and sons, drink-
trends in the interpretation of Islam as they were being developed through- ing whiskey, and terrorizing the countryside with the help of a privare army
out the nineteenth and early twentieth century. Perhaps more than anybody of armed strongmen (dacoits). This talk probably has its roots in the power
else, the poet-philosopher Muhammad lgbal has linked Islamic moderniza- struggle that has been going on in Pakistan between the ciry-based bureau-
tion to Pakistani Muslim nationalism. His ideas have been very influential in cracy and military, and regional elites of landlords (Jalal 1995).
Pakistan, so much so that he is generally considered the national thinker of Furrher, several forros of exclusivist solidarity that may divide the Muslim
Pakistan. The most prestigious university of the country is named after him, communiry have come under attack as being un-Islamic. This is most pro-
and his grave in Labore, located next to the Badshahi Mosque and other foundly the case with the concept of caste (zar), which is widely condemned
monuments of the Moghul Empire, has become a place of secular pilgrimage as being parí of a Brahmin ideology. As a result, many Pakistaní have dropped
for many Pakistani citizens. Iqbal argued that the future of an independent their indigenous family narres, most of which refer to a caste-speeific occu-
Muslim state in South Asia would rise or fall depending on the extent to pation, such as mochi (cobbler) or gasai (butcher), and Nave adopred a so-
which South Asian Muslims would be able to modernize Islam and free them- called ashraf (noble) name with an origin in Islamic territories such as Arabia,
selves from what he saw as erroneous innovations and regional pre-Islamic Persia, or Central Asia. To a lesser extent, kinship solidaria, as in so-called
traditions that had come tu pervert the original revelations of the Prophet. biradaris (endogamous groups), are condemned as being in violation of the
The new state of Pakistan was thus not simply based en Islam, or the ideal of ideal of Muslim unity, as symbolized by the hajj as well as the hijra, the exodus
Muslim unity, but, more precisely, en a particular interpretation of Islam that that marks the leaving behind of tribal ties in favor of a new religious com-
stressed the importance of modern education leading to an individual Bildung munity. Finally, regional solidarities, commonly labeled "ethnie," are often
into modern citizens and the ability of independent reasoning concerning said to endanger the all-Muslim unity.
religious matters (ijtehad). In nationalist ideology, particular practices were In the early decades of Pakistan independence, cultural and political elites
rejected as being traditional, backward, syncretic, a remainder of Hindu or depicted the state as a major initiator of social and cultural reform-literally
colonial influence, and therefore un- or not-quite-Islamic. as a set of parents (roa-bap) educating their children (Ahmad tg64: 103 -20).
Three sets of practices and mentalitics were considered especially unwor- Modernization was seen as a liberating gift presented by a social avant-garde
thy of a modern Pakistani citizen. The one that has been given most attention to a backward society steeped in tradition. Modernization was a mixture of,
in academic writing is popular Sufism, especially saint worship, which is dis- on the one hand, a mimicry of things Western, including ideas of military
missed by many Pakistani as either a sign of backward mentality or a Hindu discipline and economic development, a bureaucratic style of clothing, a
perversion (e.g., Ewing 1983, 1997; Gilmartin 1988; Werbner and Basu igg8). knowledge of British philosophy and poetry (at a minimum, one was ex-
Two others have been less noticed rhus fax. "Fcudalism" has been identificd pected to be able to quote from Russell and Wordsworth), and on the other
as being in contradierion with the Islamic ideal of social equality (musawat). hand, a training in a sharif or noble, mostly urban culture, centered around

350 STATES OF IMAGINATION THE CAPTIVE STATE 351

 the Urdu language as it had been developed in the nineteenth cenmry in Partly in Chis context, young Muhajirs in Karachi and Hyderabad mobilized
northem India. Condemning saint worship, "feudalism," and particularistic themselves en the basis of a new Muhajir identity from che late 1970s on-
loyalties as tradicional and un-lslamic, members of urban elites tried to com- wards. They became the most vocal of [hose groups that have not been will-
pete with established regional and rural political nerworks centered around
ing to identífv with one of che tour recognized ethnic groups. In 1978, che
powerful landlords and spiritual leaders. The effort to make che state of Paki-
All-Pakistan Muhajir Student Organization (APMSO) was formed, followed in
stan prosper, as well as the attempt to purify Islam from Hindu and colonial
1984 by che Muhajir Qaumi Movement (MQM). The MQM has since then made
influences, were rheir claims ro legitimacy.
headlines in both che Pakistani and international press as it has been a major
From the late rg6os onwards, a counterdiscourse emerged that developed player in the dratnatic mercase of ethnic polarization and violente in urban
and promoted regional or "ethnic" identities. With some exceptions, region- Sindh. It is a movement with many different faces. From an academic stand-
alista have not rejected che idea of Muslim narionalism as such, but have tried point, it is movement of remarkable social mobilization based on a rare per-
to give a new meaning to the concept through a much more positive evalua- ception of ethniciry, as the MQM does not claim a separare ethnic identiry for
tion of regional syncretic Islamie traditions. Especially Sufism or tasawwuf, as Muhajirs en the basis of a shared homeland but rather en che basis of a
the mystical, less shari'at-oriented branch of Islam, has been reevaluated as a
shared experience in exodus and migrarion. For many of its ssupporters, the
form of Islam that is more at honre in che various provinces of Pakistan than MQM is a revolutionary movement that has given voice to large sections of
che reformist notions of the urban modernist elite. There has been more at- the underprivileged population of Karachi and Hyderabad. Its critics associ-
tention to ecstatie practices, as well as more physical expressions of faith-
ate the Party mainly with armed terror, ethnic discrimination, nepotism, and
music, dancing, crying-which in che rg6os and 1970s especially attracted crime. The Parry, which has won elections in urban Sindh with large margins
che urban, vaguely left-wing students, intellectuals, artists, and others, who since 1987, has seriously suffered from an internal split in 1991 and even more
found in these passionate and swinging practices of devotion more opportu- from attempts of various governments to break che back of che party's srreet
nities for social rebellion, fun, and identity construction than in che mod- power in a series of so-called Operations Clean-up, which have been criticized
ernist emphasis on che importante of reason ('aql) and discipline (nazm). In by human rights commissions for rhe extrajudicial methods that reportedly
the province of Sindh, which together with forrner East Pakistan became the were being employed.
most rebellious province in the late rg6os, che shrines of two local saints, Lal
Shahbaz Qalandar and Shah Abdul Latif, became importan[ centers of re-
gional resistance and che construction of a separare Sindhi religious and po- Corruption and Violente
litical identiry.
In the neighborhood where 1 did fieldwork, the experiences people have with
The rediscovery of syncretic traditions and Sufism was [alcen up by che state functionaries clearly do not meen their expectations. Expectations are
Pakistan People's Party (Ppp), which in 1970 won che first democratie elec- based en che ideal of an imparcial, rransparent, autonomous starc, which,
tions in che history of Pakistan. During che administration of the PPP, ethnic as a benevolent and just parent, rules over sociery. That is the standard-
or regional identities were promoted as intrinsically parí of che complex iden- the way things should be. In practice, however, that ideal is felt to be con-
tity of a Pakistani citizen. Whereas a Pakistani identity had always had a na- stantly corrupted by dishonest and insincere state functionaries with whom
cional and religious component, an ethnic one was now added. Four ethnic the people of che neighborhood have to deal daily. As Gupta ('995) also ob-
groups were more or less officially recognized; [hese coincided with provin- serves in his northern Indian village, there is a profound discrepancy berween
cial boundaries, so that a Pakistani was required to identity as either Punjabi, expectations of che state and everyday experiences with local state function-
Sindhi, Pakhtun, or Baluchi. Moreover, a guata system, which distributed po- aries, a discrepancy that serves as che basis of a discourse of popular prorest
sitions in che civil service as well as places in educacional institutions en the and resistance.
basis of ethnic background, allotted certain rights to members of these vari- Most residents of my neighborhood-all of them are Muhajirs-subscribe
ous ethnic groups.
to the legitimacy of state authority. In Chis respect, they consider themselves

352 STATES OF IMAGINATION THE CAPTIVE STATE 353

different from, for instance, the Pathans in the mountainous north, who, as vants, and most look ir state instirutions as impenetrable basrions. It is gen-

a tribal people, are said to be m unruly, unpredictablc lot who do flor recog- erally considered a oraste of time and energy to apply ter a 01) 111 ti], civil

nize rhe authorirv of rhe state. Norwirhstanding rhe caer rhat Pathans are to service unless you are willing and able to pae a bribe as high as one vcarb

sorne estent admired for their proverbial cense of independence, thev are also salary. Purther, rhe implementariou of governmental deveiopment schemen

condemned for their tribal laves, espeeially rhe custom of blood feud, which (Iealizarion of illegal housing, consrruction of sewage systems, ete.) depend,

are leen as evidence of an un-Islamie and baelavard mental tv and a violation on personal conracrs or high bribes and tales a frusrraringly Long time. 1112

of borh die shari'at and rhe Constiturion. Very often. rhe discourse of modcr- police in particular are considered unrrusttivorthy, hostile, and benl iliatiug:

niry is being used in chis respecc dte tribal laws are raid to bclong to a pasr when in troublc, stav away from rhe police ifyen do nor want more problem
that educated people have lefi behind. These traditional laws, based en exclu- Nevertheless, it is nor always possible to avoid rhe police, who regularly ser

sivisr loyalties of tribe and family, should now be replaced by the impartiality up checkpoints to relieve passers-by of their money and valuables.

of the shari'at and the bureaucratic laws of the modera nation-state. More- Over rhe past few years, corruption has become a major theme in polirical

over, as Muhajirs who claim to have left their homelands in India for rhe ideal debates-nor only en a popular level, but also in newspapers, en Tv, and in

of Pakistan, rhe residents of the neighborhood feel that they have profoundly rhe Nacional Assenibly. Prime ministers are usually dismissed en tbe charge

contribured to die establishment of Pakistan, which was expected to make of corruption: Muhammad Khan lunejo in tg88, Benazir Bhurro in 1990 and

real che ideals of Muslim uniry and social equaliry and put an end m par- 1996, Nawaz Sharif in 1993 and r999. In November 1996, after Biturto had

ricularistie solidarities and tradicional law systems. Generally speaking, the been dismissed, a major operation was launched by the interim government

residents identify rhe srare as an agent of modernity and reform and hold that to overcome corruption. With great fanfare, an Accountability (Ehtasab) Com-
mission was insralled thatwould check the bank accounrs of every major poli-
Muhajirs- as an educated people and a people who went through rhe chas-
tening, if flor purifying, experience of exodus-are especially equipped to tician and bureaucrat. During rhe lame period, Pakistan appeared second on

understand and promote rhe ideals of Islam and moderniry. Their everyday some international lisr of most-corrupt countries, which beeame a popular
talk is full of condemnations of rhe identified antipodes of moderniry, flor copie of publie debates and jokes. Some raid thar it oras to be Pakistan's tragic

only tribal laws, but also Sufism, feudalism, and Gaste and kinship loyalties. fase never to be rhe first in anything; others counrered that Pakistan had ini-

Although always softened by an ideology of Muslim uniry and nacional soli- tially been on rhe top ofthe list, but then some government ofñcial had bribed

dario, rhe idea thar Muhajirs are the champions, if not the chosen people, of rhe makers of rhe list to persuade rhem to put some African state first. Mean-

Pakistan regularly pops up and most often comes with a feeling of solitude while, rhe nacional relevision network daily reported en the findings of the

and displacement, of having been betrayed and left in rhe cold by rhe other Ehtasab Commission, showing the commissioner behind a huge pile of docu-

peoples of Pakistan, who have barrered away the ideals of Pakistan for their menta on his desk, or solemnly talking wirh rhe president of high court

regional or ethnic interests. The personal experience of migration and che judges againsc rhe background of high-ranking uniformed soldiers. Ir created

stories that are told about this in rhe family and rhe neighborhood have al- the illusion rhat corruprion and nepotism were notan intrinsic part of state

ways been linked to rhe nacional experience of independence and, subse- power but, en rhe contrary, a foreign, if contagious and dangerous, disease

quently, rhe building up from scratch of Pakistan. The sacrifices that are said rhat had affected many a state body and alar was now being fought wirh all

to have been made by Muhajirs have never been properly appteciated. On rhe the power the state could muster. Interestingly, rhe Ehtasab Commission it-

contrary, rhe state, once established, has been eonquered and barricaded by self has since become a subject of controversy and allegations of political

orhers who are less receptive to the liberating ideals of Islamic modernism. intrigue , but although this has affected the aceountability of yet anorher state
With such arguments, people in rhe neighborhood evaluare and talk about institution , it has hardly damaged the belief in the possibility of a transparent
their daily dealings wirh local state officials. Very few of them work in state and impartial state as such. Ir has, however, seriously added ro a general pub-

institutions, although some work as policemen or other low-grade civil ser- lic distrust of politicians and bureauerats, who are commonly believed to "ear

354 STATES OF IMAGI NATION THE CAPTIVE STATE 355

 your brain" (dimag khana) with hollow phrases of Islam and nacional solidariry to visir anybody likely to be an isi informer, to drink a cup of tea, have a chal,
while ruthlessly exploiting che common or peor people (am log, gharib log). and leave a good impression. It took os a couple of days inside travel agen-
The popular perception of the state is shaped not only by corruption, but cies, petrol stations, post offices, tea stalls and restaurants, shrines, offices
also by violente and humiliation. According te che neighborhood residents, of lawyers and NGOS, che Press Club, and several shops and workshops in
state violence has become salient and more brutal since che 1970s. This in- che bazaar. Most intriguing of this limited study was to find out that the iden-
eludes everyday instantes of harassment by che police, but also more violen[ tity o£ intelligence agents was known te, or at leas[ suspected by, a large num-
state persecution of potencial militants of the MQM. Baclced by massive pro- ber of people. When I returned to che neighborhood of my fieldworkwith the
paganda that depicts che series of Operations Clean-up as battles against "ter- exciting idea that 1 had qualified information to tell, it appeared that most
rorisrs" funded by foreign enemies such as India, stare forces have killed a informers 1 had been drinking tea with were already known there.
large number of MQM militants in what rhe press commonly defines as I do not want to suggest that Pakistan is at par with che state of Rumania
"police encounters." At times, the army has proudly, amos[ as a trophy, pub- prior to che ig8g revolution in which, as Katherine Verdery (r9g6) describes,
lished che monthly number of dacoits (bandits) it has shot dead. Severa¡ news- an astonishingly large percentage of the population was involved in secret
papers and magazines have published critica¡ stories about [hese practices intelligence work. But I am convinced that intelligence agencies are an estab-
together with pictures of maimed bodies. Fureher, the youth of che neighbor- lished aspect o£ Pakistani politics and that the isi is not the only such ageney
hood rell each other stories about peers who have been arrested and put in operating in Pakistan. In fact, several 1aw-and-order forres run thcir own
jail, where they are reportedly raped, urinated opon, made to squat for pro- ageney and often appear te be interested in each other as much as in anybody
longed periods of time, and othermise humiliated. else. This also means that the state cannot be regarded as Big Brother, spying
Despite these experiences and rumors, the people in the neighborhood en its subjects through secret activities penetrating privare places and thereby
have not altogether lost their belief in the ideal of a just and imparcial state. effectively keeping sociery under its thumb. It instead resembles a troubled,
Rather, they have found explanations that ascribe the moral roa to particular fragmented family of several brothers who are deeply distrustful of each other
groups of people who are said to corrupt the ideal in a concealed manner, and-cannot rely too much un each other in their dealings with the outside
according te secret deals and hidden loyalties. One of these explanations world. Partly for chis reason 1 think that the influence of the intelligence agen-
hinges en che presente of intelligence agencies. A second builds en the no- cies should not be exaggerated, even though there is little doubt that che isi
tion of ethnic exclusivism. in particular is an importan[ player in Pakistani politics.
Nevertheless, many wild speculations can be heard about their supposedly
Popular Explanations of Corruption: far-reaching power. Almost ala major political developments are regularly as-

Intelligence Agencies and Ethniciry cribed to che secret manipulation of "che agencies," a category that usually
remains unspecified. Large-scale riots are ofren believed to be masterminded
A great deal of political rumor in Hyderabad is about agencies. Intllligence by "che agencies." This also goes for remarkable election results. It is com-
agencies are believed to be the real rulers of Pakistan, che secret manipulators monly believed that "the agencies" work for particular economic, political,
of politics. They are believed to be omnipresent, and to some extent that and military elites who try tu defend their "vested interesas" through a policy
seems true. This was brought home to me one day when a friend told me that of divide and rule. The idea is that the ideals of Pakistan-social equality,
a man who had identified himself as an agent of the Inter-Service Intelligence Muslim uniry, che abolition of feudalism, and so on-would harm therr po-
psi), che army's intelligence ageney, had visited him. The agent had asked sition, but because they cannot openly declare their unpatriotic conservatism,
about my doings in Hyderabad. The bes[ way to get rid of him, my friend they need secret agencies to spread hatred and distrust among che common
suggested, was to casually inform the right persons about che harmless char- people of Pakistan. In Chis way, the problems that pester Pakistan-poverty,
acter of my research. The way te do this was to make a tour around the ciry corruption, ethnic polarization-are seen to be che result of the wheelings

356 STATES OF IMAGINATION THE CAPTIVE STATE 357

and dealings of a happy few who are in control of che secret niches of the Agra, che Sindh looked like an uncivilized backwater and an unhcalthy wil-

Pakistani state, rather than of inequalities in social relations or ethnic or re- derncss of extreme hear, dangerous anlmals, strange diseases, mad mullahs,

ligious prejudices, which are deeply ingrained in society at large. and fbnacical fakirs. Such notions 'vete also prometed in colonial writings.

Comparatively speaking, t1115 discourse en intelligence agencies is rather fhere are many exaniples of chis, but che most well-known is probahly Su

optimistie and less dangerous rhan thc other explanarion fbr suite corruption, Richard Burton's travel aeeounts. In an early wodc Burton virote, "Ihe darle

which revolves ¿round ethnic conipeticion. Conipared ro the ethnic discourse. complexion of che Sindhi points hiel out as an instance of arrested dcccl-

argumcnts abour che prevailing pnwer ot intelligence agencies put che blame opment.... He is idle and aparhetic, unclean in bis person, and addiered to

un an almost invisible and unidenrifiable enemy. According ro rhis talle, dif- roxication; notoriously cowardly in times of danger, and proportionally inso-

ferent social groups within Pakistani sociery would live peacefully together if len[ when he has nothing tu fear; he has no idea of truth or probiry, and only

ir were flor for che dividing scheming of unknown intelligence agents. In con- wants more talenr to be a model of treachery" (1988: 283-84). To be fair,

rrast, the argument of ethnic differences potentially blames anyone who be- Burton did malee in exeepcion for che Sindhi woman. In a later work, which

longs to che rival ethnic group. also contains a chapter entitled "The Sindi Man: His Charaeter, and Especi-

In the beginning of chis essay 1 sketched che contents of che argument ally What He Drinks" and that was recently reprinted by che Department

about Punjabi biradari solidarity. Punjabis, however, are not the only ethnic of Culture and Tourism of che Governmcnt of Sindh, Burton describes the
group said tú have capmred and polluted che state by unpatriotic and un- Sindhi woman's attitude as "nor ungraceful, she carnes herself well, she never

Islamic means. In a slightly different argument, Muhajirs also accuse Sindhis stoops and, observe, she has high but nor round shoulders" (1993: 328).

of dominating che state through tradicional and backward forros of loyalty Although informed by older, pre-Independence traditions of prejudice, it is

that damage nacional solidarity. lf Punjabis are condemned for their exclusiv- clear that popular images of a state captured by a backward ethnic group to a
ist kinship mentaliry, Sindhis are considered an inveterately feudal people. large extent foltow che discourse of modernity and tradition as it has been

The Sindhi way to domination is through che vertical networks of feudal pa- promoted by che mosdy urban elites, who have considerably shaped che pro-

tronage headed by either a landlord (wadera) or a spiritual leader (pír). These cess of state formation and nation building in Pakistan. In che popular imagi-

rural big men are primarily blamed for che prevailing feudalism that is said to nation, we see a similar notion of a reforming state fighting che evils of tra-

eonrinue to dominare Pakistani politics and that is especially believed to char- ditions, such as feudalism, superstition, or, in che case of che anti-Punjabi

accerize che PPP, which is led by che Sindhi landlord family of che Bhuttos. argument, a roo exclusivist kinship solidariry. From a nacional perspective,

The ordinary Sindhi, however, is nor considered free of guilt either. He is said such tradicional mentalities have always been depicted as che crait of particu-

to be of a lazy, unmanly, submissive, and superstitious disposirion, which lar ethnic groups. In che hands of underprivileged social groups, such as che

does nor befit a modern Pakistani citizen. In a gesture of welcome, he puts Muhajir shoemakers and rickshaw drivers in the neighborhood of my field-

bis hands before his chest like a Hindu instead of offering a fellow Muslim work, such ideas are appropriated to lay bate che hypocrisy of che privileged

his right hand. He bows in front of his master. He begs rather than demands. straca of society. These norions feed into a sense of betrayal and displacement

Some residents in my neighborhood also said that che Sindhi prefers fish for so pivotal to present-day Muhajir identiry.
dinner rather than mear and thar chis disputable taste symbolizes his fishy,
unreliable, deeply feudal character. A Geography of Secret State Power
It would be an interesting study to trace down che genealogy of such ideas.
I am not in a position to do Chis, bur it is clear that negative images of the The captive state, then, is seen as che truc, but hidden, image of power that

backward Sindhi Muslim (che Sindhi Hindu, in contrast, has che reputation is coneealed by empry words about bureauerarie rransparency, Muslim frater-

of an energetic and efficient moneylender and trader) were already circulating niry, and self-effacing patriotism. Iris imporcant to realize that the hypocrisy

in northern india during colonial times. From cides like Delhi, Lucknow, and that is considered self-evident and for everyone ro see is at che same time said

358 STATES OF IMAGINATION THE CAPTIVE STATE 359

 to cake place within a domain of secrecy that cannot bear che light of day. It unlikely places, such as air-conditioned restaurants, which had the additional
is in this sense that che image of che captive state contributes to che repro- advantages of coolness and expensive beverages, to put me in the know. Two
duction of a separare Muhajir identiry. It is not simply che fact thatcorruption, places were especially favored for this purpose. One was a poorly lit restaurant
nepotism, and abuse of power have rendered talk about che impartiality of on one of Hyderabad's main srreets, owned by a well-known man who him-
che state meaningless-a development that has been going on for several
self had a reputation of being close to the secret center of things. He liad
years and is now accepted as a fact by probably most Pakistani citizens and, been a candidate in local elections in the 1970s and had spent brief periods
for that matter, most foreign observers. More important, such practices are in jail for his political activities. In those days, the restaurant had been a meet-
said to cake place in an atmosphere of secrecy and conspiracy of which che
ing place of politically active students. Since his retreat from active politics,
common people (am lag) are excluded. This sense of exclusion from what the owner usually sat outside che restaurant, mostly accompanied by one of
actually and rightfully belongs to che nation-a notion that, in the hands of
his former friends and comrades. Inside, the fans on the ceiling were ram-
Muhajirs, has an ethnic as well as a class component-is an important theme shackle enough to make plenry of noise so that privare conversations could
in present-day Muhajir identit. not be overheard.
It is, of course, well-known that secrets, rumors, and gossip can be impor- The other place was a coffee house, equally centrally located and also
tant means of communication through which to establish a sense of com-
known as a place where radical politicians, students, journalists, and, also,
munity or loyalry. From the theoretical works of Simmel (1950) and Goffman
intelligence agents mer. Like che air-conditioned restaurant, it maintained a
(1959) to, for instance, a recent study of ethnic confiict in south Senegal (De reputation as a place of rumors and secret knowledge. Amazingly, then, to
long 2000), we find che notion that secrets must first be identified as secrets exchange secrets, my informants would often propose going to places thar
to become secrets at all. Second, secrecy creares a community of people who
rnade one suspect by just being there. These places served as identified hot-
are "in the know," as opposed to che ignorant, who are therefore outsiders. beds of classified information; being seen there to some extent made one feei
In this way, a secret, regardless of its contents, may become che glue that involved in the secret center of local politics. ti made these places, which also
binds a group together. But secrecy can also generare insiders and outsiders included certain shrines of Sufi saints, the Press Club compounds, and stu-
in a different way, by creating che fiction of a secret syndicate that controls dent hostels, awe-inspiring. In addition, they were characterized by an ano-
and penetrares sociery through covert means. Rumors about Freemasonry nymity that went well with the rumors that were being exchanged. These were
and a worldwide Jewish conspiracy are historical examples. typically without an author; they carne from nowhere in particular but were
In Hyderabad, secrecy and che communication ofrumors worked both ways. believed by everybody.
On the one hand, rumors about intelligence agencies or invisible biradari These places, then, formed a geography of power complementary to che
or feudal networks depicted opponents as scheming conspirators connected prestigious and often colonial architecture of official state buildings, such as
through tradicional and less than transparent nerworks that are anathema che court buildings, the head office of che police, and the army bungalows in
to che ideals of a free, democratic, Muslim nation-state. On the other hand, the lush Cantonment arca. Considered the darle workshops of local politics,
however, such rumors were seldom communicated openly. Although every- rhey were seen as places not only where conspirators would meet, but also
one was aware of the existente of such rumors to che point that they can where che secret bodies of che stare gathered information, bribed informants,
rightly be called public secrets, most of my informaras took care to contribute and hired people for ¡Ilegal activities. If conspiracies were masterminded
to che notion of secrecy by first crearing a conspirational atmosphere before within che drawing rooms of Punjabi biradaris or che residences of Sindhi
telling me what 1 had already been told a thousand times. Ata minimum, the feudal lords, they were believed to be carried out in the workshops of covert
person would lower his voice, take me by che arm, and look at me signifi- state power where one could fiad che people willing to do the dirry work. On
cantly to indicare that the words ro come were both valuable and dangerous. visits to such places, my informants would never fail ro point out to me che
Other people were not contenr with chis and would propone to recrear finto men who had been involved in spying activities.

360 STATES OF IMAGINATION

THE CAPTIVE STATE 361
 I t s i not
l my inrention ro argue rhat rumors about covert ncrworks of power take tare of thcir inrerests in a chaotic and opportunistic manner. In Pakistan,
and illegal state activities stem from collective tantasies that have uo basis in ir is often recalled that rhe nrost violent cpisodes in rhe hisrory of ncc country
actual fact. On rhe conrrary, inrclligence agencies do play an important role occurred during rhe collapse of an authodtarian militan regime, in rho late
in Pakistani politics, jusr as biradarí and parronage norworks continuo ro be rg6os and die late 1g8os. Democratic rcgimes, ir is argued, are weuker and
ini portan[. However, rhe annosphere olí secrecy in which such activities are less willing or able to maintain law and arder through violen[ means.
discussed makes it possible ro overestimate thcir influence. Probably rhe mosr 1 have difficulties with riese assumptions. Surely ir is difficult to maintain
well-knowu example of this is rhe contention rhar rhe army, through rhe ille- rhat die Pakistani state has becmne a wcak stare in rhe r99os. A largo number
gal and hidden activities of the isi, is responsible for the rise of rhe mQM. of hcavily arnied milirary, paramilirary, and police tones have beca broughr
This is a rumor that has become a widely accepted fact among rhe erirics of out irlo rhe streets of Karaehi and Hyderabad since rhe late t98os. 1 am not
rhe MQM. The idea is that in the mid-rg8os rhe milirary regime needed a so sure that rhe transition to dernocracy since 1988, an arrested process lince
popular political parry in the province of Sindh to curb the power of rhe pie, rhe coup of tgg9, lowered rhe readiness of government leaders to use force
1
the regime's main opponent. Although it is possible that rhe isi had some against civilians. All the same, it is possible to argue that rhe state is inereas-
dealings with the MQM in its early years, it seems rather farfetched to make ingly characterized by fragnrentation. Various state institutions are constanrly
rhe army responsible for the massive popular support the Parry subsequently in comperition with each orher. The most dramatic expression of chis is the
received. The argument contributes ro the notion of an omnipresent, all- violent confrontations between police forres and the inilitary in a number of
powerful state fiar may produce popular movements wherever it wishes. Such incidents (Verkaaik 1999: chap. 7). Much of this is the result of the fact rhar
runiors thus rcinforce die notion of a state that is both separared from society new administrations often establish new governmental bodies and law-and-
as well as capable of manipulating society totally. Without wishing to belitrle order forces and staff rhem with known allies. rather than work with, or rey
rhe force that state officials can mobilize through both overt and covert to rearrange, exisring institutions staffed by civil servants whose loyalty is
means, it is also olear rhar the rise of such popular movements as headed by doubtful. In this way, social and political rivalry is firmly established within
the MQM are, among other things, forras of protest against state crime and rhe state apparatus, but that in no way implies that these various state insti-
corruption ratlier [han rhe planned outcome of state manipularion. Rumors tutions are less willing or able to use effective force.
that make rhe isi responsible for the movement, however, enable one to close Rather than look at rhe state as an autonomous, unified, and monitoring
one's eyes to rhe complex changes that are going en in society at large and agency, then, it appears more fruirful to study rhe state as a particular parí of
that offer much more plausible, if alarnring, explanations for ethnic conflict. society that is characterized by a relatively high degree of political activities,
Instead, they enable one to interpret the crisis as a crisis of the state alone. including competition over resources, exchange of information, ideological
By way of conclusion, 1 will say a few words about this tendency to explain debates, and rhe use of physical coercion. One should not look for causes of
complex social rransformations solely in [erras of a crisis of governability. this increasing fragnrentation of the state merely in the state apparatus itself,
but rather take it as a reflection of a more far-reaching transition of society
at large. To quote Sandria Freitag, who has made a similar argument in thc
Conclusion case of India, present-day social unrest and political violence are symptoms
A very influential, almost commonsensical, theory of rhe state explains tic of "a much larger and more challenging process under way: the redefinition
recen[ rise of political violence and social unrest in Pakistan in terms of a of Indian civil social space and who will be allowed to participare publicly
"crisis of governabiliry" or a "collapse of state power" (e.g., Nasr 1982). Ac- inside that discursive space" (1996: 232). The same applies to present-day
cording to widely held public opinion, the trouble is caused by a decay of Pakistan. In other words, rhe argument that social unrest is caused by a col-
political institutions and an erosion of state authority. When rhe state loses lapse of state aurhoriry fails tú acknowledge two important facts. First, ir ac-
lis grip on socicty, the subjeers, according to an almost Hobbesian logic, will cepts rhe idea of a separation betwecn state and sociery, though there is no

362 STATES OF IMAGINATION THE cAPTIVE STATE 363

 evidence that such a strict separation exists outside the domain of the political
imagination. Second, it reduces social unrest lo a crisis of the state, whereas PUBLIC SECRETS, CONSCIOUS AMNESIA, AND
it is an expression of changes that are taking place in a much larger part of THE CELEBRATION OF AUTONOMY FOR LADAKH
society.
In this essay, then, 1 have tried to initiate a more integrated approach of
Martijn van Beek
sntdying present-day relarions between state and society in Pakistan by look-
ing at the state "from below," that is, by examining popular perceptions of
the state within an "ethnic" group that has been at the heart of social transi-
tions and political unrest in urban Sindh since the mid-tg8os. Building en
che work of Eickelman and Piscatori (r996), 1 have argued that the popular
imagination of the state is linked lo a modernist concept of Islamic nation-
alism so that the state carne to be regarded as the leading agent of religious
and social reform. This notion of the state as a benevolent parent was itself
based en the myth of che state as separated from and impartial to society.
In turn, this ideal of the state has become the basis for a popular critique Die Welt ist einmalig. Das blofle Nachsprechen der Momente , die immer und immer
of state practice. Expecting the impossible, namely, a truly transparent and wieder als dasselbc sich aufdrüngen , gleicht eher ciner vergeblichen und zwangshaf-
democratic state sincerely devoted to the ideals ofMuslim fraternity, Muhajirs
ten Litanci als dem erlósenden Wort. Klassifikation ist Bedingung con Erkenntnis,
have grown deeply disappointed with and distrustful of che state, which they nicht sic selbst , und Erkenntnis lóst die Klassifikation wiederum auf.-Horkheimer
feel fails lo live up lo its own high standards. Instead of reforminga backward and Adorno , Díalektik der Aufklarung
tradition, the state has been captured by social groups who, through secret
and invisible means, hold on to their particularistic sense of loyalty and use Bureaucrats work on the categories of social existence much che same way as sor-

state power to enhance their self-interests. The growing attention of the cerers are supposed to work on the hair or nail clippings of their intended victims.
-Herzfeld, The Social Production ofindi$erence
theme of corruption in Pakistan must therefore be seen primarily as a sign of
emancipation rather than an increase of immoral behavior among state offi-
cials. The practices now increasingly labeled as corruption are not new. What Paradoxes of Recognition

is new is the growing sense of indignation about [hese practices in Pakistani

On a sunny morning in September rggg, one of Ladakh's political leaders and
society.
1 were sitting in his garden, sipping tea. A week earlier, he had been aniong
the twenty-six elected and tour nominated members of the new Ladakh Au-
Note tonomous Hill Development Council in Leh (LAHDC), who were sworn in
during a ceremony presided over by the governor of the State of Jammu and
This study was parí of a dissertation project. which has resulted in a dissertation entitled "lo-
Kashmir (J&K), General K. V. Krishna Rao, PvsM (Retd). Congratulating the
side che Ciudel : Pon, violence, and Religious Nationalism in Hyderabad , Pakistan ."
The re-
people on their use of "peacef il and democratic means," the governor said,
search was financed by the Netherlands Foundation for che Advancement of Tropical Research
(worxo) and was carrled out at che Amsterdam School for Social Science Rescarch (ASSR). "This is a shining example for everyone in out democratic country, that one
can achieve one's legitimate demands through constitucional means."' That
morning che young political leader and I shared a sense of closure: for me,
the creation of the Hill Council coincided nicely with the end of severa) years
of research on the movement that led to its constitution. I could hardly have

364 STATES OF IMAGINATION

wished far a berter conclusion ro nty doctoral project. Far him, autonomy ership of rhe Ladakh Buddhist Association.' This movement, not quite so
signaled Che end ofa long struggle whose final, suceessful phase he had beca peaceful and dcmocraric as Che governor made out, relied on a eommunalisr
instrumental in planning and implemeuting. The Hill Council signihed an srrategy, arguing ruar Buddhist Ladakh could prosper onKit it were liberated
opportunity to hnullo Cake charge and show Che people, Che state government, from Che yoke of Muslim Kashmir. Iadaldi, then, was Bttddllisu Buddhists
Che world. rhat Ladakh could develop roward greater prosperity while protect- are patriots; logically, Muslims are not proper Ladakhis, and-as Che sepa-
ing its cultural heritage. Even communal harmony among rhe dominan[ reli- ratisr violenee in rhe Vallcy testified-dice were not parrlots and eould not
gious connnunirl es of thc regios which had suffered during the agitation, be trusted.
could be restored, he was cure, as Muslims liad been given adequate reprc- The purpose of Che prescnr essay is te) explore crhnographically Che ways In
sentation on Che council. Where generations of previous political leaders hall which oficial and illegitimate hegemonic grammars of idenrifieation, based
failed, bis associates and he had succeeded. 1 asked him whether he regretted en reifying and often ferishized imaginings of communiry and identity, do
anything about Che agitation. He expressed regret over some of Che "arco not necessarily evoke either compliance or resistance, bur more complex
twisring" that liad been necessary, "but we couldn't have done it any better practices of representation, performance, and belonging. The celebration of
way. Even at the risk of blowing our own horn: we did it well." autonomy for Ladakh is premised on Che public recognition of a unique iden-
Although the amiosphere among the spectators and invited guests at the tiry, which must remain unspecified and fuzzy to be recognizable tú bureau-
swearing-in ceremony on Che polo ground was one of celebration, rbc gover- crats, politicians, scientists, and ordinary people. The unambiguous speei-
nor's eseort of "Black Cat" eommandos, constantly scanning Che crowds, fin- ficiry of Ladakh's officially recognized eight Cribes is singularly unsuired for
gers en Che trigger of their automatic weapons, provided a reminder of Che thc kind of shared notion of Ladakhi identiry that is purportedly being cele-
fact that elsewhere in I&K there was considerable resistance to India's clainis brated, and has no basis in local understandings of identiry and belonging.
of " Unity in Diversiry" and secularism. Yct, while sections of Che population Atthe same time, Che dark realiry of Che communalist strategy thar, as Ladakhi
of Kashmir had been fighting for decades m break away from India, Ladakhis political leaders and many others are convinced, at last led to Che reeognition
had for an even longer period been struggling to "Free Ladakh from Kash- of the legitimacy of Ladakhi demands for autonomy is regarded as illegiti-
mir." The Ladakh region covers an area of about sixty thousand square kilo- mate and unsuited to public recognition and certainly celebration. Both Che
meters in rhe rain shadow of the Himalayas ro the northeast of Che Kashmir public reeognition of Che tribes in the straight-faced applieation of a socio-
Valley.' The estimated rwo hundred thousand people who inhabit Che region logically impossible set of criteria, and Che denial of communalism as a real
constitute almost equal shares of Muslims and Buddhists, yet Che region is and effective strategic resource in Indian polirics as expressed in Che celebra-
commmily characterized as Tibetan in tercos of culture, religion, and tan- tion of Ladakhi identiry and the granting of autonomy, illustrate the saliente
guage. Muslims predominare in Kargil District, and Buddhists constitute Che of everyday dissimulations that make Che impossible puriry of exclusivist
overwhelming majority of Che population in Leh District. At leas[ since the categorizations practicable. Neither communalism sor rribalism captures Che
I93os Ladakhi grievances had been expressed in tercos of Che "unique condi- "realiry" of social identification in Ladakh, but their realiry as sources of em-
tions" of the region. As Secretary to the Govenunent, )ammu and Kashmir powermenr-a realiry premised on rhe shared imaginings of bureaucrars,
Affairs, K. Padmanabhaiah wrote in rhe "Reasons for Enactment" for Che politicians, activists, and scientists-requires dissirnularions thar make them
Ladakh Hill Councils Act: "Ladakh region is geographically isolated with a appear as such expressions of Che "truths" of Ladakhi social life and Indian
sparse population, a vas[ area and inhospitable terrain, which remains tand- polirics. The following is intended te show how such dissimulations are nec-
locked for nearly six months in a year. Consequently, the people of tile arca essary to live with Che irresolvable ineommensurabilities of communalism
have had a distincr identiry and special problems distinct from those of Che and exclusive official systems of classifieation, and rhe fluidity, multipliciry,
other arcas of Che Srate of )ammu and Kashmir."' The granting of a Hill and indeed "disorder" of social practices of identificarion.
Council was Che ourcome of in agitation Iaunched in Ig8g under Che lead- 1 first discuss rhe "tribalization" of Ladakh. With communalism out of rhe

366 STATES OF IMAGI NATION AUTONOMY FOR LADAKH 367

 question, officially, as a basis for recognition and empowerment, orher legal In a recen [ article, Michael Herzfeld notes that today "it becomes increas-
classificarions were explored, in particular Scheduled Tribes legislation. Sec- ingly difficult to sustain che idea that facts exist independendy of che social
ond, 1 briefly trace the hisrorieal production of communalist imaginings and world" (1998: 8o). Indeed, in che wake of che onslaught by hermeneuticists,
representations of Ladakh. The links of communalism with colonial and na- deconstructionists, and "other invading bordes" (Skinner 1985: 6), the "fae-
tionalist imaginings of religious "identiry" as central to Indian society and tieity of facts" (Herzfeld 1998) is no longer a heretic proposition. As Taussig
culture have been the subject of many excellent smdies, and the story in points out, Chis means that we live, all of us, in a space berween fact and
Ladakh follows a similar pattern.° Communalism, of course, is an officially fiction, between the real and the really made up (19936: xviii). The ways
illegitimate discourse, at odds with che Republic's secular, democratic prin- people negotiate chis irreducible gap, then, is che substance of culture; it is
cipies, and could not be used as a formal justification for affirmative action [hese acts of classification, che production of their logics, and che fliling of
for Ladakh after Independence. Ladakhi political leaders, however, perceived che spaces of che matrices thus devised, as well as the ways in which the
communalism to be che narure of che Indian political system and hence ulti- objects of classiflcation engage them. Anthropologists and other social scien-
mately adopted a communal strategy to demand autonomy. And it worked, tists have studied the classificatory practices of colonial and metropolitan
they say.
states, focusing on the transformations of space and identities through tech-
Both the formulation and implementation of the Scheduled Tribes nologies of governmentaliry such as statistics, mapping, and censos taking
scheme-publicly and officially-and the recognition and empowerment The contras[ berween lived practices of community-multiple, fluid, overlap-
of religious community-mostly "secretly"-are possible only through the ping-and the formal serialiry of communities instare and bureaucratic rep-
compliciry of bureaucrats, scientists, politicians, and ordinary people. Their resentations -singular, exclusive, fixed-has been che subject of many illu-
survival and effectiveness as principies of organization of Ladakhi identities minating studies. With respect to India, Chatterjee (1986, 1995) and Kaviraj
rests on che willingness of [hese different actors to ignore, in che case of tribal (1992) stress che qualitative change in conceptions of identity and belonging
identities, che impossibility of their existence in most aspects of everyday so- that occurs in che formation of che modern nation-state.°
cial life and, in the case of communal identities, their reality and effectivity. Governmentaliry is often treated primarily or exclusively as a technieal is-
However, also in che case of communalism, che efficacy of the strategy was sue, and one imposed by the state. The technologies of state agents and in-
premised in large part on che perceived reality of communalism. While com- stitutions threaten to become treated as hegemonic in a narrow sense, che
munalism was performed effecrively in Ladakh, it had to be policed and im- metaphor of che panopticon is literalized, and che objects of governmental
posed with sanctions on a reluctant population. There was, as I have argued technologies are confined to live within ineluctable "¡ron cages." This, of
elsewhere, no natural and lasting mobilization of the general Buddhistpopu- course, is far removed from Foucault's (1991) understanding of governmen-
lation in communalist opposition to Muslims in Ladakh (van Beek'996)7 taliry. O'Malley et al. Nave noted that resistance and contestation as consti-
In particular. 1 take up examples of che ways in which not only "ordinary" tutive dimensions are rarely given serious attention in che literature: "The
citizens, but also che state, political leaders, and bureaucrats ignore, violare, separation of contestation from rule," they argue, "leaves little space for theo-
or creatively interpret in their daily practices che various demands of order rizing che productive engagement berween them" (1997: 5r1). They call for morc
placed on them.
attention to the "messy implementation" of programs and recognition that
[hese are not "wrirten by one hand" but "multivocal, internally contested and
Interstices thus, in a sense, always in change and often internally contradictory" (5'z-
13). This multivocaliry and lack of internal coherence is particularly clearly
Might it rurn out, then, that not che basic truths, not ihe Being or che illustrated by the process of constructing official identities for the people of
ideologies of che cencer, but ihe fantasies of che margined concerning Ladakh, applying different coneeptions of race, casre, Cribe, and communiry.
che secret of ihe cencer are what is most politically importan[ to the state At che same time, speeific forms of governmentaliry, through che production
idea and hence state fetishism. (Taussig 19932: 23) of praetieal political rationalities in Weber's cense, contribute to che consti-

368 STATES OF IMAGINATION AUTONOMY FOR LADAK}{ 369

tucion of specihe forros of political practico: in Chis case, "tribalism" and co be classified precisely. The classihcation of Ludakhs population reflectad
00111111 n 11a lisa 1. °. changos in colonial and scientific discourse, as central organizing cacegories
such as cace were substituced with rcligion, casto, tribu, eom nlunicv, and h-
nally, once again, tribe (van Beck 1997b). llte collection ol inlirrmaciou, of
The Importance of Being Tribal
faces about che country was, of course, "scernly practica1," as che geographer
When the British govcrnment sold Kashmir in 1846 to che Dogra Maharaja Hugh Robert Mill pur it (quoted in Livingscoue 1992 213). And its adoption by
Golab Singh, Chis marked che emergence of a new kind of srace in the re- che Kashmir governmen t, urguably wích considerable British con:phcin. re-
gion. Ladakh, the cruts-Himalavan region benveen I limalaya and Karakoranl ncas a process of chango in the reason ol cate, inicially colonial. then ua-
broadly comprising che upper valleys of Shyok and Nubra, Zangskar, Sutil, tional developmental, coward che kind of modero governmemality described
Dras, and Indus, had been conquered by the Dogras several years earlier. The by Foucault.
British were keen to determine and demarcate che boundaries of not only As Gupta has argued, "To be a nacional subject in an 'underdeveloped
their own new possessions (e.g., Lahaul, Spiti, and Kinnaur), but also those country'-for example, to be a citizen of India-is to occupy an overdeter-
of the lands now under control of che maharajah. The British also developed mined subject position interpellated by discoutses of che nation and by dis-
an interest in long-distance trade through che region with Central Asia, and courses of development to which that nation is subjected" (1998: 34). The
in a eommereial treary of 1870 gained joint control over the mala trade route notion of backwardness, isolation, and cultural difference expressed in La-
through Ladakh, while ensuring tax exemption un certain goods to be carried dakhi representations, well before Independence, was already the product of
along these roads. The growth of trade and other colonial effeets, such as che a reimagining of Ladakh aceording co new, colonial/national perceptions of
travel of officials, sportsmen, and advenmrers, liad an impact on the lives of normalcy and development derived from a selectiva reading of European his-
some Ladakhis, especially those living along che roads who were subject to tory." A central element in the legitimizing scrategies of the late colonial and
obligatory labor requisitions (Grist 1994), but ofhcial interest in the region postcolonial Indian state has been the "upliftment" of "backward" sections
was largely limited to the imposition and collection of taxes. The explorations of the population. Groups deemed to be characterized by particular disadvan-
of the Boundary Commission did lead to a number of publications, such as tages, such as so-called Scheduled Castes and Scheduled Tribes (sclsT), were
Cunningham's encyclopedic Ladak: Physical, Statistical, and Historital (1854). singled out for different forms of affirmative action and protection. Once Chis
Another Englishman, Frederic Drew, produced a comprehensive survey of the prineiple was accepted, the identification of castes and theit enumeration be-
realm, published in 1875 as The jummoo and Kashmir Territories. Drew had joined carne an important object of governmental technologies. In accordance with
the services of the maharajah in 1862 as a geologist, and ended his career a British colonial perceptions of the nature of Indian sociery, standardized lists
decade later as wazir (governor) of Ladakh. Drew's book included some tables of tribes and castes were drawn up, each assigned specific characteristies, and
from the first censos of Kashmir published in 1873 and an attempt to cate- for each district the number of people belonging to each determined." Cer-

gorize the population of Kashmir and Ladakh aecording to roce. Drew ad- rain castes and tribes were designated as "scheduled" gtoups, eligible for
uritted race comed out racher like nationality in his proponed scheme, but he certain kinds of proteccions and benefits. Arcas with a significant share of
suggested that his book contains "ample illustration of the principies of Eth- tribal populations were "excluded," boch as a strategy of control and out of a
nological Science" for one "who would work it out in detail" (1976: 6). paternalistie desire to protect the "primitivo and backward" tribal populations
The spate of publications describing, mapping, and quantifying the faces (Metcalf 1994; Sonntag 1999). Aftet Independence, this colonial practice was
about the region reflects the emergence of a modera concern with population continued and expanded. The Constitution of India contains provisions for,

and economg with governmemality. With the growing desire to know the among ocer things, reservations in education and govcrnment service, and
country and as population, cartographic horror vacui accompanied the fear of a host of national and state-level programs exist to funnel additional re-
cmbarrassment of anibiguity of che censos takers: che population needed sources into the development of scisT. The links among data collection, bu-

370 STATES OF IMAGINATION AUTONOMY FOn LADAKH 371

 reaucratic operarion, and the nacional project are clearly illustrated by Home (Wartal 1948) and "dumb driven cattle" (Kashmir Raj Bodhi Maha Sabha
Minister C. Rajagopalachari's exhortation to the census takers in che first enu- [KRBMS] 1935). As Nehru wrote in 195t to Karan Singh, son of che last ma-
meration afer Independence:
haraja of Kashmir: "The real difficulty about Ladakh is its terrible economic
Yours is notan isolated local inquiry. You are one of about six hundred backwardness" (Singh 1982: 141). The forms of demands and their justifica-
thousand parriots, 211 of whom will be engaged on an identical task at tions have varied, but che content of che demands, che concrete measures
the same time. All of you are jointly responsable for enumerating all the sought, have been basically che same: more resources, faster development,
people. Collectively you will prepare a record of basic facas relating to and greater control over decision making.
the life and livelihood of all che citizens and families in our Republic. If With Ladakh's difference from other parís of che state generally recognized
this record is to be correer and complete, the parí that you contribute and well established in both academic and official texts, its backwardness and
should, in itself, be correct and complete. You should master the simple isolation widely accepted asan accurate designation of its condition, itwould
instructions you Nave received, and apply them uniformly and conscien- seem reasonable that Ladakhi political leaders expected that special provi-
tiously. These instructions are based on a common plan for the country sions would be made to promote education and development in che region. A
as a whole. Like a swarm of bees that build a beautiful hive according to number of measures were taken, including che creation of a special office for
che laws of geometry, each doing its pan in obedience to a mystic urge, Ladakh Affairs headed by Kushok Bakula Rinpoche, and severa! development
you should do your parí according to conscience and che sense of truth schemes were introduced. However, none of chis was deemed adequate by
inherenr in us all. (Census of India 1951: vi). Ladakh's leadership, and by che early rg6os a local opposition was forming
that accused Bakula Rinpoche and his associates of corruption and a general
Ladakh and its people, too, were mapped, measured, counted, and classi-
lack of initiative in promoting che interesas of Ladakh. The office for Ladakh
fied te enable the central planners and administrators to devise and imple-
Affairs was, according to one £ormer Ladakhi civil servant, little more than a
ment the correct measures to bring Ladakh finto che mainstream. As Nehru
"post office" where files passed rhrough bur no action was taken. Over che
himself had told che Ladakhjs in 1949, "4n Ladakh you are backward and
next years, local resentment grew, and in early 1969 a large-scale agitation
unless you ¡caro and train yourselves you cannot run the affairs of your coun-
was launched for local autonomy along che unes of che Northeast Frontier
try" (Arnríta Bazar Patrika, 8 )111Y 1949), and so the task had been entrusted, as
Administration that existed en che other extremity of che Himalayan range. It
in colonial days, te others more advanced. Ladakh would be guided along
was in rhe context of (his agitation that che demand for recognition as ST was
che path to progress and prosperiry by other sympathetic members of che
first voiced.
nation. Although the practice of reservations for sc/sr populations has come
It may seem odd that it took Ladakh's leaders rwenry years to raise Chis
under increasing eritieism, it continues to be a central element in central and
demand, which, as mentioned earlier, could provide such a range of benefits,
stare governmenr policy, and given che benefits that come with it, it was a
including possibly greater autonomy. According ro several people who were
long-standing goal of Ladakh's political leaders to have che population clas-
sified as ST." active at che time, Nehru had actually offered ST status for Ladakh te Bakula
Rinpoche shortly after Independence. There is no documentation or indepen-
Since 1931, when Ladalchi representatives first presented demands to an
dent corroboration of chis srory, but it is said that he declined it because he
official Commission of Inquiry, theie has been a steady flow of petitions,
thought it would be degrading to be ranked with savages and untouchables
memoranda, and reports in which local leaders sought to present Ladakh's
and in general was unaware of che potencial benefits flowing from sT status.
condition in such a way that it might be recognized as entitled ro special
In any case, che demand was rejected in 1969 and only occasionally raised in
treatment. The simation of Ladakh was routinely described as "extremely de-
che course of the next decade until in 1980 an All-Ladakh Action Committee
plorable" (Buddhists 1932: 127) and "most underdeveloped, poor and back-
for Dectaring Ladakh a Scheduled Tribe was formed. The goal of this move-
ward,"" and its people were dubbed "ignorant, illiterate, addicted to drinks"
ment was te achieve greater autonomy through Scheduled Tribe Arca status.

372 STATES OF IMAGINATION AUTONOMY FOR LADAKH 373

In the wakc of a hloodv clash benveen protesmrs and security forces in ig8i, 'jable 1 Scheduled Trlbes by Rellgion

die demmi id uas reponed to hace been aecepred be lile statc governmenG bar
no acrion was talen until severa) rcan lato, :md si declaration did not take Bol; l;rol:p.!.
Changpa Garra Maa Ferlgp.a
place nntil ig8q. Bal[i Beda Bono cte.

2,000 3,511 827 873

Buddhist 319 76.493
Muslim 10 ,272 11,205 24386
Identifi'ing'Iribes

Alrhough diere muy hace been many political and other reasons for the slow
process of recognition of Ladakhi claims to tribal status, there were a number Dard/Shin, Changpa, Garra, Mon, Purigpa. The rhree smallest tribes-Beda,

of issues to resolve that followed from the ways tribes are conceived in the sT Garra, and Mon, each with fewer than a thousand members-are groups of

legislation. 'lo qualify, Ladakh's population would need to meet a set of cri- "gypsies," blaclcsmiths, and musicians, which are regarded as more or less

teria of tcibaliry, which were derived from colonial and nationalist imaginings polluted groups of low status. Changpa are mostly semisedentary sheep,

of subeontinental tribals. Here the identification of tribes and their distinc- goat, and yak herders concentrated in Rupshu and Changthang arcas en the
border with Tibet. Balti is used to designare Shia Muslims sertled around Leh
tion from castes was fraughtwith difficulty: "The only safe criterion to distin-
in parts of Kargil and those living in the Batalik arca bordering en Baltistan;
guish the tribes from the castes is the former's territorial affiliation" (Majum-
by eontrast, Shia Muslims of Kargil and the Suru Valley are classified as Pur-
dar 1961: 390). Despite die absence of olear-cut criteria, there were a number
of more or less explicit assmnptions tied to en evolurionary vision. As a for- igpa, afrer one of the narres for the region. Brokpa and so en are the labels
for a number of different groups of people, some Muslims, some Buddhists,
merly independent kingdom with considerable literary and religious tradi-
tion and involvement in long-distance nade and diplomatic relations, Ladakh generally referred to as Dards in some of the literature.
Although the identification of [hese tribes is surprising given the lack of
hardly matched popular, off sial, and scientific imaginings of tribality. De-
substance of their tcibaliry in Ladakhi identifications, their fabrication is not
spite the considerable ereativity in identifying tribes, because of the absence
the main concern hete. Rather, the creatively selective identification and ap-
of a fixed set of criteria there were lintits to the possible. por one, itwas soon
evident that "Ladakhi" was an unsuitably inclusive category, as it includ'ed plication of criteria to classify the population is what is strikingly clear-

Shia and Sunni Muslims as well as Buddhists, sedentary farmers, traders, and despite the fact that both the report of the minicensus of 1986 and 1987
and the social scientists' report on the tribes remain classified inforniation.
businesspeople, as well as nomads and sections of the population who were
Whereas che formal creation en paper of [hese tribal identities was, aftr
treated as lower-status groups. Sume kind of subdivision of the population
all, an academic and bureaucratic matter, the implementation of The Con-
was necessary.
stitution (lammu and Kashmir) Scheduled Tribes Order, 1989 requires even
Finally, in the mid-tg8os the process of recognition gained pace and the
greater creativiry, as many of these idemities have little or no grounding in
registrar general, responsible for censos operations, was charged with the
social practice in Ladakh and were not recorded anywhere as formal identities
task of ascertaining how many people belonging to which tribes were wor-
thy of ST status. Ladakhi representatives drew up a list of proposed tribes prior to the Act.
With regard to identification of the population, the first obstacle was ro
to be recognized, which was presented to the government (see Table 1)."
ensure that the Indian scientists sent tu investigate the proposed lis[ of tribes
Social scientists were dispatched to Ladakh to investigare the conditions of
[hese tribes and found eight of diem worthy of recognition. One category, would find a reality matching those clainis. As the experience of census takers

Shamma, was rejected on the grounds that these were basically identical with prior to Independence liad shown, when caste, class, tribe, and communiry

the Boq Boto tribe, which was said to constitute the bulk of the population of were still enumerated, self-identification led to a bcwildering proliferation of

Leh district.- The other tribes recognized were Balti, Beda, Brokpa/Drokpaj identities. The igir eensus, which allowed self-identification, produced lit-

374 STATES OF IMAGINATION AUTONOMY FOR LADAKH 375

 erally thousands of caste names and more than twenry thousand subcaste communalism, [heir success in cornering a substancial share of che profitable
names for che frontier region as a whole (inciuding Leh, Kargil, and present- modern economy, as well as [heir identification, righdy or wrongly, with con-
day Pakistaní-controlled Baltistan)." Ladakhi activists were also only roo version and the Sunni population of che Kashmir Valley, made them a primary
aware of the fact that people were fiable to identify themselves inappropriately target of Buddhist activist ¡re. Thus, although che "technical" reason used to
when asked what they were. Although 1 have heard different stories aboutthe exclude them was, apparently, that they were nota stable community but one
process, it is clear thar both people to be surveyed and che interpreters on that anyone who was of mixed descent would automatically belong te, it is
whom che scientists were dependent were insrrucred en appropriate answers clear rhat other political considerations played a parí in [heir exclusion.
and findings. Villages and households included in che supposedly random Although Argons were thus formally excluded ([heir case continues to be
samples were also cautioned not to display roo much wealth and "moder- appealed at various central and state levels), che operationalization of che ST
nity." Two groups in particular feared to be excluded in view of che manage- order proved to require creativity of a different order and to invite yet more
ment of che process of identification and recognition: che small communiry bending of rules and regulations. Because tribal membership was not regis-
of Christians at Leh, and che larger, powerful group known as Argons. tered previously, che documentation of belonging was an obvious challenge.
The "chairman" of che Christian community, S. S. Hischey, wrote to che This problem was resolved, en paper, by taking che records of che Land Settle-
depury commissioner of Leh to express his concerns. His primary complaint ment of rgo8 as a basis. Hence, if one could show ancestry in che rgo8 rec-
was che organization of che survey "performa" (sic), which restricted the pos- ords in, say, Karnak, one would automatically be Changpa, whereas if one's
sible choices for Buddhists and Muslims. Certain categories, in otherwords, house was recorded in, say, Batalik and one were a Shia Muslim, one would
were apparently reserved for either one. Mon, Beda, and Garra were wrongly be Balti. In principie, then, things were olear. In pracrice, however, Chis poned
limited to Buddhists: "Such possibilities can also be under the Muslims and several problems arising from people's shifting identificarions and locations.
Chrisrians." He concluded his letter with a pica for an entirely new caregory A striking example is che case of a Muslim who converted to Buddhism upon
te be added: "My submission to you is to kindly broaden che scope of admis- marriage in che lgzos. This person's ancestry were Argons, as was he himself
sibility of Ladakhis finto che Schedule Tribe and not relígion as che latrer might in lgo8. Therefore, his offspring, although Buddhist, were technically ineli-
produce che fruits of bitterness in the soil of peace and brotherhood."'a gible for recognition as ST. To resolve chis dilemma, che presiden[ of che La-
In rhe end, Christians were encouraged to identify themselves as BotlBoto, dakh Buddhist Association issued a statement ro [he effect that X, opon his
che most "generic" caregory available, and so there is no Christian Cribe today. marriage and conversion, had "fully adopted che way of lile, customs and
The late of che Argons was less easily settled. When che Cribes were declared traditions of che BotlBoto Cribe. Thereafter, al¡ che children born out of che
in 1989, 89.71 percent of che population was said to have been included. Hin- marriage and che grandchildren have followed che BotlBoto way of life and
dos and Sikhs were among che excluded, but also Argons, said by theirleaders have been following all che Customs and Traditions of the Cribe in che exact
no constimte up tu 7 percent of che population of Leh. The oficial, public manner that any other member of che Cribe would do. Consequent upon [heir
reason for [heir exclusion was, as Leh member of che Legislative Assembly complete assimilation finto che BotlBoto Cribe, all che descendants of X and Y
(M LA) Tsering Samphel wrote in a letter to che editor published in theJammu- have been accepted by the BotlBoto tribe as members of che said Cribe and
based newspaper The Daily Excelsíor, "technical." The term Argon is used to have been and are being treated as such members of che tribe."'°
designare families of mixed Ladakhi-Kashmiri or Ladakhi-Yarkandi ancestry. Hence, for all practical purposes, che Ladakh Buddhist Associarion (LEA)-
Leh, as an importan[ entrepót on che trace mutes connecting Central Asia a religious, social organization, according to its statutes-certified che au-
with Kashmir and beyond, attracted many foreign setrlers, often members of thenticiry of che tribal identiry of [hese applicants by referring to a set of
the large trading families of Central Asia." The trace was ended by che clos- apparently well-defined "customs and traditions" of che equally substancial
ing of borders in che rggos and rggos, and Argons shifted te other forros of social entiry of the BotlBoto tribe. Again, although che mode of existence of
business, including contracting. As we shall see below in che discussion of che BotlBoto tribe would be a mpic worth exploring, the point here is te

376 STATES OF IMAGINATION AUTONOMY FOR LADAKH 377

illustrate Che ereativigv wirh which Che realiry of Che Cribe, its customs, and the Territory status - implying direcr rule doro New Delhi-or eur r ighr state-
Itood as ultimare gouls. Au[onomy , in wliarever toril. proved to be an elusiva
notion of membership are dissimulated, and how Che case is then cemented
goal until Che compromise of die liill Developutent Council was reached in
bv refrence ro odier cases ol creative interpretation. lile impossibility of
Botl Boro rribaliry, or at lcast of unamblguous notions of membership, cus- the wakc of Che tg8q agitation.

tom and tradition. is overconte by creative action.

Creativire of a diffcrenr sorr is albo said ro characterize lile issue of Si lile Secret oi Conununalism
cc rti Reates. which are necessarv ao gain access lo che benefirs of si status.
Ir would he a rragedy if... the people of Ladakh remaiu nnhuppc ard di,-
such as lujos agaiust advantagcous conditions and quotas fm jobs and cdu-
contented, and thus in easy prey lo all sorrs oí exploitation, borh commu-
carian. As explained, ir is Che Tchsildar's (distrier revenue officer) Office that
nalist and communist- (Karan Singh, letter ao Nehru, 1951; citad in Singh
issues rhese certificares, and given the benefits of ST status, it is not surpris-
ing that there are numerous stories about corruption. People complain that it 1982: 141).
The All-Ladakh Action Committee in the 198os reflects local tradirions of
is almost impossible for villagers to get rhe certificare, as they do not know
noncommunal activism in the region. This all-Ladakh ethos is exemplified
the systcm and are confronted with demands for bribes. Sonre Buddhistactiv-
also by the fact that Kargil's majoriry Shia voters have conmonly supported
ists are now convinced that the Argon issue is no longer acate, because most
the Buddhist candidate from Leh in elections to nacional parliament, while
Argons llave obtained ST certificares anyway-presumably as Bot/Boto Cribe
sending a Muslim to represent Che district in rhe J&K Legislativa Assembly."
members, who can be either Muslim or Buddhist. With larger percentages of
Parallel to this broad identification wirh Ladakhi interesas, there has also been
Che population of I&K now in possession of the coveted I.D., for example, as
a tradition of political activism along communal fines, pitching Buddhist La-
Gujjar have now also been recognized as ST, Che benefits in gaining access to
dakh as the oppressed victim of Muslim Kashmiri discriminarion. Commu-
quotas are in any case now limited.
nalism is widely believed to be the true character of Indian politics in general.
Although ST was granted in 1989 and Ladakh's two districts nave been
Always controversia), and certainly after Independence publicly illegitimate,
eligible for a range of affirmative action programs, their impact has been
eommunalism was commonly regarded as the nrotivating force of I&K poliey
limited.To be sure, individual opportunities such as cheap loans have en-
couraged hundreds of people ro take out loans to buy taxis or to open tele- toward Ladakh.
Whereas ST status was reluctandy and belatedly pul on Che agenda in La-
communications businesses, many of which are likely to go bankrupt in the
dakh, communalist diagnoses of Che plight of the region were introduced as
coming years. Yet severa) people who had been in the forefront of the agita-
early as the 193os around the time of the Glancy Commissiou of Inquiry luto
tion for si, status argue that roo little has been done to attract largor amounts
Grievances and Complaints. This commission was instituted by the govern-
of central resources through, for example, the Tribal Subplans and other spe-
ment in Che wake of severe unrest in the Kashmir Valley targeting rhe Hindu
eial funds. Hence, dependence on state programs and development initiatives
maharaja. Glancy's brief was specifically to solicit representations en the ha-
remains high, and the state government's performance continues to be re-
sis of community, and hence Che coniposition of his commission reflected the
garded as insufficient. Among factors contributing to the relative lack of ac-
officially recognized communal arithmetic of Che state. Buddhists, however,
tion en the part of the state government may have been Ladakh's very mar-
were not invited to sit en Che committee, and to ensure thaa their voice would
ginality in economic, geographic, and electoral terms, but for many Ladakhi
be heard, a group of Kashmiri Pandits (Hindus), some of whom had con-
leaders the explanation was simple: communalism. Ladakh's Buddhist and
verted ro Buddhism, founded the Kashmir-Raj Bodhi Maha Sabha (KKEMS)
Shia Muslim populations were discriminated against by Che Sunni Muslim-
and asked m be allowed to make a representation on behalf of all Buddhists
dominated Kashmir government. This complaint was voiced time and again
in the state, almost all of whom, of course, were Ladakhis. This was granted
in the course of the past sevenry years, and Ladakh's political leaders consis-
and on 21 November 1931, theee Kashmiris and a young Ladakhi student,
tently strove for some forro of auronomy froni Kashmir, with either Union

AUTONOMY FOK LADAKH 379

378 STATES OF IMAGINATION
 Sonam Norbu, presented the Buddhist community's submission to the com-
resentation, image management, and arm twisting, as one agitation leader
mission, emphasizing the backwardness of the region aud demanding suit-
able measures. called it, were necessary. As in the case of the ST identification, a series of
contradictions and inconsistencies needed to be "forgotten," while the pub-
Among the initiatives proposed by the KRBMS was a ban en polyandry, the
licly illicit character of communalism posed important restrictions on its use
practice of several brothers marrying the same wife. Polyandry, in the view of
as a political strategy.
the Pandits, was a major factor in the low growth rate of the Buddhist popu-
A first obstacle to be overcome was the absence of a cohesive Buddhist
lation. Its practice would lead to the demise of chis last remaining stronghold
community in Ladakh. To be sure, a distinction is routinely made betwcen
of Buddhism in the state, as Muslims not only were deemed to be reproduc-
Buddhists (referred to as nang.pa or "insiders") and non-Buddhists (phyí.pa
ing at a higher rate, but also to be actively converting Buddhists, for example,
or "outsiders"), but there are a host of socially, economically, politically, and
through marriage of "surplus" Buddhist women. The focos en polyandry,
geographically significant differentiations that militate against any easy all-
although certainly also reflecting a general disapproval of such immoral prac-
Buddhist identification beyond this immediate, coarse dichotomization. La-
tices, powerfully illustrates the "general scramble for percentages that char-
dakhi society, like most Tibetan societies, is historically characterized by a
acterized the politics of the State at that time" (Ganhar and Ganhar 1996:
strong hierarchical division of the population. There continue to be strong
216). In 1934. as a consequence of the Glancy Commission's work, a legisla-
antagonisms among different sections of the nobility, who are also com-
tive assembly, the Praja Sabha, was constimted with nominated members for
monly a(Tiliated with monastic establishments of different Tibetan Buddhist
Ladakh, as elecrions were deemed impossible in view of the geographic and
sects. Although doctrinal differences benveen seets and arguably sectarian-
edueational situation of the region.22 Around this time, a local Buddhist Edu-
ism in general are generally not as important in Tibetan societies as they are
cation Society was founded, succeeded a few years later by the Young Men's
in Christianity, the intersection of economic and political power with mo-
Buddhist Association, which later changed its narre to che Ladakh Buddhist
nastic establishments-for example, through landholdings, taxation, and
Association. Both were established very much at the encouragement of local
religious patronage-sustains rivalries among households. The politicalsali-
Pandits and KRBMS members." Hence, by the late 193os Leh's developinental
ence of the clergy is also evident in the central role of religious authorities
and political landscape was already thoroughly communalized.
such as Kushok Bakula Rinpoche and his longstanding rivals Khanpo Rin-
While communalist motives were routinely attributed to the state govern-
poche and Togdan Rinpoche.'5
ment, local and regional relations between Muslims and Buddhists were gen-
Many villages in Leh District have no or very few Muslim inhabitants, but
erally cordial. During Partition, although Pakistani "raiders" carne to within
a few miles of Leh, tensions between Muslims and Buddhists in Central La- several thousand Shia Muslims are settled along the Indus River near Leh,
and there is a sizable Sunni community in the town itself As intermarriage,
dakh remained relatively low.24 In the subsequent period, one of the argu-
especially with Sunnis, had not been uncommon, many people had kinship
ments commonly used by Ladakhi political leaders in justifying their claims
ties across "communities," and social and economic ties between households
for autonomy or other kinds of special treatment was precisely the popula-
similarly crosscut religious divisions. In addition, there were rather banal re-
don's patriotism and noncommunalist ethos (van Beek r998). It is only in
alities to consider, such as that al] butchers and bakers were local Muslims,
tg69 and especially in 1g8g that sustained communalist strategies were used
and the vegetable market was an exclusively Kashmiri affair. There were, in
in agitations for regional autonomy. The 1989 agitation for Union Territory
other words, social as well as economic relationships that made mobilization
status, which culminated in the granting of a Hill Council in 1995, was led by
on the basis of religious difference difficult.
the Ladakh Buddhist Association (LBA) and involved a social boycott of the
There was also much difference in perception between village farmers and
Muslim community for three years. The agitation was presented by the LBA
urban professionals of che state government's policies. Whereas [hose with
and the media, both domestic and international, as a natural uprising of the
modern education or business interests would lament the lacé of government
discriminated Buddhist community. But for this image to stick, creative rep-
jobs, infrastructural development, or differential treatment, villagers com-

380 STATES OF IMAGINATION

AUTONOMY FOR LADAKH 381
 soon this pretense was dropped and a social boyeott was imposed Oil all Mus-
monly applauded the government as a provider of services, roads, canals,
lims in Ladakh. Officially, this move was said ro have beca necessary be-
ferti lizers, cheap rations, and other developmenral boous. In the villages,
cause of che reluctante or refusal of Shias to join the I.EA-led movemenr. but
diere was no feeling of discrimination, and in che absence of Muslims in
diere are indications rhar che communal strategy targering :ill Muslims ma'
mane villages, che issue of communalism simple did not sor up much cmo-
have been planned from tila start.- Within a shortrime, Buddhisr bakers and
rion. Hence, che LEA leaders firsr built up a network of yourh organizations
butchers set up shops in Leh, and vegetable Sellers appcared along che newlt'
and cultural 1ssociarions iu ncC villages m prepare nce gromtd for che agit -
opened road from Mamali in Himachal Pradesh. Aftcr diree Buddhists were
riou, and in igSS-rgbq Ieadership and Vouth WVing activista toured the vil-
shot during a demonstration in I.ch's lime bezaar on _¡ Augusr tgSg, au
lages ro "educare che people," as one acrivist per ir, about the faca rhat Bud-
ulrimatum was -¡ven to all Kashmiris to leave the disrrict Bforts were made
dhists were, in fact, victims of discrimination. Despite rhese efforts, diere was
to mark Buddhisr and Muslim spaces, structures, speech, and bodies unam -
a real danger that a communal agitation would not be sustainable over time.
biguously. Painters went to non-Muslim-owned shops to write the narres in
Popular supporr could not be taken for granted, and a strong unified leader-
Tibetan script, helping shoppers to know where te go and which establish-
ship was crucial , as che rg6g example had shown when rhe coalition of Balcula
ments to avoid. Radical youths started sporting earrings to signal their
and his opponents fell apart within months.
Buddhismess, while many Muslims stopped using che local greeting, "Julley,"
Hence, two broad tasks lay before the LEA: first , te ensure and maintain
and instead offered a "Salaam." And en religious holidays loudspeakers en
popular compliance wirh che communal agitation , or at leas[ ro ensure that
the maro temple in Lela rang with che rape-recorded recitations of monks.'e
such broad support appeared to exist; second , te ensure that outside Ladakh
At che swearing-in ceremony of che LAHDC in 1995, Governor Krishna Rao
public and political perceprions of che agitation were rhatthe movement was
echoed Indian news repores and many academícs' assessment of che agita-
noncommunal and democratic . Although communalism was seen as neces-
tion: "I am happy to note that your seruggle has been peaeeful and democratic
sary in getting che attention and support of the central government, LEA
means were used rowards this end."29 John Crook noted that "incidenes of
leaders were well aware that communalism could not be officially rewarded
violence have been few, and have been denounced by che leadership" (1981:
with recognition of their demands . Or, to push the point a little further: al-
238). This pereeption bears witness to che success with which che LEA man-
though communalism was not much of a reality in daily Efe of most people
aged che public image of che agitation, rather [han che practice of che agita-
in Ladakh , it must be represented as such, and although communalism
tion. The LEA imposed a strict code of behavior en all Buddhists, with sane-
needed to be representad as central te Ladakh' s plight, it must be represented
tions enforced by young vigilantes ranging froni a fine ro beatings and social
as involuntary, passive, imposed . Ladakh's Buddhists needed to argue like
boyeott.1' Signs of public support, such as che groups of volunteers-mostly
communalists and behave like communalists to fit che dominan [ discourse of
women-who demonstrated daily in Leh Bazaar during 1989-1990 or who
Indian politics as they undersrood it, while simultaneously deploring com-
went on relay hunger serikes, or che thousands of people who showed up ar
munalism and insisting en che "age - old traditions of communal harmony."
mass rallies, were also realized through "arm twisting" such as che demand
that each (Buddhisr) household send a representative for sutil occasions. Non-
Identifying the Buddhisr Communiry compliance would incur a fine of several hundred rupees.
Having ehus ensured some stabiliry at honre, tila LEA sought to ensure thar
Soon afier dile agitation for Union Territory status was launched in June 1889
che correct view of Ladakh's predicament and aspirations were reported in
in che wake of clashes berween Buddhisr and Muslim youths, tila LEA carne
India. Hence, they carefully groomed certain journalists froin nacional news-
out as che driving force behind che agitation. Initially, cace liad been taken to
papers, sometimes fiying diem in tu Ladakh at their expense, ensuring they
distinguish between (Ladakhi) Shia Muslims en the one hand, and "Pakistani
met with che desired spokespersons, witnesses, and politicians, and keeping
agente" (Kashmiris) and their supporters (Sunni Muslims, Argons) en che
Ihem away from others. Consequently, che LEA rarely received negative press.
other. A publication documenting the grievances and demands claimed to be
not even during che explicidy communal piase of che agitation between 1988
che manifest of a Ladakh Pcople's Movement for Union Territory Status, but

AUTONOMY FOR LADAKH 383

382 STATES OF IMAGI NATION
 and 1gg2. To add pressure on the central government and perhaps to exploit the series of specific tribal identities in Ladakh has had che paradoxical effect
the escalating secessionist violence in the Kashmir Valley, che LEA also sought
that the Ladakhis no longer exist. The recognition of the eight Scheduled
support from Hindu nationalist organizations such as the Vishwa Hindu Par- Tribes has meant that almost all people in Ladakh now possess a picture
ishad (vHP) and Bharatiya Janata Party (BJP). Contacts with these organiza-
I.D. staring unambiguously their identiry: Bot/Boto, Balti, Changpa, and so
tions dated back to at least 1988, when the LBA had received support from the on. At the same time, the success of che agitation for regional autonomy has
VHP in its campaign against the alleged abduction of Buddhist children by been at the cost of a thorough communalization of Ladakhi sociery, dividing
Christian missionaries in Srinagar. Overt political support from the EJP for
the population mato Buddhists and Muslims. While the LBA leadership con-
the Union Territory demand was extended by Atal Behari Vajpayee at the BJP sciously chose a communal strategy in 1989 tu get the artention of the govern-
congress at jammu in September r99o. However, in the wake of the demoli-
ment (because communalism was seen as the secret, because illegitimate, but
tion of the Babri Mosque in Ayodhya in December 1gg2 and the return to real character of the Indian political system), a younger generation has come
power of the Congress, the LEA leadership no longer maintained these con- of age who are unfamiliar with che historical all-Ladakh antecedents of the
tacts intensively."
struggle for Ladakh. Communalism for them is a way of life in accordance
Neither subsequent opposition-led governments, such as those of Y P. with reality. Yet, as 1 have sought to illustrate here, it would be a mistake to
Singh, nor the Congress regime of Narasimha Rao would want to be seen to think that local identifieations-che ways activists, politicians, bureaucrats,
reward communalist campaigns, and so the LEA was told that it should lift lawyers, and common people live their lives-are reducible m either "tribal"
the boycott of the Muslim community and that negotiations over the future or "communal" identities. Nor can these two discourses of identiry be differ-
status of Ladakh would need to be conducted also with Muslim representa- entiated by their degree of spuriousness, as both are products of imaginings
tives. Although che boycott remained in force until late 1992, che political with specific historical, scientific, and legal-bureaucratic antecedents.
leaderships of LEA and Muslim organizations" gradually began to cooperare In a recent article, Arjun Appadurai (x998) examines some of the con-
more closely and eventually a Coordination Committee was formed that also sequences of "new frameworks of identity, entitlement, and spatial sover-
included a representative of che small Christian community in Leh. Itwas Chis eignry." Echoing Anderson (1992), he argues that globalization has produced
joint Ladakhi representation that negotiated the final compromise solution a new order of uncertainty in life, which is expressed, among other ways, in
of an Autonomous District Council, giving everyone the possibility to ignore erhnic violence. Joining a host of anthropologists who draw on Mary Doug-
or at least downplay any communalist element in che process or outcome of las's discussions of "matrer out of place" to analyze ethnic violence, he uses
the agitation. Thus, the president, the prime minister, the governor, and the Malkki's work en the genoeide in Rwanda, in particular che problem of am-
celebrants gathered en Leh's polo ground could all congratulate themselves biguity of ethnic bodies, thar cannot in the real world be consisten[ with the
and one another en the democratic and peaceful way in which India's plural- cosmologies they are meant to encode: "It is this reversal of Douglas' cos-
ist, secular democracy had once again prevailed, allowing Ladakh's Hill Coun- mologic that might best explain macabre patterns ofviolence directed against
cilors ro "bring out che full potencial of this unique part of che J&K State and the body of che ethnic other; to eliminare the flux introduced by somaticvaria-
the Union of India.""
tion, by mixture and intermarriage; and to evict the possibility of further so-
matic change or slippage" (Appadurai 1998: 232): ethnic violence, in short,
Conclusion: The Impossibility of Being a Ladakhi as a means of eliminating uncertainry.
There are, in my opinion, two issues here. First, as Appadurai allows in a
There is a profound irony in the fact that each discursive straregy used to state
footnote, "Of eourse, nor all forros of abstraetion in social lile conduce te
Ladakh's "legitimate" case must resr on che exclusivist categorizations of bu-
violence, nor have such potentially violen[ forms of abstraetion as rhe map,
reaucratic and scientific formalism: ethnicization and communalization as
che censos, and models of economic development always led to coercion or
the priee for entry inro the liberal democratic system, to join the world of
eonflicr" (1998: 240). Mindful of che facticity of al¡ facts, elassification and
nations en the march tu freedom and prosperity. The official recognition of cosmologic are pervasive aspects ofhuman existence, alrhough the modes of

384 STATES OF IMAGINATION

AUTONOMY FOR LADAKH 385
 genc,ally refnined t ron'. identihing dm >ourcc it specific ros rials and irh rn.uion. Re-
nccir j ustification may difter (c.g., divino order, rationality, nature). Second, s hrch in ,9g;-n)95 was made possihle bc , luu rngs Raudolp 1 Pcaru xholar .hrad Prurn
as 1 lcrzfcld reminds us, "Cultural identiry is thc material of national rhetoric, tilo United Sanes bisonte of Pe, c, and ,epplema ned se-pP I as provid ci i c thc I'uaCc

social variation tila[ ot evcrvday eyperiencc" (1992: ¡o8). We all live In Chis imdic^ Pmgra n,. Inwr ucionel I ,I leal Ilcc.umi I g; '11 _ s -cun u Lo

-'silly if nos desperate place benveen lile real and tlic really made-up" ("Faussig ncll L mersit}- ke.ponsi hilln' tú, i, has is presc I c hoce ii, , Icor
Addie by Gencnl K. r Kriishua Rao ti's ti, Lh cd I1 Lcccll , Goc rn i I;n
¡993b: xviii). lhere is, in othcr words, little reason to treat complexity, disor-
nid h,:hntir Stalt r 1adalch Aumnonono f1(110.elupnie,tC nu c1. ^i s ir n.,c.
der and insecutin, as experiences or conditions rhat are particu larly new, er
u)1)5 t p icript-
ro assumc tila¡ human bcings are ill-cquippcd to handle choro. As is well- _. U0, nll, lile sise ,t th egiou 1 y km su chis ügu:c idcderrorrilirc,cl)cCir_.-
known, corruption, turning a blind oye, aire] other forros of ereative inrerpre- under Pakistaui sud Chis,,, adnsinisrration_ 1 vu populan n '^gunz. _i i hcnec prca^.
tation of rules are integral paros of tilo ftmctioning of any bureaucratic order." relufive abarca of che differeut religious communities, are unwailablc, as no regular censos

Chatterjee's conclusion thar the modern stare °cannor recognize within jis has buen held in ¡ctc since i981.
g. The Gazme of India, extrapart II, section 1, 9 May 1995, p. i9.
jurisdiction any forro of community except rhe single, determinare, demo-
4. Kargil Disrrict deelined che offer of a Hill Couneil sor sise rime being. alrhough rhe Act
graphically enumerable fono of the nation" (Chatterjee 1993: 238) is rather
provides for it, as people here feel rhar nceir greater proximity ro and dependence cal rhe
roo pessimistic, I believe, for two reasons. First, although nationalism fre- Vallen malees it less politic m risk anragonizing me Kaehmiris. Regular shelling of Kargil
quently tends toward ethnic/culturalist exclusivism, chis is, as Anderson tocan and tire border arca in recent years and che ineunion tilas took place from aeress use
(1992) reminds us, nos an inevitable movement." Inclusivist conceptions of fine of control in 1999 are unders tood by pcople in Kargil lo be rerribution for their lacé of
nations, even fuzzy ones are possible (Kaviraj 1992). Second, as 1 have sought support for sise militant secessionists in Kashmir proper.
5. In South Asia, rhe terne "communalism" refers co sise evils of religious community par-
to demonstrare rhrough experiences from Ladakh, it is important lo bear in
tisanahip.
mind that bureaucrats and politicians also must dissimulate. The purity de-
6. See, e.g., Chandra (1984); Hansen (1999); laffrelot (1996); Kaviraj (i997a); Pandey (1990):
manded by legal and other formal systems o£ classification of populations
Tambiah (i996); van der Veer (1994); Vanaik (1997). 1
cannor exist on rhe ground. To be sure, as Appadurai and others have argued, 7. 1 realize chal chis is a tridry proposition. 1 do nos mean to suggest tila¡ only politicians chiné
that impossibiliry may lead lo ethnic cleansing and genocide, bus it may also, communalism is rhe name of the game, but 1 question die universaliry of the cominunalisr
as is rather more eommonly rhe case, lead to a constant violation, creative logic implied by its recognition as a potent force in Indian political practice. My experience

interpretation, and experiential moderation of the hard, sharp roles of imag- in Ladakh and elsewhere suggests that "identiry fetishism" is especially a vise of sciencists
and politicians , of the national(isq elites in general , pcrhaps.
ined orders. It may no longer be possible lo be a Ladakhi, bus ¡here is no
8. The literature is vaso, bus important examples are Abraham (1996); Anderson (1991, 1998);
reason ro think that therefore Ladakhis can no longer exist. Indeed, rhe ab-
Appadurai (1993); Cohn (i9gi); Hacking (1990); Iones (1981); Herzfeld (1992); Mitchell
sence of a formal-legal Ladakhi identiry and rhe varied practices of identifi- (i988); Pedersen (1986); Visvanathan (1988). Per a review of
rhe anrhropological literature

cation that temper the officially sanctioned as well as unofficially praetieed with reference to colonial staves, see Pele (19971.

exclusions and inclusions may provide rhe space for Cae kind of social en- 9. In Ladakh, it is possible to trace the lame "ethnicizing" or "communalizing." Thanks to che
eneyclopedic urges of as adminisrrators, students, and missionaries, rhis procesa can be
gagement that only roo often js now located in romantic longings for primor-
naced in great detall. See, e.g., van Beek (1996, 1997a).
dial, original, authentie, cultural identiry.
io. 1 cake chis point from Hindcss (1997).
u. Por genealogies of development, see, e.g., Escobar (j995); Esteva (i992); McMichael (1996

i2. An excellent historical discussion is offered by Metcalf (1994). Anthropology and sociology
Notes were not only insrrmnental in Che designing and fdling of me matrices, but aleo in the for-
nndation and iniplemcnration of policies targeting [hese populations. Discussing academic
'Ibis essay dt-aws oil research conducted in Lada kh since 1985. 1 particularly want to thanE
anthtopology in India, D. N. Majumdar noted in 1944 thhat °rhe craining of penonuel for
current and prcvious leaders asid activisrs of Buddhisr and Muslim organizations, as well as
manning the welfare services sud feeding vise research instimtes has been given priodty, fis
niany other ind ividuals in Iadakh for their eourage and opeuness in shari ng docu menos and obvious reasons" (1961: en).'lo he sure, the racial interpretation afeaste and tribe exempli-
personal reflections. In view of che sensiriviry of sonie of rhe matices discussed hete, I Nave

AU'IONOMY FOR LADAKII 387

386 STATES OP IMAGI NATION
 red by Risley 's influendal work ( e.g., 19(59 ) lost influence after independence , bur he (1 ices,
was raised te help defend rhe region and Muslims joined, albeit alter ¡nidal hesitation or
partlcularly its essentializing of such "identities ," continue to have considerable inlluence
exclusion , depending on whose account one follows . There is as yet no thorough historical
en Indian ethnographies , also those of Ladakh ( e.g., Mari, 1986 ; and Bhasin , 1992).
account of the Partition period in Ladakh . Kaul and Kanl (1992) offer an overview, albeit
13. In particular, rhe Mandal Commission report in the late rg8os triggered massive violence,
from a partisan perspective.
including self-immolations by uppenmiddle - class youth on tire streets of New Delhi. The
25. These three continue to play a prominent role en che political stage in Ladakh. In recent
controversy over reservations was one of che elements that led to rhe downfall of rhe ]anta
years Bakula Rinpoche has been serving as Indian Ambassador te Mongolia : presentí y,
Dal-led coalition government of V. P. Singh. See , e.g., Hansen (
1999: 140-45). Thikse Rinpoche is a member of rhe Rajya Sabha and 'logdan Rinpoche isa Minister ofSrate
14. The quote is from a speech by Kushok Bakula Rinpoche , rhe represenrative of Ladakh's Bud-
for Ladakh Affairs in rhe 15K government oE Farooq Abdullah. The impossibility of entirely
dhists in che Kashmir Legislativa Assembly during che budget session en 12 May 1952. The
reducing rivalries to secearian differences is exenrplified by che fact that Thikse Rinpoche has
late Sonara Stobdan Lachumir provided me with a manuscript version of rhe text, and laved
herir an arden [ opponent ofóakula Rinpoche lince che early rgóos , although borla belong to
.Mir Qasim helped to translate Chis from Urdu.
che lame Tibetan Buddhist school , rhe Gelugpa (dge.lugs.pa).
is. These data from rhe classified report en che Special Censos conducted in 1986 and 1987 were
26. There is no "hand" evidente of chis in die form of docmnents , bur personal acmunts of
published by Ladakhi Meinber of rhe Legislative Assembly of JsK Tsering Samphel in a letter
Buddhist activists show thar ostracization ofall Muslims was intended from che aran, par
to rhe editor ofThe Daily Fxrelsior newspaper, i6 March ig91.
, ricularly by che urban Buddhist youth. During secret meetings in rhe winter of 1988-1989.
16. Sham ( gsham ) is che name of rhe region of rhe lower Indus Valley roughly between rhe towns
for example, severa) youths swore oaths rever to have social contad with Muslims again.
of Basgo and Khalatse.
Toda,, [hese oaths are che subject of some amusement am ng activista who did not go so
17. The ntess was "resolved " by drafting a limired list of sixry-odd categories ,
which was further fax and sea some of their friends vainly struggling to keep theni.
reduced to ahout a doren in che next few decades . By tire time of Independence , che enurner-
27. As a convoy. including buces carrying tourists , leaving che region was atmcked mar Khaltse,
arion was abandoned eltogether , except for se and sT- which did not exist in Ladakh-and
Che news of che Buddhist up ri si ng even nade che internacional presa.
religion. Por religion, census takers were instrucred nor to ask " What is your religion?" bur
28. In 1995 Chis practice escalated ¡ rito a competirion benveen rlic muezzin of the Leh Sunni
"Are you a Hindu , Muslim, Sikh , Jain, Buddhist, or Christian ?"
These, tuo, are good ex- mosque and che mped ntonks of Che Chokhang, jusr acrosa che street from ore inuther.
amples of rhe creativiry with which impossible purities are put roto practice and nade real.
Whenever rhe Muslims were called to prayer , che sound of rhe monks rang out as evell. A
Fue a demiled discussion of tire census classification procesa in Ladakh before Independence,
Buddhist activist joked thar Buddhists also were now called tu prayer five tintes a day. This
sea van Beek ( 1997a).
still continued ¡ n che summer of 1999, despite much popular resentmenr.
18. Both quotes from a letter Croas S. S. Hischey m rhe D.C., Leh , datad 20 Atigusr 1986; empha-
29. Address by General K. V. Krishna Rao to LAHDC, 3 September 1995, p. 7.
sis in me original.
30. Bach offense earried a spee ific penalty, inereasing with reptar offenders. Of course, such
19. Por a diseussion of aspects of [hese rrading nervorks , see che autobiographical account of
nteasures were kepí secret, asad even a document for interna ) use listing these sauctions
Abdul Wahid Radhu (1981 ). only recently published in English with an excellent incroduction
merely stipulares " if transaers with x" as an offense, carefully avoiding naming che con,-
by José Cabezón (Henry 1997). On trade in Ladakh in general, see , e.g., Rizvi
( 1994, 1997, nunity. There ¡ s no such shyness about penaltles ; "transaction with x" ¡s punishable with a
' 999); Warikoo ( 1985. 1994 1995 ); Criar ( 1985).
fine up to Rs 500, and a "beating " is added in case of repetition . Other fornts of punishmenc
zo. The sratement was issued in June 1993 . Atrached to ¡t is an atfdavit prepared by a Iawyer,
include rhhagc ( prostradons), beating, and "sound beating." The unrepenting offender
who offers che genealogical pedigree of the family, as well as cirations of legal precedents
would ultimately face social boycott.
from other parís of India . 1 have , of course , left out the narres ofthe pcople concerned.
31. ilistorically, Ladakh has had clase links with the Congress Parry, bur many Buddhists, espe-
2t. This is no longer che case. Sine e ag89 , Congress has only once, in 1996 , managed to win Che
cially youths , resignad their membership in tg89. Prior to rhe elections for che Hill Council,
Lok Sabha seas. bur lost control again by a landslide in rg98. Most recently, in 1999, che
most rejoined Congress and only cavo seats were contesced by non-Congress candidates. All
Nstional Conference Party candidate from Kargíl defeared formar IBA supreme Thupstan
seats en rhe first Hill Council were initially held by Congress members. Links with tire Bil,
Chhewang . See van Beek (1999) for a diseussion of chis
and orher political developmcncs in
and other Hindu narionalist organizarions liase been revived in recent years, particularly
I adakh since che granting of regional autonomy.
lince che risa to power of che aje at che center (van Beek x999).
22. The Commission of inquiry was followed by a Constitucional Reform Conference, aleo
32. Although in times of crisis statentents are made on bchalf ola Ladakh Muslim OrganizaNmt,
chaired by Glaney, and a Franchise Commitree, which recommended tire eonstitution of Chis
che LMA is no equivalen [ of che [BA, lacking es ir does any formal status, mernbership, or
parliament . por rhe recommendations , see Jammu and Kashmir ( 1932a: 1932b
; 1934). sratutes . There are cwo maro Muslim organizacions in Leh: che Anjuman - e Mu"in -e Islam
23. See, e.g ., Bertelsen (1997 ) for a diseussion of che role of Kashnliri Pandits in Buddhist or-
is rhe main Sunni social organization ; tire Anjuman - e Imanria ¡s its Shia counterpart. See
ganizaóon in Ladakh in Chis perod.
Pinault ( x999 ) for more detail.
24. Local accounts vary, bur apparendy t here was agrcement ro remain wirh India . A local milicia
33. Krishna Rao, address coche Hill Council.

388 STATES OF IMAGINATION

AUTONOMY FOR LADAKH 389
 n dcuv tl:v valiente of thc links Eciwcen senranác panty .rol social altion hija
to yue+ón:: ihe rnnccnr c.:sv causal linking oJ ,,,1, urders svirl, edad, v,olunce. Indeed,'te r EIBLI O GRAPI 1Y
Ibrms of idanri nt caloicmenr, and ¢ Iwrial sobrr .tn" are cruda] In che mediaban o:
CO]] ncetion. as I ht 'remad c9.c 99L.:1.
xu aL<o cha c>cell, ,1 eollettion cdl:cd bc Cheah and Robbinv 5'9951.

Abraham, Itry . 1996. Scien ce and power in che posteolonial stare. Alternativas 21141: ;z1-39.
Abrams, P 1988. Notes on die difficulry of studying the srate, Journal of Histmical Soclo logy 1(1):

58-89.
Abu-Lughod, Lila :99o. The romance of res istance:'Ira e: ng Ira n sfo rmadono ni power rhrongil

Bedouin eeomen. Anterimn Ethnologist 17(1): 41-55.

Acr (South Africa). 1995. Statutes of che Republic of South frica. Constitutional law- Promottan ot national

untty and reconciltation . Gocernment Gazette. Act no. 34 of 1995 as amended by Act no. 87 of
1995. Soud: Africa.
African Nacional Congress (ANC). 1996. Resolutivo of ANC peam and securih) swnnttt ¿4ugusr-
September.
. December.
-. 1997- Strategies and tactits : Resolution adaptad at 5oth nacional congreso in Mafekeng
. htnll.
Available froni www.ancorg . zalanedoeslhistorylconflconferenec5olstrategyamend
mg.co.za.
Afrikanerbond forges cíes wirh ANC. 199q. Elettronic Mail and Gunrdinn, 27 August. www.
Agamben, Giorgio . 1898. (domo cocer : Soveretgn power and liare life. Sranford, CA: Stanford Universiry

Press.
Agnew, John. 1999. The new gcopolitics of power. In Huasa, geography todas, ed. D. Massc}t
J. Alíen, and P. Sarro, Cambridge, England: Polio, Press.
1995. Mas:ering space: Hegemon5, territory and internatto nal poi ¡r-
Agnew, ) ohn, and Stuart C orhridge .
ica! economg London : Roudedge.
Ahmad, Muneer. 1964. The civil servant in Pakistan: A study of che bad<ground and altitudes of public

se"anes in Lahore . Karachi, India : Osford Universiry Press.

- psis: 1'!uríml-
Albán Gómez , E., A. Andrango , and T. Busramenre . eds.:9g3. Los indios u el estado
turalidad y multi-etnícidad en el Ecuador . Quito, Ecuador Abya-Pala.
Alíen, J. 1989. Span al assemblagos of power From dominante co enlpowerme ir . lo Hummrgeag-
raphy todas, ed, D. Massey, 1. Alíen, and P. Sarre. Cambridge, England: Poliry Press.
Ahncida. I., ad. 1991. ]odios: Reflexiones sobre el leva atamiento i ndígesa. Quito, Ecuador: 1 LO15.

390 STATES OF IMAGI NATION

 Alonso , Ana Mario. 1994 . The político of space, time and substance
: Suite formation , narional-
. 1995. L fe in fragments : Essays in postmodem morality. Oxford : Blackwell.
ism, and ethnicity. Animal Romeo ofAnthropology 23: 379-405.
1995. Thread of blood_ Colonialism , revolution and gender
Rayart, J. F. 1991. Finishing with che idea of the third wodd: The concept of political trajectory.
on Meeíco's norihern frontier. Tucson:
Universiry of Arizona Press. In Rethinking third world politics, ed . James Manor. London: Orient Longman.
Alcheker, A., and J. Steinberg . Bayart, J. F., S. Ellis, arad B. Iiibou, eds. 1998. The criminalization ofthe state in Africa . London: James
1998. Reavon and representarion in National Party discourse,
19go - 92. In South Africa in '1'ransition : Currey.
New Theeretical Perspectiva, ed. D. Howarth and A. J.
Norval . London: Macmillan. Beasley-Murray, J. 1999. Learning from Sendero : Civil sociery theorv and fundamentalism . Journal
1 he ANC triumphs at Mafekeng . 1997. Electranic Mail and Guardian of Lacia American Cultural Sodios 8(t): 75-88.
, 2; Deeember. s+nvwmg.eo.2,a.
Anderson , Benedict . 1991 (1983 ). Imagined mmmuníriec : Bertelsen, Kristoffer Brix. 1997 . Protestant Buddhisin and social identification in Iadakh. Archives
Refections on the origin and spread ofnation-
alism. London : Verso, de sciences sociales des religions 99 Jnly-September : 229-51.
Ehasin, M. K . 1992. Cold desert : Ladakh. Emlogy and develupmene Delhi: Kamla - Raj Enterprises.
1992. The new world disorder, New Left Reuiew (193):3-13.
1998. Nationalism , identiry, and che logic of seriality. In The spectre Billig, Michael , 1995- Banal nationalism . London: Routledge.
of coreparisons: Nation-
alism, Southeast Asía and che wodd , London: Verso, Blasting a hole in crime smts, 1997 . Electmnic Mail and Guardian , 19 September. www.mg . co.za.
Andolina , Roberr. 1999 . Colonial legacies and plarinatlonal imaginarias Booth, David , and Bernardo Sorj. 1983. Military reformism and social classes : The Peruvian experiente,
: Indigenous movement politics in
Ecuador and Bolivia . Ph.D. dios, Minneapolis : 1968 - 8o. New York : St. Martin's Press.
Universiry of Minnesota.
Boraine, Alex , 1996a. Alternativa and adjuncts te criminal prosemtions. Paper presentad at seminar
Angel], A. 1982, Classroom Maoists: The polides ofPeruvian schoolteachers andar military gov-
emmenr. Bulletin of Latín American Raearch 1 (2): 1-20. Justice in cataclysm : Criminal tribunals in che wake of mass violente . Brusscls, Belgium,
Ansión, T. 1g89 . La escuela an la camun,dad campesina . 20-21 July.
Lima: Proyecto Escuela, Ecología y Conumi-
dad Campesina. . 19966. Justise in cataclysm: Criminal tribuials in the wake of mass violente . Paper delive red in
Ansion, J., D . del Castillo, M, Piqueras . and 1. Zegarra . Brussels, Belgium , seo-21 July.
1998. La escuela en tiempos de Guerra: Una
Boraine, Alex , and Janet Levy. 1995 . The hoaling of a nation . Cape Town, South Africa : Justice in
mirada a la educación desde la crisis y la violencia . Lima, Peru : Tarea, Asociación de Publica-
ciones Educativas, Trensition.
Appadura i, Arjun. 1993. Number in The colonial imagination Boraine, Alex , Janet Levy, and Ronel Scheffer. 1994- Dealing with che past: Trath and reconciliarion in
. In Orientalism and che postcoloaial
predicament : Perspectiva en South Asia , South Africa. Cape Town , South Africa : 1DASA.
ed. Carol A. Breckenridge and Peter van dar Veer. Phil-
Bourdieu , Pierre. 1977 . Outline ofa theorv ofpractico . Cambridge , England: Cambridge Univcrsity
adelphia: Universiry of Pennsylvan la Press,
-. 1996. Modernity at larga: Cultura l dimensions ofglobalizaGOn Press.
. Minneapolis : Universiry ofMin-
nesota Press. 1990. The logia of practica. Cambridge, England: Polio, Press.

-. 1998. Dead cerrainry : Ethnie violence in che era of globalizarion 1999- Rethinking che state: Genesis and structure of clic bureaucratic field. In Statei
, Public Culture 10(2):
Cultura : State-formation afeen che cultural furo, ed, G. Steinmerz . Iliaca, NY: Cornell University
225-47
Arce Borja , L., ed, 1989. Guerra popular en el Peru : Press.
El pensamiento Gonzalo . Brussels: El Diario.
Arendt, Hannah , 1958. The origino of tatalitarianism . Breuillg J. 19g3. 2d ed. Nationalism and the state _ Manchester, England: Manchester Univcrsity
London: Allen and Unwin.
Arnion , Cynthia L ., Press.
ed. 1999. The popular referendum and the fisture of che peace pmcess in Guatemala.
Bricker, Victoria R. 1981. The Indian Christ , che Indian king: The historiral substrate of Maya myth and
Latín American Program, working paper no. 241 . Washington ,
DC: Woodrow Wilson In-
ritual. Austin : Universiry ofTexas Press.
ternacional Center for Scholars.
Asad, Talal. 1993. Genealogía of religion: Discipline and reason Brito, Alexandra Barahona de. 1997. Human rights and demacratization in Latín America : Uruguay and
of power in Christianity and Islam. Balri-
Chile. Oxford : Oxford Universiry Press.
morc lobos Hapkins Universiry Press.
Asmal, K „ L. Asmal, and R. S. Roberts . 1997 . Reconciliation through truth Buddhista , Representativas of Kashmir. 1932. Memorandum of che Kashntir Buddhists . The Me-
: A redmning ofapartheid's
criminal governance . Cape Jówn , South Africa : David Phillip. habodhi40 (3):127-31.
Baig, Tara Ali . 1979- Out children . Burga, Manuel , and Alberto Flores Galindo , 1979. Apogeo y crisis de la república aristocrática. Lima,
New Delhi: Governmenr of India, Ministry of Inforntation and
Broadcasdng , Publications Division. Pero : Institnm de Estudios Peruanos.

Rastos, Santiago , and Manuela Camus . 1995. Abriendo Caminos Burton, Richard F. 1988 ( r8g). Scndh and the Taca that iuhabit the valley of che Indus. Kornchi, India:
: Las organizaciones mayas desde el Norel
hasta el Acuerdo de derechos indígenas . Ciudad de Guatemala , Indus Publishers.
Guatemala : Fiasco,
Bauman , Zugmunr. 1989 . Modernity and che Holocaust. Ithaca . 1993 (1877), Sindh revisited : Government of Sindh , Karachi, India : Deparnnent of Culture
, NY: Cornell Universiry Press.
and Tourism.

392 BIBLIOGRAPHY
BIBLIOGRAPHY 393
 1997. Unidad del ,,tostado e ruglune: coa
Fuerl,,,, 1996 1,,¢.enes _ j s. te iru setir1 les South afmo. l}at11 c.r.d Feto m silla ti o a Can I111issl a s mol
p:lmesu mrbng u2.u 1 'l atunes ai1 rt mra um-.an 1 l lidola: Bnk t.iadacl cla
thelrllmes 11ititurini tus lodern a homstate. Foper delivered at [he intcrnatioual seminal
Guate Halo, Gunleora la: 1'rieddel 1 berl btiticni
Scucs 0f Imagiu:lt 1n. C p nhagrn - Dcnmaek 13-15 February.
Is a 1 ,.'atl 1,. 1 t Ahil 11! stork a; ice Sc .. h d¡1.u:: mita aad Colbus, loar I_ 1ySS. l n 1.c 1..a, mg n Fuml la (1
1491- nnun;rl :.
Ilecom I:atton con. m .asa . Papcr preserlred .11 che intern.uional eontereure " Che IRG ( casi Unlversiuv Presa.

ulissioning ihe par ." lohau nrsburg . Universiry ot tice VVirwatcrsraud , 11-14 Ju le, Cnmisian dt Isdmidmienm Ilisrr
hrd 1.u mas.urgl ell a.porrlspo r5h'tochrrnL
t.mctau . 1ran k. 1ybq - 1 ,r, , a nd "le, :¡ ", 1, 1 I kl:l9an, r ty u .y:.lsmlJ af tie nuspot aryo sgslea
Connoilj. A4. IS. .9yt.II ntIIUID rrans Iundol 1 Corncll Lcir
I n 7 U anratan . Sautlm d t ,A : Stanrnc U nuesity cs
Gmxj o de Org:ulisaesl Aav,u dm l2.ulul nal:l ir usr .9q1 llr ,rlmal
Can nev lord : nhip :,.l h c ANl? 19y C.1-uanic \!, ti a Laardlcl-. v. DecemLer. suxrc_
Cuidad de Guatemala, Guatemala: coMC.

Cape activista resist truth hearing. 1996. El eltronie Mail sud Gtiardian, 26 April. www.mg.eo.za. Consejo de Planificación Nacional para Indígenas y Negras le:u,\ P1dDEtN). 1gy at 1 te ac G:s
tedn del PRDDEPINE: Dommento de trabajo. Manuscripe. Quito, Ecuador: ceN t ADts1N,
Carnlaek, R. M. 1995. Rebela of highland Cuaterna la: '1 1t e Quiché-Mayas of Momostenango. Norman:
Contreras, C. 1996. Maestros, mistas y campesinos en el Perú rural del siglo XX. Lima. Pera: Docunsenm
Universiry of Olciahaola Presa. I
Cassirer, Ln1st. 1946.'fhe recta of the state. New York; Doubleday Anchor. de Trabajo, Instituto de Estudios Peruanos.

Cawrhra, G 1993. Policing South Africa. The SAP and che transition from apartheid. Cape Town, Coronil, Fernando 1997 . The magiml state: Nature, money era modernity in Venezuela. Chicago: Uni

South Africa: David Philips. versiry ofChicago Press,

Coronil, Fernando, and Julie 5kurski. 1991. Dismembering and rememberi ng tire natou: The
Censos ofIndia. 1951. New Delhi: Goveroment o[ India Publicaticures,
1 semantics of political violente in Venezuela. Camparative Studies in Sodety and Cultme 33(2):
Chandavarkar R. 1994. The origins of industrial capitalina in India. Cambridge, England: Cambridge

Universiry Press. 289-337

Corrigan, Philip. 1994. State formafion. In Cveryday forma ofslatejornmtion: Reuolution and che ne,qo'
1998. Imperial poseer and popular polities. Cambridge, England: Cambridge Universiry Presa,
Chandra, Bipan. 1984. Communalism in modero India- New Delhi: Vileas. Italia, of role in modero Meeico, ed. Gilbert Joseph, and Daniel Nugene. Durham, NC: Dulce

Chatterjee, Partha. 1986. Natívnalict thought and the colonial world: A derivative discourse? London: Zed Universiry Press.
1
Corrigan, P., and D. Saycr. ig85.1he great aren: English stateformation as cultural revolutton. Oxford:
Books.
1993. The nation and itsfragmenta: Colonial and posteolonial histories, Princeton, NJ: Princeton Basil Blackwell.
Cortés, S. ig6o. Breve reseña histórica del instituto geográfico militar In Instituto Geogrdfice Militar
Universiry Presa.
1995. Religious minorities and the secular sute: Reflections en an indias] impasse. Pubhc 1928-1960. Quito, Ecuador: IGm.
Crain, M. 1990. The social construaion ofreational identiry in highland Ecuador, Anthropological
Culture 8:u-39.
. 1998. Beyond ihe nation ? Or within? Social Text 16(39): 57-69, Quartedy 15(3):43-59.
Cheah, Pheng, and Bruce Robbins, eds. 1998. Cosmopolities: Thinking and feeling beyond tire nation. Cmok, John Hurrell. 1991. Buddhist ethics and tire problem of ethnic minorities: The case of
Ladakh. In Buddhist ethia una modero sodety, ed, C. Wci-Hsun Fu and S. A. Wawrytko. New
Minneapolis: Universiry of Minnesota Press.
Cliquct, Robea, and Kristiaan Thienpont. 1995. Populatton and development A messagefrom ele Cairo York: Greenwood Press.
Dandeker, C. 1990. Suteeillanre, power and modernity. Cambridge, England: Polity Press.
conference. Boston: 1Cluiwer Acadentie Pub lis hers.
Dasgupta, Shahana. 1g9o. Child welfare legislation in India: Will Indian children benefit froto
Coetzee, C. 1998. IGotoá remembered: A mother of uniry a mother of sorrows? In Negotiating he
ihe Untted Nations convention ora the rights uf the child? Mírhigsu journal qC lnternatioual
post: The maleing of memory in South Africa, ed. S. Nuaall and C. Coetzee. Cape Town, South
Africa: Oxford Universiry Press, Laso 11:1301-16.
1 Davidson, P. 1998. Museunss and tire reshaping of memory. In Negotiattny sise pase ihe unaittng of
Coho, Bernard. 1y83. Representing authoriry in victorian India. In Tice invention of tradition, ed.
memory in South Ajhta, ed. S. Nuttall and C. Coetzee. Cape Town, South Africa: Oxford
Erie liobsbawm and Terence Ranger. Canbrdge, England : Cambridge Universiry Press.
Cohn, Bernard S. 1987a. An anthropologist among ihe historiaras unid otlter essays. Delhi: Oxford Uni- Universiry Press.
Davia, Shellou H. 1997. la tierra de memos antepasados: Estudio de la herencia de tierra en el altiplano de
versiry Press.
a987h. The censos, social strucrure, and objectifier tion in South Asia. In An anthropologist Guatemala. Antigua, VT: e1RMA and Plumsock Mesoamerican Smdies.
1
Deaeon, H. 1998. Renlembeting ttagedy, constmeting modernity: Robben Island as a nationaj
among the historiaras sud other emays, ed. B. S. Cohn New Delhi: Oxford Untversity Press.
Co tj CCusi1, DemeIrio. 1996. The polities ufMaya revirad icatio o. In Maya cultural settvíswn ¡Ti Guate- tnonuosent. In Neg oti ati ng ele pasta The tuaking of memory in South Afitm. ed. 5, Nurcall and

...ala, ed. Edward F. Fischer and R. McKeuna Brmcn. Austin: Texas Untversity Press. C. Coetzee. Cape Town, Soutic Africa: Oxford Universiry Press.

BIB LI OGRAPHY
BIBLIOGRAPHY 395
394
 Dean, Mitchell . 1999, Gouernmentaltty : Power and role in modero society. Thousand Oaks, CA: Sage
Esquit, Edgar, and Iván García . rggg. El derecho consuetudinario y la reforma legal en Guatemala . Ciudad
Publications.
de Guatemala, Guatemala : Flasco,
Dean, Mitchell , sud 8arry Hindess, eds . 1998. Governing Australia :
Studies in mntemperary rabonalities Esteva, Gustavo. rggg. Developmenc . In The development dictionary: A guide m knowledge and power,
of government . New York: Cambridge University Preso.
ed. Wolfgang Sacho. London : Zed Books.
Dearh and drudgery on die beat . 0997, Elettronic Mail and Guardian ,
20 Junte. www.mg.co.za. Evans, P., D. Ruschemeyer , and'IL Slmepol, eds. 1985. Bringiug the state bario in. Cambridge. En-
De Brito, A. B. 1997 . Human rights and democratízation in Latín America
: Uruguay atad Chile. Oxford: gland: Cambridge University Preso.
Oxford University Preso.
Ewing, Karherine P. 1983. The polides of Sufism: Redefining the sainrs of Pakistan. ]ournal of
de Certeau, Micbel. 1984. The practico of everyday 1 fc. Berkeley: Universiry of
California Preso. Asían Studies 42(2): 'SI-68.
Degregori, C. 1. 1990 . Ayacacho 1969 - 1999: El surgimiento de Sendero Luminoso
. Lima, Pero: Instituto 1997. Arguing sainthood : Modernity , psychoanalysis, aod Islam . Durham , NC: Dulce Universit,
de Estudios Peruanos.
Preso,
. 1991. How difficult it is te be God :
Idenlogy and political violente in Sendero Luminoso. Ex-spy Williamson clashes vith Pik. 0995. Elettronic Mail and Guardian , 27 December. www.mg.
Conque of Anthropology x(3): 233-So.
De Jong, Ferdinand . 2000. The power of secrecy : Performance and Che politics of
locality in Casamance, Falla, R. 1978. Quiché rebelde . Cd. de Guatemala: Editorial Universitaria de Guatemala.
Senegal. Ph . D. dios., University of Amsterdam.
Peldman, Alían . 1991. Formations of violente , Chicago: University of Chicago Preso.
De Klerk, E. W. 1gg7. Submission te the Truth and Reconciliation Commission
. httpaJwww. truth.org.zaj Felstiner, W., R. Abel, and A. Sarat. 0g8o . The emergente and transformation of disputes: Nam-
np-truth.htm.
ing, blauring , elaiming. Law and Secreta Reviro 05 (3-4):631-54.
. r9go , Eensgesindhcid binne bereik . RSA Beleidsoorsig (7).
Ferguson , James. 1990 . The anti-polilla machina : " Deoelopment," depolítícizatlon , and bureaecratic power
Derrida , J. 1973, Speech and phenomena. Evanston ,
IL: Northwesrern University Preso. in Lesotho . Minn espolio : University of Minnesota Press.
de Sardan , Olivier. 1999 . A moral eeonomy of corruption in Africa. Journal of Modero African Studies
Femandez , Lovell, 1991 . Poli re abuses of non-political criminal suspects: A survey of practices in abc Cape
37(1): 25-52.
Peninsula . Cape Town , South Africa : University of Cape Town, Instiote of Criminology.
Dhareshwar, V., and R. Srivatsan . 1996. Rowdy sheeters :
An essay un subalterniry and político. Fischer, Edwin R, and R. McKenna Brown , eds. 0996. Maya cultural activism in Guatemala . Austin:
Subaleern Studies 9 , Delhi: Oxford University Preso.
University ofTexas Preso.
Dijkink, G. 19g6 . National identite andgeopolitical visions. London: Rontledge.
Fivaz Cakes en Mbeki Police Squad. 1997 . Elettronic Mail and Guardian , I Octobec www.mg.co.za.
Dlllon, Michael . 1995. Sovereignty and governmenrality : From che problematics
of Che "New Pog Olwig, Karen . 1997. Cultural sites: Sustaining horno in a deterritorialized world. In Siting
Wodd Order" o Che ethical problemaric of Che world order. Alternativa 20: 323-68.
cultura: The shiing anthopological object ed. ICaren Fog Olwig and Kirsten Hastrup. London:
Dirks, Nicholas . 1987 . The hollow crown : Ethnohístory of en Indían kingdom
, Cambridge , England: Roudcdge.
Cambridge University Preso.
Poucault, Micho] . 1977. Discipline and punish : The birth ofthe prison. London: Alían Lane.
Douglas, Mary. 1966. Panty and danger : la analysis of concepto of pellution and taboo
. London: Rout- . 1978. ' I'he History of Sexuality . Vol. 1. New Yor lc: Random House.
ledge and Kegan Paul.
. 1979a. Discipline and punish : The birth ofthe prison . New York: Vintage Rooks.
Urew, Frederic , 1976 (1875 ). The Jummoo mid Kashmir terr itorios,
New Delhi: Cosmo Publications. . 197gb. The Histery of Sexuality . Vol. 1, London : Allen Lane,
Drysdale , R., and R . Mvers , 1975- Conrinukv and chango :
Peruvian education . In The Perueíae . oy8o. PowerlKnowledge , ed. Colin Gordon . Brighton , England: liarvester Wheatshea£
experiment es, Abraham Lowenthal. Princeton ,
NJ: Princeton University Preso. . 1982. Afterword : The subject and power . In Beyond structuralism and hermeneutia , ed. Hu-
Duueld , Mark. 1998 . Aid policy and post- modero conflict : A critica[ policy reciew
. School of Publie bcrt Dreyfuss and Paul Rabinow . Chicago : University ofChicago Preso.
Policy, discussion paper no. 1g . Binningham , England: University ofBiriningham.
. 1984, On Che genealogy of erhic. In The Poucault reader, ed . Paul Rabinow. Harmonds-
. zoos Global governance and Che new wats : The merging af deuelepment and secudty. London: Zed
worth, England: Penguin-
Books.
1984. Space, knowledge and power . ni The Faucault reader, ed. Paul Rabinow , Ilarnlonds-
Duntonn Louis. 1980. Homo hierorchicus_ Chicago: University of Chicago Preso.
worth, England : Penguin.
Eickelman , F, Dale, and J. Piscaori. agg6 , Muslim politirs.
Princeton , NJ: University Preso. . 1988a. The ethic of Che cara of the self as a practica of &eedon0 . In The Final Fourault, ed.
Ellis, Srephen . 1998. The historical significanee ofSouth Africa
's Third Forre . J aurnal ofSouthere J. Rernauer and D. Rasmussen . Cambridge , MA: mIT Preso,
African Studies, vol. 24(2): 261-99,
. 09886: The political rechnology of individuals. In 'Ledmologtes of the ielf A seminar toith
Escobar, Arturo . Igg5. Encountering devclopntent . Princeton, NJ :
Princeton University Preso. Michel Poucault, ed. L. H. Martin , H. Gumlan, and P. H. Hntton. London : Tavisock.
Escolar, M., S . Quintero , and C. Reboratti .
1994, Geographical identiry and patriotic representa- . t988e Politics and reason . In Politice, philosephy , culture: Interviews and other oeritings 1977-
tion in Argentina . In Geography and national identity, ed. D. Hooson ,
Oxford: Blaclavell. 1984, ed, L. D. Kritzman. New York : Routledge.

396 B1 BLIOGRA11 HY
BIBLIOGRAPHY 397
 5995- Govennnr Itrlitl_ In'1 íle F ar.'tt t edrrt sandios m gelernmratnlity, ed_ Grahant Bu Grantaci, Antonio 1971. .1k-c11011o Jrtu dt urt' mlehocl.. Ion Jora lrts r 11,1 i AYlsh

chdl C.IID Gan10n.,1ld P,t, 1,ller. Londou:1 ,ter Whaaahcaf. Greenhalgh, Sisan- 5999 I'launed I n ¡ti] mrpla:lile r (rrsms [lopulalisrti " In Ti
5995. lt essevt, ur-Is 1sHr Ig3,. vol. I, Eth n ariec ti v llq rnd tmd:. ecl. 11u11 Rnhinosr. hin,od,roitu_ Unpublishe(I nnunrseri pt.
11
G rülilhs. done U,, O. 1)9-. 'a 1'e'1101 r, 5 urnnl. r 1...

gg,h II Jnw:1 t sa ,,ti:, Pan, G.dlintard:Suull. Chi,asi, Us:iversit ,t (hieygo Pess

Las
Fmitag, Sandria- 5996. Couuating ín pub)¡o Colonial I o grcies r nd mmm^tpotara communalisnt- Gdliiths, loltn- 5986- Wbar i_ .egal plur ,isla' loan a-1 L (.
In C n r t i n q I h c na t hrl die[. c I t w N,,. ¢d 1 t :rs el de t oro ul in 111 dl ed. David 55.
Ledcl-n. I'hilsd, phn Cnmcrsitr 10 ladcani,I Pns> siriso, Ni vota. 1985. L.u:alch , a rmdiu ruc :n I.Idal:

¡en 1 , d , 1 1 , 1' 1!. Ihc a 1,1 011..1test tt' ' 1c 17, r nJ ioe;. ..i
Gauhar L N., and l.. N. Ganher. 11)56. Bu.dhism. in Kaslnnir attd Ladakh. New Delhi: Aurhors. letche and P. KuplaniaTI. Pan, F1lile e. Centre P)rúuhen de G

Garci;r González, G. 0991. R sumen de geografía, historia y cívica. Ciclo basico. 1. curso. Quito, Gc- Moutagncs.
nador: Editora Andina. 1994. Thi use ofohligatory labour for portcrage in Ere-independune; Ladakh. lo Tibetan
Ti leoan Studies, Fagemes
Geertz. Clifford . 198o . Negara : Tire theatrc state in nineteenth - rentoro Sal¡ . Prineeton , NJ: Pri ncemn oto dios: Pmreedings of Che 7th seminar of rhe International Assatiation for

Universirye Press. 1992, ed. P. ICvaerne . Oslo: Institute for Conlparative Research in Human Culture.

Giddens, Anthony. 1985. The nation-state and violence . Cambridge, England: Polity Press. Guha, RanapL 1999 ( 1983) . Elementary aspects oj peasant insungenñl in tilo total India. Foreword by
. 1992. -1 he consequentes of modernity . Stanford , CA: Stanford Universiry Press. lames Scotr- Durham, NC : Doke Un¡versiry Press.
Ihe discourse ofcoauptiort che culmre of pohdes, and
Gillis, 1. R. 19940 . Memory and identity : The history of a reladonship. In Conrmemorotions: The Gupta. Akhil . 1895. Blurred houndaries :'
palitirs of nacional identity, ed. J. R . Gillis. Prineeton , NI: Prineeton Universiry Press. che imagined 'tate. AmeriTan Ethnologist 22(2): 375-402.

Gilmartin, David . 11)88- Empire sed Islam: Punjab asid tire mak ¡ng of Pakistan. Berkeley: Universiry of . 1998. Postro lo vial developmenn. Da rham, NC: Dulce Universiry Press.

Califamiro Press. Gupta, Alchil, and lames Perguson. 0992. Beyond culture: Spaee, identity and une polides ofdif-

Girard, René . 0977. Vtolence atad rhe socred . Baltimore: Johns Ilopkins Universiry Press. ference. Cultural Anthropologg 7(O: 6-23.
. Hmnannies in Societu 5(3):
Gitlita. Ni 1979. Conflictos políticos en la sierra norte del Perú: La montonera Gene] contra Hacldng, lar . 1982. Biopower aod rhe avalancloe oí ptitired nttmbca
Leguía, 1924. Estudios Andinos 1)[16). 279-95.
Gledhill, John, 1994. Poner and its disguises : Anthropologíml perspecttoes en polttia . Boulder, CO: Plato . 199o. The tato i ng oí chance In ideas in co ntext. sol. Richard Rorty. (,amb ddge, Gngland:

Press. Cambridge Universiry Press.

Goffman, Erving . 1959. The presentation ofself ¡Ti eueryday tifo. Harmondsworrh, England : Penguin Hados, Pierre . 1995. Philosoph9 asa way of Iife. Oxford: Basil Blackwell.
, 1870 - 1930. In Ideas and
Books. Hale, Charles . 1996. Political ideas and ideologies in Latín Anoerica
Goldblatt, E ., and S. Meintjes . 1996. tienden ami rhe Trullo and Reconciliaban Commission . A subm¡s- ídeologies in tnentieth century Latín Amenas, ed. Leslie Bedaell. Camb ddge , En gland: Cana-

sian to che Tmth and Reconciliation Commission . May. http:llwww. truth.org. z2lsubmirl bridge Universiry Press.
. CurrentAntlrro-
geuderhtnt. . 1997 . Conscio isness , violence, and dite polities of memory in Guatemala
G oodeno ugh , C11ervl. 19990 . Richmond: Orce ntan ' s war ? Críroe and Confiict t5, aumnm. palagy 38 (5): 817-38-
. r9ggh. Who kitled Sifiso Nkabinde? lo Hele, Suzman Foundation , Kwa-zuna natal brisng 14. Hansen, Thomas Blom. 1996. Recupcradng ntasculiniry : Hindn nationalism, violence and che
fleten Suzman Dunbac Helen Suzman Foundation. exorcism of rhe Muslim "odie[." Critique of Anthropology 16(2): 137-72.
Gootenberg, P. 1989. Between silvermnd guano : Cmmnertial policy and che atare in post - independence Pero. 19970. Incide rhe romanticisi episteme . Thesis Eleven 48: 21-41.
. Paper delivered at
Prineeton, NJ: Prineeton Universiry Press. 1997b- Segmented worlds : W ork , livelihood anal identity in central Mumbai
0993. Ionagleiny develop mente Ecanomte ideas in Peru's'htritious p rosperíty." Berkeley: Univer- rhe conference Workers in Mumbai, Mumbai , 20-23 November.

siry ofCalifornia Press. . 1998. BJP arad che politics of Hindutea in Maharashrra . In The BJP and compulsions of politice

Gorrid, G. 09go. Sendero : Historia de luguerra mtlernaria en d Peru . Linfa: Editorial Apoyo. in India, ed. T. B . Hansen and C. la(frelot. Delhi: Oxiord Universiry Press.
Hindn nationalisnt and demacracy in modera India . Prineeton, NI:
. 0999. The Shining Path: A history oí rhe millenarian mar in Peru . Chape ) Hill: Universiry of 1999. Tic sa¢roe arate:
Norrh Carolina Press. Prnceron Universiry Press.
. Jouroal of
Gosq P. 1994 . Deathly wraers asid hunyry mountains : Agrarian ritual ami craso firnmtion in an Andean - zooo. Predicamento oí secularism : Muohm idendties and polocies in Mumbai
tocan. Toronto: Unh'ersiry oí Toronto Press. rhe Roya¡ Anthropologtml hutitute 6 (2): 255-72.
and che ideology of Sendero Luminosa Bulletiin ofl.atin
Uovernment of india. 5985. A Decide of ICDS: Inlegrated Child Deee lopment Serutcec New Delhi: Min- Harding , C. 1987. Antonio Diaz Martinez
istry of Humnn Resoorees Development , Departmenr ofwonoen ' s Welfare. Americio Research 7(0: 65-73-

398 B1 B LIOG RA P H Y B1 B LIOG R APHY 399

 Handg )in. 1994. Reuolution in the muntryside: Rural conjlict and agracian reform un Guatemala, 1944-
Huntington , Samuel. 1968. Political arder in changing secietíes. New Haven: Yale University Presa.
1954. Chape) Hill: Universiry of North Carolina Press.
Ibarra, Alicia- 1991. Los indígenas y el estado en el Ecuador . Quito, Ecuador: Abya-Pala.
Hayner, Priscilla. 1994. Fifteen truth commissions, 1974 to 1994: A comparativa study. Human Institute of Democratic Alternativa in South Aicca (IDASA). 1998. IDASA Opinen Pall 99. Cape
Rights Quarterly 16: 597-655.
Town:1DASA.
-. 1996. Cmnmissioning che tmth: Further research quesrions. Third World Quarterly 17(1): Instirte for Security Studies (ISS). 1998. Crime in Cape Town: Resulte of a [íty victim su,vey. Mono-
19-29.
graph series no. 23. Pretoria, South Africa: esa.
-. 1997. International guidelines for che creation and operation of
the truth commissions: A prelimina ry lyob, Ruth. 1995. The Eritrean struggle for independence: Domination, resistance, nationalism 1941-1999.
proposal. New York: World Policy Institute, New School for Social Research. Camhridge, England: Camhridge Universiry Press.
Heaver, Richard- 1989- Improvingfamily plannioq, healih, and nutritivo outreach in India. World Bank Jaffcelot, Christophe. r9g6. The Hindu natimmlist movement in Indino politice. New York: Columbia
Diseussion Papero no. 59. Washington, DC: World Bank.
Universiry Press.
Henry, Gray, ed. 1997. Islam in Tibet and che illustrated norrative Tibetan cascaos. Louisville, KY: Fono
)alai, Ayesha. 1985. Democracy and authoritarianism in South Asia. Cambridge, England: Cambridge
Vitae.
University Press.
Hepple, Leslle. 1992. Metaphor, geopolitical discourse and che military in South Anierica. In James, P. 1996. Nation formation: 1'owards a theory of abotract commun(ty. London: Sage.
Writing worlds, ed. J. Duncan and T. Barnes. London; Routledge. Jammu and Kashmir Governnlent o932a. Report of che commission appointed under the ondeen of H.H.
Herzfeld, Michael. 1992. The social production of indii ferente: Exploring che symbolic naves of Western the Maharaja Bahadur, Nouember 12, 1931, te enquire into grieuanres and complaints. Jammu, india:
bureauaacy. Chicago: Universiry of Chicago Press.
Ranbir Govemment Press.
1998. Factual fissu res: Clailns and eontexts. Annals ofthe American Academy of Political and . ig32b. Report of che Kashmir constitucional reforma conference. Jammu, India: RanbirGoveen-
Social Sciences s6o: 6q-82.
ment Press.
1 lindes, Barry. 1996. Discourses of power: From Hobbes te Foucault. London: Blackwell.
. 1934. Report of che Franchise Committee. Jammu, India: Ranbie Govemment Press.
1997. Politics and governmentality. Economy and Society 26(2): 257-72. Jensen. Steffen. rg9ga- Discourses of violente: Coping with violence en che Cape Bata. Social
. Igg8. Divide and govern. Europea»]ournal of Social Theory 1(1): 57-70. Dynamics 25(2): 74-97.
Hinojosa, I. 1898. On poor relations and che nouveau riche: Shining Path and che radical Peruvian 19996. Privatizing apare: On social control and dio emergente al priuate serurity in South Africa.
lefr. In Shining Path and odien paths: War and socirty in Peru, 1980-1995, eds. Stern. Durham, Cape Town, South Africa: Universiry of Cape Town, Institute of Criminology.
NC: Duke Universiry Press.
Jensen, Steffen, and Simon Turner. 1996. A place called Heideveld: Strategies and identities among rol-
Hi rete, Sttsan P. Igg8. Pronesincrg and persevering: Gender and che discourse of disputing in en Afrimn oureds in Cape Towq South A frica. Research report no. 112. Roskilde: Roskilde Universiry,
klamic court. Chicago: Universiry of Chicago Press.
Institutos of Geography and International Development Smdies.
Hobbes, Thomas. 1968 (1651). Leeiathan. London: Penguin.
Jessop, Bob. 1990. State theory: Puneng capMtoli,t states in their place. Cambridge, England: Poliry
Hobsbawm, Erie. 1965. Primitive rebela: Studies in archaic forros of social movement ir rgth and 2oth
Press.
canturías. London: Normn.
Dones, K. W. 1981. Religious identity and che Indian census. In The censos in British india: New
1990. Nations and nationalism sine 1780. Cambridge, England: Cambridge University perspectives, ed. N. G. Barrier. New Delhi: Manotear.
Press.
Joedan, Z. Pallo. Jgg8. The nacional questíon in post-1994 South A frica- A discussion paper in prep-
1loliday, David, and William Sanley (foethcoming). From fragmeneation to a nacional projeas reate
aration for che ANC's Soth nacional conference . Available at www.ane.org.zalanedocsl
implementatian in Guatemala. Stanford, CA: Stanford Contar for International Smdies, Inter- diseussion Inatq uestio n. html.
nacional Peace Academy.
Joseph, Gilbert, and Daniel Nugenc. 1994a. Popular culture and state formation in revolucionar)'
Holomisa, Banco. 1998. Richmond violence: The UDM position. Sunday Independent, 2 August Mesico. In Eueryday forma of atare formation: Revolution and the negotiation of rulo in modero
1998.
Mexieo. ed. G. Joseph and D. Nugenc Durham, NC: Duke University Press.
Hooker, M. E. 1975. Legal pluralism: An intmduction te colonial and neo-colonial laws. Oxford_ Claren- , eds. 19946. Everydag fornns of acate formation: Revolution and che negotiation of role io modem
don Press.
Mexico. Durham, NC: Duke University Press.
Hooson, David, eci. 1994. Geography and nacional identity. Oxford: Blackwell. Kantorowicz, Ernst 1957. The king's two bodies. Princeton, NI: Princeton Universiry Press.
Horkheimer. Max, and Theodor W. Adorno- 1969 (1944). Dialelcrik den Aufkldrung: Philosophische Kashmir Raj Bodhi Maha Sabha (KRaMS)- 1935- Triennial repon ofthe Kashmir Raj Bodhi Moho Sabha.
Fragmente. Frankfurt am Main: S. Fischer Verlag.
Seinagar KREMS.
Human Rights Committee and Nerwork of Independent Monitors (HRCINIM). Igg8. Richmond: Kaul, Shridhar, and H. N. Kaul. 1992. Ladakh through the ages: Towards a new identity. New Delhi:
Role of security forres. Comm issioned repon. Johdnneshurg: HRCINIM. Indos Publishing Company.

400 BIBLIOGRAPHY BIBLIOGRAPHY 401

Kariol, Sudipra. 1992. The murginary insti ni tion ot India. In Subalten : Su c i-s 7 ed P. Chane pvt Lossenth;tl Ahralra ul. ry,a Fcni x amh¡yu n. rutolurissit n l ie l tnr! r t t u:'. (n l u t 1 1

ndev. Uelhe Oxfind Poirorsin' P[am. drmt 4i nri lit^. rr ruk. cd. A. Lo veoih.rl. Trina 'tn, NI. Prlna st , 1 u craity .'re>:

tIn ronsl 'u-in... ofarlo.. .d pasten Sie --aura. discmv:c hcgenumv.In.onLSi i,ui-Brighr tbth t. u)97- Intcru.ui onallly r ic n.l. invcr ig nt vcrnl rei tahry :md rrr ..t ioud

Ic^nm:Lrcd btp... a-I'.ngcls aud chula lulo. Lo ndon Bnnsh Aatdemic lten. reFanr m lmlra h.ri u r^-d'.d ^oulcmnc1 9--

7a . I h c modero rtc n hidra I n L 1, 1 nmmts o] r t ]ormatio,, In',', aud Europ c,,parnl, .taiguasha, Juan. r983. I a ----.rion res 'val Cn 'ci hl _ i Caaton -NT r "'-

ed. ,N. Doenboos aud t . has inrj. Ne w Delhi: Saga I ul heations. Historiadi Emador.'al.11, -mayal t.^.ad_F \.nlaMora. l2u> ti do::A,ptr.ciún

r99,•L. is ligin n ans. dentir ) in 111111,1, Lrlnir mili ir a 5tu11:11 20(2) 5-44 Ldiror.t Nacional.
- - u)98- Pild: and Ih, nrhlir sph, -, ` Concepts ; m il practicas abour s pace in rhe Cien nl M',jouidar, U. N. 19hr. R r n,i cah;ue, 7 rr,',a- 3omit r. tira I'uhüshinA na^u.

Stalin, Rian. ig9o. dhl truita'; 3ear. London: Vianrgc.

Kavc. st. igg¡-' 1he role of nnrh commiscimrs iti tito sanlr lar justice, reconeiliarion and de- Mtllon, Florencia. 1983. The d,tlite ofmmnurnta m Feru S Cert-al iirghlu t. Pero rton, N:"fnecunt

moerarisafion:fhe Sltlvadorean and Hondura-- ases. ]ournal ofLatin Americarn Studies4. Universiry Press.
fitrms ofsrare 1:0rntariun in o itt ecce nch-cenmry
Khilnani , Sunil. 1997.'1he idea of India. I larm ondsworch , England: Penguin. -. 1994. Reflcctians c un tus roins: Everyday
Mexiro. lit G. M. Joseph and D. Nugent, eds. Bveryday fonos aforase Formation: Reoolutiott and
Klarén, p 1986. ihe origins of modero Itero: ,SSo - 193 o. In The Cambridge history of Latin Amerim.
Dulce Universiry Press.
Vol. 5. Cambridge , England: Carnbddge Universiry Press. rhe negotiation ofrole in modero Mexiro . Dudham, NC :
1995. Peamnt and narion: The malting of postcolonial Meollo and Pera. Bedceley: Uviversity of
1973. Modernization , dislocation una aprieto . Latin Amcrican Monographs no. 32, Austin:
Universiry of Texas, Instirute of Latin Arnedcatt Studics. California Press.
, Mcdca and Chi le in rom-
. 2000. Peru : Society and nationhaod in tire Andes . New York ; Oxford Uviversity Press. -. 1999. Decoding rhe parehments of the Latin Amentnn nati onatatc Peru
Paper presenred m rhe conference The Formation of rhe NationState in
Kobrak. P H. R 1997 . Village tmublesr The civil putrols in Aguamtdn , Guatemala . Ph.D. diss. Universiry parative perspectree .
of Michigan. Latir Amcrica, Inscitute of Latin American 5todies, Universiry of London. May.
Nl: Princeton Universiry Press.
En i shnaswa nty, 1. 1966. A riot in Bombay, Aug. n, 1Sgs: A stud9 of Huida-Muslim relations in Western Mamdani, Mahmoud . 1996a. Cittzen atol subjeet . Princetan ,
India in ihe nineteenth eentury . Universiry ufChicago : Deparrment of History ( tu iero El m). . 19966. Reconciliation without justice. Southern Reuiew of BocW, no. 46, November-

Krog. A. 1995. The South African road. In Tire Healing ofa Nattonl, ed, A. Boraine and 1. Levy, Cape Dccember.
oL che African cxpericoce . In Crisis
Tocan, South Africa: Jusrice jo 'Ira o sir ion. r997 . From justice te reronclllation : Making sense
ed. Colin Lete and Mahmoud Mamdani. Discussion
Lada--, Ernesto . 1990. New refiecttons en the reeolution of out time . London: Verso. and remnstruction : Afrimn perspectivo ,
. 1996. Emancipation (.s). Chicago : Universiry ofChicago Press. paper 8. Uppsala : Nordiska Afrikainsdtute[.
of South Afiia.
Laclau, Ernesto , and Chantal Mouffe. 1985 . Hegentany ama socialist strategy :'fowards a radical dento- Mandela, Nelson . 1994. Que Nation , many cultores. Pretoria: Govertnnent
Sindico in political sociology. Oxford: Basil Blackwell.
natic polities . London : Verso. Mann, M. 1988 . States, war and tapítalism :
Lan, David. 1885 . Guns and ruin : Guerrillas and opine ntedimns in Zimbabwe . London: James Currey. Mano, R. S. 1986 . Thc Ladakhis : A cultural ecological perspective . In Emlogy, economy and religion

Latour, Bruno . 1986. The powers ofassociation . In Power, artion and belief. A new sociology oflmawl- of rhe Himalayas, ed. by L. P. Vidyarthi and M. Jha. Delhi: Orient Publicarions.
t99sa. La calda de la cuarta espada y los senderos que bifurcan. Mdrgnes 8 ( 13-14):
edge, ed. John Law. London : Rourledge. Manrique, N .
. 1qg,,. We bate neuer leen ntad ertt - Catt bridge . MA: Harvard Universiry Press. t1-42.
of Latin
Iawie, 5., and T. Swedenburg, eds. 1996. Displacentent , diaspora and the geogmphies of identity. Lon- -. 1995b. Folitical violente , ethnieiry and raeism in Pero in time of war. ]ournal

don: Duke Universiryy Press. Ameriten Cultural Stadtes 4(1):5-18.

eds. 1999. [f uman geography today . Cambridge,
Lefebvre , H. 1991- [19741 . The Pmdection ofspece . Oxford : Blackwell Publishers. Massey, Doreen , lobo Allen, and Phillip Sarre ,
Lefort, Claude . 1988. Demorrary and pahtical theory . Cambridge, England: Poliry Press. England: Poliry Press.
Lele, Jayant. 1995. Saffronizarion of Shiv Sena : The political economy of ciry , arate and nabo... In Mbeki, 'fhabo. 1999. 5catemenr co tu, Afrikanerbond. Pretora, 27 fuly. Available at tvwtt.anc.

Bombay : Metaphor of modero india, ed. Sujata Patea and Nice Thorner. Bombay : Oxford Uni- org.za.
Afinca 62 ( 1): 3-37.
versity Press. Mbembe, Achille . 1992. Provisional naces on rhe posteo1ony .
Letter to Soweto: Close to fire. 1995 . Elemonir Mail and Guardia.. , 24 FCbmari www.mg.co.za. McCrcery, David. 1984. Rural Guatemala 2760-1940. Staoford, CA: Stanford Universiq' Press.

Linares, A. 1988 . La mararia bureaucrarico - legislativo 7 la situación de los llamados " Polos de McMichael, Philip D. t996. Development and social ch.ange: A global perspective. In Socio logy for a

Desatollo y Servicios ," In CIEDFG óuatemala : Polos de Desarallo . Ciudad de Mexiro , Mexico: new century, ed. W. G. Charles Ragi and Larry Grifi n, Thousand Gaks, CA: Pire Finge Press.

Editorial Praxis. Meidng G. L. N.d. Address te rhe Truth and Reroncdiatian Connnission by rhe chief of hte South Afmmn

Livi ngsm n e, David N. 1992. ihe grogrophiml I raditioti: Episodes in rhe history al a contestad enterp risa. Nationai Defenre Force, General 6. L. Metring. Seo vv+vv.tru[11,org.za)sobmigsandt.h[na.

Osford: Blackwell. Merry, Sally Englc. 1,88. Legal plumlisnt. Lar, atol Soúetu Reuiew 22: 869-96.

402 BIBLIOGRAPHY BIBLIOGRAPHY 403

 1992. Anthropology, law and processes. Annuel Reeie,e ofAnthropology 21: 357-79
. inand for a Volkstaot. In South Africa in transition: New theoretical perspenives, ed. D. Howarth
1993. Legal pluralism and transnacional culture. In Human rights, culture and context, ed. and Aletta J. Norval. London: Macmillan.
Richard Wilson. London. Plum Press,
. 1999a. Rethinking ethniciry: Identification, hybridiry and democracy. In Ethnicity und
Merwe, H. van der, P. Dewhirst, and B. Hamber. 1999. Non-governmental organisarions and rhe
nationalísm in Africa: Cmtstructivist reejlections and mntemporary politics, ed. P. Y eros. London:
Truth and Reconciliation Commission: An impact assessmenr. Politiken 26(10).
Macmillan.
Memalf, Thomas R. 1994. Ideologies of che Ra). In The new Cambridge history of
India, vol. ;, ad. . 19996. Truth and reconciliation: The birrh of rhe present and rhe reworking of history.
G. Johnson. Cambridge, England: Cambridge University Press.
Journal of Southern Afrimn Studies 25(3): 499-519.
Midgley, James. 1975- Crome and punishment in South Africa. Johannesburg: McGraw-Hill.
Nugent, David. 1988. The mercantole transformation se' provincial urban ¡¡fe: Labor, value and time in rhe
Migdal, Joel. 1988. Stnmg sodetoes and weak atores. Princeton, NJ: Princeron Universiry Press.
northern Peruvian sierra. Ph.D. diss., Columbia University.
Miller, Peter, and Nikolas Rose. 1ggo. Governing eeonomic life. Emnomy and Socirty 1g(r): 1-3 i.
1994. Building rhe state, making rhe nation: The bases and limits of state centralizarion
? - 1997. Nacional evaluation of ontegrated child development services. New Delhi: Nt PCCU. ,1¿-
in "modem Pero." American Anthropologist 96(2): 333-69.
Mische, Ann. 1995. Projecdng democracy: The formation of citizenship across yourh networks
- 1995. Artisanal eooperation. fornts of labor, and rhe global economy: Chachapoyas,
in Brazil. Cítozenship, ldentity and Social History, Intemational Reviera of Social History 40(3):
1g3os ro rhe 1gg0s. Journal of Historical Sociolagy 8(1): 36-58.
131-58.
. 1996. Prona devil pacts to drug deals: Conimerce, unnatural accumulation and moral
Mision de Las Naciones Unidas en Guatemala (MINUCUA), ed. rgg5. Acuerdo sobre identidad y
community in "modern" Peru. American Eehnologist 23(2): 258-90.
derechos de los pueblos, 31 de marzo de 1995, Guatemala: m N000A.
. 1997. Modernity at tic edge of empire: State, individual, and nation in Ihe novelero Peruvian Andes,
-. 1988. Unpublished data en lynchings in Guatemala.
Mitchell, Timorhy. 1988. Colonozing Egypt. Cambridge, England: Cambridge University Press.
1885-1935. Stanford, CA: Stanford University Press.
. 1899. Stateand shadow state in turn-of-rhe-cemuryPeru: lllegal political networks and the
PJ^I;t^
19go. Everyday metaphors of power. Theory and Sociay 19: 545-77.
problein ofstate boundaries. In States and lllegal Practicas, ed. Josiah Heynian. London: Bceg. ' 'C~•
1991. The limits of
rhe stare: Beyond statist approaches and their critics. American Political . n.d. Beyond modernity and the nation-state: Post and transnattoual processes in the Northern Peru-
Science Reviere 85(1): 77-0-
vian Andes. Unpublished manuscripr.
_. 1999. Economy and che state effect. In State)Culture: State formation after ole cultural turn,
Nuijten, Monique. x998. In the narre of rhe land: Organization, transnationalism. and rhe culture oftire
ed. George Steinmerz. Ithaca, NY: Cornell University Press.
,tate in a Mexican ejido. Ph.D. diss., University of Wageningen, Holland-
Moore, Sally Falk. 1986. Social facts and fabricatíons: Customary lave en Kilimanjaro, 1880-s98o. New
Ochoa Garcia, C., ed. rgg3. Los contextos actuales del poder local: Governabilidad y municipalosmo. Gua-
York: Cambridge University Press.
temala City, Guatemala: IRIPAZ.
Montero, C. 1990. La escuela rural: Variaciones sobre un tema. Lima: PAO.
O'Donnell, Guillermo. rg96. The orate, democratizarion and some conceptual problems. In Latir
Nader, Laura. 1ggo. Harmony ideo legy, justice und control in a Zapotes mauntain village. Stanford, CA:
American political emnomy in the age of neoliberal reform: Theoretiml and comparative perspeetivesfar
Stanford University Presos.
de 19gos. Miami: University of Miami, North-South Center.
Nasr. Se}ryed Vali Reza. 1992. Democracy and the crisis of governabiliry in Pakistan. Asian Surrey
O'Malley, Pat. 1998. Indigenous governance. In Governing Australia: Studies in contemporary ratiotmli-
32(6):521-37.
ties of government, ed. Mitchell Dean and Rarry Hindess. New York: Cambridge University
Nacional Crime Prevention Srraregy (NCPS; Sooth Africa), 1996. Government Printer_ Pretoria.
Press.
Nedcor Pmject. 1996. De Nedmr Pmject en crome, violente and investmen4 Final reparo Johannesburg:
O'Malley, Par. Lama Weir, and Clifford Shearing. 1997, Govemmentality, eriticisni, politics. Eron-
Nedeor Project.
omy and Sodety 26(4): 501r7.
Nelson, Daaie. 1999. A fingen in rhe wound: Boda politics lo quíncentennial Guatemala. Berkeley: Uni-
Omar, Dullah. 1995. Building a new fiuure. In The healing ofa nation?. ed. A. Boraine and J. L65'.
versity of California Press.
Cape Town, South Africa: lustice in Transidos.
Nina, Daniel. 1gg5. Rc-thinking popular justice: Sef
regulation sal civil society in South Africa. Cape Ortner, Sherry. 1995. Resistance and rhe problem of ethnographic refiusal. Comparative Studies in
')'own, South Africa: Community Peace Foundation.
Sonety and History 37: 173-93.
Norval, Alerta 1. 1990. Posrscript: Post-apartheid? In New reflections un Che revoh:tiou of sur
time. ed. Osiel, M. 1997. Mass etmcities, collective nremory and the law. New Bmnswick, NI: Transacrion
Ernesto Dslau. London: Verso.
Publishers.
1gg6. Deconstructinq apartheid disceurse. London: Verso.
Padgaonkar, Dilip, ed. 1993. When Bombay burned. Delhi: UPSO Press.
1gg8a. M emory, idenriry and che (im)possib ility of reconciliadon: The work of rhe "rrurh
Pointer, Jo, and Chris Pulo. rgg5. Spaces of citizenship. Politiml Geography 14(2): 107-70.
and Reconciliation Commission in Sourh Africa. Constellatoons 5(2):250-65.
Palencia, Tania. 1997. Entre los hilos de la nueva cultura. In Guatemala 1983-1997: Hacia dónde va la
-. 19986. Reinvenring che politics of cultural recognition: The freedom front and che de-
transición?, ed. IeremyArmon eral. Cuadernos de Debateno. 38. Guatemala Ciry: FLACSO.

404 El BLIOGRAPHY
BIBLIOGRAPHY 405
 bridge, V.ugLin,] C,mlbri
Palmer, David Sato 1994. sl:inlnq Pal11 o Peru. znd J. Nr,e Sodc St. Martín s Press. Pricu. Panda- 1996. ko ,Inp sud polittcal pr

- -- igeS D n- cocha 1l o I I la y p o l i l l nfonna1 cn i I , n 1 El orno dr Ayaoaho. Paper p resa1md Lnt clvry' Press

.ir che 1as,. congres.. Chicago. a5'-3o Scpa-mber. Primov , 0.1920. The puliticll role ofm iza ziunrlt :hcrs lu ledo n.oo.nnmitica. 1
_l,,l, m, ee . i . r. u l 13c::
I ' 1 ,dup. Gt, n,ndi.t. 1995.1 J I I J I o r[1 l i i otm'i, 1 1 a co:ora 11 N 1¿ hldia. Del'ni: Ox
torcí Un overo ce' Preso. and ((Loo CustneL Nc,c York : 1 )]e el and e1111I
cepatriation ara
la;cnu, elado. 199;. Los fusiles de 1 [.ubre Loo los y : e.m u. militares sobre ,a recelaron guatunalt¢e Pr ahard, Dtnna . 1 995 Fhe Icgacy u( mnflicr-. Rc( gec
cal Amcriu- In . ;r; ,mar. n I' ra ; tLr. 1. R:¢hel Sir'I ; de s.ugsr k . hng,aod:
;qS5 SS V1exim Cas Inan Puhlos Ediror-
edvrser. 'oul 1926. Khau1 .aishca. or h,h n . 1' sn ,sao 1 eolcui d admtusmuion aud cul-
cora111 aura11 ab419 1. Prouwrion ut Nationa L utc znd Re. m.u en ion Act uc

Pele. Petera ;9q7. Thc antitropologs of coloiiiahs:u: Culunc, liisorc, ardí clic emcrgence ot \Nest- Prtnler . Prerori2.

eru pocero meo tality. Annual Reuiew ,fAnthtopohoyy 26: 163-83. Quintero, Luis, and Brika Silva. 1990. Ecuador. Una na-son no ciernes. Quilo: rLacss.

Perelli, C. 1993. The power of memoryand tire memory af power. In Repression, exile and dentonacy: Rabinow, k, ed.:984.'1he Foucault reader. ¡la nnondaworth, England: Penguin.
Radchffe, Sarah A. 1996. Imaginativo geographies, posccolontalism and nacional idendcies: Con-
Umguayan adture, ed, S. Sosnowski and L. Poplcin. Durham, NC: Dulce Universiry Preso.
Pelees, Mtchael. 0995. "Alter Auschwitz": Ethics and educational policy. Disrourse 16(2): 237- temporary discourses of che narion in Ecuador. Ecumene 3(1): 23-42.
-. 'qg8, Frontiers and popular nationhood: Geograp hies of identicies in che :995 Ecuador-
51.
Peuker[, Detlev. 1989. Inside Nazi Gerntan4. Harmondaworth, England: Penguin. lloro border dispute. Politiral Geoyraphy 17(3): 273-93.
1999. Popular and st-ate discourses of power. In Human geography today, ed. D. klacep',
. 1993- 111C genesis of tire `Final Solution" froto che splrit of science. In Reevaluatíng tire
Third Reich, ed. T. Childers and L Captan. New York: Holmes and Meier, J. Allen, and P. Sade. Cambridge, England: Polity Press,
Kadcliffe, Sarah A., arad Sallie Wesnvood- 0996. Re-making che nation: Place, politics and identtty :Ti
Philport, Sisen. 1997. Knowing Indonesia: 0rientalísm and che diseaurse of Indonesias politice. Ph.D.
doce,, Australian Nacional Universig, Latín Am ere m- Cundo o: Roudedge-

Pioaulr, David. )999. Muslim-Buddhisr relativos in e ritual amrexr: An analysis of che mulate- Radhu, Abdul ANahid.:981. Carevane tibétaine: Adapté n) frantais par Roger du Pasquier I,a Bibho-

ram procession in Lch Township, Ladalch. In Ladakh: Culture, hist:,4, and development between théque des Voyageurs. Paros: Fa,ard.

Htmalaya and Karakoram. Recent researrh Da Ladakh 8, ed. M. van Eeek, K. Bertelseu, and Ranson:, John. 0997. FouraeIt's discipline. Durham, NC: Dulce Universory Preso.
P. Pedersen. Aarhus, Denmade Aarhus Universiry Preso. Recinos, Adrian. 1913. Monografía del Departamento de Huehuaenango, República de Guatemala. Gua-

Pine, Lisa. 1997. Nazi fantily poliry 1933-45. Oxford: Berg. temala Ciry: Tipografm Sanchez and de Guise.

Pinnock, Don. 1984. Tic bmtherhoode Street gangs and cuate control in Cape Tawn. Cape Towa, Sourh Reno, W illianr. 1998. Warlord poluta and , ion stales- Soulder, CO: Lynne Ricnnec.
Africa: David Philip. Restrepo, M. 1993. El problema de la frontera en la consrruccifin del espacio amazónico. In

Plano, Roger. 1998. Ethniciry and che Guatemalan peace process: Conceptual and practica] chal- Amazonía: Escenarios y conflictos, ed. Luey Ruiz, Quito, Ecuador: Cedime.

lenges, In Guatemala after che peoce aaords, ed. Rachel Sieder. London: Institute of Latin Richards, Paul. 0996. Foghting for che ratnferes : War, youth, and resources in Sierra Leone. London: James

American Scudies. Corre y.

Paley, J. A., ed. 1985. Seleeted wrtttngs ora tic freedom charter 1955-1985. London: African Natlonal Risley, Herbert. 1969. (:915). The prople of India. Delhi: Orienml Books Reprinc Co.
Congreso-
Rizvi, Janec. 0994. l'he Trans-Karakoram nade in che none:eenth and twendech cencories. Indtan

- ed. 1988. The jieedom charter and the future. Johannesburg: A. D. Donker. Economtc and Social Hiseory Revtew 3:(1).
. 1997. l,eh co Yademd: Travelling che areno-Karakoram trade route. In Recent Recearch en
Police `sal oto' Nkabiude docket. 1999. Mail and Guardias, 20-26 August.
International Associatton for Ladalch Studies,
Pools, D., and G. Renique. 1991. The new chroniclers of Peru: U.S. scholars and rheir''Shining Ladalch 7: Proceedtngs of ihe 7th Colloqutum of tire
Bonn/SI. Augustos, :2-15 )uno 1995, cd, T. Dodin asid H. Rádaer. UIn1, Germany: Uniter-
Parlo" of peasant rebellion. Bulletin of Latín Amerimn Research m(1): 133-91-
-. 1992. Pera, time offear. London: Latin American Burcau. sitdt Ulm,
Porrocarrero. G. 1998. Razones de sangre: Aproximaciones a la violencia politím. Linfa: Fondo Editorial, . 1999.1 oreshíntalayan caravans: Merchant priores and peasant traders in Ladalch. Delhi: Oxford

Pontificia Universidad Católica del Pero. University Press.

Portoearrero G., and P. Oliarte. 1989. El Perú desde la escuela. Lima, Peru: Instituto de Apoyo Robins, S. 1998. Silence in my farhei s house: Memore, nationalism, and narracives of che body.

Agrario. In Negottating ihe past The making el memory in South Afrtm, ed. S. Nuaall and C. Eoerzec,

Poscl, D. 1990. The maletng of apartheid 1948-1061.Oxford: Clarendon Press. CapeFown, Sourh Africa: Oxford Universics' Press.
Ronsbo. H. 1997. State fornsation and property: Reflecrions ora che pahfical cechnologies of epace
Pratt. Gerr. 0992. Spatial menaphors and speaking positions. Environmene and Planntng D: Sodety
,red Space lo: 241-44. in Central Amcricc. 1Ournal of Historial Sociology :o(c J: 56-74.

406 B1 BLIOG R A P HY El BLIO G RAP HY 407

 Rose, Nikolas. 1985. The psyehological complex: Polítia, psychology and sociny in England, 1869-1939.
opment Services programme and profile of suggested indicators in the social components.
London: Routledge and Keegan Paul.
Indica joernal of Social Work 47(3): 303-13.
-. 1996. 'che death of the social? Re-figuring the territory of govemment, Emnomy and
1990. More rhan roo million women are missing. New York Reeíew of Books, 20 Deceniber.
Soeiety 25(3):327-56.
. 1994. Population: Delusion and realiry. New York Review of Books, 22 September.
Roseborry, William. x994. Hegemony and the language of conten don. In Eoeryday forros of Sta te
Serrano, F. 0993. Las organizaciones indígenas de la Amazonía ecuatoriana. In Amazonfa: escenarios
fornmtion: Revolution and the negotiation of rule in modere Mexico, ed. G. Joseph and D. Nugent.
y conflictos, ed. Lucy Ruiz. Quito, Ecuador: Cedime.
Durham, NC: Duke Universiry Press.
Sharina, Urmil. 1986. A critica) assessment of monitoring and evaluation in Integrated Child
. 1996. Hegemony, power, and languages ofcontention. In The politics of dii erence: Eflinic
Developmenr Services Programme sud profile of suggested indicators in the social com-
premisos in a world al pewer, ed. Edwin N. Wilmsen and Patrick McAllister. Chicago: Univer-
ponente. In The Indianjournal ofsocial work 47(3): 303-13.
sity ofChicago Press.
Show, Mark. 1997. South Africa: Crime in transition, in Terrorten asid political violente 8 (4): 156-
Rowe, William, sud Vivían Schelling. 1991, Memory and modernity: Popular culture in Latín America.
75.
London: Verso.
Shearing, Clifford, and Mike Brogden. 1993. Palidng for a new South Africa. London: Routledge.
SatT. Grant. agg8. Changíng Cape Town: Urban dynamics, policy, asid planntng during the political transi-
Sieder, Rachel. 1896. Derecho consuetudinario y transición demoadtim en Guatemala. Ciudad de Guate-
tion in South Africa Boston: Universiry Press ofAmerica,
mala, Guatemala: Facultad Latinoamericana de Ciencias Sociales.
Sahlins, Peter. ig8g. Beundaries: The makinq of Franco and Spain in the Pyrenees. Berkeley: Universiry
. ed. rgg8. Guatemala alter che peace acmrds. London: Instimte of Latín American Studies.
ofCalifornia Press.
Sieder, Rachel , and Jessica Wirchell. zoci. Advancing indigenous demande through che law: Re-
Said, Fdward. 1883. Orientalism. London: Verso.
fleetions on the Guatemalan peace process . In Culture and rights, ed. Jane Cowan, Mario
Sanchez, R. 1981. Toma de tierras u conciencia política campesina. Lima, Peru: Instituto de Estudios
Dentbour. and Richard Wilson. Cambridge, England: Cambridge Universiry Press.
Campesinos.
Sigley, Gary. 1996. Governing Chinese bodies: The significante ofscudies in the concept ofgov-
Santana, Robert, 1995. Ciudadanos en la etnicidadi Los indios en la política y la política de los indios.
ernmentality for the analysis of government in China. Economy and Society 25(4): 457-82.
Quito, Ecuador: Ahya-Pala.
Sinunel, Georg. 1950. The secret and tlte secret socicry. In The sociology of Georg Simmel, ed. K. II.
Santos, Boavenmra de Sousa. 1987. Law: A map of ntisreading. Toward a post-modern concep-
Woltt, New York: Free Press,
ti on of la w. journa 1 of Law and Son mi 14(3): 279-302.
Singh, Karan . 1982. Heir appareno An autobiography. Delhi: Oxford Universiry Press.
1995- Toward a new tommon sonso: Law, serenes and pelillos in the paradigmatie transition. New
Skinner, Quentin. 1985. Introduction: The rentrn of grand theory in tire human sciences. In The
York: Routledge,
roturo ef grand theory in the human sciences, ed. Q. Skinner. Cambridge, England: Cambridge
Sarakinsky, 1. 1988. A web en national manipulation. Unpublished manuscripc.
Universiry Press,
Sehjrf, wilfried. 1997. Specialist courea and communíty courts. Position paper submhted to the Mim
Slater, David. 0997. Re-thinking the spatialities of social movements: Quesrions of (b)orders in
istry ofJustlce, the Planning Unit, Instituto of Criminology, Cape Town Universiry.
global times. In Cultura of politia: Política ofculture, ed. S. Alvarez et al. Princeton. NJ: Prince-
Schirmer, Jennifer. 1gg8. The Guatemalan military projeer: A violente called democracy. Philadelplua:
ton Universiry Press.
Universiry of Pennsylvania Press.
Smith, Anthony . 1991. National identity , Harmondsworth, England: Penguin.
Schmidt, Lars-Hendk. 1993. Setting the values. Working paper no. 0. Anrlms, Denmark: Center for
Smith. Carol A., es. 19c o. Guatemalan Indian arad the state 1540-1988. Austin: University, ofTexas
Kulrurforskning.
Press.
Scott, James C. 1985, Weapons of the weak: Eoeryday fornis of resistance. New Haven: Yale Universiry
Smith, Gavin. 1989. Livelihood and resistance: Peasants and the politia of )and ti Peru. Berkeley: Univer-
Press.
siry of California Press.
. 1990. Damination and the art of resístante. New Hacen: Yale Universiry Press.
Sonntag, Selma. 1999. Auronomous comsils in India: Contesting the liberal nailon-srate. Alter-
-. 1994. Foreword. In Eoeryday forros of state formation: Reoolutian and the negoeiation of rulo ni
nativos 24(4): 415-34.
modere .Meneo, eal. G. Joseph and D. Nugent. Durham, NC: Duke Universiry Press.
South African Police Service (SAPS). 1997x, Police plan, 1996-97. Pretoria: SAPS.
-. 1998. Seeing like a surte: How eertain sobemos te improve che human mndítimi nave failed. New
. t997b. Status repon: The transformatien ofthe South Afrimn Police Sereice. Pretoria: SAPS.
Haven: Y.de Universiry Press.
Sparks, Allister. 1990. The mind of South .Africa. London: William 1 leineman.
. 1999. Foreword. In Elementary aspecto ofpeasant insurgency in colonial India, ed. Ranajit Guha.
Srikrishna, B. N. 1998. Report of che Sdkrishna Commissiou. Vols. 1 and 2. Mumhai, India: Pu nwani
Durham, NC: Duke Universiry Press.
and Vrijendra,
Selected writings en the freedom charter 1055-1985. 1985. London: African National
Congress. Starn, O. 1991. Missing che revolucion : Anthropologists and me war in Peru. Cultural Anthropology
Sen, Amartya. 1986. Critical assessment of monitoring and evaluation
in Integrated Child Devel- 6(0): 466-504.

408 BIBLIOGRAPHY
BIBLIOGRAPHY 409
 Cambridge, Fngland: Cambrid), L' nvesiiv Presa, Cenan ol 1 arel Antenem Smmrs.
Star.1u00..o ni l.mr C'.ollier. 1989. 111,pn; id piloso' 111 hC sludy afin le .Neu:directionsinle galnnthro-
+ niogy- Irhaca, NY': Co nell Univcrsim fresa. ;998. Counmr 11 1 surge n er stnt'R. Olio' l'CI'-Sendero 1 1111i11vo m!'.hcci •s u :0 Peru

tu, n hasu ; , Rodoltir. 19ph. Indigcnour righrs: Same cvuccpuml prohlems. Iu Cowm¢ting dt- 1y8o-1996.1(1:11 n uf Lun. Amc; n e¢md' r-h.-3S-59-

^;oo 1, t Wman nghl h .nsh:p. i 1 nn5 :r. La n n:r,u1 cci. k1,111eth Jelin aud Frie
lheisett Gano:. 999 un 1 h 11 i lrrirli tm 1-.d Re : 1 a zni i r ^.uP.
l lershb c,(. houlder C (J wcvmicw Pres..
:1f,±, lulo: a}teul I.per pr suud at elle Stre.ldnonal eontcruue Lite .RC ( u nn,m'-r
Socio, Sre, tuso. Populism n. Perca' fhe ernergenee fine mas `es a td tire poülirs of Sorial control. Madisnn:
ing che Past. 1r141utic . eniver,ity of the A1'itsvarersri nd, lotr. tneah erg.
Univcrsm ot WSCOnsin Preso.
Third Por sí tito Cape liar. te9,. L c,:cn , Mal: t' (uerd.:o;. 31
Srt;ppuml, 1 ton. 1992. Reyo a et' Life n t esa emanan ratuya ser. ,nenl ;n Mesom - Ph U- dios..
lbrpcc )oh, 2000.11 : r t-. ! le 1w,p01 .- Sn¢ 1 1 1e. :.:e les tt 1 des
r, ir ten hagan.
Uuivcr,
Lngland: Cambridge Utuver. iry Presa.
Igg9a . Pillines of displaccntent in Guatemala. Joumal el Hutorical So7olog9 12(9:
Irinquier, Roger.1964(1961) . Modem wafare: A Frend, uew of munl orhtaurg enty- Net 5orl:segar
54 -So.
frurh and Reconciliation Commission of South Africa 1986. Interim repon. Pretoria: Gov-
19996 . Repatriados and everyday formo oí trace forntafion in Guatemala . In lile atad of
che refugee cycle? Refugee r,patranco , and reeonstruction, ed. Richard Black and Khalid Koser. ernment Printer.

New York: Berghahn Books. ig98a. Final repto . Cape Tocan, South Africa: Jura and Co.
. zoilo. At the frontiers of che modero state in post-war Guatemala, In Anthropology , devel- a998b. ' Che South African Truth en(] Reconrilintion Commission Proposal fór Reporatien and Reha-

opment atad modernities : Exploring díscourses , contener-tendencias and violente , ed. Norman Long and bilítation . Pretoria: Government Printer.
, 2ooo.'1R6: Current status and statisti¢. llecember g, 0999. (TRC Mailing Lis'[. h¢pilwwsv.
Arturo Arce . London : Roudedge.
Stern, S., ed , igg8. Shining Path and othet paths: War and setieco in Peru, 1980 - 1995. Durham, NC: truth.org.za)

Duke Universiry Press. --. website , t998. CO, Noventber.

Stoler, Ann Laura. 1985, Rice and rho' education of desiro. Durham, NC : Duke Liuiversity Preso. . Tmth and Remncil iatí on Commission : l nfo rmation padc. Pretoria: G uvermne nt Pri mea.
the-way place . Prince-
Stoll, David. 1993. Betwem tuvo fires in tire leal atoas of Guatemala. New Yodo' Colombia Universiry 'Lsing Anna , L. 1993. In che reaten of che diamend queen : Marginality in en out-of-
ma, NI: Princemn Universiry Preso.
Press.
Tully, James, 1995. Shange lnultiplicity: Constitutionalism in on age of dwersitg Cambridge, Bngland:
Strudsholm , Jesper. 1997. Man lod en leder gel amok. Politihen , llecember (Copenhagen).
Cambridge Universiry Press.
. 1999 . Arrenrat pi en voldelig vendekábe . Politiken , January (Copenhagen).
Submission ro che TRC by M. F. W. de Klerk, leader of che Nacional Party . http:llwww. Tutu, D. ogg6a . Healing a nailon . Indcx sin Cersorsltip. s.

trutli.org. zalnp-truth. htinl. Accessed February 15, 1997, . [9966. Letter to che Sundag Times of South Africa , 4 Decentber.
New York: United
Sulmon . Denis. 1977 . Historia del movimiento obrero Peruano , 0890 - 1977. Lima ;'lárea. United Nations Developntent Prograni ( UNDP ). 2000. Human developntent repon .

Swilling, M. , and M. Phillips . 1988. Surte power in die og8os: Froni "total strategy " lo "counter- Nations.

revolutionary warfare. " In War and society : The militarisation of South Africa, ed. J. Cock and Uquillas, Jorge . 1993. Estructuración del espacio y actividad productiva indígena en la Amazonia

L. Nathan. Cape Tocan , South Africa : David Phillip. ecuatoriana , In Amazonia : Escenarios y conflictos, ed. Lucy Ruiz, Quito, Ecuador: Cedime.
. Austin: Univer-
Tambiah, Stanley 5. 1996. Levelling nowds, Berkeley: Universiry of California Preso. Urban , Greg, and Joel Sheraec, eds . 4990 . Nation-states and Indiano in Latin America

Taudon, S. N „ K. Ramechandran , and S . Bhamagar. 1981. Integrated Child Development Services siry of Texas Press.

in India: Objectives , organization and baseline survey of che project population. Indias Valverde, Mariana . igg6. "Despotism " and ethical governance . Emnomy and Snriety 25 (3):357-72-
van Beek, Mattiju. 1996. 1dentity fetishism and the ara of representation:'I'he long straggle for regional
Jaurnal of Medical Research 73: 374-84-
Tapia, C. 1997. Lasfuerzas armadas y Sendero Luminoso. Dos esarategiasy unJPnal . Lima, Peru: Instituto autonomy in Ladakh. Ph.D. diss.. Cornell Universiry, Irhaca, NY.
. 19972. Begond identity fetishísm: "Communai" confito in Ladakh and che Botito of autouamy.
de Estudios Peruanos.
Paper presented at che SSRC MacArthur Foundation Program (in Peace and Internarional
'l'aracena, Artrno . i 995. Nación y republica en Centroamérica ( 1821-1865 ). In Identidadesnacionales
Security in a Changing World, Comell Universiry, Ithaca, NY.
y estado moderno en Centmamérim , Arturo Taracene and Jean Piel, eds., San José: Cosco Rica:
19976. Contestcd classifications of people in Ladakh : An analysis of rho censos of
Editorial de la Universidad de Costa Rica.
'faussig, Michael . i9932. Maleficiare : Sra re ferishism . In Retichisn as cultural distourse , ed. E. Apter Kashmir, 0873 - 1941 . In ']'ibetao studies : Praceedings of tire sevenih seminer of the Internatienal

and W. Pietz . Ithaca, NY: Cornell Universit Preso. Associatien for Tsbetan Sludtes , Graz 1995, ed. ll. Kra es er, M . T. Much, E . Ste inkelln e r, and

1993 b. Mimesis ami altero ly : A particular histonl of the sonsea. London: Roudedge. H. Tausche . Vicnna: Verlag dar 0stereieh en Alcademie den Wi s s enscliaften,
- : 1998. True patriota: bustif'ing autonomy for Ladakh. 1limelayan Researdt Ealletin 18(1]'.
1997. The magia 0.f the orate New York: Roudedge.
helor, I.ewis. 1986. &mdits and politirs in Peru : Landlord and pesaste violeuce in Haalgnyoe igoo-agio. 35-45-

410 1318 L1 OG RAP HY B1 BLI OGRAP HY 411

 . 1990 , Hill councils , developmenr, and democracy :
Assumptions and experiences trozo
Ladakh. Alternativas 24(4): 435-59. 2oth internacional congress of che Latín American Studies Assoeiation , Guadelajara, Mex-
van der Veer , Petar. 1994 . Rels9ious nationalism : Hindus and Muslim, in India ico, eg April.
. Bed<eley: University of
California Press, Wattal, C. L. 1948. Buddhists in Kashmir . Maha-Bodhi , May-June: 181-84.
van Krieken, Robert 19g6 . Proto- Weber, Max . a968a. Bureaucraey. In From Max Weber : Essays in sociology
govemmentalizacion and che hisrorical formados oforganiza - , ed. H. H. Gerth and
tional su bjectivity. Economy and Sodety 25(2): 195-221. C. Wright Mills. New York: Oxford University Press.
Van Neiuwkoop , Manten. and Jorge Uquillas . x9686. Economy and saritty : An outline of interpretive sociology - Vol. 3, ed. G. Roth and
2000. Defining ethnodevelopnient in operational terma:
Leasons from cha Ecuador lndígenous asid 11,c-Eluadoratl Prop i os C. Wittich. New York: Redminister Press.
Derelopment Project . Washington.
Werbner, Pnina , and Helena Basa Igg8. Embodying charisma : Modernity ,
DC: World Bank. lorality and che perjórmance
of emotion in Sufi culis. London: Routledge.
Van Oss, Adrian C. 1986, Catholia mlonialism. A parish history ofGuatemala 1524-1821. Cambridge,
Whitten, Norma u , ed. 1981. Cultural eransfonnations and ettnicity
England: Cambridge Universiry Press. in rnodern Ecuador. Chicago: Uni-
Vanaik, Achín , 1997, Communalism contestad; Religion , modernity versity of Illinois Press.
and secularizatíon . New Delhi:
Will ke, Helio ut . 1992. trocla des Staaees: 6 ru ndli cien amar Seaatstheo tia polyzentricher Gesellcchaft
Vistaar. . Prank-
furr anm Main: Suhrkamp Verlag.
Vazquez. M. 1965. Educación rural en el valle del Callejón de Huaylas . Vicos . Lima: Instimro de Estudios
Wilson, Piona . 1982, Properry and ideology : A regional oligarchy in che Central
Peruanos. Andes in tire
Verdery, Katherine . 1996. What was socialism , nineteenth century. In Ecology and exchange in che Andes, ed. David Lebrerann . Cambridge,
and what ruines next? Pdnceton , NI: Princeron Univeo-
sity Press. England: Cambridge Universiry Press.
Verkaaik, Oskar. 1999. Incide tal citadel : Pun, violente , . 2000. Representing che srare? School and teacher in posr-Sendero Peru. Bulletin of Latin
and religiosa nationalism ir Hyderabad , Pakistan.
Ph.D. diss., Universiry of Amsterdam. American Research 19(3):1-16,
Wilson , Richard. 1981 . Machine guns asid mountain spirits: The
Vincent, Joan. lago. Anthropology and politice .- Visions, traditions , cultural effects of arate repres-
and orando . Tucson: University of
Arizona Press. sion among che Q'eqchi ' ofGuatemala . Critique of Anthropology 11(1): 33-61,

Visvanathan , Shiv. a988a , On lile annals of che laborarory atare, In Scienre . 1995a. Manufacturing legitimaey: The Truch and Reconciliadon Commission and rice
, hegemony ond otetence.-
A requiemfor modernity , ed. A, Nandy. Delhi: Oxford University Press. role of law. Indícator SA 13(1) .
. 1995 b. Mayar resurgenae in Guatemala : Q egchi experiences . Norman: University of Oklahoma
19886. The early years . In Poul play : Chronicles ofmrruption , 1974-1997 .
New Delhi: Banyan
Books. Press.

Visvanarhan , Shiv, asid Harsh Sethi, eds. 1998 . . 1996. The Siswe will not go away : The Tmth and Reconciliadon Commission. Human
Paul play: Chronídea of mrruptien 1947-97. Nov,
Delhi: Banyan Books. rights and cation-building in South Africa . Afrimn Studies 55(2): 1-20,

Wagner, P. 1994. A sociology of modernity : Liberep and discipline . Wolf, Erie R. 1966. Peamnts. Englewood Cliffs, NJ: Prentice-Hall,
London: Routledge.
Walt, B. J. van der. and T. van der Walt . 1986 . Die waarheids : . 1969. Peasant wars in che twentieth century . New York: Harper and Row.
En versoeningskornmissie . Potchefsrruom,
Woods, D. 1997. Can rice cillero ever find peace ?The Guardias, 1 Pebruary.
South Africa: Insdmre for Reformarional Studies.
Warikoo, Kulbushan . 1985. Ladakh: An entrepór of Indo-Central Asian trade during che Dogra Woodc, M. D, 1994. Developing a noticia ofvillages: Rural commuoiry as srate forman in Sr¡

rulo, South Asian Studies 20(2). Lanka. Critique of Anthropology 14(1): 77-95-

. 1990. l,adakh ' s rrade relations wirh Tibet under che Dogras. China
A word, a nudge, a drink, a gun . , . then dcath . 1999. Smlday Argus (Cape Town), 24-25 April.
Report z6 ( z): 137-43.
*World Bank . 1997. The state in a changing world . New York: Oxford Universiry Press.
. 1995. Gateway Co Central Asia : The transhimalayan trade ofLadakh , 1846 -
1947. In Recenr
Young, Crawford . 1994. The African colonial state in mmparatíve perspectiva
research en Ladakh 4 and 5 : Proreedings ofthe Fuerais asid Pifih Colloquía en Ladakh, ed . New Haven: Yale Univeo-
. H. Osmas-
siry Press.
ton asid p Denwood . Delhi: Motilal Ba narsi dass.
Young, J. E. 1902 . 'fhe counter- monlimenr : Memory against itselfin Germany
Warren, K. E. 1978, The symbalism of subordination: lidian identity in a Guatemalan tocan roday. Crítica) In-
. Austin:
quiry 18.
University of,l'exas Press.
Warren , U. B. 1988 . lndigenous mouements asid their critico Pan-Mayas actioism in Guatemala Zamose, Leon . 1994. Agrian pcetesr and tire Indian movement in che Ecuadorian highlands.
. Prince-
Latín American Research Review 21(3): 37-69 ,
ton, NJ: Pdnceton Universiry Press.
Watanabe , J. M. 1992. Mapa carota and savia in a changing world . Zartman, W., ed. 1194. Collapsed states.- The disíntagration and restoratíon of legitimare authority. Boul-
Austin: University ofTexas Press.
der, CO: Lynne Rienner.
. 1997. Forgi ng identities ir an emerging nation-state : Bureaucratic procederes , policireg ,
and pi nish-
ZiTek, Slavoj. ag8g. The sublime object of ídeology. London : Verso.
ment in Maya communities of late nineteenth - century western Guatemala . Paper presentad at che

412 BI B LIOGRA P HY
B1 B L 1 OG R APHY 413
CO NT RI LlUT O RS

11IOSMAS RLom IIANSeN is a reader in dhe Dcparmrcnt of Socal An[hropology at rhr University

ofEdinhurgh.

LARS BIT II E ie n researeh frllow al che Centre 10r Develo p ment Research 'Ti Copenhagen

MTTCnELI. DEAN is a profesor of Sociology Ti J4aequarl e Uniersip in Austmha.

A KI ¡TI, GU PrA is an asociare profesor in the Cepa erial en[ of Cntcnral and Social Audaropo]ogy

at Stanford Unive rsirp.

ST E EEEN 1 ENS EN is are search fellow ar [he Centre Loe Devcl opine nc ReseareII la Copenhagen.

ALFTTA 1. NORVAL le a senior lecturer in rhe Deparrmenr of Governmenr ac the Universiny of

Essex.

DAVID NUGENT is an associate profesor in rhe Department of Anthropology a[ Colby College.

SAICA1r A. RAROL[ FFE is a lecturer in the Deparnnenc oí Geography at the University of

Cambridge.

RACeEI, elED en is a senior Ieecurer ac rhe InsGm[e oE Larin American Studies ar rhe Unversity
of London.

FINN s rePEUTAT le a senior researcher at che Centre far Ceselopntent Research ni Copcn-
hagcn.

AlAEEl]N VAN EEEK ie an ascociste profesor in rhe Departmcnt of Ethnoe aphe in,¡ Social
Anthropology at Aarhus University in Denmarle.

Che Eacolw nfSocial Studies at Vrife University la Amsterdam.

OSKAR VERKAAIK is a IeCrorcron

1 IONA W t1.SON 15 a profesor at che Centre fiar Ceselopinenc Research in Copenhagen.

 - 1999. Hill councils, development, and democracy: Assuniptimrs and expericnces from
Ladakh- Alternativas 24(4); 435-59. 2oth internacional congreso of the Latin American Studies Association, Guadelajara, Mex-
ico, a9 April.
van dar Veer, Peter. 1904. Religious nationalism: Modus and Muslime in India. Berkeley: University of
California Preso. Wartal, C. L. 1948. Buddhists in Kashmir. Maha-Bodhi, May-lime: 181-84.
Weber, Max, tg68a, Bureaucracy. 1,1 Fra. Max Weber : Eseays in sociology, ed- H. H. Gerth and
van Krieken, Robert 1996. Proto-governmentalization and the historical formation oforganiza-
tional subjectiviry. Economy and Socíety 25(2): 195-221. C. Wright Mills, New Yode Oxford University Preso.

Van Neiuwkoop, Martien, and Jorge Uquillas- 2000. Dejining ethnodevelopmenf in operacional termo: . 1968h. Eronomy and sodrry: An andina al interpretíve sociology. Vol. 3, ed. G, Roth and
C. Wittich. New York: Bedminister Preso.
Lessonsfrom ihe Ecuador Indigeneus and Afro-Ecuadoran Peoples Development Prajeet. Washington,
DC: Wodd Bank. Werbner, Pnina, and Deleite Baso- 1998, Embodying charisma: Modernity, locality and the performance
of emotion in Sufí culto. London: Rourledge.
Van Oss, Adrian C. 1986. Catholic roionialism. A pariste history of Guatemala 524-r8z1. Cambridge.
England: Camhridge University Preso. Whitteu, Norman, ed. tg81. Cultural transformations and eehnicity in nrodern Ecuador, Chicago: Uni-
versity of Illinois Preso.
Vanaik, Achin. 1997. Communa lísm (mtestad: Religion, modernity and secularization. New Delhi:
Vis raar. Willlee, Helnmt. 1992. 1rmt ie des Staates: 13 rundlini en am ar Staatstheorie poluzcoen rher Gesellsrhafe. Franle-
furt am Main: Suhrkantp Verlag.
Vazquez. M. ag65. Educación rural en el valle del Callejón de Huaylas, Vicos. Lima: Instituto de Estudios
Peruanos. Wilson, Fiona. 1982. Propcrty ;md ideology: A regional oligarchy in che Central Andes in the
ninetcenth century. In Emlogy and exchange in the Andes, ed. David Lchmann. Cambridge,
Verdery, Katherine. rgg6. What seas sosialism, and what comes rece? Princemn. NJ: Princeton Univer-
sir, Preso. England: Cambridge University Peces.

Verkaalk, Oskar. 1999. Inside che cí ta deó Fun, violonce, and relig ites na tíonalisin in II yd era bad, Pakismo- . 2000, Represendng the state? School and teacher in post-Sendero Peru. Brdletín of Latin
Ph.D. dise'., University of Amsterdam.
Ameriran Reseurch 19(3): r-16.

Vincent, loan. 1g9o. Anthropology and politics: Visions, traditions, and trends. Tucson: University of Wilson, Richard. 1991. Machine guns and inountain spirits: The cultural effecis of orate repres-
sie n among the Q'egchi' of Guatemala. Critique of Anthropology 11(1): 33 -61.
Arizona Preso,

Visvanathan, Shiv. r988a. On the annals of che laboratory sute. In Science, hegemony and uiolenre: . 19951. Manufacturing legitimacy: The Truth and Reconciliation Commissioo and che
role of law. Indirator SA 13U)
A requiere for modernity, ed, A. Nandy. Delhi: Oxford University Preso,
resurgence in -.ig5hMan Guatemala: Qegchi experiences. Norman: University of Old:rhonta
-, ug88b, The early years. Di Leal play: Chronides ef conuption, 1974-1997. New Delhi: Banyan
Preso.
Books.
Visvanathan, Shiv, and Harsh Sethi, eds. 1998, Foul play: Chronides of corruption 1947-97. New . i996. The Siswe will nos go away: The Truth and Reconciliation Commission, Human

Delhi: Banyan Books. rights and nation-building in South Africa, Afrimn Studies 55(2): 1-20.
Wolf, Eric R. 1966- Peasants. Englcwood Cliffs, NJ: Prentice-Hall.
Wagner, ti 1994, A sociology of modernity: Liberty and discipline. London: Roudedge.
. 1969. Peasant wars in che twrntienh century. New York: Harper and Row.
Walt, B, 1, van der, and T. van der Walt. s9g6. Die waarheids: En verseeníngskommissie. Potehefstroom.
Woods, D. 1997, Can che killers evcr fiad peace? The Guardias, 1 Pebruary.
Sourh Africa: Instituto for Reformational Studies.
Woost, M. D. 1994. Devcloping a nation of villages: Rural community as serie formation in Sr¡
Warikoo, Kulbushan, 1985. Ladakh: An entrepót of Indo-Central Aojan trade during the Dogra
role. South Asían Studies zo(2). Lanka. Critique afAnthropology 14(13: 77-95.
A word, a nudge, a drink , a gen . , . then death. 1909. Sunday Argos (Cape Town), 24-25 April.
. 1ggo. Ladakh's trade relations with'tibet under die Dogras. China Report 26(2): 137-43.
World Bank. 1997. The orate in a dranging werld. New York: Oxford University Preso.
. 1995. Gateway ro Central Asia: The transhimalayan nade of Ladakh, 1846-1947. In Recent
resean'h ce Ladakh 4 citó 5: Proceedings of che Fourth and Fifth Colloquia in Ladakh, ed. H.Osmas- Young, Crawford. 1994. The Afrimn colonial state in comparatíve perspectioe. New Haven: Yale Univer-
sity Preso.
ron and P. Denwood. Delhi: Motilal Banarsidase.
Young, J. E. 1992. The counter-mouumenn Memory egainet itself in Germalry today, Critical ter
Warren, K. B. 1978. The symholism of submdinatiom Indian identity in a Guatemalan tosen. Austin:
quiry 18.
University ofTexas Press.
Zamose, Leon. 1994. Agradan protest and the lndian movemenr in the Ecuadorian highlands.
Warren, U. B. 1gg8. lndígenous movemenes and chau critico: Pan-Magan aetivism in Guatemala. I'rince-
Lacro Americon Researrh Reviese 21(3): 37-69.
ton, NI: Princeeon University Press.
Zartrnan, W., ed. 1994. Collapsed states: The disintegration and restoration of IegiGmate authmity. Boul-
Watanahe, J. M. 1992. Maya saints and solo in a changing world. Austin: University of'fexae Preso.
der, CO: Lynne Rienner.
. 1997, Forging tdentit:es in al emerging nation-state: Bureaucratic procedurea, policing, and puresh-
Zizek, Slavoj. 19,89. The sublime object of ideolagy. London: Verso.
ment in Maya communities of late nineeeeuth-century western Guatemala. Paper presenred at che

412 BLBLIOGRAPHY
BIBLIOGRAPHY 413
CON 1PIBUPORS

THOMAS CLON IIANSEN isa reader in tire Departen ent ofSocial AnthrOp 0lOge at lile Unloerslt'
of Edinburgh.

LAR S Ru U R is a researeh fcllow at the Centre for Developmenc Relea-.ch in Cope nha gen

.M ITC HE.1. DEÁN ie a profese 0U ot Sociology at Macga.arie Un iversi¡y in Australia.

AKHIL CUe'rA jean asociare profesor in che Departen enr ofCulmral and Social Anthropo lo p.

al Sranford University.

fellow at che Centre for Devclopmenc Restareis in Copenhagen.

STEFREN TENSEN is a research

in che Department of Goveen m ent al the University ot

ALci rA ]. NO RVA L ni a senior lecturer

Essex.

DAVID N U C R N T lean associate profesor in the Department of AnthroNology al Colby College.

SARAII A. RADCLIERE is a lecturer in che Depa¢mcnt of Geography at che Unnersity oí

Cambridgc.

RACHEL SiFUER is a senior lecturer atete InshRRe of Latin American Studies at che Universill

ofLondon.

FINN STE1pUTAT is a senior reecarcher at che Centre for Development Rescareh in Copen-
hagen.

MARTIIN VAN RFEK is in associate profesor in the Department of Ethnography and Social

Anth ropol ogy at Aarhus Un iversi ty in Den marle.

(SS KAR VERKAAIE ie a leetureron clic Eaeulty uf Social Smdiesar Vrijc Universiry in Amsterdam.

1 IONA WILSON S o a prof eeor ar che Centre for Dcvclopm<nl Reecarch in Copenhagen_
 INDEX

Abrams, PhiliP , 346-48

135, 297-98; and cducation , 328; and mili-
Abstracrions: of the popular , 24; of space, mry rule, 274-76, 327- 33, 363; and radical
131, 285, 289 ; ofsrate, 17 , 19, 262-63, 266, populist discourse , 274-75; cure of 327-
273-75, 281, 345-64 ; and violence, 385- 33; and state formation , 130-31, 288. Seo
86 aleo Insurgency ; National security
Accountability, 74-75 Authoritarianism: forms of mle , 51, 4611.5,
Administration : teehnology of, 7. See alto
71-72 274-76, 327-33: regimes , 46, 209,
Po pulatiou 290
Advocacy, 140 Authodry: to authorize , 8, 294; centers of, 8;
Affrmative action, r5on.2, 195 , 206, 352, 371, language of, 7, 8, 227; ofsmte, 226, 348; to
378 validare, 6 , 348-49
Agamben, Giorgio, 44 Autonomy: ofcommunity, 287-88; demande
ANC: and che apartheid state, 195 - 96; and for, 365-86 ; erosion of, 324-27; and neo-
local politics , foz-5; and the National liberalism, 286; ofsmte institutions , 11-12,
Democratic Revolution , 99; and nace, 183- 30, r59-60 , 163, 172-73
84, 1g5; and sute reform , zg, 97-122.
See aleo Apartheid; Political parties; South Biopolitics , 8; in China, 6o -61, 71 n7; in
Africa India, 65-94; internacional , 47; and non-
Anti-Communism , 208, 293 liberal rulo, 41- 63; and political econ-
Antistate movements , 23, 207 306, 319, omy, 48; and racism , 54-60. See algo
331-33 Govermnent
Apartheid: and Afrikaner nationalism, roo; Birth control. Ser Biopolirics ; Population
and counterinsurgency strategies , m1; and control
policing, 101-2; resistance te, 195; security Borders: history of, 131
apparanls , 97-98. See algo ANC Bourdieu , Pierre, 1. 5 , 6. 175n.42
Appadurai , Arjun, 385 Brokers, 26 , 294, 301, 305-6, 314, 318. Seo
Aristocratic role, 263 - 65. See algo Elites; algo Cliente ; Patronage
Sovereignty Eureaueracy : and categorization , 24,162,
Armed forces : counterinsurgency straregies,
166-69,171-72.369,371-79,382,383-
m1, 209, 287. 294 - 95; and development, 86; coping vith, 93 ; depoliticization of,
 tare workeru yo. S-alw Bu rc.racracp. oi ps -c.. :5,'y t: ovcrc,gnr;.
Pubh II s. 7S ;,i Levolutiou S Imurgcncy
5 ch ool tctcher
I ,"-.49
Pulla lctf rman S State -ap, i,de Rights: and itir n'hip, 5 .284: die se ot,
Snl te[iehisn;. 1S. ': 36S 559 a. ,a>lwh
Pubhr........ t I 12 3 22- - ,. 'S. 42. 204- 5, '12. and ti, Icmi-
o(nc^ sum
SI:,ic timnanon.-tgan;o JI.2,.,.. aol-
Puaht,p¢c t,S -.90.30rn. sr:.Gso Pn_
1
culonia l , r1- , Eandor 45'. in
guanntnr ot. ,8; violad .... s 149. t50, z , r-auto ourvcs of. '. mpoln_g05.
;3; iu ut I
l .gept. 6; cvcr 1 t f,rnm
Sc, talco Cirizcuship: Hunrm rights tumnlizatl: u .. 312-_-
o[v a85: anl ;dcr qr.; Cuu vdr- .c
I epresemmmons rditiunah n as;2: r_„
_o,-h, 2s54 uhin-uur tl tcom,,
Puhb hc vndmn x, S1, 1, cuq,t' m Rn-1 ,nit rota- ritu.Js ot Starc spcctadc

Pu len dort, S;un O Clson_ 1S Rumor '1 13, „356,360-62

313 44; and p ihnes, 302 tnd posiriv
Rural a11a1, 65-96.257-83 185-,12- 313- lcansition SeaSCttc retnrnt
idcologv, 127 rnd rcgiomd mtcgration
Race: and biopolitics , 54- 60; sud crime, u3; 44: elites in, 267 11 11 urna ti 0lral nenvorks, 12,5 t39 210
267-73; and space. 12345,284-312;
discourse uf, z63-64; and social order, Transparency, 172-73
theory of, z85-87 , 288-89, 308-10, 315-
314, 329; in South Africa, 99-toa Set alsu School tcacliers, 89-9 130, 1951 313-44. T7uth and Rcco ncilia rion Cam mis'sioq 18, 22,
19; in Venezuela, 14, zo; in Z i m b a b we, 20 -
Biopolitice; Governntentaliry See alca Education 108 . 149-81, 182-2021.. and categodzation,
21. See siso Nation-state
Rcason ofstate , 49 Scott, Jaines, 33
State idea: as a moral enrity, 242-43; pro fane 162, t66-69; creation ofan evil odrer, 152- 1
Recognition: of conimuniry , 84 , 297; ofiden- Shadow ecoriomies, 36 57; critique of, 183, 188n.33, 193-94; and
and sublime dimensions of, 225-26, 240-
tiry, 367; "tal- of, 21, 300, 365; ofstates, Social movcmente, 138-39, 141; and human data collection, 159-6o, 163-66, 233:
41, 242, 250-51; universalizztio u of, 6,
8,36 rights, 204-5: quest for citizenship, 267-
everyday pracrices in, 149-81; in Guau-
124-25
Reform: agrarian , 327-29; educational, 327- 73; and state geographies, 13q -42. Sec aleo
State policy. See Policy mala, 205; hcarings, 152-55, 190-91,
3: . See aleo Stare reforlIl Pohtical panties 230-36; in India, z26, 228-51; bcavecn
Sute racism, 57, 60. See cIso Racc
Re g ime . 23; apartheid , 99-102 , 195-96: au- South Africa: Afrikaner representations of, sute sud socicry, r59-6o, 163, 17z-73
Sute reform, 23, 97-122; m Guatemala, 203-
thoritarian , 46 , 274-76 , 327-33; dema- 152-57: crime, 97, 98; institution building
in, 109 -15; National Parry, 188- 89; negoti- 20; in Pakis un , 363; in Peru, 327-33; and
cratic , 10; neoliberal , z86; praeturian, 10. Unions, 304, 330
politice, z9; in Sourh Africa, 23, 24, 230-
See elsa Sute reform atcd settlement, tOS, 150, 162; sute reform
United Nations , taz, 307. See a150 Biopolitics
In, 97-122; Third Force, 115- t8; transition,
36, 97- 122. See also Reform
Regionalism , 135 , 257-83, 314-44. Sce aleo Urban elites , 351-52
185-86, 189-go; Truth and Reconciliation State socialism, 53, 6o
Provincc Urban governance , 246-49
State specucle, 15, 17, 19, 20, 37, 106, 118,
Registeis , 75-83, 300, 323-26; and gunder 93 Comnlission, 97-122, 109- 15. See aleo
229-30, 240, 334-41. See aleo Ritual:
Reification ofstate, 346-48 ANC; Apartheid; Tmth and Reconciliation Vagrancy laves, 285, 289, 29o
Syuabolism
Religion and religious movements, 292-93; Commission Village: as disciplinary space, z96
Sute systent after 1945, 10, 48
Buddhistpolities, 366-68, 377-82; Carho- Savereignry, 48; arismeratic, 263-66; democ-
State territory. See Te rrimry Violence, 35 - 36, 230 - 36, 306; colonial, 196-
lic Church , 288 , z9z-93; Hindus and Mus- ratization of, 49; discourses of, 48; and
State rransformation. See Sute reform sliolence i)
lime , 221-51: Islam and Muslim politics, hurtan rights, 48; language of, 55; and 97; communal , 221-51 ; elite, 264; ethnic,
Sutistics, 74-75, 84-88; and census ukers,
345 -64; Muslims and Buddlrists , 366-68. nonliberal role , 41- 63; popular, 263, 267- 385; foundational , 15; personified , 153--57;
86, 371-72; and data collection, 84, 324-
374; Shiv Sena and Hindu politics, 221- 73, 294; and terrirop; 8, 123, 126, 131; un-
25, 371-72; and local know l e d ge, 8 - 8 7: as a problem ofgovernance , 97, 1e8-9,
24 , 237-38. Ser aleo Communalism; Na- dermining of, 2, 127; in Western Europe, 306; sud che state , 54. 332 - 33. 356 . Sea siso
and resisunce, 87-88
tionalism; Politica1 organizations 49. See also Territorialiry Human rights
Stoler , Ann , 42, 54-55
Representations: ofahscract space, 131; ofthe Space: ofstate, 11, 123-45, 257-83, 285, Vulnerable group, g3
Suhaltern smdies, 24; and resisunce, 24
Afrilcaner , 152-57; ofapartheid state, 195- 315-19: and geopolitics, 129-32; and sci-
Subjeces: and eitizens, u, 208, 289; die pro-
g6; ofindigenous populations, 326; public. ence, 124. See also Privare space; Public
duction of, 66, 74, z86, zg6-97, 31S Warlords, 15, 36, 105
150 , 158 , 162; ofstate 90wer, 19-20, 221- space
Surveillance, 66, zg6: and carrographic work, Weak status, 1r
State capital, 5-6
54 , 275. 345-64. See aleo Myth of rhe state: 132; and gender, 66-95; and the Indian Webca 4$, 173
State specucle State collapse, 249-51. 269, 276, 347, 363-64 Wclfare, 298; programe, 65, 73; state, 45
sute, 83; limite to, 84, 87-88, 306-7
Resistance, 4: ambiguity of, 33: constinttive, Sute control: disputes over, z63-66; and
Survcy. Seo Statisties
World Bank, 1, 139-41
89: cdmography oF 66; Foucault and, 4 cducation, 322- 27
Svmbolism: ofsurhoriry, 6, 15, 333; ofnation,
32: and govern mentaliry, 69, 89; as reinter- Sute cmployees, 70, 81--83, 89-9r, 228, 272,
pretalion, 92: theory of. 32-34. See aleo 298, 314, 348-49. 3S3-55: aumnomy,
Popular imagination 348-49; geographers as. iz8; and volun-

INDEX 423
422 INDEX