Playing Different Games

INTEGRATION AND CONFLICT STUDIES

Max Planck Institute for Social Anthropology, Halle/Saale

Series Editor: Günther Schlee, Director at the Max Planck Institute for Social
Anthropology

Editorial Board: John Eidson (Max Planck Institute for Social Anthropology), Peter
Finke (University of Zurich), Joachim Görlich (Max Planck Institute for Social
Anthropology), Jacqueline Knörr (Max Planck Institute for Social Anthropology),
Bettina Mann (Max Planck Institute for Social Anthropology)

Assisted by: Cornelia Schnepel and Viktoria Zeng (Max Planck Institute for Social
Anthropology)

Volume 1
How Enemies are Made – Towards a Theory of Ethnic and Religious Conflicts
Günther Schlee

Volume 2
Changing Identifications and Alliances in North-East Africa
Vol.I: Ethiopia and Kenya
Edited by Günther Schlee and Elizabeth E. Watson

Volume 3
Changing Identifications and Alliances in North-East Africa
Vol.II: Sudan, Uganda and the Ethiopia-Sudan Borderlands
Edited by Günther Schlee and Elizabeth E. Watson

Volume 4
Playing Different Games: The Paradox of Anywaa and Nuer
Identification Strategies in the Gambella Region, Ethiopia
Dereje Feyissa
 Playing Different Games
The Paradox of Anywaa and Nuer Identification
Strategies in the Gambella Region, Ethiopia

Dereje Feyissa

Berghahn Books
New York • Oxford
 First published in 2011 by
Berghahn Books
www.berghahnbooks.com

©2011 Dereje Feyissa

All rights reserved. Except for the quotation of short passages

for the purposes of criticism and review, no part of this book
may be reproduced in any form or by any means, electronic or
mechanical, including photocopying, recording, or any information
storage and retrieval system now known or to be invented,
without written permission of the publisher.

Library of Congress Cataloging-in-Publication Data

Feyissa, Dereje.
Playing different games : the paradox of Anywaa and Nuer identification strate-
gies in the Gambella region, Ethiopia / Dereje Feyissa.
p. cm. – (Integration and conflict studies ; v. 4)
Includes bibliographical references and index.
ISBN 978-0-85745-088-3 (hardback : alk. paper)
1. Anuak (African people)–Ethiopia–Gambela Astedader Akababi–Ethnic
identity. 2. Nuer (African people)–Ethiopia–Gambela Astedader Akababi–Ethnic
identity. 3. Ethnicity–Ethiopia–Gambela Astedader Akababi. 4. Gambela
Astedader Akababi (Ethiopia)–Ethnic relations. I. Title.
GN652.E75F49 2011
305.896'509633–dc23
2011018006

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Printed on acid-free paper

ISBN 978-0-85745-088-3 (hardback)

E-ISBN 978-0-85745-089-0
 Dedicated to
my wife meron zeleke and our son naol
 Contents
List of Figures and Tables ix
Preface xi
Acknowledgements xv
List of Acronyms xvi

Introduction The Regional Setting of Ethnic Identification and

Ethnic Conflict 1

Part I Theory and Methodology 9

Chapter 1 Theoretical Orientation and Arguments 11

Part II The Contrast 29

Chapter 2 The Anywaa Primordialist Ethnic Identity Formation 31
Chapter 3 The Nuer Constructivist Ethnic Identity Formation 53

Part III The Encounter 75

Chapter 4 In the Riverine Lands 77
Chapter 5 The Cultural Contestation 95
Chapter 6 Differential Incorporation into the Ethiopian State 119
Chapter 7 The Anywaa Response to Ethiopian Ethnic Federalism 145
Chapter 8 The Nuer Response to Ethiopian Ethnic Federalism 167
Chapter 9 Civil War in the Sudan and Ethnic Processes in the
Gambella Region 193

Conclusion Modes of Ethnic Identification 211

Glossary of Local Terms 217

References 223
Index 233
 List of Figures and Tables
Figures
1.1 Districts of Gambella 3
1.2 Ethnoregional states of the FDRE 5
2.1 Routes of Lwoo migrations 33
2.2 Anywaa woman with dimui necklace 37
2.3 Nyinya Adongo’s coronation at Utalo village, 28 March 2001 43
3.1 Kir’s shrine where Mut Wiu is kept, Dorong village, Itang 58
3.2 The Jikany Nuer divisions 61
3.3 The Cieng Buoy in the Thiang genealogical structure 67
3.4 A wut – Ochom village 70
5.1 Jingmir Anywaa within the Gaat-Guang Nuer genealogical structure 98
8.1 To whom does the land belong? 182
8.2 Bil Puk – a national hero 186

Tables
1.1 Demographic distributions in the Gambella region 2
1.2 Refugee camp population in Gambella in 2002 8
2.1 Anywaa villages and clan distribution along the Baro River 40
4.1 Land size and population density of districts 82
4.2 Distribution of animal population by district 83
7.1 Political representation in the GPNRS Regional Council, 1991–2000 148
7.2 Allocations of ministerial posts in the GPNRS 148
7.3 Ethnic profile of civil servants in the GPNRS in 2000 148
7.4 1984 Census result of Illubabor province 151
 Preface
In the years to come, there will be no Anywaa left in the Gambella region.
Nasser was ours, the Nuer took it. Jikaw and Akobo were ours, again the
Nuer took them. Wherever we go, they will follow us. They take our land,
they take our rivers, and they take our people. Now they want to take
Gambella town. Where else should we go? We should stop them doing so.
The Nuer behave like that because they think that they are many and because
it is their nature to be aggressive. For us the word Nuer means something bad.
(Anywaa civil servant, Gambella town, June 2000)

Why should the Anywaa deny us access to land that they do not use? Why
do they think the land belongs to them? Land is for Kuoth [God]. Land is for
kume [government]. Because kume is not coming to us, we go where there is
kume. The Anywaa cannot stop us. If they say that we are Sudanese, let them
go back to Malakal, where they originally came from. The Anywaa are not
good people. They want to live alone. They kill you even if you are related,
bär cie wat [the Anywaa cannot be relatives]. (Nuer elder, Jikaw district,
September 2000)

This book explores ethnic phenomena through a comparative study of two ethnic
groups, the Anywaa and the Nuer. It examines two interrelated issues: the varying
configurations of ethnic identity and the causes of ethnic conflict. The geographic
locus of the study is the Gambella region in western Ethiopia, one of the main areas
of interaction between the Anywaa and the Nuer; but the issues raised, the actors
involved and the social space within which the ethnic process occurs extend beyond
the Ethiopian state and are intimately related to the political process in neighbour-
ing countries, particularly in southern Sudan.
There is a fair amount of confusion about how properly to name the Anywaa.
Most authors have followed Evans-Pritchard in spelling the name of both the people
and the language ‘Anuak’, with a final velar stop. Other terms used are ‘Anywaah’
(Crazzolara 1950), ‘Anyuaa’ (Bahru 1976), ‘Anyuak’ (Perner 1994; Hutchinson 2000)
and ‘Anyua’ (Reh 1996). In this book I have spelled it ‘Anywaa’, which sounds closer
to the way the people call themselves. Kurimoto (1992) has used a similar spelling.
The Anywaa and the Nuer are two of anthropology’s ‘classic’ people, made
famous in Evans-Pritchard’s Nuer trilogy (The Nuer, 1940a; Kinship and Marriage
among the Nuer, 1951a; Nuer Religion, 1956), in Evans-Pritchard’s book on the
Anywaa nobles (1940b) and in Godfrey Lienhardt’s articles on the Anywaa village
headmen (1957, 1958). The Nuer were revisited by Douglas Johnson and Sharon
Hutchinson in the 1980s, which resulted in the seminal books, Nuer Prophets (1994)
and Nuer Dilemmas (1996), respectively. All of these studies were carried out in
xii Preface

southern Sudan among either the Anywaa or the Nuer. To date, there has been no
study that focuses on the inter-ethnic relations between the Anywaa and the Nuer in
general and in the Gambella region in particular. Kurimoto wrote a book (in
Japanese) and numerous articles on the subsistence economy and politics of the
Anywaa in Ethiopia. Although I carried out fieldwork among the Anywaa and the
Nuer in southern Sudan for a short period of time, my study is focused on the
Anywaa and the Nuer in the Gambella region of western Ethiopia, where my field-
work was extensive.
The study upon which this book is based was initiated in the context of the
explosive emergence of ethnicity as a focus of identification and political action
worldwide – a context in which the escalation of ethnic conflict in Ethiopia repre-
sented one variant among many others. Gambella is one of the most conflict-ridden
regions in Ethiopia. The most protracted conflict in contemporary Gambella is
between the Anywaa and the Nuer, who are caught in a deadly struggle to determine
their political futures. Although there have been elements of reciprocity and comple-
mentary socio-economic exchanges, the dominant pattern of inter-ethnic relations is
conflict. In interpersonal and intergroup relations, friends and foes are represented
in ethnic terms, and tension and violence are expressed in various fields of social
interaction: from villages to churches, from schools to political parties. In cities,
inter-ethnic hostility has resulted in segregated ethnic neighbourhoods. The manifes-
tation of violence ranges from the complete destruction of villages to rioting in the
schools; from the targeting of minors and the raiding of public transport to the
burning in effigy of individuals to symbolize group humiliation. In recent times, the
conflict has assumed a more violent form, involving bombings, massacres and the
circulation of inflammatory so-called ‘confidential papers’ that tend to fuel ethnic
animosity. The intensity and magnitude of the violence lends an existential quality
to the conflict. It also involves mutual stereotyping that is premised on different con-
ceptions of the world in which one’s sense of justice is interpreted as evidence of the
moral corruption of the other.
On the basis of ethnographic and historical data collected through systematic
fieldwork and archival research in the Gambella region and neighbouring southern
Sudan, this book examines the contrasting modes of ethnic identity formation
among the Anywaa and the Nuer – modes which I call, respectively, primordialist
and constructivist; and it explains the causes of conflict between the Anywaa and the
Nuer as it is acted out in various domains of social life, becoming increasingly
violent. The conflict is also explained with reference to ‘significant others’ in
Anywaa–Nuer relations, including especially the Ethiopian state and the
Highlanders, the category of people most closely associated with it. The book
addresses the following questions: what are the conditions for the emergence of the
contrasting modes of ethnic identity formations? To what extent is this contrast acted
out in ethnic conflict? Why, among alternative and coexisting units of identification,
has the social struggle found expression in ethnic terms? For those who profess such
identities and invest in them, what does it mean to be an Anywaa or a Nuer? Who
entertains what kind of fears, and how do such fears enter into the definition of the
conflict situation, as, for example, in the opening quotation, in which an educated
 Preface xiii

Anywaa speaks of the extinction of his people? Considering that a similar discourse
about ethnic extinction was expressed by various travellers who visited the Anywaa
in the late nineteenth century and by historians of Nilotic peoples, do Anywaa fears
of extinction have a real basis? Does the statement made by the Nuer elder indicate
changing power relations between the Anywaa and the Nuer? In addressing these
questions, this book describes and analyses the multiple concerns of the Anywaa and
the Nuer in different social fields in which interactions are increasingly competitive
and in which the power of one provokes fear in the other.
The book is organized into three parts and nine chapters, preceded by an
Introduction that presents an overview of the location, demographics, historical
background and political context of the study area. The single chapter of Part I out-
lines the theoretical orientations and major arguments and describes the research
methodology and field experiences. Part II, consisting of two chapters, depicts and
analyses Anywaa and Nuer varieties of ethnicity, arguing that they are not simply
similar forms with variable contents but two fundamentally different ways of con-
ceptualizing, representing and practicing ethnicity. Chapter 2 describes and analyses
the Anywaa ethnic identity formation and its primordial configuration. The chapter
also discusses new socio-political processes that have reinforced Anywaa’s primordial-
ist conception of ethnic identity. Chapter 3 describes and analyses the Nuer form of
ethnic identity formation and its constructivist configuration. The chapter also dis-
cusses the incipient forms of a primordialist reconfiguration of Nuer ethnic identity
formation in new socio-political contexts.
Part III describes and analyses the encounter between the Anywaa and the Nuer
in various domains of social life, an encounter which not only has reinforced the two
contrasting modes of identity formation but also has resulted in violent conflicts.
The more the Nuer elaborate on their constructivist approach to their own ethnic
identity, the more primordialist the orientation of the Anywaa becomes. In the
remaining six chapters, the conflict between the Anywaa and the Nuer is discussed
in terms of three interacting variables: competition over resources, struggles for cul-
tural identity, and fluctuating power relations.
Chapter 4 examines the resource dimension in the conflict between the Anywaa
and the Nuer. It argues that competition for natural resources is relevant in
Anywaa–Nuer conflicts, but that scarce resources are not in themselves sufficient
factors to cause ethnic conflict. The ethnic framing of the resource conflicts is dis-
cussed with reference to the contrasting modes of ethnic identity formation and the
new political context of ethnic federalism in Ethiopia. Chapter 5 examines the iden-
tity dimension of the conflict and the cultural contestation that follows. It describes
and analyses processes of ethnic conversion – how in some areas Anywaa have
become Nuer – and the different ways in which most Anywaa respond to the process
of ethnic conversion.
Chapters 6 to 9 follow the fluctuating power relations between the Anywaa and
the Nuer in the context of the changing policies of the Ethiopian and Sudanese
states. Chapter 6 discusses the differential modes of incorporation of the Anywaa and
the Nuer into the Ethiopian state system during two political regimes – imperial
(1898–1974) and socialist (1974–91) – and their impact on local identification
xiv Preface

processes. In this chapter, I argue that the Anywaa were integrated into the state
system in a way which caused their economic marginalization, political decline and
social dislocation. Meanwhile, the Nuer ‘benefited’ from being neglected by the
state. This political process has generated an Anywaa narrative of loss which has, in
turn, reinforced their primordial form of ethnic boundary making.
Chapter 7 discusses the Anywaa response to ethnic federalism in Ethiopia since
1991, specifically, how the Anywaa have sought to contain Nuer expansion by using
administrative power in the new regional state of Gambella. To that end, they have
employed various strategies to gain political power: contributing to regime change,
claiming ‘indigenous’ status on the basis of settlement history, and framing ethnopoli-
tics in national terms. The chapter shows how the Anywaa’s radical notion of territori-
ality emerged in the context of Ethiopia’s unique experiment in ethnic federalism, which
has offered new incentives to invest in primordialist boundary making.
Chapter 8 discusses the Nuer response to ethnic federalism – how the Nuer have
contested Anywaa political dominance in the Gambella regional state in the 1990s
through counter-narratives and creative ideologies of entitlement. Nuer politics of
inclusion have drawn on various sources of legitimation, including the use of
mythology as an ideological resource, invocation of a longer historical frame of ref-
erence, census-based claims to ‘ethnic majority’ status, and reframing ethnopolitics
in national terms. Nuer political strategies vis-à-vis the Ethiopian state are embedded
in their cultural world. They draw on their constructivist concept of identity in relat-
ing to the state, for example, by projecting their flexible concept of localization onto
the state concept of citizenship.
Chapter 9 describes and analyses Anywaa–Nuer conflicts in the context of the
Sudanese civil wars. This issue is explored in terms of the following topics: changes
in the regional demographic structure; the differential access that the Anywaa and
Nuer have to the military power of the various rebel groups and to the government
of the Sudan; their differential access to resources delivered by aid agencies; and the
link between the experience of civil war and the reconfiguration of identity concepts.
The concluding chapter recapitulates key aspects of the two central issues
addressed in this book: the conditions for the particular (re)configuration of ethnic
identities and the causes of ethnic conflict. From an ethnographic perspective, it
becomes clear that ethnicity and ethnic conflict do not correspond to media images
of African ‘tribal violence’ and cannot be reduced to simple reflections of local eco-
nomic and political changes. On the contrary, this book demonstrates how the exist-
ing ethnic attributions (‘Anywaa’ and ‘Nuer’) are transformed and consolidated in
the context of translocal economic and political changes.
 Acknowledgements
I am grateful to the Max Planck Institute for Social Anthropology in Halle/Saale,
Germany, for providing the intellectual climate in which this book could be pro-
duced. On a personal note, I would like to thank Günther Schlee especially. As a
member of the Department of Integration and Conflict at the Max Planck Institute
for Social Anthropology, which Schlee heads, I have gained valuable insights that I
have applied extensively throughout the book. Schlee also visited me in the field
while I was doing the research upon which this book is based. One of the research
tools that I used – the village census – was inspired by this field visit. Thanks also go
to John Eidson, who contributed in many ways to the successful completion of the
book, particularly in the area of refining and smoothing the flow of argument, as well
as providing extensive editorial support.
I also thank Richard Rottenburg, Professor at the Institute for Social Anthropology,
Martin Luther University Halle-Wittenberg in Halle (Saale). I benefited greatly from
the postgraduate colloquium that he organized and also from the thoughtful discus-
sions that I had with him. Some of the insights from these discussions were used to
develop my arguments, particularly in my attempt to make sense of the dynamic inter-
action between manifest and latent functions of social action.
I am indebted to Boris Nieswand, a colleague and a friend at the Max Planck
Institute, with whom and from whom I have learned a lot. The stimulating discus-
sions I had with him, and his keen interest in the research project, sustained me
during the extended process of undertaking the research for and writing this book.
Among experts in the field area, those to whom I owe special thanks include,
above all, Eisei Kurimoto, Sharon Hutchinson and Douglas Johnson. I am grateful
to them for their encouragement and for their generosity in sharing with me their
ideas, observations and relevant fieldnotes.
I would like to extend my heartfelt gratitude to all residents of Gambella who
shared with me their views, concerns and worries. Special thanks go to my research
assistants: Peter Lual, Uchan Kayier, Okello Akuway, Maria Nikolai, Agape Nikolai,
Okello Awiti, Oman Agwa, Simon Shenkuoth, Chuol Gew, Chuol Ruey, Luk Kuey,
Amanuel Bahru and Sisay Tilahun, all of whom have actively contributed to this
book in various ways. May I also take this opportunity to thank my wife, Meron
Zeleke, and Ashebir Mesfin for their encouragement and editorial support. Finally
my deepest gratitude goes to my father, Feyissa Dori, and my mother, Askale Wechu,
whose unfailing support and keen interest in my academic development have always
been a source of inspiration.
 List of Acronyms
ACANA Anywaa Community Association in North Africa
ACORD Agency for Cooperation and Research in Development
Anyanya First southern Sudanese liberation movement (1960–72)
Anyanya II Second southern Sudanese liberation movement (1978–89)
APM American Presbyterian Mission
BGPNRS Benishangul-Gumuz Peoples National Regional State
EGBS Eastern Gambella Bethel Synod
ELF Eritrean Liberation Front
EPLF Eritrean Peoples Liberation Front
EPRDF Ethiopian Peoples Revolutionary Democratic Front
ETAM Ethnic Transmission and Acquisition Model
FDRE Federal Democratic Republic of Ethiopia
GLF Gambella Liberation Front
GPLF Gambella Peoples Liberation Front
GPDC Gambella Peoples Democratic Congress
GPDF Gambella Peoples Democratic Front
GPDM Gambella Peoples Democratic Movement
GPDU Gambella Peoples Democratic Union
GPDUP Gambella Peoples Democratic Unity Party
GPLM Gambella Peoples Liberation Movement
GPLP Gambella Peoples Liberation Party
GPNRS Gambella Peoples National Regional State
MPDO Majangir Peoples Democratic Organization
NGO Nongovernmental organization
NIF National Islamic Front
NPDO Nuer Peoples Democratic Organization
OETA Occupied Enemy Territory Administration
OLF Oromo Liberation Front
PDRE Peoples Democratic Republic of Ethiopia
REYA Revolutionary Ethiopia Youth Association
RPG Rocket-propelled grenades
SNNPRS Southern Nations, Nationalities and Peoples Regional State
SPLA Sudan Peoples Liberation Army
SSIM Southern Sudanese Independence Movement
TPLF Tigrean Peoples Liberation Front
UNHCR United Nations High Commissioner for Refugees
UNICEF United Nations Educational Fund
WPE Workers Party of Ethiopia
 Introduction

The Regional Setting of Ethnic

Identification and Ethnic
Conflict
Location and Demography
This study of the varying ways in which the Anywaa and the Nuer conceive of their
own ethnic identity and of the causes of conflict between these two groups is set in the
Gambella region, located in western Ethiopia about 780 kilometres from the national
capital, Addis Ababa. Currently, this region covers 34,063 square kilometres, and con-
sists of nine weredas (hereafter, districts):1 Gambella, Itang, Jikaw, Akobo, Abobo, Gog,
Jor, Godere and Dimma.2 Two features stand out in defining Gambella, not only as a
physical space but also as a socio-political unit. First, Gambella is one of the hottest
lowlands in the country, having an average temperature of 37 degrees Celsius at an alti-
tude of only 500 metres above sea level – in contrast to the neighbouring highland
regions, which rise as high as 3,000 metres (Ellman 1972: 2). Second, Gambella is a
peripheral region, situated along Ethiopia’s long international border with the Sudan.
These two features explain in part Gambella’s socio-economic marginality and politi-
cal sensitivity. Gambella is one of Ethiopia’s poorest regions in terms of infrastructure
and social services. Because of its location along the border with the Sudan, Gambella
is also susceptible to wider geopolitical processes. In fact, identification processes in the
region are intimately related to the civil wars in southern Sudan.
The population of Gambella has been variously estimated. The results of the 1994
national census indicated that it had at that time 181,862 inhabitants, more than eighty-
five per cent of whom lived in rural areas.3 Aside from recent arrivals from elsewhere in
Ethiopia, members of five ethnic groups – referred to, officially, as ‘national minorities’ –
coexist in Gambella: the Anywaa, the Nuer, the Majangir, the Opo and the Komo. The
Anywaa and the Nuer speak languages belonging to the Nilotic language family, whereas
the languages of the Majangir, the Opo and the Komo belong to the Koman language
group (Bender et al. 1976). In recent decades, these populations have been supplemented

1. Cf. Bureau of Planning and Economic Development (2000), Conservation Strategy of the
Gambella Region. Gambella: Gambella Peoples National Regional State.
2. The term ‘Gambella’ stands for three things: Gambella region, Gambella district and Gambella
town. Throughout the remainder of the book, if not specified as the district or town, the word
Gambella stands for the region.
3. The 2007 census results have not yet been made official but a preliminary report suggests that
the Gambella region had at the time of the census a total population of 306,916.
2 Introduction

by internal migrants from the Ethiopian highlands and by refugees from the Sudanese
civil wars. The internal migrants refer to themselves collectively with the generic term
degegna (Highlanders), indicating their places of origin, or with the more prestigious term
habesha, a cultural identity associated with the Ethiopian Orthodox Church. The major-
ity of the Highlanders came to the Gambella region in the mid-1980s as part of the gov-
ernment’s policy of resettling famine-affected people from the northern and southern
highlands to the western lowlands. Most are ethnic Amhara, Oromo or Tigreans, but they
also include a variety of ethnic groups from southern Ethiopia.
Beginning in the 1960s but with much greater intensity since the late 1980s, the
number of Nuer in Gambella has increased due to the arrival of refugees from south-
ern Sudan. For the same reason, the Anywaa population has also risen but to a much
lesser degree. The consequences of these processes is made evident in the census of
1994, according to which the Nuer constitute forty per cent of Gambella’s popula-
tion, the Anywaa twenty-seven per cent, the Majangir six per cent, and the Komo
and the Opo, taken together, three per cent. The same census identified twenty-four
per cent of Gabella’s population as Highlanders.4

Table 1.1 Demographic distributions in the Gambella region

Group Urban Per cent Rural Per cent Total Per cent
population of urban population of rural of total
population population population
National Minorities
Anywaa 9,831 36 34,750 26 44,581 27
Nuer 3,014 11 61,459 45 64,473 40
Majangir 64 0 9,286 7 9,350 6
Opo and Komo 1,067 4 3,735 3 4,802 3

Internal Migrants (Highlanders)

Amhara 4,639 17 7,927 6 12,566 8
Southerners 1,334 5 12,170 9 13,504 8
Oromo 5,890 22 4,635 3 10,525 6
Tigrean 1,341 5 1,255 1 2,596 2
Total 27,180 100 135,217 100 162,397 100

(Source: The 1994 Population and Housing Census of Ethiopia. Results for Gambella Region. Addis
Ababa: Central Statistics Authority, 1995.)

The various groups of people in Gambella pursue different livelihood strategies: the
Anywaa, the Opo and the Komo are predominantly cultivators; the Nuer are agropas-
toralists; and the Majangir combine hunting and gathering with shifting cultivation
(Stauder 1971). In the villages, the Highlanders make a living as cultivators, and in
the towns they comprise the majority of the traders and civil servants. Patterns of reli-
gious affiliation also seem to reinforce social boundaries between groups. Thus, most

4. According to the preliminary report on the results of the 2007 census, the Nuer constitute
forty-six per cent and the Anywaa constitute twenty-one per cent of Gambella’s population.
 Introduction 3

of the Highlanders are followers of the Ethiopian Orthodox Church, while the
Anywaa and the Nuer are members of various Protestant denominations, principally
the Presbyterian Church, or they are traditional believers.
In Gambella, the majority of the Anywaa live along the major tributaries of the Sobat
River: the Baro, the Gilo, the Akobo, the Alwero and the Pibor. In the current administra-
tive structure of the region, the Anywaa live in eight of the nine districts, the exception being
Godere; and they constitute the majority in five of those districts (Abobo, Gog, Dimma,
Gambella and Jor) as well as occupying the largest part of the mixed-settlement district of
Itang. The Anywaa also live in pockets of settlements in Jikaw and Akobo districts. Most
Nuer, on the other hand, live in the two outlying districts of Jikaw and Akobo. More cru-
cially, the representation of the Anywaa and the Nuer in Gambella town shows a great dis-
parity: thirty-six per cent and eleven per cent, respectively. Most of the Majangir inhabit one
district, Godere, although some of them are dispersed among the Anywaa in Abobo and
Gambella districts as well. All of the Opo live in Itang district, whereas pockets of Komo
settlements are found in Gambella district. The Highlanders live predominantly in three
districts: in Godere district they live together with the Majangir; in Abobo district they live
together with the Anywaa; and they constitute the majority in Gambella town.

Historical Background and Political Context

Gambella was incorporated into the Ethiopian state at the end of the nineteenth
century. The current international boundary between Ethiopia and the Sudan was set

Figure 1.1 Districts of Gambella

4 Introduction

by the Anglo-Ethiopian Boundary Agreement of 1902 (Bahru 1976). This agreement

also gave the British an important economic concession to establish a commercial
enclave in Gambella town on the Baro River. The British claimed sovereignty over the
western half of the town, while the eastern half remained under Ethiopian sovereignty.
Gambella remained an economic hub of the country, serving as an important station
in Ethiopia’s international trade via the Sudan for the first three decades of the twenti-
eth century (Bahru 1987). According to the terms of the concession, the enclave was
transferred to the Ethiopian government upon the independence of the Sudan in 1956.
The security situation in southern Sudan in the 1960s and 1970s undermined the
viability of the trade route passing through Gambella. The town lost its economic signif-
icance as trade in the former enclave declined because of competition with the eastern
railway route through Djibouti, and trade finally ceased because of civil war in the Sudan.
Gambella regained its strategic importance for the Ethiopian state in the context
of the geopolitics of the Horn of Africa in the 1980s. After 1974, when a military-
socialist government (the Derg) replaced the imperial Ethiopian government, inter-
state relations between Ethiopia and the Sudan became tense. By the early 1980s,
both countries were actively supporting each other’s rebels: the government of the
Sudan supported the Eritrean secessionist struggle, whereas the Ethiopian govern-
ment supported the southern Sudanese liberation fronts (Johnson 2003). Gambella
hosted southern Sudanese military training and refugee camps. The strategic impor-
tance of the Gambella region led to its promotion from a district to an administra-
tive region in 1987, when the Peoples Democratic Republic of Ethiopia (PDRE) was
established by the Derg. With the region’s enhanced political status, the infrastruc-
ture and social services in Gambella improved and, above all, a new political space
was created within which Anywaa and the Nuer elites competed for political power.
The 1990s brought tremendous political changes to Gambella. In 1991 the Derg
regime was overthrown by the EPRDF (Ethiopian Peoples Revolutionary Democratic
Front). The EPRDF adopted ethnicity as the official state ideology, an ideology that
was institutionalized in the 1995 constitution in the form of ethnic federalism.
According to this constitution, every ethnic group (‘nations, nationalities and peoples’
in Ethiopian parlance) was accorded the right of self-determination up to and includ-
ing secession (Art.39/1) in order to develop its culture and language (Art.39/2) and to
form a state of its own within the federation (Art.39/3). On the basis of these consti-
tutional provisions, the Federal Democratic Republic of Ethiopia (FDRE) was estab-
lished. The FDRE consists of nine ethnoregional states (see Figure 1.1), including the
Gambella administrative region, which was renamed as the Gambella Peoples National
Regional State (hereafter the GPNRS). The GPNRS is one of three multi-ethnic
regional states of the FDRE. The remaining six regional states are allocated to ethnic
majorities despite the heterogeneity of the populations of all the regional states.5

5. The multi-ethnic regional states are the GPNRS, the BGPNRS (Benishangul-Gumuz Peoples
National Regional State) and the SNNPRS (Southern Nations, Nationalities and Peoples Regional
State). The six regional states allocated to ethnic majorities are the ANRS (Amhara National
Regional State), ONRS (Oromia National Regional State), ANRS (Afar National Regional State),
SNRS (Somali National Regional State) and HNRS (Harari National Regional State).
 Introduction 5

Figure 1.2 Ethnoregional states of the FDRE

Ethnic federalism is the contemporary political context of Anywaa–Nuer relations.

The Anywaa and the Nuer constitute the two largest ethnic groups in Gambella,
accounting together for sixty-seven per cent of the total population of the state.
Politically, they are also the two most dominant groups in the GPNRS. As a result,
the post-1991 political process in the GPNRS has been shaped largely by social inter-
action and political relations between the Anywaa and the Nuer and also by the way
in which both relate to the EPRDF and the Highlanders.

The Highlanders, the Ethiopian Government, the Anywaa and the Nuer
The relationship between the Highlanders and the other inhabitants of Gambella,
especially the Anywaa and the Nuer, is multidimensional. Most commonly, the bound-
ary between these two categories is constructed in terms both of regional origins and
skin colour: the ‘black’ Anywaa, Nuer, Majangir, Opo and Komo are contrasted with
6 Introduction

the ‘red’ Highlanders. The category of Highlander is socially elastic insofar as all non-
Nilotic and non-Koman people with brown skin pigmentation (‘red’ in local parlance),
no matter where they come from, are classified as Highlanders. Thus, the term makes
sense only in the context of Ethiopia’s borderlands, such as Gambella (see Table 1.1).
When understood with reference to Anywaa–Nuer relations, however, the term
‘Highlanders’ signifies not only a diverse group of newcomers from other parts of
Ethiopia but also and especially the Ethiopian state itself. Ever since its first representa-
tives arrived in the Gambella region at the end of the nineteenth century, the Ethiopian
state has been introduced through, represented by and identified with the Highlanders.
It is for this reason that the Anywaa and the Nuer use the same word – gaala or buny,
respectively – to refer both to the Highlanders as people and to the Ethiopian state.6
Once politically dominant and numerically the third largest group, the Highlanders
have been stripped of formal political rights under the terms of the constitution of 1995,
because, as migrants, they now ‘belong’ to various other ethnoregional states in Ethiopia,
depending on their respective ethnic identities. Despite the EPRDF’s new ideological spin,
however, there has been marked continuity in centre-periphery relations in Ethiopia from
one regime to the next. In the current government, as in the past, all authorities who wield
substantial clout in the politics of the region are Highlanders. This enduring special polit-
ical status of the Highlanders makes them the ‘significant other’ in Anywaa–Nuer relations.
Possibly due to the large population of Highlanders, who have been frustrated with
their lack of representation in regional politics under conditions of ethnic federalism, the
EPRDF strategy towards this constituency has changed significantly since the advent of
violent conflict between the Anywaa and the Highlanders in 2003. Particularly since the
contested regional and national elections of May 2005, the EPRDF, in an attempt to
reclaim ‘protest votes’, has chosen the Highlanders as their chief target of reconciliation.
Accordingly, the Highlanders are being invited to join EPRDF’s member organizations
on the basis of their respective ethnic identities.7 This is likely to alter the existing power
relation between the ‘indigenous’ and the settlers/highlanders in Gambella.
Despite occasional rifts between the Highlanders and the Ethiopian government,
they also share an ‘organic’ link, insofar as both are positioned to benefit from the
continued conflict between the Anywaa and the Nuer, given the ethnic basis for the
organization of the Ethiopian federal system. It is thus not surprising that tensions
between either the Anywaa or the Nuer and the Ethiopian government also result in
tensions between the corresponding Nilotic groups and the Highlanders.

6. Etymologically, the term gaala is associated with the word galla, a pejorative term used by the
Amhara and the Tigrean in reference to the medieval Oromo (Hultin 1996). The Oromo were
the first Highlanders with whom the Anywaa came into contact. The term buny was coined by
the Nuer prophet Ngundeng in 1898 during the first encounter between the Nuer and the
Highlanders. Buny, in the Nuer language, means ‘those who bow down’, in reference to the
Highlanders’ reverence towards authority figures. Both the Anywaa term gaala and the Nuer
term buny signify the same things: skin colour (red) and the Ethiopian state.
7. The Amhara National Democratic Movement (ANDM) has taken the lead in recruiting the
Amhara in the Gambella region, followed by similar attempts by the Southern Ethiopia Peoples
Democratic Movement (SEPDM) among members of ethnic groups from Southern Ethiopia
who live in Majangir Zone. The TPLF and the OPDO are moving in the same direction.
 Introduction 7

Admittedly, there also seems to be an ‘organic’ link between the Anywaa, the
Highlanders and the government on the basis of the Anywaa’s greater integration
into the Ethiopian state and their higher degree of competence in Ethiopian national
culture – in contrast to the Nuer. In other domains, however, a more competitive
relationship is observable, particularly with regard to Anywaa political empower-
ment. Over an extended period of time, the Highlanders and the various political
regimes in the Ethiopian state, though they have differed significantly in their modes
of governance, have tended to view Anywaa political empowerment with consterna-
tion, as it is perceived to undermine the economic dominance of the Highlanders in
the Gambella region. The Anywaa’s potential and actual political claims to owner-
ship of the region, as ‘first-comers’, are also considered to be a threat.
The relationship of the Highlanders and the Ethiopian state to the Nuer is also
ambiguous. On one hand, the greater presence of the Nuer in the Sudan and their lesser
political and cultural integration into the Ethiopian polity make the Nuer ‘unreliable’
citizens. As will be shown in the chapters that follow, there have even been occasions
when the Nuer have been viewed as a ‘national security threat’ (see Chapter 8). On the
other hand, the Nuer display a number of characteristics that facilitate the development
of cooperative relations with the Highlanders. In the villages, the Highlanders make a
living as cultivators, for whom the Nuer serve, via the regional livestock market, as the
sole suppliers of cattle. The Highlanders also share with the Nuer the politically insecure
status of being ‘late-comers’ to the Gambella region, which the Anywaa are keen to high-
light. In fact, the Highlanders appreciate the more inclusive and ‘cosmopolitan’ orienta-
tion of the Nuer, in contrast to the more parochial tendencies of the Anywaa.
Finally, it must be noted that the protracted conflict between the Anywaa and the
Nuer has benefited the Highlanders and the Ethiopian government, at least indi-
rectly, insofar as it enhances their own respective economic and political standing in
a highly contested region. This political motive may help to explain the Ethiopian
government’s apparent lack of a strong political will to find a sustainable resolution
to the conflict between the Anywaa and the Nuer. Maintaining and at times fanning
the conflict could have its own political rationale for those who might lose power or
influence under conditions of inter-ethnic peace.

The Effects of the Sudanese Civil Wars

The recurring civil wars in southern Sudan are another important political context of
Anywaa–Nuer relations. Indeed, the political process in Gambella has been intimately
tied to the politics of liberation in southern Sudan ever since the first civil war broke
out in 1955. As early as the mid-1960s, war-fleeing southern Sudanese refugees crossed
the border and settled in different parts of Gambella. By the mid-1980s, the refugee
population outnumbered the local population by far. Even if the actual number of
refugees amounted to only about half of the official estimate of 366,000, as some
experts have suggested (Kurimoto 2005: 344), and even though many refugees
returned to southern Sudan before the regime change in Ethiopia in May 1991, these
population transfers had lasting effects on demographic relations in Gambella. Since
the Nuer far outnumber the Anywaa in southern Sudan, the same is true of the rela-
tive proportion of members of these two groups among the refugees who remained in
8 Introduction

Gambella. As a result, the Anywaa ‘first-comers’ are now outnumbered by the Nuer
‘late-comers’ – a fact which adds to the Anywaa’s demographic anxiety.
Although the exact size of the Nuer population in southern Sudan has not yet been
established, it was estimated to be 500,000 in the Sudanese census of 1956. Basing his
estimates on this census and on the population growth rate of the Sudan, Duany (1992:
20) writes that, in southern Sudan, ‘the population of the Nuer people in 1990 should
lie between 1.5 and 3 million people.’ The size of the Anywaa population in southern
Sudan is less disputed: several scholars have estimated it at 25,000 (Johnson 1986; Perner
1994; Kurimoto 1997). This demographic imbalance is significant, as the cross-border
settlement pattern is one of the social contexts within which contemporary identity pol-
itics in Gambella are being fought out. The expansion of the Jikany, the specific group of
Nuer covered in this study, from the western Upper Nile region in the Sudan to the
Gambella region several hundred miles further east, has been intensified in the context of
the Sudanese civil wars, but it is also a general pattern which began over a century ago.
The escalation of the civil war in southern Sudan in the late 1980s and the conflicts fol-
lowing the regime change in Ethiopia in 1991 drove many Nuer and Anywaa to refugee
camps, located either in Gambella or in Kenya. Surging dramatically in the late 1980s, the
refugee population in Gambella was estimated in 2002 to be 51,374, or twenty-two per cent
of the regional total. At that time, the refugees, forty-five per cent of whom were Nuer, were
distributed among three refugee camps, all of which were located in Anywaa territories.
Many other refugees of the civil war in Sudan were resettled by the United Nations
High Commissioner for Refugees (UNHCR) in North America and Australia. The
number of the Anywaa and the Nuer in the diasporas dating from the late 1980s is
estimated to be 10,000 (Shandy 2001; Falge 2006) and 3,000 (Dereje 2006b), respec-
tively. Members of both the Anywaa and the Nuer diasporas are actively engaged in
the identity politics of Gambella, as shall be shown in subsequent chapters.
In sum, the Sudanese civil wars have produced new political actors and alternative
centres of power, thus making access to transborder political networks one of the most
important factors in determining the outcome of the Anywaa–Nuer conflict. The sig-
nificance of these facts will be explored in detail in subsequent chapters of this book.

Table 1.2 Refugee camp population in Gambella in 2002

Refugee camps Pinyudo Dimma Bonga

Ethnic groups Number Per cent Number Per cent Number Per cent
of camp of camp of camp
population population population

Nuer 12,086 57 11,083 79 – –

Anywaa 7,216 29 300 2 – –
Uduk – – – – 13,832 95
Others 3,565 14 2,490 19 689 5
Total 22,867 100 13,873 100 14,521 100
(Source: Abraham 2002: 2)
 Part I
Theory and Methodology
 Chapter 1

Theoretical Orientation
and Arguments
Debating Ethnicity
The terms of the debate in ethnic studies do not allow for a comprehensive under-
standing of ethnic processes in the Gambella region of Ethiopia. Scholarly discourse
concerning ethnicity and ethnic conflict exhibits a high density of polemic, produc-
ing what Richard Jenkins (2001: 4826) has described as ‘more heat than light’. What
an ethnic group is, how it is related to culture, and when it is mobilized by whom
and for what purposes – these questions are still largely unanswered.
In the literature on ethnicity and nationalism, it is common to distinguish between
two general approaches known as primordialism and constructivism; but, as scholars
use them, both terms are ambiguous. The term ‘primordialism’ is ambiguous, insofar
as some writers fail to distinguish between ‘naturalizers’ and ‘analysts of naturalizers’
(Gil-White 1999: 803). Some analysts of ethnicity are themselves ‘naturalizers’ of
ethnic phenomena, perhaps the most obvious example being those sociobiologists who
suggest that ethnicity should be understood as an extended form of kin selection (van
den Berghe 1981). Probably, however, most authors who are commonly categorized
among the ‘primordialists’ are not themselves ‘naturalizers’ but, rather, ‘analysts of nat-
uralizers’. This is certainly true in the case of Clifford Geertz, who is often regarded as
a key spokesperson for the primordialist approach to ethnicity (see also Foster 1991:
237; Jenkins 2001: 4826; Fenton 2003: 73–90). Nevertheless, Geertz and his senior
colleague, Edward Shils, may be regarded as primordialists, insofar as they assume that
the people studied by anthropologists and other social scientists attach ‘a certain inef-
fable significance … to the tie of blood’ (Shils 1957: 142). In Geertz’s much-cited state-
ment, primordial sentiment is defined as

One that stems from the ‘givens’… – or the assumed ‘givens’ – of social exis-
tence: immediate contiguity and kin connection mainly, but beyond them
the givenness that stems from being born into a particular religious commu-
nity, speaking a particular language, or even a dialect of a language, and fol-
lowing particular social practices. These congruities of blood, speech, custom
and so on, are seen to have an ineffable, and at times overpowering, coercive-
ness in and of themselves. One is bound … [to these loyalties] in great part
by virtue of some unaccountable absolute import attributed to the very tie
itself (1963: 109).
12 Playing Different Games

Thus, those who emphasize primordial attachments to ethnic identities maintain

that ‘ethnicity is a phenomenon rather than an epiphenomenon’ (Horowitz 2002:
73). From this perspective, ethnic identification cannot be explained solely in terms
of a ‘pressure group that mobilizes in order to compete for scarce resources or for the
ambitions of a few leaders’ (Connor 1994: 73). Accordingly, ‘ethnic affiliations
belong to the class of fundamental affiliations that reflect feelings of intense solidar-
ity and are capable of inducing selfless behaviour on the part of group members’
(Horowitz 2002: 74). The object of analysis in the primordialist conception of eth-
nicity is thus ‘the powerful emotional charge that appears to surround or to underlie
so much of ethnic behaviour’ (Epstein 1978: xi).
Constructivism, on the other hand, is an approach based more obviously on ‘an
understanding of ethnicity and nationhood as products of human thought and action’
(Yeros 1999: 1). In this view, ‘ethnicity is the product of a social process rather than a
cultural given, made and remade rather than taken for granted, chosen depending on
circumstances rather than ascribed through birth’ (Wimmer 2008: 971). According to
advocates of constructivism, ‘the process of social construction proceeds at an individ-
ual as well as at a group level; in the innumerable transactions of daily life, individu-
als are engaged in a constant process of defining and redefining themselves’ (C. Young
1993: 24). Like primordialism, constructivism comes in many varieties (Comaroff
1996; Yeros 1999). For the purposes of my argument, however, I will focus on a
variety that differs most clearly from Geertz’s form of primordialism; namely, the
instrumentalist variety of constructivism formulated by Fredrick Barth (1969).
The hallmark of the approach to ethnicity introduced by Barth was the shift from
the cultural to the social function of ethnic groups. According to Barth, ethnic groups
are not ‘culture-bearing units’ but rather ‘a form of social organization’ (Barth 1969:
11–13), meaning that people who share a culture can have different ethnicities and,
conversely, people who are culturally different may have the same ethnicity. This shift
from the cultural to the social in the study of ethnicity was part of a theoretical move-
ment in the wider anthropological scholarship in which culture was no longer taken
for granted but was seen, rather, as part of the explananda. Barth’s paradigm of eth-
nicity emphasizes the role of boundaries in the formation and reproduction of ethnic
identity: ‘the critical focus of investigation from this point of view becomes the ethnic
boundary that defines the group, not the cultural stuff that it encloses’ (ibid.: 15). The
cultural content is defined as variable, depending on which aspect of the culture is
selected by its members to construct the ethnic boundary. Thus, Barth emphasizes the
subjective meaning of ethnicity, for ethnic groups are conceived as categories of ascrip-
tion and identification by the actors themselves on the basis of their own standards of
evaluation, expressed in the form of diacritical markers and core values: ‘the features
that are taken into account are not the sum of “objective” differences, but only those
which actors themselves regard as significant’ (ibid.: 14).
In this book, I analyse Anywaa and Nuer modes of ethnic identity formation
with reference to Geertz’s primordialism and Barth’s constructivism, but with an
important difference: in my analysis, the primordialists and the constructivists are,
respectively, the Anywaa and the Nuer themselves. In formulating my argument, I
employ the distinction between emic and etic points of view, which was introduced
 Theoretical Orientation and Arguments 13

to anthropology by Kenneth Pike and elaborated in somewhat altered form by

Marvin Harris among others (Headland, Pike and Harris 1990). Harris (2001:
571–75) glosses these terms as follows:

Emic statements refer to logico-empirical systems whose phenomenal distinc-

tions or ‘things’ are built up out of contrasts and discriminations significant,
meaningful, real, accurate, or in some other fashion regarded as appropriate
by the actors themselves … Etic statements depend upon phenomenal dis-
tinctions judged appropriate by the community of scientific observers. Etic
statements cannot be falsified if they do not conform to the actor’s notion of
what is significant, real, meaningful, or appropriate.

I call the Geertzian native model of ethnicity ‘emic primordialism’ and the Barthian
native model of ethnicity ‘emic constructivism’. By emic primordialism I mean the
actors’ belief that ethnic identity is a natural part of ‘the givens of social existence’ in
the Geertzian sense (1993: 58). For emic primordialists, the reference point for
inclusion in or exclusion from an ethnic group is common origin. Belonging is
framed in biological terms, however subjectively constructed these terms might be.
By emic constructivism I mean the actors’ belief that ethnic identity is acquired and
that ethnic membership is open to all on the basis of cultural competence. This dis-
tinction is useful in avoiding the conflation of etic terms with emic categories. Of
course, my basic framework is constructivist, insofar as I assume that both emic pri-
mordialism and emic constructivism are constructed in the context of specific histor-
ical, social and political processes.
To the extent that Barth and Geertz were engaged, in their seminal works, with
subjective meanings, they both may be seen as constructivists, despite the conflation
of ‘naturalizers’ and ‘analysts of naturalizers’ under the heading ‘primordialism’ in the
secondary literature. They differ in their views of actors’ understandings of the basis
of their own ethnic identity. For Geertz, the subjective belief in a common origin,
which Roosens (1994: 86) calls the ‘genealogical dimension of ethnic belonging’, is
central to the construction of ethnic identity. Barth’s ethnicity paradigm is also a
native model of ethnicity to the extent that actors themselves set the criteria that
define the ethnic boundary, so that membership in an ethnic group is determined
less by common ancestry than by cultural competence. Barth’s ethnicity paradigm is
based on emic constructivism, but his ideas of flexibility and manipulation contain
elements of rational choice theory, insofar as he assumes that actors are self-interested
and often seek to maximize ‘utilities’ (Hechter and Kanazawa 1997: 194). This is
evident in his concept of identity switching: individual actors change their ethnic
identity in order to tap into the resources of other ethnic groups.
Barth’s theory draws on ethnographies demonstrating that positive bonds may be
established between ethnic groups on the basis of complementary differences (Barth
1969: 18–20). Inter-ethnic relations based on complementary differences may create
especially favourable conditions for identity switching on the part of individuals.
What the ethnic boundary marks is the complementary differentiation between the
groups, while individual actors straddle the boundary as best fits their interests. In
14 Playing Different Games

this sense, Barth’s ethnicity paradigm has an instrumentalist dimension. What

matters for membership in an ethnic group is competence in value standards that
allows the flow of persons across the ethnic boundary, while the boundary itself per-
sists. The continuity of an ethnic group, therefore, depends on the active mainte-
nance of the boundary that differentiates it from other ethnic groups, not on policing
individuals who cross the boundary.
In sum, the perspective I adopt shifts the choice between primordialism in the
Geertzian sense of the word, and constructivism in the Barthian sense, to the emic
level where these two approaches to ethnicity can coexist. Whether an ethnic group
under investigation is primordialist or constructivist in the self-understanding of
group members is sometimes, perhaps often, an empirical question. The task of
scholars should be, then, to identify modes of identification – how ethnic member-
ship is defined by the actors – and to explain the conditions of a particular identity
configuration. This perspective entails a shift from Weber’s notion of ideal types to
Wittgenstein’s idea of family resemblances among various phenomena. To illustrate
the idea of family resemblances, Wittgenstein (1983: §66/§67) uses the example of
games: ‘Consider … the proceedings that we call “games”. I mean board-games,
card-games, ball-games, Olympic-games, and so on. What is common to them all?
… For if you look at them you will not see something that is common to all, but
similarities, relationships, and a whole series of them at that. I can think of no better
expression to characterize these similarities than “family resemblances” … And I shall
say: “games” form a family’.
As Llewelyn (1968: 344) notes, Wittgenstein’s concern is ‘to show that not all
words function in the same way’ and to challenge ‘the strong essentialist idea that
where a general term is used univocally there is a property or set of properties neces-
sary to anything to which the term is correctly applied and sufficient to entitle one
to apply that term’. In this sense, I argue that various ethnic groups, like games, form
a ‘family’. The ideal typical definition of an ethnic group privileges common origins
as the quintessence of ethnicity, but differences between ‘thick’ and ‘thin’ under-
standings of common origins among members of various ethnic groups, and the
implications of these differences for identification processes, have not yet been suffi-
ciently acknowledged. In his important work Ethnic Groups in Conflict, Donald
Horowitz has argued along similar lines:

Many of the puzzles presented by ethnicity become much less confusing once
we abandon the attempt to discover the vital essence of ethnicity and instead
regard ethnic affiliations as being located along a continuum of ways in which
people organize and categorize themselves. At one end, there is voluntary
membership; at the other, membership given at birth … Both principles of
membership – birth and choice – are capable of accommodating fictive ele-
ments … Ethnic groups can be placed at various points along the birth-
choice continuum. (Horowitz 1985: 55)

On this conceptual basis, it is possible to discover different kinds of ethnic groups

showing only family resemblances to one another. This is so because in real-life situ-
 Theoretical Orientation and Arguments 15

ations – for example, in Gambella among the Anywaa and the Nuer, the two ethnic
groups at the centre of this study – ethnic phenomena may take different forms.
Judged in terms of their own ideas of relatedness and the nature of their ethnic
boundaries, the Anywaa may be called ‘primordialists’ and the Nuer ‘constructivists’
– today and in the recent past, at least. In that sense, the Anywaa and the Nuer play
different ethnic language games, as Wittgenstein might say – language games, which,
to conceive of the ‘givenness’ and ‘constructedness’ of ethnic identity in somewhat
differently, seem to be played in terms of ‘being’ and ‘becoming’.1
Being Anywaa is defined with reference to exclusive criteria: descent from an
apical ancestor, having Anywaa parents, belonging to a specific territory, sharing a
particularistic reproductive regime (regulated by scarce bridewealth beads) and val-
orizing social order, especially in the form of territorialized ethnicity. Anywaaness is
something which one is born into. The same is not true for the Nuer: Nuer identity
can be acquired, even if one comes from a different ethnic background. Scholars of
a primordialist persuasion, who posit that ‘ethnic actors believe that membership is
a matter of shared biological descent’ (Gil-White 1999: 803), might not consider the
Nuer to be an ethnic group at all. For constructivists in the Barthian sense, Anywaa
ethnic identity formation and the various institutional arrangements they have
designed to police the ethnic boundary might contradict their understanding of
ethnic groups with permeable and manipulable boundaries. But if we adopt the
concept of family resemblance, we get a different picture: the Anywaa and the Nuer
are examples of different modes of ethnic identity formation.
The concept of family resemblance allows for the possibility of discovering dif-
ferent ‘kinds’ of ethnic groups, without obscuring their structural comparability.
Various ethnic groups may be similar enough to be compared, but they may also
exhibit variations in central features, such as ideas of common origin, criteria of
ethnic membership and degrees of shared sociality. Conceptualized this way, ethnic
groups share a ‘family resemblance’ with one another but are not necessarily identi-
cal in the degree to which they accord relevance to the different dimensions of eth-
nicity. On Horowitz’s ethnic continuum, the Nuer occupy one end, whereas the
Anywaa, with a stronger belief in common origin and an ideology of ethnic
endogamy, occupy the other. To the extent that they emphasize different criteria of
membership for their respective ethnic identities, the Anywaa and the Nuer are,
indeed, playing different ethnic ‘language games’. Using the terms ‘Anywaa’ and
‘Nuer’, they refer to different ideas of ethnic identity and to divergent conceptualiza-
tions of what an ethnic group is. These divergent conceptualizations correspond in
turn to different rationalities and create a potential for conflict in situations of
increased interaction. The Anywaa and the Nuer communicate in various domains
of social life, but the messages conveyed are frequently not understood by those for
whom they are intended.
In explaining the two ‘emic’ modes of ethnic identity construction, I find it nec-
essary to adopt an historical approach. My own ethnic language game is, like that of

1. Compare the similar distinction in Hutchinson 2000 between the ‘primordialist’ and ‘performa-
tive’ approaches to ethnicity on the part of the southern Sudanese Dinka and Nuer respectively.
16 Playing Different Games

the Nuer, constructivist; but it differs from the Nuer understanding of constructive-
ness insofar as in adopting an ‘etic’ position, I do not attribute any intrinsic worth to
a constructivist mode of ethnic identity formation as the Nuer do. Both the Anywaa
primordialist ethnic identity and the Nuer constructivist ethnic identity are con-
structed under specific historical circumstances and in response to certain socio-
political processes. These identities differ in their modes of construction on the basis
of the nature of social experiences, in the conditions of their mobilization and in
their historical depth. Viewed from a diachronic perspective, neither the Anywaa nor
the Nuer mode of identification has remained stable, despite assumptions to the con-
trary on the part of the Anywaa and the Nuer themselves. In fact, the Anywaa iden-
tity discourse seems to have moved over time from an assimilationist ideology to an
ideology of ethnic purity.
The conditions under which the transformation of the identity discourses has
occurred can be specified with reference to differing interactional settings and
various social experiences. Based on the available evidence, I hypothesize that the
objective conditions for the construction of these contrasting ethnic groups are
related to differing processes of ethnogenesis. Historians of Nilotic society have
noted that the Lwoo, a category of people among the western Nilotes including the
Anywaa, incorporated various groups of people during what Crazzolara (1950: 34)
called their ‘epoch-making migrations’ from the thirteenth to fifteenth centuries. In
the postmigration period the emergence of the Anywaa as a distinct people among
the Lwoo seems to have been conditioned by relative isolation or by the presence of
weak neighbours, who could not serve, in the process of self-identification, as ‘ethnic
others’. It is for this reason, perhaps, that the relevant ‘other’ in Anywaa ethnogene-
sis is an oppositional spiritual force known as Jwok. In Anywaa ethnic identity dis-
course, Jwok is a belligerent force posing a constant threat to their existence.
Anywaa primordialism has been reinforced by various socio-political processes,
including the confrontation of Anywaa primordialism with Nuer constructivism as
well as the territorial and cultural encroachment of the Ethiopian state on the
Anywaa cultural world. Apart from a brief moment of politico-military empower-
ment in the early decades of the twentieth century, the Anywaa have been weakened
progressively with their incorporation into the Ethiopian state. The centre-peripheral
mode of relations between the Ethiopian state and its minorities has undermined the
possible emergence of an alternative national identity that is both meaningful and
inclusive. In fact, the Anywaa have practiced their ethnic identity within the
Ethiopian state as a form of resistance, further elaborating on their primordialist
imagination of ethnic identity. The new political process related to ethnic federalism
has further validated and reinforced their primordialist concept of ethnic identity.
Nuer conceptions of ethnic identity, on the other hand, seem to have developed
in the opposite direction, i.e., from an earlier ideology of ethnic purity to an elabo-
rate assimilationism in the context of their dramatic territorial expansion in the
second half of the nineteenth century. Nuer ethnogenesis seems to have taken place
in a context characterized by a high occurrence of competitive intergroup interaction
with neighbours, particularly the Dinka (Evans-Pritchard 1940a; Sahlins 1961). This
appears to have produced an externalized boundary-making process in the Barthian
 Theoretical Orientation and Arguments 17

sense of the term. As a result, Nuer ethnicity is strongly pragmatic: ethnic member-
ship is defined in terms of competence in ‘the cultural standards of evaluation’ rather
than being based on origin. This entails an inclusive identity discourse that places a
high premium on ethnic conversion (becoming Nuer). As in Barth’s version of con-
structivism, there is an instrumentalist current in Nuer emic constructivism. The
emphasis on ethnic recruitment among the Nuer is consistent with a strategy of
resource extraction. As will be shown, however, new social experiences have recently
induced an incipient form of primordialist reconfiguration of the Nuer identity
concept (cf. Hutchinson 2000).

Ethnicity and Conflict

Understanding the link between ethnicity and conflict is even more problematic than
debating the nature of ethnicity. The existence of an ethnic identity does not neces-
sarily entail the political mobilization of group members in conflicts. Horowitz (1998)
has identified a wide variety of theories of ethnic conflict, including primordialism as
advocated by Shils (1995) and Issacs (1975). The primordialists are principally con-
cerned with describing the density and intensity of ethnicity as a form of collective
identity. Horowitz (1998: 5) has summarized their views on the link between ethnic
identity as a form of Gemeinschaft and ethnic conflict in the following manner: ‘a sense
of community of this sort … necessarily generates awareness of other communities,
and this spills over (by mechanisms unspecified) into conflict and violence’.
Taking the primordialist notion of the coercive nature of ethnic emotions as their
point of departure, some sociobiologists have developed theories of ethnic conflict
based on evolutionary conceptions of kin selection and inclusive fitness (van den
Berghe 1981; Brewer and Miller 1996; Gil-White 2001). From this perspective, eth-
nicity has an adaptive value ‘because individuals historically were unable to survive
alone. The benefits of cooperation, however, decline as groups expand, and so there
is an optimal level of group distinctiveness beyond which groups lose the loyalty of
their members ... Outside the boundaries lie ethnocentrism and hostility’ (Horowitz
1998: 12). The most explicit formulation of a primordialist theory of ethnic conflict
has come from anthropologists inspired by evolutionary psychology. Taking birth as
the essential criterion for group membership, Gil-White (2001: 535) recently
advanced a normative theory of ethnic conflict, rehabilitating the old concept of in-
group amity as a condition for out-group enmity: ‘[the] baseline attitude towards
out-group ethnics, in the best of times, is one of at least mild distrust’. Interacting
with members of groups who do not share norms about social exchange may entail
inclusive fitness costs, he argues. Thus, a cognitive framework that sharply demar-
cates ethnic groups as if they were natural species has, in his view, survival value.
In another theory of ethnic conflict, ancient hatred between groups is empha-
sized. This idea, which is especially popular in the mass media, explains contempo-
rary conflicts in much of the less-developed world with reference to the resurgence
of ancient hatreds (Varshney 2002). While some hostilities are very old, however,
many are not (Horowitz 1998: 6). Even in the more distant past, apparent
antipathies were often constructed by means of undercommunication of cooperative
intergroup relations. Although traditional antipathy may play a role in some con-
18 Playing Different Games

flicts, it is far from sufficient in explaining all current conflicts. It is true that the
expansion of the Nuer in the second half of the nineteenth century and their
encroachment on Anywaa territory provides a historical reference point for the con-
temporary Anywaa–Nuer conflict, so that, in this sense, past enmity enters signifi-
cantly into contemporary actors’ definition of the conflict situation. However, this
past enmity is once again relevant not in its own right but because the Nuer continue
to expand. Institutionalized identity politics within the Ethiopian state have also
influenced the selective memory of the Anywaa, which is based on a narrative of loss.
Advocates of the clash of cultures theory argue that conflicts arise between ethnic
groups ‘whose values are in conflict, who want different things, and who do not
really understand each other’ (Horowitz 1998: 6). This theory of ethnic conflict was
originally a reaction to the ‘contact hypothesis’. Proponents of the contact hypothe-
sis argue that the key feature of prejudice and intergroup conflict is the existence of
unfavourable stereotypical attitudes and related behaviour. The best way to reduce
tension and hostility between groups, these same authors suggest, is to bring them
into systematic contact with each other in various ways (R. Brown 1995: 236).
Contact alone, however, is not a sufficient condition for reducing inter-ethnic hos-
tility, especially in situations of unequal relations. Forbes (2004: 74) identifies three
variables that condition the nature of contact between ethnic groups: (1) ‘the equal-
ity or inequality of status of the groups in contact’; (2) ‘their cooperative or compet-
itive interdependence in the pursuit of common goals’; and (3) ‘the presence or
absence of social norms supporting intergroup contact’. It will be shown that the
conflict between the Anywaa and the Nuer is not related to a lower level of interac-
tion but rather to changes in the nature of their contact. After the initial conquest of
Anywaa territories by the Nuer, the relationship between these two groups evolved
into a seasonal symbiotic interaction that was built upon complementary exchanges
of products from their respective agrarian and pastoral economies. This seasonal
contact was well regulated by a local power differential: Nuer settled temporarily in
Anywaa territories along the banks of the rivers with the consent of Anywaa local
leaders. Changing power relations and the transition from seasonal to permanent
contact have made the cultural differences between the two groups more visible.
In his insightful article ‘The Cultural Contexts of Ethnic Differences’, Thomas
Eriksen (1991: 140), drawing on Wittgenstein’s (1983) concept of language games,
identifies three inter-ethnic contexts where the degree of shared meaning is variable: (1)
language games within a single language, which ‘implies agreement over constituting
and strategic rules of interaction … [through which] the agents understand each other
when they are playing the same language-game’, (2) overlapping language games, when
‘there is agreement as to the form and content of only some relevant aspects of the
interaction’ and (3) incommensurable language games, when ‘interaction is difficult
and its regulating rules will normally be defined by the most powerful agent.
Misunderstandings and highly divergent definitions of the situation will be common’.
In conceptualizing ethnic identity, the Anywaa and the Nuer play incommensu-
rable language games, in Eriksen’s sense, which makes for an inter-ethnic context with
great potential for conflict. The Anywaa define the inter-ethnic context as a Nuer con-
spiracy to eliminate them. This interpretation is partly generated by the profound cul-
 Theoretical Orientation and Arguments 19

tural differences between the two societies, including a divergent conceptualization of

what an ethnic group is or ought to be. If, however, the Anywaa are angered by what
Hutchinson (2000: 9) calls ‘the sticky grasp of the Nuer on their neighbors’, the Nuer
are also bewildered by attempts of the Anywaa to achieve ‘closure’.
Cultural differences between the Anywaa and the Nuer find expression in con-
trasting ideas of the link between identity and the land. Anywaa identity discourse is
rooted in an historical sense of place; thus, the land has an important symbolic
meaning. Nuer identity discourse is mobile, and territorial expansion is a function of
their material conditions of life, especially of the exigencies of the pastoral economy.
The Anywaa experience Nuer territorial encroachments less as a threat to their eco-
nomic well-being than as a symbolic violation of their cultural world. In fact, one
aspect of the struggle in the Anywaa–Nuer conflict is the struggle for cultural iden-
tity. The Anywaa are trying to prevent the Nuer from making territorial and also cul-
tural encroachments. It is a struggle that is fought at different levels and with
different means, encompassing the masses and the elites in different social arenas.
Competition for natural resources is another factor that is often cited in explain-
ing ethnic conflict (Fukui and Markakis 1994; Schlee 2004). Riverine lands repre-
sent a contested resource at the inter-ethnic level in the Gambella region. Access to
and control of riverine land is, therefore, a genuine material concern for those who
participate in ethnic conflict. Nevertheless, not all the resource conflicts are based on
ethnicity. Even when a resource conflict assumes an ethnic dimension, not all the
Nuer are antagonistic towards all the Anywaa. Therefore, competition for natural
resources is not in itself a sufficient factor to generate the conflict between the con-
temporary Anywaa and Nuer. There are internal divisions within each ethnic group,
and there is a buffer zone occupied by groups that have links to both sides and, there-
fore, would benefit from inter-ethnic peace. In fact, in Gambella, resource conflicts
within ethnic groups often take more severe forms than those between ethnic groups,
because the various subgroups are competing for the same economic niches.
Other theorists argue that elite competition for political power drives ethnic con-
flict (Vail 1991; Brass 1999). According to Brass (1999: 13), ‘elite competition is the
basic dynamic which precipitates ethnic conflict under specific conditions, which
arise from the broader political and economic environments rather than from the
cultural values of the ethnic groups in question’. Brass further notes (1999: 25) that
‘ethnic communities are created and transformed by particular elites in modernizing
and in postindustrial societies undergoing dramatic social change. This process
invariably involves competition and conflict for political power, economic benefits,
and social status between competing elite, class, and leadership groups both within
and among different ethnic categories’.
Not surprisingly, the elite competition model has been criticized for creating an
inaccurate picture of ‘evil’ politicians and ‘innocent’ masses and for leaving too little
room for individual acts by those members of the masses who engage in conflict behav-
iour. The model sheds light on the ethnic process, but it cannot provide definitive
answers to the questions posed by Horowitz (1998: 9), including the following: ‘Why
does inter-elite conflict proceed along ethnic lines, and why do the followers of elites
follow them if the benefits flow solely to the elites whose interest motivates the strug-
20 Playing Different Games

gle?’ Varshney (2002: 29) criticized the elite competition model in similar terms: ‘If the
masses were only instrumental with respect to ethnic identity, why would ethnicity be
the basis of mobilization at all? Why do the leaders decide to mobilize ethnic passions
in the first place?’ Furthermore, the depiction of the elites as manipulators leaves no
room for their occasional articulation of societal concerns. In my analysis of the
dynamic relations between the Anywaa, the Nuer and the Ethiopian state, I will show
that elites, while often pursuing their own self-interest, often also articulate and engage
in vital collective interests. While doing so, they are not, however, in total control of
the effects of their own actions. They are, indeed, rational actors, but the end result
does not always match their intentions. If the empowerment of the Anywaa in the
1990s led to the political marginalization of the Nuer elites, it also meant an increase
in Nuer expansion into Anywaa areas, which allowed them to gain access to social serv-
ices, most of which were concentrated in these Anywaa areas. The unintended conse-
quences of political action can also be illustrated with reference to the varied results the
Anywaa and the Nuer elites obtained by framing local issues in national terms. The
Anywaa initially succeeded in framing their conflict with the Nuer in terms of
‘Ethiopian citizens’ troubled by the ‘Sudanese Nuer’. But this also encouraged the Nuer
to adopt the same political discourse in the politics of entitlement. Indeed, the Nuer
utilized the Anywaa strategy of political entitlement with greater success than the
Anywaa, but it also resulted in increased intra-ethnic political fragmentation, as some
Nuer tribes or clans appeared to be ‘more Ethiopian’ than others.
In the wider debate over causes of ethnic conflict, I go beyond monocausality,
explaining the conflict between the Anywaa and the Nuer with reference to three inter-
acting variables: identity, resources and power. The multiple causal links and the inter-
action between the variables that generate conflict are formulated in this book as follows:

• Contrasting ethnic identity formations create the potential for ethnic conflict
(the identity variable)
• Unregulated access to and control over natural resources lead to ethnic con-
flict (the resource variable)
• Differential patterns of incorporation into an ethnically stratified state gener-
ate ethnic conflict by altering inter-ethnic power relations and preventing the
emergence of an alternative national identity (the power variable)

The ‘identity’ variable is related to the cultural opposition inherent in the contrasting
ethnic identity formations, which find expression in different ethnic language games.
For the agrarian Anywaa, ethnic identity is territorialized, and this is the basis for inter-
ethnic exchanges. The identity discourse of the Nuer, embedded in their pastoral
lifestyle, is mobile – a contrast which is also evident in the constitution of their local
communities. Unlike Anywaa villages, Nuer local communities can be transplanted
into new territories with new members who can claim full membership. The inherent
formation of Nuer culture is geared towards the dissolution of other ethnic boundaries,
resulting in their own territorial and demographic expansion and eliciting from the
Anywaa an even more extreme reiteration of their primordialist self-understanding.
This has produced mutually antagonistic stereotypes. The contrasting ethnic identity
 Theoretical Orientation and Arguments 21

formation has also created a certain degree of value dissonance. For example, the eco-
nomic value the Nuer give to the land and the symbolic value the Anywaa give to the
land make it difficult to negotiate a settlement of interests at the inter-ethnic level.
The ‘resource’ variable is related to access to and control of riverine lands. This
type of land covers only a minimal percentage of the total land area of the region,
but it has outstanding agricultural and pastoral value. Imbalance in the geographical
distribution of riverine land also contributes to resource conflicts. Most land of this
type is located in Anywaa territories. Individual Nuer and Nuer local communities
have a strong desire to gain access to riverine land, which is vital for their agropas-
toral livelihood. Thus, the economic stakes are high for both the Anywaa and the
Nuer. For the Anywaa, however, the economic expansion of the Nuer into the river-
ine lands also poses an additional threat, insofar as land serves as a symbolic resource
in their identity construction.
The resource dimension of the Anywaa–Nuer conflict, as it is experienced through
the lens of varying identity structures, must be understood with reference to fluctuating
local ‘power’ relations and translocal political processes – particularly the various ways in
which the two ethnic groups are incorporated into the state system. Anywaa and Nuer
elites do compete for political power, but power is valued not only for its materiality. In
the case of the Anywaa, it is also of vital importance in their attempts to maintain their
ethnic identity and support their status claim. This is all the more true in the context of
post-1991 Ethiopia, where state politics are organized in terms of ethnic federalism.
Anywaa elites seek political power in order to contain Nuer expansion and to prevent
what they perceive as the threat of the extinction of Anywaa society. For their part, Nuer
elites seek power to create and expand a political space that allows them to make use of
the new opportunity structure that has accompanied ethnic-based decentralization in
Ethiopia. Basically constituted in terms of centre-periphery relations, the ethnically
stratified Ethiopian state has failed to integrate its historic minorities such as the Anywaa
and the Nuer in a positive way. The strategic co-optation of one or the other ethnic
group by the various political regimes has created fluctuating inter-ethnic power rela-
tions. Rather than providing opportunities for transcending ethnic division by con-
structing an alternative national identity, the Ethiopian state has thus become an arena
for inter-ethnic competition. Simultaneously, the competitive relations between the
Anywaa and the Nuer has allowed state actors to seek political control in the spirit of the
old dictum, ‘divide and conquer’. The lack of political will on the part of the govern-
ment to resolve the conflict in a sustainable manner might have its own political ration-
ality, namely, that of enhancing state control over its periphery.
This book identifies the causes of conflict between the Anywaa and the Nuer.
What is more, it also shows how these causes interact. In competing over natural
resources, the Anywaa and the Nuer orient themselves with reference to different
identity systems. The interplay between the resource and identity variables is
expressed in the different values that the Anywaa and the Nuer give to the land and
in the inevitable misunderstanding this creates. In inter-ethnic relations, the Nuer,
pursuing their own economic interests, overlook the symbolic concerns of the
Anywaa. Conversely, the economic imperatives of Nuer territorial expansion are inter-
preted by the Anywaa in terms of a conspiracy theory: They see it as a deliberate
22 Playing Different Games

attempt by the Nuer to eliminate the Anywaa. The interplay between the identity and
the power variables, on the other hand, is expressed in the Anywaa’s discourse of
ethnic extinction. The progressive economic and military decline of the Anywaa
within the Ethiopian state has undermined their capacity to contain Nuer territorial
encroachments. The failure of the Ethiopian state to integrate its minorities and its
recent institutionalization of ethnicity have also put a premium on primordialist
imaginings of the social universe. Thus, in post-1991 Gambella, the three variables to
which I refer in explaining the Anywaa–Nuer conflict – identity, resources and power
– reinforce one another. Nuer expansion has assumed a new dimension, as it has a
direct bearing on the politics of ethnic entitlement. Nuer pastoral mobility is now
associated with the expansion of a political constituency, which, in turn, leads to a
magnification of the Anywaa discourse of ethnic extinction. It is in this new political
context that the issue of Anywaa ethnic rights becomes prominent.
The subsequent chapters substantiate the main arguments of the book. The two
chapters of Part II describe and analyse the contrasting ethnic identity formations of
the Anywaa and the Nuer. Chapter Two shows what makes the Anywaa primordial-
ists, describing the conditions for the primordialist configuration of their ethnic
identity formation; while Chapter Three shows what makes the Nuer constructivists,
describing the conditions for the constructivist configuration of their ethnic identity
formation. The first two chapters of Part III – Chapters Four and Five – explain the
conflict between the Anywaa and the Nuer as resource-based and identity-based con-
flicts, respectively, and they show how the two variables interact. Finally, Chapters
Six to Nine explain Anywaa and Nuer conflict in terms of their differential pattern
of incorporation into the Ethiopian and Sudanese states.

Notes on Methodology
This book has grown out of ethnographic fieldwork that I carried out from 2000 to
2006. Systematic fieldwork was preceded by two years of residence in Gambella,
from December 1997 to September 1999, when I worked there as a research officer
and programme coordinator for a nongovernmental organization, the Agency for
Cooperation and Research in Development (ACORD). The working conditions and
the staff composition of the organization made it an ideal site for preliminary obser-
vations. Members of each of the three major groups in Gambella (the Anywaa, the
Nuer and the Highlanders) worked in the organization in various capacities. My own
social position as a Highlander was a ‘methodological resource’. In Anywaa–Nuer
relations, the Highlanders, given their identification with the Ethiopian state,
become an audience for their political debate.2 The two years when I lived in
Gambella were times of heightened conflict between the Anywaa and the Nuer. In
fact, it was the magnitude of one of these conflicts, the 1998 conflict in Itang, that
inspired me to carry out research along the lines developed in this book.

2. It was relatively more difficult for me to do fieldwork in Gambella after 2003, when a major
conflict broke out between the Highlanders and the Anywaa.
 Theoretical Orientation and Arguments 23

As a researcher in the Department of Conflict and Integration at the Max Planck

Institute for Social Anthropology in Halle (Saale), Germany, I carried out thirteen months
of fieldwork in Gambella between June 2000 and May 2001 and in January and February
2002. In March 2002, I undertook further fieldwork among the Anywaa and the Nuer
refugees in Kenya and the Sudan as part of a research project, ‘Social and Political Settings
of Refugees’, organized by the Graduate School of Human Sciences, Osaka University.
From May 2003 to February 2005, as a Research Fellow in the Department of
Anthropology, Osaka University, I carried out fieldwork in Gambella: in the Anywaa and
Nuer refugee camps in Ruiru, Kenya; in Pochalla County in southern Sudan; and in
Khartoum. From December 2005 to March 2006, as a Research Fellow at the Max Planck
Institute for Social Anthropology, I was occupied with additional fieldwork in Gambella.
Throughout 2007, I conducted research among the Anywaa and Nuer diasporas in the
United States and Canada, also following up on their homeland ties.
The nature of the research topic required a multi-sited ethnographic project for
which specific sites were chosen on the basis of the issues being explored.
Nevertheless, much of the data used in this book is derived from four sites: Gambella
town, Pinykew village in Gambella district, Makot village in Itang district and
Wechdeng village in Jikaw district. The choice of these sites was largely dictated by
the nature and density of the social and political interactions of the people being
studied. Gambella town is the main area of interaction among the Anywaa, the Nuer
and the Highlanders. According to the 1994 census, the town had a population of
18,263, of which thirty-four per cent were Anywaa, ten per cent were Nuer and fifty-
six per cent were Highlanders. Above all, most of the contemporary conflicts
between the Anywaa and the Nuer have occurred in Gambella town.
Pinykew village is located eight kilometres west of Gambella town. Traditionally,
Pinykew was an Anywaa village, but since the mid-1980s it has hosted a growing
Nuer population, the members of which were attracted by the availability of social
services in nearby Gambella town. The Nuer in Pinykew live in a compact settlement
area called Ochom. According to the 1994 census, the population of Pinykew village
was 2,500. Although the numbers of Anywaa and Nuer in Pinykew are roughly
equal, the Anywaa are politically dominant. This partly mirrors the new power rela-
tions at the regional level and is based largely on Nuer recognition that it is Anywaa
land. The subordinate position of the Nuer is expressed in the power of the Anywaa
to appoint and demote local Nuer leaders. The Nuer also pay tax to the government
on behalf of the Anywaa, which de facto gives them something equivalent to a ‘resi-
dence permit’. This power relation echoes the seasonal encounters that occurred
when the Nuer were hosted by Anywaa nobles and headmen prior to the 1970s.
Unlike other rural villages and Gambella town, where the nature of social and polit-
ical relations are characterized by contestation and intermittent conflict, social rela-
tions between the Anywaa and the Nuer in Pinykew village during the study period
were largely peaceful.3 It was this peacefulness and my interest in comparing the dif-

3. The only major confrontation between the Anywaa and the Nuer in Pinykew occurred in 2002,
itself a spillover from the power struggle between the Anywaa and Nuer elites in Gambella town.
24 Playing Different Games

ferent processes in different villages that led me to choose Pinykew as one of the vil-
lages to study. The nature of the social and political relations in Pinykew is used to
illustrate the changing power relations between the Anywaa and the Nuer in the
context of the new ethnopolitics in the Gambella regional state.
The third main field site was Makot village in Itang district. Makot is also a
village with recent Nuer migrants, who came to Itang in the early 1980s, largely from
southern Sudan. Unlike Pinykew, however, Makot had been at the forefront of con-
flicts between the Anywaa and the Nuer, though social and political relations have
also included cooperative economic exchanges. One of the extended case studies (the
1998 conflict in Itang district) is drawn from Makot and the neighbouring villages.
The fourth main field site was Wechdeng village in the Kurtony area, which is the
Thiang Nuer wet-season settlement area in Jikaw district. Most of the data used to
describe the Nuer identity discourse and practices is derived from the microcensus I
conducted in Wechdeng village. Wechdeng village serves as my main example in
demonstrating the flexibility in Nuer recruitment of ‘outsiders’ for membership in
their local communities. Wechdeng village was founded by Deng Buoy, an assimilated
Dinka. The integrative thrust in Wechdeng village is contrasted with the persistence
of the insiders-outsiders boundary among the Anywaa in Akedo village on the Baro
River. To illustrate the process of ethnic conversion, especially among the Nuer,
further intermittent fieldwork was carried out in the following localities: Akobo vil-
lages of Pone, Burbey and Jingmir; the villages of Pol, Itiel and Pinyman in Itang dis-
trict; and the mixed-settlement villages of Edeni, Pijwo and Teyluth in Jikaw district.
The research area included four districts, Gambella, Itang, Jikaw and Akobo,
where the majority of the Anywaa and the Nuer live together. The Anywaa who live
in these districts are the Openo Anywaa, whereas the bulk of the Nuer covered by
the study are the Gaat-Jak, with the exception of Pone and Burbey villages, which are
home for the Gaat-Jok Nuer. In Gambella town, however, all of the Anywaa and
Nuer sections are represented. In order to get a more accurate picture of the refugee
population, I also carried out fieldwork in the refugee camps, particularly in Pinyudo
refugee camp, located in the Anywaa district of Gog, some seventy kilometres south
of Gambella town. In March 2001, I visited Utalo village in Pochalla County in
southern Sudan, where I attended the coronation ceremony of Adongo Ageda, the
current nyiya (king) of the Anywaa. In August and September 2002, I carried out
fieldwork in Nairobi and Khartoum among the Anywaa and Nuer refugees and polit-
ical leaders. The fieldwork in Nairobi, where I managed to interview some of the key
political actors who were in Gambella prison during my earlier fieldwork, was
crucial. Most of the data for the analysis of transnational political networks in the
Anywaa–Nuer conflict is also based on fieldwork in the Sudan and Kenya.
I used a variety of tools to collect data. One of the research tools was localization
through host families. In Gambella town I stayed in the guest house of the church
compound of the Eastern Gambella Bethel Synod (EGBS), also known as the
Mekaneyesus Church. This was also the site of the church’s administrative office.
There were three families living in the compound, two Anywaa families and a
Highlander family. Being a ‘normal’ resident in the compound gave me the degree of
invisibility that was necessary to observe everyday interaction between the Anywaa
 Theoretical Orientation and Arguments 25

and the Highlanders. I also cultivated friendly relations with my Anywaa and Nuer
field assistants, who often stayed with me in the compound. The rapport they devel-
oped over a period of seven years contributed to the development of an atmosphere
that allowed me to engage them in discourse and counter-discourse. I gained consid-
erable insight into the complexity of inter-ethnic relations in Gambella by observing
the interaction among my research assistants.
I also had a host family in Gambella town, George Nikola and his wife and chil-
dren, an interesting family with a mixed history. George Nikola is a descendant of
one of the Greek merchants who came to Gambella during the enclave period
(1902–56). He is a Highlander on his mother’s side and is married to an Anywaa.
George is very knowledgeable about the early history of Gambella, and his wife,
Ariat, an Anywaa from Gog area, commands a considerable degree of respect among
the Anywaa in different walks of life. Apart from enjoying the advantages of my long-
term acquaintance with George’s family, I also benefited from being in his family
home, which was an important social meeting place. His compound was frequented
by notable Anywaa opinion-makers, who were also my interviewees.
In Pinykew village, I was hosted by Riek Tuany, a Gaat-Jak Nuer from the Cieng
Cany clan. Following Riek’s recommendation, I built a hut in his compound. I chose
Riek as my host for two reasons. First, he is the first Nuer settler in Ochom, the Nuer
section within Pinykew village, and so my association with him helped me to gain
recognition among the others more easily and more quickly. As the representative of
the Nuer, Riek had a compound that served as a site for socializing, particularly in
the days before and after court cases. Court cases are preceded by factional politick-
ing and followed by a drinking party. My stay in Ochom subvillage enabled me to
learn about the different facets of Anywaa–Nuer relations, not least the economic
basis of Nuer local power (the cattle economy), which is at the heart of the assimila-
tion process and ultimately leads to ethnic conversion, particularly in areas far from
administrative centres, where there is less reflexivity on ‘identity maintenance’. The
Anywaa political influence in Pinykew village is not supported by economic power,
and so it is not unusual in everyday life in Ochom to see Anywaa asking Nuer for
favours or entering into unequal economic exchanges. Riek’s family is also interest-
ing, insofar as it illustrates one of the micro-processes through which the Nuer
expand, i.e., instrumental inter-ethnic marriages. Riek’s second wife is a widowed
Anywaa, and my intimate and open discussions with him and his Nuer first wife,
Buk, about inter-ethnic marriages provided most of the data I used to explore the
grey area between manifest and latent functions of social acts.
I employed similar research methods in Makot village. My first attempt to visit
Makot village in 1998 was frustrated because, by then, it was a ‘no-go zone’ after
months of bitter conflict between the Anywaa and the Nuer in Itang district. This
resulted in the imprisonment, for one year, of Kong Diu, the leader of the Cieng
Reng Nuer community in Makot village, as one of the ‘ringleaders’ of the conflict.
When I went back to Gambella to carry out fieldwork in June 2000, I learned that
Kong had been released and had left for Addis Ababa to appeal to the national par-
liament for citizenship for his community. I followed him to Addis Ababa. After
some initial hesitation on his part, we established a rapport and, over a period of
26 Playing Different Games

intermittent meetings, the relationship further developed into a friendship. When

Kong left Addis Ababa for Gambella, I followed him to Makot village, where Kong,
who had been impressed by my courage in visiting him in what was still a ‘no-go
zone’, introduced me to the Cieng Reng Nuer as his trustworthy friend. This per-
sonal relationship was crucial for my fieldwork in Makot village. Without it, it would
have been virtually impossible for me to earn the confidence of the Makot villagers,
who, living through an extremely volatile and uncertain situation, had good reason
to be suspicious of strangers. The data I used for one extended case study (the 1998
conflict in Itang) is largely derived from formal and informal exchanges I had with
Kong in various places.4
I conducted extensive semi-structured interviews with people of various social
categories among the Nuer, Anywaa and Highlanders in Gambella town and in the
villages. These interviews were conducted in various ways. Interviews with ordinary
Anywaa and Nuer men and women in rural settings were carried out in vernacular
languages with the help of my Anywaa and Nuer field assistants, who also worked
with me as translators. I tried to enhance translation accuracy through cross-check-
ing with other native translators, particularly for the narratives included in this book.
Furthermore, my admittedly limited knowledge of both languages still enabled me
directly to follow some of the conversations and to communicate during casual
encounters in everyday life. I conducted interviews with the Anywaa and Nuer polit-
ical leaders and ordinary men and women in the urban areas either in Amharic or in
English, whereas all of my interviews with Highlanders were conducted in Amharic.
A few of my interview partners agreed to be identified with their real names, but
most of the names of interviewees have been replaced with pseudonyms in order to
ensure their anonymity.
The choice of the interviewees was largely dictated by their degree of relevance to
the case studies selected. I organized the interviews in a way that allowed me to elicit
discourses pertaining to the definition of the conflict situation. Explorations of the
life-histories of interview partners helped me to determine the extent of the overlap
between individual and group interests. I also conducted random interviews and
informal discussions. By recording interviews in narrative form, I came to under-
stand the views, worries and concerns of my various Anywaa and Nuer informants
and was able to document their own stories and definitions of the conflict situation.
The bulk of the argument in the cultural analysis of the conflict is derived from these
narratives. I tried to minimize the limitations of selectivity by including ‘negative
cases’, that is, alternative representations of the conflict that did not fit into the dom-
inant discourse of the group. Thus, in addition to interviewing the main political
actors in the conflict, who, not surprisingly, subscribed to an essentialist discourse of
ethnic differences, I also talked to ordinary Anywaa who had encountered new forms
of social experience. These categories of interviewees included both educated Anywaa
and church leaders, who, despite their presumed modernist and universalistic
outlook, were often at the forefront of ethnic radicalism.

4. Kong passed away in February 2003. I would like to take this opportunity to express my con-
dolences to both his family and the entire Cieng Reng community in Makot village.
 Theoretical Orientation and Arguments 27

I also used a novel method that I call ‘engaged conversation’. Throughout the
fieldwork period, I participated in political debates as a former resident of Gambella.
On the basis of past experience, I could initiate everyday conversations on issues
related to the conflict situation without appearing to be an interviewer. In the course
of ‘engaged conversations’, I mediated a discursive dialogue, provoking the Nuer
with Anywaa arguments, and vice versa, in order to elicit their counter-arguments
without naming the originators. This methodology was intended to evoke counter-
discourses in order to gain access to what went on in the minds of the relevant actors,
i.e., what they thought the conflict was about and how they were involved in the
conflict. This methodology had its own rewards and risks. Except for one major inci-
dent, I was successful in my approach. In Makot village, I was once confronted by a
Nuer elder who thought that I was defending the Anywaa in the discursive dialogue
I was conducting. In the middle of our conversation, he became very angry with me
and said, ‘You are talking like the Anywaa’. In the tense atmosphere that followed,
my reputation was restored, thanks to Kong’s intervention. In his opinion, I was a
friend of the Cieng Reng, evidenced by my courage in visiting his troubled fellow vil-
lagers and listening to their worries.
Interviews were supplemented by systematic observation and participation. This
included attending conferences, political meetings and celebrations on public holi-
days, and getting the ‘feel’ of the emotional landscape of those who participated in the
conflict. I was in Gambella during the 1998 Itang district conflict between the Nuer
and the Anywaa. I visited some of the Anywaa villages burned by the Nuer, talked to
the displaced Anywaa in Itang town, and met many Nuer in Itang and Gambella
towns who were deeply angry about what they called unacceptable Anywaa domina-
tion. I was horrified by the magnitude of the conflict and the atrocities committed by
both sides. During my systematic fieldwork, the emotional investment of both the
Anywaa and Nuer in the conflict later became a part of the problem for which this
study seeks to provide solutions, at least analytically. What generates such strong emo-
tions? Were such atrocities mere creations and manipulations by the elites as is often
claimed in the literature on ethnicity? While preparing for the field, therefore, one of
my main research inspirations was Horowitz’s (1985, 1998, 2002) perennial question:
‘Why do followers follow in situations of ethnic conflict?’ My personal encounter with
the people who took part in the conflict convinced me that there is more to ethnicity
than a mere ‘false consciousness’ – an observation which became vivid as I got to know
more about the attitudes and the differences between the Anywaa and the Nuer in
their definition of the conflict situation.
Another important research tool I used was the microcensus. I conducted a com-
plete microcensus in Pinykew village in order to determine demographic trends, eco-
nomic status and push-and-pull factors for migration.5 An additional neighbourhood
census was conducted in Wechdeng village to determine the social composition of a
local Nuer community, which became the data I used to determine Nuer social
organizations and their mode of recruitment into local communities. Collection and

5. I learned about microcensus as an anthropological research tool from Professor Günther Schlee
during his visit to my field sites in February 2001.
28 Playing Different Games

analysis of songs, proverbs and sayings was another method of data collection. I
found it useful to learn which aspects of social interaction are judged to be typical
and which historical events are remembered by whom and how. I collected and
analysed the songs and sayings particularly to learn about stereotypes, collective
memory and cultural representations of the political process.
As the research project was basically diachronic, I used historical methods in
order to examine the impact of the various political regimes on local identification
processes. Towards that end, I carried out archival research in Gambella town and in
the two highland towns of Gore and Metu. I supplemented my own archival research
and the oral accounts of the interviewees by drawing on two seminal dissertations on
the history of the region and its people: Bahru Zewde’s Relations between Ethiopia
and the Sudan on the Western Frontier (1976) and Gabriel Jal’s History of the Eastern
Jikany before 1920 (1987). I also consulted the writings of Robert Collins extensively,
particularly Land Beyond the Rivers: The Southern Sudan, 1898–1918 (1971) and
Shadows in the Grass: Britain in the Southern Sudan, 1918–1956 (1983). Data for the
Nuer concept of ethnic identity was generated from my own observations as well as
from three sources: Evans-Pritchard’s Nuer trilogy: The Nuer: A Description of the
Modes of Livelihood and Political Institutions of a Nilotic People (1940a); Kinship and
Marriage among the Nuer (1951a) and Nuer Religion (1956); Douglas Johnson’s Nuer
Prophets (1994); and Hutchinson’s Nuer Dilemmas: Coping with War, Money and the
State (1996). For the discussion of the Anywaa discourse of ethnic identity, I supple-
mented my own material with four sources: Evans-Pritchard’s Political System of the
Anuak of the Anglo-Egyptian Sudan (1940b); Lienhardt’s two articles on the Anywaa
village headmen (1957 and 1958); two volumes of Perner’s meticulous ethnography
on the Sudanese Anywaa, The Anyuak: Living on Earth in the Sky (1994 and 1997);
and numerous articles by Kurimoto on contemporary Anywaa politics and their sub-
sistence economy.
 Part II
The Contrast
 Chapter 2

The Anywaa Primordialist

Ethnic Identity Formation
Once one concedes that the term ‘ethnicity’ refers to a range of phenomena linked
loosely by certain underlying family resemblances, in Wittgenstein’s sense, it becomes
evident that ethnic groups may exhibit considerable variation in their specific manifes-
tations. This chapter is devoted to a description and analysis of the formation of ethnic
identity among the Anywaa – a formation which may be characterized generally as pri-
mordialist. In this context, the term primordialism is used as a gloss for the ‘emic’ dis-
courses and practices of the Anywaa themselves: the Anywaa believe that ethnic identity
is among the ‘givens of social existence’ (Geertz 1963: 109) that are conferred by birth.
The primordialist aspects of Anywaa ethnic identity are observable in various domains
of social life, including in particular their ideas of origin, their marriage practices, their
ideology of purity, their village-based political organization, and a belief system based on
particular understandings of territoriality. A primordialist style of establishing ethnic
boundaries may be inferred from Anywaa interactions with neighbouring groups, from
Anywaa images of divinity and also from new socio-political processes, including espe-
cially the reactions of the Anywaa to the hegemonic projects of the Ethiopian state.

Emergence of the Anywaa – Common or Diverse Origins?

While the Anywaa tend to view themselves as essentially homogeneous, due to their
supposedly common origin, the evidence from language history, ethnohistory and
the analysis of mythology is ambiguous. The following, very brief review of the lit-
erature on Anywaa ethnogenesis is offered here to provide some insight into the com-
plexity and controversy surrounding the origins of a supposedly ‘pure’ ethnic group
and to serve as a backdrop to the subsequent exposition on the various aspects of the
‘emic primordialism’ of the Anywaa.
Anywaa – sometimes spelled Anuak or Anyuak by other scholars – is the name
of the people. Pach Anywaa and dho Anywaa are, respectively, terms which the
Anywaa use to refer to their country and their language. Dho Anywaa is part of the
western Nilotic language family, and within that family it is classified under the cat-
egory of the Lwoo (Crazzolara 1950; Ogot 1967).1 The Nilotes are said to have come
from West Africa and to have occupied the Nile valley in central Bahr al-Ghazal
‘perhaps before the birth of Christ’ (Collins 1971: 51). Crazzolara (1950: 34)
believes that Anywaa ethnogenesis occurred in the context of the ‘epoch making
migrations’ of the Lwoo.

1. Lwoo is spelled by other scholars as Lwo (Collins 1971) or Luo (Ogot 1967).
32 Playing Different Games

The Lwoo group consists of nine major divisions, including, among others, the
Anywaa and the closely related Päri and the Shilluk of the Sudan, the Acholi of
Uganda and the Luo of Kenya (Crazzolara 1954; Reh 1996). Crazzolara (1950: 31)
suggests that ‘the land of origin of the Lwoo must apparently be sought for south-
east of Wau and south of the Bahr-al-Ghazal’. Collins describes the great Lwoo
migrations as follows:

In the fifteenth century … the Lwo began to migrate from the Bahr al
Ghazal. The Lwo marched south and east toward the Bahr al-Jabal. [One
group] struck off to the north … Another group led by Gilo also disengaged
themselves from the main body, marched north and east to the Sobat River,
and made their way upstream to their present location at the base of the
Ethiopian escarpment. They are known today as the Anuak … in the seven-
teenth century, a splinter group moved from the Anuak country to Lafon Hill
where they are called the Pari, while a second clan, the Pajook, penetrated
southward into Acholi territory in to northern Uganda. (Collins 1971: 53)

Jal concurs with Collins in locating the original homeland of the Lwoo, their routes
of migration and the reasons for the migration. He further identifies the last settle-
ment of the Lwoo before their split into various groups as Wipach, east of the Bahr
el-Ghazal, near the Lake No:

Their tradition recalls that the Luo [Lwoo] managed to remain in harmony
at Wipach until the three cousins, Nyikang, Dimo and Gilo placed a quarrel
[sic] which led to the splitting up of the Luo migrants into three groups …
Dimo led a party of adherents and marched southwards … Nyikang led
another group … to the west bank of the White Nile and Gilo led a third
party and went eastwards, directly to the Bahr el-Jebel which they seem to
have crossed to the east. (Jal 1987: 111–12)

The underlying causes of the Lwoo migrations may have been, Crazzolara (1950: 33)
speculates, internal population pressure or displacement by the Nuer-Dinka group.2
The Anywaa emerged, as Collins (1971) argues, from a Lwoo splinter group.
Crazzolara (1950, see annex) identified the complex routes of Lwoo migration as
they are depicted here.
The views of Crazzolara (1950) and Collins (1971) reflect those of the Anywaa
themselves, who trace their origin to Gilo, the leader of one of the Lwoo splinter
groups. It is for this reason that contemporary Anywaa call themselves dibuoc gilo
(followers of Gilo) or kwar nyigilo (descendants of Gilo). They refer to a popular
myth to establish their common origin. This myth is about the aforementioned
quarrel among three brothers – Nyikang, Gilo and Dimo; it is about ‘getting one’s

2. The fifteenth and sixteenth centuries had seen waves of population movements involving the
Somali and the Oromo as well as the Nilotes in much of East Africa (Bahru 2001).
 The Anywaa Primordialist Ethnic Identity Formation 33

Figure 2.1 Routes of Lwoo migrations

34 Playing Different Games

own thing back’.3 In this narrative, the quarrel between Nyikang and Dimo is given
as the reason for the rupture between the Anywaa (Nyigilo), on one hand, and the
Shilluk (Nyikang) and the Jur-Luo (Dimo), on the other.
Some analyses in the secondary literature suggest, however, that the contempo-
rary Anywaa had diverse origins. For example, Perner (1997: 134–35), attempting to
reconstruct ethnohistory using clues provided by mythology, hypothesizes that
Anywaa history started with three leaders: Cuai, Gilo and Othieno. The existence of
people called Jowatcuai (a clan named after Cuai) seems to lend support to this
hypothesis. Cuai is mentioned in other myths as the chief of the Anywaa before the
emergence of the divine nobles. Another origin myth documented by Perner suggests
a separate origin for one of the Anywaa clans, the Jowatnaadhi:

[Out of the ashes of the skin of a divine goat] grew two calabashes: one had a
long and narrow neck (awido) while the other had no neck (amulo). People of
the village belonging to the chief of the Maro clan tried to open the calabashes
in order to make vessels out of them, when a voice cried from inside that the
knife was hurting. People got very afraid and threw the calabashes into the river.
The calabashes were carried away by the water. The awido calabash went down-
stream and reached the country where the Nuer people are living today; it was
cut and out came the ancestors of the Nuer. When the people tried to open that
awido gourd in the Nuer country, a voice warned them not to cut it. But the
people insisted and opened it by force. It appears that the calabash cried very
loudly, as the people had to put the knife in it many times until it was finally
split open. Because of the people’s disobedience and because they had inflicted
great pain on the calabash, the Nuer were cursed by it: ‘Was it you, Nuer, who
cut me?’ the calabash said. ‘Well then, your children’s heads shall be marked
forever’. The amulo gourd was seen floating in the river by children who took
it on land. Inside the calabash, there was a voice singing … The children
brought the calabash to the chief of the village, Cwai [Cuai] … It took a long
time to split the calabash, as the people tried not to penetrate with their spears
inside the calabash. Finally, it was split into two halves. In the right half of the
gourd, there was a human being and its placenta. In the left half of the gourd a
small lizard was found. This lizard is known as digwi Watnaadhi (lizard of the
Watnaadhi people). The child found in the gourd was taken home for care …
and he was adopted. [He was named Abek and became the chief of the village].
Abek prophesised that after his death his grave would turn into a mountain. On
the fifth day after his death, Abek turned into a mountain. The people evacu-
ated the village because they feared Abek’s spirit. (Perner 1994: 80–82)

The origin myth of the Jowatnaadhi clan is intriguing for a number of reasons. First, it
establishes a common origin for this Anywaa clan and the Jikany Nuer on the basis of

3. The idea of ‘getting one’s own thing back’ is featured in many Nilotic mythologies. In this
myth, Nyikang and Dimo quarrelled because they demanded, respectively, the same beads and
spear that they had lost to each other.
 The Anywaa Primordialist Ethnic Identity Formation 35

their shared name, Naath, an ethnonym of the Nuer. Evans-Pritchard (1940b) provides
further evidence of links between these two groups. There is, he notes, a similarity
between the honorific name of the Jikany Nuer and the honorific salutation of the
Jowatnaadhi clan: ‘The honorific name of the Jikany tribes is gatyou (nyayou) while the
honorific salutation of the Jowatnaadhi is nyooyu and … the Gaatgankiir, the dominant
clan of the Jikany tribes, like the Jowatnaadhi, believe that their ancestor came out of a
gourd’ (Evans-Pritchard 1940b: 32). So it seems possible that the Jowatnaadhi clan was
originally a proto-Nuer group which was later detached from the broader Nuer back-
ground; or it could have been a Nuer division that was incorporated into the Anywaa
during the Lwoo migrations or at a later stage. But the really intriguing question is why
this myth accounts for the origin of one Nuer section (the Jikany) but not for all of the
Nuer, who also call themselves Naath but do not share this myth of origin. Generally,
however, the calabash mythology is significant for seeming to suggest that the contem-
porary Anywaa have diverse origins – a point contradicted by most Anywaa discourse.
The supposition that contemporary Anywaa actually have diverse origins – which
receives further support from Evans-Pritchard’s suggestion that the Anywaa clan named
Jowatjaango may be traced back to the Dinka – might well be suggestive of an assimila-
tionist past probably made inevitable by the large-scale migration of the Lwoo. But despite
or perhaps because of these apparently diverse origins, the Anywaa espouse an identity dis-
course based on a strong ideology of a common origin, namely, common descent from
Gilo. In this sense, their own self-understanding corresponds to the standard definition of
an ethnic group as a community that is based on a subjective belief in a common origin.

Anywaa Social Organization and the Ideology of Purity

Anywaa primordialism also finds expression in an ideology of ethnic purity, which is
linked to distinctive practices of ethnic distancing, of descent and descent group
affiliation, and of alliance. At first glance, the evolution of the Anywaa identity dis-
course of ethnic purity seems paradoxical, given their Lwoo roots. According to
Crazzolara (1950: 45), ‘the Lwoo could never afford to be too particular in their
choice, and they have never been. The Lwoo do not despise a man because he is an
alien. These dispositions guided the Lwoo in their migratory march and have become
the base of their growth and greatness up to the present day’. Crazzolara does note,
however, that Lwoo assimilationist tendencies have their limits:

The Lwoo have never incorporated aliens into one of their clans as the Nuer
used to. Prisoner-slaves are joined or added to a family or clan group and
treated as blood-relatives. Such men would be given wives, or cattle to marry,
as members of the family; but with their children and descendants they started
their own sub-clan and social life … The real Lwoo autochthons of the group
always remained distinct and the real masters, the ruling clans; while the
assimilated, added clan groups always remained in the mind and tradition of
the people, descendants of ancient slaves. (Crazzolara 1950: 47–48)

Similarly, recent war captives were incorporated into Anywaa society, but they were
subjected to identity-distancing by the ‘pure Anywaa’. Descendants of Murle and
36 Playing Different Games

Nuer captives taken during the large-scale Anywaa counter-offensives against the ter-
ritorial expansion of the Nuer in the 1910s are still recognizable as a separate group
of people, referred to as Tung Akwei, named after the Anywaa king who famously
carried out the raids.
Anywaa notions of ethnic purity seem to be especially evident in social organiza-
tion and in practices linked to descent and descent group affiliation. The Anywaa are
divided into twelve clans called dho-oto: the Jowatcuai, Jowatnaadhi, Jowatmaaro,
Jowatjango, Joponguu, Jowatmaalo, Jowatong, Jowatluaalo, Jowatkaanyo,
Jowatyuaa, Jowatnaamo and Jowatmuongo (Evans-Pritchard 1940b).4 Besides these,
there is a royal clan called the Jowatnyiye, which is related to the Jowatong clan,
whose members lost their noble rank because of their failure to acquire the royal
emblems. Each dho-oto has its salutation, or math, i.e., the mode of address used for
both male and female members of the clan by members of other clans.
Membership in any given dho-oto is determined through patrilineal descent, but
other aspects of Anywaa kinship pertain to relatives of the mothers. For example,
each dho-oto is divided into descent groups called tung, defined by Evans-Pritchard
(1940b) as patrilineages; but a tung as understood by the Anywaa also includes the
maternal kin (Kurimoto 2000: 3). Each dho-oto also has an honorific title, or paak.
Girls take the honorific title of their father’s clan, and boys take the honorific title of
their mother’s clan. Thus, the patrilineal affiliation of the mother is relevant for her
children’s social identity; and in order to be considered fully Anywaa, one should be
born of Anywaa on both the father’s and the mother’s sides. In this sense, the Anywaa
conception of social status institutionalizes an identity discourse that centres on the
ideology of ethnic purity.
Ideally, then, a person needs to be Anywaa on both paternal and maternal sides
of his or her family in order to claim full Anywaa ethnic identity; otherwise, he or
she will be socially disadvantaged, deprived of both honorific title and clan saluta-
tion. This emphasis on purity discourages inter-ethnic marriages. Children of mixed
ethnic background are subjected to social discrimination. They are referred to by the
derogatory term jur, which literally means ‘foreigner’. In sum, marriage practices
have served as mechanisms for the active maintenance of an ethnic boundary. The
names of children born of inter-ethnic marriages often signify their ‘foreign’ back-
ground. As a result, few of the ethnically mixed children live with their Anywaa
parents. This stands in contrast to the practices of the Nuer, who give names featur-
ing the integrative motif to children of ethnically mixed parents.5
The ‘pure’ Anywaa employ naturalistic arguments in justifying their own prac-
tices of ethnic purity. For example, in the following statement, an informant from
Pinykew village explains Anywaa scepticism regarding adoption and assimilation
with reference to supposedly self-evident truths derived from biological origins:

4. The prefix Jo- means ‘people of ’.

5. Nyanuer is, for example, a common Nuer name given to a daughter with a mixed ethnic back-
ground. It means ‘daughter of the Nuer’. Ugaala, on the other hand, is a typical name that the
Anywaa give to a person born of an Anywaa father and Highlander mother. It means ‘he has
become a Highlander’.
 The Anywaa Primordialist Ethnic Identity Formation 37

This is the way we were created. We do not take other people’s things.
Stealing of whatever sort is bad. It kills you. The Anywaa ask about origins.
You might adopt a child, but there is always a danger that when he grows up
people might bother him. Although the adopted child grows up with the
family, people might remind him of his background. For that reason the
Anywaa do not take other people’s children. It is different with the Murle and
Nuer. They steal people, just like they raid the cattle. (elder Ujulu Omot,
Pinykew village, June 2000)

A Particularistic Medium of Social Exchange

Unlike their pastoralist neighbours, such as the Nuer, the Dinka and the Murle, the
Anywaa do not ascribe a crucial role to cattle in creating and maintaining social rela-
tionships. In traditional Anywaa society, fundamental social relationships between
leader and commoner, elder and junior or husband and wife were created through
the medium of beads. The traditional Anywaa bridewealth is not cattle but the blue
glass beads called dimui (see Figure 2.2). The social significance of the beads is so
fundamental that Evans-Pritchard called the Anywaa the ‘beads people’ (1940b: 20).
The origins of dimui are obscure. In their oral traditions, the Anywaa say that the
ancestors brought the beads with them in ancient times. According to other state-
ments, dimui were brought to the Sudan from Egypt by Ottoman traders. Be that as
it may, dimui is for the Anywaa the scarce good par excellence. There is a finite
supply of dimui, which cannot be replenished and which are transferred from one
kin-based group to another in the form of bridewealth, bloodwealth, ransom or
other kinds of payment. Traditionally, the number of dimui one possessed deter-

Figure 2.2 Anywaa woman with dimui necklace (photo: Dereje Feyissa)
38 Playing Different Games

mined one’s status. The beads possessed by a family formed a kind of heritage, which
was controlled by elders and which, therefore, gave them tremendous power over
juniors. Men were also dependent on their sisters, inasmuch as they could usually
only gain access to dimui through the bridewealth that they received when their
sisters married. With the beginning of wage labour in the 1950s, especially in coffee
plantations in the neighbouring highlands, and with the beginning of salaried jobs
and gold mining, it became possible for men who lacked sisters or daughters to buy
dimui; but because of its fundamental scarcity, it always eluded commercialization.
Sometimes, perhaps often, men without sisters could gain access to dimui, if only
indirectly, by placing themselves in the service of a noble (nyiya, pl. nyiye) or a
headman (kwaaro, pl. kwaari), who then served as their patron. When the client had
reached the age that entitled him to marry and had rendered sufficient service to his
patron, especially through agricultural labour in his gardens, then the patron would
assume responsibility for paying the bridewealth. The nobles and headmen were able
to accumulate dimui by means of imbalanced reciprocity, to modify Sahlins’s famous
term. While the nyiye and the kwaari received dimui for their daughters’ marriages,
they were not required to pay dimui for their sons’ marriages; rather, it was the sons’
maternal uncles who assumed responsibility for these payments. This created a one-
way flow of the resource on which power was based, filling the treasury of the nyiye
and the kwaari with one of the most desired cultural objects. At the same time,
however, the nobles’ and headmen’s dimui was considered to be public property,
insofar as these offices served as redistributive centres for dimui-poor families. In
return for their crucial support in helping poor people marry, the nyiye and the
kwaari could build a constituency based on their networks of clients.
Besides serving as a medium of exchange among the Anywaa, the dimui are a
marker of Anywaa identity in the eyes of the Anywaa themselves and in the eyes of
their neighbours.6 Because the dimui are valuable and scarce, the Anywaa make
efforts to prevent their loss to the community through bridewealth payments to out-
siders. In this way, the dimui and the practices related to them encourage ethnic
endogamy, place a limit on inter-ethnic marriages and, more generally, reproduce the
ethnic boundary between the Anywaa and others. One consequence of this effective
policing of the ethnic boundary, however, is to limit the demographic size of the
Anywaa, particularly in comparison to assimilationist neighbours with an expanding
cattle-based bridewealth system.

Descent and Territory as Structuring Principles

There were two kinds of political communities in traditional Anywaa society: the ji-
nyiye (‘peoples of the nobles’) and ji-kwaari (‘peoples of the headmen’). The jurisdic-
tion of both the nyiye and the kwaari were confined to specific territories. Leading
scholars have found Anywaa territoriality to be worthy of special commentary.
According to Evans-Prichard (1940b: 37), for example, ‘the Anuak are strongly

6. Besides the dimui, there are at least five other kinds of beads that are peculiar to the Anywaa and
that have specific meanings and uses; for example, the uchuok, which serve as royal emblems.
 The Anywaa Primordialist Ethnic Identity Formation 39

attached to the sites where their ancestors lived and often tenaciously occupied them
in the face of extermination.’ In a later work, this same author illustrates the close
identification between the Anywaa village and the dominant lineage, or jobur, of the
village, by emphasizing the precarious status of newcomers: ‘however long strangers
and their descendants live in a village and however much they intermarry with …
[the jobur], they can never become members of it but remain welo, strangers’ (Evans-
Pritchard 1947: 93). According to Kurimoto (1992: 4) ‘the Anywaa possess a clear
notion of their territory’. Perner (1997: 144) described Anywaa territoriality in a
similar way: ‘Because of fighting – but also because of floods – the history of Anyuak
settlement never ended. But if one site had been chosen as the definite home of one
community and had been inhabited for several generations, its people would never
give it up definitively but would always try to return and re-occupy it’.
Anywaa territoriality is embedded in the institution of the earth priest known as
the wat-ngomi. The wat-ngomi is selected from the jobur. In Anywaa cosmology, the
sphere of the earth has a spiritual dimension, a quality which, according to Perner
(1997: 51), is expressed in their daily greetings: piny bede nidi (‘how is the earth/your
land?’, ‘how are you?’) to which one answers: piny be`r yak (‘the earth is just fine’) or
piny rac (‘the earth is bad’, ‘things are bad’). The relationship between earth and the
people living on it is a very intimate one. The earth is respected, as it is considered a
site for the piny kwari (ancestors). According to Perner (1997: 77), the wat-ngomi is
‘the representative of the earth and of the earthly substance of human life and can,
because of his authority, urge the matter of earth to fulfil its natural duty of protect-
ing a foreigner who has come with merely friendly intensions’. This ritual office is
hereditary but with no political rights derived thereof. Besides ensuring fertility
(human and agricultural), and maintaining the ‘dignity’ of the earth, the wat-ngomi
also ensures the separation between the human and animal territories. When wild
animals threaten a village, the wat-ngomi performs a ritual to protect the villagers.
The village is the principal unit of social identification among the Anywaa. A
village is identified with a dominant lineage but, despite a strong preference to stay
in one’s own lineage village, there are always strangers, the welo. It is the lineage of
the first-comers who become the dominant lineage of a village. A clan per se is not
associated with a particular territory. The twelve Anywaa clans are dispersed nearly
in all the regions, although there might be a concentration of one clan or the other
in a certain region. The Jowatcuai and the Jowatnaadhi, for instance, are prevalent,
respectively, in the Adongo and Akobo regions. With the exception of the noble clan,
a commoner clan as a whole has no precise genealogical structure (Evans-Pritchard
1940b: 28). As shown in Table 2.1, more than six clans are spread along the Baro
River, collectively known as the Openo Anywaa.
Members of the Jowatnaadhi, the Jowatcuai and Jowatnaamo clans live in three
different villages each. The Jowatnaadhi are also prominent in the Akobo region,
whereas the Jowatcuai are also found in the Adongo region and so on.
Clusters of villages form regional groupings. As Perner (1997: 88) noted, ‘when
Anyuak think of their country, they do it in terms of regions and distinguish eleven
factions’. These Anywaa regions are Openo, Rawanye, Jor, Ciro, Lull, Bat Gilo,
Nyakani, Ojwa, Adongo, Tiernam and Thim. A region is not a political unit and it
40 Playing Different Games

Table 2.1 Anywaa villages and clan distribution along the Baro River

Village Dominant lineage Name of clan

Emetho Pelow Jowatnaamo

Pinymau Pubala Jowatnaamo
Etiel Pompai Jowatnaadhi
Pole Tung Gilo Jowatnaadhi
Pinymole Pakini Jowatnaadhi
Pijuwo Ejuwo Jowatcuai
Emetho Pelow Jowatnaamo
Pimoli Pinygilo Jowatcuai
Enyuwey Pumalo Jowatjango
Edeni Puguta Nyguni Jowatcuai
Itang Kir Tungwit Jowatcuai
Pinymou Pinykuey Jowatongo
Abol Tumbela Jowatmaaro
Akedo Ekan Jowatmaaro
(Source: author’s field notes)

primarily designates ecological conditions. However, at times, the regions or cluster

of regions might form a political block or units of social identification. In contem-
porary Anywaa politics in the Gambella region, for instance, political alliances are
often framed in terms of the ecological differences and separate political history of
the Openo and the Lull. The Openo are those who live along the Sobat River and its
tributary Baro River. When used in the Anywaa language as an ecological category,
lull means ‘forest’, specifically the area between the Baro and the Gilo Rivers, ending
in the east at the foot of the Ethiopian escarpment. In present-day Gambella the dis-
tricts of Abobo and Gok are considered Lull regions, but Lull is also used by the
Openo Anywaa to refer to all the groups outside of the Baro/Sobat basin.
Ideally, the Anywaa fix descent with territory, but practice has it that there are
always welo in many of the villages. Nevertheless, the welo permanently remain as
guests and are always treated as ‘second-class’ village citizens. The Anywaa say, ‘One
must be a very bad person to leave his natal village in the first place and settle with
others’. Guests of a temporary or permanent nature are welcome but they are not
fully integrated into the jobur; they might contribute to the economic or military
strength of the village, but they can also leave at any time. As will be discussed in
Chapter 3, the Anywaa and the Nuer differ in their conception of a stranger. A
stranger is marginal among the Anywaa and he is not fully integrated into the centre.
Among the Nuer, a stranger plays a crucial role in the affairs of the centre, particu-
larly in the competition for vital resources and political power.
Anywaa territoriality was originally a general conception of the spiritual dimen-
sion and supportive nature of the Earth. But the longer the familiarity with a partic-
ular territory, not only does the attachment to it become greater, but it also becomes
a more constitutive part of local identity. Unless there are good reasons to leave, such
as epidemics, pressure by neighbours or crop failure, the ideal is to remain perma-
nently in one’s natal village. In extreme cases, the Anywaa can be re-rooted to a new
 The Anywaa Primordialist Ethnic Identity Formation 41

village, but this occurs only as a last resort. Even when they leave their villages, the
Anywaa keep the names of their former villages. Resettlement in a new village
requires an elaborate ritual of ‘befriending’ the new earth.
The exclusivity of an Anywaa village is expressed in an anthem known as agwaga.
Each Anywaa village as a territory has boundaries (kew), known to both its own
inhabitants and those of other villages. According to Perner (1997: 180–81), ‘each
village in fact does have its territory with boundaries, well known by everybody […]
the borders of a village’s territories were outlined by runners who went to circum-
scribe the limits of a site, fixing certain points (such as trees, mouths of rivers, etc.)
as boundary posts. The clear demarcation of a territory is extremely important as it
helps to avoid conflicts between people of different territories’.
The Anywaa’s strong emotional ties to their own territory contrasts with apathy
towards other people’s land. Evans-Pritchard (1947: 64) reported that there were
more than one hundred and twenty Anywaa villages in Ethiopia, and Perner (1997:
86) mentioned that there were forty to fifty such villages in the Sudan. Until the
1970s, there were more than forty autonomous villages alone along the Baro River
in Gambella, which formed distinct local identity units, each associated with a dom-
inant lineage. The Anywaa notion of territoriality is different from the notion of
autochthony as applied, for instance, in West African cases, where groups claim to
have originated from a particular territory or where some lands could assume a sacred
status (Lentz 2000). The Anywaa have no such holy places. Some parts of a territory,
such as nobles’ graves, streams and extraordinary rocks, assume a spiritual dimension
and are believed to be inhabited by a supernatural being, Jwok. However, these
abodes of Jwok in ‘human territory’ are not revered, but avoided.
More than half a century after Evans-Prichard’s (1940b) observation, a similar dis-
course on territoriality still exists among the contemporary Anywaa. In March 2001,
for instance, the Adongo Anywaa were evicted by the Openo Anywaa from the village
of Akedo by the Ekan lineage after an interpersonal quarrel became politicized on the
basis of origin. The Adongo Anywaa belong to the Jowatcuai clan whereas the Ekan
dominant lineage in Akedo belongs to the Jowamaaro clan. The Adongo Anywaa first
came to Akedo village in the 1950s, when the American Presbyterian Mission estab-
lished a clinic there. Like the Adongo Anywaa, many other Anywaa and Nuer flocked
to Akedo village in order to have access to the new welfare services provided by the
church, giving the village a more multi-village and multi-ethnic composition. But
neither the Nuer nor the Anywaa migrants were ever fully integrated into the Akedo
village community, although a significant number of them intermarried with the
Ekan. After they were evicted from Akedo village, the Adongo Anywaa resettled in
Gambella and Itang towns. Interestingly, they did not use their half-century of local-
ization as an argument for entitlement to remain in Akedo village. In not doing so,
both the Akedo and the Adongo villagers played the same language game. When they
use the term Akedo, they mean it belongs to those Anywaa whose ancestors come
from Akedo. Identification on the basis of territory is often described in the literature
as inclusive (civic) as opposed to an exclusive ethnic form (Brown 2007). The Anywaa,
however, intertwine descent with territory, which makes their mode of identification
a ‘thick’ primordialism. ‘Anywaaness’ is not only defined in terms of purity of blood
42 Playing Different Games

but also in having a location in a particular village. Residence in a certain village does
not mean ‘belonging’ to it, a mode of identification which sharply contrasts with the
Nuer. As will be shown in the next chapter, residence in a village among the Nuer
guarantees one a secure local identity, whereas newcomers are viewed as political
resources in the competition over community leadership.

The Anywaa Political Organization

Anywaa territoriality also finds expression in characteristic forms of political organi-
zation, which are quite distinct from the political organization of their neighbours.
The relatively centralized ‘village state’, which has been described in detail by Evans-
Pritchard (1940b) and Lienhardt (1957/58), contrasts sharply with the political
organization of their pastoralist neighbours such as the Nuer, whose decentralized
political system has been described by anthropologists as ‘ordered anarchy’ and which
is more contemptuously referred to by the Anywaa as ‘chaotic’.
As has been seen in the preceding section, there were two kinds of political com-
munities in traditional Anywaa society: the ji-nyiye (people of the nobles) and the ji-
kwaari (people of the headmen). Despite some differences in their political status,
both the nyiye and the kwaari were attached to specific villages, and they rarely
embarked on missions of territorial expansion. While inter-village warfare was rife
among the Anywaa (Evans Pritchard 1940b; Shumet 1986), none of these inter-
village fights resulted in territorial encroachments. The object of the struggle between
the various nyiye and kwaari was not territory but royal emblems or emblems of
authority. Traditional Anywaa politics is noted for its rituals (see Figure 2.3). The
insignia of the offices of headmanship and kingship consisted especially in additional
forms of precious beads; in particular, a string of beads called the abudho for the
kwaaro and a necklace called the uchuok for the nyiya. Unlike the dimui of the
common people, the abudho and uchuok cannot be obtained by exchange; they must
be inherited within ruling lineages. In addition to the uchuok, there is a second royal
emblem, the walo (the royal stool). The Tooth Drum, three spears and iron fork are
also regarded as royal emblems. The nyiya who controls the uchuok and the walo is
called the nyinya. The nobles were found in the east and south-east part of the Anywaa
land and that of the headmen in the rest of the country (Evans-Pritchard 1940b: 38).
The royal emblems, however, are not monopolized entirely by the nyinya, as he
is required occasionally to allow other nyiye to wear the uchuok and to sit on the walo
to demonstrate their nobility. Nevertheless, the competition for the position of
nyinya was a constant source of warfare. In the sixteenth generation after the foun-
dation of the Anywaa kingdom by the descendants of Goora, the royal line was
divided into two: the tung udola and the tung goc.7 By the late nineteenth century,
the tung goc had left the Adongo region and established a new political centre at the
foot of the Alwero River on the eastern fringes of Anywaa land (the present-day
Abobo district of the Gambella region), thus extending the nobles’ political sphere of

7. Goora had two sons, Udiel and Apur. The mother of Udiel was called Akew Nyigoc and the mother
of Apur was called Ochuro Nyiudola. The two lineages were named after the respective mothers.
 The Anywaa Primordialist Ethnic Identity Formation 43

Figure 2.3 Nyinya Adongo’s coronation at Utalo village, 28 March 2001 (photo: Dereje Feyissa)

influence farther east. The royal emblems alternated between the tung udola and the
tung goc until, in the mid-1920s, the British fixed the royal centre in the Adongo
region in present-day southern Sudan (eastern Anywaa land), further dividing the
two royal lines.
The political system of the Anywaa is characterized by a kind of ritual kingship,
which Evans-Pritchard (1940b: 137–38) describes:

The Anuak kingship is a good example of the ritual character of African king-
ship … The royal emblems have only a ritual value. Their composition sym-
bolizes some of the more important objects in Anywaa culture, beads, spears
and drums …The kingship is indivisible. It can circulate, but there cannot be
more than one king at the same time … The kingship is not an office with
ritual duties but is itself a ritual object … It is the acceptance of a common
value, and not corporate action, which constitutes the polity.

Comparatively, the nyiye hold more power than the kwaari, but the office of the
headmen is also buttressed by ritual authority. For example, men bow low and
women kneel down in front of the kwaari – a courtesy called gungi. As in the case of
the nyiya, courtesy is due primarily to the institution of headmanship, not to the
headman himself (Lienhardt 1957). In return for the prestige and respect that the
incumbent of the office of kwaaro receives, he is expected to provide regular feasts.
Neither the nyiya nor the kwaaro wields absolute power. In fact, the instability of
leadership is characteristic of Anywaa political history. Leaders are respected inas-
44 Playing Different Games

much as they continue serving their people by sponsoring feasts, helping the poor to
marry and leading their followers to victory. In extreme cases, Anywaa leaders, par-
ticularly the kwaaro, may be removed violently from office for failing to offer such
services, either in fact or in allegation. This is known as agem, which Lienhardt
(1958: 31) called ‘village rebellion’, i.e., violent efforts on the part of members of
certain factions aimed at replacing a discredited leader by a new favourite.
The Anywaa mode of governance, kingship and headmanship ties people to a spe-
cific territory (villages), and its distinctiveness reinforces their sense of who they are. The
Anywaa take pride in their political organization as an index of ‘civilization’ and often
contrast it with ‘stateless’ pastoralist neighbours such as the Nuer and the Murle. Anywaa
traditional leaders, particularly the nyiye, provide ideals of social behaviour and are the
bearers of the ideology of purity. Ideally the nyiye do not marry into other ethnic groups.

Anywaa Cosmology
The Anywaa primordialist identity discourse is also embedded in their cosmology. In
the post-Lwoo migration period, the Anywaa seem to have developed an elaborate
worldview that has, to a large extent, defined their sense of who they are. In Anywaa
cosmology the relevant ‘others’ serving to help orient self-understanding are Jwok
and animals. The Anywaa self-image is expressed in the concept of luo. The term luo
refers to human purity. In Anywaa cosmology, luo is contrasted with Jwok, the world
of spirituality. Jwok is the principle of creation and is represented by its dual nature,
the forces of creation (nyingalabuo) and the forces of destruction (nyidungu). The
Anywaa belief system recognizes three ‘spheres of existence’, which Perner (1994)
calls the Human Person, the Sphere of the Earth, and the Sphere of Spirituality
(Jwok). Each is entitled to ‘exist’ within its own specific sphere. The Anywaa do not
subscribe to the concept of a supernatural being that ‘governs’ the natural world.
They acknowledge the precedence and primacy of Jwok, but the dominant pattern
of relationship between Jwok and human beings is oppositional. In most Anywaa
creation myths, human beings appear to have been an accidental product of Jwok’s
creative work and to have survived not because of Jwok’s will but despite it
(Lienhardt 1962: 77–78; Perner 1994: 57–68).8 In Anywaa identity discourse, there-
fore, human existence, particularly that of the Anywaa, has to be defended against
‘encroachments’ by Jwok. Jwok is not envied for its power, but is merely contrasted
with humans. In this belief system, there is no ‘original sin’ and subsequent ‘fall’ as
in the Judeo-Christian tradition. The source of human suffering is attributed directly
to the malevolence of Jwok (nyidungu). The only positive aspect of Jwok is creation;
but, even so, Jwok regretted having created a conscious being that was similar neither
to the animals nor to the spirits. In fact, in the creation myth, the Anywaa owe their
survival to the caring Medho (dog), not to the creator Jwok.
The persistent themes in Anywaa–Jwok relations are those of confrontation,
abandonment and hostility. When the destructive side of Jwok is manifested, as in

8. For a full account of Anywaa creation myth, see Perner (1994), Living on Earth in the Sky: The
Anyuak, Volume I, The Sphere of Spirituality, pp. 57–68.
 The Anywaa Primordialist Ethnic Identity Formation 45

droughts, epidemics, infertility and floods, people openly revolt and chase Jwok out
of their villages. Such acts of resistance against Jwok are expressed in collective
rituals. One of these rituals is pö, which Perner describes (1994: 143) as a ‘manifes-
tation against the presence of Jwok in human territory … it takes place in the case
of general death in the village, when Jwok has become a common disease’. The
Anywaa ‘chase’ Jwok from the human territory by making a human noise. Perner
(1994: 143) presents the ritual as follows: ‘At the sight of the new moon, the Odolo-
drum will be taken out to the dancing place of the village and beaten by elders;
women will take the skins out of their huts and beat them with a stick, thus clean-
ing them from the dust; the men of the village and the children will clap their hands
and shout “ci-ya, ci-ya, ci ki moi ya” (Go-yaah, go-yaah, go with your own)’.
The Anywaa rebellion against the evil Jwok alternates with appeasement. When
a woman is pregnant and about to give birth, Jwok might appear to her in dream.
When this occurs, the Anywaa say that Jwok is jealous of the creative capacity of
humans. Ultimately, however, the creative principle is attributed to Jwok: ‘The
women will in such a case symbolically give her child to Jwok, naming it after that
particular spirit; beads of certain colours particular to that Jwok are to be tied around
that child’s neck, thus demonstrating its affinity with Jwok but also to concentrate
Jwok in those particular beads and to prevent it from entering the child’s body’
(Perner 1994: 115).9 The Anywaa are ambivalent about Jwok, which they recognize
as the ultimate source of power. They also attribute ultimate justice to Jwok, at least
in its benevolent capacity (nyingalabuo). But the Anywaa do not pray to Jwok for
justice; rather, they invoke it, even against Jwok’s destructive features. As Lienhardt
(1962: 75) puts it, ‘Anuak have produced headmen and nobles, but not significant
priests or prophets’. Although they actively resist Jwok, the Anywaa also connect
with it discursively, for example, in the name they give to themselves: nyilwinyjwok,
‘unique among Jwok’s creations’.
The Anywaa worldview and the identity discourse it entails are expressed in dom-
inant symbols such as the sexual features marking the boundary between humanity
– luo or the Anywaa – and Jwok. If Jwok is primarily defined as pure spirituality, luo
is physical, a quality that consists principally in having ‘proper’ sexual organs.
‘Proper’ sexual organs must not exhibit any anomalies, such as peculiarities in the
number and shape of the testicles, an incompletely formed vagina, or a penis that is
disfigured naturally or through circumcision.10 In Anywaa identity discourse not all
of those who seem to be human satisfy the aforementioned criteria. There are those
who ‘look like’ humans but who in reality work for Jwok (and its various manifesta-
tions) and participate in the destruction of humanity. Such anomalous beings are
called padhano, ‘nonhuman’. The padhano are greatly feared because they work

9. In the Anywaa belief system Jwok can manifest itself as various spirits at various times.
‘That spirit’ and ‘that Jwok’ refer to a particular manifestation at a particular time.
10. The Anywaa believe that Othieno, an Anywaa culture hero who is believed to have brought
technical knowledge and equipment to the Anywaa country, was socially rejected because of
his circumcised penis. An Anywaa oral tradition has it that even when Othieno turned into a
lizard after his rejection, the lizard was still recognized as circumcised (Perner 1994: 182).
46 Playing Different Games

within ‘the human territory’ as fifth columnists, to borrow a military metaphor. The
padhano are not only destructive but also indistinguishable, for, although their fea-
tures are known, it is difficult to detect them, except in their childhood. Since adult
padhano are not easily recognizable, children are carefully watched. One of the dom-
inant events after childbirth is checking the testicles or the vagina of the infant. If
any of the aforementioned signs are observable, the child has to be disposed of, lest
it grow up to be a padhano (Perner 1994: 182). From the perspective of the Anywaa,
the witchcraft of the padhano is not motivated by personal envy or by any of the
factors known from standard sociological explanations for witchcraft. For the
Anywaa, witches simply work for Jwok in furthering the ultimate destruction of
humanity. That is why the padhano are called ci-Jwok, ‘wives of Jwok’, modelled on
Anywaa gender relations, which subordinate women to their husbands.11 The
padhano operate by spiritually transplanting bones in their victims. This is variously
called ramo or aramo. The Anywaa believe that when a ci-Jwok looks at a human
being, a bone will be transplanted into that person’s body. The bone can only be
removed through an elaborate ceremony performed by an ajuwa (ritual expert).
If the boundary between humanity and Jwok is drawn along natural lines (sexual
organs), the boundary between humans and animals is marked by artificial means
(Perner 1994: 247). The second important symbol of humanity (‘Anywaaness’) is
naak, the extraction of the lower incisors of humans in order to distinguish them
from animals. The canines are destroyed early in infancy, and the four lower perma-
nent incisors are removed at about the age of ten. Many Nilotic societies, including
the Nuer, practice dental evulsions, but none of them attribute a fundamental
meaning to it as do the Anywaa.12 Although the relationship between animals and
humans is not fundamentally oppositional, animals are represented in the creation
myth as part of Jwok’s conspiracy against humanity. Unlike humans, animals are not
rebels against the ‘kingdom of Jwok,’ and their relationship to Jwok is believed to be
peaceful. Still, animals share an ‘earthly existence’ with humans, and some animals,
such as dogs and chickens, have a close relationship with the social world of the
Anywaa (Perner 1994: 216).

Constructed Primordialism
The Anywaa primordialist ethnic identity formation, like any other forms of identi-
ties, is socially constructed. Various social identities may differ, however, in their
mode of construction in various historical contexts and in response to specific socio-
political processes. In this section I discuss the particular circumstances that under-
lie the construction of the seemingly primordialist ethnic identity of the Anywaa.

11. This does not mean that all ci-Jwok are women. It is used metaphorically to express the power
relationship between Jwok and humans; men can also serve Jwok as ci-Jwok.
12. Schlee (1994: 133) has analysed extraction of the teeth among the Rendille as an emblem
serving to enhance visibility in intergroup interaction: ‘The Rendille refer to Gabra or Boran
as “the enemies with teeth” because these northern neighbours do not remove the two central
lower incisors’.
 The Anywaa Primordialist Ethnic Identity Formation 47

Relative isolation and social introversion

It seems likely that the ethnogenesis of the Anywaa was affected by the peculiarities
of the interactional setting in which it occurred. In contrast to the settlement history
and the corresponding intergroup relations of other Nilotes, which were character-
ized by protracted warfare, there is no historical account of major military confronta-
tions between the Anywaa and the pre-Anywaa settlers of the present-day Anywaa
settlements in the Sobat basin. According to Jal, ‘on the Sobat the Anuak migrants
appear to have stayed peacefully for a long time. They did not experience any threat
of attack from their immediate neighbours – the Thoi Dinka who were far deep
inland off the south bank of the Sobat and the Shilluk who appear to have concen-
trated their settlements on the west bank of the Nile’ (Jal 1987: 116).
Oral traditions also suggest that the pre-Anywaa settlers of the present-day
Anywaa territories were the Burun, a non-Nilotic Koman-speaking people. According
to the late nyiya, Agada (quoted in Perner 1997: 138), ‘the Burun were originally in
possession of Anyuak land. They were fought, decimated and enslaved’. Much
remains to be learned about the early history of the Burun, but there is no indication
that they were incorporated into Anywaa society. Instead, they seem to have resisted
the Anywaa but to have been defeated for reasons that are not yet fully established.
This is true particularly for those Burun (Opo) who live near Gambella town, to
whom the Anywaa refer with the derogatory term lango, or slave (Perner 1997: 138).
Given the weakness of the neighbours of the Anywaa in their new homeland, it
seems fair to say that Anywaa ethnic identity evolved in relative isolation – unlike
that of other Nilotic groups in the region, such as their closest Lwoo relatives, the
Shilluk. None of the ethnographers and historians of the Anywaa have reported pro-
tracted fights and hegemonic struggles between the Anywaa and any of their neigh-
bours before the nineteenth century. At the same time, much is known about
inter-village warfare among the Anywaa, which has shaped the political identities of
the villages to a significant degree.
The Anywaa’s neighbours to the north and east are the Oromo Highlanders and
the Majangir, respectively. The social and political relationship between the Anywaa
and the Oromo before the arrival of representatives of the Ethiopian state was char-
acterized by reciprocal exchange (Kurimoto 1992). The neighbouring Illubabor and
Wellega highland regions of the Oromo are rich in resources. In contrast to the
northern highlands, whose inhabitants have incessantly encroached on bordering
lowlands in the north-west, there is no land shortage in the western highland
regions.13 The other eastern neighbours of the Anywaa are the Komo and the
Majangir, who are numerically and militarily weak. In fact, by the 1940s, the
Anywaa of the Abobo and Gog districts had reduced the Majangir to the status of
vassals (Evans-Pritchard 1947: 73). This contrasts quite sharply with the ethnogene-
sis of the Nuer: in the formation and reproduction of Nuer ethnic identity, inter-
group warfare played a central role.

13. For a historical and comprehensive account of the land pressure on the north-western low-
lands, see Wolde Selassie Abute (2002), Gumuz and Highland Settlers, Ph.D. dissertation,
University of Göttingen.
48 Playing Different Games

While it is wise to be suspicious of theories of ethnogenesis positing a general

development ‘from isolation to interaction’, it still seems that significant differences
in the nature and intensity of intergroup relations in the broader interactional milieu,
as it changes over time, should be taken into account when explaining the develop-
ment of particular ethnic groups and their self-understanding. During the Lwoo
migration, the Anywaa were part of a larger Lwoo community, the various subgroups
of which made contact with different peoples in various ways. Following the ‘epoch-
making’ migrations, as they have been described in the secondary literature, the
Anywaa seem to have experienced a relatively low density of inter-ethnic interaction
(Crazzolara 1950/54; Ogot 1967; Collins 1971). This ‘low’ interactional density and
the corresponding absence of a ‘relevant other’ in the form of threatening neighbours
seems to have induced an internal mode of ethnic boundary formation, namely, the
Jwok/luo relational template, as described above. It was not until the second half of
the nineteenth century that the Anywaa began to experience sudden and massive
encounters with two of their new, powerful neighbours: the Nuer in the west and the
resettled Highlanders in the east.

Changes in the material conditions of life

Other factors contributing to the primordialist self-understanding of the Anywaa,
particularly to Anywaa territoriality as described above, include changes in the mate-
rial conditions of life. Anywaa territoriality has, it seems, been strongly affected by the
change from a pastoral to an agrarian mode of production. Many Nilotic people are
pastoralists, but the significance of pastoralism varies from one group to another. The
Anywaa are, arguably, the group that has moved furthest away from pastoralism and
the associated cultural practices; this, in turn, appears to have induced a significant
shift in values. Major indices of this shift in values have already been described. One
is the change from a bridewealth system based on cattle to one based on scarce beads;
another is the development of a strong attachment to the land, which has gained sym-
bolic significance within an ideology of territoriality. One of the most noticeable fea-
tures of the Anywaa vis-à-vis other Nilotic societies is their agrarian economy, based
largely on riverine maize and sorghum cultivation, an economy which is very much
embedded in the local ecology. Most of the Anywaa villages are found along rivers. It
is for this reason that the Anywaa are often called ‘River Nilotes’ (Perner 1997: 103).
In some areas, such as Jor district, the Anywaa still keep some cattle, but pastoralism
has a minimal significance in their livelihood and social relationships.
Various scholars have attempted to explain why the Anywaa stand out as ‘cattleless’
people and have become cultivators among their pastoralist Nilotic neighbours, despite
indications that they were once pastoralists themselves. Evans-Pritchard (1940b: 20) was
the first to note the absence of cattle values among the Anywaa, a ‘lack of that deep feeling
truly pastoral peoples have for their herds’. Perner (1997: 21), however, refutes Evans-
Prichard’s representation of the Anywaa as ‘uncaring towards their cattle’. Through oral
testimony, he establishes the recent conversion of the Anywaa from an agropastoralist to
a largely agrarian life, and he also provides other kinds of evidence, most notably ecolog-
ical: the presence in Anywaa territories of the tsetse fly (Trypanosomiasis), which infects
 The Anywaa Primordialist Ethnic Identity Formation 49

the cattle. Large parts of Anywaa land are reputed to be unfavourable for cattle keeping,
owing to the hot and humid climate, which promotes a great number of other diseases.
But such accounts fail to explain why the Anywaa remain cultivators even in areas where
cattle could be kept, as the new Nuer settlements in Anywaa areas have demonstrated.
The abandonment of the pastoral economy, notes Perner (1997), might have been caused
by pressure from the Anywaa’s pastoralist neighbours, who, presumably, raided existing
Anywaa herds and stole whatever cattle the Anywaa had.
Finally, Perner (1997: 21) suggests an additional economic reason for the decline
of the pastoral economy, namely, the presence of game and fish, which provided
alternative sources of nourishment: ‘the seasonal presence of thousands of white-
eared cob antelopes [called anger in Anywaa] in their country may well have influ-
enced the Anywaa’s decision to lead a more sedentary existence and to concentrate
on agriculture, hunting and fishing’. The skin of these antelopes is still an important
individual possession, even in areas where anger is no longer found. The abundance
of fish supplies, particularly for the Openo Anywaa, who live along the Baro River,
may have had a similar effect. Fishing is one of the crucial economic activities for the
Anywaa and fish forms a significant part of the Anywaa diet. Thus, regardless of
precise relations of cause and effect, Anywaa territoriality corresponds to their mate-
rial condition of life. Their agrarian lifestyle ties people to specific territories.

Changes in the nature of the inter-ethnic setting

The sudden and massive encounter with new and powerful neighbours appears to
have reinforced Anywaa proclivities for primordialist forms of self-understanding.
After centuries of relative isolation, the Anywaa began in the nineteenth century to
experience territorial encroachments by two of their pastoralist neighbours, the Nuer
and the Murle. In subsequent struggles, the Anywaa lost large expanses of territory
along the Pibor, Akobo and Baro Rivers to both of these groups. These pastoral ter-
ritorial encroachments, primarily driven by the desire for access to the riverine lands
in the Sobat basin, represent not only a loss of land but also a violation of a central
symbolic resource in Anywaa identity construction. Reacting to these expansive pas-
toral systems, the Anywaa have invested in primordial forms of boundary marking.
Primordialism here emerges as a form of resistance, a counter-hegemonic project
extending beyond an ‘innate’ feeling developed on the basis of self-reflection.
The expansion of the Ethiopian state and the arrival of its representatives in the
areas occupied by the Anywaa have also meant territorial and cultural challenges to
the Anywaa way of life.14 The expanding presence of the Ethiopian state has entailed
changes in the demographic structure. In contrast to the reciprocal relations between
the Anywaa and the neighbouring Highlanders in the pre-state era, the recent reset-
tlement of Highlander farmers from different parts of the country, which peaked in
the 1980s, has resulted in losses of Anywaa territories and in subjection to various
forms of cultural hegemony. Pressed by pastoral expansion from the west, and facing

14. For a fuller account of the Ethiopian state territorial and cultural encroachments into the
Anywaa mode of life, see Chapter 6.
50 Playing Different Games

a continuous influx of Highlanders from the east, the Anywaa have ‘gone on the
defensive’, exhibiting, in the process, clearly xenophobic tendencies.

State-related socio-economic processes and the consolidation

of Anywaa primordiality
After briefly improving in the early decades of the twentieth century, the position of
the Anywaa in regional politics has declined progressively with their incorporation
into the Ethiopian state. Anywaa regional power was curtailed by the joint military
pressures they faced from the Ethiopian imperial government and the British colo-
nial government of the Sudan (see Chapter 6 for a fuller exposition). Their political
power was further undermined by the Ethiopian government’s projects of control in
the 1970s. For the Anywaa, the so-called cultural revolution of the socialist govern-
ment was equivalent to cultural uprooting. Throughout these periods, the Anywaa
experienced and articulated their ethnic identity as a form of resistance to state
encroachments, thus further consolidating their primordialist imagination.
State related economic processes have also led to the dispossession and disem-
powerment of the Anywaa. Over a period of a century the Anywaa gradually lost
their status as producers of their own subsistence and active participants in the
regional exchange economy, becoming instead consumers of relief food and
Highlander products. These processes of economic marginalization have fostered
Anywaa’s narratives of loss and introspection.
Domestic politics and wider geopolitical processes have brought hundreds of
thousands of settlers from the Ethiopian highlands and refugees (mostly Nuer) from
southern Sudan to Gambella. All of these settlers and refugees were resettled in
Anywaa territories. Moreover, the new neighbours appeared to be more empowered
than the Anywaa, since both the settlers and the refugees had stronger ties to state
actors. These new state-related socio-political processes have generated a kind of
anxiety among the Anywaa, which is in turn linked to a ‘growing sense of territori-
ality … and autochthony’ (Kurimoto 2005: 351). Since 1991, this consciousness has
intensified in the new political context in Ethiopia in which ethnic identity has been
institutionalized in formal ethnopolitics. Ethiopia’s unique experiment in ethnic fed-
eralism has legitimated ethnicity and, beyond that, it has established ethnic identity
as the preferred idiom of political action. Responding to the new political structure
and hoping to make use of it , the Anywaa, ‘like other nationalists, … recognize their
own identity from an essentialist/primordialist point of view and have an exclusive
attitude towards other peoples living in Gambella’ (Kurimoto 2005: 351).
The preceding discussion has shown the ‘construction sites’ of the Anywaa pri-
mordialist configuration of ethnic identity formation, ranging from an emphasis on
presumed common origins to an ideology of endogamy and to pronounced notions
of territoriality. It has also been seen that primordialism, like other modes of social
identification, should not be taken for granted; rather, the conditions of its dynamic
construction need to be explained. The conditions for the particular primordialist
configuration of Anywaa ethnic identity include a whole series of variables: the
absence of a ‘relevant ethnic other’ during the early period of ethnogenesis; changes
in the material conditions of life, i.e., the shift from a pastoral to an agrarian life-style
 The Anywaa Primordialist Ethnic Identity Formation 51

and to corresponding forms of territoriality; changes in the conditions of the inter-

ethnic setting – from relative isolation to sudden and massive confrontation with
more powerful neighbours; and the state-related socio-economic processes such as
political disempowerment and economic marginalization in a dominant society.
Ethiopia’s unique experiment in ethnic federalism, itself based on a primordialist
understanding of ethnic identity, has reinforced the Anywaa’s own primordialist
imagination of the social world. In the following chapter, however, a radically differ-
ent concept of ethnic identity is presented and analysed, namely, the constructivist
configuration of Nuer ethnic identity.
 Chapter 3

The Nuer Constructivist Ethnic

Identity Formation
In this chapter, I describe and analyse the Nuer mode of identity formation, which, I
argue, is not primordialist, as among the Anywaa, but constructivist. In this context,
‘constructivism’ is used in an ‘emic’ sense, that is, to refer to the Nuer conviction that
ethnic identity is not ascribed but achieved – that being a Nuer is based on cultural com-
petence rather than on shared origins. Hutchinson (2000: 9) anticipates my argument
in her distinction between ‘primordialist’ and ‘performative’ aspects of Nuer ‘concepts of
identity’. Still, it is important to emphasize at the outset that this view of the Nuer is at
variance with received interpretations from the classical era of social anthropology. The
chapter also discusses the incipient form of primordialist reconfiguration of Nuer ethnic
identity formation in new socio-political contexts (cf. Hutchinson 2000: 9–12).
Ever since the publication of Evans-Pritchard’s classic study, The Nuer (1940a), the
larger community known by this name has become synonymous in the ethnographic
literature with the so-called segmentary system based on patrilineal, or agnatic, descent.
The Nuer, in Evans-Pritchard’s presentation, are a Nilotic ‘people’, who are divided into
a number of ‘tribes’, or groups of related tribes, such as the Gaawar, the Lou, the Jikany
and so on. These tribes or tribal groups are, in turn, divided into clans, just as the clans
are divided into lineages. Finally, Evans-Pritchard distinguishes among ‘maximal line-
ages’, which are most inclusive, ‘major lineages’, which are less inclusive, ‘minor line-
ages’, which are relatively exclusive, and ‘minimal lineages’, which are most exclusive.
He refers to all of these units, which represent different levels of inclusiveness and
exclusiveness in the organization of descent groups, as genealogical ‘segments’ or some-
times as ‘sections’; hence the term ‘segmentary system’ (Evans-Pritchard 1940a: 148).
What impressed Evans-Pritchard most, and what he sought to convey to his readers
in his famous monograph, was that this segmentary system provided the basis for coor-
dinated political action in a community of circa 200,000 souls that was totally lacking
in hierarchical structures and centralized political institutions. I shall return to Evans-
Pritchard’s analysis of decentralized political coordination among the Nuer throughout
this book. At this point in my argument, however, it is important to emphasize that the
segmentary system is based on the principle of patrilineal descent. When descent is
patrilineal, as it is among the Nuer, all individuals, whether male or female, belong to
the segments – the minimal lineage, minor lineage, major lineage, maximal lineage, clan
and tribe – to which their father belongs; and, ideally, all Nuer should be able to trace
their ancestry through a chain of males back to a common male ancestor.
If, however, group membership among the Nuer were determined only or even
primarily by descent, as Evans-Pritchard claimed, then we should expect them to be
54 Playing Different Games

‘emic’ primordialists, much as the Anywaa are. In fact, it is well known that the Nuer
regularly incorporate people into their communities who began their lives as out-
siders, transforming these former outsiders into Nuer in the process. What is more,
the Nuer are fully conscious of their practices of incorporating outsiders, and they
provide extensive commentary on them. Indeed, it is largely because of their aggres-
sively assimilationist policy that I describe the Nuer mode of ethnic identity forma-
tion as constructivist. My task in this chapter, then, is to qualify the descent-based
model of the segmentary system, which has shaped our understanding of the Nuer
since the publication of Evans-Pritchard’s cogent and very influential study. Clearly,
patrilineal descent is important in Nuer society, as Evans-Pritchard insisted, but it is
equally clear that the Nuer do not live by patrilineal descent alone. Evidence of the
constructivist character of identity formation among the Nuer may be found in
various domains of social life, e.g., in representations of origins in mythology, in a
change in the identity discourse from purity of descent to assimilation, in multiple
modes of affiliation, and in the bodily construction of manhood. I also argue that
the Nuer constructivist mode of identification has a predominantly materialist char-
acter, in contrast to the ideological orientation of the Anywaa. The Nuer assimila-
tionist drive may be understood as a social strategy for resource extraction and also
as a function of competition for leadership within descent groups and under condi-
tions of political decentralization at the national level.
In revising our view of the Nuer, I draw on my own field research and on the
recent secondary literature. It is a tribute to the quality of Evans-Pritchard’s research,
however, that any revisionist approach to the Nuer must rest to no small degree on
his data, which reveal more facets of Nuer society than one might suspect, if one is
familiar with his central theses alone.

Representations of Origins in Mythology

Because of similarities linking Nuer and Dinka languages, scholars have placed them
together within the Western Nilotic language family (Evans-Pritchard 1940a;
Lienhardt 1956; Hutchinson 1996). On the basis of glottochronological analysis,
McLaughlin (1967: 87) has suggested that the separation between the Nuer and the
Dinka occurred in the first century AD. Nuer definitions of Nuer ethnogenesis,
however, often deviate from the organic imagery of language history, with its flowing
‘roots’, ‘stems’ and ‘branches’; and, consequently, Nuer self-understanding also devi-
ates from standard definitions of ethnic groups as collectivities based on the subjective
belief in common origins and essences. In this section, I review various Nuer and
Anywaa myths for evidence of indigenous theories of origin. While some of the
authors whom I cite are interested in attempting to draw conclusions from myths
about actual historical occurrences, I am concerned primarily with Nuer and Anywaa
self-understandings and their relation to their respective modes of identity formation.
Some Nuer myths trace all human beings or all Nuer back to a single origin, but
these exist only in fragmentary form. Southall (1976: 481) provides the following
very brief account of common origins: ‘The tamarind tree itself was the mother of
men, who either emerged from a hole at its foot or dropped off its branches like ripe
fruit … this tree was in Jagai country … the Jebal Ghazal triangle … called naath
 The Nuer Constructivist Ethnic Identity Formation 55

cieng, the homeland’. According to one version of the myth that I collected from
Gambella, this tamarind tree is known as Jiath Lie, which is found in a village called
Maar in Ler district. Jackson (1923: 70–71), on the other hand, establishes a
common origin of the Dinka and Nuer, based on a popular myth:

In the dim and distant past Deng Dit, the Great God of the Dinka …
married a woman called Alyet … While living in an aradeib tree Alyet gave
birth to Akol (or Aqwol) who married Garung from whom are ultimately
descended Deng and Nuer, the respective ancestors of the Dinka and Nuer
tribes … When Garung died he left behind him a cow and a calf, the former
being bequeathed to Deng and the latter to Nuer … Deng stole the calf of
Nuer who was the youngest brother and not able to retaliate. Nuer conse-
quently left the family and, when he had grown to man’s estate, returned with
some friends and retook his calf. From that day to this the Nuer and Dinka
are constantly raiding and counterraiding one another for cattle.

Evans-Pritchard, in turn, summarizes mythic representations of the descent of all

Nuer from a common progenitor as follows: ‘Many Nuer regard their ancestors Gee
and Ghaak as the progenitors of all true Nuer … Gee and Ghaak and Gwea are rep-
resented as brothers, sons of a mythological ancestor, sometimes called Ghau, the
World, and sometimes Ran, Man, whose father is said to be Kwoth, God’ (1940a:
238–39). The idiom ran mi ran (the real person), which the Nuer employ to bolster
their self-esteem, is related to this origin myth. The Dinka also use the term ran to
denote a person, but only among the Nuer does it confer prestige, while denying it
to members of other ethnic groups, who are not ‘real people’. Gee and Ghaak are said
to have formed moieties through a ritual called rual (cutting a cow into two halves),
and to have subsequently forbade marriage or sexual intercourse between their
members (Jal 1987: 14). The place where rual was first practiced is called Koat Lieh,
the site of a tamarind tree in Kuer Kuong (the present-day Bentiue area in southern
Sudan), which is still regarded as a holy place. Koat Lieh is also called naath cieng,
from where the Ghaak expanded southwards and the Gee northwards (Jal 1987: 15).
In many other instances, however, Nuer origins are represented in myth as having
been diverse. Evans-Pritchard (1940a: 240) acknowledges that ‘the Gaawar clan have
… an independent origin, their ancestor having descended from heaven’. ‘A number
of clans are not associated with tribes’, he adds. Jal (1987: 15) also cites evidence of
Nuer beliefs in the diverse origins of Nuer clans: ‘Many of the ancestors of some
Nuer clans appeared after Aak [Ghaak] and Geah [Gee], and according to most tra-
ditions they appeared miraculously’.
The separate origin of the Jikany, a name referring to a group of related ‘tribes’
within the larger Nuer community, is recounted in a rich mythology. In the Jikany
origin myth, Kir is considered to be their apical ancestor.1 Contemporary Nuer dis-

1. The term Jikany is said to have come from ‘ci Kany wic ke ciok rew’, which in Nuer language
means ‘someone who has come with two legs’, in reference to Kir’s mystical power with which
he killed two of his wives.
56 Playing Different Games

tinguish between descendants of Kir (Gaat-Kir) and descendants of Gee (Gaat-Gee).

In accordance with a strict genealogical principle, Gee is recognized as the ‘original’
Nuer. On the other hand, ‘the Jikany tribes’, as Evans-Pritchard (1940a: 239) notes,
‘have dominant lineages of Dinka origin, descended from Kir’. He identifies Kir with
a being found in a gourd by a Ngok Dinka named Gyng [Jing] and reared by a
Ruweng Dinka (1940a: 231). On the basis of this origin myth, Evans-Pritchard
(1940a: 239–40) suggests that the Jikany Nuer and some lineages of the Ngok Dinka
have an agnatic relationship (buth) and that the two tribes ‘have, by analogy, a fra-
ternal relationship’. In this mythology, Kir is said to have left the Ngok Dinka and
gone to the Gaat-Gee after he killed his foster brother. Jackson (1923: 74), on the
basis of genealogical reckoning, traced the birth of Kir to the beginning of the six-
teenth century. With reference to such evidence in oral tradition, Johnson (1982:
185) concludes that ‘the Jikany were a group of mainly Dinka origin, closely linked
to the Ngok and Rueng Dinka’.
Jal, on the other hand, documents a different version of the myth, which con-
nects Kir (of the Jikany Nuer) with the Anywaa. In the following I present Jal’s richly
detailed version of this foundational myth, to which I shall return in subsequent
chapters when referring to Nuer ideas of relatedness at local and national levels:

The ancestral founder of the Jikany, Kiir [Kir], was an Anywaa nyipem
(prince) who, after a defeat of his party by his half brother in an agem [sic,
angem] (rival grouping), was taken out and made to escape in a floating
gourd by his mother who said a chien (invocation), ‘thou shall go, multiply
and thou shall avenge the defeat and destruction of thy family’. Incidentally,
the gourd in which Kiir was hidden was discovered by a Ngok Dinka man,
Yiol Kuot, who broke it into two parts and brought out the young nyipem
and his ritual belongings – tuach kuach (a leopard skin), deeth (iron smith’s
tools), dual (piece of skin) and Mut Wiu (the divinity spear) … Subsequently,
Yiol took Kiir along together with his ritual belongings to the Ngok Dinka
country, on the Zeraf island, where the young nyipem was brought up to
manhood. But when Kiir grew up to manhood, the Ngok Dinka began to
refuse to accept him … in their country on the ground that the nyipem was
a powerful peath (witch), with the ability to employ evil powers to kill
animals and people. [They] expelled Kiir from their country and Jing
Mareang, his newly adopted uncle, together with a small party of men,
decided to accompany the young man to an undisclosed destination. In their
flight westwards … Kiir and his uncle landed on the east bank of a naam
duong (large river) … But there was no canoe to cross it … As they struggled
to find their way out, Kiir saw a guan yier [master of the water], Tiek [Tik].
Tiek struck the water in the river and opened up a passage for Kiir and his
party to cross the Bahr el-Jebel where they were joined up by many groups of
strange peoples … probably arriving in Bentiue area at about the fifteenth
century … In Bentiue, however, the new arrivals did not join the Geah [Gee]
or the Aak [Ghaak] groups, but they set up their own independent settle-
ments … the Nuer sacrificed a black bull … to neutralize the power of witch-
 The Nuer Constructivist Ethnic Identity Formation 57

craft in Kir … a party of men was sent back to the Bahr el-Jebel to bring Tiek
… Tiek decided to put up with Jing Mareang who sacrificed a thok mi reang
(bull with all colours) to welcome the new comer. After this ritual ceremony,
Tiek was shaved and incorporated as a brother to Jing. (Jal 1987: 15–18)

This Jikany origin myth provides evidence of complex forms of relatedness among
the Nuer that go beyond the belief in common origins and common descent. Tik,
the mystical figure who helped Kir cross the river, and Jing, Kir’s classificatory uncle,
adopted the name of Reng.2 Tik and Jing joined the Jikany tribe called Gaat-Jak as
Cieng Reng, but they both continued to acknowledge their separate origins. Various
scholars identify Kir and Jing as a Dinka (Evans-Pritchard 1940a; Johnson 1982). Jal
(1987: 18–19), on the other hand, establishes the Anywaa background of both Kir
and Tik – or at least their connection to the Lwoo group to which the Anywaa
belong. Tik’s ‘spear name’, mut pini duong, seems to be Anywaa in origin, as it is
similar to the Anywaa name for a large river, naam duong.3 Thus, the Gaat-Jak is a
Jikany tribe of diverse origins, including groups of people who trace their origins
ultimately to Nuer, Anywaa or Dinka.
According to another myth, Kir married three wives from Gee and Ghaak families.
With his first wife, Nyakuini, Kir begot Thiang; with his second wife, Nyabora, he
begot Khun; and with his third wife, Duany, he begot Jok. Kir died after the birth of
Jok, and his eldest son, Thiang, begot him a child named Nyang with Jok’s mother, i.e.,
his own stepmother, Duany. The children of Kir founded the four descent groups of
the Gaatgankir clan, namely, Thiang, Khun, Jok and Nyang. The half-brothers Thiang
and Khun lived together, forming, along with Reng, the Gaat-Jak and adopting Khun’s
‘ox name’ (Jak) as their common designation.4 Jok and Nyang, who had the same
mother, lived separately, however, forming two additional tribes, Gaat-Jok and Gaat-
Guang, respectively. This suggests, in mythological discourse at least, that some seg-
ments of the Jikany have a single origin, while others have heterogeneous origins.
I collected similar mythologies about Kir, which also connect the Jikany with the
Anywaa:

There was a mysterious person called Lekor who lived in the small pool near the
Anywaa of Nyiche.5 He had a golden canoe and a fishing spear. One day, an
Anywaa kwaaro came to him and asked him to come to the village. Lekor
agreed. In that village there was a beautiful girl who had declined many marriage

2. Reng is the colour of the sacrificial bull that symbolizes the new bond between Tik and Jing.
3. Among the Nuer the spear stands for masculinity. According to Evans-Pritchard (1953: 12),
‘each Nuer clans has a spear name … This spear name is shouted out by a representative of the
clan, which he brandishes a spear in his right hand, in war, at weddings, and in other public
occasions when the clan as a whole is concerned’.
4. An ox name among the Nuer is ‘the name of men, males who have passed through the rite of
initiation to manhood’ (Evans-Pritchard 1953: 183).
5. The Nuer name Nyiche resembles that of the chief of the Anywaa Maro clan, Nich, further
suggesting Anywaa connections. In another myth, Lekor is portrayed as a man of the Tapoza
tribe from the equatorial region of southern Sudan.
58 Playing Different Games

offers. When she saw Lekor she said, ‘That is the man I have been waiting for’.
Lekor knew all languages, as he was a man of the river and was related to Kuoth.
He slept with that girl and he said to her, ‘When you give birth, take my child
(Kir) to the bank of the river and ker (calabash) will come and take him away’.
Lekor told her to put Kir on a kom (chair) together with a kuac (skin) and a mut
(spear). The next day, he returned to the water. When she gave birth, she did as
she had been told. Kir had already started speaking while he was in his mother’s
womb. Kir was a strange child. His eyes were bleeding and he killed people when
he looked at them. His three wives were survived by four children: Thiang,
Khun, Jok and Nyang. When they became many, they were called the Jikany.
(Tap Gatwech, elder from Makot village, Itang district, 12 April 2001)

Thus, the various origin myths of the Jikany Nuer provide ample evidence of Nuer
beliefs in the separate origins of the various Nuer divisions. This contrasts starkly
with Anywaa conceptions of Anywaa origins. Despite some indications that their
origins are in fact diverse, contemporary Anywaa believe that they all descend from
Gilo. On the other hand, contemporary Nuer openly embrace the idea that their
own origins are diverse.
The separate origins of the Jikany Nuer find symbolic expression in the cult of
the sacred spear known as Mut Wiu (see Figure 3.1). It is important to understand,
however, that perceptions of separate origins of the Jikany Nuer do not affect their
fully-fledged membership in the wider Nuer society.

Figure 3.1 Kir’s Shrine where Mut Wiu is kept, Dorong village, Itang (photo: Dereje Feyissa)
 The Nuer Constructivist Ethnic Identity Formation 59

From Descent to Assimilation: Criteria of Ethnic Membership in

Historical Perspective
Notwithstanding the incorporation of the Jikany into Nuer society when Kir joined the
Gaat-Gee, supposedly in the beginning of the sixteenth century, the Nuer of that earlier
period do not seem to have been as thoroughly assimilationist in orientation as they are
today. In fact, there is some evidence suggesting that, prior to that legendary event, the
Nuer recognized group membership largely on the basis of descent, perhaps even espous-
ing an ideology of ethnic purity (Jal 1987: 17). In some versions of the Jikany origin myth,
the affinal link between Gee and Kir is depicted as having been weak. All the wives given
to Kir by Gee were either half-Nuer or handicapped. In Jikany oral history, one of the main
reasons for the eastward expansion of the Jikany tribes in the nineteenth century was the
social discrimination that they experienced at the hands of the Gee. Members of Gee, sup-
posedly the ‘original’ Nuer group, are said to have excluded immigrants such as Kir and
his followers from collective rituals and associated cultural practices, such as cattle sacrifice:

Nuer call themselves ran mi ran because of Gee, who was the first Ran. Gee
was the dil [senior male of the dominant descent group] of Nuer. It was Gee
who first opened the calabash where Kir was found. Kir and his people lived
with Gee, but they were not allowed to participate when an ox was killed.
Whenever the Nuer kill an ox, they usually curse the ox, putting all the bad
things on it. Gee and his people used to call Kir ran mi jaang, not original
Nuer. During the time of Latjor, Kir’s people said ‘let us go to a place where
there are no diel’. In these new places they started cursing the ox [i.e., became
real Nuer] like the people of Gee. (Kong Diu, leader of the Cieng Reng Nuer,
Makot village, Itang, 2 February 2002)

Given the significance of cattle sacrifice among the Nuer, the Jikany’s grievances
about their exclusion from this important ritual are understandable. According to
Evans-Pritchard (1956: 220), ‘most sacrifices … have a peculiar purpose. They are
made in times of trouble and their general object is always the same, to get rid of the
evil or threatening the evil by offering God a victim whose death will take it away’.
The victim explicitly plays the role of the scapegoat. Since the ‘original’ Nuer pre-
vented newcomers from participating in vital cultural practices such as cattle sacri-
fice, it seems valid to conclude that their earlier mode of identification was exclusive,
being based on genealogical descent and, in this sense, on ethnic purity.
In the early nineteenth century, an ambitious man named Latjor Dingyian mobi-
lized the dissatisfied Jikany, leading them in a large-scale conquest of new lands,
where they could attain the status equality that they sought and he could attain the
political position that he desired. The followers of Latjor are called Jikanydoar
(‘Jikany of the bush’) and those who remained in Bentiue are called Jikanycieng
(‘homeland Jikany’). The Nuer shift from an ideology based on descent – and, by
extension, ethnic purity – to one based on assimilation seems to have taken place in
the second half of the nineteenth century as the Jikany migrated, conquering exten-
sive Dinka and Anywaa territories in the process. By assimilating those whom they
captured, particularly the Dinka, instead of killing them, the Jikany were able to
60 Playing Different Games

double the size of their group in a relatively short period of time. The assimilation of
new members was practically a necessity, because of the composition of Latjor’s
group of followers. Because of the perceived risks, only a few women had joined the
adventurous eastward migration (Pal 2006: 223). The more Latjor’s followers
married among their enemies and adopted captured children, the larger their popu-
lation became and the greater their need for further expansion.
The success of the Jikany Nuer seems to have encouraged other sections of the
Nuer to migrate, to conquer new lands and, in the process, to adopt the assimilation-
ist strategies of the Jikany. What is more, even those who remained at home turned
increasingly to assimilation in the context of competition over local leadership.
According to Hutchinson (2000: 9), ‘individual Nuer men competed with one
another for positions of political leadership and independence by gathering around
themselves as many co-resident Dinka clients and supporters as possible’. She goes
on to say that ‘the “enduring loyalty” of these clusters of co-resident Dinka was
secured, primarily, through the generous provision of Nuer cattle and Nuer wives’
(Hutchinson 2000: 9). Of course, cattle and wives are linked through the institution
of bridewealth, about which more will be said later in this chapter.
Thus, under the new social circumstances that arose with migration or with com-
petition for local positions of leadership, the Nuer mode of identification was recon-
figured with an accent on integration rather than exclusion. Contemporary Nuer,
regardless of the diversity of their origins, all claim and have access to group honour,
principally expressed in their self-esteem as nei ti naath (distinct people) with a
unique language (thok Nueri) and culture (ciang) and with unique rights to claim the
status of being ran mi ran, i.e., a real person (Hutchinson 1996: 76). On this basis,
the Nuer have redefined the criteria of ethnic membership in terms of cultural com-
petence; and, by acknowledging the diversity of their origins, they facilitate the
absorption of outsiders through various mechanisms of assimilation.
These various mechanisms for absorbing outsiders, which were alluded to,
directly or indirectly, in the Jikany origin myths cited above, can be seen in Figure
3.2. At the top of the diagram, the broken lines indicate the non-blood relationship
between the Nuer (Gee and Ghaak) and Kir, between Kir and Tik, and between Tik
and Jing. The Jikany became Nuer through matrilateral links, i.e., through marriage
alliance and descent from Nuer women, as represented in the relationship between
Kir and his Gee and Ghaak wives and children. Tik and Jing, in turn, became broth-
ers through the creative power of cattle sacrifice. The bond between the Cieng Reng
and Gaat-Kir is based, mythologically, on the reciprocal favours underlying the
friendship between Kir and Tik. It is this mythical friendship, in which Tik appears
as a kind of Jikany ‘Moses’, that contemporary descendants of Cieng Reng cite in
making claims to resources and citizenship.6

6. For the relevance of mythical friendship to the Cieng Reng politics of entitlement, see
Chapters 4 and 8.
 The Nuer Constructivist Ethnic Identity Formation 61

Figure 3.2 The Jikany Nuer divisions (source: author’s field notes)

Beyond Agnation: the Critique of Evans-Pritchard’s Model in the

Secondary Literature

The question is not so much whether the Nuer value links through men,
patrilateral affiliation, and perhaps even something approximating patrilin-
eality. They clearly do. But these are not the sole structuring forces of Nuer
society. (McKinnon 2000: 68)

Although patrilineal descent is a major principle of social organization among the

Nuer, it is important to emphasize that it is supplemented by other modes of filia-
tions such as locality and matrilateral ties (Hutchinson 1985; McKinnon 2000) as
well as friendship rooted in mythical history. These multiple modes of relatedness
and the principles that underlie them are not contradictory, nor do they operate in
different domains; rather, they are mutually constitutive of Nuer social reality. The
perspective I adopt regarding the nature of Nuer descent groups differs from that
which privileges patrilineal descent and neglects other principles of social organiza-
62 Playing Different Games

tion or relegates them to other domains of social life (Evans-Pritchard 1940a; Sahlins
1965). But I also disagree with those who dismiss the relevance of descent in Nuer
social organization entirely (Kuper 1982). Instead, I argue that the mechanics of
Nuer constructivism become intelligible only when the interplay among the various
modes of relatedness is taken into account. The discussion starts with an outline of
the various interpretations of Nuer descent groups and continues with ethnographic
examples serving to substantiate the perspective I adopt.
The study on Nuer descent group was inaugurated in the pioneering writings of
Evans-Pritchard (1940a, 1951a). Through his research, the Nuer have become a par-
adigmatic case in studying political organization in ‘stateless societies’ in which patri-
lineal descent serves, in the absence of institutions of political power, as a basis of
solidarity. In the model developed by Evans-Pritchard, which is widely known as the
segmentary system, or segmentary lineage system, descent through the male line
(patrifiliation or agnation) is the fundamental value upon which the political system
is built. According to Evans-Pritchard (1940a), the lineage system and the political
systems operate in two different domains, corresponding to patrilineal descent
groups (clans and lineages), in the first case, and to territorially based tribes, in the
second. The fundamental units upon which the lineage system is based are the seg-
ments known as thok dwiel, whereas the fundamental residential or territorial units
within the political structure are known as cieng. Evans-Pritchard argued that,
although the two structures operate in different domains, the thok dwiel takes prece-
dence as a principle of social organization, and the cieng is predicated on the thok
dwiel. Apart from serving as an ideological frame of reference for Nuer political com-
munities, thok dwiel are also relevant points of reference in articulating the rules of
lineage exogamy, and further serve as basic units in the performance of rituals. Evans-
Pritchard accounted for what was later called the ‘E-P paradox’ (Southall 1986: 1),
the discrepancy between the primacy of agnation and the prevalence of matrilateral
ties in residential units, by relegating the two principles to different domains: the
political domain, based on agnation, and the domestic domain, based on bilateral
kin ties. In attempting to resolve this paradox, Sahlins commented somewhat later
on Evans-Pritchard’s ethnography, saying that, in the case of the Nuer, ‘a descent doc-
trine does not express group composition but imposes itself upon the composition’
(Sahlins 1965: 104).
Other scholars have criticized Evans-Pritchard’s notion of the primacy of agna-
tion among the Nuer and reduced it to the level of ideology (Holy 1979; Southall
1986). Holy (1979: 12), for example, distinguishes between representational and
operational models: ‘Among the Nuer, lineage organization is representational and
the spatial relations between the various cieng (residential units) as well as the
genealogical relations between the dominant lineages associated with them are oper-
ational’. Southall (1986) also demotes agnation from a fundamental principle of
social structure to the level of ideology, which, in his view, serves as a kind of social
cement that enhances solidarity among diverse groups in their collective efforts to
secure their material well-being in a harsh environment. Southall attributes the flex-
ibility of group formation and the fluidity of the units of identification up and down
the scale of social organization, from the most exclusive to the most inclusive seg-
 The Nuer Constructivist Ethnic Identity Formation 63

ments, to the very fictionality of the agnatic principle: ‘Naath [Nuer] social organi-
zation is based on localized groups recruited by cognatic and affinal bonds, linked
together by genealogical charters of fictional agnation’ (Southall 1986: 16). Pushing
his materialist interpretation further, Southall notes: ‘Such charters in themselves
have little force with which to bind people together. They can only justify and add
some sentiment to links between groups whose material interests have already
brought them to desire an ideological basis of alliance’ (ibid.: 16). In his view, ‘these
genealogical charters reflect the actual relationships of persons when alive or that
they were as agnatic as represented. The very fictionality makes them flexible and
able to express and give primordialist value to the really determining factors … [i.e.,]
material culture’ (ibid.: 8).
The strongest objections to Evans-Pritchard’s structural functionalist approach
have been formulated by Kuper (1982). Kuper rejects the role of descent as a princi-
ple of social organization among the Nuer with reference to the discrepancy,
acknowledged by Evans-Pritchard, between descent groups and territorial groups: ‘It
is more reasonable to conclude that the Nuer model provides reliable guidance
neither to Nuer social behaviour nor to Nuer values’ (Kuper 1982: 84). But Kuper
has gone too far in rejecting the role of descent as a principle of social organization
among the Nuer. Schlee (2002: 261) has recently argued in favour of an empirical
approach to determine the social and political relevance of patrilineal descent in con-
temporary African societies and has criticized both the ‘overly one-sided accent on
the segmentary lineage system’ and the ‘anti-structuralist wave’ that condemned the
segmentary lineage model ‘on the grounds of principle, not on empirical grounds’.
Recent studies on Nuer descent groups have proceeded in this spirit, painting a
much more complex picture of the place of patrifiliation in Nuer society
(Hutchinson 1985, 1996; McKinnon 2000). These studies have identified different
forms of affiliations as principles of social organization, not least of which are matri-
lateral ties, which play an important supplementary role.
Building on data provided by Evans-Pritchard (1951a: 109), Hutchinson demon-
strates the importance of matrilateral ties with reference to the special relationship
between two categories of male kin, namely, gaatnaar (sons of the mother’s brother)
and gaatwac (sons of the father’s sister). Evans-Pritchard noted that the relationship
between gaatnaar and gaatwac is ‘one of “easy companionship,” stripped of the reserve
that difference of generation foster in their relations with each other’s parents’
(Hutchinson 1985: 633, quoting from Evans-Pritchard 1951a: 167). When, however,
one or the other of these kin groups has difficulties in producing heirs, a basic asym-
metry in the gaatnaar/gaatwac relationship becomes evident. If a man or a man’s son
has difficulty in conceiving offspring, the gaatwac, the son of the man’s sister, may
marry in the name of his uncle or his uncle’s son and, in this way, provide the uncle
or uncle’s son with an heir. But the gaatnaar cannot provide the same service for his
gaatwac – hence Hutchinson’s characterization of this relationship as one of ‘genera-
tive asymmetry’ (1985: 633). In Hutchinson’s interpretation, the gaatwac is obliged to
help out in cases of infertility, because, earlier, his mother’s brother had provided him
with cattle and thus helped him to pay the bridewealth required for his marriage. For
our purposes, however, this asymmetrical relationship between gaatnaar and gaatwac
64 Playing Different Games

shows how a patrilineal descent group, in the minimal form of the relation between a
father and his son, can reproduce itself by drawing on the services of relatives who are
related matrilaterally, i.e., through the man’s sister. One might say that the patrilin-
eage ‘recruits’ new members by employing the reproductive powers of men belonging
to other lineages who are related to them only through female ties.
Even more recently, McKinnon (2000) has revisited the debate on the nature of
the Nuer descent groups, highlighting the centrality of cattle, given as bridewealth
(twoc ghok), and the supplementary role of links through women in agnation. Success
in building a strong agnatic group, or patrilineal segment, depends on the nature and
level of bridewealth payments. Ideally the groom’s party pays the bridewealth in full,
but not infrequently the groom or his party fails to do so, because they lack the
resources, i.e., the cattle. In such a case the child resulting from the union is likely to
be affiliated to the thok dwiel of its mother’s brother: ‘it is bridewealth/childwealth,
not birth and genealogical connection, that establishes membership in a thok dwiel’
(McKinnon 2000: 62). It is for this reason that the bridewealth payments are proces-
sual among the Nuer, occurring in three stages: larcieng (betrothal), ngut (wedding)
and mut (consummation). Each phase marks the progressive transfer of cattle from
‘wife takers’ to ‘wife givers’ (Evans-Pritchard 1951a: 51). The ideal bridewealth is
forty cattle, but currently among the Jikany it is set at about twenty-five. There are
many Nuer men who are affiliated with their mother’s brother lineage because they
were unable to pay the bridewealth in full.
The ethnographic data that I collected during fieldwork lend support to the
claim that agnation is both a core value and an important operational model among
the Nuer. At the same time, my data indicate that agnation is not a lived experience
for all Nuer. Patrifiliation among the Nuer is not automatic; rather, it is a practical
achievement, a form of filiation that must be earned through fulfilling rigorously the
obligation to pay bridewealth. All Nuer seek to establish strong agnatic groups, but
practical limitations in achieving this ideal lead to alternative forms of filiation, most
commonly to attachments through matrilateral ties. Once we acknowledge that
agnation is a fundamental value and operational model among the Nuer, it is still
important to ask what happens when men fail to establish strong agnatic ties and,
thus, strong patrilineally based groups. In fact, failure to create strong agnatic ties is
not denigrated among the Nuer; instead, there are institutional mechanisms that
provide Nuer men with what might be called a ‘Plan B’. For those who fail to create
strong agnatic ties, matrifiliation provides the possibility of an alternative social iden-
tity and support system centring on the mother’s brother. In the context of the close,
very sentimental relationship between the mother’s brother and the sister’s son,
matrifiliation even acquires a very real degree of dignity.
In sum, identification among the Nuer is much more complex than the segmen-
tary system of ‘balanced opposition’ between structurally equivalent lineages and res-
idential units, as in Evans-Pritchard’s structural functionalist interpretation. Alliance
formation does not always follow genealogical structure. In addition to descent, itself
constructed through the supplementary role of matrilateral ties, friendship, rooted in
mythical history and in ecological considerations, is an important element in the
repertoire upon which the Nuer draw in building meaningful ties.
 The Nuer Constructivist Ethnic Identity Formation 65

The Mechanics of Nuer Assimilation

Descent is important for the Nuer, but it works in a way that allows the integration
of those who were previously outsiders. Group formation and identification operates
through three interrelated Nuer concepts: dil (pl. diel), rul and jang (pl. jaang).
Evans-Pritchard (1940a: 220) defined dil as ‘an aristocratic clan’, though the term
also applies to individual men within the clan or descent group in question. In
Evans-Pritchard’s analysis, the diel, though a minority, provide a lineage structure on
which the territorial, tribal organization (cieng) is built. A Nuer is a dil only in a tribe
where his clan is the dominant lineage. As soon as he is outside of the tribe where his
clan has the superior status he becomes rul. A rul is a Nuer immigrant who attaches
himself to a dil through affinal ties, marrying either the sister or the daughter of a
dil. Jaang are non-Nuer captives or immigrants. The jaang are assimilated into the
diel through either adoption or marriage. Sons of diel are called gaattutni, while the
sons of rul or of jaang who have married women from the local dil are called gaat-
nyiet, i.e., sons of the daughter of a dil (Howell 1954: 81–82).
Evans-Pritchard mentioned the status differences among the diel, rul and jaang, but
he saw these differences as characteristics of kinship or of the ‘domestic’ domain, which,
in his view, was distinct from the political domain. Indeed, by separating politics from
kinship, Evans-Pritchard was able to represent the political life of the Nuer as radically
egalitarian. In subsequent studies, however, scholars have often rejected or at least qual-
ified Evans-Pritchard’s analytical strategy, emphasizing instead the relevance of suppos-
edly purely domestic distinctions for the construction of hierarchical relations within
the Nuer political order (Gough 1971; Verdon 1982; Hutchinson 1996; McKinnon
2000). The rul and the jaang and their patrilineal descendants ‘were initially at a dis-
advantage vis-à-vis the descendants of the original Nuer conquerors (the diel …)’,
Hutchinson (1985: 626) argues. Status differences between the diel, on one hand, and
the rul and jaang, on the other, were indicated by the use of the terms tut, or ‘bulls’, for
the former and yien, or ‘tied’, for the latter (Hutchinson 1985: 626). Gough (1971:
113) mentions that bulls have rights over land and possess considerably more cattle
than rul or jaang do, which makes it easier for them to pay bridewealth and to achieve
patrilateral affiliation and patrilocal residence. McKinnon also questioned the egalitar-
ianism of the Nuer political system, citing the important distinction ‘between those
who are able to achieve some kind of lineal continuity through successive patrilateral
affiliation to a thok dwiel and those who are not and who, therefore, affiliate to other
thok dwiel through matrilateral ties’ (McKinnon 2000: 68). The ‘tied’ are important in
the politics of the ‘bulls’, because they increase the size of his group and hence his
power base: the more a dil manages to attach rul and jaang to his immediate group of
supporters, the higher his chances of wielding political power among his agnates.7

7. Evans-Pritchard recognized that the ‘elders with most influence are the gaat twot, the children
of bulls’ (1940a: 179), but his description of Nuer as a ‘stateless’ society operating through ‘bal-
anced opposition’ between its various units prevented him from understanding the dynamics
of competition for leadership among agnates. This may have been because Evans-Pritchard
understood political power in a very restricted sense, namely, as ‘the authority and power to
force compliance with laws’ (McKinnon 2000: 67).
66 Playing Different Games

While the diel are concerned primarily with the intra-agnatic competition for
bridewealth cattle and leadership positions, these activities also create the contexts in
which assimilation occurs. Often, a rul attaches himself to a dil by marrying into his
family. Then, in the course of generations, his descendants gradually assimilate as
gaatnyiet. The jaang, on the other hand, are often integrated into a dil through adop-
tion. Thus, their assimilation is often more complete than that of the rul, since the
jaang, in contrast to the rul, are completely cut off from their homeland ties. Evans-
Pritchard described the assimilation of the jaang into the diel through adoption as
follows: ‘Captured Dinka boys are almost invariably incorporated into the lineage of
their Nuer captors by the rite of adoption, and they then rank as sons in lineage
structure as well as in family relations, and when the daughters of that lineage are
married they receive bride-cattle […] People say “caa dil e cieng” or “caa ran wec”,
“he has become a member of the community”… Adoption gives him a position in
lineage structure, and thereby ceremonial status, for by adoption he becomes a
member of his captor’s thok dwiel, lineage’ (Evans-Pritchard 1940a: 221–22).
Jaang are assimilated into a dil lineage through a ‘stick and carrot’ policy. If
people of foreign origin remain culturally ‘other’ and if their commitment to their
affinal lineage remains precarious, they could be labelled jaang in the derogatory
sense of the term, which connotes not only a foreigner but also a lower position. This
is the assimilation ‘stick’. At the same time, jaang are actively supported in becom-
ing Nuer (caa naath) insofar as they receive contributions towards their marriage pay-
ments, the size of which are often substantially reduced, and insofar as they are
rewarded with leadership positions, should they be heroic enough to contribute to
the strength of the local community. That is the assimilation ‘carrot’. A fully assimi-
lated jang may assume a high social standing; and it is not ‘culturally correct’ among
the Nuer to remind somebody who is fully assimilated and upholds local standards
of his foreign origin. In fact, whoever is discriminated against on the basis of his
origin can easily elicit sympathy. In this sense, the Nuer identity system encourages
assimilation by rewarding excellence in fulfilling Nuer standards.
The more complete assimilation of jaang into the diel explains in part why Nuer
seeking to augment their own groups are more interested in outsiders than in their
fellow Nuer. The loyalty of fellow Nuer to the thok dwiel and the cieng they become
attached to is precarious, because they could defect and rejoin their natal communi-
ties. Nevertheless, both potential rul and potential jaang are encouraged to join the
diel on the basis of an ideology that makes it possible to create binding social and
economic ties. Unlike the Anywaa ‘strangers’ (welo), who retain the permanent status
of being guests, the Nuer rul and jaang are temporary classifications. Having married
into the dil family, they become full members of the local community (cieng) and
they also play a crucial role in the politics of their host. Thus, and perhaps somewhat
paradoxically, the Nuer notions of descent and first-comer status provide an ideolog-
ical framework for recruiting and incorporating newcomers.

Census of Wechdeng village

In order to shed further light on the highly debated nature of Nuer descent groups,
I conducted a census in Wechdeng village in Kurtony, one of the major wet-season
 The Nuer Constructivist Ethnic Identity Formation 67

settlements of the Thiang Nuer in Jikaw district. The dominant lineage in Wechdeng
village is Cieng Buoy, the founder of which is a member of the fifth generation of the
Thiang descent group. The Cieng Buoy is further subdivided into segments, the
most important of which is Cieng Deng, after which the village is named. By exam-
ining the relations among the diel, rul and jaang in this context, I illustrate the
various forms and processes of affiliation in a Nuer local community (see Figure 3.3).
In Wechdeng village there are seventy-eight families. Census result show thirty-
one (forty per cent) of the male heads of families are agnates and forty-seven (sixty
per cent) are non-agnates, i.e., not related patrilineally to Deng. Of the forty-seven
male heads of the non-agnates thirty-five (seventy-four per cent) are matrilaterally
affiliated with the family of Deng, six (thirteen per cent) are rul who followed their
Cieng Deng friends, and six (thirteen per cent) are affines, out of which four are rul
and two are jaang (Dinka).
The diverse composition of Wechdeng village reveals the principles of group for-
mation and the multiple means of social inclusion among the Nuer. In this particu-
lar village, however, the flexibility of Nuer identity may also be illustrated with

Figure 3.3 The Cieng Buoy in the Thiang genealogical structure (source: author’s field notes)
68 Playing Different Games

reference to the origins of the village’s namesake. Deng began his life as a Dinka; he
became an orphan while still very young and then grew up with his uncle in a village
in the Ngok Dinka country. As the story is told, one day Deng hunted a giraffe in
his homeland, but his mother gave the trophy to her brother. Deng became angry
and left his original home. Upon reaching Kurthony village as a jal tang (guest), he
was given food by the Nuer. He made contact with Buoy, who later adopted him as
his son. Thenceforth, Deng was called Deng Buoy. Deng proved his merits in
hunting and fighting. Once he led the Wechdeng villagers in raiding the Anywaa and
brought back many cattle. After he returned with his plunder, Deng married, in
accordance with established rules of exogamy, and was fully assimilated into the thok
dwiel of Buoy, ultimately succeeding in creating his own lineage and a large local
community named after him, the Cieng Deng.
The centrality of the thirty-one agnates in Wechdeng village who are descended from
Buoy or from his adopted son, Deng – i.e., of the diel, to which the remaining forty-seven
families are attached – shows the relevance of patrilineal descent as a framework of social
inclusion among the Nuer. But this example also makes clear that Nuer local communi-
ties are based on various forms of affiliation. Local communities are constituted not
through two principles operating in different domains – the agnatic principle in the polit-
ical domain and matrilateral affiliation in the domestic domain, as Evans-Pritchard
argued – but through the interpenetration of the two principles in a unified socio-polit-
ical order. A village such as Wechdeng is composed of agnates, cognates, affines, friends
and immigrants, and mythically related groups of people, all assuming the common
name of the dominant lineage. For all practical purposes, it is their identity as Cieng
Deng that matters in both their social relationships and in their political mobilization as
members of their locality. The descent ideology, the very adoption of the collective name
Cieng Deng, and the custom of referring even to non-agnates as if they were descendants
of Deng have had a real effect, in as much as the non-agnate residents are either adopted
or, more frequently, encouraged to marry into the dil. Consequently, most of the inhab-
itants become related to one another over a period of time, even if they are not agnates.

On the relative strength of agnatic and local ties

The common name of any particular village – e.g., Cieng ‘X’ – is maintained not by
the collective effort of lineage members, most of whom are dispersed among other
localities, but by a local community of diverse people, united in their pursuit of their
livelihood and collective security in an economy built through, and defended against,
cattle raiding. This pragmatic local community is cemented through a web of kinship
ties. There is always a dominant descent group in a given village, but its members, the
dil, are typically in the minority. Most community members, as has been shown
above, are related to the dil through friendship, marriage or matrifiliation. Village res-
idents belonging to the dominant lineage may also maintain contact with agnates
living elsewhere, especially within the immediate vicinity. But the same applies to
village residents who do not belong to the locally dominant lineage. I was interested
to learn what kind of relationships still exist between those rul who have become
members of the Cieng Deng and their agnates who live in other villages, for this
would be a real test of the relative strength of agnatic ties and local affiliation. What
 The Nuer Constructivist Ethnic Identity Formation 69

I found were dispersed lineages, the members of which still participate in collective
rituals promoting the spiritual well-being of the shared patrilineal descent group. I
had an opportunity to observe a ritual in Wechdeng for colwic, a member of a class of
divinities associated with lightning and identified by Hutchinson (1996: 138) as
‘guardians of specific lineages’. In the case that I observed, colwic was the guardian of
the family of Jock. Agnatically, Jock is a Cieng Nyajani from Makuey village. He came
to Kurthony, married a dil ’s daughter, and his children become gaatnyiet to the Deng
family. Usually, lineage rituals are held at the place where the senior members of a
lineage live. As the eldest son in his lineage, Jock made a sacrifice to the colwic in
Wechdeng village in 2001. This ritual was attended only by those people who are
agnatically related to Jock, most of whom came from the far-off village of Makuey.
With the notable exception of such agnatic rituals, most of Jock’s social relation-
ships are with the Wechdeng villagers and with the Thiang at large; and, ultimately,
Jock and his family owe their loyalty to the village and the lineage to which they are
attached, i.e., to Cieng Deng. If Wechdeng village went to war against a village in the
Cieng Nyajani area, where Jock was born, Jock’s primary loyalty would still be to the
Thiang or to Cieng Deng, not to the Cieng Nyajani. In such cases of conflicting loy-
alties, however, Jock would probably be given a face-saving option, so that he would
not end up killing his agnates in Makuey village. He would, thus, avoid directly con-
fronting them in the event of war and, instead, attack people in Makuey village to
whom he was not related. Indirectly, however, he would still be joining his local com-
munity of affiliation in annihilating his local community of origin.

The Bodily Construction of Nuer Men

The Nuer’s constructivist approach to ethnic identity formation is also manifest in a
male initiation ritual called gar. Gar refers, more specifically, to the scars that an adult
male Nuer bears on his forehead as a symbol of his initiation into manhood between
the ages of fourteen and sixteen (see Figure 3.4). Evans-Pritchard described gar as
follows: ‘All male Nuer are initiated from boyhood to manhood by a severe operation
(gar). Their brows are cut to the bone with a small knife, in six long cuts from ear to
ear. The scars remain for life, and it is said that marks can be detected on the skulls
of dead men’ (1940a: 249). Nuer age sets, known as ric, differ from comparable insti-
tutions in other East African societies, insofar as they do not constitute successive age
grades (cf. Kurimoto and Simonse 1998). Nor do ric serve to organize men in mili-
tary units, as is sometimes the case elsewhere. As Evans-Pritchard notes, ‘The age set
system of a tribe is in no way its military organization. Men fight by villages and by
tribal sections and not by sets. The war companies are local units and not age-set
units’ (1940a: 254). Gar is, rather, a one-time initiation that marks the difference
between men (wut) and boys (dhol). After initiation, a man’s domestic duties and
privileges alter radically: ‘At initiation a youth receives from his father or uncle a spear
and becomes a warrior. He is also given an ox, from which he takes an ox-name and,
becomes a herdsman’ (Evans-Pritchard 1940a: 254).
Gar, which the Nuer describe as ‘shedding blood together’, creates an enduring basis
of friendship among the Nuer. It also promotes an inclusive ethnic consciousness:
‘Initiation rites, more than anything save language, distinguish Nuer culture and give
70 Playing Different Games

Figure 3.4 A wut – Ochom village (photo: Dereje Feyissa)

Nuer that sense of superiority which is so conspicuous a trait of their character’ (Evans-
Pritchard 1940a: 260). Within Nuer society, gar creates male identity and status through
the dramatization of the experience of pain; but it also serves an important ideological
role in the process of assimilating non-Nuer through the articulation of a particular con-
ception of masculinity. In this radical formulation of male identity, non-Nuer men are
loosely referred to as dhol, or boys, for they have not gone through the litmus test of
bearing pain in the initiation rite. Conversely, however, non-Nuer men may become
Nuer by going through gar. In the secondary literature to date, one may find suggestive
remarks regarding the role of gar in the assimilation process, but this role has not yet
been adequately explored. Hutchinson (1996: 289) provides the following commentary
on the integrative role of gar during the expansion of the Eastern Nuer, which resulted
in the absorption of many non-Nuer: ‘It would not be surprising if the eastern “first
comers” [Jikany and Lou Nuer] began to view initiation rites as a powerful means of
recruiting and definitively affiliating new community members’. This presumed integra-
tive function of gar was confirmed in many conversations that I had with the Nuer
during my fieldwork. The following narrative by Reverend Stevenson throws light on
the ideological and instrumental dimensions of gar in creating a ‘boundary of inclusion’:

Once you have gar, you cannot get rid of it whether you like it or not. You per-
manently become Nuer. You can change your residence, you can learn a new
 The Nuer Constructivist Ethnic Identity Formation 71

language, but once you have gar, it would always remain with you. Those who
were captured in the war were immediately made gar. Take the case of Joshua
[the administrator of Gambella during the 1980s]. His family is Bär Jingmir
[Anywaa from Akobo] who were captured during a fight, and he was made a
gar [became Nuer]. No Anywaa considered him as Anywaa, let alone the Nuer.
But his brother is still Anywaa. Gar is one of the important means of becom-
ing Nuer. The other is through marriage and the Nuer language. With these
three you become a Nuer over time. When the Nuer migrated, they came to
Anywaa land. The Anywaa left and those who remained were invaded and were
assimilated into the Nuer. Many Anywaa from Akobo became Nuer this way.
Similarly, many Opo were also made Nuer. That is how the Nuer became many.
Above all, if you do not have gar, you will not be able to marry. Many Anywaa
who are living with the Nuer started taking gar because of that. (Reverend
Stevenson, Nuer Mekaneyesus Church, Gambella town, September 2000)

Gar, viewed as a practice with ideological implications, operates at different levels in

the process of assimilation. For one thing, non-Nuer men are encouraged to partic-
ipate in the Nuer version of masculinity. Typically, when a Nuer girl declines a mar-
riage offer by a non-Nuer, she says that she does so because the would-be groom has
not been initiated. Many Anywaa men, particularly in the Akobo area, where some
Nuer neighbourhoods are located, have been initiated. The Nuer tease those who
refuse to undergo gar in a rather rough fashion. Those who go through the initiation
confirm not only their masculinity, but also their lifelong commitment to the Nuer
mode of identification. Gar is thus a vivid example of the constructivist variety of
ethnic identity formation that is characteristic of the Nuer. It defines an area of cul-
tural competence in which people are required to perform in a certain way in order
to claim Nuer identity. The complete integration of the conquered Dinka and
Anywaa into Nuer society was possible partly because the men underwent an initia-
tion rite in which becoming a man is synonymous with becoming Nuer.

New Challenges to the Constructivist Mode of Identity Formation

In this chapter, I have reviewed the ways in which ethnic identity formation among the
Nuer might be understood to be constructivist in orientation, and I have offered some
explanations for this orientation, or presented materials that might contribute to explaining
it. As Hutchinson (1985, 1996) has argued, quite cogently, the assimilationist tendencies of
the Nuer are consistent with their transhumant pastoralist adaptation, especially with their
use of cattle as a form of wealth, as a medium of exchange, as a kind of political capital and
as a religious symbol. ‘The cow creates the person’, the Nuer say (Hutchinson 1996: 60),
and we have seen many examples of this, including the paying of bridewealth, the assump-
tion by the diel of bridewealth debts for rul or jaang, the creation of new kinship links, based
on friendship, through the sacrifice of a bull and so on. Clearly, this generally open orienta-
tion towards the incorporation of outsiders was encouraged further by particular historical
and political circumstances, such as the eastward expansion of the Jikany Nuer and the com-
petition in all Nuer groups among potential local leaders. And, clearly, the assimilationist
orientation has been a boon to the Nuer – one might even call it the secret to their success.
72 Playing Different Games

The constructivist orientation of the Nuer is evident, however, not only in the
emphasis on assimilating non-Nuer but also in recognizing the potential for the loss of
ethnic identity. In a reversal of the trend discussed so far, some groups of Nuer have
indeed been assimilated to other ethnic groups. The Atuot and the Kuok are two exam-
ples of the loss of ethnic identity that are frequently recounted by the Nuer. Both were
Nuer sections that became Dinka and Jur-Luo, respectively. In fact, a popular saying
reminds the Nuer of the need actively to maintain ethnic identity: cien bi juor cet ke kuok
(‘I should not disappear like Kuok’). All of this is consistent with the interpretation that
contemporary Nuer local communities are constructed, not only on the basis of patri-
lineal descent, but also through ritual practices and flexible modes of recruitment cross-
cutting descent-group and ethnic boundaries. The Nuer concept of identity is not fixed.
Identity is something that is acquired and has to be actively maintained, lest one lose it.
Under new circumstances, however, Nuer understandings of their own ethnic
identity have begun to change. In the contexts of ethnic federalism in Ethiopia and
civil war in the Sudan, an incipient form of primordialism has reemerged among the
Nuer or, rather, among various Nuer groups (cf. Hutchinson 2000: 9–12).
The new political structure in Ethiopia is Janus-faced. If ethnic federalism has rein-
forced inter-ethnic boundaries, it has also fostered intra-ethnic political fragmentation. The
flexible yet effective identity system of the Nuer has been undermined since 1995 with the
introduction of electoral politics on the basis of group identity. The construction of pri-
mordialist boundaries within the Gaat-Jak amalgamation is a case in point. Although there
was already an economic incentive for the Thiang to move towards a separate identity vis-
à-vis the other ciengs of the Gaat-Jak, it is electoral politics that has encouraged Thiang
elites to articulate a separate identity in the competition for political power.
As noted above, the Jikany Nuer have, in recent memory, been divided into three
tribal groups: Gaat-Jak, Gaat-Jok and Gaat-Guang. Drawing on Kir mythology and
thus creating for themselves a new genealogical charter, the Thiang have advocated
splitting Gaat-Jak into two parts, namely, the Thiang and Gaat-Guong, including
only the remaining former members of Gaat-Jak (Cieng Cany, Cieng Wau, Cieng
Nyajani and Cieng Reng).8 The result has been the emergence of a new, four-fold
tribal division: Thiang, Gaat-Guong, Gaat-Jok and Gaat-Guang.

8. In propagating a separate Thiang identity, Thiang elites refer to Kir mythology as follows. In
the Jikany origin myth, it will be recalled, the apical ancestor, Kir, had three sons: first Thiang,
the eldest, and then Khun and Jok. After Kir died, Thiang married Jok’s mother and begot
Nyang in Kir’s name. Despite his seniority, Thiang joined Khun and, together with the Cieng
Reng, they formed a wider amalgamation. This amalgamation was given the ox name Jak, and
the descendants of Thiang, Khun, and Reng came to be known as Gaat-Jak, children of Jak.
The descendants of Jok and Nyang came to be known as Gaat-Jok and Gaat-Guang, respec-
tively. To mark the new boundary constructed with reference to this myth of biological
descent, the Thiang have recently reinstated the older ox name of Khun, viz. Guong, when
addressing the three groups descended from Khun (the Cieng Cany, Cieng Wau and Cieng
Nyajani) and the group affiliated with them, the Cieng Reng. This serves to reinstate the orig-
inal distinction between Thiang, the eldest son of Kir, and Khun, one of the younger sons,
thus making it possible to create a new division within Gaat-Jak between Thiang, on one hand,
and Guong, on the other.
 The Nuer Constructivist Ethnic Identity Formation 73

This re-categorization of tribal groups, which elevates the Thiang by placing

them on an equal footing with Gaat-Guong, Gaat-Jok and Gaat-Guang, bore fruit
during the May 2005 election, when the Thiang managed to get the lion’s share of
what had previously been the Gaat-Jak’s quota of seats in the council of the Gambella
regional state, almost in parity with the Gaat-Guang and Gaat-Jok tribes.9 Aside
from shaking up the political organization of the Jikany Nuer in Gambella, the
success of Thiang identity politics might be a sign of future developments, for there
are other units of identification within the Jikany Nuer that are also vulnerable to
‘deconstruction’.
Institutionalized identity politics have also inserted primordialist currents in
Nuer identification through the politicization of the diel /rul /jaang distinction. The
terms of the political debate and the mobilization of political support by the Nuer
electoral candidates in the May 2005 election featured this aspect of the reconfigu-
ration of Nuer identities as well. The political competition between Peter Lual and
Yie Chuol in the nomination process is a case in point. Both competed to represent
the Cieng Jenyang/Thiang. When Peter obtained the party’s nomination, Yie
appealed to the Cieng Jenyang constituency, framing the political competition in
terms of the purity of Cieng. He also claimed the status of a dil while labelling Peter
as an outsider, a rul, whose father originally came from the Cieng Nyajani. Yie
managed to persuade the representatives of the Cieng Jenyang elders to write a peti-
tion to the Nuer Peoples Democratic Organization (NPDO) to reconsider its
approval of Peter as the representative of the Cieng Jenyang. Peter, however, secured
the party’s backing, which put pressure on the Cieng Jenyang elders to persuade Yie
to accept Peter’s nomination. Yie withdrew his candidacy from the party, ran as an
independent, and campaigned extensively on the ‘dil ticket’. In one of the Cieng
Jenyang’s public meetings, Peter defended his Thiang identity as gaatnyiet and high-
lighted Yie’s ‘dubious’ genealogy as follows:

In that meeting I raised the issue of dil. I reminded the Cieng Jenyang elders
that my mother is a dil in Cieng Jenyang. I gathered that Yie is not a dil as he
claims to be. In fact, his mother is not a Cieng Jenyang, and the origin of his
grandfather is not clear. On that basis, I appealed to the elders that there are a
lot of people who are not originally Cieng Jenyang and the issue of dil will bring
division among the people. Many people in Cieng Jenyang were originally
Gaat-Guong. Raising the question of who is pure would undermine the
Thiang’s position in Nuer politics. (Peter Lual, Wechdeng village, 3 April 2005)

9. The distribution of political power in the Gambella regional council during the May 2005
election may be summarized as follows. While the total number of Members of Parliament
(MPs) in the regional council is 81, 33 of these positions are allocated to the Nuer. Previously,
these 33 seats were distributed among the tribal groups as follows: the Gaat-Jak had 17 MPs,
the Gaat-Jok had 9, and the Gaat-Guong had 7. Following the establishment of a separate
Thiang identity, the number of MPs among the cieng of Gaat-Jak were redistributed as follows:
the Thiang received 6, the Cieng Cany 4, the Cieng Nyajani 4, and the Cieng Wau 3.
74 Playing Different Games

As the microcensus from Wechdeng village indicated, more than fifty per cent of a
given village at any point in time is composed of families other than, but affiliated
with, the dil. In the short term, the discourse of cieng purity, though an effective
strategy for individuals who can claim the status of dil, is likely to undermine the via-
bility of Nuer society, for it could ultimately result in an authenticity discourse along
‘pure’ and ‘non-pure’ Nuer lines. Although Peter managed to be elected as a repre-
sentative of the Cieng Jenyang, even he framed his identity claim in the language of
purity: while drawing attention to Yie’s pretence as a dil, he drew equal attention to
his own blood connections with the Cieng Jenyang through his mother.
The civil wars in southern Sudan and their steadily increasing ethnic framing,
both by the government of the Sudan and the various southern Sudanese political
actors, have also served to foster a primordialist reconfiguration of Nuer ethnic iden-
tity. The various southern Sudanese politico-military organizations that claim to rep-
resent the Nuer and the Dinka, the two largest ethnic groups in southern Sudan,
have injected an ethnic current into the politics of the liberation movement. In this
context, the Nuer have departed from their traditional assimilationist drive, insofar
as they have begun to view ethnic strangers not as assets but as liabilities. In the new
politico-military game, there are, after all, no guarantees that assimilation will
proceed as it did earlier, as this would require that, in situations of conflict, captives
identify with their captors (see Chapter 9 for a fuller exposition of the impact of the
Sudanese civil wars in the reconfiguration of Nuer ethnic identity formation).
In this chapter, I have described and analysed the constructivist mode of Nuer
ethnic identity formation. Nuer constructivism is evident in the flexible use of
descent, alliance and assimilation in the formation of local communities and in the
ritual construction of manhood, which is open to outsiders. Becoming Nuer is not
merely a possibility; rather, outsiders are actively assisted in taking this step.
I have also attempted to explain the conditions of the Nuer constructivism. Beyond
the intrinsic superior moral value that the Nuer attribute to their concept of ethnic
identity, there is also an instrumentalist and materialist current. That is, the assimila-
tionist drive has a political and economic rationale, insofar as it allows each cieng to add
to its numbers and, hence, to gain strength vis-à-vis intra- and inter-ethnic rivals.
The discussion in this chapter has also shown the dynamic reconfiguration of
identity discourses, which, in the case of the Nuer, has involved an earlier shift from
a primordialist to a constructivist form of identification and, most recently, a ten-
dency to revert to primordialism under new socio-political circumstances.
Taken together, the two chapters of part two serve to demonstrate the contrast-
ing modes of ethnic identity formation of the Anywaa and the Nuer and their diver-
gent trajectories. The following chapters in part three explain the causes of Anywaa
and Nuer conflict and determine the extent to which the contrasting ethnic identity
formation are implicated in the conflict situation.
 Part III
The Encounter
As has been shown in the preceding chapters, causal explanations for ethnic conflict
are often one-dimensional and, therefore, of limited relevance in explaining the con-
flict between the Anywaa and the Nuer. The following six chapters explain that con-
flict with reference to the interaction among three variables: first, contrasting modes
of ethnic identity formation (the identity variable); second, competition over scarce
natural resources (the resource variable); and, third, the different ways in which
ethnic groups are incorporated into an ethnically stratified state system (the power
variable). Having characterized the contrasting modes of ethnic identity formation
in Chapters 2 and 3, I turn now to the resource variable and its interplay with the
identity and power variables.
 Chapter 4

In the Riverine Lands

Conflicts are often explained in terms of the interests of groups involved,
especially their competition for resources or gains … What people are fight-
ing about is a fundamental question in conflict analysis, but there is another
equally fundamental question that remains poorly understood, namely, who
is fighting whom and why? (Schlee 2004: 135)

This chapter examines the resource dimension of the conflict between the Anywaa
and the Nuer. The basic argument of the chapter is that competition for natural
resources is relevant in Anywaa–Nuer conflict but cannot be taken as the sole cause of
ethnic conflict. One of the objects of the Anywaa–Nuer struggle is access to and
control over the riverine lands. Riverine lands in Gambella form a minimal propor-
tion of the total land surface but have an outstanding agricultural and pastoral value.
This has created scarcity of a specific type of land. Indeed, the Nuer expansion into
Anywaa-inhabited territories is primarily driven by the attempt to gain access to and
control over the riverine lands. But the same is true of the conflicts among the various
Nuer tribes. Identifying the riverine land as a bone of contention does not imply that
all Nuer are antagonistic towards all Anywaa. On the contrary, there are internal divi-
sions in each of the groups, and there are groups occupying a buffer zone and main-
taining links to both sides. To speak of the Anywaa–Nuer conflict strictly as a resource
conflict is therefore an oversimplification. In fact, with regard to natural resources,
intra-ethnic competition is even more severe than inter-ethnic competition.
After briefly recapitulating the long anthropological debate on the Nuer expan-
sion, I proceed by outlining key natural resources and their distributional pattern.
The stark imbalance in the distribution of such resources firmly establishes the
resource dimension in Anywaa–Nuer conflict. The following two sections, however,
serve to complicate the link between resource and ethnicity, showing that not all
resource conflicts are fought at the inter-ethnic level or involve ethnic mobilization.
I argue that there are two intervening variables between resource conflicts and eth-
nicity, i.e., the diverging Anywaa and Nuer schemes of interpretation and the eth-
nopolitical structure in Ethiopia. The diverging schemes of interpretation of Nuer
expansion emanate from the contrasting identity formations of the Anywaa and the
Nuer, which is discussed at length in the previous chapters. The Anywaa scheme of
interpretation converts individual Nuer economic motives into a Nuer ‘ethnic con-
spiracy’. The consolidation of ethnic boundaries in the context of institutionalized
ethnopolitics in post-1991 Ethiopia also links up with local resource conflicts and
often leads to an ethnic representation of those conflicts.
78 Playing Different Games

On the Nuer Conquest

The reasons for the nineteenth-century Nuer expansion have long been a subject of aca-
demic debate. The Jikany and Lou Nuer migrated to the east from the original Nuer
homeland (Bentiue) in present-day southern Sudan, ultimately forming large settle-
ments that are now referred to as Eastern Nuer (Jal 1987; Hutchinson 1996). At the
beginning of the nineteenth century, the Nuer were confined to the area west of the Bahr
el Jebel River (Jal 1987: 36). Successive waves of Jikany migrants appeared on the Sobat
in the 1830s, and by the end of the century, they advanced as far east as the escarpment
of the Ethiopian highlands. Several scholars have remarked on the dramatic expansion
of the Jikany Nuer at the expense of Anywaa territories in the second half of the nine-
teenth century. According to Evans-Pritchard (1940b: 8), ‘a century ago the Anuak had
occupied what is now Jikany Nuer land to the north of Sobat, parts of which are now
Jikany and Lou Nuer country to the south of that river, the banks of the Pibor to its
junction with the Sobat, and the banks of the Sobat to within a few miles of Abwong’.
Jal (1987: 125) also describes the scale of Nuer territorial expansion in a similar way: ‘It
is clear from their traditions that when the Nuer migrants arrived on the Sobat, they
found the Anuak in possession of the country immediately along the banks of the Sobat,
from Abwong in the west to the mouth of the Khor Jokau in the upper reaches of the
Sobat in the east, the whole of the Wading along the Pibor in the south and the Adura
island in the triangle country’. Perner (1997: 145) also mentions the extensive territorial
loss of the Anywaa to the Nuer, the area that extends ‘from Tobai on the Sobat to Thor
in the west of Kigible to the Adare and Obela (Mokwai) rivers in the south’. Based on
the magnitude of the territorial losses described by Evans-Pritchard, Kelly (1985: 1) esti-
mates Nuer territorial gains as fourfold of what they had before the expansion: ‘Nuer
displacement of the Dinka (and Anuak) represents one of the most prominent instances
of tribal imperialism contained in the ethnographic record’.
Generally speaking, there are five explanations for the dramatic Nuer expansion
to the east at the expanse of Dinka and Anywaa territories in the second half of the
nineteenth century. The first is structural functionalism. Propounded by Evans-
Pritchard (1940a), structural functionalism identifies the Nuer’s segmentary lineage
system, built on a ‘structural antagonism’ between the Nuer and their neighbours, as
a basis for the Nuer’s ethnic mobilization and subsequent successful expansion.
The second explanation is evolutionary, mainly associated with the works of
Marshall Sahlins. Sahlins (1961) elaborated on Evans-Pritchard’s analysis of the ‘seg-
mentary lineage’, situating it as a ‘tribal’ type within a typological sequence of evolu-
tionary stages of political organization, beginning with the band and proceeding to
the tribe, the chiefdom and the state (Sahlins 1963; 1968). In a widely read article,
‘The Segmentary Lineage – An Organisation of Predatory Expansion’, Sahlins
(1961) analysed the Nuer social organization and its expansion as a form of adaptive
mechanism developed by tribal groups invading territory already occupied by other
tribes: ‘The segmentary lineage system is a social means of intrusion and competi-
tion in an already occupied ecological niche. More, it is an organization confined to
societies of a certain level of development, the tribal level’ (1961: 323).
A third explanation is ecological: the population pressure that resulted from high
floods, particularly in the Nuer-inhabited areas (Southall 1976; Johnson 1989;
 In the Riverine Lands 79

Hutchinson 1994). According to Hutchinson (1994: 644), ‘periodic and sustained

flooding is highly significant in the nineteenth century advancement of the Nuer
eastwards across the White Nile’. Kelly (1985) proposes a cultural explanation. The
key to the dramatic Nuer expansion at the expense of their neighbours, he suggests,
is the difference in their bridewealth systems: ‘Nuer bridewealth requirements are
roughly twice as large as those of Dinka … Both the larger size and particular com-
position of Nuer herds imposes significantly greater grazing requirements on Nuer
local populations than those of the Dinka’ (Kelly 1985: 112–13). The link between
Nuer expansion and their bridewealth system, argues Kelly, is through grazing land
requirements and the recurrent shortage of dry-season pasture. Hutchinson (1985)
and McKinnon (2000) suggested a different type of cultural explanation for Nuer
expansion, the tension within the agnatic structure, which is discussed at length in
Chapter 3. To recap the thrust of the argument, the struggle over cattle inheritance
and competition over community leadership among the agnates generated the assim-
ilationist drive both intra-ethnically (significance of matrilateral ties) and inter-eth-
nically (assimilation of ethnic strangers).
In a fifth and final explanation, competition for political leadership provides the
impetus for the Jikany expansion (Johnson 1982; Pal 2006). According to Johnson
(1982: 185), ‘the best-known Nuer migration, that of the Jikany under Latjor early
in the last century, was the outcome of political disputes caused by land scarcity
under conditions of the expansion of swamplands in the Jikany homeland … Latjor
came into conflict with the existing leaders of the Jikany and decided to migrate’. Pal
(2006: 2) also describes the Jikany expansion in a similar way: ‘Latjor had minimal
chances to ascend to the Nuer leadership due to the fact that he belonged to a minor-
ity group that could not succeed in competing with other dominant figures of the
community’. In the Nuer oral traditions I collected from Gambella, the role of Latjor
in the expansion of the Jikany is also a central theme, but other factors such as a
search for better land and discrimination of the Jikany by the ‘original’ Nuer are also
indicated. The following is one of these oral accounts:

The land was not enough for all of the Nuer in Bentiue. Together with his
Dinka friends, Latjor went out to explore new lands. A person called Padiel
Gakgak was sent by Latjor to faraway places to check the area. He wondered
why the birds, which came from the east, were fat. He went in the direction
where the birds came from. On the way, he met a lot of lät [a Nuer term for
a subhuman creature1] in the Toch, Uriem and Lare areas. Padiel discovered
Yom, the only dry place in the area that was not flooded. Padiel managed to
kill the lät and took the tail of a giraffe with him to show Latjor that the place
was very good. Latjor was very much impressed by the richness of the area. But
his people were not willing to leave their area and go to unknown places. Then
Latjor wanted to consult a ghok [prophet] to convince the people to follow
him. In order for him to know which ghok was stronger, he planned a trick.

1. In the Nuer belief system, the term ‘subhuman’ refers to beings that are half-human and half-animal.
80 Playing Different Games

He hid his jeop (axe) in his hair. Latjor’s hair was very long. He pretended that
he had lost his axe. He went to many ghok. They all told him that someone
had stolen it and he should be careful of evildoers. Finally, he met a powerful
woman ghok named Nok.2 She immediately recognized his trick and said to
him, ‘Why are you lying? Didn’t you hide the jeop in your hair?’ Latjor was so
impressed by her knowledge that he told her all his secrets. He wanted to
marry her, although she was very old. He paid hundreds of cattle. Nok per-
suaded the people to follow Latjor. Latjor’s people went in the direction of
Malakal. When they reached Kakak, they prayed to Kir and said, ‘Kir, show
us the way to cross the river’. There was a big mountain called Cham Kuntar.
It broke by itself. It made a bridge so that the people could cross the river. Nok
collected some human faeces, which she miraculously turned into a canoe. But
Latjor’s people did not know how deep the river was. In the middle of the river
they saw a ngok (blue heron) standing. Then they realized that the river had
become shallow and they could easily cross it. After crossing the river they
reached a place called Maluts. They called it was-ngok, named after the bird.
In Maluts they found the Dinka. The Nuer fought with the Dinka, but they
were defeated. Nok advised Latjor to half-kill some important Nuer in order
to deceive the Dinka. The Dinka came and killed the wounded. They were
very happy that all the important Nuer were killed. The Nuer then attacked
the Dinka and the Anywaa, but some of them became Nuer after they made
gar [male initiation mark]. The Anywaa were then living in Malakal (Thiang
Louny, Gaat-Jok Nuer elder, Gambella town, November 2000).

All of these explanations indicate in one way or another that access to and control
over natural resources was an important factor in the nineteenth-century expansion
of the Jikany. The land factor is particularly prominent in Nuer’s expansion into
Anywaa areas.

Key Natural Resources and Their Distribution Pattern in the

Gambella Region
The rangeland in Gambella and the adjacent Upper Nile region of southern Sudan
are subject to severe flooding from three major rivers: the Baro (Openo), Gilo and
Akobo, which together form the Sobat River basin, itself a tributary of the White
Nile. Flooding is caused by the high amount of rainfall that flows into the rivers of
Gambella from the western Ethiopian highlands. These highlands, where the rivers
originate, receive some of the heaviest rainfall in Ethiopia. The floods cover the fields
for months, finally leaving behind a fertile (alluvial) soil. This soil can support con-
tinuous cultivation in both rainy and dry seasons due to the annual replenishment.
Agricultural production on the riverbank is stable and fruitful, affected by neither
drought nor soil exhaustion (Kurimoto 1996: 44–45). The banks of the Baro River
‘are lined with some of the richest and most coveted agricultural lands … because

2. In a similar myth documented by Jal (1987: 37), this diviner is called Nyagueach.
 In the Riverine Lands 81

they yield lush crops of maize and tobacco during the height of the dry season with
only the moisture that rises from the river’ (Hutchinson 1996: 114). Though detri-
mental when they are intense, the floods nevertheless create and regulate the distri-
bution of key natural resources in the Gambella region. The flooded land is crucial
for cultivation and pasture during the early period of the dry season.
These key resources are scarce. Although there are vast amounts of arable land in
Gambella, currently only 2.4 per cent of it is being cultivated.3 Exploited through
hoe cultivation, rather than with the plough, as in most other regions of Ethiopia,
landholdings in Gambella are, on average, 0.5 hectares. Cultivation in the region
involves three farming systems: sedentary rain-fed cultivation, flood-retreat cultiva-
tion along the banks of the rivers, and shifting cultivation. Of the total cultivable
land, sixty-five per cent is savannah, thirty per cent forestland and 4.5 per cent
marshland. Only 0.5 per cent is suitable for flood-retreat cultivation, but neverthe-
less it supports a significant percentage of the farming population (Ellman 1972: 12).
Nearly sixty per cent of this moisture cultivation is concentrated along the Baro
River, largely in Anywaa-inhabited areas in Itang district.4 Thus there is a severe
shortage of riverine land.
The plain of Gambella is one of the most suitable areas for raising cattle (Dereje
2006b). The region’s livestock population is largely restricted to two of the nine dis-
tricts, Jikaw and Akobo, accounting for over eighty-five per cent of the livestock.5
Ellman (1972: 13) classified the range vegetation into three major categories: season-
ally flooded pasture, pastures on the plain, and pastures on the piedmont and adjoin-
ing uplands. The major sources of livestock feed are the open woodlands, riverine forest
and woodland during the wet season, and savannah grassland, riverine forest and
woodland during the dry season. No other food supplement is provided to livestock.
Savannah grassland, a relatively scarce type of pasture, nonetheless provides the main
source of animal feed during the dry season. Settlements near the major rivers are best
positioned to access these lands, involving shorter transhumance than villages far from
the rivers. Of the total land area classified as natural grazing area in Gambella, only
sixty-four per cent is currently utilized by livestock.6 The best rangelands are located
along the Baro and Gilo Rivers. These rivers have a high overflow capacity, which
creates suitable conditions for dry-season grazing lands and flood-retreat cultivation.
For this reason, the Baro basin, most of which is inhabited by the Anywaa, has been
a pastoralist magnet, drawing towards it what Sahlins (1961: 63) once referred to as the
Nuer version of the Drang nach Osten.7 Towards the west, the alluvial strip is narrower
and gradually fades. The scarcity of key natural resources is accentuated by the distri-

3. Cf. Bureau of Planning and Economic Development, Gambella Regional State, 2000.
4. Cf. Disaster Prevention and Preparedness, Gambella Bureau.
5. Cf. Livestock Research Report, Pastoralist Livelihood Programme, Agency for Cooperation
and Research in Development, 1999.
6. Cf. Conservation Strategy of the Gambella Region, Bureau of Planning and Economic
Development, Gambella Regional State, 2000.
7. Drang nach Osten is a German expression for ‘Drive towards the East’, a term first used by
‘19th century intellectuals, and later by Nazi propaganda to explain Germany’s desire for land
and influence in Eastern Europe’ (Carlson 1937: 233).
82 Playing Different Games

bution pattern. At first the idea of land scarcity in Gambella sounds contradictory,
because Gambella exhibits one of the lowest population densities in the country.8 A
close examination of the settlement pattern by district, however, reveals a different
picture. The population density of the districts varies from the lowest (2.8 people per
sq kilometre) to the highest (25.5 people per sq kilometre) as shown in Table 4.1.

Table 4.1 Land size and population density of districts

District Area (km²) Population Density Inhabitants Claimant

Itang 1,837.04 25,175 13.7 Anywaa Anywaa

Nuer
Opo
Gambella 2,859.85 41,867 14.6 Anywaa Anywaa
Highlanders
Nuer
Komo
Majangir
Abobo 3,515.78 18,618 5.3 Anywaa Anywaa
Highlanders
Majangir
Gog 7,138.60 20,285 2.8 Anywaa Anywaa
Jor 2,488.13 9,181 3.7 Anywaa Anywaa
Akobo 3,830.47 32,862 8.6 Nuer Nuer
Anywaa
Jikaw 2,192.82 55,922 25.5 Nuer Nuer
Anywaa
Godere 1,939.32 45,090 22.2 Majangir Majangir
Highlanders

(Source: Central Statistics Authority, 2005. National Statistics: Gambella Region, supplemented in the
column with the header ‘Claimant’ by data from the authors’ field notes)9

The variation in population density becomes even more acute if we divide the dis-
tricts by ethnic group. Of the nine districts, five are inhabited by the Anywaa and
two by the Nuer. Itang is a mixed-settlement area with a roughly equal population
of both groups. However, the 1994 census put the Nuer population at forty per cent
and the Anywaa at twenty-seven per cent, indicating a stark imbalance in the ethnic-
based land-people ratio. As a result, the two Nuer districts, Akobo and Jikaw, show
the highest population densities in the region compared to districts inhabited by the
Anywaa. The situation is further complicated because the population pressure is cur-

8. According to the Ethiopian Central Statistics Authority, the national population density in
2005 was 70/km2, which contrasts sharply with the population density of Gambella, which
was 9.57/km2.
9. This does not include the population density of the Anywaa-inhabited district of Dimma.
 In the Riverine Lands 83

rently concentrated in the Itang district, which has a relatively high Anywaa density.
It is no coincidence, therefore, that the major site of conflict between the Anywaa
and the Nuer is in and around this district, which combines the best rangeland with
the riverine lands.

Table 4.2 Distribution of animal population by district

District Number of cattle Number of sheep Number of goats

Itang 13,100 62,000 11,000
Gambella 5,100 6,000 1,400
Abobo 1,500 1,000 1,100
Gog 1,300 1,000 1,700
Jor 6,600 4,400 7,100
Akobo 135,000 324,000 43,700
Jikaw 187,000 410,000 36,300
Total 349,600 808,400 102,300

(Source: Bureau of Planning and Economic Development, Gambella Regional State. Conservation
Strategy of the Gambella Region 2000: 33)

The distribution of the animal population is also uneven, as shown in Table 4.2.
According to the Bureau of Planning and Development of the GPNRS, the natural
grazing area of Gambella covers an area of 1,804,800 hectares. Slightly more than
half of that area is extended over the eastern part of Gambella (Abobo, Gambella,
Gog and Jor districts), which supports a minimum livestock population. The other
forty-seven per cent, on the other hand, is utilized and seasonally overstocked as the
transhumant herds range over the area, trekking from the rivers during the wet
season to use the accessible upland grazing area and, during the dry season, moving
back to the perennial rivers for drinking water and the new supply of sprouting
grass.10 Of the cattle-rearing districts, Jikaw and Akobo have the highest cattle den-
sities, and there is continuous pressure on the relatively less-populated district of
Itang. As indicated in Tables 4.1 and 4.2, the highest human and animal densities are
in Itang, Jikaw and Akobo districts, where the majority of the pastoralists live.
Currently human and cattle population pressure gravitates towards the Baro River in
and around Itang district, for it has the ideal combination of prime rangeland and
riverine lands suitable for flood-retreat cultivation. The various groups of Nuer use
different strategies to gain access to these vital natural resources, most of which are
located in Anywaa territories.

Inter-ethnic Conflict: an Overview of Past Resource Conflicts

The Anywaa have responded to the Nuer territorial expansion in different ways.
Initially the Anywaa resisted, but they were eventually overwhelmed by the Jikany

10. Cf. Bureau of Planning and Economic Development, ‘Conservation Strategy of the Gambella
Region’ (2000: 33).
84 Playing Different Games

Nuer. During the first two decades of the twentieth century, however, political and
military fortunes reversed in favour of the Anywaa, who, thanks to their geographic
proximity to the highlands, took advantage of the arrival of the representatives of the
Ethiopian state and the resultant access to firearms. This led to the emergence of
powerful nobles, and ‘the Anywaa were catapulted from near extinction to ascen-
dancy’ (Bahru 1976: 107). In 1911–12 the Anywaa carried out a series of large-scale
raids against the Jikany and the Luo Nuer. Although the Gaat-Jak Nuer ultimately
prevailed over the Anywaa in 1912 at the Battle of Biot, they learned the important
lesson of modifying their strategy from violence to inter-ethnic negotiations to gain
access to resources. Prior to the 1990s, Biot was the last instance of total mobiliza-
tion in the conflict between the Openo Anywaa and the Gaat-Jak Nuer.
After the battle, the Anywaa–Nuer conflict was localized, and various groups of
the Gaat-Jak Nuer made peace with the Openo Anywaa at the Baro River. The first
initiative was taken by the Thiang Nuer, who negotiated a peace agreement with the
Pinyman Anywaa, cemented by inter-ethnic marriages and economic ties. The socio-
economic ties between the Thiang Nuer and the Openo Anywaa are described by a
Nuer elder in the following narrative:

Peace was made between the Thiang and the Anywaa at Tiengdeng [an important
lagoon along the Baro River near Pinyman village in Itang district]. The Anywaa
women from Pinyman village were sent to Tiengdeng to give the Nuer maize flour
as a peace gesture to end the conflict. Thowat Gac, the leader of the Cieng Jenyang,
took half of the flour and gave the remaining half back to the Anywaa, saying ‘as
we share this food, may we always live together in peace’. The Cieng Jenyang
Thiang renamed Tiengdeng as Banubdak, which, in the Nuer language, means
‘dividing the flour’. Thowat Gac and kwaaro Deng Cuia of Pinyman village
became friends. Thowat gave kwaaro Deng a black ox and kwaaro Deng allowed
the Cieng Gac to come to the Baro during the dry season. Thowat also made
contact with the kwaaro of Pinymoo. After the Banubdak agreement, the Pinymoo
kwaaro gave his daughter Abang to Thowat as his wife. The Nuer named her
Bonge. Thowat gave the Pinymoo two oxen and a feast was made in Pinymoo
village. (Elder Gatluak Choul, Wechdeng village, August 2000)

Since this time the Thiang Nuer have managed to establish and maintain closer links
with the Anywaa than have any other Nuer tribe. The two have developed not only
social and economic ties but also forged politico-military alliances. The traditional
Anywaa political system tied people closely to particular villages. Warfare was the
mechanism through which the Anywaa constructed their village identity. In inter-
village warfare, the kwaari sought military alliances with Nuer clans. In such warfare
among the Openo Anywaa, for instance, the Thiang Nuer participated by support-
ing one or the other of the villages depending on the existing socio-economic ties.
The following narrative illustrates these ties across ethnic boundaries:

A long time ago a group of Gaat-Guang Nuer migrated to an unknown place

in Anywaa country but they did not know that they were near to the Anywaa.
 In the Riverine Lands 85

One day their bull went to a place called Walo where he ate the maize harvest.
His Nuer owner realized from its faeces that his bull had eaten maize and
wondered where it had got the maize from. The Anywaa owner of the maize
also wondered what was eating his maize. The Nuer followed his bull and dis-
covered the garden from where it got the maize. The Nuer told the Anywaa
that they were staying in the nearby bush. The Anywaa offered to let the Nuer
come to Walo and settle near them. The Nuer went to Walo and settled there.
One day there was war between the Walo and Pekade [Pokedi]. The Nuer
sided with the Walo Anywaa and together they defeated the Pekade Anywaa.
The neighbouring Pol Anywaa heard about the victory of the Walo and asked
the Nuer to help them fight their Anywaa neighbours. The Walo kwaaro
refused but the Nuer went on their own and fought on behalf of the Pol
Anywaa. The kwaaro of Pol offered to let the Nuer be his people in return for
the land which he gave them to cultivate. The Nuer accepted the offer and
they went to Pol. During the time of Litgach [the age-set initiated in the
1920s], the kwaaro of Pol, Jiokdwor and a Cieng Mek Nuer called Lee Ruey
became friends because they were both good hunters. One day, Lee went to
Pol and stayed for few days. While he was in Pol, a war broke out between
Pol and Malwal. Lee participated in the war and he managed to kill many of
the enemy because he was a sharpshooter. For that reason kwaaro Jiokdwor
blessed the relationship between the Cieng Mek and the Pol Anywaa. After
Lee left, a second war broke out with Malwal when Jiokdwor was killed.
Before his death, Jiokdwor reminded the Pol Anywaa to keep the relationship
with Lee’s people and demanded that Lee sacrifice a cow on his grave.
Jiokdwor also said ‘in case you [Pol Anywaa] leave the land, give it to the
Cieng Mek because we do not have any relatives other than them’. After the
death of Jiokdwor, a Cieng Mek man called Bol Juc married an Anywaa
woman from Pol who already had two children from another Anywaa. Bol
took them to the Nuer village and they were renamed Maker and Dol and
they became diel. (Elder Tut Gatwac, Wechdeng village, November 2000)

In the politico-military alliances that have emerged, the Anywaa have also regarded
the Thiang Nuer as a buffer zone sheltering them from less familiar and often more
aggressive groups of Nuer, while the Thiang consider all ‘unoccupied’ Anywaa land
in Itang district as their hinterland, which they call nyamdoar. Following the Thiang,
various groups of the Gaat-Jak Nuer made similar peace agreements and exchange
networks with the neighbouring Anywaa villages along the Baro. In the first half of
the twentieth century, the interaction was seasonal. The Nuer went to the Anywaa
areas along the Baro River at the height of the dry season (February to April), and
during the wet season (May to March) they occupied camps in the upland areas. This
seasonal movement was regulated in such a way that it clearly featured the symbolic
ownership right of the Anywaa over the rangelands around Itang. A kwaaro would
tell the Nuer when to come, where to temporarily settle, and how far to keep their
cattle from the Anywaa fields. In fact, as long as the Nuer recognized the authority
of the kwaari and their presence did not cause any trouble, their seasonal movements
86 Playing Different Games

to the riverine lands provided the Anywaa with both an economic opportunity and
a military power resource; consequently, the Nuer’s coming was something to look
forward to. With an incipient ethnosystem based on symbiotic exchanges along the
Baro, the zone of tension gravitated along the Gilo River to the Jor district, where
other groups of Nuer put pressure on Anywaa villages due to the scarcity of land.
Some groups of the Gaat-Jak such as the Cieng Nyajani, the Cieng Reng and the
Gaat-Guang were relative latecomers to Gambella. As such, these groups of Nuer,
finding most of the fertile land along the Baro already occupied, oriented their
expansion towards the Gilo River, the home of the Jor Anywaa. By the time the
Cieng Nyajani came to Jikaw, most of the dry-season settlements along the Baro
River had been largely occupied by the Thiang. A group of the Cieng Nyajani settled
southwards in the direction of the Gilo River, where they encountered the Jor
Anywaa, some of whom they displaced. Given the acute desire to gain access to the
rivers, the Cieng Nyajani used various strategies, from local economic exchanges,
inter-ethnic marriages and the use of occult power to translocal political networking.
During the Italian occupation of Ethiopia (1936–41), the Cieng Nyajani leaders, for
instance, appealed to the Italians for help in their fight against the Jor Anywaa
(Ojullu 1987: 43).11 In 1965, nine Cieng Nyajani Nuer leaders also appealed to the
Ethiopian government to support their claim for access to the Gilo River.12 They did
not get as much support as they had expected. When they were unable to manipu-
late the translocal political networks, the Cieng Nyajani resorted to an effective inter-
nal mobilization. To that end, a famous Cieng Nyajani prophet, Nyachay Nuer,
organized a military campaign. Nyachay was killed by the Jor Anywaa in 1976. The
Cieng Nyajani then organized another force led by a well-known elder, Yang Thuan.
Yang was also killed by the Jor Anywaa. The death of these two prominent figures
angered the entire Cieng Nyajani community, which launched yet another attack on
one of the resource-rich Jor Anywaa settlements at Angela village on the Gilo River.
The Cieng Nyajani took over Angela and drove the Anywaa deeper into the Jor hin-
terland. In 1977, the Anywaa of Angela village, with the support of other Jor villages,
recaptured Angela, losing forty of their number (Kurimoto 1997: 806). In 1978, the
Ethiopian government intervened by putting pressure on both sides to end the con-
flict. With increased government intervention in local conflicts, direct military con-
frontation became costly. In the changing ‘rules of the political game’ set by
government power, the Cieng Nyajani had to reorient their strategy. A section of
them entered into peaceful economic exchanges with Anywaa villages in Jor district,
and another group settled in Itang district along the Baro River.
The violent expansion of the Lou Nuer into the Akobo region of Ciro Anywaa in
present-day southern Sudan is another example of the Nuer forcibly occupying
Anywaa lands after the nineteenth-century Nuer expansion. Originally the Lou Nuer
settled in the Waat area, which is far from Akobo and poorly drained. The first

11. Evans-Pritchard also reported on the Italian support for the Nuer in their fights against the Jor
Anywaa (1947: 72–73).
12. Letter written by the Ministry of Interior to the Governor of Gambella District, Metu Archive:
2066/1, 2/8/1949, Ethiopian calendar, C/1957.
 In the Riverine Lands 87

encounter between the Ciro Anywaa and the Lou Nuer was in 1911–12, during which
the famous Anywaa king, nyiya Akwei Cham, checked their territorial expansion and
raided their cattle. The British intervened to protect the Lou Nuer from Anywaa raids
by establishing an administrative centre at the confluence of the two tributaries that
form the Akobo River (Collins 1971; Bahru 1976). Driven by their perennial desire to
access the rivers of the Akobo region during the dry season, and placed by the British
in the same administrative district with the Ciro Anywaa, the Lou Nuer gradually
expanded into traditional Ciro Anywaa territories. Initially the Lou paid tribute to the
nyiye for the use of the Akobo River. Some sections of the Lou also bought land from
the Ciro Anywaa. In due time, the socio-economic exchanges resulted in the complete
occupation of western Akobo by the Lou Nuer. The Ciro Anywaa were not happy
about this territorial loss, and resisted Lou expansionism through all possible means.13
Since the 1960s, the Ciro Anywaa hold over the Akobo region has been further
undermined by the Sudanese civil wars. During the first civil war (1955–83) most of
the Anywaa villages around Akobo town were burned to the ground by the Sudanese
Government Army, which considered the Ciro Anywaa ‘collaborators’ and support-
ers of the southern Sudanese rebels. As a result of these military campaigns, many
Ciro Anywaa crossed the border and resettled on the Ethiopian side. This state of
affairs facilitated further expansion of the Lou into Anywaa territories. The outbreak
of the Second Sudanese Civil War (1983–2005) brought more repression of the
Anywaa by the government of the Sudan. This was followed by a major conflict
between the Ciro Anywaa and the Lou Nuer in 1983. The Lou Nuer used their con-
nections with the Sudanese army to insure access to the Akobo River. As a result of
these two wars, Akobo became de facto Lou Nuer land, which they now claim on the
basis of the right of conquest, known in Nuer as camun nak. The Anywaa were left
with two options: assimilate or vacate. Nearly all the Anywaa vacated and joined
their kin on the Ethiopian side of the border. The rise of a Nuer-led politico-military
organization in the 1990s, the Southern Sudanese Independence Movement (SSIM),
and its connection with the Lou Nuer further pushed the Anywaa out of the Akobo
area.14 The SSIM ultimately joined the government of the Sudan after the 1996 peace
agreement between the two. The military power and the political standing of the Luo
Nuer in the Akobo region were thereby greatly enhanced.
It is to be noted, however, that neither the conflict between the Cieng Nyajani
Nuer and the Jor Anywaa, nor between the Lou Nuer and the Ciro Anywaa, led to
total ethnic mobilization. These conflicts were localized and, in fact, resulted in a
more severe intra-ethnic resource conflict than an inter-ethnic conflict. The expan-
sion of the Lou into the Ciro Anywaa areas brought them closer to the Gaat-Jok
Nuer. Relying on their connections with the SSIM, the Lou encroached into Gaat-
Jok grazing lands. The Gaat-Jok responded by forging closer political ties with the
Sudan People’s Liberation Army (SPLA), which was in conflict with the SSIM. This

13. This includes putting spears in the river to discourage Lou movements (Jal 1987).
14. For a fuller account of the impact of the Sudanese civil wars on Anywaa–Nuer relations, see
Chapter 9.
88 Playing Different Games

brought about new local power relations. The Lou claimed the entire Sobat basin as
a lasting solution for their uncertain access to vital natural resources. The tension
between the Gaat-Jok and the Lou escalated into a major war in 1993, commonly
known as the Lou–Jikany war. The Gaat-Jok were overwhelmed by the Lou. Over a
thousand people are said to have been killed in this conflict (Wal 1992), which also
produced a massive displacement of the Gaat-Jok, most of whom resettled in the
Anywaa-inhabited district of Itang.

Contemporary Inter-ethnic Resource Conflict: a Case Study

The 1990s saw intense resource conflicts between the Openo Anywaa and various
groups of Gaat-Jak Nuer along the banks of the Baro River. In January 1998, there
were a series of clashes between Anywaa and Nuer villagers in Itang district that
lasted over a period of six months. The crisis threatened to escalate into an all-out
ethnic confrontation at the regional level. Sixty people were killed, more than ten vil-
lages were burned, and around three thousand people, largely Anywaa, were dis-
placed. The conflict came to an end in June 1998 after a state of emergency was
declared by the government and the ‘ringleaders’ of both sides were imprisoned. The
trigger of the conflict was an inter-ethnic divorce case. A Cieng Nyajani man who
was married to an Anywaa asked for the return of the bridewealth after their divorce.
The Anywaa family either refused or was unable to pay it back. In response, the Nuer
killed his father-in-law, igniting the conflict between the two communities. The
resource dimension of the conflict is obvious. There was, however, more to the con-
flict than mere inter-ethnic resource competition. For one thing, resource competi-
tion is stronger intra-ethnically than inter-ethnically. When the conflict took an
ethnic turn, political factors featured prominently. Whereas the politicization of the
conflict toned down in due time, the pattern of alliance in the continued conflict
took a non-ethnic form once again.
For a better understanding of the resource conflict, the following discusses the
four sets of actors involved in the 1998 Anywaa–Nuer conflict in Itang district. One
set of actors was the Cieng Reng Nuer from Makot village. The Cieng Reng is a
section of the Gaat-Jak Nuer who live mainly in Yom, a locality in southern Sudan.
In 1984, a small section of the Cieng Reng came to Itang district and established a
new settlement known as Makot. This was partly due to the Cieng Reng’s perennial
desire to gain access to the Baro River. This group of the Cieng Reng was led by a local
leader called Kong Diu. With the intensification of the civil war in southern Sudan in
the 1990s, the Cieng Reng settlement at Makot grew considerably. By 1998, Makot
village was the biggest Cieng Reng settlement in Ethiopia. Over time, the Cieng Reng
managed to create links with the neighbouring Anywaa communities and were able
to obtain access to riverine land through inter-ethnic marriages, exchanges of gifts and
purchase of land. Towards that end, Kong himself established alliances by marrying
extensively, not only with local Anywaa, but also among the various sections of the
long-time Nuer residents.15 As a result, the Cieng Reng in Makot village became very

15. Kong married sixteen wives from the various Nuer clans, the Anywaa and the Dinka.
 In the Riverine Lands 89

prosperous, combining their pastoral economy with increased rain-fed and flood-
retreat cultivation, as well as developing new market outlets in Itang and Gambella
towns. In 1998, the population of Makot village was estimated at three thousand.
Initially the local Anywaa did not feel the Nuer presence as a threat because it did not
result in any Anywaa displacement. Although Makot village falls within a tradition-
ally Anywaa territory, the Cieng Reng established their settlement by clearing a forest.
In addition, the local Anywaa benefited from the Cieng Reng Nuer in terms of the
flow of cattle wealth (bridewealth) as many Cieng Reng married into the Anywaa. For
these Anywaa, economic considerations took precedence over symbolic or political
concerns. The wealth and demographic expansion of the Cieng Reng in Makot village
attracted, instead, the attention of the long-time Nuer residents of the area. For this
group of Nuer (the Thiang), the new Cieng Reng settlement at Makot village meant
competition for a vital natural resource, namely, riverine land. The problem was com-
pounded when a group of land-hungry Thiang, the Cieng Dung, put in a claim for
the areas now settled by the Cieng Reng by reactivating their older alliances with the
neighbouring Anywaa.
The second set of actors in the conflict was therefore the Thiang Nuer. The
Thiang felt threatened by the Nuer migrants to Anywaa areas that they considered
as their hinterland. It is against this background that the Cieng Dung became
extremely hostile to the Cieng Reng, whom they envied because of the latter’s cattle
wealth and access to the Anywaa’s riverine land. In the resource competition between
the Cieng Reng and the Cieng Dung, the Anywaa supported the Cieng Dung on the
basis of their longer familiarity with the Thiang and their desire to buffer the steadily
growing Cieng Reng community at Makot village. Having outmanoeuvred the
Cieng Reng, the Cieng Dung adopted an exclusionary political discourse in their
quarrel with the Cieng Reng. Like the Anywaa elites in the regional power game, the
Cieng Dung labelled the Cieng Reng ‘foreigners/Sudanese’. Authorized by their
demotion of the Cieng Reng to noncitizens, groups of Thiang and Anywaa raided
Cieng Reng cattle. In this local pattern of alliance, the Cieng Reng found themselves
increasingly vulnerable, more so because they had faced an increasingly hostile local
administration. The Itang District Council was dominated by the Anywaa power
elites, who were apprehensive about the demographic growth of the Cieng Reng that
ultimately threatened the Anywaa’s political domination of the district in particular
and the region at large.
The third set of actors was thus the aforementioned Anywaa power elites. The
Cieng Reng settlement at Makot dates back to 1984 but has been politicized only
since the introduction of ethnic federalism in Ethiopia in 1991. On the basis of the
new national political structure, Anywaa and Nuer political parties were established,
competing fiercely for political power in the newly created Gambella Regional State.
Itang district was one of the most politically contentious districts. The number of
Anywaa and Nuer in the district were about the same, but the distribution of the
kebele, the smallest unit of government administration, was disproportionate.
Because they could claim plausibly to be the original inhabitants of the area, the
Anywaa were allocated fourteen of the district’s twenty kebeles. They insisted fer-
vently that Itang district was their territory and considered the Nuer in general to be
90 Playing Different Games

latecomers. As a pre-emptive move, the Anywaa politicians in Itang district took

various administrative measures to discourage the growth of the Cieng Reng popu-
lation in Makot village. Above all, they defined them as refugees, thus denying them
access to administrative justice. They also discouraged the local Anywaa from enter-
ing into social and economic exchanges with the Cieng Reng, thereby denying them
access to the riverine land that the Cieng Reng had managed to obtain through
various forms of social networking and payment. The Cieng Reng were even prohib-
ited from using the Baro River for drinking water, and attempts were made to relo-
cate them to a refugee camp. In reaction, the Cieng Reng developed two
counter-strategies. First, they sought new local allies, in particular forging military
alliances with other groups of recent Nuer immigrants such as the Cieng Nyajani. It
was against this changing background of local alliances that the Cieng Reng swiftly
became involved in the conflict on the side of the Cieng Nyajani when the latter
fought the Anywaa because of the divorce case. Second, and with wider repercus-
sions, the Cieng Reng made contact with the Nuer power elites, who readily
responded to their plight and showed interest in their case regarding issues related to
the demographic politics in the regional power game.
The Nuer power elites, vying for political power with their Anywaa counterparts,
constituted the fourth set of actors. These power elites in Itang district and in the
regional council took the Cieng Reng case seriously. A number of Nuer officials
visited Makot village and lobbied for the recognition of Makot as a kebele. The final
link between the Cieng Reng villages and the Nuer power elites was found in the
political discourse of the Anywaa, whose own power elites used the same discourse
vis-à-vis the Cieng Reng and the Nuer power elites in the politics of exclusion: as
with the Cieng Reng, the Nuer power elites were defined as noncitizens. Virtually all
of the Nuer officials and civil servants had been educated in the southern Sudanese
refugee camps in Gambella. It was the involvement of these power elites that turned
the resource conflict into an ethnic conflict. Encouraged by the Nuer officials, the
Cieng Reng demanded six kebeles ‘in proportion’ to their demographic size. Having
a kebele thus became both a symbol for citizenship and a code word for resource enti-
tlement. At the height of the conflict, Kong Diu and other ‘ringleaders’ were impris-
oned for a year. Attempts were made on Kong’s life. After the issue became
deadlocked at the district and regional levels, it was elevated to the federal level.
When the Anywaa-dominated regional council produced a document to prove the
‘noncitizenship’ status of the Cieng Reng, Kong Diu travelled from Makot village to
Addis Ababa, the national capital, to appeal to the federal government. The federal
authorities compromised by granting the Cieng Reng a ‘residence permit’. Kong was
told to return to Makot without further ado.
From this case study, the resource dimension of the Anywaa–Nuer conflict is
evident. The Cieng Reng’s politics of recognition centres on ensuring access to the
riverine lands and seeking legal protection for their property. In the economy of rel-
ative scarcity, however, some groups of Nuer such as the Thiang are strategically posi-
tioned to benefit from inter-ethnic peace. As described in the case study, the
competition between the Cieng Reng Nuer and the Thiang Nuer over the riverine
lands was one of the underlying reasons for the conflict. The non-ethnic roots of the
 In the Riverine Lands 91

resource conflict in Itang became even more evident during the 2002 conflict
between the Anywaa and the Nuer power elites in Gambella town and in how these
same roots acted themselves out in the resource conflicts in the villages of Itang dis-
trict. In the struggle for power at the regional level between the largely Thiang power
elites and the Anywaa, the pattern of alliances at the local level once again changed.
Until 2002, the Thiang had dominated Nuer politics in the GPNRS. Keen on repair-
ing the strained relations with the Anywaa, the Cieng Reng kept aloof from the con-
flict. In fact, the Cieng Reng and the Cieng Nyajani gradually disassociated
themselves from the Thiang and made a separate peace agreement with their Anywaa
neighbours in return for their renewed access to the riverine lands. This brought
about a new round of hostility between the Cieng Reng and the Thiang that erupted
into a major conflict in February 2004.

Divergent Schemes of Interpretation as a Complicating Factor

It is true that the Nuer covet the Anywaa riverine lands along the tributaries of the
Sobat, but access to and control over these lands can be sought by various strategies.
Mobilizing ethnic identity is just one possibility. What transforms the resource con-
flicts into ethnic conflicts is the way in which politics is organized in post-1991
Gambella as ethnopolitics and the contrast between the two identity formations. In
the following I discuss how the divergent modes of ethnic identity formation of the
Anywaa and the Nuer are acted out in the resource conflicts.
Resource conflicts assume an ethnic dimension partly through the use of a similar
strategy of accessing resources by individual Nuer. At this level, ‘Nuerness’ emerges as
a cultural model. These strategies of resource extraction are the instrumental use of
inter-ethnic marriages and the manipulation of kinship ties. The Nuer also employ a
similar cultural mode of legitimizing resource claims. This gives the impression of there
being an organized pan-Nuer economic interest. The Nuer instrumental use of inter-
ethnic marriages is eased by the gap between the Anywaa identity discourse and its
social practice. The Anywaa identity discourse ideally prescribes ethnic endogamy.
Inter-ethnic marriage is discouraged for fear of corrupting ‘ethnic purity’ (luo). In
reality, however, pragmatism occasionally compromises ideology, the first such com-
promise being made by Anywaa leaders who had a political incentive to make a mili-
tary alliance with groups of Nuer in the inter-village fights. In so doing, the kwaari and
the nyiye anticipate the flow of cattle wealth to their treasury. This compromise is called
bilo (exchange marriage). Through bilo, a kwaaro gives one of his daughters to a Nuer
in return for cattle, forming a basis for interpersonal friendship or a Cieng/village-level
alliance between the two. The commoners followed the kwaaro model and entered into
socio-economic exchanges with the Nuer. As marrying a Nuer girl is hardly possible for
the Anywaa because it requires a bridewealth payment of twenty-five cows, over a
period of time the Anywaa have been reduced to ‘wife-givers’. Asymmetrical marriages
at the individual level produce a pattern, for each individual Nuer shares the same cul-
tural model. Where a Nuer cieng and an Anywaa village are intertwined in kinship rela-
tions and economic ties, the Anywaa tend to be drawn into the Nuer cultural orbit.
There are also cases when Anywaa (foreigners in general) are helped to marry a Nuer
girl in order to hasten the process of assimilation. This often involves economic and
92 Playing Different Games

social incentives: a reduced bridewealth payment and the provision of food for needy
foreigners. These acts of Nuer ‘generosity’ are attractive particularly to the poorest
section of Anywaa society. In the long term, inter-ethnic marriages have brought about
the economic and cultural expansion of the Nuer (Dereje 2005).
In most cases a Nuer man marries an Anywaa woman. This is initially beneficial
to both parties. For the Nuer, it is cheaper to marry an Anywaa as the bride wealth
payment is lower, and for the agrarian Anywaa, the marriage ensures the flow of
cattle wealth. The Nuer anticipate additional gains from such exchanges: marriage
ties are used as a legitimizing discourse to establish settlements in Anywaa territories.
These settlements gradually serve as a nucleus for more immigrants; over time, Nuer
immigrants outnumber the Anywaa, who are then left with the option of either
joining the Nuer kinship and political structures or leaving their villages in order to
maintain their ethnic identity. The intentionality and instrumental nature of Nuer
marriages operate at the individual level. Individuals apply the model as much at the
intra-ethnic level as they do inter-ethnically. The birth and growth of Nuer settle-
ments in Makot and Ochom amidst the Anywaa territories in Itang and Gambella
districts is related to these social dynamics. Kong married not only a Dinka and an
Anywaa but also many women of the neighbouring groups of Nuer. Riek Tuany, the
founder of Ochom subvillage in the Anywaa village of Pinykew, married an Anywaa
‘in order to create social ties with the neighbouring Anywaa’, as he explicitly put it.
I arrived in Ochom subvillage shortly after Riek’s Anywaa wife had left him because
of a personal disagreement. I talked to Buk, Riek’s Nuer wife, about what she felt
about Riek’s Anywaa wife. She answered me in strongly pragmatic terms:

Nyapini is a very bad person. She is often drunk and even insults Riek. A
Nuer woman would never do that. I also had problems with her. But it is
good that Riek married her. Had it not been for Nyapini, Riek would not
have become the dil of Ochom. Nyapini’s father also gave us a plot of land to
grow tobacco.16

Similar intentionality and pragmatism are attributed to inter-ethnic marriages by a

Nuer elder from Akobo district.

When you marry or give your daughter to an Anywaa, you know that they
will all belong to you. The idea is to be a big cieng in order to be able to
defend yourself. No cieng wants to be small. It strives to be bigger. If these
two boys [my Nuer research assistants] were my children, when they grow up
they would compete to form their own cieng and become bigger. They may
even quarrel. The smallest cieng might be defeated. So you recruit followers
from different places with different means. You foresee in the future that a
problem might happen somehow, so that you make sure that they will remain
loyal to you. (Thiang Louny, Gaat-Jok Nuer elder, Ochom village, July 2000)

16. Tobacco grows on the most fertile part of the bank of the Baro River. It is also the most lucra-
tive cash crop.
 In the Riverine Lands 93

The Anywaa conflate the economic interests of individual Nuers with the idea of an
‘ethnic conspiracy’. This can be understood through the diverging schemes of inter-
pretation that the Anywaa and the Nuer employ to define the interactional situation.
Here, I draw on Robert Merton’s concepts of motive and function. Merton (1968)
makes the distinction between motive and function while establishing the link
between motive and action. He defines motive as ‘the subjective aim-in-view’ and
function as ‘the objective consequence of action’ (1968: 51). He then goes on to
identify two kinds of function: manifest and latent. Manifest functions are ‘objective
consequences that contribute to the adjustment or adoption of the system that are
intended and recognized by participants in the system’, whereas latent functions are
‘neither intended nor recognized, are unanticipated consequences of action which are
functional for a designated system’ (ibid.: 51). Applying Merton’s concepts to
Anywaa–Nuer relations, one might say that the motive of a Nuer man who marries
an Anywaa girl or gives his daughter to an Anywaa man, without demanding that
bridewealth or the full amount of bridewealth be paid, is to secure economic bene-
fits and physical safety. This motive corresponds to a manifest function, insofar as it
is intended and recognized by the actors involved (note the statements by Riek, Buk
and Thiang Louny above). Merton’s notion of function applies, in this case, to the
individual level. Simultaneously, however, the Anywaa tend to view the same act, an
individual inter-ethnic marriage, at the level of inter-groups relations, interpreting it
as an expression of Nuer expansion. In Merton’s terms, a latent function is mistaken
for the motive.
In Merton’s model, the fine distinctions between motives, manifest functions and
latent functions does not, however, encompass a further aspect of social reality: the
grey area in between them. Beyond the individual level, inter-ethnic marriages and
the recruitment of outsiders do have a manifest function for members of Nuer local
communities: creating a bigger and stronger cieng, as Thiang Louny put it.
According to Merton, therefore, this is not latent because it is intended and recog-
nized. It is latent only where it leads to the displacement of Anywaa and to ethnic
conversion, i.e., to Anywaa becoming Nuer. It is common to hear a Nuer arguing as
follows: ‘It is the fault of the Anywaa to abandon their culture and adopt the Nuer
culture or leave their villages when a Nuer comes. We are not forcing them to believe
in Kuoth, speak our language, make gar or become cattle keepers. This happens
because they do not have confidence in their culture and their culture disintegrates
when it makes contact with other people’s culture’. Given the increased emphasis on
identity politics under the new Ethiopian constitution, contemporary Nuer seek to
convert the ‘latent’ to the ‘manifest’; for even if inter-ethnic marriages have unin-
tended consequences – that is, even if they are not expressions of a ‘conspiracy’ – it
is recognized that they contribute to the expansion of the Nuer. The Nuer openly
acknowledge their expansion and take pride in it. The growing interest of the con-
temporary Nuer power elite in demography is also likely to close the gap between the
manifest and the latent, as their active involvement in the Cieng Reng’s quest for
kebele in the case study above shows.
Nuer microprocesses, on the other hand, are made intelligible by an Anywaa
macroscheme of interpretation. The demographic and political implications of
94 Playing Different Games

micro-economic practices between individual Anywaa women and Nuer men led to
the ethnic framing of interpersonal and intergroup relations. It is this microlevel
dynamic that has eluded the Anywaa political actors who operate at the ethnic level.
Despite their attempts and occasional campaigns to stop the socio-economic
exchanges between the ordinary men and women of Anywaa and Nuer, business
often continues as usual at the local level due to the gap between individual and
group interest. Group interest is not always served by the pragmatic flow of the indi-
vidual’s everyday concerns.
Nuerness emerges not only as a strategy for gaining access to resources but also a
key element in post-hoc modes of legitimizing claims. When the Anywaa say ‘the
land is ours’, the Nuer reply ‘there is enough land for us all’ – a discourse that res-
onates with the colonial discourse of ‘waste land’, i.e., land not ‘properly’ utilized by
indigenous peoples and, therefore, inviting exploitation by settler capitalists (cf.
Lester 2002: 30–40). The Nuer claim is, however, embedded in their notion of prop-
erty, itself related to their contrasting ethnic identity formation. Indeed, the Nuer
have a sense of individual and collective property, but property rights are tightly
defined. For instance, effective occupation is needed to make a claim on land.
Anything that is not possessed either pertains to Kuoth (God) or is there for the
taking. This obviously has potential for conflict. For the Anywaa, the land is as much
an economic as a symbolic resource. For the Nuer, areas that are not actually settled
by the Anywaa are considered as part of ‘the economy of the commons’.
In the 1970s, there were only a few pockets of Nuer settlements between Jikaw
and Itang districts. By the end of the 1990s, Nuer settlements along the banks of the
Baro River extended as far as the Anywaa villages of Pinykew and Abol in Gambella
district. The Nuer have also established a very large settlement in Gambella town,
popularly known as Newland. From Newland, the Nuer have continued their expan-
sion to the east in the Bonga area, in close proximity to Tier Agak, the last Anywaa
settlement at the foot of the highlands. This large-scale territorial expansion of the
Nuer has become a source of concern for the Anywaa. Facing continual Nuer terri-
torial and cultural encroachments, the Anywaa have contested Nuer hegemony both
discursively and more directly.
In the aforementioned discussion, it has been shown that access to and control
over natural resources play important roles in the Anywaa–Nuer conflict.
Nevertheless the link between resource competition and ethnic conflict is mediated
by two intervening variables: first, the contrasting modes of ethnic identity forma-
tion and the divergent schemes of interpretations they have generated; and, second,
a political structure that puts a premium on ethnic identity as unit of political action.
On one hand, the Nuer’s strictly economic interest in land clashes with the symbolic
significance that the Anywaa attribute to it; and, on the other hand, within the
framework of ethnic federalism, resource conflicts tend to be interpreted as ethnic
conflicts. In the following chapter the conflict between the Anywaa and the Nuer is
examined in the cultural domain in order to show how the contrasting ethnic iden-
tity formations are implicated in the conflict situation and how they interact with the
resource and power variables.
 Chapter 5

The Cultural Contestation

In the previous chapter we have seen how contrasting formations of ethnic identity
mediate between resource competition, on one hand, and ethnic conflict, on the
other. In this chapter we discuss the ways in which these contrasting formations
themselves may cause ethnic conflict, particularly in processes of ethnic conversion.
The constructivist ethnic identity formation of the Nuer is inherently expansionist.
It not only allows ethnic membership for outsiders but it also actively encourages and
supports the process of becoming Nuer, for example, by providing material incen-
tives. Ethnic conversion is especially common in those situations in which the Nuer
enjoy military and economic advantages, in comparison with their neighbours. Thus,
in some areas of contact with the Nuer, the Anywaa are unable to sustain their pri-
mordial imagination of ethnic identity and, as a result, they undergo a process of
ethnic conversion. This has embittered Anywaa in other areas, who persist on
defending their way of life; and it has generated a cultural contestation, in which
both the Anywaa and the Nuer reflect on their contrasting modes of ethnic identity
formation and promote their respective worldview, while demonizing and criminal-
izing the other’s way of life.
The Anywaa have responded to the territorial and cultural encroachments of the
Nuer in various forms: avoidance, ethnic conversion and resistance. Some Anywaa
have coped with the situation in the first manner, by leaving their villages and estab-
lishing new villages with the old names. Ideally an Anywaa would prefer to remain
in his natal village, but living with the Nuer or having the Nuer as nearby neighbours
is felt to constrain their way of life. For these Anywaa, animal husbandry is not desir-
able – ‘cattle do not mix with maize’. Others recognize the existential needs of the
Nuer and tolerate them on moral grounds. This is possible as long as there is enough
vacant land for resettlement. What this category of Anywaa does not like, or what
angers them most, is when these ‘acts of generosity’ are taken by the Nuer as a state-
ment of weakness, and thereby encourage further encroachment.
Another section of Anywaa society has responded differently by conforming to
the precept, ‘If you cannot beat them, join them’. Anywaa assimilation in Nuer
society is facilitated by the Nuer identity discourse, which actively encourages
becoming Nuer and only requires cultural competence. For this category of Anywaa,
becoming Nuer implies an enhanced way of life, as this also involves conversion to a
different means of livelihood, an agropastoral economy. Unlike other instances of
identity switching that can be reversed (Leach 1954; Barth 1969), becoming Nuer
involves a process of ethnic conversion and a lifelong commitment to the Nuer mode
of identification such as through gar, the physical manifestation of male Nuer iden-
tity discussed in Chapter 3.
96 Playing Different Games

Processes of Ethnic Conversion

In some areas Anywaa–Nuer socio-economic exchanges have resulted in identity
change. Nowhere is this process of ethnic conversion epitomized more than in the dis-
trict of Akobo. Originally, Akobo district was Anywaa territory, but today only a few
pockets of Anywaa settlements have survived. Even these pockets are being increas-
ingly drawn into the Nuer cultural orbit. The three Anywaa villages of Jingmir,
Burbey and Pone, for instance, have recently joined the three primary Gaat-Jok Nuer
tribal sections of Cieng Wang, Cieng Nyanjiok and Cieng Kuek, respectively.
The assimilation process started in the 1940s when the Gaat-Jok Nuer from the
Nasser area made contact with the Ciro Anywaa along the Akobo River during the
dry season. As the number of Nuer coming to the Akobo area increased, many
Anywaa started moving out, while others resisted, mainly in the form of surprise
attacks on those Nuer who went to Anywaa villages to buy grain. Repeated attacks
resulted in serious battles between the Gaat-Jok Nuer and the Akobo Anywaa, as a
result of which many Anywaa were displaced and resettled in the neighbouring Jor
district and others concentrated in the district’s capital, Tiergol. The kwaari of the
three Anywaa villages of Jingmir, Burbey and Pone decided not to get involved and
remained in Akobo, where they were immune from attack by the Nuer. This group
of Anywaa intermarried with the Nuer, initially Nuer men marrying Anywaa
women, but later on Anywaa men were also given Nuer girls, either freely or for
reduced bridewealth in order to strengthen the social ties. As they were cut off from
other groups of Anywaa, the Burbey, Jingmir and Pone villagers were drawn into the
Nuer cultural orbit. The following two examples illustrate the process of ethnic con-
version, one from Pone (Tut Obang) and the other from Jingmir Anywaa (Ochiere
Oruach) villages.
Tut belongs to the first generation of ‘Nuerized’ Anywaa from Pone village. Tut’s
father, Obang, and Dong, a Cieng Kuek Nuer from a neighbouring village, were
friends. Dong offered his daughter to Obang as his wife, with whom Obang subse-
quently had five children. Dong, like so many other Nuer in Pone village, followed
his daughter and lived with Obang’s family. In the company of his Nuer in-laws,
Obang became competent in Nuer culture, a competence that became even higher
among his children. Obang now speaks perfect Nuer and has initiated (gar) his sons,
who all bear Nuer names. Tut, the most successful of Obang’s sons, owns the biggest
cattle byre in Pone village and has married two Nuer wives. Pone is still recognized
as an Anywaa kebele in the now Nuer-dominated Akobo, but for all practical pur-
poses social life in these pockets of Anywaa settlements is effectively ‘Nuerized’. By
August 2000, there were seven hundred and fifty families in Pone village, out of
which six hundred and fifty were Nuer.
An even more dramatic process of ethnic conversion has occurred among the
Jingmir Anywaa: a change of ethnic identity among individuals with only Anywaa
parentage. Take the example of Ochiere Oruach. Ochiere, a Jingmir Anywaa, is an
Anywaa on both his father’s and mother’s sides. Like his counterpart Obang in Pone
village, Ochiere’s father, Oruach, brought him up in a Nuer cultural milieu. Ochiere
changed his Anywaa name to a Nuer name, Wako, went through the Nuer male ini-
tiation, although he was not even half Nuer, and married a Nuer. Wako, as Ochiere
 The Cultural Contestation 97

prefers to be called, narrated his personal experiences to me, which sheds light on the
incentives and ambiguity that surround the assimilation process:

The first Nuer who made contact with the Jingmir Anywaa was Wangkech.
He was a Gaat-Jok. Wangkech was a friend of a Jingmir Anywaa called
Pokunya. That was during the time of the Litgach age group. Wangkech’s son,
Lual, married the daughter of Pokunya called Dul. The children of Lual and
Dul formed the Cieng Pokunya lineage. When the kwaaro of Jingmir died,
the Burguany Anywaa wanted to take the kwaaro of Jingmir. That was the
problem between the Jingmir and the Burguany. The daughter of Lual’s
brother, Nyajok, married Wal. They had three children: Diu, Buk and Tar.
They are called Cieng Nyajok. The Cieng Nyajok were allowed by the
Jingmir Anywaa to permanently settle in their village because of the marriage
between Lual and Dul. The Cieng Pokunya joined the Cieng Nyajok. The
Burguany Anywaa left the area when a conflict broke out between the Luo
Nuer and the Gaat-Jok Nuer in 1992. We, the Jingmir, remained because of
our relationship with the Cieng Nyajok. As a result, we were cut off from the
Burguany and other groups of Anywaa. Some Jingmir people tried to keep
their Anywaa identity by keeping dimui and marrying Anywaa. Two of my
sisters were married to Anywaa. My father thought that one day we might go
to the Anywaa areas. He hoped that all of his three sons would marry Anywaa
girls. One of my brothers married an Anywaa and went to Malakal and the
other two of us married Nuer. The first Anywaa generation after the contact
with the Nuer did not have gar. Gar started with my generation. I am the
only one in the family who has gar. I asked my father whether we would go
to the Anywaa areas or not. When he said no, I was initiated. It was a bit late
for me. I was already twenty. I wanted gar because I could not marry other-
wise. All my friends were Nuer. One cannot marry a Nuer girl in Akobo if he
does not have gar. When I went to Utalo village to visit my Anywaa relatives,
they considered me as Nuer and they said I should go back to the Nuer,
which I did. (Wako Oruach, Pinyudo refugee camp, 4 May 2001)

The genealogical location of the Jingmir Anywaa within the Gaat-Jok Nuer is shown
in Figure 5.1.
The Jingmir Anywaa in general and the Cieng Pokunya in particular (to which
Wako belongs) identify themselves as Nuer. When I asked Wako how he identifies
himself he replied to me in evaluative terms:

Those Anywaa who became Nuer would not leave the Nuer because the con-
ditions of living are good. If the Nuer see that you are a good person, they give
their girls freely. You do not have to cultivate for anybody if you want to marry
as a poor person. That is the case among the Anywaa. Those who do not have
dimui cultivate the land of the kwaaro in order to be able to marry. If you
offend somebody, the Nuer will collect cows for you so that you can pay com-
pensation. The Nuer give you food if you go to them. The Anywaa do not do
98 Playing Different Games

Figure 5.1 Jingmir Anywaa within the Gaat-Guang Nuer genealogical structure (source: author’s notes
compiled from Wako Orauch’s narratives)

that. My father compared the two people and decided to stay with the Nuer.
The Anywaa also have a culture of beating (utak). The Nuer marry with cattle,
but the Anywaa marry with dimui. But dimui is not food. The Nuer marry
with cattle and cattle bring you food. Now the Anywaa [in Akobo], too, have
realized how important cattle are. Even if the Anywaa become rich, we will
remain Nuer because we married Nuer girls and have gar.

Wako is not alone. Similar stories can be heard from other Nuer of Anywaa origin,
even among the educated and among those from urban settings, where the ethnop-
olitics in Ethiopia has generated a greater degree of reflection on origins. The follow-
ing statement by an Anywaa with a mixed parentage who nevertheless identifies with
the Nuer is a case in point:

My mother is an Anywaa from Pukumu village. She married a Nuer from

Luel village. He met my mother when they came to Pinymou during the dry
season. Both my Nuer father and grandfather died when I was a child. I grew
up with my mother’s relatives in Pinymou [Anywaa] village. I speak both
Anywaa and Nuer, but I identify myself as a Nuer. I do not have gar but this
 The Cultural Contestation 99

is not because of my Anywaa background but because my father is from Luel.

Luel village is a Christian village [Seventh-day Adventist]. Many Nuer from
Luel do not have gar. I like the Nuer because they are richer and stronger. As
long as the Anywaa continue to be poor, the children of mixed parents prefer
to be Nuer. The Nuer are so rich that they can even buy RPGs [rocket-pro-
pelled grenades]. An RPG costs fifty cattle. It can destroy an aeroplane.
(Uchan Omot, elementary school teacher, Birhaneselam village, May 2001)

Uchan is half Anywaa and grew up in an Anywaa cultural milieu, and yet he identifies
himself as Nuer. To him, one of the main attractions of being Nuer is the economic
status it confers, the scale of which is given in terms of the affordability of RPGs.1
Examples of ethnic conversion abound in comparative ethnographic literature:
Fur becoming Baggara in the Sudan (Haaland 1969), Pathans becoming Baluch in
Pakistan (Barth 1969), Kachins becoming Shan in Burma (Leach 1954) and Gunga
becoming Hausa in Nigeria (Salamone 1974). This process of ethnic conversion has
recently been disputed by Gil-White (1999), who argues that most of these ethno-
graphic examples represent signalling, not switching, of ethnic identity. Gil-White
instead proposed an ethnobiological model which he calls ETAM (Ethnic
Transmission and Acquisition Model), a model that attributes a primordialist con-
ception of an ethnic group to ethnic actors. Reworking Barth’s data on the Swat
Pathans, Gil-White argues: ‘In Swat, considerable “playing” with the ethnic labels is
allowed, so those who wish to articulate with particular networks may use the labels
that signal the expectations associated with them. But if ethnic actors are in general
primordialists, and also essentialists, that is, one’s ethnicity implies an inalienable
“essence”, then I would expect the Barthian flexible signalling system to be the excep-
tion, and to find that in most times and places, one cannot simply “grab” a new
ethnic label and begin interfacing with another community as a full member’ (Gil-
White 1999: 812).
Doubting the efficacy of the Nuer assimilation system as described by Evans-
Pritchard, Gil-White further noted that ‘Dinka absorbed by the Nuer are not “real
Nuer” – they have a special name: Jang Nuer … [likewise] Nuer who marry into the
Dinka ethnic are not considered “real Dinka” either and are called Nuer-da, “our
Nuer”’ (ibid.: 811). Certainly, ethnic conversion occurs transgenerationally and ideas
of origin might survive for a while. Nevertheless, Gil-White should have taken the
Nuer as yet another ‘exceptional case’, which, I suspect, together with many other
assimilationist societies, undermines his generalized primordial thesis. None of the
Nuer with Anywaa or Dinka backgrounds to whom I spoke showed any interest in
a hybrid identity or resented assimilation. Nor are they categorized by the Nuer as
not ‘real Nuer’. The Nuer refer to the Jingmir Anywaa as Bär Jingmir. Unlike Gil-
White’s reference to the Nuer idea of ethnic purity, however, Bär Jingmir are not
putative Nuer, but ‘Nuer under construction’. By calling them Bär Jingmir, the Nuer
are making a distinction between ‘kinds’ of Anywaa, the others being generally

1. Weapons such as rocket-propelled grenades (RPGs) are acquired by Nuer villagers from the
various rebel groups in southern Sudan.
100 Playing Different Games

referred to as luuch naath (murderous people). Although they are still referred to as
Bär Jingmir, they are also addressed as Cieng Nyanjiok or Cieng Pokunya. The prefix
‘Bär’, the Nuer name for the Anywaa, is not used here in a derogatory sense.
It is rather the gradual but complete ethnic inclusion, such as in the case of Wako
becoming Nuer, that is discursively negated by the contemporary Anywaa, who are
engaged in the struggle for cultural identity. The one-way process of ethnic conversion
undermines the Anywaa’s status claim as ‘better than’ the Nuer, the basis of this being a
local discourse on civilization that refers to the ‘superiority’ of the agrarian life and the
prestige derived from their political organization, an organization that produced nobles
and headmen. The Anywaa contrast their ‘orderly’ political system with the ‘chaotic’
social and political organizations of the Nuer. The Nuer, on the other hand, take pride in
their integrationist social system. Unlike other assimilationist systems, Nuer assimilation
has not produced ‘internal others’.2 Neither an assimilation gap nor elements of class are
observable. The Nuer do not make status distinction between the internal strangers (rul)
and the ethnic strangers (jaang). Ethnic strangers could be either adopted into a domi-
nant lineage, or attached to it through affinal or matrilateral links, just like the rul are.
Contemporary Nuer are aware of their diverse ethnic backgrounds and for that
reason they are keen on emphasizing the inclusivity of ‘Nuerness’. Assimilation
among the Nilotes might appear to be relatively easy, given their shared origins and
phenotypical similarities; but assimilation to the Nuer way of life occurs even across
‘racial’ boundaries. In the 1980s, the Nuer incorporated some Highlanders (the ‘red’
people) who were forcibly resettled in Gambella as part of the government’s policy of
combating the recurrent problem of draught and famine in the highlands. The
appalling living conditions and the high degree of coercion by the Ethiopian govern-
ment that characterized the resettlement programme caused many such Highlanders
to flee to the Sudan. Some of these Highlanders, particularly women and children,
were captured by the Nuer and, over time, effectively became Nuer.3

Anywaa Discursive Resistance to Nuer Cultural Hegemony

For those Anywaa who are outside the Nuer cultural orbit there is strong bitterness
towards that which Hutchinson (2000: 9) called ‘the sticky grasp of the Nuer on their
neighbours’. The Anywaa are well aware of the assimilation process and lament for
their people ‘lost’ to the Nuer. Commenting on this social fact, an Anywaa elder once
said to me, ‘All short Nuer are Anywaa and all tall Nuer are Dinka’. The assimilation
process undermines the Anywaa status claim that their way of life is as good as, or even
better than, their neighbours. This disjunction between the Anywaa self-image and
the realities of inter-ethnic relations has magnified their ‘fear of extinction’, an anxiety
that increasingly develops into the form of a political project of containment.

2. Even the much-acclaimed Oromo assimilation contains within itself ‘internal others’, such as
the Wata and the Gabra among the Boran Oromo (Fekadu Adugna, personal communication)
or the Omotic and the Nilo Saharans among the Wellega Oromo (Tesema 2006).
3. I met some of these assimilated Highlanders in Jikaw district. Many of them showed no inter-
est in me as a Highlander; while others ran off so as not to be taken away from the Nuer.
 The Cultural Contestation 101

The Anywaa resistance to the contemporary territorial and cultural expansions of

the Nuer has taken different forms. The first is a discursive resistance. This is
expressed by ‘demonizing’ the Nuer cultural form. While commenting on the nature
of ‘being Nuer’, the Anywaa often start by searching for ‘the hidden agenda’:

The problem between the Anywaa and the Nuer is that the Nuer take other
people’s land and steal their children. I have also heard that part of the
Anywaa’s land belonged to other people but we do not know who these
people are and they are not around these days. But the Nuer found us when
they came to Akobo. The people who were there before the Anywaa were
encouraged to join their relatives elsewhere instead of being assimilated into
the Anywaa. But the Nuer take other people. God gave each people a lan-
guage, a land and a system. The Anywaa have a system called kew [bound-
ary]. There is kew between neighbours, between brothers and even between
father and son. But the Nuer just move and take other people’s land. The
Nuer have no system. The Nuer move like the Felata [Fulbe] and the
Kibabish. The Nuer, the Felata and the Kibabish have no system. The
problem with the Nuer is their behaviour. Nuer for the Anywaa means some-
thing which is not good. Maybe God created them like that, just like He
created black and white people. In tribes like the Murle and the Nuer, if you
steal and bring something, you are considered brave and get respect, but for
the Anywaa such things are bad. We call thieves kuruach. A kuruach will not
get a wife and he will be cursed. (Pastor James, head of Anywaa congregation
in Akobo, Presbyterian Church of Sudan, Khartoum, 17 August 2002)

A similar interpretive scheme was employed by an Anywaa church leader in

Gambella: ‘How is it possible that I give you my daughter or even my wife free if it
were not for ulterior motives? That is what the Nuer are doing. This is part of the
Nuer master plan to eradicate the Anywaa’.4 Interestingly there seems to be little dis-
agreement between the Anywaa and the Nuer regarding the form that Nuer culture
takes. The Anywaa recognize, sometimes to their bewilderment, how ‘democratic’
and powerful the Nuer are. The difference lies in what it is meant for:

The problem with the Nuer is too much democracy. Anybody can go to a
Nuer village and live with them, and become Nuer. That is what happened
with the Jingmir Anywaa. Jingmir were originally Anywaa but now, kalas, they
have already become Nuer. Some even have goro [gar]. Their Anywaa culture
is gone and they behave like the Nuer. They know that they are Anywaa and
still speak Anywaa, but they prefer to speak in Nuer. They dance like the Nuer.
We are fighting with the Nuer in Akobo, but the Jingmir support the Nuer. If
I go now to Bentiue, I will get a wife free. The Anywaa look down on those

4. An extract from an informal talk with Omot Agwa, ex-president of the Eastern Gambella
Bethel Synod, Gambella town, 12 February 2002.
102 Playing Different Games

who leave their villages. Only those people who are not good or have problems
leave their villages. That is because the Nuer have no system and that is why
they do not respect systems where these exist. We got our system from our
kings. We respect our kings; wives respect their husbands. Other people say
the Anywaa are selfish and do not want to live with others. That is not true.
We respect system because we do not want to create problems. The Anywaa
got their system from nyiye. Kings bring system, and you find kings in the
Bible as well. There were kings in Israel. Jesus came from the house of King
David. Haile Selasssie was called the King of Kings of Ethiopia, just like our
nyinya which means nyiya of the nyiye. (Pastor James)

Anywaa traditional leaders, particularly the nyiye, are expected to provide ideals of
social behaviour. They are considered the repository of the ideology of ethnic purity.5
Ideally the nyiye do not marry wives from other ethnic groups. In the nyiye origin
myth, Ucuudho, the divine ancestor of the nyiye, is represented as the ‘legal mentor’
of Anywaa society. The pre-Ucuudho period is viewed as a time marked by chaos,
when people took the law into their own hands:

Before Ucuudho, Cuai was the leader of the Anywaa. He did not govern the
people properly. In fact, when people asked his advice, he used to tell them to
solve it on their own. One day a person came to Cuai to complain about a neigh-
bour who troubled him. Cuai told him to ‘spear him in his stomach’. That person
went to his neighbour and speared the offender. But Cuai did not mean it liter-
ally. He rather wanted to say ‘take his cattle’. The family of the deceased in turn
speared one of Cuai’s sons. Anywaa law remained an eye for an eye until the
coming of Ucuudho. (nyiya Adongo Agada, Utalo village, Pochalla, 3 April 2001)

As the legal mentor of Anywaa society, Ucuudho is well expressed in the origin myth
of kingship. In this myth, Ucuudho, son of the river god, put an end to the peren-
nial dispute among Anywaa boys over a fish catch. According to Reh (1996), who
extensively documented the myth, Ucuudho suddenly appeared from the river
during one of these disputes and asked, ‘Who caught the head, and who caught the
tail?’ to which he added, ‘You holding the head, take your hand off it’. The person
holding the head took his hand off and the fish slipped away. When the person
holding the tail did so, the fish died. ‘The person who catches the head owns the fish
because the strength of a fish is in its head’. When the villagers heard about
Ucuudho’s judgement, they wanted to make him their leader. Ucuudho was ‘cap-
tured’ and brought to the village. He was given a wife. Ucuudho ‘went back to the
river’ after his wife became pregnant and gave birth to the first nyiya, Gilo. Gilo is
said to have continued his father’s judicious rule and helped the Anywaa construct a
new social order (Reh 1996: 498–503).

5. The nyiye do not always live up to expectations. One of the wives of the late nyiya Agada
Akwei, for instance, was a Nuer from whom he had a son.
 The Cultural Contestation 103

The political subtext of this myth is the legitimation of the power of the nobles as
the descendants of Ucuudho, who introduced the Anywaa to the basic principles of
social justice. Because of his ‘enlightenment’, Ucuudho’s relics became symbols of
Anywaa identity, sovereignty, cohesion and power. His necklace, ucuok, is one of the
royal emblems. The constant reference to ‘system’ in Pastor James’s narrative above
ultimately goes back to the social order established by Ucuudho. Ucuudho’s son, Gilo,
became the first nyiya and continued to govern the Anywaa in a more orderly way
than the ‘chaotic’ administration of Cuai. The contemporary Nuer are likened by the
Anywaa to the ‘primordial chaos’ of Cuai’s time. Nuer cultural hegemony is also con-
tested through scriptural references. Pastor James’s reference to the Jewish kings of the
Old Testament and to Emperor Haile Selasssie of Ethiopia, who legitimized his power
through a mythological connection with King Solomon, is a discursive link with what
Redfield (1955) called ‘great traditions’, whereas the Anywaa’s repeated references to
the Nuer’s lack of a ‘system’ reduce the latter to people of ‘small traditions’.
Other forms of Anywaa discursive resistance to Nuer territorial and cultural
expansions explain Nuer power by attributing intrinsic strength to its cultural form:

The Nuer are dangerous people. Even those Tigreans [resettled Highlanders]
who fled to the Sudan during the girgir [the turmoil during the regime
change in 1991] became Nuer and have gar. If you are smart or intelligent,
the Nuer want a son from you but they become the father. This is possible
among the Nuer. Look at the case of Joshua. He was originally Anywaa from
Akobo, but he became completely Nuer. His brother and relatives are still
Anywaa. The Nuer have the skill of deceiving people. (Okello Obang,
Anywaa NGO worker, Gambella town, 15 July 2000)

Okello is referring to the process of ethnic conversion among the Jingmir Anywaa in
Akobo district, which was described earlier. Specifically Okello refers to the family of
Lual. Lual is a Jingmir Anywaa, with two sons, Opal and Odol. While Odol retained
his Anywaa identity, Opal became Nuer and changed his name to Joshua. Growing
up in a Nuer cultural milieu, Opal was initiated and became fluent in the Nuer lan-
guage. Here is an interesting case of one family with two ethnic identities. This
process of ethnic conversion has produced conspicuous evidence supporting the
Anywaa discourse of ethnic extinction. For all practical purposes, Opal, renamed as
Joshua, identifies himself and is identified by others as Nuer. Fear of extinction, thus,
is not the mere rhetoric of political mobilization, but an overarching theme among
the ordinary Anywaa and their political leaders. The descriptions of the conflict sit-
uation between the Anywaa and the Nuer by two of the principal Anywaa political
actors in the 1990s bear this out:

The Anywaa and the Nuer conflict has always been there. It will continue in
the future. Whatever alliance existed, it was tactical when the Nuer were
numerically weak. I speak the Nuer language fluently but I will never let them
know that I speak their language. There are only two ways of solving our
problem. Either the Anywaa will cease to exist [when they abandon their lan-
104 Playing Different Games

guage and culture completely] like the Opo and other people, or the Anywaa
will become strong enough to resist the Nuer. The Nuer are born aggressors.
It is in their blood. The Anywaa will not go the Opo way because we like to
keep our culture. The only difference between the Anywaa political system and
the modern state is that the Anywaa kwaari do not pay salaries to their follow-
ers. Everything else is the same. We do not have any inferiority complex. Why
should we adopt other people’s culture? The Opo now speak Nuer and they
have started adopting Nuer names. (Opamo Obaya, former head of Bureau of
Education in Gambella Regional State, Uriru, Kenya, 27 August 2002)

The Nuer come as guests, very humble, and become the hosts [the leaders].
The Nuer come to a new place, they marry into that area, and the area
becomes Nuer. The Anywaa do not like such things. In the years to come
there will be no Anywaa left in the Gambella region. Nasser [Sudan] was ours,
but the Nuer took it. Jikaw and Akobo were ours, again the Nuer took them.
Wherever we go, they will follow us. They take our land, they take our rivers,
and they take our people. Now they want to take Gambella town. Where else
should we go? We should stop them doing so. The Nuer behave like that
because they think that they are many and it is their nature to be aggressive.
For us the term Nuer means something bad. (Abula Obang, former head of
curriculum department and founder of Gambella Peoples Democratic
Congress, Uriru, Kenya, 28 August 2002)

Opamo’s and Abula’s narratives reveal the tension between the two dominant motifs
in Anywaa identity politics: the status claim that the Anywaa are ‘better’ than the
Nuer and the fear of extinction because of Nuer expansion. These are vividly
expressed in Opamo’s ‘ban’ on the Nuer language that he speaks fluently, and the
constant reference to the Opo, a numerically small ethnic group in Gambella who
have also come under strong Nuer assimilationist pressure in recent times. Although
he bitterly resents it, Abula, like Pastor James, recognizes the assimilation process,
but admitting that to an external audience would be self-defeating. From Opamo’s
and Abula’s remarks we also learn about the subtleties of Nuer expansion, a social
process that evokes suspicion and hostility among the Nuer’s neighbours.
The Anywaa share this scheme of interpretation of Nuer expansionism with their
Lwoo ‘cousins’. Centuries of separation, war and displacement from Ethiopia and
southern Sudan, as well as civil war in northern Uganda, have driven various groups
of the Lwoo into Kenya: the Anywaa of Ethiopia and the Sudan, the Shilluk of the
Sudan and the Acholi of Uganda. In Nairobi, the Anywaa have ‘re-encountered’ their
‘cousins’, the Luo of Kenya. Much to the regret of the Anywaa, the Lwoo are the
most widely dispersed of the western Nilotes. They regard the dispersion of the Lwoo
as a result of Nuer expansionism. In Nairobi, I had the opportunity to discuss the
issue of Nuer expansionism with some Sudanese Anywaa and Shilluk refugees and
their leaders. In their narratives, like so many other Anywaa narratives on Nuer
expansionism, the Lwoo sense of order, justified in reference to the kingship of the
Anywaa and the Shilluk, was contrasted with the Nuer ‘chaos’. Nairobi offers a new
 The Cultural Contestation 105

venue of ‘Lwoo networking’ where the Kenyan Lwoo, the Luo, are one of the major
power blocs in the political landscape of Kenya.6 The following is a summary of nar-
ratives by a Shilluk and a Sudanese Anywaa resident of Nairobi:7

Nuer expansion and assimilation is not sustainable. Look at the Arabs. They
have used Islam as a vehicle of Arabization. This is failing because those who
are the object of Arabization are resisting it. All people who became Arabs are
going back to their origins. As far as the Nuer are concerned, it is even worse
because they have nothing to offer. Culturally the Nuer are zero. Assimilation,
if it works at all, is into a higher culture, the backward into a developed culture.
That is why the Nuer were assimilated into kingdoms such as the Shilluk when
they met cultures higher than theirs. Agrarian cultures are more developed than
nomadic cultures. You still find many Nuer going naked, wealthy as they are.
They still paint their face with ashes. Above all, they have no sense of order. For
a Shilluk, a Nuer is anybody who is chaotic and an outlaw. There is a Shilluk
saying: Yi ba Onwar, which means ‘Are you a Nuer?’ when someone is not
behaving well. An unruly child is also referred to as ‘Nuer’, ‘cie nwar to bie
tuokto’. There is a famous decree by a Shilluk king: ‘There are two living
animals that should not go to Puchudho [the royal centre] – a live chicken and
a Nuer girl’. This is because during the time of Latjor, a Nuer girl was married
to a Shilluk prince. Her child turned out to be not of a royal nature. Like the
chicken, the Nuer tamper with royalty. The Nuer are expanding not because
they have a higher culture, but thanks to their raw power (foolhardy) and their
connections with state power. They managed to annex most of the territories
from Dinka and Anywaa thanks to the support they got from the Sudanese
army. Governments like Nuer chaos because they can be easily manipulated.
The Nuer are like bubbles. You can blow them easily. They are changing sides
like nobody’s business. Is there any society which has developed horizontally?
The Nuer system is in a primordial form, the idea that people are all equal.
Some people should lead. The Nuer have been caught up in a situation where
they cannot evolve. The concept of state is not in their heads. (Peter Adwok,
Shilluk SPLA veteran, Nairobi, August 2003)

The Nuer contest the Lwoo representations of them as ‘chaotic people’. In fact they
take pride in their constructivist mode of ethnic identity formation.

The Anywaa have a certain problem in dealing with outsiders. They tend to
be suspicious with foreigners. With us it is different. We take everybody as
long as they are good people. We even favour outsiders in leadership posi-

6. In August 2003, a delegate of the Anywaa diaspora in the US made contact with Reila Odinga,
a major Luo political actor in Kenyan politics, in order to secure political and financial support
on the basis of Lwoo solidarity. In 2006, the Anywaa diaspora hosted a meeting with Reila in
Minnesota as part of the wider Lwoo political networking.
7. I received permission from both to use their narratives in this book.
106 Playing Different Games

tions. To be a dil is not enough to be a leader. During the election in

Gambella in 1987, for instance, Gatwech Pal and Getachew Mulat competed
in the local election in Jikaw district. Gatwech was the district’s administra-
tor and was a dil from Cieng Cany. Getachew was the district’s secretary and
was an Oromo [Highlander]. The Nuer elected Getachew because he was a
more able leader than Gatwech. The Nuer value people who can unite them
and lead them to victory during war. As a good leader, Getachew had no
problem collecting taxes and implementing government policies. (Bichok
Wan, former Secretary of the Workers’ Party of Ethiopia in Jikaw district
during the election, Nairobi, 25 August 2002)

Contemporary neighbours of the Nuer, however, have a different view of Nuer

assimilationism. Educated Anywaa even refer to Evans-Pritchard’s description of the
Nuer as ‘ordered anarchy’; however, when doing so, they conveniently omit the
adjective ‘ordered’ and replace it with the more derogatory ‘chaotic’. Nuer chaos is
then contrasted with the Lwoo sense of order. Above all, Nuer power is discursively
negated by attributing it to state power, such as in the statements by the Shilluk
above. Thus, the microprocesses by which most of the contemporary Nuer expan-
sion has occurred are reinterpreted as ‘favours’ granted by governments in return for
Nuer ‘subservience’, a logical outcome of the lack of a ‘system’. In this scheme of
interpretation, the Nuer appear not as a powerful people, but as ‘a people without
culture’. Even while the Anywaa acknowledge the success of Nuer assimilationism,
its vitality is attributed to those who assimilate to Nuer ways:

The reason for the success of the Nuer is simple. Those who assimilate go with
the power. The assimilated Anywaa are given power by the Nuer, because they
are the only people in Nuer society who have a sense of justice. That is why
the Nuer system works. It is also because of the gar which locks you inside the
Nuer culture. Once you are marked you are lost as an Anywaa, but not because
you believe in the cultural superiority of the Nuer. The Nuer instead should
be assimilated into the Shilluk or the Anywaa, just as the Amharas did to the
Oromo. Unfortunately, the Anywaa prefer to be committed to so many things
that make us vulnerable to our neighbours, such as the idea of jur and the
value we give to dimui. The Nuer are solitary. Those who live near the Shilluk
are becoming Shilluk. The Dinka and the Nuer have no home. The luak
[cattle byre] is their home. They do not bury their dead, just like the Murle,
who are really wild. It is the Anywaa who introduced farming and burying the
dead to the Murle. The Nuer have become strong by taking other people and
by connecting with state power. In southern Sudan, the Arabs established the
Nuer militia to fight the southerners. The Nuer used that power to take land
from their neighbours. In Ethiopia, President Mengistu gave power to the
Nuer in Gambella, while he turned the Anywaa into a sea of unrecognizable
community. It is one thing to claim power but quite another to live up to it.
There are two famous sayings in the Sudan: ‘Don’t give power to the Nuer, he
will prove a disaster. Don’t give the Dinka the right to divide the food, he is
 The Cultural Contestation 107

not likely to share it with others’. (The late Ambassador Philip Obang, a
southern Sudanese Anywaa, Nairobi, August 2003)

The preceding statement was made by an Anywaa who is sceptical of his own culture.
When he says that ‘the Nuer instead should be assimilated into the Shilluk or the
Anywaa’ and when he reprimands his own people for not incorporating others, he is
deviating from the ‘ethnic purity’ argument of the dominant Anywaa identity dis-
course. In fact, the indispensable bridewealth system and the ideology of ethnic
purity are mentioned by Philip as examples of a cultural trap that has undermined
the demographic standing of the Anywaa in the contemporary inter-ethnic relational
scene. In that sense, what Philip contested is not assimilation per se but rather, like
Peter above, who assimilates whom. In Peter and Philip’s definitions of the situation,
we find a lamentation about the course local history has taken. For the Anywaa, the
‘uncivilised’ Nuer do not deserve to assimilate other people. That is why they believe
that history has taken an unusual course and needs to be reversed or put back on
track. This mindset has morally energized their project of containment.
The discursive struggle between the Anywaa and the Nuer is also embedded in their
respective belief systems and cultural practices. The radical difference in the imagina-
tion of the supernatural between the Anywaa and the Nuer is reflected in their diverg-
ing discourses on issues of entitlement over natural resources. Whereas the Anywaa
draw on their notion of a first-comer (more binding and inflexible) to legitimize their
rights over natural resources, particularly on the question of land, the Nuer draw on a
different tradition of knowledge that bestows ultimate ownership of natural resources
to Kuoth. Unlike the Nuer Kuoth, the Anywaa Jwok is not the ultimate owner of
‘earthly resources’. At its most instrumental, religious discourse is used by the Nuer as
an exit strategy against the Anywaa exclusionary practices. This value disagreement is
noticeable in their debate over access to and control over natural resources.
The following narrative by an elder from Makot village sheds light on the nature
of the debate:

The Anywaa do not like other people and they say all the land of Gambella is
theirs. This Baro River, is it the Anywaa who dug it? Is it not from Kuoth? Who
owns the rivers? Humans can drink, cattle can drink, and animals can drink
water from the Baro River because Kuoth gave us all these things. If somebody
says do not go to the river, is that person a Kuoth?8 You [the author] came from
Gambella town. Have you seen many people living in all the areas [between
Gambella and Itang]? The Anywaa live only along the banks of the river. Where
does Ethiopia start? You cannot finish the land. The land is bigger than the
people. (Tap Manytap, Nuer elder from Makot village, 3 February 2001)

Tap’s reference to Kuoth on the debate over access to and control over natural
resources needs to be embedded in the contrasting notions of the supernatural

8. The reference here was to the prohibition of the Cieng Reng from access to the Baro River by
the Anywaa officials in Itang district during the 1998 conflict.
108 Playing Different Games

between the Anywaa and the Nuer. The relationship between humans and Jwok in
Anywaa cosmology is confrontational. The Anywaa fix responsibility for the exis-
tence of evil squarely on the activities of Jwok. For the Nuer, Kuoth is a gwandong
(grandfather) or intimate friend (madh). Evans-Pritchard (1951b: 4–5) described
Nuer imagery of the supernatural in the following way: ‘God the creative spirit is the
final Nuer explanation for everything … The heaven and the earth and the waters
on the earth, and the beasts and the birds and reptiles and fish were made by him,
and he is the author of custom and tradition … God is thought as of the giver and
sustainer of life’.
Where the Nuer show humility towards God, the Anywaa assert humanity and
claim purity vis-à-vis God. If God is an ‘ally’ of the Nuer, it – the impersonal pronoun
is used intentionally – is a ‘problem’ to be managed for the Anywaa. Thus defiance,
not humility, is the central theme in Anywaa imagery of the supernatural. Here can
be seen a fundamental difference between the Anywaa and the Nuer worldviews.
While the Anywaa believe life is a paradox and a plot authored by Jwok, against whom
they must defend their existence, the Nuer try to surmount the human predicament
through the spiritual agency of Kuoth in which the sacrificial rite plays a central role
(Evans-Pritchard 1956; Meeker 1989). The emphatic rejection by the Anywaa of the
Nuer religious framing of the debate should also be contextualized in these contrast-
ing worldviews. In fact, the Anywaa consider this as a Nuer excuse to take their land.
Whereas land belongs to God in the Nuer religion, it belongs to the human being in
the Anywaa belief system. After all, they resist their own God’s territorial encroach-
ments, exemplified by the Pö symbolic rebellion discussed in Chapter 2.

Anywaa Practical Resistance against Nuer Cultural Hegemony

The Anywaa and the Nuer developed their worldviews independently of each other,
but in contact situations the contrast is reinforced. This has created a definite poten-
tial for a ‘clash of cultures’. The Anywaa are often angered by the ‘participation’ of
the Nuer gods in inter-ethnic conflicts. The aggression against Mut Wiu, the sacred
spear of the Jikany Nuer, by Obang Uriem, the Anywaa governor of Itang district
from 1995 to 2000, is a stark illustration of the different perceptions of the role of
the supernatural in inter-ethnic relations. The trigger for the incident was a quarrel
between two Nuer men in the district where Dhol Koryom, the guardian of Mut
Wiu, lives. One of the disputants was killed, and the issue was settled with cattle
compensation according to the customary Nuer law. When Obang, accompanied by
government soldiers, reached the area where the incident occurred, the case had
already been settled, but he nevertheless confronted Dhol Koryom. Obang accused
Dhol of being a troublemaker and labelled him an ‘evil magician’. In the fight that
followed this act of ‘humiliation’, five Nuer and two government soldiers were killed.
Fearing for his life, Dhol Koryom, carrying Mut Wiu, left his village for the Nuer-
Jikaw district for protection, where he stayed for a year. After strong efforts by the
Nuer officials to exert their influence in the regional council, the district’s budget was
frozen. The civil servants staged a demonstration in Itang town because of the sanc-
tions. Obang was eventually imprisoned, but later released and reinstated to his posi-
tion. Obang turned his attention to the project of the Anywaa irredentism begun in
 The Cultural Contestation 109

1998 – that is, to push for deportation of the Cieng Reng, who as recent migrants
from southern Sudan represented a living example of the ‘foreignness’ of the Nuer.
In fact, he made several attempts to deport them back to southern Sudan or to the
refugee camps.
Obang’s political project is typical of a group of Anywaa who still harbour an irre-
dentist sentiment towards the territories lost to the Nuer, a sentiment kept alive and
nurtured by the continuous pressure of the Nuer towards the Itang area, which
threatens to upset the demographic balance and power relations between the two in
the district. The Anywaa still dominate politics in Itang district, although there is a
sizeable Nuer settlement. Thus, any move towards proportional politics could be
viewed as a threat. In the context of the increasing link between politics and demog-
raphy, the settlement of the Cieng Reng Nuer at Makot village has become politi-
cally visible. The Obang administration began to take political measures to make the
settlement illegal, the first of which was to refuse the tax the Cieng Reng Nuer were
paying to the local authorities. The neighbouring Anywaa villagers were also advised
not to enter into any form of exchange, particularly an exchange involving land and
fishing rights. On a more personal level, Obang came from an Anywaa family that
was displaced by the Nuer. His father’s name, Uriem, means ‘one who is chased’.
Obang’s engagement with the Cieng Reng, therefore, had a biographical dimension.
Practical resistance by the Anywaa to Nuer expansionism is also evident in a prac-
tice known as chirawiya, which serves to disrupt Nuer expansionism through inter-
ethnic marriages. The more the Nuer use marriage as a means of expansion, the more
aware the Anywaa become of its political implications. Nevertheless, in the last
instance ‘the cow wins the day’ – the individual Anywaa still enters into socio-eco-
nomic exchanges with the Nuer for economic reasons. There is a gap here between
ideology and the rigours of everyday life. In the realm of pragmatism and economic
imperatives (the Anywaa’s relative poverty), no matter how hard the Anywaa try to
block asymmetrical inter-ethnic exchanges, policing individual members across the
ethnic boundary would be futile.
Economic realities encourage inter-ethnic marriages. The Anywaa compromise the
ideology of ethnic endogamy by marrying Nuer in order to ensure the flow of cattle
wealth. Although a Nuer does not pay as many cows for an Anywaa girl as he pays for
a Nuer girl, the minimum payment (up to five cows) goes well beyond the average
Anywaa bridewealth, whether in the form of dimui or in terms of cows or money.9 For
a Nuer, marrying an Anywaa is an economic rationality not only because he pays less,
but also because it allows him to gain access to natural resources. In the process of
coping with their strategic dilemma, the Anywaa have devised a mechanism that will
enable them to simultaneously have their cake and eat it. An individual Anywaa who
marries into a Nuer family is entangled with kinship responsibility and moral obliga-
tions. He will not attack his Nuer in-laws nor the multitude of other Nuer who become
his guests. Guests are highly respected in both the Anywaa and Nuer cultures; hosts
have strong moral obligations to ensure their safety. Even those Anywaa who acknowl-

9. The Anywaa economy was forcibly monetized by the Ethiopian state in the late 1970s. This
issue is further explored in Chapter 6.
110 Playing Different Games

edge the hospitality of the Nuer consider them as ‘good hosts but bad guests’, when
they reflect on how Nuer guests gradually dominate their hosts. The Anywaa seem to
have redesigned chirawiya to ‘discipline’ the Nuer into being good guests (welo).
In the Anywaa language, chirawiya means (literally) ‘lifting up the hands’, but it
is also used as a metaphor for withdrawing one’s protection. Thus, when an Anywaa
says ‘chirawiya’, he means, ‘I am not in charge’. Originally the concept of chirawiya
was used in the context of dispute settlements in which a mediator enters the scene
as an arbiter for the conflicted parties.10 While the case was being handled by the
arbiter, neither of the parties was supposed to antagonize the other. As a last resort,
when the disputants failed to reach a negotiated settlement, the arbiter withdrew his
protecting role, signalling the seriousness of the problem. The withdrawal of an
arbiter often resulted in bitter fights. Later, chirawiya was adopted as a form of resist-
ance to undercut the expansion of the Nuer by means of microprocesses. The
Anywaa who hosts a Nuer is asked to ‘lift up his hands’ in order that the non-related
Anywaa can kill the Nuer. Applied in this context, therefore, chirawiya becomes a
‘face-saving’ formula for the Anywaa host in managing competing loyalties.
The objective of chirawiya is to deny the vitality of the Nuer cultural form, for if
fewer Nuer marry non-Nuer, this would mean a reduction in the expansion of Nuer
culture. It took a while for the Nuer to recognize the Anywaa ‘plot’. Over a period
of time, such practices have caused growing irritation on the Nuer side. While the
Anywaa ‘demonize’ the Nuer cultural form, the Nuer ‘criminalize’ the Anywaa means
of resistance. They interpret chirawiya as allowing ‘cold-blooded’ and ‘indiscriminate’
killings and use it as evidence for the ‘morally corrupt’ nature of the Anywaa. It is for
this reason that the Nuer call the Anywaa ‘murderous people’ (luuch naath).
Chirawiya is contrasted with the Nuer’s more ‘humane’ notion of hospitality (neekä).
Among the Nuer, all guests, including the rul and the jaang, are categorized as
members of a minority group, and the host has a strong moral obligation to protect
all of his guests. For the Nuer, their ‘minority policy’ allows them to stand on a ‘high
moral ground’ that legitimizes their cultural form while it ‘criminalizes’ that of the
Anywaa as the following summary of a statement by a Nuer suggests:

We do not like some of the things the Anywaa do, such as killing secretly. Of
course, the Nuer also kill but we do not simply kill. We do not attack women,
children or a man who is not in a group. But the Anywaa kill indiscrimi-
nately. They are all the same wherever they are, even those in the towns.
When there is conflict in Gambella town, the Anywaa go to the forest and
kill the Nuer women who collect firewood. This would never happen with
the Nuer! Nothing happened to those Anywaa who lived with us in Jikaw
throughout all these years of conflict. During the girgir killings the Nuer
became very angry and told the Anywaa, ‘If you want to fight with us there
is no problem. Let us go to a place where there are no government soldiers to
see who will win!’ (Nuer civil servant, Gambella town, 28 October 2000)

10. Chirawiya is particularly common in the district of Jor.

 The Cultural Contestation 111

What the Nuer call ‘indiscriminate killing’ is not only chirawiya but encompasses all
Anywaa surprise attacks. The killing of Nuer women in the forests illustrates the level
of anger on the side of the Anywaa, who use unseen violence as a communicative act
intended to hurt the enemy. The Anywaa ‘cold-blooded’ killings to which the Nuer
refer could also be an evasive strategy to escape the observational gaze of the state.
No matter what the intention(s), the very act of killing unprotected Nuer women
feeds into a moral debate. In the Nuer conception of warfare, an open and regulated
battle, the indiscriminate violence of the Anywaa appears an act of cowardice:

The Anywaa kill weak people. A Nuer is killed by a Nuer; but he would tell
the parents not to look for the victim and would confess that he killed him.
If a death is not reported, it could only be a murder by the Anywaa. This is
one of the main reasons that create conflict between us. The Anywaa of today
have the same behaviour as the Anywaa in the past. They murder, and do not
fight openly. The Nuer are angry because of these killings by the Anywaa. The
Anywaa are also not good to one another. They kill each other by cursing.
(Gatwech Gatluak, Nuer elder, Ochom village, 8 November 2000)

These statements by the Nuer make sense if they are placed in a cultural context. The
Nuer make distinctions between different kinds of conflict: dwac (individual
duelling), ter (intra-tribal feuds), kur (tribal wars) and pac (cattle raiding). In all these
conflicts there is a ‘code of conduct’. Categories of enemies are clearly demarcated.
The only unpredictable fight is biem. Biem is the ‘cooling-off ’ period after a feud, in
which it is culturally tolerated for relatives of the deceased to kill any relative from
the enemy side as an act of compensation even after an agreement has been reached.
Individuals from the enemy side normally take all necessary precaution not to run
any such risk. If anybody is killed through biem, it is his own responsibility. The
institutionalization of conflict enhances predictability and the preparedness to deal
with it. In addition, the Anywaa are not recognized by the Nuer as enemies, for this
implies a certain degree of equality. In the Nuer definition of an enemy, parties to a
conflict are of the same status. For the Nuer, ‘the real enemies’ are the Dinka and the
British. From the Nuer perspective, the Anywaa are ‘too small a group’ to be consid-
ered as such. Here the self-images of the Anywaa and the Nuer are in collision. They
derive ethnic honour from different sources. Anywaa self-esteem is built on the basis
of a discourse on civilization, whereas Nuer self-esteem is on the basis of effective
assimilation, heroism and military exploits against their neighbours.
In contemporary Gambella, being an Anywaa or a Nuer implies being a particu-
lar kind of person in the eyes of the other. Thus the Nuer are principally defined as
‘born aggressors’ by the Anywaa, while the Anywaa have become ‘murderous people’
to the Nuer. Both abstractions are premised on a particular way of looking at the
world, ways which have produced their own rationality. Nuer ‘chaos’ appears incom-
prehensible to the ‘orderly’ Anywaa, whereas the Anywaa seem to be xenophobic in
Nuer eyes. The mutual feelings of anger characterizing relations between Anywaa
and Nuer have thus induced a moral contestation over definitions of humanity. This
has the effect of a new ‘primordial game’ in inter-ethnic relations. By fixing a ‘sub-
112 Playing Different Games

stance’ on Anywaa nature, the Nuer have begun to enter into the Anywaa concep-
tion of an ethnic group and, thereby, have also started to compromise their own
assimilation ideology, an ideology which has always been based on the assumption of
a certain degree of sameness. Here we find the Anywaa and the Nuer starting to play
the same ethnic language game. The conflict situation is having the effect of further
defining them, of producing similarity through opposition.11
The cultural differences and the divergent schemes of interpretation between the
Anywaa and the Nuer are further illustrated in their contrasting notions of host and
guest. For the Anywaa, being a guest (welo) is a permanent status, a concept that is also
used in intergroup (inter-village) relations within Anywaa society. It is contrasted with
the term jobur (first settlers of a village). For the Nuer, being a guest is a temporary
status, a phase in the localization process, and a concept that is also applied intra-ethni-
cally. The notion of a first-comer among the Nuer (dil) is a framework of inclusion for
newcomers, Nuer and non-Nuer alike, within which localization occurs through adopt-
ing the lineage name of the dil. For the Anywaa, the Nuer immigrants to their villages,
now related through affinal ties, are defined as welo no matter how long they stay. On a
long-term basis, however, the Nuer have largely managed to establish their definition of
being a host or guest through economic incentives, demographic power and the manip-
ulation of kinship ties. This instrumentalization of inter-ethnic exchanges, coupled with
flexible ethnic recruitment and elaborate assimilation ideology, has resulted in the one-
way process of ethnic conversion and the expansion of Nuer cultural space, a process
which has created different categories of Anywaa with different concerns.
In the context of such profound cultural differences, the Anywaa interpret the
Nuer expansion as an ethnic conspiracy. The Anywaa definition of the inter-ethnic
conflict situation as such a conspiracy can be made intelligible by regarding folk con-
spiracy theories as theories of power (Waters 1997). Conspiracy theory is a theory of
power of a particular kind – invisible power. People have a need to come up with
explanations when situations become complex and abstract. Facing complex power
systems such as the effectiveness of Nuer assimilationism, the Anywaa consider that
this power has to come from somewhere and that there must be individual agents
who carry it. The power of the Nuer involves a certain degree of invisibility, inas-
much as it is a power without a centre. This power is manifest in the expansion of
the Nuer, but this expansion occurs through microprocesses at the level of the indi-
vidual. The Anywaa, on the other hand, are used to a different kind of power, that
of centralized power, i.e., village kingship or headmanship. The invisibility of Nuer
power makes it difficult for the Anywaa to resist it. The Anywaa are therefore forced
to reflect on the situation and piece together the fragments of this power and its
mechanisms in order to make sense of the conflict situation. In doing so, they attrib-
ute an element of directionality and mission to ‘Nuerness’.
At the heart of the complication lies the units of reproduction of the Nuer social
system: the microdemographic processes. These microprocesses – inter-ethnic marriages

11, For a comprehensive analysis of ‘integration through conflict’, see the comparative study by
Horstmann and Schlee (2001).
 The Cultural Contestation 113

and friendship networks – are social strategies used by the individual Nuer to both widen
his resource base and ensure his physical safety. It is through extensive marriage ties and
the resultant social networking that the bulk of Nuer expansionism at the expense of
Anywaa territories since the conquests of the nineteenth century has occurred. This is
seen as ‘ethnic’ in the eyes of the Anywaa because the individual Nuer draws on the same
cultural model of resource extraction and social security with more or less the desired
effect. But attributing intentionality to the Nuer is an external definition, a retrospective
reading of an objective consequence (a latent function) to a subjective motivation, which
is already explained in Merton’s terms. Each individual Nuer has a stake in the cultural
model, aka Nuerness, because it is functional in solving their everyday problems. In that
sense, the Nuer cultural form is a form of power whose frontier is expanding on the basis
of its efficacy for the individual Nuer who practices it. On the other hand, the effective-
ness of the Nuer cultural form for the individual Nuer has created apprehension for the
Nuer’s neighbours. This is evident in their modes of resistance, such as chirawiya, which
targets, at a microlevel, the very basis of the reproduction of the Nuer cultural form.

The Anywaa in Search of the Nuer ‘Hidden Agenda’

The Anywaa elites seek evidence for their definition of the conflict situation as a
Nuer conspiracy in order to enhance its plausibility. They have found two conven-
ient pieces of ‘evidence’: the Nuer song of the Tier Agak, composed in 1912, and a
‘confidential letter’ that circulated in Gambella town from 2000 to 2002. In the fol-
lowing, this Anywaa ‘evidence’ of a Nuer conspiracy is discussed, as well as the extent
to which it has a bearing on Anywaa–Nuer relations.

The Tier Agak song and the ‘Nuer conspiracy’

Song in Nuer Translation in English

(1) Diu came by his own

(2) Majak, son of Tinpe gar their servant boy

Diu Majak

(3) Diu will curse you and you will have no

testicles and the curse of Anywaa will match
the Nuer curse

(4) Majak is in trouble like the fire of a

burning pipe or like burning wood

(5) When Thoch was discovered they all were-

cieng of Gaat-Jak.

(6) Our Gaat-Jak, if you defeat the

Highlanders, if you defeat the red people, who
else in the world will face up to you?
114 Playing Different Games

(7) Even if you are turuk

(8) Even if you are circumcised people

(9) Hawk bird, this is country of all Jikany,

all Jikany of bush. Bush of Kedol, good bush

(10) I don’t want to go back through wath

Ngok. The force of Wudol Anywaa was driven
back by Nuer

(11) Oh, Anywaa, we will make you shit and

chase you up to Tier Agak. Between you and
us there would be no stopping of war. Those
cattle can not be stolen by the Anywaa. We
will destroy the Anywaa unity.

(12) What do the Anywaa want from us?

What do foreigners want from us? We will
fight until the dust covers us.

(13) Wiu help me, spear of Mathiang help me.

We always killed Dinka, we Gaat-Jak, we
always killed Anywaa

(14) The people of this country have always

talked about war; that they will fight with the
Pinykew.

(15) All children of Anywaa will see the flame

of gun and we will drive the Anywaa into the
pool. Don’t be afraid, we know it is not good
to die but a nation survives on the strength of
the youth.

The song grew out of the Battle of Biot fought between the Gaat-Jak Nuer and the
Openo Anywaa in 1912. For the contemporary Anywaa, a revelation of particular
importance is verse 11, which provides evidence for the ‘hidden’ Nuer agenda. Bär is
a Nuer derogatory name for the Anywaa, primarily signifying people without cattle.
Bia lac is an abusive word (‘to make somebody shit’), and Tier Agak is the eastern-
most settlement of the Anywaa, near present-day Bonga town, some forty kilometres
east of Gambella town. In this verse, therefore, the Anywaa read not only of their
humiliation, but also of Nuer arrogance, aggression and the ultimate objective of
Nuer expansionism: the displacement of the Anywaa from all their lands. As Bonga
is the last Anywaa settlement at the foot of the highlands, the reference to Tier Agak
in the song is interpreted as a code word for the ultimate extermination of Anywaa
 The Cultural Contestation 115

society.12 Contemporary Anywaa often refer to the song of the Tier Agak when envis-
aging the future as an ‘impending danger’:

The Anywaa were living in Jikaw. The Nuer defeated them and took Jikaw and
renamed all the Anywaa villages. The Nuer will not stop until they take over
all the Anywaa lands. They even have a song for their plan. We get angry when
we hear this song. (Anywaa civil servant, Gambella town, 22 June 2000)

The Nuer, not surprisingly, have a different reading of the circumstances which led
to the Battle of Biot and the composition of the Tier Agak song:

There was a very famous Anywaa kwaaro called Diu Majak. His wife once
said to him, ‘If you want to remain as my husband, fulfil my dream. I want
to bath with Nuer milk’. Diu organized a bunam [mobilization of all the
youth]. Then a very big war broke out between the Anywaa and the Nuer.
The Anywaa defeated all the Gaat-Jak Nuer except the Cieng Wau. They
chased them all the way to Biot where Diu was killed. The Nuer were very
angry and they wanted to defeat the Anywaa up to Tier Agak. (Riek Tuany,
Nuer elder from Ochom village, Pinykew, 14 October 2000)

For the Nuer, the Tier Agak song signifies two things: justice and pride. The emphasis
is on Anywaa aggression and the decisive victory the Nuer scored afterwards. According
to the Nuer version of the song, verses 6–8 are references to the 1911 conflict between
the Gaat-Jak Nuer and the Ethiopian state: ‘We defeated the red people even if they are
powerful’. Verse 4 compares the spiritual powers of the Anywaa and the Nuer, whereas
verse 10 ‘ratifies’ the nineteenth-century Jikany conquest of Anywaa lands. In the first
two decades of the twentieth century, however, the Nuer were weaker militarily than
the Anywaa because of the Anywaa’s earlier access to firearms. Viewing the balance of
power of the day, it would seem that the Tier Agak song was sung to discourage the
Anywaa from attempting to reclaim territories lost to the Nuer at the end of the nine-
teenth century. The contemporary Nuer describe the Tier Agak song more as propa-
ganda than an actual plan, whereas the Anywaa refer to the song to validate their belief
in the Nuer conspiracy of dismembering the Anywaa’s existence.

12. The Tier Agak song is also constantly referred to by the Anywaa diaspora in North America.
In one of the peace debates in a Southern Sudanese Discussion Board, referred to as Gurtong,
an Anywaa made a reference to Tier Agak to indicate the Nuer agenda: ‘To this day the Nuer
follow wherever the Anywaa go. They have reached Gambella region in Ethiopia and want to
take the town itself. “We have reached Tier Agak; we must go to Pochalla and round up the
whole land”. That is the goal. It has always been a fight and death along the way to that goal.’
The full exposition can be read on the Gurtong website: ‘The Anywaa Need Peace and
Happiness’, posted by J. Ojoch, 3 April 2003, www.gurtong.com.
116 Playing Different Games

The ‘confidential letter’ and the Nuer ‘conspiracy’

In June 2000 an educated Anywaa gave me a two-page letter written in English so that
I could ‘learn about what the Nuer are doing to the Anywaa.’ The ‘confidential letter’,
entitled ‘The Nuer Invasion and Colonization Plan for Anywaa Lands,’ was allegedly
written by Nuer youth in the Sudan and Gambella. The letter contains sixteen points
that elaborate on ‘the colonization plan’. It begins with the statement, ‘Every Nuer must
know that the Anuak [Anywaa] are his traditional enemy and that our colonization,
invasion and assimilation plan must be applied to all Anywaa regions both in Ethiopia
and Sudan’ (#1). The letter addresses two audiences: villagers and intellectuals, who are
connected through the overarching ‘Blood Exchange Pact’ (#7) and ‘cold policies’ [Nuer
marriage practices] (#2). The following is a direct quotation from the text:

With regard to conquering the Anuaks you [the Nuer] must use the ‘Blood
Exchange Pact’ in which the Anuak foolishly believe. Haven’t we desecrated it
several times with impunity? We must fight the Anuak and their leaders who
oppose our interests in their fertile land. Avail yourselves of the Anuak greed.
They do not want to live with us but we must keep on with our invasion poli-
cies. When they leave the land for us, we will continue going after them until
we finish their race… In the past our invasions and colonization plans were not
written on paper as today. It was simply in our heroes’ minds. We are, however,
proud of our people because they have always cooperated and were occupiers.
If we apply the above principles, we are sure that the great Nuer kingdom will
be established and we will assimilate all of this small [Anuak] tribe (##1,2).

With reference to the nineteenth-century Nuer conquest of Anywaa territories, the

contemporary conflicts between the Anywaa and the Nuer are understood in terms
of ‘ancient hatreds’. On that basis, the future is envisaged in terms of impending
danger, as seen in this additional excerpt from the ‘confidential letter’:

To remind young Nuer about what your national heroes have done in the
past, here are some of the important deeds they did: we are originally from
Banteu, but our heroes found the fertile land of the Anuaks that was Nyium
Abiel (Nasser) and they invaded in 1850 and have colonized until now. We
defeated the Anuaks and Dinkas in that fertile land and we assimilated those
who did not want to move. Our heroes did not stop there. They went on to
Akobo and colonized it. Here they applied cold policies. They carried out
those invasions when there was no money and no education; so what about
now when you are all intellectual? As we said above, it is your duty to finish
the unfinished job that your heroes left behind (#2).

The Nuer reject the Anywaa allegation that a Nuer wrote the letter. They refer to the
way the Nuer place names are spelled and the mentioning of the ‘Nuer Kingdom’ in
the text as proof that the letter was written by an Anywaa in order to overcome the
bitter division of the Anywaa elites in the run-up to the 2000 regional election.
Putting aside the difficult task of identifying the author, it is more revealing to engage
 The Cultural Contestation 117

with the political life of the letter. Above all, the ‘confidential letter’ has helped the
Anywaa to focalize and objectify the Nuer ‘hidden agenda’. As W.I. Thomas reminded
us in his famous theorem, if people believe something is real, it becomes real in its
consequences. The first political effect of the letter was felt among those Anywaa who
are living in the diaspora in the US. The Anywaa Community Association in North
America (ACANA) held a meeting in Minnesota in March 2000 to seek ways to deal
with the Nuer ‘colonization plan’. A fund-raising event was organized to generate
funds to build up the military capacity of the Anywaa in Gambella. In this meeting,
the level of concern was so dominant that even indications that the ‘confidential letter’
may have been forged by other interest groups became reasons for increasing political
awareness of the ‘impending danger’ of the Nuer:

In this meeting, an Anywaa raised his hand and asked, ‘How do we know that
it [the “confidential letter”] was written by a Nuer?’ Then the organizers of the
meeting replied, ‘Whoever wrote the confidential letter, the content is true. If
an Anywaa wrote it, he could only be a patriot, because he alerted us to the
threat from the Nuer.’ Many of the participants agreed with the organizers’
position. I also believe that the issue is not new and that the paper merely made
us aware of its present conditions [aneqan, an Amharic term meaning ‘to be
awakened’]. We, the Anywaa, believe that the Nuer are exterminating us
silently, through marriage and friendship (Amharic: Nuer dimtsun atifto naw
yemiyatefan). The concern was rather, ‘What if the [“colonial”] project started
10 years ago? These days, Anywaa leave their lands to other people instead of
confronting them. It was not like this before. Now the Anywaa just sell their
land for 300 birr. What would happen in the future if Anywaa continue loosing
their land?’ Then the discussion focused on the issue of ‘how the Anywaa can
best deal with the Nuer threat?’ Finally it was decided that the only way is to
strengthen militarily those Anywaa who live next to the Nuer. After the meeting
we came to know that the secret was exposed by a half Nuer and half Anywaa
who knew about the plot. (personal communication with Ariat Ochalla, an
Anywaa woman living in Minneapolis, Minnesota, USA, 12 June 2007)

The ‘confidential letter’ was also an escalating factor during the 2002 conflict
between the Anywaa and the Nuer. At the height of the conflict, an Anywaa politi-
cal organization convened a unilateral extraordinary session of the regional council
to discuss the ‘confidential letter’. In the minutes of that meeting, the letter was men-
tioned as the official Nuer position that drove the conflict in the region.
Subsequently, Anywaa politicians actively discouraged inter-ethnic marriages and
economic exchanges while alerting the villagers about the Nuer ‘peril’.13

13. Schlee (2008) discussed a similar issue in inter-ethnic relations between the Garre and the
Boran-Gabra alliance – how a sketch map used by an academic to show the approximate set-
tlement pattern of the Garre in southern Ethiopia and northern Kenya was perceived by the
Boran and Gabra as a graphic manifestation of the Garre’s territorial ambition on their lands.
This map fuelled the conflict between them in 2000.
118 Playing Different Games

The preceding discussion has shown how contrasting ethnic identity formations
are implicated in the conflict between the Anywaa and the Nuer. This is expressed in
various ways: the unidirectional process of ethnic conversion and its discontents;
conflicting narratives of property relations; the collision of self-images; constrasting
power regimes; and the Anywaa’s search for the ‘hidden agenda’ of the Nuer . It has
also been shown that the identity variable operates through the resource and power
variables to generate ethnic conflict. The resource variable is linked with the identity
variable, insofar as the need or desire for resources leads the Nuer to elaborate their
constructivist mode of ethnic identity formation. This, in turn, leads to the contin-
ual territorial and cultural expansion of the Nuer. The gains from the resource expan-
sion are then invested in ethnic recruitment. The link between the identity and
power variables is expressed in the renewed interest that the Anywaa have shown in
resisting Nuer expansion, especially during their short-lived politico-military
empowerment in post-1991 Gambella. The following chapters show how the power
variable generates ethnic conflict in the context of the state system, and trace the
interaction of the power variable with the resource and identity variables in the esca-
lation of Anywaa–Nuer conflict. It is a story about the fluctuation of inter-ethnic
power relations within the political arenas organized by the state.
 Chapter 6

Differential Incorporation
into the Ethiopian State
This chapter provides an in-depth analysis of the ways in which the Anywaa and the
Nuer have been incorporated into the Ethiopian state in two successive regimes. The
basic premise is that different patterns of incorporation into an ethnically stratified
state and corresponding fluctuations in inter-ethnic power relations may themselves
be viewed as causes of ethnic conflict. In keeping with the fundamentally multi-
causal approach advocated in this book, however, the power variable will be viewed
in its interaction with the resource and identity variables.
One of the main features of contemporary inter-ethnic relations is the role of the
state in shaping the context of interaction. In a relatively recent addition to his pre-
vious insights on ethnic phenomena, Barth (2000 [1994]) explains why this is so:
‘Firstly, it is essential to recognize that a modern state provides a vast field of public
goods, which it may allocate to categories of persons, or leave open to competition.
Secondly, the state also deals directly with groups and categories of people, regulat-
ing their lives and their movements … Valued resources are arbitrarily allocated, or
denied, by bureaucratic action, thereby creating communities of fate – which will
next tend to emerge as social, self-aware groups – from formal legal categories’ (2000
[1994]: 19). Elaborating on the organizational effect of the state on ethnic processes,
Barth continues as follows:

So as to integrate the level of statehood successfully into our analysis, we need

to see the state as an actor, not merely as a symbol or an idea. To do this we
must employ analytical procedures that differentiate rather than lump states
in terms of their structures and the patterns of action they pursue. I suggest
that we must start by analyzing these policies of each state by linking the poli-
cies to features of the regime, that is, the state’s policy-making core. We are
then able to depict the power represented by the state as a specifiable third
player in the processes of boundary construction between groups, rather than
confound the regime, and its powers and interests, with the more nebulous
concepts of state and nation. (Barth 2000 [1994]: 19)

In making sense of the contemporary conflict between the Anywaa and the Nuer, I
rely throughout the remainder of this book on this analytical framework, which iden-
tifies the state as a ‘specifiable third player’ in inter-ethnic relations. This necessarily
entails a discussion of the role that the Highlanders, who are closely identified with
the Ethiopian state, have played and continue to play in the Anywaa–Nuer conflict.
120 Playing Different Games

The modes of incorporation of the Anywaa and the Nuer into the Ethiopian state
differ significantly, although both Anywaa and Nuer were incorporated into an ethni-
cally stratified state system. The Anywaa were hit hard by the territorial and cultural
expansion of the Ethiopian state, whereas the Nuer ‘benefited’ from neglect. This
introduced unequal power relations at the inter-ethnic level. Living in outlying dis-
tricts far from administrative centres, the Nuer were not subjected to the same level of
political control as were the Anywaa. The Nuer’s larger cross-border settlement
pattern between Ethiopia and the Sudan was also a diplomatic resource that had kept
colonial Britain and imperial Ethiopia at bay. Neglect also meant less social disruption
for the Nuer. While the Anywaa political organization was abolished and their local
economy forcibly monetized, the Nuer managed to cushion themselves from the
Ethiopian state and market forces due to the continued use of cattle in creating and
maintaining social relationships. These various modes of incorporation resulted in dif-
ferent identification strategies of the Anywaa and the Nuer vis-à-vis the Ethiopian
state. Incorporation into the state system also created a new regional power structure
within which the Anywaa and Nuer elites competed for political power. The
Anywaa–Nuer conflict, however, is not merely a reflection of the state’s political
behaviour. The state is also an arena where older patterns of hegemonic struggles are
acted out and elaborated on, and where the microlevel social struggle increases and
becomes more contested and legitimized. There is, in short, local agency.
The expansion of the Ethiopian state at the local level has also meant changes in the
quality of inter-ethnic relations. The political and economic decline of the Anywaa has
encouraged the Nuer to change their regulated seasonal settlements along the banks of the
rivers, which were established with the consent of local Anywaa leaders, into permanent
settlements with or without the consent of the latter. Nuer settlements in Anywaa territo-
ries have dramatically expanded since the 1960s, and some of the Anywaa villages have
been renamed in the Nuer language, thus signifying the change from access to control of
the riverine lands. The expansion of the Ethiopian state at the local level, on the other
hand, was not backed by an ‘integrative revolution’ in the Geertzian (1963) sense of the
term. The ethnic and racial character of the Ethiopian state has precluded the positive inte-
gration of the Anywaa and the Nuer into the national identity. The ethnicities of the
Anywaa and the Nuer emerge at the local level then as reactions to the integrative failures
of the Ethiopian state. Ironically, the local expansion of the state has meant a change in the
framing of inter-ethnic relations. Viewing the state as a new and powerful actor at the local
level, Anywaa political actors have reframed their troubled relations with the Nuer in
national terms: a conflict, respectively, between Ethiopian and Sudanese citizens. The
expansion of the Ethiopian state has also meant changes in the demographic structures of
the region. As late as 1984, the Anywaa were a demographic majority. A decade later they
had been surpassed by the Nuer and nearly matched by the Highlanders.
The first section below discusses the incorporation of Gambella into the imperial
Ethiopian state system in the late nineteenth century and its impact on the local identi-
fication process up until 1974. The following section focuses on Anywaa–Nuer relations
during ‘socialist’ Ethiopia (1974–91). In doing so, the chapter provides the background
against which the dynamics of inter-ethnic relations in contemporary Gambella since
1991 – the object of analysis in subsequent chapters – can be understood.
 Differential Incorporation into the Ethiopian State 121

The Anywaa, the Nuer and Imperial Ethiopia

Gambella was incorporated into the Ethiopian state at the end of the nineteenth
century (Bahru 1976: 25). In 1898, Emperor Menelik of Ethiopia pre-empted
British colonial interest in the region by extending his dominion as far west as the
Sobat basin in the Upper Nile region in present-day southern Sudan in connivance
with the French, who were simultaneously advancing into the region (Jal 1987:
183–84).1 Imperial Ethiopia had two major stakes in the Gambella and Upper Nile
regions. Economically it was interested in safeguarding the lucrative ivory and cattle
trade of the late nineteenth and early twentieth centuries (Johnson 1986: 222–24).
The main sources of this merchandise were the newly conquered regions in the
south, west and east of present-day Ethiopia. In fact, it was control of the trade routes
that enabled King Menelik to turn his small Shewan kingdom in central Ethiopia
into an empire that became the modern Ethiopian state (Bahru 1976: 24). Gambella
was one of the main sources of ivory and cattle for imperial Ethiopia. Ivory was
obtained both in the form of tribute and through trade with the Anywaa and Nuer
leaders (Johnson 1986: 224–30). The quest for ivory led to a flourishing gun-for-
ivory exchange, as more guns were needed to hunt more elephants. This economic
activity brought imperial Ethiopia into strong competition with the British, who had
a wider political, economic and strategic interest in the region.
The overriding British strategic interest in Gambella was their preoccupation
with safeguarding the waters of the Nile (Collins 1971). One of the main tributaries
of the White Nile, the Sobat is also fed by major tributaries, the Baro, Gilo, Akobo
and Alwero Rivers, that descend from the western Ethiopian highlands and flow
through the plains of Gambella. The British also had an ambitious economic
scheme. They aspired to tap the natural resources, coffee and rubber, of the western
Ethiopian highlands, and for that they needed a commercial enclave linking these
regions with the Sudan (Bahru 1987: 80–81). They were also in competition with
the French, who had forged close diplomatic and economic ties with Emperor
Menelik, evident in the construction of the railway that connected Ethiopia’s capital,
Addis Ababa, with the French colony of Djibouti. In order to undermine the
growing French political and economic influence in Ethiopia, the British negotiated
with the Ethiopian government to establish a trading station in Gambella on the
Baro River, the only navigable river in Ethiopia. According to Article IV of the sub-
sequent 1902 Anglo-Ethiopian Boundary Agreement, the British were allowed to
establish a trading post on the Baro River, in the western part of Gambella town,
which came to be known as the Gambella enclave (hereafter the enclave).
On the basis of the agreement, Gambella town emerged as an important eco-
nomic centre in the first three decades of the twentieth century, handling seventy per

1. As part of the Franco-Abyssinian expedition to the Upper Nile region, also called the Bonchamps
mission, Emperor Menelik sent military units to the Sobat basin; they arrived earlier than the
British and planted the Ethiopian flags at the Nuer-inhabited areas along the south bank of the
Sobat (Jal 1987: 184). Bahru (1976: 74) mentions that ‘You, the Nuer chief, had signed a treaty
– presumably of submission – and received two Ethiopian flags and some clothes. The Nuer had
also been given strict orders to have no dealing with anyone but Ethiopians’.
122 Playing Different Games

cent of Ethiopian foreign trade with and via the Sudan (Bahru 1987: 77). British sov-
ereignty over the enclave, however, was conditional on their rule over the Sudan, and
thus ended in 1956 when the Sudan became independent. For the preceding half-
century though, Gambella, with a parcelled sovereignty, occupied a unique status as
somewhat of a political anomaly in the context of independent Ethiopia.2 In addi-
tion to British attentions, the enclave was highly favoured by Menelik as an inlet for
salt and cloth imported from Port Sudan and an outlet for coffee, hides and beeswax
from the newly conquered western highlands (Bahru 1987: 82–83).
The 1902 Boundary Agreement also defined the national identities of the
Anywaa and the Nuer. Except for a section of the Jikany Nuer, the majority of the
Nuer were placed within the Sudan and except for the Adongo and Akobo Anywaa,
the majority of the Anywaa were placed within Ethiopia.3 The Anywaa and the Nuer
were thus differentially incorporated into the two states.

The Anywaa under imperial Ethiopia

The Anywaa interacted with the Ethiopian state system earlier than the Nuer because
of their proximity to the highlands and their new nationality as ‘Ethiopians’. The
Anywaa village states were initially better connected with the local representatives of
the Ethiopian state. The nyiye and kwaari responded to the new political opportuni-
ties, which above all ensured them earlier access to firearms than the Nuer, a new form
of dominance that decisively changed the balance of power in Anywaa–Nuer relations
in the former’s favour (Bahru 1976; Johnson 1986). Three powerful nyiye emerged in
the first decade of the twentieth century: Udiel, Ulimi, and Akwei of, respectively, the
Abobo, Akobo and Adongo regions. The most powerful among the kwaari were Adiu
Nyigwo of Edeni village and Urubolo of Pinykew village, both on the Baro River. In
contrast, the placement of the Nuer under the more politically assertive British
administration put them at a disadvantage. With access to firearms and a new form
of political centralization that went beyond the traditional village constituency, the
Anywaa took the offensive and, in the first three decades of the twentieth century, had
the upper hand over their once powerful pastoralist neighbours.
The balance of power during these decades contrasted sharply with the Anywaa–Nuer
power relations during the second half of the nineteenth century. Various travellers and his-
torians reported that by the end of the nineteenth century the Anywaa were said to be on
the verge of extinction after waves of displacement by the Nuer. Jessen (1905: 5) wrote:
‘There is no doubt that these people, who, sad to say, are gradually becoming extinct, are
greatly influenced by their surroundings and the peculiar circumstances in which they are
placed. Shut in on one side by the giant Abyssinian Mountains and on the other by the
warlike and ever-aggressive Nuer tribes, their existence is not much better than that of the

2. Ethiopia is the only African country that was not colonized by the European countries, except
for the brief occupation by fascist Italy (1936–41). The Gambella enclave was administrated
first by the British Custom Inspector and later on as part of the Upper Nile region.
3. As late as 1931 there were only two thousand Gaat-Jak who were permanent residents of
Ethiopia, whereas forty-five thousand Jikany crossed the border annually on their way to dry-
season grazing grounds (Hutchinson 1996: 112).
 Differential Incorporation into the Ethiopian State 123

flying fish’. Collins (1971: 203) concurred: ‘They [the Nuer] left the Anuak shattered.
Many had died opposing the Nuer advance. Others had perished from the famine which
followed, and all suffered the loss of cattle. At the end of the century, the Anywaa appeared
near extinction. They were saved by a technological revolution’.
This ‘technological revolution’ was the acquisition of firearms through the ivory
trade with imperial Ethiopia. As Bahru (1976: 112) noted, ‘About 1911, the total
number of rifles in Anywaa possession was estimated at between 10,000 and 25,000’.4
In 1911, the Anywaa, led by nyiya Akwei of the Adongo region, launched their famous
raids against the Nuer in the Akobo region.5 In the first raid, the forces led by Akwei
killed thirty-five Nuer and captured forty cattle; in the second raid, five Nuer were
killed, three hundred cattle were captured and a hundred women and children taken
as captives; in the third raid, two thousand cattle were captured (Bahru 1976: 116–17).
Akwei’s forces raided both the Lou and the Jikany Nuer and penetrated as far west as
the Bahr el-Zaraf (Collins 1971: 203–04). In 1912, kwaaro Adiu Nyigwo of Edeni
village organized a massive attack on the Gaat-Jak Nuer (Jal 1987: 269–70). Initially
Adiu and his forces managed to drive all the Gaat-Jak from Gambella. The Gaat-Jak
struck back and at the Battle of Biot (the present-day Jikaw district in southern Sudan),
Adiu was killed. Since then, Edeni village has been under Nuer cultural and political
influence and was renamed Majak by the Nuer. The Battle of Biot is the historical
context within which the Tier Agak song (analysed in the previous chapter) is situated.
The rise of Anywaa military power threatened the British and the Ethiopian state
interests in the region. In the early decades of the twentieth century both govern-
ments had multiple economic and diplomatic stakes in Gambella.6 The British were
the first to try to contain the Anywaa. With vital strategic interests to protect and an
imagined economic eldorado to pursue, the British were increasingly nervous about
the rise of Anywaa military power. ‘Disarm the Anywaa’ was the British political pre-
occupation in the 1910s and 1920s. In 1912, the British carried out a military cam-
paign against the Anywaa of the Adongo region. Although the Anywaa were no
match for the British in conventional battles, their guerrilla war inflicted heavy
damage on the British military. In the confrontation with nyiya Akwei, the British
lost four commissioned and thirty-seven non-commissioned officers (Bahru 1976:
120). The Anywaa also captured firearms from the British (Zerai 1971: 9).

4. According to Johnson (1986: 34), around 1912 a large elephant tusk could be exchanged for
ten rifles. The types of rifles used during this period were senadir (Remington), wujigra (Fusil
Gras), mascob (Russian gun) and muzzle-loader (Kurimoto 1992: 13).
5. Akwei Cham was the most powerful of the nyiye. Thanks to the abundance of elephants in the
Adongo region, as well as his ability to recruit shifta (bandits) from the highlands and defec-
tors from the Sudan army, nyiya Akwei built a very powerful army that successfully contested
regional power with the Ethiopian and British governments as well as with Anywaa’s pastoral-
ist neighbours (Bahru 1976: 109–10).
6. The political and economic significance of the Gambella region to Ethiopia in the early twen-
tieth century is evident in Emperor Menelik’s rejection of the British proposal to exchange ter-
ritories, first, Zeila in British Somaliland for the Baro salient (Bahru 1976: 112) and, later, the
British-controlled Illemi triangle at the junction of the Sudan, Kenya and Ethiopia for the Baro
salient (Collins 1983: 104).
124 Playing Different Games

Embittered by the humiliating defeat, the British determined to resolve ‘the Anywaa
problem’. To that end, they launched a diplomatic offensive to corner the Anywaa by
arguing for a joint military operation with the Ethiopian government.7 The plan did
not materialize except for a reconnaissance trip along the Gilo River. The Anywaa
problem assumed such a high political profile that it provoked a parliamentary
debate in London on whether the Anywaa country was worth the trouble (Bahru
1976: 128).
Imperial Ethiopia was initially hesitant to curb Anywaa military power. It preferred
to pursue a non-confrontational approach towards the Anywaa in order to create a buffer
between itself and the British, but more importantly it had a stake in the ivory–gun trade.
As Bahru (1976: 131) noted, ‘a vigorous policy of disarming the Anywaa would have
been tantamount to financial suicide’. However, as the political influence and military
muscle of the Anywaa leaders grew, political measures were taken against the Anywaa by
imperial Ethiopia. Both nyiya Udiel and nyiya Akwei were imprisoned briefly to curb
their growing military power and political influence (Bahru 1976: 111; Johnson 1986:
226). In 1913, the Anywaa demonstrated their political insubordination by killing Lij
Kasa, the imperial agent of Gambella (Zerai 1971: 24). The death of Lij Kasa provoked
a strong punitive campaign by the Ethiopian state. Four thousand spearmen and one
thousand riflemen were sent under the command of fitawrari (Commander of the
Vanguard) Solomon, the son of dejazmach (Commander of the Gate) Jote Tulu, gover-
nor of Sayyo, western Wellega. Fitawrari Solomon lost more than one hundred of his fol-
lowers, further boosting the spirit of Anywaa resistance (Zerai 1971: 25). Subsequent to
their military victory, the forces of Akwei Cham managed to take control over the trade
routes along the banks of the Baro and Akobo Rivers, as far as Gambella town, Bure and
Dembidolo in the highland, taxing all commercial goods that were coming through these
areas (Zerai 1971: 10). The British officials in the enclave appealed to the Ethiopian gov-
ernment. Nyiya Akwei gave an ultimatum to the Highlanders in Gambella to be confined
to an area within a kilometre of the town.
In 1914, the ambitious qenyazmach (Commander of the Right) Majid Abud, a
Druze Syrian in the service of the imperial Ethiopian government, was sent to force the
Anywaa into submission. In 1916, qenyazmach Majid launched a major military offen-
sive against the Anywaa in what came to be known as the Battle of Itang. The Anywaa
put up strong resistance but ultimately succumbed to the forces of qenyazmach Majid –
they lost five hundred and thirty-two men, and five hundred men were castrated (Zerai
1971: 26). As Bahru (1976: 142) described it, ‘in a sordid feat of carnage, he [Majid]
asserted [Ethiopian] government authority’. The Anywaa nevertheless continued their
resistance to the British colonial state and the Ethiopian government, as well as their
counter-offensives against their once powerful pastoralist neighbours, the Nuer and the
Murle. In 1931 the Anywaa once again raided the Gaat-Jak Nuer, and in 1932 they
launched two extensive raids on the Murle in the Akobo area. The Anywaa killed
twenty-seven Murle and captured eighty women and children and eight hundred head
of cattle (Bahru 1976: 156). The Anywaa escaped British reprisals by crossing the inter-

7. The British went as far as employing the service of the Abun, the Egyptian Patriarch of the
Ethiopian Orthodox Church, to lobby the Ethiopian state to join the campaign (Bahru 1976: 125).
 Differential Incorporation into the Ethiopian State 125

national boundary into Ethiopia. These Anywaa raids provoked a major diplomatic
crisis. The British demanded that the Ethiopian state govern the Anywaa and also pay
compensation for the losses. As a result, qenyazmach Majid undertook two major cam-
paigns against the Anywaa in the Akobo region during 1932–34 in order to reassert
Ethiopian government authority. The Anywaa call these campaigns laegnmajid, the wars
of Majid (Ojullu 1987: 42). They tenaciously resisted Majid, and in 1934 his forces were
annihilated by the Openo Anywaa in Pol village, and Majid himself sustained serious
injury (Zerai 1971: 32). After a series of subsequent military campaigns, however, the
Anywaa were finally subdued, and entered into a long military and political decline.
The corollary to imperial Ethiopia’s campaigns against the Anywaa was the com-
mencement of the slave trade, which left a permanent stigma on the peoples of the
borderland who were forcibly integrated into the Ethiopian state. The pejorative term
bariya (slave) originates in the borderland people’s experience of slavery. Gambella was
one of the main sources of slaves in south-western Ethiopia. B.H. Jessen, a British
traveller who visited south-western Ethiopia in 1904, described the plight of the
Anywaa as follows: ‘The Abyssinians [Highlanders], though officially their protectors,
make yearly raids on them, ostensibly to collect their tribute, but incidentally taking
away boys or women for slaves … The Nuer on the other side make inroads on their
land, in order to gain larger pasture-grounds for their cattle’ (Jessen 1905: 163). Apart
from the cultural differences used to justify slavery, the general characteristics of a slave
were dark skin and very short and curly hair. On the basis of oral history, Birhanu doc-
umented the inhumane practices of slavery and the slave trade in western Ethiopia:

After their capture, slaves were beaten and roped together, and gags put in their
mouth to prevent them from making loud noises. Their legs were also tied
tightly so that they wouldn’t escape, or run away. [In the markets] they sat in
rows on stones in what was called daga garba and [the slave quarter] was called
dabigarba. Their faces were painted with butter and a type of grass called soso was
put around their necks…to make them look healthy. If their skins were not dark
enough, they would be warmed beside a fire for long time in order to change the
pigment of their skin before taking them to the market (Birhanu 1973: 14).

The political status of the Anywaa was no better during the brief Italian occupation
of Ethiopia (1936–41) prior to and during the first two years of the Second World
War. The Italians supported the Nuer for strategic reasons (Collins 1983: 386). The
Ethiopia–Sudanese border was one of the theatres of the Second World War. The
Italians’ plan to attack the British in the Sudan counted on Nuer ‘heroism’. The
Anywaa political leaders were targeted and deposed by the Italians, who then actively
supported the Nuer in their fight against the Anywaa in Jor district.8

8. Evans-Prichard (1947) reported on the Italian involvement in the Nuer–Anywaa conflict in Jor
district with their policy of supporting the Gaat-Jak and Gaat-Guang tribes’ claim over
Anywaa lands. He also reported that the Italians deposed the nyibur (representative of a
kwaaro) of Pinyudo village, one of the largest Anywaa settlements along the Gilo River (Evans-
Pritchard 1947: 72–73).
126 Playing Different Games

Imperial Ethiopian authority over the Anywaa was further consolidated in the
post-liberation period. Police stations were established to enforce this authority and
state demands were changed from tribute to tax. As dimui was the traditional cur-
rency, the Anywaa could not meet the tax demands in monetary terms. Some
Anywaa responded to the demand through seasonal wage labour on the coffee farms
of the neighbouring highlands, particularly in western Wellega. Others responded
through military resistance. By the 1950s the Anywaa were once again politically
restive. During 1952–58, they raided police stations in Gog, Itang and Pokumu dis-
tricts. These raids posed a serious threat to the Ethiopian government, as they did to
the British commercial establishment in Gambella.9 The state of rebellion lasted until
1960, when it was finally quelled by an army sent from Gore, the provincial capital.
Pokumu village, the centre of Anywaa resistance, was burned to the ground by the
Ethiopian government troops and renamed Birhaneselam, which in the Amharic lan-
guage means ‘light of peace’. The Anywaa spirit of resistance spread into other dis-
tricts such as Akobo and Gog in the late 1960s. The main items of contention
between the Anywaa and imperial Ethiopia in the 1950s and 1960s were the legiti-
macy of state authority and the inability of the Anywaa to pay tax in cash because
their economy was non-monetized (Ojullu 1987: 43).
The double pressure from imperial Ethiopia and colonial Sudan greatly affected the
viability of Anywaa society. Politically their ‘enlightened’ nyiye were reduced to mere
collaborators. Some Anywaa kwaari and nyiye were made balabats (local imperial offi-
cials), but real political power was exercised in Gambella throughout the imperial
period by officials sent by the central government who, in the eyes of the Anywaa and
the Nuer, ‘belonged’ to the category of Highlanders. In fact, the people of Gambella
were referred to at the national level as lemma, named after the imperial governor of
Gambella in the 1960s, General Lemma. The Anywaa and the Nuer were addressed as
‘People of Lemma’, as if they did not exist before the arrival of the general.
The appointment of General Lemma as the governor of Gambella was character-
istic of the way in which the Ethiopian state related to its periphery. Most imperial
governors of the peripheral regions were sent there as a form of ‘exile’. General Lemma
came to Gambella because of his participation in the failed 1960 coup against
Emperor Haile Selasssie. The choice of Gambella for his governorate served as a dual
vendetta against him. On a personal level, the general had problems with the gover-
nor of Illubabor province, Enquselassie, and his assignment as governor of Gambella
thus placed him under the authority of a rival. On a symbolic level, the general was
subjected to a different form of slight. Though he was an ethnic Amhara, he fell on
the black side of the colour spectrum. In the Ethiopia of the day, when the discourse
about ‘purity of race’ played a prominent role in national identification, associating
him with the ‘black’ Anywaa and Nuer was intended as symbolic violence against him.
The arrival of the imperial Ethiopian state in Gambella also heralded the eco-
nomic decline of the Anywaa. Until that time, there had been reciprocal socio-eco-

9. ‘Security Situation in Gambella’: a report by the British officer of the enclave to the adminis-
trator of Gambella, 13 February 1952, Metu Archives.
 Differential Incorporation into the Ethiopian State 127

nomic exchanges between the Anywaa and the Oromo of the neighbouring high-
lands. The main trading items were cotton from the Anywaa side and beads and
grain from the Oromo (Kurimoto 1992: 14). The masculine ritual of buffalo killing
by Oromo men in the lowlands of Gambella also introduced a nascent form of pos-
itive social integration. The Oromo and the Anywaa hunted the buffalo together, the
former taking the horns and the skin, the latter taking the meat. With the establish-
ment of the enclave, cotton, one of the main Anywaa export items to the highlands,
became redundant. Cotton goods constituted the most important import item from
the Sudan, amounting to 15,029 out of a total of 27,962 pounds sterling in 1911
(Bahru 1976: 253). The military campaigns also caused the farmlands to be left unat-
tended; as a result, the Anywaa started to buy grain from the highlands, a trend
which ultimately led to the change in their position in the regional economy from
subsistence producers to consumers of highland products. Socially, the Anywaa
assumed a ‘spoiled identity’, reduced from luo to bariya and lemma. Above all, with
the political and military decline, the Anywaa villages were no longer able to contain
territorial encroachments by their pastoralist neighbours. In fact, it was this decline
that undermined negotiated access to resources between the Anywaa and their neigh-
bours, with the latter being increasingly encouraged to adopt violence as part of a
strategy of resource access.

The Nuer under imperial Ethiopia

Unlike the Anywaa, the Nuer occupied remote areas where imperial Ethiopia lacked
the administrative resources to govern them. Besides, imperial Ethiopia and colonial
Britain were vying to win over the political loyalty of the Nuer on economic grounds
(cattle wealth). Who should tax which Nuer section was a serious bone of contention
between the two states (Johnson 1986: 230). In fact, ‘there was a brisk struggle for
Nuer allegiance between the respective frontier agents of the two governments’ (Bahru
1976: 163). This created a wider political agency for the Nuer to manipulate alterna-
tive centres of power by instrumentalizing the international boundary because the
militarization of the Anywaa posed a threat to both states. As Anywaa power waned,
Nuer power waxed. Acutely aware of the new military might of the Anywaa, the Nuer
caught up with them in the local arms race (Bahru 1976: 146). After the Battle of Biot
in 1912, the Nuer began to establish trade networks with the neighbouring Ethiopian
highland towns, exchanging cattle for firearms. It was virtually impossible to secure
arms from the British, who sought to maintain the ‘monopoly of force’, while
Ethiopian imperial power was constructed through a strategic co-option of local
power structures provided that they remained loyal (Johnson 1986: 228–31).
Prior to the Italian occupation, the British toyed with the idea of a quid pro quo
with the Ethiopian government in order to claim all the Nuer and their lands and
enforce a ‘Pax-Britannica’ along the frontier (Collins 1971: 104). In fact, the British
proposed to redraw the 1902 border through an exchange of their territory, the
Illemi triangle, a triangular piece of land joining Sudan, Kenya and Ethiopia, with
the Gambella region of Ethiopia (Collins 1983: 390). When this failed, the British
pressed for a grazing agreement in 1933, in which they were prepared to pay fees for
the use of the Gaat-Jak Nuer dry-season grazing lands inside Ethiopia (Johnston
128 Playing Different Games

1986: 230). The British viewed this as a first step that ultimately would lead to an
exchange of territories, but by 1935 British frontier officials realized that the idea of
territorial exchange was not feasible, with imperial Ethiopia succumbing to fascist
Italy a year later.
As the colonial and imperial states vied for the grazing and taxing rights of the
Nuer, the Nuer learned that there were multiple claimants to the land, and that the
Anywaa’s land rights claim was just one of many. In the 1930s, Koryom Tut, a
Thiang Nuer from Kurtony in Jikaw district, became the first Nuer to attain the
high-ranking imperial title of fitawrari (Collins 1983: 378).10 Koryom, in league with
qenyazmach Majid, encouraged the permanent settlement of sections of the Nuer
who had cross-border settlements.
The Nuer were better integrated into the Italian colonial establishment in the
Gambella region than were the Anywaa. As part of the Second World War military
strategy, the Italians sought to play the Nuer ‘card’ against the British East African
colonies, employing the Nuer ‘as the shock troops of a new Italian African army’
(Collins 1983: 390). The Italians’ preference for the Nuer was based on the latter’s
numerical advantage, their larger settlements in the Sudan and their ‘warrior’ iden-
tity. Italian Nuer policy was articulated by Major Colacino, Italian commander in the
Gambella region:

It is necessary to protect and cherish our Nuer as well as the Sudanese Nuer.
It is necessary to carry out this policy … so that it will keep alive in the Nuer
the lighted torch of sympathy towards Italy with their political future in the
hands of God and our Duce. Involved in a war with the English we should
have the sympathy of a quarter million Nuer on our frontier to safely advance
into enemy territory. We should enrol under our banner thousands and thou-
sands of these magnificent Nuer … warriors at heart, frugal, dignified, solid,
faithful, and grateful. (Report by Major Colacino, 23 February 1940, quoted
in Collins 1983: 386)

The Italians, on the other hand, perceived the Anywaa as ‘unreliable’ (Collins 1983:
386). Some groups of Nuer instrumentalized the Italian attraction to them to gain
access to the Anywaa riverine lands. This is certainly true for the Cieng Nyajani, who
appealed to the Italians for military support in their fight against the Jor Anywaa
(Ojullu 1987: 43). Gambella was occupied by the combined forces of the British
East African army and the Belgian Congolese soldiers on 22 March 1941 (Collins
1983: 400). As part of the Occupied Enemy Territory Administration (OETA), the
British exercised power over Gambella and sought to annex it to the Sudan. The guilt
associated with the British policy of appeasing fascist Italy, which included the ‘sac-
rification’ of Ethiopia, made the military occupation of Ethiopia morally unattain-
able. After four months of rule under the OETA, Ethiopia regained sovereignty over

10. Qenyazmach Koryom Tut travelled to Addis Ababa where he was hosted and given presents by
the emperor so that he could serve as the agent of the Ethiopian government in the Nuer-
inhabited areas of Gambella.
 Differential Incorporation into the Ethiopian State 129

the Gambella region (Collins 1983: 403). Post-liberation Ethiopia continued its
policy of political integration of the Nuer in its diplomatic competition with the
British. More Nuer imperial title holders appeared in the 1950s and 1960s, creating
new spaces for individual Nuers to assume formal leadership positions, whereas the
traditional power bases of the Anywaa leaders were weakened.11
The political autonomy of the Nuer, however, also meant a benign neglect,
leading to their marginalization in the distribution of state-mediated goods and serv-
ices, particularly access to education. Indeed, the pattern of incorporation of the
Anywaa and the Nuer into the imperial Ethiopian state was affected by this differen-
tial access to modern education, which created an imbalance among the newly edu-
cated power elites. The first school in Gambella town was established in 1942. At
that time Gambella town was predominantly Anywaa, with no Nuer residents except
for intermittent cattle traders. In 1952 the American Presbyterian Church opened a
better-equipped mission school at the Anywaa village of Akedo on the Baro River. In
1960, a second mission school was opened in the Anywaa settlement at Gilo. It was
not until 1962 that the Nuer had their first modern school, established by the mis-
sionaries in the village of Adura. Five additional government schools were established
in Anywaa and Nuer areas in 1964. The few Nuer who attended elementary school
at the mission station and government schools went to Juba, the capital of southern
Sudan, to continue their high school education or to seek employment opportuni-
ties, particularly after the establishment of the autonomous Southern Sudanese
Regional State in 1972. When the revolution broke out in Ethiopia in 1974, there
were more educated Anywaa who were ready and willing to participate in the state
than there were Nuer.
The imperial political space in Gambella, however, was dominated by the
Highlanders. All the imperial administrators were ethnic Amhara or Oromo, two of
the main groups who constitute the category of Highlanders. These were not resi-
dent Highlanders of Gambella but rather Highlanders from the capital. Nonetheless,
in the eyes of the Anywaa and the Nuer, they fell readily into the general category of
Highlanders. The imperial regime was Orthodox Christian in its belief, Amharic in
its tongue, Highlander in its geographic base, and ‘red’ in its colour, hardly capable
of carrying out what Geertz (1963) called the ‘integrative revolution’, particularly in
border regions such as Gambella where none of its features had local resonance. The
Anywaa and the Nuer were thus not incorporated into an ideal and integrative state,
but rather into a state represented by and identified with certain categories of people.
It is for this reason that the Anywaa and the Nuer referred to the Ethiopian state with

11. Apart from one instance, the imperial administration received no major political or military
resistance from the Nuer, in contrast to the Anywaa, who resisted strongly throughout the first
half of the twentieth century. For a short period of time, the vigorous extension of imperial
authority in the 1910s caused Nuer–Highlander relations to become tense. In June 1912, the
Gaat-Jak Nuer clashed with dejazmach Gorfu, governor of the neighbouring highland district of
Gidami, who raided their cattle and killed more than a hundred. Dejazmach Gorfu himself sus-
tained heavy casualties. The hostility was brief. Dejazmach Gorfu was replaced by a new imperial
leader who resumed the economic exchanges: the cattle for gun trade (Bahru 1976: 146–47).
130 Playing Different Games

the same terms they used to refer to the Highlanders as a category of people, namely,
gaala and buny, respectively. The Ethiopian state, however the political regimes have
differed, has been perceived by the Anywaa and the Nuer in the same racial and
ethnic terms ever since.

The Anywaa, the Nuer and Socialist Ethiopia

In September 1974, the imperial regime was overthrown by a popular uprising and
replaced by a ‘socialist’ regime that lasted until 1991. The socialist regime, popularly
known as the Derg, was characterized as ‘garrison socialism’ (Markakis 1987: 202),
denoting its military background and Marxist ideological orientation. Clapham
defined the nature of the Derg regime as ‘Jacobin’, including a project which he refers
to as encadrement: ‘It amounted to a project of encadrement, or incorporation into
structures of control, which was pursued with remarkable speed and ruthlessness. It
sought to intensify the longstanding trajectory of centralized state formation by
removing the perceived sources of peripheral discontent and espousing an ideal of
nation-statehood in which citizens would equally be associated with, and subjected
to an omnipotent state’ (Clapham 2002: 14).
In a centralized one-party system, the Derg sought to ‘remap’ Ethiopia along mod-
ernist lines (James et al. 2002). The process of encadrement was pursued in earnest in
social, economic, political and cultural fields. Above all, the Derg carried out social
engineering in the context of a ‘high modernist’ project (Scott 1998: 4), known in
Ethiopia as zemecha (‘Development through Cooperation Campaign’). Under this
campaign, students from urban centres were sent to revolutionize the countryside by
liberating the rural masses not only from the yoke of the ancien régime but also from
the ‘tyranny’ of local tradition (Donham 1999: 29–35). This was to be achieved
through the Ethiopian version of the Chinese Cultural Revolution. Economically,
encadrement took the form of attempts to gain total control of the peasantry through
villagization and resettlement programmes. Fourteen million peasants were forced to
settle in new villages (Tadesse 2002: 117), and another half a million peasants were
resettled (Pankhurst 2002: 133) – classic examples of a state’s attempt to ‘capture’ the
peasantry in order to promote national development (Hyden 1980). In the political
arena, the project of encadrement meant not only a one-party system but also a violent
repression of other modes of governance, through the elimination of opposition
parties and the abolition of traditional authorities.

The Anywaa under the Derg

Various groups of people experienced this project of encadrement at differing levels.
The Anywaa were politically visible because they had a relatively higher form of cen-
tralized political system (the nobles and the headmen). Both the nyiye and kwaari
were labelled ‘reactionary’, ‘anti-revolutionary’ and ‘feudal’, as if they were an impe-
rial system writ small. Anywaa traditional leaders certainly had privileges of office,
but they hardly qualified as the kind of strong authority that the term ‘feudal’
implies. Among their privileges, they had exclusive rights over some hunting tro-
phies; they did not observe wudo, the ceremonial respect for various categories of rel-
 Differential Incorporation into the Ethiopian State 131

atives observed by ordinary men; and they did not appear barefoot because the
removal of sandals was a sign of inferiority in status. The nyiye and the kwaari were
greeted with gungi, the ceremonial low-bowing posture adopted when approaching
them. Nevertheless, as Evans-Prichard (1940b: 137) remarked on the ritual kingship
of the Anywaa, ‘it is kingship, not kings, which is sacred’. Lienhardt (1957) also
described the Anywaa headmen in a similar way: the courtesies were to the office, not
to the incumbents.
The avant-garde of the cultural revolution were the zemach – high school and
university students who ‘would be sent, like an army, to reconquer the country side
… not with force, but with knowledge’ (Donham 1999: 29). On the basis of this
‘enlightenment’ project, five hundred zemach were assigned to the districts of
Gambella, Itang, Abobo, Gog and Jikaw.12 The zemach campaigned against the bal-
abats (generic term for imperial office holders). Both the Anywaa and the Nuer had
balabats, but attacks on the Anywaa balabats resounded like an attack on Anywaa
society because it was the kwaari and the nyiye who had been converted into impe-
rial agents, whereas the Nuer balabats were largely self-made and more of an individ-
ual project. The zemach taught ordinary Anywaa men and women about the
existence of social classes among the Anywaa. The Anywaa traditional currency
(dimui) was also forcibly monetized and labelled a ‘backward’ cultural practice. A
monetized system of bridewealth was imposed. Some of the revolutionary cadres
confiscated the dimui of various villages and threw them into the river, while others
profited by smuggling them to the Sudanese Anywaa. Other ‘backward’ cultural
practices such as naak (dental evulsions) and utak (the beating of a husband by his
in-laws for failing to take care of his wife) were banned.
For all practical purposes, however, the cultural revolution meant the articulation
of a particular culture in the name of progress. The Derg reinforced the cultural
hegemony of the Highlanders (the ‘national’ culture) in Gambella, for the Derg was
still perceived as gaala or buny. Despite their revolutionary rhetoric, the Derg drew
on the cultural infrastructure of the ancien régime. For the Anywaa, therefore, being
‘not backward’ and becoming ‘modern’, meant participation in a particular
(Highlander) culture thinly disguised as ‘progressive’. The Anywaa bore the brunt of
the cultural revolution as they were near the administrative centres and made visible
by their political system.13 These impositions and state encroachment into the
Anywaa cultural world provoked a discontent which was reflected in the popular
song that captures the spirit of the time:

12. The Jikaw post was cancelled for security reasons, and the Gog-Jor post was also transferred to
Abobo because of transportation problems. As a result, more than four hundred students in
Gambella demonstrated against being deprived of their right to participate as zemach. In the
end, only the Gambella and Itang posts were operational.
13. An Anywaa village on the Baro River, Nyikwo, was selected by REYA (Revolutionary
Ethiopian Youth Association) as a model village. Nyikwo village was visited by Hailu Tujuba,
the chairman of REYA, in 1978.
132 Playing Different Games

Song in Anywaa Translation in English

Paap beete lääy ni wär jøwi wí juuru wäri As the forest is recognized as the home of
ni rany paaní këëdö aa, Dërgi ba bwøk the animals we fight against those [Derg]
who come to take our home

Ngøøm löö keeri pøø timø ni kare caarø The land will sue you

Böö carru Opääl o Lwal ni bang pogi Joshua [the Nuer administrator of
keeta Ajuna ngø no dööng ni bang pogi, Gambella who had an Anywaa back-
jöör pogi nyoodhe, ground] is a man without land, show
him where he belongs

Cänga tïme ni räny pinynyi ngati man Everybody is leaving his own land to be
kamo pogi o cïppe ji ciik mo nyään, part of a new system

Ajur man Okäla kaa ni rwänge ni kwua Where have all these foreigners come
bare ngwieny wäägi gïï leth pøthgi from? They do not recognize the
pøththø jaak problem they are creating for us.

The Anywaa discontent ultimately led to a rebellion, organized in the remote village
of Utwol in Jor district. The abolishment of dimui provided a cultural justification
for the nyiye and kwaari, who fought not only to retain their political power but also
in defence of ‘tradition’. The new ‘revolutionary’ socio-political order was referred to
by the nyiye and the kwaari as kwec gel, leadership legitimated by acquisition of
money, in which accountability is not to the people governed but to the ‘paymasters’,
a reference to the salaries the new leaders received from the government. The new
leaders elected by the Derg, the liqemembers of the peasant associations, were con-
trasted with the kwaari and the nyiye, who drew on tradition and patrimony to legit-
imize their power. The abolishment of dimui was perceived as socially disruptive. In
fact, the increasing incidence of divorce in the 1970s was attributed to the sudden
and forceful shift from dimui to money. Unlike dimui, which was difficult to repay
in cases of divorce because a brother or another relative could have already married
using the sister’s bridewealth dimui, money could be paid back relatively easily.
Contemporary Anywaa also relate the beginning of prostitution among the Anywaa
to the introduction of money as a medium of social and economic exchange.
Dissatisfaction with the new social order was described to me by a member of a
kwaaro family:

Educated Anywaa and some gaala [Highlanders] came and told us that the
problem was because we, the kwaari, got power through family heritage.
They said the new system is for all, that anybody can be a leader. They said,
‘You kwaari, you do not know how to administer, give the power to us’. We
could not say anything. It was like somebody trying to show you how to
make porridge which you know already. Finally you realize that their porridge
is no better than the one you already know. You know that this way of doing
things [imposition] does not work but you cannot do anything, for it is not
you who made the porridge in the first place. What can you say? We gave
 Differential Incorporation into the Ethiopian State 133

them the power but they did not bring us the better life they promised us.
Yours [referring to the author’s generation] is the world of money. We [the
elders] just see you using money. It has not brought any change. If you do not
have money, you are nobody. It was not like that during the kwaari. The
kwaari did not discriminate against anybody, poor or rich. If you did not have
dimui, you would just go to a kwaaro and ask him to help you marry. Yours
is a tiqim [benefit] system. (Omot Uchala, brother of the kwaaro of Gog, 26
December 2000)

The Anywaa rebellion against the Derg was led by kwaaro Omot. In 1978, Omot
chased the teachers (Highlanders) from the adult education centre in Jor and burned
all the books. The vice-administrator of Jor district (a Nuer) was also forced to flee. In
1979, ten ‘revolutionary guards’, as the armed Derg cadres were called, were killed and
seven were wounded. The rebellion spread into other areas. In 1980, three more
Highlander teachers were killed in Cham village (Jikaw district) and the cooperative
shop in Pukumu village was robbed and more schools were burned. The rebels took
control of eight villages in Jor district and made contact with the Sudanese Anywaa.14
Nyiya Agada Akwei in Pochalla district in southern Sudan was active in giving moral
and material support to the rebels.15 Emboldened by this, the rebels caused more
damage to government forces, killing twenty-nine and wounding thirty. In February
1982, the Derg organized a large-scale military campaign to put down the Anywaa
rebellion, and after a fierce battle, the defence of ‘tradition’ was broken. The campaign
ended in a symbolic act. One of the rebel leaders, kwaaro Batade Ulaw, was beheaded
and the ‘political trophy’ taken to the district’s capital (Gog) for public display.
The Derg were surprised to witness resistance from a minority group who it
claimed to represent in the popular revolution against the imperial regime. It had
expected enthusiastic support for the abolition of the ‘oppressive’ traditional politi-
cal system. In order to ‘liberate’ the broad masses from yetesasate niqatehilina (the
Amharic translation of the Marxist notion of ‘false consciousness’), Highlander peas-
ants were ‘imported’ to help create awareness among the Anywaa farmers and per-
suade them to rise up against their leaders. This revolutionary discourse can be
elicited from the ‘History of the Broad Masses’ manuals that the Derg cadres pro-
duced for the various peoples of Ethiopia. In the ‘History of the Broad Masses of
Gambella Awraja’, Anywaa resistance against the revolution is described thus:

Although there was exploitation by the local balabats, the broad masses were
not aware of its existence. Besides, there were no mechanisms which could
have served as an outlet to vent grievances. That was why the peasants strug-
gled to restore the balabats to power. (Author’s translation from the docu-
ment titled ‘Ye Gambella Awaraja Sefi Hizb Tarik’, ‘History of the Broad
Masses of the Gambella District’, 1977: 18)

14. These villages were Ulawo, Gony, Angela, Amedho, Iranga, Pakang, Ulang and Ujalo.
15. Interview with nyiya Adongo Agada, son of nyiya Agada, Utalo village, April 2001.
134 Playing Different Games

The manual further describes how difficult it was to convince the Anywaa to be
revolutionary:

Because the district was large and the people were few, there was no shortage
of land. The technique of production was primitive, but this did not create
land shortage. Because their rulers were not capable of or interested in surplus
production, there was little economic exploitation on the scale we have in
other parts of Ethiopia. It was also difficult to convince them of the benefits
of the land proclamation act. In order that the people of Gambella would
become aware of the exploitation and take up arms against the balabats,
ninety peasants were brought to Gambella from the highlands. It took three
years to prepare the people of Gambella to start the struggle. (ibid.)

On the basis of the nature of its incorporation into the Ethiopian state (namely, the
context of diplomatic rivalry and political competition between imperial Ethiopia
and colonial Britain), coupled with the lack of economic incentives because of the
unattractiveness of the lowland plains, the Gambella region was spared from land
confiscation such as that which occurred in the newly conquered regions in the
southern and western parts of Ethiopia during the imperial period. Towards the end
of the 1960s, the Amhara local governors appealed to the central government for
land in Gambella, either as a gift or through purchase. By the time the revolution
broke out, sixty noble families had acquired land for large plantations, but none of
them had started to develop it.
The cultural revolution received a mixed reception among the Anywaa. While
the Anywaa who had a stronger stake in tradition (kwaari, nyiye and the elderly) felt
threatened, others, particularly Anywaa youth, viewed it as an opportunity, because
the abolishment of the ancien régime meant not only new political space but also
positions of leadership in the new government bureaucracy as opposed to the hered-
itary mode of governance. With the ban on dimui, the youth felt ‘liberated’ from the
elders as this allowed them more freedom of marriage. In the mid-1970s, there were
many sisterless bachelors (bouth) who were unable to marry because of the scarcity
of dimui. The plight of these bouth is recounted in a popular song of the time: ‘If I
do not have a sister, I go to Dambala’. Dambala is a gold mining centre on the upper
Akobo River, where many bouth went to earn enough to pay the requisite
bridewealth.16 Others went to the highlands as wage labourers to work on the coffee
farms. The youth therefore welcomed the revolution now that their social advance-
ment was not governed by either the scarcity of dimui or the authority of the elders.
With the monetization of bridewealth, it became possible for young Anywaa to
hasten their social advancement as long as they could afford to pay for their wives.
Above all, the revolution was viewed as an opportunity by certain Anywaa to par-
ticipate in ‘progress’. The Derg ideology of Ityopiya tikdem (‘Ethiopia first’) appealed

16. According to Kurimoto (1996: 49), it takes a week on foot to reach Dambala from Gambella
town. In the mid-1980s, there were more than three thousand men permanently engaged in
gold panning.
 Differential Incorporation into the Ethiopian State 135

to the first generation of mission-educated Anywaa, who saw in the revolution an

opportunity to take their own people along this road. A significant number of edu-
cated Anywaa joined the revolutionary camp. Many of them were sent to Addis
Ababa and Eastern Europe for political training. Anywaa elites actively campaigned
against traditional dietary practices, which they believed provided conspicuous ‘evi-
dence’ of the backwardness of the Anywaa. An increasing number of Anywaa became
culturally competent in the Highlander culture.17
There was also a gender dimension to the Anywaa reaction to the revolution.
Anywaa women were receptive to the revolutionary rhetoric of chiqona (exploita-
tion). Commenting on the spirit of the time, an Anywaa woman from Pijwo village
(Jikaw) proudly described the revolution in this way: ‘We [the women] deposed the
kwaari and we can now talk freely in front of men.’ She refers here to the gender
inequality in Anywaa society, still expressed in the code of honouring men, in which
women approach their husbands on their knees.
The building of infrastructure, particularly the construction of the highly prized
Baro Bridge in Gambella town, the opening of schools and the employment oppor-
tunities in government bureaucracy were received as a welcome gesture by ordinary
Anywaa men and women, who hitherto had experienced incorporation into the
Ethiopian state as a form of stigma characterized by political exclusion and economic
marginalization. Commenting on the prevailing Anywaa mood of the period,
Kurimoto (2005: 341–42) writes as follows: ‘For the first time in history the Anywaa
were fully integrated into the Ethiopian rule … Generally speaking the local people
welcomed the development of infrastructure and education, appreciating the fact
that they were enjoying more opportunities than ever before’. For a brief period, at
least, relations between the Ethiopian state and one of its historic minorities were
expressed positively in kinship terms. The head of the Derg regime, Mengistu
Hailemariam, was addressed as Wora Ariat, the son of Ariat (Ariat was the name
given by the Anywaa to the firstborn daughter). This sense of kinship with Mengistu
echoed his self-portrayal as the black leader in the national public sphere. Although
Mengistu’s autobiography mentions his Amhara and Oromo ethnic origins, political
resentment to his brutal military dictatorship was often framed by the general public
in ‘racial’ terms. Like the Anywaa and the Nuer, he too fell on the black side of the
colour spectrum in the discourse about Ethiopian national identity. To what extent
Mengistu played with such imagery is hard to ascertain, but for the Anywaa his fre-
quent official visits to the strategically important Gambella were physical proof of his
‘connections’ with the people of Gambella.
However, high expectations soon bred disappointment as the Anywaa elites found
themselves competing with the ‘upstart’ Nuer elites. The first generation of educated

17. The highest manifestation of this cultural competence was eating tiresega (raw meat), hitherto
taken as a sign of the Highlanders’ own backwardness. In fact, in their early encounters with
the Highlanders, both the Anywaa and the Nuer were disgusted, and indeed frightened, by the
eating of raw meat, which appeared to them to be no less than ‘cannibalism’. Tiresega is one of
the three major dietary symbols of Ethiopian national identity; the others being injera (the
Ethiopian pancake-like bread) and the coffee ceremony.
136 Playing Different Games

Anywaa and Nuer started participating in regional power politics in 1978. At that
time Gambella was still a district within the highland province of Illubabor, and the
seat of the provincial capital was the highland town of Gore. A Highlander was the
provincial administrator, and he appointed an Anywaa and a Nuer as vice-administra-
tors of Gambella. In order to promote political integration after the 1976–77 bloody
conflict between Anywaa and Nuer in Jor district, Anywaa administrators were
assigned to Nuer districts and vice versa.18 The political careers of the vice-administra-
tors were intimately connected with the emerging symbolic ethnopolitics between the
Anywaa and the Nuer. Largely owing to his character and effectiveness in giving
service to the state, the Nuer administrator, Joshua Delual, rose to prominence,
whereas the Anywaa representative, Philip Opiu, showed relative weakness and was
ultimately transferred to a highland district at his own request.19 Without a competi-
tor, the Nuer vice-administrator built up his political base, and through him the polit-
ical standing of the Nuer was enhanced. The fact that Joshua was an ‘ex-Anywaa’ (Bär
Jingmir) was an additional factor that fuelled ethnopolitics. In him, the Anywaa saw
their status claim undermined. Joshua was later joined by another Nuer self-made
politician, Thowat Pal.20 Having demonstrated his efficiency in crushing the Jor rebel-
lion during the so-called cultural revolution, Thowat was appointed as the regional
security head. When the WPE (Workers’ Party of Ethiopia) was established in 1985,
Thowat was appointed regional party secretary. In 1987, the Derg introduced a
limited regional autonomy in response to the steadily growing ethno-national libera-
tion movements that demanded self-determination. Gambella was one of these newly
created administrative regions and Joshua was promoted as the administrator. With
the occupation of the two most senior political offices by Joshua and Thowat, the
Anywaa identification of the Nuer with the Derg regime solidified. This created
resentment among the Anywaa elites in two ways: first, there were more educated
Anywaa than Nuer; second, the Anywaa considered themselves to be the indigenous
people of Gambella and viewed the Nuer as foreigners and upstarts. Consequently, the
unprecedented rise of Nuer elites into political pre-eminence introduced the concep-
tion of first-comers/latecomers into the regional political debate.
Drawing on their settlement history and their relative competence in mainstream
Ethiopian national (Highlander) culture when compared to the Nuer, the Anywaa
had higher expectations of political advancement. Their educated elites found the
political advancement of the Nuer incomprehensible at best and a plot at worst.
Affronted by the ‘upstart’ Nuer, they tried to undermine the growing influence of the
Nuer elites through personal political networks. The Anywaa sent a delegate to the
office of the prime minister to bring to the attention of the central government the

18. Cham was the first Anywaa administrator of Jikaw district, followed by James Uthow. Of the
Nuer, Mark Chuol was assigned to Abobo and Thowat Pal to Gognajor district.
19. Many Anywaa admit, and in fact regret, the weakness of the Anywaa representative, whereas
Philip Opiu explains his transfer to the highland district of Sornageba (Illubabor) as a result of
the increasing insecurity he faced after the establishment of an Anywaa liberation front (Philip
Opiu, interviewed on 13 October 2000).
20. Thowat is the son of a Nuer imperial official, fitawrari Pal Chay.
 Differential Incorporation into the Ethiopian State 137

imbalance in power-sharing in the Gambella administrative region.21 Although this

resulted in the re-appointment of an Anywaa vice-administrator, their continued
subordinate political position bred strong discontent. By the early 1980s, the edu-
cated Anywaa also took up arms against the Derg regime, thus mending their differ-
ences with the ‘traditionalists’ whom they had been busy fighting only years before.
With this rapprochement between the ‘modernists’ and the ‘traditionalists’, Anywaa
discontent crystallized into the establishment of a liberation movement. In about
1980, Anywaa people who were living in Khartoum organized the Gambella
Liberation Front (GLF), which was renamed as the Gambella People’s Liberation
Movement (GPLM) in 1985 (Kurimoto 2005: 342).
Other projects of control by the Derg regime encroached on Anywaa territories
and damaged Anywaa ethnic sensibilities. In the wake of the Ethiopian famine of
1984 and 1985, the Derg organized a controversial resettlement scheme and planned
to relocate more than a million people into so-called land-abundant areas, particu-
larly in western Ethiopia.22 As part of this programme more than 60,000 peasants
from the northern and southern Ethiopia highlands (Highlanders) were forcibly
resettled in Gambella, thus increasing the population of the region by about at least
one third – previously, the total population of the region had been between 100,000
and 150,000 (Kurimoto 2005: 338), of whom, according to the 1984 census, about
28,000 were Anywaa. Four resettlement sites, designed exclusively for Highlanders,
and an integrated resettlement scheme were launched along the Baro and Gilo Rivers
and around Abobo, without first securing Anywaa consent (Kurimoto 2005: 339).
The task of the Anywaa was to cater to the needs of the Highlanders, which meant,
above all, the appropriation of some Anywaa lands and the exaction of corvée labour.
In the integrated resettlement scheme, the Anywaa were forced to join the
Highlanders and were organized into five kebeles.23 With regard to the impact on the
host peoples and the difficulties faced by the settlers, the resettlement programme
was a miserable failure. While a significant number of Highlanders did remain in
Gambella as permanent settlers, both Anywaa and Highlanders fled from the reset-
tlement villages in great numbers, going to the Sudan or, after the regime change in
1991, back to their home areas.
In Seeing Like a State, Scott (1998) uses the example of the villagization and reset-
tlement programme of the Derg to illustrate the follies of high modernism and to
show ‘how certain schemes to improve the human condition have failed’. According
to Scott (1998: 4), high modernism is characterized by ‘a strong, one might even say

21. This was done through an Anywaa-adopted son of Fikre Selasssie Wogderes, the Derg prime
minister. In 1984, when Mengistu Hailemariam visited Akedo village, an Anywaa boy offered to
translate from Anywaa into Amharic. Mengistu, head of the Derg and thus president, was
impressed by this Anywaa boy and he instructed the prime minister to educate him in Addis
Ababa. It was through this Anywaa that the political elites presented their grievances to the Derg.
22. More than a million people are said to have died because of this famine (de Waal 1997).
23. This was known as the Baro-Abol Integrated Resettlement Programme Abol Village. The total
population of the area after resettlement was ten thousand, of which thirty per cent were
Anywaa and seventy per cent Highlanders.
138 Playing Different Games

muscle-bound version of the self-confidence about scientific and technical progress,

the expansion of production, the growing satisfaction of human needs, and above all
the rational design of social order commensurate with the scientific understanding of
natural laws’. Similarly, Donham (2002: 35) describes the underlying premise of
Derg policies as ‘a kind of irrational faith in rationality, an ideological use of science’.
‘By the end of the 1980s’, Donham (2002: 35) continues, ‘the state had created new
villages laid out in ninety-degree grids for close to half of all rural cultivators’. Derg
social engineering, like so many other high modernist development projects, failed;
but what Fergusson (1990) termed ‘the effect of failure’ permeated different domains
of social life. In Gambella the resettlement programme was intimately related to the
emergence of a militant Anywaa ethnicity and the consolidation of their primordial-
ist form of ethnic identity.
In addition to giving Anywaa lands to the resettlers and causing ecological degra-
dation through large-scale deforestation, the resettlement programme created demo-
graphic anxiety among the Anywaa, who felt sandwiched between pastoral expansion
and state-sponsored resettled farmers.24 The new demographic imbalance had politi-
cal significance, for the Anywaa found themselves reduced to a minority in their own
land. For ordinary Anywaa men and women, the resettlement programme also
brought with it socio-economic decline. In resettlement areas, the Anywaa were
removed from their rivers, aquatic food sources and fruit trees. More importantly, the
Anywaa were reduced from being subsistence producers to being consumers of high-
land products and food from the refugee camps sold in the local markets. The reset-
tlement programme was connected to the state farms and refugee camps, whose
combined production drastically depressed the price of Anywaa agricultural prod-
ucts. Kurimoto (1996: 48–49) described his observation of the economic decline of
the Anywaa: ‘Between December 1988 and January 1989, I stayed for six weeks in
the vicinity of the refugee camp. There, a 90 kg of maize was available for only five
birr or even less’. As if this were not enough, additional grain was imported from yet
another state farm, Anger, in the neighbouring highland region of Wellega.
In short, the Anywaa could not protect themselves from state and market
encroachment, nor could they positively integrate into the new order (cf. Kurimoto
2005: 338). Because the Anywaa lacked entrepreneurial skills, the market economy
was destined to be dominated by the resettlers and refugees. Some Anywaa sought to
regain economic agency by adopting new income-generating activities such as distill-
ing local spirit (araqe), which they learned from the resettlers. But the resettlers still
enjoyed a comparative advantage in such activities, and the araqe produced by the
Anywaa was no match in quality for that of the resettlers. Consequently, the Anywaa
began to consume Highlanders products, rather than producing and exchanging their
own. This resulted in a long period of economic decline, culminating in the 1980s.
By this time, at the latest, all business activities in the Gambella region had come to
be dominated by the Highlanders, as those few Anywaa who strove to participate in

24. According to a 1999 UNICEF census of Abobo district, sixty-four per cent of the district’s
population of 31,700 were resettled Highlanders. Abobo district was one of the main resettle-
ment sites and a centre for state-mechanized farming.
 Differential Incorporation into the Ethiopian State 139

the business sector could not withstand competitive pressures from the Highlanders
or, for that matter, the disincentives that they experienced within their own society.
On a more fundamental level, and with longer-term effect, the resettlement programme
took a social toll on the Anywaa. Alcoholism spread in the villages, weakening both the
work ethic and productivity. Social stability in the villages was further undermined by an
increase in theft and by migration to the towns, which depopulated the villages and left
them more vulnerable to land encroachment by the Nuer. Educated Anywaa elites refer to
this social decline as the ‘four Ks’: kac (hunger), kwac (begging), kap (prostitution) and ku
(theft). The ‘four Ks’ are attributed to the arrival and expansion of the Ethiopian state, iden-
tified with and represented by the Highlanders (Kurimoto 2001: 267).
Against the backdrop of political marginalization, demographic anxiety and eco-
nomic decline, the GPLM resorted to armed rebellion, attacking in particular the
resettlement sites. In 1986, the GPLM launched military operations in Nyikwo and
Pinykew villages in Gambella district. This was followed by severe political repres-
sions by the Derg, now represented by the Nuer, in which more than eighty Anywaa
were killed. Although they were forcibly resettled in Gambella, the resettlers were co-
opted by the Derg to form local militias. Dependent on the state, and sharing its
insecurity in the new lands, the resettlers (Highlander farmers) became the national
‘flag bearers’ and increasingly adopted the nation-state discourse to which they owed
their very presence in the region. Armed by the Derg, they ‘defended’ the revolution
not only in the service of the national state but also as stakeholders in their own right.
At this point, the Anywaa and the Derg parted company. For the Anywaa,
Mengistu was no longer the Wora Ariat but part of a wider ‘conspiracy’ mounted
against their very existence as a people. The Derg, whose priority was no longer equal-
ity and progress but regime survival, defined the Anywaa as wonbede (pejorative
Amharic term for rebels) and thus unreliable, as shown in the following archival mate-
rial: ‘The main security problem in Gambella is the Anywaa bihereseb [nationality].
They are now receiving military training in the Sudan. The only solution is punitive
measures’.25 By defining the GPLM as an Anywaa movement and indiscriminately cat-
egorizing the Anywaa as wonbede, the Derg drove more and more Anywaa from its
fold. As the Derg became increasingly embroiled in the multiple wars against opposi-
tion forces at the national level, they embarked on large-scale forcible recruitment of
the youth into the army. This pushed many Anywaa youth into the GPLM camp. The
Anywaa were targeted for conscription more than the Nuer because their settlements
were nearer to the administrative centres. They were also perceived by government offi-
cials as ‘more Ethiopian’ than the Nuer. Because the agent of recruitment was the Nuer
leadership, this was seen by the Anywaa in ethnic terms. Many Nuer youth, however,
preferred to join the southern Sudanese refugee camps or to live far away.26 Thus, many

25. Quarterly Report, Gambella Awraja Administrative Office, 7 April 1986, File Number
2494/2-68/2, Gambella Archive.
26. The same discourse of discrimination can be heard from the Thiang Nuer, who were closer to
the administrative centres and complain that the Cieng Cany/Derg leadership deliberately tar-
geted them. The WPE secretary in the Gambella Administrative Region, Thowat Pal, is of the
Cieng Cany clan.
140 Playing Different Games

Anywaa perceived an imbalance in the relationship of the two peoples to the state: ‘the
Anywaa give more to the Ethiopian state but gain less than the Nuer, while the Nuer
give less but are rewarded more’.
The GPLM, whose leaders had previously been the avant-garde of the revolution
in Gambella, sought military, financial and diplomatic support from various sources.27
With support from other anti-Derg organizations based in the Sudan, ‘they [the
GPLM] began to broadcast underground radio and launched a small scale military
operation against government’ (Kurimoto 2005: 342). The Sudanese government was
receptive to the appeal and gave the GPLM a military training centre within its terri-
tory; but its attempt to fully control it (absorption into the Sudanese army) alienated
the GPLM leadership.28 The Sudanese government hoped to make use of the GPLM
in its war against the Sudan Peoples Liberation Army (SPLA), which was supported
by the Derg. As a result, the GPLM sought internal allies. By the late 1980s, there
were more than twenty ethnic-based and regional political movements opposing the
Derg regime (Tadesse 2002: 130). Of these, the strongest militarily were the EPLF
(Eritrean Peoples Liberation Front), the TPLF (Tigrean Peoples Liberation Front) and
the OLF (Oromo Liberation Front). Although the OLF initially gave military support
to the GPLM, the OLF’s similar attempts to control the GPLM as a satellite organi-
zation led to a brief confrontation between the two movements.29 As a result, the
GPLM contacted the TPLF, which, in 1989, transformed itself into the EPRDF
(Ethiopian Peoples Revolutionary Democratic Front), an ethnic coalition consisting
of Tigrean, Amhara and Oromo–based political organizations. In fact, the GPLM was
invited to attend the founding conference of the EPRDF, and on condition that it
would support regime change, it was promised political power in post-Derg
Gambella. As a result of the EPLF’s and EPRDF’s shift of military operations from
northern to western Ethiopia in the late 1980s, the GPLM emerged as an important
politico-military organization in Gambella in the 1990s.30

The Nuer under the Derg

The Nuer experience of the revolution differed significantly from that of the
Anywaa. The Nuer were not subjected to political control and cultural revolution
because, from the government’s perspective, they ‘lacked’ leaders. Unlike in other

27. One of the prominent leaders of the GPLM, Agwa Alemu, a member of the Woz League (a
Marxist group allied to the Derg), was sent to Cuba for political training. After he returned
from Cuba, Agwa was assigned as an administrator of Jikaw district. He was briefly imprisoned
during the infighting within the Derg that pitted the Woz League against the Revolutionary
Seded faction. Upon his release, Agwa joined the United Nations High Commissioner for
Refugees (UNHCR). Before he joined the GPLM, Agwa was reappointed administrator of
Jikaw district.
28. Interview with Uchan Okello, a major in the GPLM, Pochalla, 16 August 2003.
29. J. Young (1999: 326) mentioned the hegemonic aspirations of the OLF toward the neighbour-
ing border people such as the Anywaa, whom they refer to as Oromo guracha, ‘black Oromo’.
30. The GPLM attended the EPRDF’s conference held in the Sudan in 1990 to build a common
front against the Derg regime. As a result, five hundred GPLM soldiers were sent to Mekele,
the Tigrean capital in northern Ethiopia, for military and political training.
 Differential Incorporation into the Ethiopian State 141

pastoral areas, the ban on traditional resource management practices was not thor-
oughly implemented in those of the Nuer.31 On the contrary, the few educated Nuer
elites were integrated into the regional administration and party structure of the
Derg regime. What started as ‘buying’ individual services by the state had taken an
ethnic turn, as the Anywaa were increasingly defined as wonbede. The genesis of the
‘Derg-Nuer alliance’ was initially incidental. Effective service and loyalty, rather than
ethnicity, were given priority in making political alliances by the Derg (Teshale
1995). When the idea of vice-administrators was proposed in 1978, both the
Anywaa and the Nuer were perceived as not being ‘ready’ to fully exercise political
power. They were instead placed under Highlander administrators. The rise to pre-
eminence of Joshua as the governor of Gambella in the mid-1980s was as much due
to his own personal strength as it was to the weakness of the Anywaa vice-adminis-
trator. Like his counterpart Joshua, Thowat, the regional WPE Secretary, was origi-
nally a self-made politician. He made himself relevant through his contributions in
crushing the Jor rebellion in the late 1970s. From the perspective of the Derg regime,
the stakes in Gambella were high and the situation demanded people who were
capable of dealing with its multiple problems. At this stage of their regime, the Derg
systematically opted for allies.
The Anywaa believed that the imbalance in the distribution of administrative
power facilitated Nuer expansion into their territories. Notable evidence produced
by the Anywaa for this link was the resettlement of famine-affected Nuer from
Akobo district (Gaat-Jok Nuer) during the Baro-Abol integrated resettlement pro-
gramme. After the dissolution of the programme in 1991, this group of Nuer
remained at Abol and currently form one of the largest Nuer settlements near
Gambella town. According to senior SPLA officials, the Derg-supported SPLA also
encouraged the resettlement of Nuer – Gaat-Jak sections from Jikaw – near the Itang
refugee camp in order to inflate the number of refugees and thus ensure the uninter-
rupted flow of resources from the aid agencies. A joint communal farm was estab-
lished for the refugees and the locals.32 Most of these ‘citizen-refugees’ later returned
to their villages, but some managed to stay in the neighbourhoods of Itang town. It
was also during this time that spontaneous migrations from southern Sudan
occurred, such as the Cieng Reng settlement at Makot village in Itang district.
For the vast majority of ordinary Nuer men and women, however, neither the
revolution nor administrative power brought improvements in their living standards.
Like their Anywaa counterparts in the villages, the Nuer also suffered from the
SPLA’s atrocities (see Chapter 7), which led many to abandon their villages for the

31. See Ayalew (1997) and Getachew (2001) for a thorough discussion of the impact of the gov-
ernment development policies on the Kereyu and the Borana pastoralists, respectively. Both
the Kereyu and the Borana suffered from restrictions on mobility and because of the ban on
traditional resource management practices such as burning during the dry season, a widely
practiced form of pastoral resource management carried out to combat the problem of bush
encroachment.
32. This scheme was sponsored by the World Lutheran Federation (Peter Adwok, ex-SPLA senior
officer, interviewed in Nairobi, 29 August 2002).
142 Playing Different Games

apparent safety of the refugee camps. At the political level, fixing the international
boundary was an overriding policy of the Derg that adversely affected the Nuer’s
transhumant livelihood strategy. Two of the Gaat-Jak sections, the Cieng Cany and
the Cieng Wau, straddle the boundary during the wet and dry seasons, but a 1978
decree outlawed such movement and encouraged permanent settlement either in
Ethiopia or the Sudan.33 For most ordinary Nuer, therefore, the Derg period meant
insecurity and restrictions on pastoral mobility. As a result a significant number of
the Ethiopian Nuer joined the refugee camps, attracted by their safety and the avail-
ability of social services, particularly education. Nearly all of the current Nuer civil
servants and politicians in Gambella were educated in the refugee camps under the
auspices of the aid agencies.
Although the Nuer were not subjected to the Ethiopian state’s projects of control
to the same extent as the Anywaa, due to the location of their settlements in outly-
ing districts where the Ethiopian state lacked the administrative resources to ensure
political control, this also meant greater neglect in terms of access to basic social serv-
ices than in areas inhabited by the Anywaa. It was this marginality that led to the
Nuer political opportunism, expressed in the form of alternative citizenship between
Ethiopia and the Sudan. Subjected to neglect rather than control, however, the Nuer
did not take up arms against the Derg. On the other hand, emboldened by its polit-
ical recognition and military support from the advancing EPRDF soldiers, the
GPLM intensified its struggle against the regime. As the EPRDF approached the
western highlands, there was political excitement and euphoria among the Anywaa,
who now threw in their lot with the GPLM.34
This chapter has shown how fluctuating inter-ethnic power relations, generated
by the state, may themselves be viewed as causes of conflict between the Anywaa and
the Nuer. Before the arrival of the Ethiopian state in the region, power relations
among ethnic groups were, by and large, generated internally. In the second half of
the nineteenth century, the Nuer became the most powerful group among neigh-
bouring Nilotic peoples not because of military power generated externally but
because their system of segmentary lineages was well-adapted to ‘predatory expan-
sion’ (Sahlins 1961). In contrast, since the early twentieth century, inter-ethnic
power relations have depended largely on the degree of articulation between the
interests of various ethnic groups and the interests of the state. The emergence of the
Anywaa as a regional power in the first three decades of the twentieth century was
related to their privileged access to firearms provided by the Ethiopian state. If,
during that period, the Nuer reoriented their strategy in seeking to gain access to
resources through peaceful inter-ethnic social networking, it was partly because they
lagged behind the Anywaa in the local arms race. The second half of the twentieth
century saw the rise of the Nuer as a political and military power and the decline of

33. Joint Ethiopian–Sudanese Jikaw meeting, report submitted to Gambella District

Administration Office, dated 17 November 1978, File Number 218, Identification number 1,
Gambella Archive.
34. In the curfew that was imposed on Gambella in the early 1990s, no Anywaa was allowed to
walk around after 8 pm; to do so meant to be regarded with suspicion as a possible wonbede.
 Differential Incorporation into the Ethiopian State 143

the Anywaa. This is, again, directly related to the dispossession and disempowerment
of the Anywaa by Ethiopian state projects and the Anywaa’s unfavourable integration
into the market economy. The progressive decline of the Anywaa has, in turn,
encouraged members of some sections of the Nuer to be increasingly predatory, once
again changing their strategy for gaining access to resources from inter-ethnic social
networking to territorial occupation. The last decade of the twentieth century
brought a new opportunity structure for the rise of Anywaa political power. The fol-
lowing chapter shows how the Anywaa have sought to gain political power in the
context of ethnic federalism, why they have failed to sustain their power, and how
this has affected Anywaa–Nuer power relations.
 Chapter 7

The Anywaa Response to

Ethiopian Ethnic Federalism
At the end of the twentieth century, inter-ethnic power relations – which have been
identified above, together with differing modes of ethnic identification and compe-
tition for natural resources, as one of the three causes of ethnic conflict between the
Anywaa and the Nuer – were altered once again, initially in favour of the Anywaa.
In May 1991, the Derg was overthrown by the EPRDF. Upon seizing power, EPRDF
leaders and activists restructured the Ethiopian state according to ethnic federalism.
The 1995 constitution explicitly recognized ethnicity as the official state ideology
and the ethnic group as the legitimate unit of political action. Sovereignty now rested
on ‘nations, nationalities and peoples’ (Art.8), which were granted not only the right
to self-rule but also the right to secession (Art.39). On this constitutional basis
ethno-regional states were created. Some of these ethno-regional states were allocated
to and named after resident ethnic majorities. These states include Oromia, Amhara,
Tigray, Somali and Afar. Others were organized as multi-ethnic regional states with
various levels of political entitlement for the nations and nationalities living there.
These include the Gambella Peoples National Regional State (GPNRS), the
Southern Nations, Nationalities, and Peoples Regional State (SNNPRS) and
BGPNRS (Benishangul-Gumuz Peoples National Regional State (BGPNRS).
The Ethiopian experiment in ethnic federalism has generated intense scholarly
interest. Its admirers have hailed it as an expression of the democratization of state
and society (Kinfe 2001) and even as a model for other multi-ethnic African societies
(Chabal and Daloz 1999). Its detractors see behind it a Machiavellian design, i.e., a
strategy of ‘divide and rule’ which allows an ethnic minority – the Tigreans – to dom-
inate the majority (Vestal 1994). In a similar vein, other analysts emphasize the hege-
monic aspirations of the Tigrean elite, the main political actors of the new regime
(Merera 2003). In one way or another, all of these views address dimensions of the
new political order. In this chapter I discuss the practice of ethnic federalism with
reference to its ideological underpinning and to the contradictions that it has gener-
ated. Specifically, my ethnographic data raise questions as to why a policy with the
declared objective of promoting inter-ethnic harmony and enhancing a new sense of
national belonging has resulted in just the opposite, i.e., the escalation of ethnic con-
flict and greater contestation over national identity.
Attempts to form wider political communities on bases that were not ethnic did
not materialize, partly because of limitations imposed by the hegemonic EPRDF def-
inition of what constitutes a group (namely, ethnic identity) and partly because the
prevailing ethno-politics activated a particular form of social memory nation-wide –
146 Playing Different Games

in the case of the Anywaa a narrative of loss in the context of interethnic relations,
particularly with the Nuer. Attempts by some leaders of the Anywaa political organ-
ization, the GPLM, to create a political community on a wider basis of solidarity
(‘the black people of the border land’) were discouraged:

We tried to get the Mao and other Nilotic people in Anfillo [‘black’ minori-
ties in western Oromia] to join us. Initially the EPRDF supported us because
they wanted us to help them control the OLF, which was then active in the
region. Together with the Benishangul people [a north-western region that
borders on the Sudan], we had discussed the possibility of forming a political
organisation that would include all the black people in the border regions
from Gambella to Benishangul, Amhara and Tigray – all the way to Eritrea.
The EPRDF did not like the idea. Nor would the EPLF let that happen. We
sought, as a last resort, to merge with Benishangul, but we both were too busy
with internal problems to pursue such a higher goal. (Okello Oman, Ex-pres-
ident of Gambella Regional State, interview May, 2001)

Burdened with a historical mindset (i.e., the memory of Nuer conquest of Anywaa
lands), blocked by political imposition (the EPRDF’s hegemonic definition of what
constitutes a group), and excited by the promises of ethnopolitics, GPLM leaders
seized regional power, and their subsequent attempts to redress past injustices and
the resulting dramatic changes in the balance of power between the Anywaa and the
Nuer have largely shaped political processes in post-1991 Gambella. In the new
political space created by ethnic federalism and regional empowerment, the Anywaa
and the Nuer have found themselves to be variously positioned, and they have
employed different political strategies of entitlement, following new avenues of social
mobility and new strategies in the politics of recognition.
In the context of ethnic federalism, the GPNRS was conceived as the homeland
of ‘indigenous’ peoples or ‘Minority Nationalities’.1 The EPRDF has classified five
national minorities in Gambella: the Anywaa, the Nuer, the Majangir, the Opo and
the Komo. The Highlanders are neither politically recognized as a group nor differ-
entiated along ethnic lines as Oromo, Amhara or Tigreans. In fact, in this new polit-
ical dispensation, the Highlanders appear to have been reduced to the status of ‘guest
workers’.2 Under ethnic federalism, the peripheral region of Gambella suddenly

1. According to the 1995 Election Law of Ethiopia, ‘“Minority Nationality” means a community
determined, by the House of People’s Representatives or its successor, to be of a comparatively
smaller size of population than that of other nations/nationalities’ (National Election Board of
Ethiopia, ‘Election Law of Ethiopia 1995’, Proclamation No.111/1995).
2. This is despite the criteria of candidature specified in the Election Law. According to this law
(Art.38/1.b), ‘Any person registered as an elector shall be eligible for candidature, where he is
versed in the vernacular of the regional state of his intended candidature’s election law’. There
are many Highlanders who are fluent in Anywaa and/or Nuer languages, but the incongruity
between constitutional theory and practice is justified on the basis of protecting the political
rights of the national minorities.
 The Anywaa Response to Ethiopian Ethnic Federalism 147

became a regional state and thus a new political centre. Thus, both Anywaa and Nuer
elites enjoyed the prospect of new leadership opportunities, including the chance to
fill up the regional ministries dominated hitherto by the Highlanders. New political
offices, most notably the regional presidency and vice-presidency, were created, and
for the first time Anywaa and Nuer elites occupied ministerial positions at the
national level.3
In the competition to dominate the new political space created by ethnic feder-
alism, ethnic groups are variously positioned according to their degree of ‘localness’
(Dereje 2006a: 217). This is determined in part by differences in settlement history,
settlement patterns, demography and in degrees of incorporation into the national
centre; but at the national level the politics of entitlement are not based on an
explicit standard. One has to read between the lines to get a feel for the modus
operandi of Ethiopia’s post-1991 political structure. Which particular ethnic group
is qualified for which political status and on what basis is not spelled out. The hap-
hazard way in which the EPRDF applies the terms ‘nation’, ‘nationality’ and ‘people’
to the various groups gives the impression that demographic size is the defining cri-
terion. The five regional states that are designated as ‘mother states’ of the five largest
ethnic groups in Ethiopia seem to have been formed according to this principle. But
the allocation of a regional state to the Harari, one of the smallest ethnic groups in
the country, is, as Vaughan (2003: 229) describes it, a ‘constitutional oddity’ that
seems to contradict this generalization.4 In the three multi-ethnic regional states,
including the GPNRS, the various groups of people compete for dominance. In the
context of this competition, however, the Anywaa and the Nuer have adopted differ-
ent strategies.

The Anywaa Project of Containment

The GPLM seized state power in Gambella in 1991, and consequently the Anywaa
dominated political processes in the GPNRS throughout the 1990s. This is evident in
their disproportionate political representation in the regional council, in ministerial
posts and among civil servants in the regional bureaucracy (see Tables 7.1, 7.2 and 7.3).
These new relations of dominance had a direct effect on the nature of inter-ethnic rela-
tions, insofar as political representation in the regional council determined access to
new objects of struggle: employment opportunities in the regional bureaucracy and
access to modern goods and social services delivered by the federal government.

3. Anywaa and Nuer politicians have served as state ministers and Ethiopian ambassadors to
Kenya, Uganda, Ghana and Japan for the first time since the incorporation of Gambella into
the Ethiopian polity at the end of the nineteenth century.
4. Out of Harari Regional State’s population of 131,139, the Harari constitute only fifteen per
cent. This is ‘officially justified with reference to the unique historical and religious significance
of the holy city of Harar’ (Vaughan 2003: 229). This ‘constitutional oddity’ is often referred
to by larger ethnic groups in other multi-ethnic regional states in their attempts to legitimate
their claim for the status of a regional state.
148 Playing Different Games

Table 7.1 Political representation in the GPNRS regional council, 1991–2000

Group 1992 1995 2000

Anywaa 19 25 29
Nuer 12 10 19
Majangir 5 5 4
Opo 1 1 1
Komo 1 1 1
Highlanders – – 1
Total 38 42 55
(Source: author’s field notes)

Table 7.2 Allocations of ministerial posts in the GPNRS

Year Number of positions Anywaa Nuer Majangir Opo Komo Highlanders

in regional ministries
1992 20 15 3 1 – 1 –
1995 19 14 4 – – 1 –
2000 19 13 5 – – 1 –
2002 18 11 6 – – 1 –

(Source: author’s field notes)

Table 7.3 Ethnic profile of civil servants in the GPNRS in 2000

Ethnic group Male Female Total Percentage

Anywaa 1053 344 1397 36.3
Nuer 205 38 243 6.3
Majangir 29 5 34 0.9
Opo 7 3 10 0.2
Komo 1 – 1 0.02
Highlanders 1476 684 2160 56
Total 2,771 1074 3845 100

(Source: Bureau of Civil Service, Gambella Regional State)

Gambella’s political representation at the national level has also shown an imbalance.
Until 1995, the GPNRS was represented in the national House of Representatives
solely by the GPLM. There was an internal debate within the GPLM regarding the
extent to which the Nuer should be excluded from political power. Extreme elements
within the GPLM advocated a total exclusion; others opted for strategic co-option of
some of the Gaat-Jak clans, especially the Thiang. This resulted in the limited inclu-
sion of ‘Ethiopian’ Nuer in the GPLM, though in a subordinate political position.
As is indicated in the tables shown above, Anywaa political dominance found
expression in their hold on the office of the presidency and their presence in upper
 The Anywaa Response to Ethiopian Ethnic Federalism 149

echelons of power in the regional council. The Anywaa have interpreted the seizure
of power by the GPLM and the placement of one of their own in the office of the
president as a symbolic confirmation of their ownership rights over Gambella and
thus a validation of their claim to indigenous status. The resulting pattern of power
distribution has remained unchanged: an Anywaa president, a Nuer vice-president
and a Majangir secretary. Reserving the upper echelons of power for the Anywaa
further dramatized the subordinate political status of the Nuer in regional politics.
Thus, political power not only increases access to material rewards; it is also useful as
a means of renegotiating group status. As Horowitz (1985: 217) notes, ‘when uncer-
tainties exist, efforts are made to obtain authoritative allocations of prestige’. On the
basis of their de facto political empowerment and with reference to various ideolo-
gies of entitlement, the Anywaa dominated political processes in the GPNRS
throughout the 1990s.

Contribution to regime change and Anywaa empowerment

Anywaa political actors legitimized their dominant position in the GPNRS through
the language of liberation. In doing so, they mimicked the Tigrean Peoples Liberation
Front’s (TPLF’s) own ideology of power at the national level. The TPLF is the main
political force within the EPRDF. It has dominated the political process in Ethiopia
since 1991, despite having a much smaller constituency than political organizations
that claim to represent the Oromo and the Amhara, two of Ethiopia’s largest ethnic
groups. One of TPLF’s ideologies of power is its ‘greater’ contribution to regime
change during the armed struggle against the Derg. TPLF/EPRDF’s consent to
GPLM’s seizure of power in the GPNRS was granted on the same basis. The Nuer
were excluded from the regional power structures because they did not participate in
the armed struggle. In fact, they were identified with the defunct regime.5
The new political order has markedly changed the existing power relations
between the Anywaa and the Nuer. The seizure of power by the GPLM was perceived
by both Anywaa and Nuer as the empowerment of the Anywaa. Shortly before the
GPLM occupation of political power in Gambella, the Nuer Derg officials and all of
the Sudanese refugees left the camps in Gambella, embarking on a long and arduous
‘exodus’ to southern Sudan. The refugees were joined by Ethiopian Nuer citizens
who were persuaded to leave Gambella for fear of an ‘Anywaa reprisal’ (Kurimoto
2005: 352). With the elimination of the Nuer in the regional power game during the
transitional period (1991–95), the Anywaa elites monopolized power in the new
regional state, seizing all the key offices from the presidency to the regional ministries
and the security apparatus.
In the period from 1991 to 1992 there were a series of violent conflicts among
various interest groups. Radical GPLM members committed atrocities on the
remaining Nuer civilians in Gambella town and Itang district. Nuer-based southern

5. A similar pattern of political alliance was evident in the Benishangul-Gumuz Regional State,
where the Berta-based Benishangul Peoples Liberation Movement was initially rewarded by the
EPRDF (Young 1999: 335).
150 Playing Different Games

Sudanese rebel groups, in concert with the Nuer ex-Derg officials who fled to Nasser,
launched counterattacks on the GPLM/EPRDF establishment and burned Anywaa
villages in Jikaw and Itang districts. An independent and objective account of the
number of casualties on both sides is hard to come by. The Nuer version of events
puts the count of the dead as high as a thousand, and refers to the period as the ‘mass-
killing’ of the Nuer. The Anywaa casualties are often measured in terms of the
number of villages burned and people displaced. According to Anywaa accounts,
nearly all of the Anywaa villages between Itang and Jikaw districts were burned twice
by the Sudanese-based armed Nuer groups. The conflict also involved elements of
symbolic violence, epitomized by an incident in Itang town. In 1991, GPLM soldiers
killed a Nuer civilian and then ‘crucified’ him with maize flour in his mouth: this was
meant to represent a hungry Nuer who came to Anywaa areas in search of food, and
thus to serve as a symbolic act of group humiliation. The Nuer regrouped in Nasser
where they established a political party called GPDUP (Gambella Peoples
Democratic Unity Party) in 1992.
Supported, during the regime change, by the EPRDF, emboldened by the new
discourse of ethnic equality, and encouraged by the EPRDF’s own primordialist
concept of identity, the Anywaa power elites sought to control the new regional state
of Gambella. Politically excited by the sudden turn in the balance of power, they seri-
ously contemplated an irredentist project, laying claim to what they regard as tradi-
tionally Anywaa territories that are now de facto occupied by the Nuer. Seeking to
capitalize on their initial promotion by the EPRDF to new forms of administrative
power, the Anywaa aimed to renegotiate their asymmetrical local power relations
with the Nuer.

Settlement history and the indigenous claim

The Anywaa political actors also justified their dominant political position in the
language of indigeneity. They advanced their political ownership right over the
Gambella region on the basis of settlement history. The exact time when the Anywaa
settled in present-day areas is not yet well established. Various scholars have suggested
that the process of ethnogenesis within the Lwoo conglomeration might have
occurred from the fifteenth to the seventeenth century (Crazzolara 1950; Collins
1971). Perner (1997: 138) places the genesis of the Anywaa kingship in the seven-
teenth century. Nor do we know who the pre-Anywaa inhabitants of Gambella were.
Anywaa oral tradition makes reference to the Burun-speaking people at the time of
their settlement. It also suggests that parts of their present-day settlement areas were
formerly occupied by the Majangir, particularly in the south-east in present-day Gog
and Abobo districts (Evans-Pritchard 1947; Kurimoto 1994). Evans-Pritchard
(1947: 73) described the relationship between the Anywaa and the Majangir of the
1940s as patron and vassal, respectively. Apart from allusion and gradual incursions,
‘there is practically no proof that the immigrants [the Anywaa] should ever have met
military resistance and taken the land by force’ (Perner 1997: 138). The Anywaa
deny, play down or defend their migration because it did not result in the large-scale
displacement of other groups of people. Nor are there groups of people who could
contest the Anywaa land claims.
 The Anywaa Response to Ethiopian Ethnic Federalism 151

Table 7.4 1984 Census result of Illubabor province

Major ethnic Rural Per cent Urban Per cent Total Per cent
groups population population
Anywaa 25,486 2.8 2,558 3.7 28,044 2.9
Amhara 34,522 3.8 19,628 28.7 54,150 5.6
Bencho 1,712 0.2 90 0.1 1,802 0.2
Dizi 1,010 0.1 31 0.1 1,041 0.1
Guraghe 2,171 0.2 4,913 7.2 7,084 0.7
Keffa 28,724 3.2 1,770 2.6 30.494 3.2
Majangir 10,748 1.2 5 0.1 46,305 4.8
Mocha 44,303 4.9 2,002 2.9 10,753 1.1
Nuer 26,406 2.9 668 1.0 27,074 2.8
Oromo 705,144 78.2 32,357 47.4 737,501 76.0
Sheko 6,910 0.8 62 0.1 6,972 0.7
Tigrawi 1,362 0.2 1,938 2.8 3,300 0.3
Others 12,333 1.4 2,189 3.2 14,522 1.5
N/S 1,129 0.1 72 0.1 1,201 0.1
Total 901,960 100 68,283 100 970,243 100
(Source: Office of the Population and Housing Census Commission, Population and Housing
Census, 1984: 32)

The Nuer settlement history in Gambella, on the other hand, is less disputed. It was
part of their massive migration from southern Sudan to the east which started in the
second half of the nineteenth century. By the early twentieth century, groups of
Jikany Nuer (Gaat-Jak) settled in present-day Jikaw district. In the 1960s the demo-
graphic size of the Nuer in Gambella continued to increase, partly due to the influx
of Nuer refugees from southern Sudan, some of whom managed to settle as
Ethiopian citizens. As shown in Table 7.4, the Anywaa held a very slim margin over
the Nuer in Illubabor province as of the 1984 national census.
By the mid-1990s, however, the Anywaa were overtaken by the Nuer; this is
evident from the 1994 census, which established the Anywaa as twenty-seven per
cent and the Nuer as forty per cent of the region’s population. This disparity has
further pushed the Anywaa to re-introduce the historical and citizenship cards in the
politics of entitlement. The more the Nuer emphasize their political status as an
‘ethnic majority’, the more the Anywaa frame their power claim in the language of
indignation. Anywaa political domination is also legitimated by the number of dis-
tricts they control. In the new regional states, political representation is district
based. Of the nine districts of Gambella in the 1990s, six were inhabited predomi-
nantly by the Anywaa, only two by the Nuer.
Adopting the language of indignation in the politics of entitlement has activated
a particular kind of collective Anywaa memory that centres on the narrative of loss.
As a result, actual peaceful socio-economic exchanges and shared cultural features are
neither remembered nor acknowledged. For example, forms of inter-ethnic integra-
tion, such as the socio-economic exchanges and politico-military alliances between the
Openo Anywaa and the Thiang Nuer, are played down by the Anywaa power elites,
152 Playing Different Games

except when it serves their purposes to magnify the political divisions among the Nuer
by glossing the Thiang as ‘real’ Ethiopians, in contrast to other ‘Sudanese’ clans.
The event which serves as an historical validation of the Anywaa narrative of loss
is the nineteenth-century Nuer expansion at the expense of Anywaa territories. In the
politics of memory, ancient hatreds are invoked in the form of songs and sayings
from past conflicts in order to produce evidence of loss. When the past is thus recon-
structed, the future emerges as inherently dangerous. It is in this atmosphere that the
Anywaa political actors have presented the Tier Agak song, discussed in Chapter 5,
as evidence of the Nuer hidden ‘colonial’ agenda. Here we find the Anywaa seeking
political power not only for material rewards but for collective preservation, an
agenda that features prominently among the ordinary Anywaa as well.
As indicated in previous chapters, Nuer expansion occurs largely at the individual
level through microprocesses. To the Anywaa, this Nuer expansion becomes the
expansion of a life world, a cultural form and a speech community rather than solely
an organized political movement or concerted military effort, except in places like
Akobo where violence has periodically erupted. In the post-1991 identity politics,
Nuer demographic trends (assimilation and associated practices) are represented as
ethnic conspiracy. Every Nuer man and woman suddenly appears to be an ‘ethnic
soldier’. Here is a clear interplay among the resource, identity and power variables in
Anywaa–Nuer conflict. The new political process has brought local forms of power
and asymmetrical integration into what Verdery (1994: 46) termed ‘the realm of
notice’: ‘We might … see state- making as a process that raises “difference” from the
realm of doxa, the assumed, into the realm of notice, where disputes could occur
between the orthodox and the heterodox, the normal and the strange’. With the
increasing relevance of settlement history and who came where and when, the Anywaa
even fix dates and personal names to assert their ownership right over Gambella:

The Nuer started coming to Gambella in 1976. The first person who settled
in Newland [Nuer neighbourhood in Gambella town] was a person called
Tumoro. Tumoro settled at a place called Aba Joro veranda. At that time the
Nuer used to come to Gambella, stay for couple of days, sell their cattle and
go back to Jikaw. (Okello Akwei, resident of Gambella town, 3 June 2000)

In this and other similar narratives the word Gambella refers to the town and the
region. For the Anywaa who do not border the Nuer, such as in Abobo and Gog dis-
tricts, their contact with the Nuer is indeed recent, as late as the 1970s when an
incipient form of urbanization started in the region. For this group of Anywaa the
‘latecomer’ status of the Nuer is very conspicuous, more so because Gambella town
itself is regarded as an Anywaa territory, an extension of the neighbouring Pinykew
village. For the Anywaa who have a longer contact with the Nuer (the Openo
Anywaa), the generic use of the name Gambella and the recentness of the Nuer is at
times used as a rhetorical device to mobilize the other sections of Anywaa society to
participate in the project of containing the Nuer.
 The Anywaa Response to Ethiopian Ethnic Federalism 153

Framing local concerns in national terms

Settlement history has also introduced the issue of national identity into the politi-
cal debate, where the Anywaa emerge as ‘more Ethiopian’ than the Nuer who are
labelled as Sudanese. The Anywaa’s evidence for their definition of the Nuer as for-
eigners comes not so much from oral history or history books as from the Nuer prac-
tice of alternative citizenship: switching between Ethiopian and Sudanese national
identities. This was particularly true in the 1980s, when being a refugee was more
advantageous than being a citizen in one’s own country. The multiple aid agencies
that operated the relief industry provided a variety of goods and services to Sudanese
nationals and those Ethiopians who could claim a refugee status. At that time, the
Ethiopian Nuer could pass as southern Sudanese more easily than the Ethiopian
Anywaa, because more Nuer than Anywaa lived in southern Sudan. On that basis,
the aid agencies accepted the Ethiopian Nuer as refugees prima facie, while the
Ethiopian Anywaa had to undergo tight screening procedures.
In the 1990s, when claiming Ethiopian citizenship became materially rewarding,
this state of affairs enabled the Anywaa to frame their exclusionary political practices
in national terms. Despite differences in settlement history, some groups of Nuer had
already arrived in the Gambella region at the beginning of the twentieth century; but
the Anywaa elites have defined nearly all the Nuer as Sudanese and thus as foreign-
ers, creating a new homogenizing element in the consolidation of Nuer ethnicity.
The genesis of this debate, however, goes back to the colonial period when the British
proposed a division of subjects with the Ethiopian state, claiming all the Nuer and
offering all the Anywaa to be placed within Ethiopian domain on administrative and
economic grounds (Collins 1983). This, coupled with the differential access to
NGO-mediated resources, created a certain envy on the Anywaa side in the 1980s,
and made the Nuer vulnerable to the politics of exclusion in the 1990s.
Cognizant of the cross-border settlement pattern of their competing neighbour,
the Nuer, the Anywaa constantly refer to the state border as a frame of reference to
define the parameters of inclusion/exclusion in the GPNRS. The state border as a
reference point has enabled the Anywaa to frame an ethnic concern (fear of extinc-
tion) in national terms, in which the Nuer do not just encroach on Anywaa lands but
become ‘outsiders’ – ‘foreigners’ troubling ‘citizens’. Framed this way, the Anywaa
anticipate mobilizing the Ethiopian state in a local struggle, and invoke the state dis-
course (sovereignty) more than the state itself does. Their call for a rigid state border
serves, however, as a catalyst to talk about ethnic security. Principal Anywaa political
actors extend the discourse of an ‘impending danger’ to the Ethiopian state itself, to
which they hope Nuer pastoral expansion will appear as a ‘national security threat’:

The Nuer are now taking over our lands. Tomorrow they will go to the high-
lands. They are already in Dambidolo, even Addis Ababa. What do they do
there? During the SPLA time [in the 1980s] the Nuer commanders used to
say Gambella was southern Sudan because there were no black Ethiopians.
The Nuer are not just expanding into Anywaa territories. They have a hidden
agenda of annexing Gambella to southern Sudan as well. If the Nuer do not
154 Playing Different Games

stop pushing us, we will also finally go to Bure, even to Gore [the two nearest
highland towns]. Where else could we go then? What makes Ethiopia a
country if it does not secure its border? (Opamo Oboya, Ruiri, Kenya, 12
August 2002)

Although the Anywaa were politically dominant in the GPNRS throughout the 1990s,
they did not manage to determine the international border as much as they would have
liked. Management of the international borders is currently a federal mandate, and in
securing the border areas, the Ethiopian state has multiple concerns rather than focus-
ing solely on a show of sovereignty. Lack of interest by the federal government in polic-
ing the international border so that Nuer border-crossing could be checked is
interpreted by the Anywaa as evidence of their own status as ‘second-class citizens’.

Some borders are well protected and the government provides security to the
border people. Hasn’t the entire nation gone to war with Eritrea [1998–2000]
because of Badime? The government swiftly declared war on Eritrea because
the people who live in the border town of Badime are the Tigreans; the same
people who also rule the country. The Gambella border is 360 degree open.
Sudanese Nuer could cross the border any time and take over Ethiopian land.
(Abula, Nairobi, August 2002)

The Ethiopian government, however, remains ambivalent towards the Anywaa’s call
for rigidification of the border. More generally, Ethiopia has often seen itself as sur-
rounded by potential enemies, and worried about the extent to which its minorities
would be loyal in the event of war with its neighbours, a classic example of what
Kymlicka (2006: 39) called ‘the securitization of ethnic relations’. The regional states
in the border areas are more susceptible to geopolitical dynamics. In fact, the extent
to which the EPRDF manages to integrate these states into the new political order is
the real test for the viability of the risk-laden ethnic federalism. Two of these border
regional states – the Somali and the Benishangul – pressed for secession in the mid-
1990s on the basis of political grievances and in reliance on the Constitution (Young
1999; Hagmann and Khalifa 2006). With a larger cross-border settlement of the
Nuer in the Sudan, the EPRDF is also wary of actual and potential cross-border
political networks. This source of concern can also be regarded as a diplomatic asset,
as it offers the possibility of influencing the politics of neighbouring states by way of
the same cross-border settlement. A clue to EPRDF’s ambivalent position can be
found in the following narrative by a government cadre.
As part of a policy to enhance political legitimacy, the EPRDF opened a politi-
cal indoctrination school at Tatek, some seventy kilometres west of the capital, par-
ticularly for elites in the peripheral regional states such as Gambella who were not
directly governed by the EPRDF. During the political training of Anywaa and Nuer
cadres in 1998, the issue of citizenship was addressed:

One of the issues which we discussed in this training was ‘What would
happen to Gambella should Southern Sudan become independent?’. The
 The Anywaa Response to Ethiopian Ethnic Federalism 155

Nuer cadres said, ‘We will remain as Ethiopian citizens’, but the Anywaa said,
‘No, the Nuer will be the first to go to the Sudan because even now they iden-
tify as Sudanese’. After a long debate, the EPRDF told us the answer: ‘It is
not right to say the Anywaa are Ethiopian and the Nuer are Sudanese. Both
of you come from the Sudan [in reference to Nilotic migrations] and nothing
will change the fact that Gambella is an Ethiopian territory. Both the Anywaa
and the Nuer have the right to live in Gambella’. The EPRDF cadre then
posed this question to us: ‘But what is this Nuer Christian prayer which says
“May God liberate our land”? Are you not living in an already liberated land?
Why should you pray for Southern Sudan?’ We told them that we pray like
that because the Nuer church came from the Sudan and we simply adopted
their prayer. (Peter Kayier, GPDUP cadre, Nairobi, August 2002)

A similar ambiguity is observable among the Highlanders, who are categorically

identified with the Ethiopian state by the Anywaa and the Nuer. Represented as
such, the Highlanders emerge as ‘significant others’ in evaluating the depth of
Anywaa and Nuer commitment to Ethiopian national identity, although they have
been politically marginalized in post-1991 Gambella. The following narrative by a
Highlander from Gambella town illustrates how the Highlanders position them-
selves in the citizenship debate between the Anywaa and the Nuer:

The Anywaa are familiar to us. We have a history of living together. You can
trust them as Ethiopians. The problem with the Anywaa is their politics.
They claim Gambella and regard all the other people as outsiders. The Nuer,
on the other hand, are foreigners [Sudanese]. But the good thing about the
Nuer is that they are hard working and they don’t discriminate against people.
Economically they are also useful. We don’t get anything from the Anywaa.
At least we see the Nuer cattle in the butchery. But you cannot trust the Nuer.
Their heart is with the Sudan. You do not find them in Ethiopian history
books. How can you trust somebody whom you do not know? (Highlander
civil servant, Gambella town, July 2000)

In the context of very volatile relations with the Sudan, the Nuer are occasionally
regarded by the Ethiopian government as a ‘security threat’. In 1997, for instance,
the two countries were on the brink of war, with Ethiopia accusing the Sudanese gov-
ernment of ‘exporting’ Islamic fundamentalism to Ethiopia (Medhane 2007). With
a Sudanese Nuer rebel leader (Riek Machar) then allied with the government of
Sudan, the Ethiopian Nuer were considered a security threat in the event of a war
with the Sudan. As a result, the Ethiopian government, allied with the Anywaa,
undertook a tight screening procedure to distinguish between Nuer ‘citizens’ and
‘refugees’. Towards that end, the Newland, the Nuer settlement in Gambella town,
was raided by the federal army and the regional police dominated by the Anywaa.
For their part, the Anywaa present themselves to the Ethiopian government and the
Highlanders as ‘genuine’ Ethiopian citizens who represent the national interest better
than the Nuer. In return, they expect government protection from the cross-border Nuer
156 Playing Different Games

territorial expansion and, in their eyes, the ultimate ‘extinction’ of Anywaa society. In
effect, the Anywaa have appealed to the Ethiopian government to live up to their expec-
tations of it, by determining and policing the international border. Throughout the 1990s
the Anywaa in Gambella, desiring enforcement of the international border to help them
contain further Nuer territorial expansion and to act as a guarantee of their ethnic sur-
vival, overstated their Ethiopian national identity and were conspicuously silent about
their own Sudanese connections, at least until they fell out with the EPRDF in 2002.
The Anywaa’s appeal to the Ethiopian state, and by implication to the Highlanders,
has brought them mixed results. Although many of the Highlanders were resettled
involuntarily by the Ethiopian government in Gambella in the 1980s, they have, in due
course of time, developed local interests, which they defend vis-à-vis the Anywaa.
Because of the strengthening of Anywaa rights to land and their political empowerment
in post-1991 Gambella, the Highlanders feel insecure on the land that they have appro-
priated. Highlanders enjoyed a hegemonic political position in pre-1991 Gambella, and
many of them have not yet come to terms with the subsequent political empowerment
of the ‘indigenous’ peoples, whom they look down on. Highlanders also resent the
reverse discrimination and insecurity that their new status as a regional ‘minority’ has
bestowed upon them. The tension between the Anywaa and the Highlanders erupted in
two major violent conflicts in 1991 and 2003. Kurimoto (1994:16), who has reported
extensively about the incidents leading to the 1991 Ukuna massacre of the Highlanders
by groups of Anywaa in Abobo district, provides the following description:

Soon after the capture of Gambella town at the end of May 1991, fleeing
former government soldiers passed through Ukuna, an area to the east of
Abobo district where 770 Anywaa and 3,000 resettlers [Highlanders] lived. A
few days before their arrival, a notorious Anywaa outlaw killed members of a
settler family. Settlers brought the case to the soldiers [Highlanders] and the
chairman of the Anywaa peasant association was shot dead by soldiers. After
the soldiers fled, Anywaa villagers started to attack settlers, setting fire to their
houses and killing them indiscriminately with rifles and spears.

The Anywaa targeted all the ‘red people’ whom they met in Ukuna, as they were per-
ceived to be the same people as the soldiers who killed their leader. The exact number
of human casualties is unknown, but Kurimoto (2005: 352) mentioned that ‘it was
later estimated that between 1,000 and 2,000 settlers were killed, and numerous vil-
lages were completely devastated’. In the aftermath of the Ukuna massacre, many set-
tlers left their villages and re-settled in Gambella town, where they expected that
some security would be provided by the federal army, which was garrisoned in the
town and manned entirely by Highlanders. Local representatives of the federal state
– all of whom were (and still are) Highlanders – put pressure on the regional govern-
ment, at that time, manned entirely by Anywaa, to bring the culprits to justice.
Through the Ukuna incident, the federal government learned of its ‘need’ to limit
regional autonomy and to curb the ‘excesses’ of local empowerment.
The violent conflict between the Anywaa and the Highlanders in 2003 was of an
even greater magnitude. The events leading to the massacre of Anywaa by
 The Anywaa Response to Ethiopian Ethnic Federalism 157

Highlanders on 13 December 2003, in Gambella town was related to the complex

and changing political relations between the EPRDF and both the Anywaa and the
Nuer. Shortly after the EPRDF had helped the GPLM to gain regional political power
in 1991, the two political organizations clashed on issues relating to regional and orga-
nizational autonomy. As part of the EPRDF’s wider effort to ensure total political
control nationwide, it disarmed all liberation fronts, including the GPLM, in 1992.
The GPLM bitterly resisted and briefly confronted the EPRDF on the issue of disar-
mament. In response, the EPRDF replaced the GPLM leaders with more pliant
members, especially after the GPLM was reconstituted as a political party, the
Gambella Peoples Liberation Party (GPLP) in 1995. Anywaa political power was also
seriously compromised by the Office of Political Advisors, EPRDF functionaries who
ruled the regional states as de facto ‘king-makers’, if not as kings (Dereje 2006a: 224).
Although this office was resented by both the Anywaa and the Nuer, the Anywaa
power elites found it to be especially threatening. The fact that all the political advis-
ers were ethnic Tigreans (Highlanders) also exposed the ethnic roots of the EPRDF
regime. In 1998, in what was considered by the Anywaa to be yet another ‘plot’ to
weaken their power, the EPRDF imposed a merger between the GPLP and the Nuer-
based Gambella People Democratic Unity Party (GPDUP), forming a new ruling
front called the Gambella Peoples Democratic Front (GPDF). The merger signalled a
certain degree of parity between the Anywaa and the Nuer in regional power politics.
This was, however, seriously contested by members of the more educated
segment of Anywaa society, whose discontent finally led, in 1998, to the establish-
ment of a regional opposition party known as the Gambella Peoples Democratic
Congress (GPDC). Although the GPDC assumed a regional name and included
some Nuer and Majangir, as had been the case with the GPLM which preceded it, it
was, for all practical purposes, dominated by the Anywaa. The establishment of the
GPDC heralded a bitter intra-ethnic power struggle, with a split developing among
the Anywaa power elites on which strategies to follow in the project of containing
the Nuer. Styling itself as a party of the educated Anywaa, the GPDC not only chal-
lenged the ‘uneducated’ and ‘docile’ GPDF, but also exposed the undemocratic
nature of the ruling EPRDF in a national political debate in the run-up to the
national and regional elections in 2000. In the regional election, the GPDF and the
GPDC fought a fierce political battle, which was often characterized by violence.
Threatened by the growing popularity of the GPDC, particularly among its Anywaa
constituency, the Anywaa power elites in the GPDF threw in their lot with their
Nuer ‘partners’ in a desperate bid to maintain political power. They labelled the
GPDC as anti-Nuer and anti-Highlanders and promised a greater inclusion of both
in the politics of the region should they be re-elected. Nevertheless, the GPDC won
in most of the Anywaa districts, and some of the electoral results were initially rec-
ognized by the national election board. In what was eventually exposed as one of the
most serious frauds in the 2000 national election, however, the result in Gambella
was rigged, and the GPDF, under the auspices of the EPRDF, was accorded a ‘land-
slide’ victory.
The intra-ethnic rivalry among the Anywaa continued until 2001, when two
political events resulted in rapprochement among the competing Anywaa power
158 Playing Different Games

elites. The first event was the split of the ruling TPLF/EPRDF in March 2001 and
the ultimate removal of the Office of Political Advisers. This created a new political
space for inter-ethnic competition and added to the growing fear among the Anywaa
that the Nuer would ultimately usurp their political power. The second event was
related to the issue of succession to the office of the regional vice-president. In May
2001, the Nuer vice-president died. For more than a year afterward, the office of the
vice-president, a de facto Nuer position, was vacant because the two parties that
comprised the ruling GPDF – the Anywaa-based GPLP (the former GPLM) and the
Nuer-based GPDUP – could not agree on their respective nominees. Rejecting the
GPDUP’s candidate, the GPLP nominated their own favourite within the GPDUP.
At the height of the political rivalry, the Anywaa helped to establish a new Nuer
party, known as the Gambella Peoples Democratic Union (GPDU), which opposed
the GPDUP and fielded its own candidate for the vice-presidency. The political con-
frontation only needed minor stimulation to cause it to escalate into the bloodiest
conflict between the Anywaa and the Nuer in recent years.
It began with a personal quarrel between an Anywaa and a Nuer in Itang district
on 7 July 2002, which swiftly degenerated into violence that left more than forty
people dead from a single engagement. In the days that followed, at least twenty-one
villages were burned, hundreds were killed and tens of thousands of people were dis-
placed. The conflict soon spread to Gambella town, where isolated incidents gradu-
ally turned to systematic killing on both sides, ranging from surprise attacks and
stabbings to bombings. From Gambella town, the conflict spread into the neighbour-
ing Pinykew/Ocham village, a mixed-settlement area for Anywaa and Nuer commu-
nities, hitherto known for its inter-ethnic peace and symbiotic exchanges. In August
2002, refugees became a new target; the Anywaa in Abobo district massacred more
than thirty Nuer refugees in retaliation for the Anywaa killed in Itang district in June.
In November 2002, the Anywaa of Pinyudo district killed thirty-three Nuer refugees
in the Pinyudo camp. Schools were closed for months, and the two communities were
virtually locked in their respective neighbourhoods. The economic activities of the
town were brought to a standstill, and people fled the region seeking safety.
The regional government utterly failed to contain the situation. Indeed, some of
the main political actors fomenting the conflict were identified as high-ranking
leaders in the regional council. As the situation deteriorated, the federal government
intervened and took measures to calm the situation: Gambella was placed under a
state of emergency; federal police and the army took control of the regional govern-
ment; the ruling GPDF was dissolved, and its leaders who were involved in the con-
flict, including the Anywaa regional president, were imprisoned; members of the
regional police (largely Anywaa) who were accused of inciting and participating in
the violence were jailed or dismissed from their jobs; and the contentious multi-
ethnic district of Itang was abolished and its territories divided between the Nuer dis-
trict of Jikaw and the newly created Anywaa district of Openo-Alwero. The federal
government also identified the ‘root cause’ of the conflict situation in Gambella as
the existence of ‘too many political parties’. On that basis, all existing parties were
abolished in early 2003 and replaced by new ethnic parties modelled on the
EPRDF’s style of governing regional states through subordinate ethnic PDOs
 The Anywaa Response to Ethiopian Ethnic Federalism 159

(Peoples’ Democratic Organizations). These Gambella PDOs were the Anywaa

People’s Democratic Organization (APDO), the Nuer People’s Democratic
Organization (NPDO), and the Majangir People’s Democratic Organization
(MPDO), which were organized by the EPRDF into a new umbrella political organ-
ization called the Gambella Peoples’ Democratic Movement (GPDM).
The political measures taken by the federal government alienated a large segment
of Anywaa society. What was initially referred to as Anywaa shifta (bandit) activity,
and which gradually evolved into an armed rebellion, was largely organized by ex-
Anywaa police who had been dismissed from their jobs. Unable to support their own
families and claiming to represent Anywaa discontent, the shifta resorted to violence
against both government establishments and civilian Highlanders, due to the
Highlanders’ ‘guilt by association’ with the Ethiopian state. In September 2003, for
instance, six Highlander road construction workers were killed, and this incident was
followed by a series of indiscriminate killings of other Highlanders. Such killings
were justified by the shifta as ‘killing’ the EPRDF. Furthermore, on 13 December
2003, eight Highlander government officials were brutally killed, and their severely
mutilated bodies were conveyed openly in a vehicle to the regional council offices for
public display before being taken to the hospital mortuary. Assuming that the
murders were committed by Anywaa shifta – ‘it was widely assumed both by the
Highlanders and the government that the ambush was the work of an armed Anywaa
group or Anywaa shifta’ (Human Rights Watch 2005: 12) – and agitated by the grue-
some display of the mutilated bodies, the Highlanders went on a rampage later that
day, indiscriminately killing Anywaa males with rocks and machetes. Some locally
deployed members of the federal army, who were exclusively Highlanders, also par-
ticipated in the killings, using automatic weapons. Estimates of the casualties vary.
Anywaa sources and international human rights organizations put the Anywaa death
toll as high as four hundred and twenty-four (Human Rights Watch 2005; Anuak
Justice Council 2006), whereas the government acknowledged only sixty-seven.
A spiral of revenge killings on both sides ensued. Aggrieved by the complicity of
government agencies in the massacre, feeling vulnerable to more attacks, and disap-
pointed by the lack of protection from or public apology by the government, about
a third of the Anywaa populace crossed the border to southern Sudan, where the
various Anywaa armed groups were brought together and formed into a politico-mil-
itary organization known as the Gambella Peoples Liberation Front (GPLF). From
its base in Pochalla in southern Sudan and the adjacent Anywaa territories on the
Ethiopian side of the border, the GPLF has fought against the Ethiopian army with
varying degrees of success. In October and November 2005, for instance, the GPLF
raided the police stations in Gambella and Abobo towns. In the ensuing gun battles
between the GPLF and government soldiers, the regional police commissioner was
killed and the GPLF were able to set Anywaa prisoners free.
The existing political tensions between the Anywaa and the EPRDF were further
compounded by the prospect of discovering strategic resources in Gambella and by
related issues of economic control. The Gambella basin is one of the major potential
petroleum producing areas in Ethiopia. Currently, a Malaysian oil company,
PETRONAS, is undertaking exploration of the entire expanse. The Anywaa, spear-
160 Playing Different Games

headed by the diaspora in North America, describe the events that led to the 13
December 2003 massacre and its aftermath in the language of genocide, claiming that
it was the culmination of the genocidal intent of both the Anywaa’s neighbours and the
Ethiopian state. The Anywaa believe that they are targeted by the Ethiopian government
because of the riches of Gambella, which they regard as their home (Dereje 2007: 18).

Anywaa Ethnicity – Protonationalism?

In his seminal work, Marxist Modern, Donham (1999) introduced the concept of ‘catch-
ing up’ and urged scholars to engage with local projects of modernity: ‘Have anthropol-
ogists or historians yet appreciated the consequences that flow from the apparently
simple fact that some actors view their societies as “behind” and therefore in need of a
way to “catch up”’? (Baker 1990, quoted in Donham 1999: xv). The local discourse on
inequality and the need to catch up is conceptualized differently in various societies.
Among the Anywaa there is a growing reflexivity about their declining group status that
underpins the discourse on the fear of extinction. One of the means through which the
Anywaa have sought to contain the Nuer and catch up with the Highlanders is through
‘capturing’ the regional state of Gambella. Here we view the Anywaa quest for political
power not only instrumentally (competition for leadership) and as part of the project of
containment (the struggle to maintain cultural identity) but also in terms of statism as
an ideology and mechanism of transforming society. By ‘statism’ I refer to ‘the belief in
the form of government that involves significant state intervention in personal, social or
economic matters’ (Soanes and Hawker 2005: 1014).
Contemporary Anywaa statism has two sources of inspiration. First, it draws on
the experience of ethnic groups who the Anywaa perceive to have used state agency to
transform their societies. The primary reference group for the contemporary Anywaa
elites, and for the elites of other ethnic groups in Ethiopia, is the Tigreans, whose
political leaders (the TPLF) are believed to have used not only the regional state of
Tigray but also the national state to turn a resource-poor region into a success story:
‘If the Tigreans made it, why not us?’ is a standard remark the contemporary educated
Anywaa elites make while talking about the agency of the state.6 The sense of relative
deprivation is more acute among the Anywaa elites because they regard their society
as not only in need of transformation but also in urgent need of being ’rescued’ from
social atrophy. There is a growing reflexivity on the progressive decline of Anywaa
society, as evidenced by widespread village alcoholism, depopulation of villages due to
a higher death rate and an exodus to the towns, and the increasing peripheralization
of the Anywaa in the regional economy. Such introspection has also generated nostal-
gia, the second source of inspiration. In their quest for transformation, contemporary
Anywaa draw on the memory of their own ‘village states’ and proudly use terms from
the modern state functionaries to describe their own traditional political system.
Translated into the language of modern state functionaries, nyinya becomes ‘king of
kings’, nyikugu a political adviser, nyibur a vice-president.

6. Such reflexivity is reinforced by the working visits of officials of the regional states to Tigray. The
Anywaa and the Nuer officials were impressed by the developmental achievements of the regional
state of Tigray. The imbalance they saw has added to their acute sense of relative deprivation.
 The Anywaa Response to Ethiopian Ethnic Federalism 161

The Anywaa have sought various explanations for their growing impoverishment
and their limited access to modernity. For the majority of the ordinary Anywaa men
and women in rural settings, the discourse on inequality (between them and the
Highlanders, or between them and the Europeans) is placed in a wider cosmological
scheme: it is understood as yet another ‘plot’ in the long confrontational drama
between the Anywaa and Jwok, as attested in the following extracts from the origin
myth documented by Godfrey Lienhardt and Conradin Perner. Lienhardt (1962: 78)
refers to an Anywaa myth ‘which explains why the white (or red) men are more pros-
perous than the Anuak’:

The white men and the black men and all people were created at one time.
Divinity gave the Anuak a Dog. Divinity became sick, and said to the black
people ‘Give me a skin to be laid in when I die’. The black people said, ‘but
what can we do with a person who is going to die just now?’ [that is, what is
the use of bothering about a dying person?]. Divinity pleaded with black men,
but they were adamant. And divinity said ‘go from me, for you have refused
me this. And all the people were dressed in skins, both Anuak and white men’.
Divinity said ‘You ugly, bad white people give men a skin‘. They did so.
Divinity said ‘Good my son, you gave me a skin for death that I will die on’.
He said to the black people ‘You did not give me a skin for death, so you shall
be poor all your lives. And that skin you would not give me, you will now buy
from my white son. You belong to the dog’. Divinity’s dying wish (gweth) to
white men was that they should travel on the river in boats, and in the air.
That is why white people have received everything, because it was divinity’s
dying wish. The Anuak were left with haunting of the dead (acyeni).

Perner documented a similar myth that resonates with Lienhardt’s:

While God in the beginning had had no liking for the humans, he finally
started to become interested in the Anywaa, found them beautiful and intel-
ligent. As time went on, God began to like them, and one day he asked them
to lick his ass! The Anywaa were filled with indignation and they refused cat-
egorically: ‘you may be great God but we are pure human beings, and we
never agree to such insanity, not even on your demand!’ God got angry. He
poured water and sand in the eyes of the people to make them blind, to be
sure that they would never see, never discover anything … Then he left, he
went to the north, to the red (Highlanders and Arabs) and the white people
(Europeans). Those people fulfilled all his wishes, and he gave them every-
thing in return. (Perner 1994: 113–14)

In these myths, the Anywaa’s lack of modernity is conceptualized as deprivation by

the ‘evil’ God, but also as defiance against God’s imposition. The modernity (or
material advancement) of other people (including the Highlanders) is thus perceived
to have been gained at the expense of their dignity. Referring to a different myth,
Kurimoto (2001) discussed the Anywaa’s perception of their lack of modernity in
162 Playing Different Games

terms of seizure of knowledge and technology by the white people with whom the
Anywaa had once shared it, which Kurimoto calls ‘primordial modernity’:7

In ancient times when white and black men lived together, knowledge and
technology symbolized by ‘iron’ (nyuwei) and ‘paper’ (warakata) already
existed. When the white men left, they took the knowledge and technology,
leaving the black men without them. When white men appeared in
Anywaaland around the turn of the century, they ‘returned’ with the lost iron
and paper. In other words, during the primordial period, the Anywaa once
had access to modernity, and then lost it. Therefore, modernity is not some-
thing to be gained, but to be re-gained. Finally it was brought back by the
British, but access to it has been limited. (Kurimoto 2001: 270)

As the lure of modernity grew stronger, the Anywaa quest for it shifted from a discur-
sive modernity to a more practical engagement. The Anywaa conversion into a ‘modern’
Church – Christianity – was not so much an ‘exit strategy’ from an oppressive tradition
or attraction to a new system of knowledge as it was an opportunity to participate in
modernity. The first Christian church, the American Presbyterian Mission, was estab-
lished among the Anywaa in 1952. The APM was originally valued by the Anywaa as
an agent of modernization, an example of Donham’s (1999) ‘paradox’: a revolution (mis-
sionary Christianity) that started as a rebellion against modernity in the West turned out
to be an agent of modernization in Africa. When the church opened one of the early
modern schools, established the first clinic in Gambella and charted a plane from the
Sudan to bring in modern goods necessary for the evangelization programme, at a time
when even the national capital was new to such signs of modernity, the Anywaa saw con-
version as a vehicle by which to modernize. Christianity itself was identified with the
missionaries and their modern gadgets. It is no coincidence that the Anywaa used to call
the Christians ‘the McClure people’, named after the American missionary who estab-
lished the first church in Akedo village. Akedo village, now renamed by the missionar-
ies as Pokuwo (‘village of hope’), became a magnet for Anywaa of the various villages,
who were attracted by the school, health and agricultural facilities that the mission
station provided. The materialist dimension of conversion is aptly described by a mis-
sionary-cum-anthropologist who visited Akedo village in the 1970s:

The loss of shamanistic services of the witchdoctor holds up some conver-

sions to Christianity. This is why a Christian mission spends its money more
judiciously by establishing ten clinics rather than one centralized hospital.
Some simple, local, personal, medical service is required as a functional sub-
stitute for the shamanistic works of the witchdoctor. This service must get as
close to the village level as possible and should have a religious dimension.
The missions should train a huge army of dressers who can go forth with
medical knowledge and clear faith. (Tippet 1970: 197–98)

7. For a similar local discourse on deprivation of and ‘expulsion’ from modernity see Fergusson
(1999), Expectations of Modernity: Myths and Meanings of Urban Life on the Zambian Copper Belt.
 The Anywaa Response to Ethiopian Ethnic Federalism 163

The Akedo mission church produced the first generation of educated Anywaa elites.
But as an institution of delivering modern goods and services, the church created
rising expectations that it could not meet. In fact, the low profile of the missionary
life and its material asceticism (the mission compound was a mere hut) failed to
sustain its appeal, particularly after the establishment of government schools and
clinics, which broke the missionary’s monopoly over symbols of modernity. The
failure of the APM to meet the growing expectations of modernity led to the emer-
gence of breakaway churches, as described by an Anywaa leader of such a church:

I left the Presbyterians because the Presbyterians were not interested in devel-
opment. The Americans wanted to keep us as we were. The Norwegians came
recently but they have done a lot for us. One of the first modern buildings [the
church’s guest house was furnished with Ikea] was built by the Norwegians,
while the Americans lived like the Anywaa [in a hut]. When they left Akedo
in 1978, they threw all their utensils into the Baro River. They educated only
few Anywaa. Look what the Catholic Church is doing now! They came to
Gambella recently but constructed a very big church, they bought tractors for
the people of Gambella and now they are teaching English language. Had the
Catholics come earlier, Anywaa could have progressed much earlier. (Okok
Ujulu, leader of the breakaway Unity Church, 4 August 2000)8

Disenchanted with ‘mission-modernity’, the Anywaa sought alternative strategies.

Migration was one option. In the 1950s some Anywaa were already involved in wage
labour in the coffee farms of the neighbouring highland region of Dambidolo, not
only to be able to afford the rare and expensive dimui, but also to earn enough money
to participate in modernity by acquiring its media and symbols. By the 1970s and
1980s the attraction was gold mining, which promised a shortcut to modernity
through the relative ease with which one could make a fortune because of the
favourable national and international gold markets (Kurimoto 1996). In this instance,
the Anywaa needed the money not only to acquire the signs and symbols but also the
means for modernity: paying for their education in the towns and gaining independ-
ence from family support in the villages. When the Ethiopian revolution broke out in
1974, therefore, the first generation of educated Anywaa elite were already ready to
join the road to ‘progress’, which for them meant catching up with ‘the immediate
moderns’: the Highlanders. However, despite the euphoria that accompanied the rev-
olution, the Derg failed to deliver and meet these local expectations.
The 1990s opened up new avenues of social mobility, and the possibility of grav-
itating toward the perceived home of modernity – the Western world. Adopting the
Nuer strategy of alternative citizenship to cope with marginalization, and making use

8. With the term, ‘Norwegians’, the narrator is referring to the Norwegian Church Aid, an ecu-
menical organization that is actively engaged in community development in many African
countries; and by ‘Catholics’ the narrator means the Catholic Church in Gambella and the
Salesians of Don Bosco fathers. In the eyes of the narrator, however, there is little difference
between these organizations – both have a ‘modernist’ drive.
164 Playing Different Games

of the escalation of conflict in the Anywaa-inhabited areas in southern Sudan, a

group of Ethiopian Anywaa tried to join refugee camps in Ethiopia by ‘passing’ as
Sudanese. Even then it was difficult to manipulate the aid agencies. As a last resort
they went to refugee camps in Kenya, from where some managed to emigrate to
America, Canada and Australia through the UNHCR resettlement programme.9
The 1990s also brought new possibilities for local modernism at home as well.
Decentralization in post-1991 Ethiopia promised local empowerment in which
statism became the latest edition of the modernity project. If some Anywaa went to
the West where modernity ‘originated’, others sought to achieve modernity at home.
This was perceived to be possible for the Anywaa only if they had a state of their own,
or at least used the new regional state of Gambella as a vehicle to rehabilitate and
transform Anywaa society along modernist lines. In the anthropological literature on
modern elites, however, they are often represented as utilitarian manipulators and
instrumentalists (Brass 1999). Elites rarely feature as actors genuinely interested in
social progress. The instrumentalist approach is helpful to understand a certain
portion of social reality, but it does not tell us everything about the multiple projects
of the elites. We should allow for the possibility that the elites have a genuine inter-
est and belief in progress. Societal concerns such as ‘progress’ partly inform the polit-
ical practice of a section of Anywaa elites, as the following commentary by an
Anywaa politician suggests: ‘Society cannot develop without a state. Above all the
state makes the people work. The Anywaa will disappear because they no longer
work’.10
The belief in the agency of the state to transform society is more evident among
the Anywaa than the Nuer, who are engaged in individualist strategies to achieve the
same end. Drawing on their own experiences in village-states, but increasingly aware
of their inadequacy in the changing economic and political realities, the Anywaa nev-
ertheless believe in the transformative power of the state:

Life was better during the British [enclave] period. Agriculture was flourish-
ing, trade was expanding. Ships were coming from the Sudan. There were
even [trading] companies. British horticultural sites are still remembered by
the Anywaa as agelgacher [sic, meaning agriculture]. The British soldiers used
to export vegetables from Gambella to Dambidolo. The Anywaa were then
hard workers. Not like now. In fact, they were also hard workers before the
British. They used to have very good harvest all the year round because the

9. Members of the Anywaa diasporas held meetings with the American Presbyterian Church,
whom they reprimanded for failing to bring the Anywaa to the United States where living con-
ditions are regarded to be much better than they are in Gambella.
10. Extracted from informal discussions I had with Abula Agwa, senior official of the Gambella
Peoples Democratic Congress. Abula is an ex-Derg soldier, aged 45. He served in the
Ethiopian army for fifteen years and lost his leg fighting the EPLF in Eritrea. He came back
to Gambella in 1992 after the Derg army was disbanded. Abula then started a small business
in Gambella town. He has travelled widely in Ethiopia and has a particular liking for the city
of Asmara, which he admires for its cleanliness and for being ‘modern’.
 The Anywaa Response to Ethiopian Ethnic Federalism 165

kwaari made the people work. Things started changing after the refugees and
the safara [resettlers] came. Fifty kilograms of maize used to be sold at 15–20
birr but then 1 quintal was sold at 4–5 birr!! The Anywaa bought the maize
and started selling borde [local beer]. Anywaa stopped working and the
women became like slaves. As borde was profitable, selling 100–200 birr from
a quintal of maize, the women started overworking, pounding all the day.
Production reached its lowest because there was no incentive. Those who did
not have cash to buy the maize started catching some fish for food and selling
it to buy maize. The main problem in Gambella is neglect by the state. The
federal government is not treating all the regions equally. There is more devel-
opment in Tigray. Take the case of the Alwero dam. Why is it difficult to
build canals and complete the dam, whereas many development projects were
completed in Tigray after Alwero dam? That is why Okello Oman [the first
Anywaa president of the GPNRS] once angrily said, ‘We do not need dams
for crocodiles and fish!’ We need our own state, free from control. During the
transitional period Gambella was more advanced than Benishangul and other
regional states because it was governed by educated Anywaa. They have now
all left and been replaced by incompetent people by the federal government.
(Anywaa NGO worker, Gambella town, July 2000)

Statism is not only adopted as a means of modernization by the educated elites but it
is shared by ordinary men and women, who also reflect on the socio-economic decline:

What Anywaa need is a strong state of their own ruled by educated people
who could initiate transformation. It is only then that we could avoid the
death of Anywaa society. The British were strong and life was in order during
their time. No Anywaa from the villages was allowed to roam around the
town, and they were encouraged to work hard. They came to the town to buy
goods and went back to the villages. Now you hardly find a diligent Anywaa
farmer in the village. They are all here in Gambella town, leaving the land for
the Nuer. (Ariat Abala, Anywaa woman from Gambella town, aged 60,
September 2000)

It is in this social context that the urge to ‘capture’ the Gambella regional state by the
Anywaa should be placed, for it serves different purposes. It is the converging point
for the multiple concerns of the various categories of Anywaa. For the elites, it creates
an ideology of power as well as expressing their societal concerns. For the ordinary
Anywaa men and women, it is perceived as a guarantee to contain the Nuer and a
promise for their rehabilitation. In inter-ethnic terms, it has implied a certain appre-
hension in power-sharing with the Nuer, for the discourse has inserted a zero-sum
game: either the Anywaa keep the status quo or the Nuer will progressively corrode
their hard-won political guarantee that ensures the continuity and revitalization of
Anywaa society. As the aforementioned narratives illustrate, there is a growing
anxiety and concern about the continuity of Anywaa society, the vitality of which has
been sapped not only by external pressures (specifically land encroachments by their
166 Playing Different Games

pastoralist neighbours and the adverse impact of state expansion) but also by inter-
nal strains, i.e., the breakup of social control, the depopulation of villages and the
decline in work ethic.
This statism, coupled with a reference to settlement history, settlement patterns
and contributions to regime change, provides the Anywaa elites with political justi-
fication in claiming the GPNRS as an Anywaa regional state at best or, at worst, to
claim their status as a political majority in a multi-ethnic regional state. In the eyes
of the Anywaa this is a legitimate demand. If the Amhara, Tigreans, Oromo or the
Afar are each allocated a regional state in their respective ‘homelands’, then the
Anywaa also deserve the same political rights in a region that they regard as their
home; or, at the very least, they deserve to occupy a dominant political position in a
region that is defined as multi-ethnic. It is no wonder, then, that whenever a new
power-sharing arrangement has been proposed by the federal government in address-
ing the issue of proportional political representation, the Anywaa have interpreted it
as a usurpation of their legitimate dominant political status.
To sum up, the long-term dispossession of the Anywaa by the previous regimes
had created among the Anywaa a sense of relative deprivation, which they sought to
redress in symbolic terms as well as through practical politics. This is particularly the
case in their attempts to take advantage of the new political order in post-1991
Ethiopia. In this context, the Anywaa have claimed political ownership of the
Gambella region and formulated exclusivist discourses that are in harmony with the
ethnopolitics of the EPRDF. Although territoriality has deep roots in the Anywaa
cultural world, its radical manifestation in recent times is related to the exclusive
claims that have been generated within the new political structure in Ethiopia.
Outnumbered by the Nuer and dealing with the growing number of Highlanders,
the Anywaa have invoked autochthony as the only available political language to
justify their power claims. This has brought them into a multi-dimensional conflict
with the Nuer, the Highlanders, and the EPRDF. 13 December 2003 is a landmark
in the new process of Anywaa political decline in the regional power politics. The
Anywaa’s attempts to regain political power through guerrilla warfare against the
EPRDF, though initially successful, further undermined their political standing in
the Gambella region. As Anywaa power waned, that of the Nuer waxed. The Anywaa
political decline has made Nuer access to natural resources easier, to which the
Anywaa react begrudgingly. In the following chapter the Nuer response to ethnic fed-
eralism is discussed with special reference to the way in which they have contested
Anywaa political dominance and have sought to establish connections with the
Ethiopian state. In both cases the Nuer have drawn on the cultural repertoire in
which their political strategies are embedded.
 Chapter 8

The Nuer Response to

Ethiopian Ethnic Federalism
Building on Chapter 7, this chapter elaborates further on the power variable, specifically
on the ways in which relations with the Ethiopian and also the Sudanese state may be
viewed as causes of ethnic conflict between the Anywaa and the Nuer. It emphasizes Nuer
contestation of Anywaa political dominance in the GPNRS in the 1990s through the for-
mulation of counter-claims to power based on creative ideologies of ethnic entitlement.

The Nuer Politics of Inclusion

Political domination of the Gambella regional state by the Anywaa in the 1990s gave
rise to resentment and created solidarity among the Nuer. Anywaa dominance pro-
duced three levels of concern for different categories of people. Nuer elites were con-
cerned mainly about their exclusion from positions of political leadership and from
the rewards of office. For ordinary Nuer men and women in urban areas, Anywaa
dominance meant exclusion from or marginal access to modern goods and services.
For the ordinary Nuer men and women in the villages, it brought with it the politi-
cization of their livelihood strategy, pastoral mobility. Within the opportunity struc-
ture that was created by the new Ethiopian constitution of 1994, the Nuer sought to
assuage their grievances by pursuing a particular form of identity politics.
While commenting on the altered political status of Gambella in post-1991
Ethiopia and on the trickle-down effect of the decentralization project, an educated
Nuer, employing a pastoralist metaphor, described the situation of the Nuer within
the new structure as bi jile duoth (‘we are given the leftover’):

Duoth is what remains after you make butter out of milk. It is given to chil-
dren who are still happy getting something. Bi jile duoth means to give some-
body a small thing without offending him. The receivers are even surprised
that they get something, because they do not know that others get bigger and
better things. It is the same with ehadeg [EPRDF]. The Tigre, the Amhara and
the Oromo got many things [from the state], and the people of Gambella were
given what remained. But the people of Gambella were happy that they got
something that they had never had before. Now the Nuer and the Anywaa
have become ambassadors, ministers and have their own kilil [regional state].
(Gatluak Choul, Nuer civil servant, Gambella town, 23 May 2001)

A more positive view of the new opportunity structure was expressed by another edu-
cated Nuer from Gambella town:
168 Playing Different Games

As late as 1995, many Nuer were not interested in becoming Ethiopians,

partly because at that time there were security problems in Gambella and
perhaps partly because they all like to be southern Sudanese! Now the Nuer
are enrolled in the Ethiopian Federal Army. We have three army officers, a
police force, and all the Nuer districts are manned by Nuer police, and the
senior provincial commissioners of Akobo, Itang and Jikaw are from the Nuer.
All the Tigreans, Amharas and Oromo [Highlanders] are excluded from polit-
ical positions, except perhaps for the political advisers. For the first time in a
hundred and fifty years [since the eastward migration of the Jikany], the Nuer
have accepted that they are Ethiopians. The notion that ‘buny cie turuk’ is not
valid any longer! (educated Nuer, Gambella town, 27 January 2002)

Literally, buny cie turuk means ‘the Highlanders are not modern’. Metaphorically, it is
a negation of the Highlanders’ self-image as more ‘modern’ than the Anywaa and the
Nuer. Turuk is a generic term for state power and modernity, originally used to refer
to the Ottoman Turks, the first ‘modern’ people the Nilotes encountered in southern
Sudan early in the nineteenth century. In the eyes of the Nuer, the Ethiopian state
failed to deliver as much as it ought to have. The Nuer also used to describe the
Highlanders as turuk mi thil kade (‘civilization without salt’), in reference to the intro-
duction of salt to the Nilotes by the Turks and Egyptians, whereas the Highlanders
could not deliver it to the Nuer in Gambella before the 1990s. Unlike the Ottoman
Turks and the Anglo-Egyptians in southern Sudan, imperial Ethiopia was short on salt
supplies, particularly in the western parts of the country. Salt becomes a metaphor for
the failure of expected deliveries of goods and social services by the state.1
The 1990s, however, promised new reward structures for Gambella. As the relief
regime declined, the establishment of the Gambella regional state opened new avenues
of social mobility and individual advancement for those who could claim Ethiopian
citizenship. Marginalized by the Anywaa from the distribution of these new rewards in
the 1990s and increasingly aware that in the new political game administrative power
mattered most, the Nuer contested Anywaa political domination through ethnic coun-
terclaims and creative strategies of entitlement. Against the backdrop of such expecta-
tion, the political exclusion of the Nuer elites by the Anywaa elites in the regional state
produced a sense of relative deprivation, which became especially acute because the ref-
erence group (the Anywaa) was perceived as a political minor in the traditional rules of
the game. This sense of deprivation created a social energy that resulted in the intense
Nuer politics of inclusion from 1991 to 2002. Since then, however, as the federal gov-
ernment has delivered more than ever to its peripheral areas, the Ethiopian state, in the
eyes of the Nuer, has ‘joined’ the club of ‘moderns’. In this new context of changing
perceptions, buny cie turuk is no longer valid.
The second level of concern was characteristic of ordinary Nuer men and women
for whom the object of the struggle was access to modern goods and social services, par-

1. This is an interesting local criticism of maqinat, Ethiopia’s version of the ‘civilizing mission’.
Maqinat was used to justify the conquest of territories in the southern, eastern and western
parts of the country in the second half of the nineteenth century.
 The Nuer Response to Ethiopian Ethnic Federalism 169

ticularly education and health facilities. For this category of Nuer, identity politics deter-
mined who was and was not allowed access to the new resources delivered by the state.
They took an active interest in ethnopolitics, inasmuch as their respective political
parties claimed to represent and articulate their demands. In line with the ‘roadside’ bias
of development in the so-called Third World countries, social services in Gambella were
concentrated in and around the towns. As a result, urban areas, where the few social
services were concentrated, became the places where crucial economic networks were
established and where people engaged in politics. Consequently, there was a heavy influx
of people into the two regional towns, Itang and Gambella. Towns in Gambella region
exhibited an annual growth rate of thirteen per cent, the highest being in Gambella
town itself. According to the 1994 census, fifty-seven per cent of the urban population
and twenty-nine per cent of the rural population in the GPNRS were migrants. Rural-
urban migration involved all groups, but Nuer migrations to the urban areas (nearly all
in Anywaa territories) gained increased political visibility. Here again, the previous ref-
erences to Merton’s concepts are useful where we have different interpretive schemes: the
manifest function – the subjective motivation of individual Nuer movements to the
towns to have access to social services – is interpreted through its latent function – Nuer
expansion at the expense of Anywaa territories. The Anywaa even planned to change the
regional capital from Gambella town to Abobo town on the grounds that Gambella
town was very close to Nuer settlements. For the Anywaa, the rural-urban migration
produced further evidence of the Nuer ‘hidden agenda’. For the ordinary Nuer men and
women, migration to the towns was part of their quest for modernity:

Towns are not Oromo, Amhara, Anywaa or Nuer. People mix and something
more comes out of it. It is only the Anywaa who prevent people from coming
to the towns because they want to have the good things in the town for them-
selves alone. The Nuer like to go to the towns because there is no peace and
development in their areas. The only town we have is Kurgeng, and even that
is only three to four years old. But the Anywaa have Gambella, Itang, Abobo
and Pinyudo towns. (James Tut, a nuer law student in Civil Service College,
Addis Ababa, March 2001)

Part of this quest for ‘the good things in the town’ was the growing realization that indi-
vidual and group advancement was no longer measured and defined by tradition alone,
but also by new means of social mobility, particularly modern education and connec-
tion with the state system. This is best illustrated in the decline of gar in defining social
status. In the context of integration into national identities and in the context of the
diaspora, gar (like other initiation marks) was increasingly seen as a sign of backward-
ness. Hutchinson (1996: 270) discussed the emerging debate among the Sudanese
Nuer concerning gar as follows: ‘During the early 1980s there was a tremendous debate
brewing among Nuer concerning the ultimate significance of male initiation, the his-
torical conditions that gave rise to this rite, and its contemporary socio-political rele-
vance … Is gar primarily a means of ethnic identification during a period of intense
inter-tribal warfare that is now no longer necessary? What distinguishes the Nuer as
people? Is garring an indispensable element of this distinction or not?’
170 Playing Different Games

According to Hutchinson (1996: 270–71), the first move away from gar began
among the western Nuer in south Sudan in the 1940s in situations of increased
contact with Arabs and others, who do not prescribe the same kind of initiation
marks. In this new contact situation, gar came to be viewed as a badge of backward-
ness. The introduction of modern education and the spread of literacy also produced
a new generation of uninitiated, literate Nuer who competed for local political power
(chieftaincy) with initiated, non-literate traditional elites. In this new political
context, having gar or not having it became a means of inclusion or exclusion from
the power game. The initiated, non-literate Nuer called adult males who were unini-
tiated and literate tuut dhoali – in Hutchinson’s (1996: 270) words, ‘a marvellous
oxymoron that clearly conveyed their liminal status’. The term may be translated as
‘adult-boys’ or, as Hutchinson suggests, ‘bull-boys’. The ‘social deficiencies’ of the
tuut dhoali, as described by the initiated, were manifest in three crucial domains: wut
(the ideology of masculinity); rich (membership of an age set on the basis of the gen-
eration of the initiated), and mut (the lineage spear that is transferred from father to
son at the time of initiation).
The tuut dhoali have managed to renegotiate their ‘social deficiencies’ through
the growing realization in the general public that education now serves to buffer the
arbitrariness of the state and allows people to enjoy its rewards. The social position
of the tuut dhoali was strengthened in the 1980s during the second Sudanese civil
war. In their quest to transcend ‘tribal divisions’, some of the leaders of the southern
Sudanese liberation movement attempted to ban initiation marks. This was champi-
oned by none other than Dr Riek Machar, a prominent Nuer commander of the
SPLA and himself a tuut dhoali (Hutchinson 1996: 296–97).
A similar debate, drawing first on the discourse on ‘backward cultural practices’
of the imperial period and then on the revolutionary rhetoric of the socialist period,
arose among the eastern Jikany Nuer in Gambella, beginning in the 1960s. The def-
inition of the non-Amhara, non-Christian cultures of Ethiopia, particularly of the
border regions, as ‘backward’ led to self-deprecation and low self-esteem among
those local elites who internalized the hegemonic discourse. In fact, in their drive to
‘catch up’ with the ‘cultures of the advanced people’, some educated Nuer cam-
paigned door-to-door to discourage the practice of gar. The local administration of
the Jikow district even went as far as prohibiting those Nuer who were freshly initi-
ated from entering the district’s capital. Caught between the modernist pressures of
the state and their own educated people, the ordinary Nuer men and women came
up with an ideological ‘exit option’ that mediated the transition. One of these ideo-
logical ‘exits’ was the representation of gar as an imposition by colonialists. In fact, a
myth circulated among educated Nuer that gar was a British invention to make a dis-
tinction between Nuer and Dinka. An extreme version of this myth took the form
of a conspiracy theory and trickled down to the villagers:

Gar is to be wut [man]. But what is wut if a Nuer cannot match the Anywaa or
the buny [Highlander]? If a Nuer fights with a buny, the buny could defeat the
Nuer, though he does not have gar. Who is wut then? I say that gar is nothing,
it is useless. It is possible to be wut without gar. Why can we not make turbil
 The Nuer Response to Ethiopian Ethnic Federalism 171

[car] or nhial bany [airplane] if we are wut? People who made all these things
do not have gar. Gar started with the British during the age-set of Rok. Their
mothers did not want their children to have gar. They thought their sons would
die. When the British started gar, the Nuer did not know about the objective.
The British deceived us while they were sending their own children to schools.
If I were to be born again, I would never have gar because it is possible to be
wut without gar. Our forefathers did not know anything. That is why they took
other people [assimilation]. Rather they should have formed their own kume
[government]. We do not have our own kume because we are not educated.
The milwal [red/Highlanders] and mibor [white/Europeans] have boum
[power] because they have their own kume. During the time of gar, we did not
know about cloth. Now our children are wearing clothes because they are edu-
cated. We are just following them. (David Doup, Christian Nuer elder from
Nipnip village, Addis Ababa, 23 November 2000)

David is a well-known Cieng Cany elder in Jikaw district. I met him in Addis Ababa
while he was waiting for the remittance that the Cieng Cany community in the US
was sending to fund the building of a school in Nipnip. With the term ‘kume’, David
is referring to the role of the state in distributing goods and services, which were
lacking, however, in his remote village. The desire ‘to have what other people have’
had generated a sense of relative deprivation expressed, in this instance, in the form
of differential access to education. Underlying the changing attitude towards gar was
the role of modern education as a new avenue of social and personal advancement,
which, at the same time, provided an alternative discourse on manhood. There is an
overarching air of realism in the above narrative by David, in that the educated [the
Highlanders] have better conditions of life and are politically dominant at the
national level. David and his contemporaries saw sending their children to school as
the way to catch up with the powerful.
The Nuer quest to ‘catch up’ has also resulted in attraction to the urban centres.
The Nuer influx to the towns was paralleled by the emergence of settlements in the
nearby Anywaa villages. The emergence and expansion of the Nuer settlement at
Ochom in the Anywaa village of Pinykew is a case in point. In 1985, Riek Tuany, a
Cieng Cany Nuer from Nipnip village, made contact with Anywaa leaders in
Pinykew village. Riek initially came to Ochom because his child was ill with back
pain. The child was admitted to Gambella hospital, but Riek could not stay in the
town for a follow-up visit because he was unable to cover the medical expenses.
Instead he thought he could settle in Gambella town, only to realize that life there
was very expensive and not suitable for keeping cattle. As a last resort, he returned to
Ochom, with which he was already familiar as a transit stage for the cattle trade in
Gambella town. Riek was well received by the Pinykew Anywaa. He rented land for
cultivation to support his family and, a little later, brought some of his cattle there
from Nipnip village in Jikaw district. Riek was soon joined by his close relatives, out
of which a small Cieng Cany settlement emerged in Pinykew.
The successful establishment of the Cieng Cany encouraged other sections of the
Gaat-Jak to settle in Ochom. Both the Anywaa hosts and the Nuer guests defined the
172 Playing Different Games

situation in moral terms: the legitimate right to have access to the social services in
Gambella town. In addition, the settlement of the Nuer in Anywaa territories was
received partly as an opportunity by the local Anywaa. For the ordinary Anywaa, it
meant access to cattle wealth, either in the form of bridewealth payment or by
rearing cattle in sharing arrangements with their Nuer friends. For the Anywaa local
leaders, it meant increased political power. It was a great relief for the Anywaa leaders
in Pinykew village not to have to force their fellow Anywaa to pay tax, which was
covered instead by the Nuer. Besides this, the fines collected as a result of Nuer court
cases in Ochom village were shared by the Anywaa and Nuer leaders. The Nuer in
Ochom left Pinykew in 1991 during the regime change and the unrest that followed.
Riek returned to Ochom in 1993. In order to re-establish ties and develop confi-
dence with the Pinykew Anywaa, Riek married an Anywaa woman. Other Nuer fol-
lowed suit, particularly the representatives of each Nuer clan. The marriage ties
facilitated socio-economic exchanges, encouraging more immigration. In the new
context of ethnopolitics dominated by the Anywaa power elites, the Nuer combined
local strategies with new forms and ideologies of ethnic entitlement, in order to gain
advantages in the regional power game and to stake claims on local resource.

Projecting the local onto the national

Confronting the Anywaa exclusionary political practices, the Nuer first looked
inward to make sense of their new situation. When the Anywaa said ‘the land is ours’,
the Nuer replied ‘there is enough land for us all’, setting up a distinct ‘clash of cul-
tures’. For the Anywaa, land is as much an economic as a symbolic resource in iden-
tity construction. For the Nuer, land that is not actually settled and effectively
occupied by the Anywaa is considered to be part of the economy of the commons.
The Anywaa settlements concentrated along the banks of the rivers are largely
respected by the Nuer; in most cases, the Nuer use peaceful strategies to have access
to the Anywaa’s riverine lands. Beyond the riverine area, however, the Nuer do not
consider themselves to be encroaching on other people’s land.
In the resulting entitlement debates and contestation of values, the Nuer impose
their pragmatic and situational identity discourse and their concepts of land rights
on the Anywaa, and they also project them onto and contrast them with the state’s
fixed concept of citizenship. This is evident in the way that the Cieng Reng formu-
lated their claim to Ethiopian citizenship, as the following narrative extract suggests:

When I first came to Makot, it was forestland. There was nobody living there.
[Note: Traditionally, the Makot area has been part of the Anywaa village of
Pinyman.] When the other Cieng Reng heard that the area was good, they
came to Makot. That is how Makot became a big village. Now it has already
been eighteen years since we settled at Makot. It has become our wech [village].
It is not only we who move. Many people are going to America: the Dinka,
Anywaa, Nuer, and buny. But the America kume does not say ‘Go back to
your country’. And if we leave Yom and come to Makot, this should be
allowed. You can change kume as you like. If Ethiopians want to go to Sudan
and stay there, the Sudan kume cannot prevent them. That is the case I am
 The Nuer Response to Ethiopian Ethnic Federalism 173

representing. We left the Sudan when that kume took our cattle and forced us
to make roads [corvée labour]. That is why people are now coming to the buny
kume. If the people of Yom want to be Ethiopians, they can, as other Nuer
did. It is also the same with the American kume. They accept people because
they want to be many. If we are Sudanese and want to be Ethiopian, what then
is the problem? The kume still accepts people. Our children left Sudan when
the problem started with the Jalab [Arabs]. Previously the Nuer were with the
British kume. But later on they were divided. Part of the Nuer became buny.
That is why we supported the British and the buny when they fought the
Italians [during the Second World War]. When the British left, we became
Sudanese. When the war with the jalab [Sudanese Arabs] started, we became
Ethiopians. We got education and food from the buny. We are happy because
our children are getting education in Gambella. The Ethiopian kume has
become responsible for our children. Up to now, we are happy. That is what I
know. (Kong Diu, Addis Ababa, November 2000)

One of the arguments put forward by Kong in the Cieng Reng’s politics of recogni-
tion is the eighteen years of residence in Ethiopia which, in Nuer terms, is ‘more than
enough’ to claim local identity. This pragmatic current of flexibility in identification
is well expressed in Kong’s narrative: ‘If we are Sudanese and want to be Ethiopian,
what then is the problem?’ There is no fixation in Nuer identification that contains
a strong demographic bias: one can change tribal identity as the situation demands.
In this identity discourse, immigration is something to celebrate, not a threat. This
contrasts strongly with the apprehensive attitude of the Anywaa towards immigra-
tion. The Anywaa justify their immigration concerns by referring to the European
model of the nation-state and the resilience of its border:

It is migration which is affecting politics in Europe. The German and the French
are angry because a lot of people are going there and disturb their system. They
are concerned because if more and more people go there, who would the land
then belong to? They fear that they would be a minority in their own country.
That is exactly what we are saying. We are not saying that Nuer should not be
allowed to use the land and the water or even live together with the Anywaa but
they should respect that the land belongs to the Anywaa. Germany and France
are concerned with immigration because they know that democracy favours
majorities and more foreigners would mean more power to them. Once they are
in, you cannot say no because they start claiming. Even after the EU was estab-
lished the state borders are still valid in Europe. (Abula Obong, senior official of
the Gambella Peoples Democratic Congress, Ruiru, 22 February 2002)

Among the Nuer, national identification, like tribal and ethnic identification, is per-
ceived to be a matter of individual choice, rather than an ascription:

At the beginning there were two kume: the British were with the Nuer, and
the buny kume was with the Anywaa. That was the difference. Then some
174 Playing Different Games

Nuer became Sudanese, others became Ethiopian. The Nuer who live with
the buny are buny. Those in the Sudan call themselves Sudanese. The kume
likes everybody. They dislike only those people who work against them. If the
buny kume and the Sudanese kume fight, if the Sudanese kume rejects us, if
they treat us badly, and if we come to the buny kume, the Sudanese kume
cannot follow us because we are no longer Sudanese. If the buny do the same,
we will be men of Sudan. (Kong Diu)

In Kong’s narrative, the Nuer project their concept of localization onto the nation-
state, as if the latter were characterized, like the Nuer, by a constructivist mode of iden-
tity formation, but on a larger scale. The Nuer attribute paramount importance to
localities with an expressed interest in newcomers. By the same token, the Ethiopian
state is expected to ‘celebrate’ when new people join, since, for the Nuer, politics is
essentially viewed as number politics: the bigger you are, the stronger you become.

Myth as an ideological resource at the local level

Confronted with the new Anywaa political power, the Nuer, particularly those in the
new settlements in Anywaa lands, have overstated a narrative of self that enhances
relatedness with the Anywaa. If the Anywaa political discourse is organized along the
narrative of loss and separateness, the Nuer are engaged in a selective memory that
enhances connectedness: with the accent being on the shared. Towards that end, the
Nuer draw on the mythohistory that fits their inclusive strategies.
The reinforcement of Anywaa territoriality in the context of the ethnopolitics organ-
ized by the Ethiopian state has generated a greater reflexivity on ideas of origin. If the
ethnogenesis of the Nuer was a subject of an anthropological debate decades ago, it has
now become an imperative in everyday inter-ethnic interaction. In order to familiarize
themselves with the local setting, and as a counter-argument to the Anywaa’s exclusive
strategies, the Nuer often refer to their mixed background and show a keen interest in
myths that connect them with the Anywaa. One such myth is the Jikany origin myth dis-
cussed at length in Chapter 3. Whereas in many versions of the Kir mythology the con-
nection between Kir and the Dinka is emphasized (Evans-Pritchard 1940a; Johnson
1982), currently the Jikany Nuer in the Gambella region highlight Kir’s connection with
the Anywaa. This is particularly true in Makot village where the Kir mythology offers areas
of relatedness between the Cieng Reng and other Jikany Nuer, as well as with the Anywaa.
It is as part of the Jikany Nuer that the Cieng Reng have renegotiated their ‘foreignness’,
both vis-à-vis the Anywaa and their economic competitors, the Thiang Nuer, in their pol-
itics of recognition, as discussed in the case study in Chapter 4. In the Kir mythology, the
Nuer in general and the Cieng Reng in particular have found a discursive resource.
When, during the 1998 conflict in Itang, both the Anywaa and the Thiang Nuer
framed local politics and the local struggle for the control of resources in national
terms by defining the Cieng Reng as Sudanese, the Cieng Reng countered this
attempt to exclude them by invoking ‘shared origins’ with the Anywaa and various
types of relatedness with the Thiang Nuer. The Cieng Reng believe that Tik, the mys-
tical ancestor of the Cieng Reng, was originally an Anywaa. Jal (1987: 18) also estab-
lished the link between the Cieng Reng and the Anywaa. The Cieng Reng construct
 The Nuer Response to Ethiopian Ethnic Federalism 175

relatedness with the Thiang not only through reference to the history of friendship
between Kir and Tik, which is discussed in Chapter 3, but also through kinship links
with Thiang through the Cieng Nyaruny, the secondary division within the Cieng
Reng. The founder of the Cieng Nyaruny lineage is a Thiang called Gil. Gil left the
Thiang after his mother inherited Mut Wiu, Kir’s Divinity Spear, from his half-
brother, Tär, on the basis of the latter’s heroic act of avenging the killer of Gil’s full
brother. Gil married a Cieng Reng named Nyaruny and his descendants are called
Cieng Nyaruny. In the eyes of the Cieng Reng, the ‘Sudanese’ label that was invoked
by the Anywaa and the Thiang to justify denying them access to resources and polit-
ical power lacks historical legitimacy. They are all so closely related, argue the Cieng
Reng, that it makes no sense to call some Ethiopian and others Sudanese.

Myth as an ideological resource at the national level

Another myth currently invoked by the Nuer that serves as an ideological linkage to
the Ethiopian state is what I call ‘the tale of the tail’. This is a story about how a Nuer
prophet, Ngundeng, gave the Highlanders cattle when they were starving.
Ngundeng advised the Nuer to cut off the tail of the black-and-white brindled bull,
tut kernyang, and give it to the Highlanders, as evidence to which the Nuer could
point when they became in need of the Highlander’s support. This myth connects
the Nuer with the Ethiopian state via a discourse of reciprocity between Ngundeng
Bong, the renowned Nuer prophet, and Emperor Haile Selassie. The full account of
the myth is given by Johnson (1986: 242–43), which was taken from a conversation
between a prominent Lou Nuer politician and the grandson of Ngundeng:

It was said that the Ethiopians were coming from Fashoda … they went to a place
called Jiör [Jor]. They were dying of starvation and people were suffering from
malaria in the water. They did not reach the village ... Their leader (kuar) came
through Jiör and found that the Dinka were coming … to the Mound [of
Ngundeng]. He followed the people until he reached the Mound … Ngundeng
said, ‘Why do you look at him? He was brought by Divinity (Kuoth). What he
came for, he will tell us. Give him something to eat’. A cow was killed, and he was
given other things to eat … Then Ngundeng told the Lou that the Ethiopian
should be given some cattle. He gave him about eight oxen to feed them. [The Lou
said] ‘Waa! Son of Nyayiel … how can we give the Lou’s cattle to someone who
comes from a place we do not know? He may kill people because of those cattle
one day’. ‘Sons of my mother’s brother,’ [Ngundeng replied,] ‘you will follow those
cattle one day. If you give a person a good thing, one day, sons of my mother’s
brother, if something happens you will join with those people’ … Ngundeng said,
‘I will give him the … tut kernyang and four other oxen’… He said, ‘Do you know
why I give him the tut kernyang?’ ‘No’. ‘You will one day follow it’.

The historical background of the ‘tale’ is the 1898 Fashoda incident between the
British and French colonial forces, which were competing for control of the Upper
Nile region (Bahru 1976: 74). Emperor Menelik of Ethiopia sent his own small con-
tingent, ostensibly as diplomatic support to the French, but also to try to extend his
176 Playing Different Games

own political influence in the area (Jal 1987: 184). The small Ethiopian contingent
might have met Ngundeng’s people in their difficult journey on the way back home.
The word buny, the Nuer term for Highlanders, was coined by Ngundeng in refer-
ence to a misunderstanding between him and fitawrari Haile. When fitawrari Haile
met Ngundeng, he considered him to be the ‘king’ of the Nuer, which he was not,
and bowed down to him as a sign of respect, which was in line with the authority
code in Highland Ethiopia. In ‘the tale of the tail’ it is indicated that Ngundeng won-
dered why Haile bowed down (a gesture that is called buny in the Nuer language), as
this was alien to him and to the Nuer in general. This is the etymology of the word
buny which the Nuer use to refer to the Ethiopian state and to the Highlanders.
In what appears to be eclipsing an historical gap of more than two decades
between Emperor Menelik (1889–1906) and Emperor Haile Selassie (1930–74),
international and national history are locally interpreted to make sense of present-
day political realities and to legitimize power and resource claims. Jal (1987: 188)
throws light on the confusion of historical names. According to Jal, the leader of the
Ethiopian contingent in Fashoda in 1898 was fitawrari Haile, one of the command-
ers under ras (head) Tassama, the governor of the western Illubabor province. The
similarity of names between Haile and Haile Selassie, coupled with the Nuer’s close
knowledge of Emperor Haile Selassie/Ethiopia during the Second World War, seems
to explain the historical incongruity. Putting aside the burden of proof, the Nuer
focus in this story is on how prophet Ngundeng (and through him the Nuer) ‘saved’
the Ethiopians/Highlanders during a difficult time.
This mythohistorical event was first employed politically to include recent Nuer
experiences related to the civil war in the Sudan and their influx into Ethiopia. As
Johnson (1986: 244) put it, ‘the gift of survival [to the starving Ethiopian contin-
gent] is repaid by survival [the influx of Nuer refugees into Ethiopia/Gambella]’.
Historical authenticity aside, the myth has become an important ideological resource
that discursively connects a local community with the national state. As contact with
the Highlanders has increased, ‘the tale of the tail’ has incorporated new signs and
symbols to further connect the Nuer with the Ethiopian state. A recent addition to
the myth brings the Nuer to the centre stage of core national symbols – among the
gifts that Ngundeng is supposed to have given to the buny is a lion:

An Anywaa person called Werjegor gave a present to Ngundeng called tony

[long, ivory-decorated pipe]. Ngundeng was happy and said, ‘You Bär
[Anywaa], go and live with the buny where you will get lots of things’. When
he slaughtered the ox for the buny and gave them a lion to help them hunt on
their way, the Anywaa and the Komo were there. It was Werjegor who brought
the buny to Ngundeng. When the Nuer took the tail of the tut kernyang as a
remembrance of their relationship, Ngundeng said, ‘You Nuer, when you are
in trouble, you should go to the Wer Jegor people [Anywaa] and the buny’.
And because of the lion, which Ngundeng gave them, the buny became pow-
erful. That is why Haile Selassie put the lion on the flag and on the birr [impe-
rial currency]. You also find tut kernyang on the birr. (Thiang Luony,
Gaat-Jok elder from Akobo district, Gambella town, December 2003)
 The Nuer Response to Ethiopian Ethnic Federalism 177

In this narrative, the Nuer are not strangers to the Anywaa and the Highlanders.
Even if it was the Anywaa who connected the Nuer with the Highlanders (the role
of Werjegor), the relationship between the Anywaa and the Highlanders is, neverthe-
less, encouraged and blessed by the Nuer (through the prophet Ngundeng). The
subtext of the narrative is how the Anywaa ‘owe’ the Nuer for ‘the better conditions
of the life they enjoy’ by virtue of the latter’s living with and proximity to the
Highlanders/Ethiopians.

The Nuer contesting Anywaa indignity

When the mythological linkages failed to achieve the connectedness to which the Nuer
aspired, they began to contest the Anywaa’s discourse about separateness. The history of
migration of the Nilotes and the debate on shared origins created an opening for the
Nuer to deconstruct the Anywaa claim of being the indigenous people of Gambella. The
Nuer counter-arguments were based on two strategies, both of which served to relativize
Anywaa seniority and to undermine their claims to autochthony in the Gambella
region: first, negotiating the scale of Nuer foreignness and, second, emphasizing the
common origins of the Anywaa and the Nuer. It has already been pointed out that the
Anywaa sweepingly define the Nuer as newcomers. This is as much a political exagger-
ation as it is an aspect of the sudden encounters between groups of Anywaa and Nuer
who do not have a history of living together. For the Lull (Anywaa who live in forested
areas), the Nuer are people whom they first met in the 1970s and 1980s, when urban-
ization started in Gambella town. Even then, interaction was in the marketplace and
ephemeral. This was also a time when hundreds of thousands of Nuer from southern
Sudan came to Itang as refugees. Therefore, a certain margin of ‘error’ might be allowed
in the Lull Anywaa’s definition of the Nuer as ‘foreigners’. But for the Openo Anywaa
who border the Nuer, the foreignness of the Nuer is the rhetoric of political mobiliza-
tion. The more the Anywaa emphasize settlement history, the greater the Nuer contest
it: ‘We both came from the Sudan’ is a typical Nuer answer to the Anywaa claim of
indignity. As an aspect of negotiating their ‘foreignness’, the Nuer are also engaged in
new myth-making to counter the Anywaa’s exclusive ownership claim over Gambella.
This is epitomized by the variants concerning the etymology of the word Gambella:

The term Gambella was first coined by the Nuer. It originated from two
words: Gam (half ) and bela (sorghum). After Latjor died, a group of Nuer
travelled to Dambidolo. The leader told his followers to make sure that there
would be enough food supply for the journey because they did not know how
long it would take to reach Dambidolo. When they reached Gambella, he
asked the keeper of the stores how much sorghum was left. He answered him
gam-bel, half of the sorghum. That is how Gambella got its name. (Thiang
Lony, a Gaat-Jok elder, Gambella town, 12 November 2000)

One of the Anywaa versions of the etymology of the word Gambella is similar to that
of the Nuer version:
178 Playing Different Games

A long time ago there were no canoes or a bridge to cross the Baro River. It
was difficult to cross it, especially carrying bell (sorghum). The Anywaa used
to cross the Baro River by throwing their sorghum in it. One person waited
at the other end of the river and the one who was crossing threw in the
sorghum. That is how Gambella got its name, from ‘catching the sorghum’.
(Ujulu Obang, Anywaa elder, Gambella town, 21 January 2001)

Another Anywaa myth locates a different source:

Bell in the Anywaa language means leopard. There were a lot of bell in
Gambella. One day Morri [Morris, the British commissioner at Gambella
Enclave] met a group of Anywaa who just captured a bell. Morris was very sur-
prised that they could capture the bell. He called the area where the bell was
captured gam-bella (the place where the leopard was captured). Gam in Anywaa
means to catch. (Gilo Abula, Anywaa elder, Pinykew village, 3 March 2001)

The terms of the debate on origin and settlement history complicate and obscure the real
issues at stake: the Anywaa have framed their ethnic concerns in national terms, as if their
conflict with the Nuer is an interstate conflict between Ethiopian and Sudanese citizens.
When the Anywaa say the Nuer are foreigners, they use the term Ethiopia to
mean Gambella, although many of the contemporary Anywaa political actors actu-
ally come from the Adongo region of southern Sudan. The citizenship debate is
therefore not a statement of political fact in the sense of who is and is not Ethiopian,
but exists because it helps to frame local concerns in national terms. Framing the
entitlement issue in such terms serves the Anywaa in two ways: it provides rhetoric
for mobilizing the Ethiopian state in their (the Anywaa’s) project of containment and
it serves as a strategy of ethnic entitlement in the Gambella regional state. This
double-edged discourse, however, backfires in some cases. It seems contradictory to
say ‘the Nuer are Sudanese’ and ‘Nasser was ours’, which the Nuer are keen to capi-
talize on. As the Anywaa increasingly rely on the citizenship card, acknowledging
their Sudanese connection appears to be politically incorrect. This has undermined
their ability to create cross-border political networks as much as the Nuer have.2

2. In March 2001, for instance, Adongo Ageda came from Canada to assume the position of
nyinya in Utalo village. Nyiya Adongo was keen to meet the Anywaa president of the Gambella
regional state in Addis Ababa. The president declined the offer for fear that this would be
‘politically embarrassing’ for the Anywaa in Gambella in general and, in particular, for the
Anywaa officials who were busy framing the Anywaa concerns in national terms. As a result,
nyiya Adongo had a very low-profile reception at Gambella airport, being met only by Anywaa
church officials. Adongo’s attempts to contact the Anywaa political leaders in Gambella town
were also in vain. Instead, he organized a meeting of ‘Anywaa intellectuals’, most of whom were
members of the regional opposition party, the Congress (GPDC).
 The Nuer Response to Ethiopian Ethnic Federalism 179

Party politics
The political use of the inclusive myths can be seen as a self-serving discourse to
present the Nuer as locals in the new areas of settlement. Aware of the significance
of the number of districts in the politics of entitlement, the Nuer strove to increase
their number of districts from two to at least four. Towards that end, in 1992, they
established a political party called the GPDUP, the Gambella Peoples Democratic
Unity Party. The GPDUP was delegitimized as a political organization by the GPLM
because it was allegedly established by people from the defunct regime (the Derg)
and noncitizens (Sudanese refugees). It was only in 1995, during the first regional
election, when a Nuer was appointed vice-president, that the GDPUP obtained
political recognition.
The GPDUP proposed the creation of two additional districts for the Nuer
(Nyinyang from Jikaw and Matar from Akobo) to add to the existing districts of
Jikaw and Akobo. Neither the regional nor the federal government heeded their plea.
Consequently, the Nuer resorted to alternative arguments for political empower-
ment, combining the struggle at the level of mythohistory with the appropriation of
the state’s own discourse. Although a political minor in the regional government, the
GPDUP capitalized on the 1994 census results. The census – the objective of which
was to ‘generate data for designing and preparing the development plan and for mon-
itoring and evaluating the impact of the implementation of the development plan’3
– estimated the population of Gambella at 181,862, of which the Nuer comprised
forty per cent and the Anywaa twenty-seven per cent.4 Seemingly overnight the Nuer
were transformed from a largely ‘foreign’ group into an ‘ethnic majority’. The census
offered new capital for the Nuer political actors on the basis of a new ideology of
entitlement: majoritarianism. Using democratic rhetoric, the Nuer challenged what
now appeared not only as exclusion, but as ‘domination by a minority’:

A glance at the 1994 national census reveals that there is a direct imbalance in
resource allocation throughout the country. It was only in 1995 that the Nuer
got the position of vice-president in the Gambella regional state. This arrange-
ment did not take into account the numerical size of the ethnic groups, for had
that been the case, the Nuer would have been given the top position in the
region because they are numerically the majority in the region. The situation
in Gambella is not different from that of Rwanda and Burundi where the Tutsi
minority dominated the Hutu majority in all political spheres, and we all are
aware of the consequences of that policy. (Nyang Baitiok, Nuer official in the
regional council, Addis Ababa, 6 September 2001)

The Anywaa seriously disputed the census results on several grounds. First, they per-
ceived the census as having grossly undercounted the Anywaa areas, for most of the

3. The 1994 Population and Housing Census of Ethiopia, Central Statistics Authority (1995: 1).
4. The areas not covered by the census were parts of Akobo, Gog and Jor districts, whose esti-
mated joint population was about twenty thousand.
180 Playing Different Games

kebeles that were not counted fall within traditional Anywaa territories.5 More impor-
tantly, the Anywaa considered the size of the Nuer population to be inflated by the
influx of refugees:

The increase in the Nuer population involves political turmoil across the inter-
national border, producing a nonstop influx of refugees into neighbouring ter-
ritories in the Horn of Africa. Since the birth of civil conflict in southern
Sudan, Gambella has been hosting hundreds of thousands of refugees from
across war-torn southern Sudan, the majority of whom belong to the ethnic
Nuer. This makes it very difficult for the authorities in Gambella to differen-
tiate between who is Ethiopian or not in the struggle for power and resources.
Prior to the mid 1980s, only few Nuer lived in Itang and Gambella. (Nyigowo
Oman, educated Anywaa, Gambella town, 14 October 2000)

The Nuer have pursued majoritarianism as a counter-strategy to the Anywaa indige-

nous claim. The more the Nuer insist on majoritarianism, however, the more the
Anywaa invoke the Nuer annexation of their lands since the nineteenth century. This
also means a special interest in immigration issues, for they have a direct bearing on
the demographic equation. It is against this backdrop of the politics of numbers that
the GPDUP officials took an interest in the Cieng Reng quest for kebele as described
in Chapter 4. The GPDUP managed to significantly increase Nuer political repre-
sentation after the 2000 regional election. By 2005, the Nuer had achieved parity
with the Anywaa in representation on the regional council. Out of the eighty-one
seats in the regional council during the 2005 election, the Anywaa and the Nuer were
allocated thirty-three each.

Going national to be local

The more the Anywaa frame their resource and power claims in national terms
(‘Anywaa are Ethiopian and Nuer are Sudanese’), the more national the Nuer
become. Here we find the Nuer using constitutional arguments of entitlement; as
much a perception as it is a real interpretation of the legal documents. The Nuer con-
stitutional argument for inclusion draws on fragments of history: Ethiopia’s loose
frontier policy during its competition with the British colonial forces in the first half
of the twentieth century, as well as recent constitutions that have tended to recognize
an inclusive argument.
Ethiopia currently lacks a law or policy on refugee matters. During the imperial
regime and the early period of the Derg, the trend was towards a more integrative
approach to refugees and local people. Refugees needed only three years’ residence to
claim Ethiopian citizenship (Klinteberg 1977: 2). It is this provision that is remem-

5 . Anywaa doubts concerning the results of the census were based particularly on figures from
Akobo district where, according to the census, only one Anywaa was counted. My own village
census in Pone kebele alone shows that there were more than one hundred Anywaa living there
together with seven hundred Nuer, and that there were as many as six such mixed-settlement
kebeles in Akobo district.
 The Nuer Response to Ethiopian Ethnic Federalism 181

bered by contemporary Nuer as ‘the constitution’. The Ethiopian Electoral Law

(1995), which states that citizens who reside for three years in a locality have the
right to elect and to be elected, is also used by the Nuer as a ‘constitutional’ argu-
ment to claim Ethiopian citizenship. According to the new Consolidated Version of
the Election Law (2005), ‘Any person who has been residing within the constituency
for at least six months may be registered as elector [voter]’ (Art.19.1). It is also stip-
ulated in this law that ‘a candidate is one who is versed in the vernacular of the
national region of his intended candidature’ (Art.38.a) and ‘who has been regularly
residing in the constituency of his intended candidacy for two years preceding the
date of election’ (Art.38.d). In reality, the electoral law does not raise the citizenship
issue; it takes it for granted that those who are eligible to vote are already Ethiopians.
The constitution of the GPNRS is vague about land rights, although it adopts
the land policy enshrined in the Ethiopian constitution, which asserts state owner-
ship of land on ‘behalf ’ of nations and nationalities: ‘All rural and urban lands as well
as natural resources belong to the regional state and its peoples. This is not subject
to sale and it is the common property of nations, nationalities and peoples of the
region’ (Art.34: 3). In line with the Ethiopian constitution, the regional constitution
also explicitly recognizes the land rights of the farmers/pastoralists and the state: ‘The
farmers of the region have the right to obtain land and protection from eviction’
(Art.34: 4); ‘the nomads of the region have the right of access to grazing and culti-
vation land and protection from eviction’ (Art.34: 5); and ‘with all due respect to the
land ownership rights of nation, nationalities and peoples, the state has the right to
allocate land to private investors with legally recognized payments’ (Art.34: 6).
The mode of implementation of these constitutional articles, however, is to be
‘specified by law’. Despite the constitutional provisions, the state’s claim has not yet
made an impact on customary laws. As a result, land sales and informal land
exchanges continue to be unregulated. So far, only a small portion of land has been
apportioned to private investors and two areas have been designated as parkland. The
state discourse on the land becomes relevant at local levels only where it connects
with the Nuer discourse. Aware of the scarcity of key natural resources in their own
areas, now formally recognized as their districts (Jikaw and Akobo), and finding
themselves vulnerable to the Anywaa’s politics of exclusion in the new settlements in
Anywaa areas, the Nuer discursively empower the state as the ultimate owner of the
land. This entails a switch of reference from God (Kuoth) to the state (kume).
Commenting on the Anywaa exclusive land claim, Kong Diu said:

As you know, land is owned by the kume. The kume is the father of all
people. If there is no kume, those people who do not like Nuer [the Anywaa]
say ‘the land is ours’. Everything is from the kume. If there is hunger, if there
is no rain this year, the food comes from other places. You can contact the
kume and it brings you food. The kume is like Kuoth. Everything belongs to
the kume. Land is for the kume. The people are for the kume. That is why it
asks for people [conscription] when there is war. Even this tree belongs to the
kume. There is nothing that the kume cannot do. Nobody can take away the
land of the kume. (Makot village, 17 November 2000)
182 Playing Different Games

Such discursive empowerment of the state neither recognizes the state claim over the
land nor represents a commitment to national identity, but rather serves as an argu-
ment to nullify the Anywaa’s exclusive claim over the land. This argument, however,
has the effect of reproducing state ideologies at the local level, an aspect of what
Joseph and Nugent (1994) called ‘everyday forms of state formation’. There is more
in Kong’s narration. The word kume is used not only in reference to the Ethiopian
state but also to the Highlanders who represent it. In fact, the Nuer often make the
explicit statement that the alluvial soil, after all, comes from the highlands:

Even all the Gaat-Jak cannot finish this soil. Anywaa and Gaat-Jak together
cannot finish this soil. After all, this river [Baro] comes from the buny area.
Pine [alluvial soil] is from buny. When it rains in the highlands, the rivers
bring all the soil to us. It is red there, but when it reaches us, it becomes black.
This soil is important for us all. Pine is for all, Anywaa, Nuer, buny. It is food.
If we did not work on it, we would all be hungry. If we sat idle, like we do
now, we would all be hungry. You cannot stop a hungry man. (Kong)

The multiplicity of the land claims can be seen in Figure 8.1. The photo depicts a
new Nuer village in Itang district, on land that historically has belonged to the
Anywaa. The poles separating the Nuer village and the land in the foreground rep-
resent the boundary between the village and the land designated by the state as park-
land, one of the two undeveloped parklands of Gambella.

Figure 8.1 To whom does the land belong? (photo: Dereje Feyissa)
 The Nuer Response to Ethiopian Ethnic Federalism 183

Local legitimizing discourses are primarily relevant in animating actors to pursue

their interests. The EPRDF’s response to the Nuer integrative rhetoric, however, has
been ambivalent. On one hand, the EPRDF countered the growing political power
of the Anywaa with progressive promotion of the Nuer in regional politics. On the
other hand, the Nuer and their cross-border social and political networks are not
regarded positively. Nuer resentment of EPRDF’s political favouritism on behalf of
the Anywaa in the early 1990s and its alignment with the Sudanese government
against the SPLA resulted in violence in 1992. In fact, the massacre of Highlanders
in 1992 and the attempt by a Nuer prophet to engage militarily with the EPRDF are
often referred to as ‘evidence’ for the ‘unreliability’ of the Nuer as Ethiopian citizens.
The extreme violence in Itang in 1992 was related to the activities of Wutnyang
Gatkek, who was at that time one of the most powerful Nuer prophets in southern
Sudan. Wutnyang sought to legitimate his prophetic claim through the ‘revelations’
he had, as well as through establishing a spiritual link with Ngundeng Bong, the
greatest of all Nuer prophets. Wutnyang’s mission to and his military operation in
Gambella are still shrouded with obscurity. Some Nuer, Anywaa and Highlander eye-
witness accounts suggest that he combined a personal project (building up his
prophetic career) with political objectives (promoting an alternative form of south-
ern Sudanese nationalism). Though accompanied by SPLA soldiers, Wutnyang came
to Itang on a personal or religious mission – to collect a sacred axe at a place called
Rewmenyang, a village near Itang town. This sacred axe is called jiop naath.
Ngundeng is believed to have hid the jiop naath on the branch of a tree in Itang.
Wutnyang paid visits to various smaller prophets in Jikaw and Itang districts to per-
suade them that he was possessed by the Ngundeng spirit. Wutnyang also attracted
a big following, thanks to the miracles he was believed to have performed, such as
detecting thieves and dazzling the enemy.
Wutnyang’s religious mission was not successful, despite his popularity, as he
failed to collect Ngundeng’s jiop because of a military clash with Ethiopian troops. It
was, rather, Ngundeng’s political project that left a lasting imprint. By 1992
Wutnyang had raised his own personal army, known as the ‘white army’, consisting
of several thousand loyal Nuer recruits, who inflicted heavy damages on the Sudanese
army. Referring to Wutnyang’s attempts to promote solidarity among various south-
ern Sudanese peoples, Johnson (1994: 348) describes him as ‘a renowned peacemaker
who had repeatedly sought to defuse intra- and inter-ethnic conflicts that had been
developing during the first decade of the second civil war between the various Nuer
communities and between them and their Dinka, Anywaa and Uduk neighbours’.
Similarly, Hutchinson (1996: 339) writes that ‘as an unswerving advocate of politi-
cal independence for the South, Wutnyang had played an especially crucial role in
uniting and galvanising continued Nuer military resistance against the Khartoum
government’. Wutnyang delivered a divinely sanctioned call for greater peace and
cooperation among the southern Sudanese (including the Nuer and the Anywaa),
whom he urged to regain their pride as jinubni (southerners) and as black people (nei
ti caar) capable of progressing by their own right.
Wutnyang explained his coming to the Anywaa and Nuer in Itang, Gambella, in
racial terms, i.e., as part of the struggle against the ‘red’ people (the Arabs of the
184 Playing Different Games

Sudan and the Highlanders of Gambella). His first step in this struggle was to organ-
ize a peace ceremony between the Anywaa and the Nuer, who were then locked into
violent conflicts. Not only Nuer and Anywaa, but also Highlanders attended. The
Anywaa sat on the eastern side and the Nuer on the western side, reflecting the geo-
graphical location of the two peoples. The Highlanders sat beside the Nuer.
Wutnyang then said, ‘We shall bless ourselves so that there will be no more fighting.
We are all black people. We are one and brothers.’ Then two oxen, a black and white,
were put between the Anywaa on one side and the Nuer and the Highlanders on the
other. Wutnyang continued his speech, ‘When I spear the ox, and if it falls down
with its head eastward, I would say that the fighting between the Anywaa and the
Nuer is caused by the Anywaa. If it falls down with its head westwards, I would say
that the fighting is caused by the Nuer and the Highlanders.’ Wutnyang first speared
the white ox and then the black ox. Both the white and the black ox started falling
eastward but suddenly they turned around and fell down with their heads westwards.
Wutnyang announced that the Nuer had caused the fighting. In the afternoon of the
same day, a truck arrived from Gambella town carrying flour to Itang town. It was
stopped by Wutnyang’s followers, who took the goods and then went on rampage in
Itang town, where they looted the shops of the Highlanders. The small EPRDF con-
tingent in the town attempted to stop the looting, but Wutnyang‘s followers pre-
vailed and annihilated all fifty-eight of the EPRDF representatives, all of whom were
Highlanders. The next day, Wutnyang proceeded to Gambella town and then openly
declared his mission to be the liberation of Gambella from the Highlanders. Halfway
from Itang to Gambella town, EPRDF forces, well armed with sophisticated
weapons including a helicopter gunship, clashed with the forces of Wutnyang.
Finally, Wutnyang managed to flee back to southern Sudan safely, although his fol-
lowers sustained significant losses.
The Wutnyang incident had long-term repercussions for inter-ethnic relations in
the region. Subsequently, the Highlanders in Gambella town took reprisal measures,
indiscriminately killing ‘black’ people, some of who happened to be Anywaa. Above
all, the incident deeply scarred the resident Highlanders who felt threatened not only
by the Anywaa but also by the Nuer. It also set a precedent in the distribution of mil-
itary power in the region. Ever since then, the army has consisted exclusively of
Highlanders, whereas the regional police force is manned by Anywaa and Nuer.
Ostensibly, the exclusion of Anywaa and Nuer from the army serves to establish its
neutrality, but, when conflicts arise in the region, the army, like other state institu-
tions, is in fact identified with the Highlanders. This was amply demonstrated, for
instance, when the army sided with the Highlanders during the Anywaa massacre on
13 December 2003.
Given the strained relations between the Nuer and the EPRDF, and due to the
latter’s geo-political considerations, Nuer could not obtain a national audience for
their demands for political inclusion until something changed. The EPRDF began to
respond to the citizenship debate between the Anywaa and the Nuer only after a
national event that significantly altered the status quo of regional politics: the out-
break of the border conflict between Ethiopia and Eritrea in May 1998. In a throw-
back to ‘trench warfare’, the outcome of which was largely decided by sheer numerical
 The Nuer Response to Ethiopian Ethnic Federalism 185

strength (Negash and Tronvoll 2000), Ethiopia prevailed over the Eritrean forces in
the war to ‘restore’ its sovereignty. Towards that end, the Ethiopian government
needed ‘popular participation’ to offset the fortified positions of the Eritreans. In the
context of extensive mobilization, peripheral regions such as Gambella entered into
the national limelight. Both Anywaa and Nuer were recruited into the army. The
GPNRS contributed money from its annual budget, as well as ‘volunteers’ and cattle
during the war preparations.6 Twelve people joined the army from Makot village, three
of whom died on the battlefield.7 Consequently, the Cieng Reng politics of recogni-
tion were connected with the Ethio-Eritrean border conflict. The Nuer ‘enthusiasti-
cally’ joined the army, partly as an employment opportunity and partly because the
rhetoric of mobilization fed into GPDUP’s politics of inclusion. The GPDUP cadres
campaigned in the villages, including the Cieng Reng village of Makot, to persuade
them to join the army should they hope to gain political recognition.
Above all, the Ethio-Eritrean border conflict produced new local heroes. One of
these heroes was Bil Puk, a Gaat-Jak Nuer from Jikaw district. Bil was credited with
killing many Eritrean soldiers in one of the critical military engagements in the first
phase of the war and subsequently decorated for his contribution to the victory. As a
political reward, four Nuer were promoted, becoming palace guards.8 A victory poster
and music cassette called Wufer, featuring a photo of Bil, were produced by the
EPRDF for the ‘Badime Heroes’ (see Figure 8.2). The Badime victory poster was auc-
tioned in Gambella to contractors for 30,000 birr. The winner of the bid donated a
large framed copy of the poster to the Anywaa leadership of the ruling regional party.
Some of these posters reached the villages, and were later reproduced manually. The
Nuer were hailed for their heroic performance in the ‘restoration’ of sovereignty. In
turn, the Nuer instrumentalized a translocal political process as a new argument for
their politics of recognition – i.e., they ‘went national’ before they fully became ‘local’.
Such actions had immediate political repercussions in inter-ethnic relations.
Contributions to the lualawinet, ‘war for sovereignty’, in concert with the Nuer claim
to majoritarianism based on the 1994 census, became important discursive resources
for the Nuer to contest the Anywaa historical argument for political entitlement. The
Nuer power elites and ordinary people claimed: ‘We are Ethiopians because we, too,
have shed our blood for Ethiopia’, a reminder of the axiom that war makes nations, and
nations make war. Kong Diu used the same argument for citizenship when he appealed
to the federal government in Addis Ababa in November 2000: ‘Why should the

6. In fact, contribution in kind was one of the distinct features of the war, which made it appear to be
a ‘popular’ war. The Nuer, like other pastoralist communities, contributed cattle to feed the army.
7. The National Army, which was identified with the defunct Derg regime, was disbanded in
1991, and the new army was reduced in size from one hundred sixty thousand to a modest
ninety thousand troops. When the war broke out, there was a sense of urgency to increase the
army to two-hundred and fifty thousand (Negash and Tronvol 2000). That partly explains the
openness in recruitment procedures.
8. Bil is a relative of Kong Diu. One of the palace guards was also a Cieng Reng from Makot village.
Many Nuer in Gambella proudly recount such stories to legitimize their ‘Ethiopianness’. As far
as the Nuer are concerned, their selection as palace guards confirms their political recognition.
186 Playing Different Games

Figure 8.2 Bil Puk – a national hero (Source: Wufer cassette cover)

Anywaa say we are not Ethiopian? Did not we fight for Ethiopia as well? Even our cows
have become Ethiopians. They went to Badime. Okay, we go back to the Sudan, but
let them give us our people who died in Badime.’ The sympathy that Kong Diu
received from the federal government, discussed in Chapter 4, becomes intelligible
when viewed against this translocal political process. The Anywaa countered the Nuer
power claim on the basis of national heroism by recalling their own national heroes
who predated the Nuer. A case in point is James Uchan, an Anywaa who was promoted
by Emperor Haile Selassie for his heroic performance during the 1960s Ogaden war
with Somalia: ‘James fought against the Somali during the Ogaden war. Haile Selasssie
gave him a kokeb [medal] for his heroic performance. He was also appointed by the
emperor as the traffic officer of Addis Ababa. That was in the 1960s!’9
The Nuer claim for power on the basis of national heroism was warmly received
by EPRDF officials in Gambella. As they grew more confident due to a sense of legit-
imate entitlement, the Nuer demanded the creation of two more districts, which
became a reality following the 2005 election.

The whole of Ethiopia knows what the Nuer achieved at Badime. You are
great people. Not only for yourselves have you proved what you can do at the
national level. You should be proud of your son, Bil Puk. Now demonstrate

9. Interview with Philip Opiew, vice-administrator of Gambella in the 1970s, Gambella town,
October 2001.
 The Nuer Response to Ethiopian Ethnic Federalism 187

your greatness by making peace between yourselves. We are saddened by the

scale of destruction caused by the conflict between the Cieng Nyajani and
Cieng Wau. On behalf of my party, the EPRDF, I urge you to work for peace
and develop your areas. (Yohannes, EPRDF representative and political
adviser to Gambella Regional State, extract from speech given at a peace
meeting in Nyinyang village, 12 February 2000)

Public pronouncements by EPRDF officials, such as this one concerning the ‘Nuer
contributions’ to the war for sovereignty, emboldened the Nuer political actors, who
became more politically assertive in the years following the victory at Badime.
From the council to the schools and villages, Bil Puk became the new ethnic hero
and a code word for entitlement. In 2000, circumventing the district-based political
representation, the Nuer gained nine additional seats in the regional council through
the ‘grace’ of the federal government. It is against the backdrop of this national polit-
ical event that Kong had the courage and legitimacy to travel all the way to Addis
Ababa to appeal for citizenship. Although he was not given all that he asked for,
‘going national’ at least earned him a ‘residence permit’.
As could be expected, the success of the Nuer politics of inclusion increased
Anywaa anxiety:

In every election, Nuer representation has increased. At the beginning there

were no Nuer in the council. But, by 1995, all of the representatives from
Jikaw and Akobo were Nuer. Who will represent the Anywaa that are still
living in these districts? In the 2000 election, the Nuer even got one of the
five seats from Itang district. If they can claim representation of Itang, what
will stop them from claiming Gambella district? There are already a lot of
Nuer who are living there! (Opamo Obang, Nairobi, August 2002)

One of the new areas of contention in the political struggle between the Anywaa and
the Nuer was education. During the period 1996–2001, there were a series of incidents
and occasional violent clashes between Anywaa and Nuer students in both schools and
institutions of higher education. The first incident occurred in 1996, when a group of
Nuer students applied to the Teacher Training Institute. They were rejected by the
Anywaa officials in the regional bureau of education on the grounds that the applicants
were Sudanese refugees, not Ethiopian citizens. This was protested by the Nuer stu-
dents, not by denying their education in the refugee camps, which was stated in their
certificates, but by challenging the ‘double standard’ that the Anywaa officials
employed by accepting Sudanese Anywaa students and Ethiopian Anywaa who studied
in the Sudan. This particular incident revealed the distribution of power in the regional
government. Embittered by their rejection, the Nuer students established a student
union and appealed to the Nuer officials in the regional council, turning a sectional
interest into an ethnic issue. After a bitter political struggle between the GPDUP and
the GPLM, the Nuer students were admitted to the institute.
In 1999, schools once again came to the political forefront, this time on the question
of language choice, i.e., who should be taught in which language and where. According
188 Playing Different Games

to Article 39 of the 1995 constitution, ‘Every nation, nationality and people in Ethiopia
has the right to speak, to write and to develop its own language; to express, to develop
and to promote its culture; and to preserve its history.’ Multi-ethnic regional states such
as Gambella have adopted Amharic as the language of regional government, but they use
the vernacular in local government and schools. This was partly favoured by the federal
government in order to avoid the contentious issue of language choice. In implementing
the vernacularization of the educational system, however, a complication occurred in
Gambella town. The Anywaa were not only allocated more schools than the Nuer, but
one of the schools was Ras Gobena Junior Secondary School, the only junior secondary
school in the town. Upon finishing elementary school, all Nuer students from Gambella
town had to attend this junior secondary school, where the Anywaa language was both a
medium of instruction and a subject. In the first educational cycle (years 1 to 4), the
medium of instruction in the school was Anywaa, and Amharic was taught as a subject
from the third year on. In the second educational cycle (years 5 to 8), the medium of
instruction was English, and Anywaa was taught as a subject. This meant Nuer students
who joined the school had to learn Anywaa. In fact, Nuer students were obliged to declare
their willingness to learn the language. Some of the Nuer students either refused to sign
the declaration or started dropping out of lessons. They demanded to be taught in Nuer,
not Anywaa. The Anywaa defended the measure by asserting ownership rights over
Gambella town, which they considered to be an extension of the neighbouring Anywaa
village of Pinykew. Overall, this created a generalized sense of deprivation among the
Nuer, reminding them of who was in charge politically in the GPNRS.
A similar problem occurred in the same year in Itang town, where earlier the
Anywaa language had been introduced. When the Nuer asked for their own school
where they could develop their own language, the Anywaa officials gave them a room
outside the school compound in the former clinic for refugees, a symbolic action that
emphasized the underlying political assertion by Anywaa officials that the Nuer were
Sudanese. The Nuer believed this was an attempt to make the Anywaa language the
regional language, and hence to reinforce the political marginalization of the Nuer.
As the political crisis escalated, people were confined for weeks in their respective
ethnic neighbourhoods. ‘Ancient hatreds’ between the Anywaa and the Nuer were
invoked and traditional war songs were chanted to inject a sense of continuity with
past enmities. The crisis was abated only after a general public meeting was convened
under the auspices of the federal authorities, which looked into the root causes of the
problem. After a heated debate, it was decided to teach in both languages in Ras
Gobena Junior Secondary School.
The new language and educational policy has had various impacts on the multi-
ethnic regional states. In Gambella it has magnified the already-existing political
competition between the Anywaa and the Nuer. Which language is taught where has
raised the question of the political ownership of the GPNRS. The Anywaa’s attempt
to promote their language in the regional capital, Gambella town, which they con-
sider as their territory, was contested by the Nuer on constitutional grounds. Once
again the Anywaa and the Nuer drew on different sources of legitimizing claims and
diverged in their schemes of interpretation. For the Nuer, the struggle to participate
in the vernacularization of the educational system was part of their politics of recog-
 The Nuer Response to Ethiopian Ethnic Federalism 189

nition and quest for modernity. For the Anywaa, it was a manifestation of the Nuer
‘hidden agenda’.
Over time, the Nuer managed to renegotiate their subordinate status in the
changing political alliance between themselves and the Ethiopian state. This gave rise
to a renewed sense of betrayal and feelings of relative deprivation on the part of the
Anywaa, whose authenticity discourse (i.e., the Anywaa are more reliable citizens)
was increasingly being corroded by the state’s own pragmatism. Once again, as
Anywaa political power waned, Nuer political power waxed. This new situation,
which favoured the Nuer over the Anywaa, became more pronounced after the
violent conflict between these two groups in 2002, especially in the context of the
new power-sharing arrangement introduced subsequently by the EPRDF. The
advancement of the Nuer both in political power and in education, however, has
made them more competitive vis-à-vis the Highlanders, who hitherto had made up
more than fifty per cent of the region’s skilled labour. The increase in the number of
educated Nuer in the regional bureaucracy has also meant the emergence at the
regional level of a new ‘critical mass’ that is less amenable to EPRDF’s directives and
is capable of raising the issue of EPRDF’s accountability for its failures of governance
in the region. As might be expected, this has not been well received by the EPRDF
and has produced new tensions between the Nuer, on one hand, and the EPRDF and
the Highlanders, on the other.
To sum up, the discussions from Chapters 6 to 8 have examined the significance
of opposition in the transformation of ethnic categories into political communities
within recent and contemporary state systems. In Gambella, this transformation ini-
tially occurred among the Anywaa, because they were the first to experience opposi-
tion both from the Nuer and later from the Ethiopian state. At the same time, the
reactive, primordialist ethnicity of the Anywaa, and the monopolistic closure that it
entails, have evoked a new form of identification among the Nuer, compromising
their constructivist identity discourse. This can be inferred, for example, from Nuer
attempts to affix an ethnic ‘substance’ to the Anywaa, which, by implication, leads
them to essentialize the ‘natural’ differences that set them apart from each other.
The significance of opposition as a condition for ethnic consciousness can be
illustrated with reference to contemporary Anywaa and Nuer ethnicities. Anywaa
ethnicity features territoriality (expressed in a land-based political strategy) and the
politicization of cultural forms (demonizing the Nuer culture), whereas the Nuer
tend increasingly to reconstruct their identity in contrast to a ‘normative other’, artic-
ulating their moral contestation in response to the exclusionary political practices of
the Anywaa. As a result, the Nuer in contemporary Gambella grasp their own eth-
nicity in terms of their exclusion from political participation and power; and their
political practice gravitates towards inclusive strategies of ethnic entitlement.
Contemporary Anywaa ethnicity is significant for various categories of people
occupying different positions within a stratified political order. Employing ethnicity
as an ideology of power, the Anywaa elites have sought to exclude the Nuer elites
from the new structures of rewards in the state system by politicizing the settlement
history of Gambella. The historical frame of reference for the Anywaa’s expression of
(in their view) justified indignation is the Nuer migration into and conquest of
190 Playing Different Games

extensive Anywaa territories since the second half of the nineteenth century. This
selective memory reduces complex inter-ethnic relations to a narrative of victimiza-
tion and loss. As an ideology of power, then, ethnicity has been translated into a
political practice through which the Anywaa elites, aspirants to the modern sector
(‘elites in the making’), have sought to capture the regional state of Gambella. The
other, perhaps more inclusive, aspect of contemporary Anywaa ethnicity is what I
have called the project of containment. With the institutionalization of ethnicity in
the official Ethiopian policy of ethnic federalism, the Anywaa have sought to contain
what they perceive increasingly to be the ‘Nuer peril’, i.e., territorial encroachments
on the part of the Nuer and the corresponding expansion of Nuer cultural space.
Here macro-politics meet micro-politics: the chirawiya of the Anywaa villagers is sup-
plemented by the administrative power of the Anywaa officials. Still, there is growing
anxiety about the diminishing political position of the Anywaa in the regional eth-
nospace. By capturing the Gambella regional state, the Anywaa have tried to contain
the Nuer ‘tide’.
In light of such narratives of loss and projects of containment, Anywaa ethnicity
may be seen as a protonationalist project, an ethnic revivalist movement in changing
socio-economic and political contexts. The global triumph of neoliberalism and the
deconstruction of state agency in social transformation do not have local resonance
here. What we can observe in Ethiopia, and perhaps also in neighbouring states, is
an enchantment with the idea of the state and a belief in the ability of the state to
realize local projects of modernity. The Anywaa elites are, however, not only manip-
ulative maximizers but also visionaries engaged in articulating societal concerns. In
their self-understanding, they are attempting to ‘rescue’ their society from decline
and ultimate extinction through recourse to the agency of the state. But the actions
of Anywaa elites do not always bring about the desired result. In some ways, the
Anywaa project of containment has resulted in more ethnic insecurity, insofar as it
has provoked a reactive and more successful Nuer ethnicity and has strained relations
with the Ethiopian state. These two factors have combined to undermine the politi-
cal standing of the Anywaa in Gambella. Framing the local political contest in
national terms, the Anywaa have failed effectively to connect with and mobilize the
Ethiopian state on their own behalf. The Nuer, on the other hand, have achieved
greater gains by reframing their regional power claims in national terms.
The consolidation and militancy of Anywaa ethnicity has produced a reactive
Nuer ethnicity. Having an assimilationist social system par excellence, the Nuer were
not the first ‘to go ethnic’. Their initial response to Anywaa exclusionary political
practices was bewilderment. Subsequently, however, the inverted power relation, and
the Anywaa attempts to exclude them and achieve closure, led the Nuer to develop
strategies of usurpation that have drawn creatively on available means of gaining
entitlement. Nuer ethnicity contains within it two levels of concern. For the elites,
ethnic politics make it possible to share power and to get their share of the ‘federal
pie’; for ordinary men and women, ethnicity has become the medium through which
their quest for modernity can be realized. As the Anywaa were dominant in the dis-
tribution of political offices throughout the 1990s, and, thus, were able to wield
more administrative power, the Nuer exerted more pressure in new avenues of social
 The Nuer Response to Ethiopian Ethnic Federalism 191

mobility, particularly education, as a strategy of opening up political space.

Educational institutions have thus emerged as new sites for contestation and negoti-
ation of power relations, where the Nuer politics of inclusion collide with the
Anywaa project of containment. This has created an explosive situation that erupts
intermittently in violent conflict. The following chapter takes the power variable into
a different political context: the civil wars in the Sudan and the impact that they have
had on Anywaa–Nuer relations.
 Chapter 9

Civil War in the Sudan and

Ethnic Processes in the
Gambella Region
The transborder settlement pattern of both the Anywaa and the Nuer has made their
interrelations susceptible to wider geopolitical processes. The larger presence of the Nuer
in southern Sudan and their affiliations with various centres of political power has
enabled them to acquire greater military power than the Anywaa. In some cases at least,
this imbalance in military power has led the Nuer to abandon the practice of negotiat-
ing with the Anywaa over access to natural resources in the latter’s territories, and it has
encouraged groups of Nuer to resort increasingly to violence in inter-ethnic relations.
Postcolonial Sudan has been plagued by civil wars because of contentious issues
related to imbalances in regional development, to the narrow social base from which
the ruling elite is recruited, to an exclusive nation-building process, and to the greed
of the ruling elite in monopolizing the strategic resources of the country (Hutchinson
1996; Johnson 2003; J. Young 2007a). Most members of the ruling elite in postcolo-
nial Sudan, coming from the riverine areas of central Sudan, ‘favoured the interests of
those from the riverine core, and that in turn fostered dissent in the peripheries’ (J.
Young 2007b: 6). The various political regimes in the Sudan attempted to overcome
their narrow power base by imposing Arabism and Islam in a country that is marked
by high cultural and religious diversity (Deng 1995). Political power has been exer-
cised by these ruling elites within racial and religious frameworks. This has enabled
them to conceal structures of inequality and cultural differences in the western and
eastern parts of the country which, together with the central riverine lands, make up
the so-called ‘Islamic and Arabic North’ as opposed to the ‘Christian and African
South’. Until recently, the contradictions within the Sudanese state resulted in pro-
tracted armed resistance, coming predominantly from the southern part of the
country, where none of the rhetoric of the ruling elites has local resonance.
The various southern Sudanese armed struggles were waged from bases along the
Ethiopian-Sudanese border. They were supported by successive regimes in Ethiopia, in
response to Sudanese support for Eritrean liberation movements and for various dissi-
dent groups in Ethiopia (Johnson 2003). The doctrine of mutual intervention, that is,
‘the practice of governmental or other forces supporting opposition groups in neigh-
bouring states’, is a hallmark of conflicts in the Horn of Africa (J. Young 2007a: 5).
This is double-edged. On one hand, ‘local conflict in the Horn’ tends to pose ‘a threat
to inter-state relations and security’ (J. Young 2007a: 5); and, on the other hand,
‘neighbouring states use [local] disputes to pursue broader political objectives’ (J. Young
194 Playing Different Games

2007a: 2). The ethnic and racial nature of the Sudanese and the Ethiopian states pro-
duces domestic identity-based conflicts, which can also be exploited by adjacent states
seeking to gain advantages by meddling in the internal affairs of their neighbours.
These domestic conflicts become interwoven with interstate conflicts because of
the transnational political ideology of the Sudanese state. Successive regimes in the
Sudan have supported Islamic political groups in Ethiopia (Medhane 2007). This
was the case, for instance, with their active support of the Eritrean liberation move-
ments, as they considered Eritrea to be a part of the Arabic and Islamic world. In
recent times, the Sudanese government has also sought to export political Islam to
Ethiopia. The Ethiopian government has reacted by actively supporting armed strug-
gles in southern Sudan. The interplay between local disputes and interstate tensions
has created a dynamic configuration of power structures and system of alliances
between peoples of the borderland on the one hand, and the Ethiopian and Sudanese
states on the other (J. Young 2007b).
Both the Anywaa and the Nuer have cross-border settlements that are intimately
related to the ethnic process in Gambella. This can be seen in a full range of devel-
opments, including the following: the refugee phenomenon and the problem of dif-
ferential access to NGO-mediated resources; changes in the demographic structures
of Gambella; differential access to the transborder military power of the rebel groups
and the Sudanese government; the militarization of society and the corresponding
decline in inter-ethnic negotiated access to natural resources; and the significance of
the international border in framing the political debate in the Gambella region.
What is more, the Sudanese civil wars have encouraged the development of new pri-
mordialist tendencies among the Anywaa and the Nuer as well. In the following sec-
tions, I discuss the impact of the two Sudanese civil wars on the identification
strategies of the Anywaa and the Nuer in Gambella.

Anywaa–Nuer Conflict in the Context of the First Sudanese Civil War

(1955 to 1972)
The Anywaa–Nuer conflict in Gambella is intimately related to political processes in
southern Sudan. The first civil war produced a liberation movement called the
Anyanya (snake venom), which waged a guerrilla war against the Sudanese state and
its practices of religious discrimination and political exclusion (Nyaba 2001; Johnson
2003). The Anyanya was active in cross-border Anywaa and Nuer communities, for
which integration into the Ethiopian polity largely meant loss of political autonomy,
economic marginalization and social discrimination. Ethiopian Anywaa and Nuer
community leaders worked closely with the Anyanya leadership; in fact, during the
initial stage of the rebellion, the conflict was framed by the Anywaa and the Nuer in
Gambella in racial terms: the ‘black’ against the ‘red’. The objective of the rebellion
was said to be the creation of a new state – jenubi – that would include all ‘black’
people in the Sudan and Ethiopia.1 Transborder political networks were established,
with ‘black’ jenubi as the imagined political community, in contrast to the ‘red

1. The word jenubi means south in Arabic.

 Civil War in the Sudan and Ethnic Processes in the Gambella Region 195

Muslims’ (the northern Sudanese Arabs) and ‘red Christians’ (Ethiopian

Highlanders). This new identity discourse created political excitement among
Ethiopia’s borderland population. The imperial government of Ethiopia, identified
with the Highlanders and, thus, perceived to be on the ‘red’ side of the colour spec-
trum, initially defined the Anyanya rebels operating along the border as a national
security threat.2
The first instance of political activity by the Anyanya in Ethiopia was led by a
group of southern Sudanese students who had gone to Ethiopia ostensibly to pursue
their education in institutions of higher education. Under the auspices of Nuer and
Anywaa local leaders, the students established military bases in Gambella. These
camps were located at various places in Jikaw, Itang, Jor/Gog and Akobo districts.
The local Anywaa and Nuer imperial officials also helped the students to establish a
local militia (bura). When the Ethiopian government took notice of this clandestine
transborder political network, the students left Gambella for other African countries
for their own safety. A section of them went back to southern Sudan. In 1962, the
Ethiopian and Sudanese governments signed a treaty of mutual extradition of ‘crim-
inals’ – the former had labelled the Anyanya as shifta – posing a direct threat to the
political activities of the southern Sudanese students in Ethiopia.
These political stirrings along the border became the first wave of opposition in
the 1960s to Ethiopian imperial rule, based on an ethnonationalist and class plat-
form (Addis 1975; Gebru 1991). The Eritrean Liberation Front was established in
1960, and the postcolonial Somali state advanced an irredentist project claiming the
Somali-inhabited region of the Ogaden. Peasant rebellions were also rife in Tigray,
Gojam and Bale provinces. Against the background of growing opposition to impe-
rial rule, in 1963 the government established the ‘Gambella Security Study Group’,
which recommended a comprehensive approach to thwart the imagined cross-border
political community. The following archival material throws light on the securitiza-
tion of state-society relations along Ethiopia’s western border:

All over the world, we find trouble spots where people of the border area are
mobilized on the basis of cross-border ethnic connections, linguistic affinity
and culture, and challenge the lawful governments. We have seen similar
troubles in the Ogaden. Before a second Ogaden is created in western
Ethiopia, we should nip the political developments in the border region of
Gambella in the bud. This should include various measures. For one, the
armed force in the region needs to be strengthened. The armed personnel
should be raised from the current 404 to 797. We should provide social serv-
ices such as health facilities, schools, roads, motor boats, and occasional
flights to Tiergol. In order to maintain the loyalty of the Anywaa and the
Nuer balabats, they need to be salaried, like their fellow officials in the Sudan;
occasional visits should be organized to bring the balabats to Addis Ababa and

2. Letter written to the Ministry of the Interior by the Governor of Gore, 15 July 1967, obtained
from the Gore Archives.
196 Playing Different Games

show them big factories in order to impress upon them the grandeurs of
imperial power. Also, to make them similar to the Ethiopian people, evangel-
ists and teachers from the Orthodox Church should be sent, and the activi-
ties of the missionaries [the American Presbyterian Church] who sympathize
with the shifta should be regulated. (Author’s translation from Amharic, The
Gambella Security Study Group Report, 1963: 8–13)

This state of affairs and the pattern of political alliance changed after it became clear
to the Ethiopian government that the Sudanese government was giving political and
military support to the Eritrean Liberation Front (ELF). The ELF launched an
armed struggle in 1961 ‘against the regime of emperor Haile Selassie from bases in
western Eritrea and eastern Sudan after the emperor eroded international agreements
protecting Eritrea’s autonomy’ (J. Young 2007b: 3). The ELF, largely recruited from
the Muslim populations of Eritrea, framed the liberation movement in religious
terms and claimed Arab identity. This earned the ELF political and military support
from the wider Arab world, particularly the Sudan. The Ethiopian government
responded by befriending the same Anyanya that it had earlier considered a security
threat. This mutual interference ‘would bedevil relations between the two countries
for the next three decades’ (J. Young 2007b: 3). General Lemma, the imperial gov-
ernor of Gambella in the 1960s, coordinated the logistic support to the Anyanya.
The ELF enlisted the support of the Arab world via the Sudan, and the Ethiopian
state turned to Israel for support, which it readily gave. Israel also gave support to the
Anyanya (Johnson 2003). In that sense, Gambella became one of the regions where
the Arab–Israeli conflict played out through proxy wars.
With this new sponsor, the Anyanya delocalized its social base and became increas-
ingly predatory towards the very people who had given it unconditional support. In
Gambella, relations between Anyanya, on one hand, and both the Anywaa and the
Nuer, on the other, began to deteriorate after 1965, when the rebels started demand-
ing corvée labour and imposing taxes. Control over the flourishing gold mining in the
Dambala region (lower Akobo) was also a bone of contention, particularly with the
Anywaa. When Ethiopia strengthened its support for the Anyanya and Israel became
involved in the conflict, the government of Sudan was keen to repair its strained rela-
tions with Ethiopia. For both the Sudan and Ethiopia, interference in the other’s inter-
nal affairs subsided, and in 1972 Ethiopia brokered a peace agreement between the
government of the Sudan and the Anyanya rebels, popularly known as the Addis Ababa
Peace Agreement, that ended the first Sudanese civil war (Johnson 2009).
This first civil war produced more than twenty thousand refugees who were
hosted in various refugee camps in Gambella (Klintenberg 1977: 158). Some of these
refugees, particularly the Sudanese Nuer, settled among their kin as Ethiopian citi-
zens. The great influx of Nuer refugees into Anywaa territory had ‘shaken the whole
micro political equilibrium along the Baro to such an extent that inter-ethnic and
inter-group fighting has become a serious obstacle to settlement and integration’
(ibid.: 158). The semi-official boundary between the Anywaa and the Nuer in the
Gambella region crossed the Baro River somewhere west of Odier, but the real border
after the refugee influx became more vague. It was also during this time that
 Civil War in the Sudan and Ethnic Processes in the Gambella Region 197

Sudanese Nuer refugee students established a settlement in Gambella town, near to

the UNHCR compound, which they called Newland. The area where the settlement
was established was previously an Anywaa settlement known as Chenquar. Currently,
Newland is a burgeoning Nuer settlement of more than five thousand inhabitants.
Nearly all the Nuer residents of Gambella town live in Newland.
The change of regime in Ethiopia in 1974 renewed the mutual interference, with
different stakes. The government of the Sudan reasserted its territorial claims on areas
of Ethiopia, including Gambella. The idea of incorporating Gambella into southern
Sudan was first entertained by the British colonial establishment in the Sudan as a
means of securing the waters of the Nile – the major tributaries of the Sobat River,
itself a tributary of the White Nile, pass through Gambella. According to the 1902
Anglo-Ethiopian Boundary Agreement, the legal status of the Gambella trading post
(the enclave) was conditional as long as the Sudan was under the Anglo-Egyptian
government. As per the agreement, the enclave was returned to Ethiopia when the
Sudan became independent in 1956.3 The Sudanese government retained a consul in
Gambella for diplomatic and commercial purposes and obtained some concessions
relating to the Baro River. According to the Gambella protocol of 1956, sovereignty
over the enclave was officially transferred to the Ethiopian state, but the imperial
government made available numerous houses for the service of the Sudanese consul
in Gambella, including a twenty-year rent-free concession.4
Sudanese interest in Gambella continued unabated, however. This interest was
translated into a political project when the government of the Sudan helped to estab-
lish a largely Anywaa-based political organization called the Gambella Liberation
Front (GLF) in 1976.5 The GLF was politically active in the late 1970s and was
linked to the Anywaa rebellion in Jor district against the Derg’s ‘cultural revolution’.6
The GLF was actively engaged in recruiting the youth on both sides of the interna-
tional boundary from its military base at Galabal. The two principal Ethiopian
Anywaa actors in this project were Uguta and David. David was later killed by the
Sudanese police in the Sudan, and Uguta was imprisoned by them for failing to
maintain order among his followers. Upon his release, Uguta surrendered to the
Ethiopian authorities and later became instrumental in forming the GPLM.7

3. Letter written by Ethiopian Ministry of Foreign Affairs to the Embassy of the Democratic
Republic of the Sudan, 2 June 1971, No. 13271/69/21, Metu Archive.
4. According to the Gambella protocol, the Sudanese government was allowed to continue meas-
uring the height of the Baro River until such time as Ethiopian hydraulic experts could assume
the task.
5. Anywaa leaders of the GLF included people who later emerged as prominent Anywaa politi-
cians in the southern Sudanese politics of liberation: Philip Udiel (governor of the Upper Nile
region); Paul Anade (MP in the Southern Sudan Regional Council); Simon Mori (minister in
the Southern Sudan government); Agud Obong (general in the Sudanese army) and Philip
Akiyu (administrator of Pochalla County).
6. This was mediated through the late nyiya Ageda, who was active in giving moral support and
protection for the kwaari who took up arms against the Derg in the second half of the 1970s.
7. The late Okello Oman, ex-GPLM leader and the first President of the GPNRS, interviewed
on 13 September 2001.
198 Playing Different Games

By the mid-1970s Sudan was sliding into its second civil war. In March 1975, a
new rebel force known as Anyanya-II was established, largely with Nuer leadership
(Nyaba 2001). Anyanya-II was active in the Akobo region. Initially, Anyanya-II,
which established one of its military bases at Bilpam in Itang district, received
support from the Ethiopian government (Riang 2005: 6); but when, by 1983,
attempts to unify the various Anyanya II factions had failed (Johnson 2003), the
Derg promoted the founding of a new liberation movement known as the Sudan
Peoples Liberation Army (SPLA). The second Sudanese civil war officially began in
1983 and lasted until 2005.

The Anywaa–Nuer Conflict in the Context of the Second Sudanese

Civil War (1983-2005)
Given the link between the SPLA and the Ethiopian state, Anyanya-II turned to the
government of the Sudan for military support. In response, the Ethiopian govern-
ment gave the military camp in Bilpham, which had previously belonged to Anyanya
II, to the SPLA (Riang 2005). Fearing, however, that the secessionist posture of
Anyanya II would encourage a similar separatist movement in Gambella, the Derg
put pressure on the SPLA to drop secessionism and adopt a unionist position
(Hutchinson 2001; Johnson 2003).
The Nuer found themselves divided between Anyanya-II and the SPLA, but an
increasing number of them threw in their lot with Anyanya-II. Although the split
was based originally on ideological differences (secessionist/unionist) and on internal
power struggles, it was also framed in the language of the ‘resurgence of the ancient
hatred’ between the Nuer and the Dinka (Hutchinson 2000). The rivalry between
Anyanya-II (with Nuer leadership) and the SPLA (with Dinka leadership) played
right into the hands of the Sudanese government, which attributed the southern
Sudanese problem to ‘tribalism’. Heavily armed by the Sudan, the Anyanya-II
attacked SPLA recruits on their way to the training camps in Ethiopia. In May 1984,
more than three thousand recruits from northern Bahr el Ghazal were ambushed and
massacred in the area of Fagak, and in 1986 over two thousand were massacred in
the Lou area (Nyaba 2001: 49).
In addition to Bilpham, the SPLA established bases in various places in Gambella.
Four camps for southern Sudanese refugees – most of whom were Nuer and Dinka –
were established in different parts of Gambella under the direct political and military
control of the SPLA (J. Young 2007a: 21). All of these military and refugee camps
were established in Anywaa territories. By 1990, the official number of southern
Sudanese refugees in Gambella reached 355,000, the majority of whom settled in
Itang and Pinyudo refugee camps (Kurimoto 1997: 799). By the mid-1980s Itang was
one of the largest refugee camps in the world (Kurimoto 2005: 344). Although the
official number of refugees was inflated, it far outnumbered the resident Anywaa and
Nuer population, estimated at fifty-five thousand according to the 1984 census.
The refugee community and the lucrative aid establishments attached to it created
further networks of interest between the Derg and the SPLA. As Kurimoto (2005: 345)
notes, ‘the refugee camps were administered by committees comprised of “representa-
tives” of the refugees, who were invariably SPLA officials … Moreover, the entire
 Civil War in the Sudan and Ethnic Processes in the Gambella Region 199

Gambella region was under a sort of informal joint administration of the SPLA and the
Derg’. Some of the resources allocated to the refugees were used by the Derg to fight
the war against armed opposition groups in the north. More such networks were
created between refugee camps and state farms. Most of the aid (grain) for the refugees
was bought from the state farms in Gambella and the neighbouring highlands.
The 1980s saw a complex pattern of alliances among the various political actors oper-
ating along the Ethio-Sudanese border. The socialist orientation of the Derg regime and
its strong Soviet military backing pushed the government of the Sudan into the arms of
the Western bloc. The United States was additionally actively engaged in oil exploration
in the Sudan (Johnson 2003: 68). This alignment of forces made the Horn of Africa one
of the hotspots of the cold war. As discussed earlier, the Sudanese government and the
Derg were at odds on the Eritrean question, the former supporting the Eritrean Peoples
Liberation Front (EPLF), a splinter group from the ELF. The Sudanese government also
supported the Tigrean Peoples Liberation Front (TPLF), which in 1989 was transformed
into the Ethiopian Peoples Revolutionary Democratic Front (EPRDF). The Derg retali-
ated by helping to organize the SPLA. The Gambella Liberation Front (GLF) fractured
into various political groupings, with one section, under the auspices of the Sudanese gov-
ernment, joining the Oromo Liberation Front (OLF), a political organization which
fought, and still fights, for the self-determination of the Oromo in Ethiopia. This section
later became the nucleus for the Gambella Peoples Liberation Movement (GPLM),
which was a response to the repressive policies of the Derg and continued to benefit from
the Sudanese connections that increased its military power and political standing.
As an autonomous political movement, however, the GPLM had to struggle to
create its own political space vis-à-vis the political projects of the Sudanese govern-
ment, the OLF and the EPRDF. The intensification of the proxy war between the
Derg regime in Ethiopia and the government of the Sudan increased the political rel-
evance of the GPLM in the geopolitics of the day. In fact, the GPLM was part of the
coalition of armed opposition forces (EPLF, EPRDF and OLF) that launched a coor-
dinated military offensive against the Derg from their base in the Sudan in the late
1980s. As part of the proxy war with the government of the Sudan, the Derg heavily
armed the SPLA, which emerged as a powerful political actor in Gambella through-
out the 1980s. However, with a strong patronage in Addis Ababa, the SPLA failed to
cultivate an amicable relationship with the Anywaa and the Nuer communities in
Gambella. From 1984 to 1987, the SPLA clashed with the Gaat-Jak Nuer, and in
1989 there was a major confrontation between the SPLA and the Anywaa.
The first serious clash occurred in Jikaw district in 1984, beginning with a minor
incident between SPLA soldiers and the Thiang Nuer in Dorong village of Itang dis-
trict.8 Tensions between them had already reached crisis level, with only a spark

8. This incident was triggered by a dispute between SPLA soldiers and Gaat-Jak fishermen over
a fish. The soldiers wanted to confiscate the fish, then fired on the fisherman as they fled.
Thereafter, there was an exchange of fire in which two fishermen and two soldiers were killed.
The next day, the SPLA attacked the village of Palkoini. As Riang (2005: 6) put it, ‘This inci-
dent became the straw that broke the camel’s back and sparked off all-out war between the
SPLA and the Gaat Jak.’
200 Playing Different Games

needed to trigger violence. The issues of contention included the Gaat-Jak Nuer’s
alleged association with the Anyanya-II, onerous taxation, cattle-looting by local
commanders, conscription by force and interference in local leadership (Riang
2005).9 The conflict that erupted gradually spilled over to all Gaat-Jak sections,
lasting for three years and costing thousands of lives on both sides. It is now remem-
bered as the ‘Gaat-Jak/SPLA War’, and contemporary Gaat-Jak describe it as the
SPLA’s ‘ethnic cleansing’ of the Gaat-Jak.10
The ‘Gaat-Jak/SPLA War’ clearly revealed the regional power configuration.11
The SPLA enjoyed a higher political profile than the Nuer-led Gambella regional
administration, who were too weak to stop the conflict because of their military serv-
ices to and political linkage with the Derg. The SPLA was also a formidable military
force in its own right because of its accumulation of high-tech weaponry. Apart from
costing the lives of many people, the conflict had the effect of the massive militariza-
tion of the Gaat-Jak Nuer, who, in the context of a state that failed to give them the
protection they needed, desperately took the law into their own hands.
SPLA–Anywaa relations did not fare any better. The 1989 clash between the
Anywaa and the SPLA is a case in point. Kurimoto (1994: 24–25) has documented
events leading to the confrontation between the Anywaa and the SPLA as follows:

On September 11, Ethiopians at Pinyudo were celebrating the Ethiopian

New Year holiday. But the market place in the refugee camp was open as
usual. An Anywaa secretary of the WPE (Worker’s Party of Ethiopia) went
there with Anywaa militia and ordered it closed. Refugees refused the order.
A quarrel developed into a fight. Militia shot in the air to stop the fight. Then
a Dinka SPLA officer in charge of the refugee camp opened the store of guns
and ammunition and distributed them to refugees. They went and attacked
Anywaa militia at the market and surrounded the village. There was an
exchange of fire between Anywaa militia and armed refugees. As the fighting
escalated, the SPLA soldiers from a nearby camp joined the fighting.
Outnumbered, the Anywaa militia was defeated and the refugees burnt the
village. Dead bodies were buried in mass graves and the casualties were esti-
mated at more than 120 Anywaa.

In the eyes of the Anywaa, the Pinyudo massacre produced further evidence of a
wider conspiracy that threatened their existence. Reflecting on the pattern of politi-

9. The SPLA charged a per unit tax of 5 birr per day for a bottle of local beer and 20 birr per day
for a cow. The sum total of the tax per family was around 9,000 birr per year.
10. SPLA sources contest the Gaat-Jak’s description of the conflict as ‘ethnic cleansing’. They
mention Kuach Kang, the chief of the Thiang Nuer Mangok, as one of the main actors in the
conflict. Kuach initially worked with the SPLA, but later on defected to Anyanya-II and the
government of the Sudan (Riang 2005).
11. Thowat Pal, the WPE Secretary of Gambella, contacted Nuer SPLA commanders to stop the
SPLA atrocities led by Commander Kerubino. Thowat also appealed to General Mesfin, com-
mander of the Ethiopian army in Gambella, but in vain (Thowat Pal, interviewed in Frankfurt,
23 November 2005).
 Civil War in the Sudan and Ethnic Processes in the Gambella Region 201

cal and military alliances among the Derg, the SPLA, the refugees and the Nuer lead-
ership of the Gambella administration, the Pinyudo massacre and the non-account-
ability of the SPLA regarding the atrocity appeared to validate the Anywaa theory of
conspiracy. Conflicts between refugees and the local Anywaa have been perceived
through an ethnic prism ever since. For one, many of the refugees were ethnic Nuer.
But, more importantly, unlike the first Anyanya movement, the Anywaa were very
marginal in the SPLA leadership.
The prevailing insecurity and the failure of the Ethiopian state to protect its own
citizens induced a political practice that I call ‘refugization’, which occurs when cit-
izens of one country attempt to pass as refugees from another country. In this case
study, refugization refers specifically to Ethiopian citizens who claim to be Sudanese
refugees in Ethiopia, if this appears to be to their advantage. However, refugization,
understood as a coping strategy, was not an option for all groups of people living in
Gambella. It was the Nuer who were in the best position to make use of it. For one,
the Nuer were perceived by the NGOs as Sudanese on the basis of their settlement
pattern. The Nuer population in the Sudan is much larger than in Ethiopia. The
Nuer were also aided by the SPLA’s political mobilization rhetoric. There was a sus-
tained ‘Sudanization’ campaign by SPLA officials among the Ethiopian Nuer in
order to enlarge their political constituency and military capacity. They propagated
the idea of cie buny michar. In the Nuer language this means ‘there are no black
Ethiopians’, a reference to the Sudanese origin of the Nuer. The SPLA used this
saying to legitimate conscription of the Ethiopian Nuer into the SPLA. There
seemed to be a tacit understanding between the Derg and the SPLA on the division
of subjects, reserving the Nuer for SPLA and the Anywaa for Derg conscriptions.
The Ethiopian Nuer influx into the Sudanese refugee camps in Gambella was thus
not as much an issue for the Derg as it was for the Anywaa. There was concern on
the side of the Ethiopian government, however, that an unchecked refugization
would ultimately undermine the newly constituted peasant associations, which
served, among other administrative controls, as agents of conscription for national
military service in the regime’s war against the proliferated ‘liberation fronts’.12
The refugization chosen by individual Nuer as a strategy for coping with insecu-
rity and marginality had, however, an impact on inter-ethnic relations. Access to
quality education in the refugee camps enabled the Nuer to renegotiate their margin-
ality in the modern sector and ‘catch up’ with their Anywaa counterparts. Dozens of
NGOs, under the auspices of the UNHCR (United Nations High Commissioner for
Refugees), operated in the refugee camps, providing social services, particularly
health and educational facilities. As a result, many Ethiopian Nuer flocked to the
camps to have access to these social services. The deteriorating security condition in
the border district of Jikaw in the second half of the 1980s was also a compelling
reason to switch national identity, with the refugee camps appearing safer than the
villages. During the Gaat-Jak/SPLA war, all the schools except one were shut down
in Nuer areas. In the refugee camps, on the other hand, it was possible to get an edu-

12. Tamirat Woldeyes, WPE cadre in Gambella, 12 March 2001.

202 Playing Different Games

cation. The educational support package included a scholarship (food, shelter and
allowance) all the way to college level. The UNHCR made an arrangement with
church-based colleges to that effect.13
The Nuer were also attracted to the refugee camps by the UNHCR resettlement
programme. Resettlement in a third country is one of the UNHCR’s durable solutions
to ensure the security of the refugees in situations ‘when individual refugees are at risk,
or when there are other reasons to help them leave the region’ (Patrick 2004: 1).
Making use of this opportunity structure within the aid agencies, a significant number
of Nuer have resettled in North America and Australia. The Nuer diaspora is estimated
at ten thousand (Shandy 2001; Falge 2006), many of whom are from Gambella. The
resettlement programme has legitimated the Ethiopian Nuer international migration,
which would have otherwise been considered ‘illegal’ had they migrated on their own.
The resettlement programme was appreciated not only for making the migration pos-
sible, but because of the benefit package that eased the process of adaptation during
resettlement. For instance, unlike other categories of refugees in the US, resettled
refugees received institutional support by the US government upon their arrival
(Shandy 2001). Pushed by dire poverty and political turmoil, contemporary Africans
saw international migration to the West as an exit option, but tight immigration poli-
cies and the high cost of self-sponsored migration made it impossible for the majority
to migrate to the West. Neither the Anywaa nor the Highlanders managed to make use
of the international border as much as the Nuer did.
Through differential access to NGO resources in the refugee camps, the Nuer
caught up with the Anywaa in the field of education and in the regional administra-
tion’s civil service sector. Not surprisingly, this was resented by the Anywaa, because
the educational advancement of the Nuer threatened to undermine their status claim
and to become a factor in the competition for political power, as the following testi-
monies from Anywaa interview partners indicate:

It was difficult for us to join the refugee camps, whereas many Nuer from
Jikaw and Akobo were allowed. In 1985, I tried to join the Itang refugee
camp but I was screened out as an Ethiopian citizen by the UNHCR. After
that I went to Gambella town to learn but I did not have any relative. Instead
I worked for a Highlander in order to be able to pay for my education. It was
my Highlander host who sponsored my education. I completed high school
in Gambella town but did not get enough marks to join college. But for the
Nuer it is different. Be hulet bila yibelalu [an Amharic saying which means
‘those who eat with two knives’ in reference to a double chance or oppor-
tunism]. They learn as Sudanese refugees and get jobs as Ethiopian citizens.
(Ujulu Akwei, Gambella town, 18 April 2001)

The Nuer are catching up with us [in Amharic eyederesubin naw] because of
the education they got from the refugee camps. Previously only a few Nuer

13. These were Kuyira Adventist College in Southern Ethiopia and the Dongoro Boarding School
in Wellega.
 Civil War in the Sudan and Ethnic Processes in the Gambella Region 203

were educated. Then they joined the camps and some came from Nassir, and
suddenly they became many in the schools. In the refugee camps, it does not
take you long to finish high school. One can be promoted from one grade to
the next if he is a good student. There are no such things in the schools in
Gambella. The quality of education in the camps is also very good. We could
not learn in the camps because ARA [Administration for Refugee Affairs] and
the UNHCR knew that we were Ethiopians. Some Anywaa had tried to
sneak in but they were screened out. (Ujulu Obong, Anywaa student in
Gambella High School, 7 April 2001)

The Nuer responded to the Anywaa accusation of ‘eating with two knives’ in prag-
matic terms: ‘the Anywaa do not know what the border offers’. The Nuer rather sar-
castically advised the Anywaa to eat with two knives as well. The Nuer also defended
their political opportunism (switching citizenship) as a reaction to marginalization:

It is not fair that we are accused of learning as Sudanese. For one, we did so
because our areas were marginalized. There was no other alternative. The first
school in the Nuer areas in the Sudan was established by the missionaries in
1922. When the missionaries were driven out by the government of the
Sudan, they came to Adura where they opened a school in 1962. In 1977, the
Ethiopian government drove the missionaries out of Adura. In the absence of
services and facilities in our area, it is no wonder that we looked towards the
Sudan. It is the same with the refugee story. What we did was very normal. All
of a sudden, services were established near us for the refugees coming from the
Sudan. As we did not have anything, we joined them. This is also good for
Gambella. We were educated as Sudanese but work as Ethiopians. Most of the
Nuer officials in Gambella were educated by the church or in refugee camps.
Who would have assumed the administrative posts in Nuer areas, had it not
been for our education in the refugee camps? (James Gadet, Nuer church offi-
cial, Western Gambella Bethel Synod, Gambella town, August 2000)

Fearing the military power and assimilationist drive of their pastoralist neighbours,
the agrarian Anywaa invoked the state border to ensure ethnic security and attain a
dominant political status in the border region of Gambella. The Anywaa, like the
Nuer, live in both Ethiopia and the Sudan, but the majority of them live in Ethiopia.
On that basis, in the 1990s, the Anywaa in Gambella viewed the rewards of this
opportunity to be higher than the cost of separation by the international border.
Through a reference to the 1902 Ethio-Sudanese border, the Anywaa sought to label
the Nuer, except for few Gaat-Jak clans, as ‘foreigners’, because the demarcation of
the border placed nearly all of the Nuer in the Sudan.
There is more in the Anywaa invocation of the border than mere strategic think-
ing. It is embedded in their traditional model of political order, which features terri-
toriality. The national state is perceived and experienced by the Anywaa through their
compartmentalized idea of a political community. Apparently, the Anywaa idea of a
fixed border connects with the nation state discourse, but the Ethiopian state has mul-
204 Playing Different Games

tiple concerns about the border beyond merely enacting sovereignty. The gap between
this discursive connection and political practice is one of the root causes of the trouble
between the Anywaa and successive Ethiopian governments. The Anywaa ‘rebuke’ the
Ethiopian state for failing to observe the foundational premise for modern states, i.e.,
that political sovereignty is identified with bounded territories. Although they were
politically dominant in the GPNRS throughout the 1990s, the Anywaa had not
managed to establish the border as much as they would have liked. Borders are federal
issues. Lack of interest by the federal government to station the army to police the
border was and still is interpreted by the Anywaa as evidence for the ethnic and racial
characters of the Ethiopian state, in which they are ‘second-class citizens’.
The Nuer, on the other hand, project their inclusive idea of a political community
and flexible notion of localization onto the national state, and by extension on the
concept of an international border. In the 1990s, however, the Nuer acceptance of
alternating citizenship became a liability: Nuer refugee students were vulnerable to the
Anywaa citizenship-based politics of exclusion in post-1991 Gambella, when the trend
was to claim Ethiopian citizenship in order to make use of the trickle-down effect of
ethnic federalism and the affirmative action programmes that were attached to it.
The ethnopolitical developments of post-1991 Gambella were paralleled by the
ethnicization of the southern Sudanese liberation movement. In 1991, the SPLA was
split into two factions, the largely Dinka-based SPLA led by John Garang, and the
largely Nuer-based Nasser faction, which later came to be known as the Southern
Sudanese Independence Movement (SSIM), led by Riek Machar. The split was trig-
gered by the regime change in Ethiopia. In May 1991, the Derg were overthrown by
the EPRDF, and the SPLA lost its rear bases in Gambella. As an SPLA veteran
described it, ‘this sent shock waves into the spinal chord and nerve centres of the
SPLA’ (Nyaba 2001: 55). The split was also on the basis of both ideology and the
power struggle. The SPLA’s vision of a new Sudan, ‘that encompasses both North and
South and assumes their coexistence in a restructured state’ (J. Young 2003: 424), did
not sit well with some of the commanders who still espoused a secessionist agenda and
whose roots went back to the Anyanya movement. The split was also related to the
narrow social base of the SPLA leadership. As J. Young (2003: 425) notes, ‘there is
little doubt that the SPLA is, as its critics claim, “Dinka dominated”, or that Bor
Dinka hold a disproportionately large number of posts in its leadership’.14 This is an
interesting political irony, an instance of a counter-hegemonic project harbouring
hegemonic aspirations. After all, it was the domination of political power by the Arab
riverine core in the Sudanese state system that generated the liberation movements.
In the struggle for power within the southern Sudanese liberation movements that
ensued after 1991, both the SPLA and SSIM mobilized their respective ethnic con-
stituencies. According to Hutchinson (2000: 6), ‘both Garang [SPLA] and Riek
Machar [SSIM] eventually reached for the “ethnic card”… mostly targeting the civil-
ian population along ethnic lines’. The extensive military campaigns by the SPLA and
SSIM in the Nuer and Dinka areas, respectively, resulted in a tremendous loss of life as

14. According to J. Young (2003: 425), of the thirteen members of the SPLA leadership council
in 2003, seven were Dinka.
 Civil War in the Sudan and Ethnic Processes in the Gambella Region 205

well as the destruction of property on a much larger scale than the destruction inflicted
on the south by the government of the Sudan (Jok and Hutchinson 1999: 126). This
was followed by ‘the rapid polarization and militarization of Nuer/Dinka ethnic iden-
tities’, as a result of which ‘growing numbers of Nuer men and women began to reject
a “performative” concept of ethnicity in favour of a more “primordialist” concept
rooted in procreative metaphors of shared blood’ (Hutchinson 2000: 8). Thus, the new
political process in southern Sudan and the conflict related to it are among the contexts
within which ethnic identity among the Nuer has become more primordialist.
Above all, the war led to the shifting ethics of Nuer/Dinka warfare. According to
Hutchinson, ‘the purposeful slaying of a child, woman or elderly person was univer-
sally perceived not only as cowardly and reprehensible but, more importantly, as a
direct affront against God as the ultimate guardian of human morality’ (Hutchinson
2000: 8). Hutchinson then went on to identify the factors that undermined the
social restraints of intra- and inter-ethnic warfare. The first was the ideological cam-
paigns of the SPLA geared towards the depersonalization of acts of homicide. The
shift from spear to gun also undermined social accountability because ‘whereas the
power of a spear issues, they reasoned, directly from the bones and sinews of a person
who hurls it, that of a gun is eerily internal to it’ (Hutchinson 2000: 10). More
importantly, the military confrontations between the SPLA and SSIM turned
women and children from, in Hutchinson’s words, mobile assets to legitimate targets:
‘The purposeful killing of women and children necessitated a major reformulation
of the presumed socio-physical roots of ethnic affiliations, particularly for Nuer com-
batants. The rationale of killing a Dinka child entailed an assumption, whether
implicit or explicit, that the child would mature into a “Dinka”. That child’s ethnic
identity, in other words, was presumed to be fixed to birth. The idea that such a
Dinka child could potentially become a “Nuer” or vice versa was thus lost in the fury
of “revenge attacks”’ (Hutchinson 2000: 11).
The reconfiguration of the Nuer identity concept in reference to their Dinka neigh-
bours in southern Sudan was relevant for changing inter-ethnic relations in Gambella.
After all, many of the Ethiopian Nuer were politically mobilized by the SPLA and
SSIM and had participated in their wars. The impact was also evident to the extent that
the gun entered the sphere of bridewealth. The power of the gun and the new culture
of violence associated with it greatly undermined inter-ethnic negotiations over access
to natural resources. In the 1980s and 1990s, some groups of Nuer seized extensive ter-
ritories from the Anywaa in the Akobo area and along the Baro River.
The reconfiguration of Nuer ethnicity also occurred in the context of new power
relations with their neighbours. After their extensive territorial losses to the Nuer at
the end of the nineteenth century, the Dinka in the west and the Anywaa in the east
renegotiated their subordinate status in the traditional power game through differen-
tial access to modern goods and services, particularly modern education. The edu-
cated elites of the Anywaa and the Dinka predate their Nuer counterparts in their
incorporation into the state system. This was reflected in the earlier advancement of
the Anywaa in the Ethiopian state system and the predominance of the Dinka in
leadership positions in the southern Sudanese liberation movements. In the new
‘ethnoscape’ the social and political advancement of the ‘cattleless’ Anywaa (bär) and
206 Playing Different Games

the ‘slave’ Dinka (jaang) accentuated the Nuer sense of relative deprivation. The
SPLA atrocities against the Gaat-Jak Nuer was made intelligible through a new
ethnic scheme of interpretation. Thus, facing the ‘SPLA-Dinka’ in southern Sudan
and the politically dominant Anywaa in the GPNRS, the Nuer sought military and
political support from politico-military organizations that claimed to represent them.
By the mid-1990s, Riek Machar had emerged as one of the key political actors in
the regional power game through the skilful manipulation of both the Nuer
prophetic tradition and the politics of clientelism (Hutchinson 2001). Lacking access
to international borders and a foreign backer, Riek developed a strategic special inter-
est in the eastern Nuer (the Jikany and the Lou).15 He sought to establish links with
the Jikany Nuer by capitalizing on the SPLA’s unpopularity among the Jikany. The
rising political influence of Riek and his military power were promising for the
Jikany as a means to ‘restore’ their masculinity. The Jikany felt ‘feminized’ by the
Dinka, who they long despised and who they identified with the SPLA leadership.
The gender discourse on inter-ethnic relations is evident in this song of praise, pas-
sionately listened to by the Nuer in Gambella in the 1990s.16

Riek Machar and Lam Akol

Even if you do not do anything for us
You made us men again
Garang Mabior sent the Bor people abroad for education
And kept us on the mouth of the big machine gun
He thought that we would not see what he was doing
But we saw it clearly
The person who we are afraid of is not Garang
But the Nuer [Commander William Nyony] who is behind Garang
Had it been in olden times
This [Nuer humiliation] would have never happened
In the future [independent southern Sudan] we the Nuer will take up the fol-
lowing positions
We will have the Prime Minister. Our major generals will be Ministers

Nuer political actors, such as Dr Riek Machar, operate according to cultural assump-
tions that are reflected in the way they define the nature of inter-ethnic relations. In
the interview that I conducted with Riek Machar in 2002, he framed the
Anywaa–Nuer conflict in the following manner:

There is something intriguing about Nuer culture. Had it not been for the
British, the whole of southern Sudan and beyond would have become Nuer.
Some cultures are strong, others are not. The Nuer are expanding not by

15. The Jikany occupy all the border areas from Jikaw to Akobo districts.
16 . Praise songs are one of the mechanisms used to create and build up personality cults among
the rebel leadership in southern Sudan. SPLA recruits would spend eight to ten hours a day in
songs of praise to John Garang (Nyaba 2001: 52).
 Civil War in the Sudan and Ethnic Processes in the Gambella Region 207

force, but through peaceful exchanges with their neighbours. It is a human

right for Nuer immigrants to ask for their political recognition. What is
wrong if they ask to administer their own affairs? Let others also expand
demographically and do like the Nuer. Why are the Anywaa not many? A lot
of Nuer die in all kinds of wars, but they are still many. Whose problem is
this Nuer expansion? Do the Nuer consider it as a problem? Not really. I do
not consider it so either. Anyway, even if I want to stop them expanding, I
cannot. They do it on their own. The Nuer also apply the same rules among
themselves. There is a recent case in Pagak. The land is Thiang, but the
majority of the people are the Cieng Reng. As the Cieng Reng have become
the majority, they have asked for the position of a chief in Pagak. The issue
was brought to me, and I endorsed the Cieng Reng demand, because this is
how the Nuer system works. But the most important thing is not who the
leader is but how the people are administered. The Nuer care for people, espe-
cially for minorities who live amidst them. During all the period of conflict
in Gambella, nothing happened to the Anywaa who still lived in Jikaw. This
was because the Thiang protected them. Would the Anywaa do such things?
I doubt it. What we see is indiscriminate killing of the Nuer by the Anywaa.
(Extract from interview with Dr Riek Machar, Nairobi, 26 August 2002)

Riek then went on to locate the ‘root cause’ of the conflict between the Anywaa and
the Nuer in their respective cultural forms:

As I see it, the problem between the Anywaa and the Nuer is a problem of an
open and closed system. The Anywaa need to open up their social system. I
had hoped that this would be possible if the half-Nuer son of Agada [the late
Anywaa king] become the king. Unfortunately he died prematurely. Perhaps
their new king [nyiya Adongo] can make a difference because he is educated.

This was basically a call for identity change for the Anywaa. From Riek’s perspective,
ethnic worth is measured in terms of its expansionist capacity. In his definition of the
conflict situation, Nuer culture is better than the Anywaa and this is evident in its
capacity to assimilate and expand. Riek’s ‘cultural Darwinism’ echoes Peter and Philip’s
contestation of Nuer assimilationism through ‘cultural evolutionism’, discussed in
Chapter 5; both perspectives are, in effect, veiled justifications for cultural hegemony.
The bulk of Nuer expansionism into Anywaa territories has occurred through the
micro-demographic processes outlined in previous chapters. But there are also cases
where groups of Nuer have expanded through violence by displacing Anywaa. This
was partly possible thanks to their connections with various southern Sudanese rebel
groups and the government of the Sudan. The expansion of the Lou into the
Anywaa-inhabited areas of Akobo region, described in Chapter 4, is a case in point.
The first attempt to resolve the conflict between the Lou and the Jikany Nuer was
organized by Riek in what came to be known as the Akobo Peace Conference. Peace,
or more accurately, a ‘suspension of hostilities’, entailed a negotiated access to natural
resources for the resource-poor Lou Nuer. In order to allay the fears of the Gaat-Jak,
208 Playing Different Games

the SSIM instead put pressure on the neighbouring Anywaa to accept the permanent
resettlement of the Lou Nuer from their homeland in Wat to the Akobo region. Riek
justified his partisanship with the Lou in cultural terms in the following manner:

During the Akobo conference, the Nuer [Lou] brought thirty-five Anywaa
families who were displaced by the war. I organized a meeting between the
Anywaa and the Nuer. The Anywaa kwaaro agreed that the Nuer could live
with the Anywaa peacefully, but the majority of the Anywaa said the Nuer
should go back to Wat. The Nuer said the land is big enough for both of us,
why should we go back to Wat, where there is not enough water? I finally
endorsed the sharing proposal because you cannot displace them from Akobo
any longer and send them back to Wat. They have lived in the area long
enough. My Anywaa comrades [Simon Morris and Paul Anade] were disap-
pointed and considered me biased. That is the Anywaa problem. They forgot
that it was the Nuer in Akobo who elected an Anywaa [Paul Anade] as the
Akobo High Executive. The Akobo Nuer elected him because he was more
popular and able than the Nuer candidate.

For his contributions in ‘resolving’ intra-ethnic resource conflicts at the expense of

inter-ethnic peace, Riek received sound political support from the eastern Nuer in
his bitter fights with the SPLA. The outcome of the Akobo conference, on the other
hand, was resented by Anywaa members of the SSIM, who ultimately left the organ-
ization because of Riek’s favouritism towards the Nuer. For their part, the Ethiopian
Nuer officials in the Gambella regional state lobbied for the recognition of the SSIM
by the Ethiopian government, ostensibly on the basis of SPLA atrocities against the
local populations, but also as an expression of ethnic solidarity, thus further alerting
the Anywaa in Gambella of the ‘Sudanese’ dimension of their conflict with the Nuer.
In fact, Riek was briefly imprisoned in 1996 by the Gambella regional government.
Riek was arrested despite the fact that he had received permission from the federal
government in Addis Ababa to cross the border through Gambella. Many Nuer from
Gambella town expressed solidarity with Riek.17 This caused consternation on the
part of the Anywaa, who were sensitized to the ‘Nuer plot’. Riek was promptly set
free because of pressure from the Ethiopian government.
The cross-border political networks widened the asymmetry in local forms of
power. Above all, it increased the Anywaa’s anxiety that politics in southern Sudan
was yet another site where the Nuer ‘plot’ was being orchestrated, whether in the
form of refugee influx, or in Riek’s grand project of establishing a ‘Nuer kingdom’
extending from southern Sudan to Gambella. The Anywaa considered Riek and the
leaders of his politico-military organization to be the masterminds of what they
viewed as the Nuer ‘colonial project’, discussed in Chapter 5. Lacking a major patron
on the Sudanese side, the Anywaa felt apprehensive about their political marginaliza-

17. Many Nuer requested the prison authorities to allow them to join Riek in prison. In a dra-
matic show of solidarity, some even brought mattresses with them.
 Civil War in the Sudan and Ethnic Processes in the Gambella Region 209

tion in the regional power game. It was in this context that ‘capturing’ the regional
state of Gambella was found to be the only available political space for the Anywaa.
For the Anywaa, overstating their Ethiopian national identity was a way of preserv-
ing their own identity and attaining a politically dominant status in Gambella, which
they considered to be their home. Southern Sudan was not only uncertain but also
perceived to be an entity dominated by their larger neighbours, the Dinka and the
Nuer. This ethnic interest was projected onto the national level in order to access and
mobilize the Ethiopian state in a local struggle. Framed in national terms, the Nuer,
had they identified with a southern Sudanese state upon its independence, would
become not only threats to the Anywaa but to the Ethiopian state itself. The invoca-
tion of the international boundary by the Anywaa in their political struggle with the
Nuer was an attempt to construct the type of discursive power resource that had
brought them varying degree of success at various times.
In the context of an increasingly marginalized position in the politics of libera-
tion in southern Sudan, the Anywaa have taken a divided position. The atrocities of
the SPLA in the late 1980s precluded any meaningful cross-border political network,
whereas the Nuer officials in GPNRS openly advocated diplomatic recognition and
military support for Riek Machar. There were attempts by some Anywaa leaders,
though, to connect with the SPLA. This became possible after the resumption of the
mutual intervention between the governments of Ethiopia and the Sudan in the mid-
1990s. The political alliance between the Ethiopian government (EPRDF) and the
government of the Sudan (National Islamic Front) changed when the latter tried to
export political Islam into neighbouring countries, particularly Ethiopia and Eritrea
(de Waal 2004: 202–209). The tension reached crisis level when, in June 1995,
Sudanese-based terrorists attempted to assassinate Egyptian President Husni
Mubarak in Addis Ababa. Following this, the EPRDF began giving large-scale
support to Sudanese armed opposition groups, including the SPLA. Ethiopia joined
the anti-Sudanese alliance forged by the US, which was comprised of Eritrea and
Uganda. In order to contain the Islamic fundamentalism of the NIF regime of the
Sudan, the US provided the ‘frontline states’ with twenty million dollars in military
equipment (de Waal 2004: 220; J. Young 2007a: 4). In this pattern of regional
alliance, Riek Machar was allied to the government of the Sudan, which made the
Nuer political standing in Gambella precarious. In 1996, the SPLA recaptured the
Anywaa-inhabited Pochalla County from the government of the Sudan. This opened
up new lines of political networks between the Anywaa and the SPLA. As Riek
Machar drew ever closer to the Ethiopian Nuer and sought military assistance from
the Sudanese government, the SPLA tried to regain the confidence of the Anywaa.
The SPLA connived with the Anywaa officials in the regional council to deport
five thousand Nuer ‘refugee-residents’ from Gambella town. What started as ‘screen-
ing’ refugees from citizens in the event of war with the Sudanese government and its
affiliate SSIM ended as a full-fledged military operation in Newland, the Nuer
neighbourhood in Gambella town. Nuer citizens and refugees alike were indiscrim-
inately taken to the Sherkole refugee camp in the neighbouring Benishangul-Gumuz
regional state. Among the ‘deportees’ were the families of Nuer officials in the
regional council. This incident is narrated by the Nuer as an ‘experience of shame’
210 Playing Different Games

never to be repeated. Aware of the growing connection between the Anywaa and the
SPLA, as well as EPRDF’s backing of the Anywaa’s political dominance, albeit
volatile, in the regional government, the Nuer politicians intensified their contacts
with Nuer-based rebel groups in southern Sudan.
The SPLA/EPRDF’s raiding of Newland was welcomed by the Anywaa, who
considered the refugees part of the Nuer plot. The fact that nearly all of the current
Nuer officials and civil servants were educated in the refugee camps as southern
Sudanese refugees, and the ease with which people shuttled between refugee camps
and villages, blurred the distinction between a refugee and a citizen in Gambella. It
is no wonder, then, that more than thirty refugees were killed in Pinyudo refugee
camp during the deadly conflict in 2002 between the Anywaa and the Nuer on issues
related to political power. In fact, the trigger of the 13 December 2003 massacre of
the Anywaa was the killing of eight Ethiopian government officials who were on their
way to open a new refugee camp for southern Sudanese refugees in an area that the
Anywaa considered as their territory.
 Conclusion

Modes of Ethnic Identification

Based on an in-depth analysis of ethnographic data, oral accounts and archival mate-
rials, this book includes systematic explorations of two central and interrelated ques-
tions in the field of identity studies. The first pertains to the definition of ethnicity
or, more particularly, to the possible degree of variation in configurations of ethnic
identities; and the second concerns the causes of ethnic conflict. In the two sections
of this brief conclusion, I summarize my arguments, making general theoretical
statements regarding the two problems that I set out to solve in this book.

Modes of Ethnic Identification

Fredrik Barth is credited with having established an innovative paradigm in the study
of ethnicity that problematized the link between culture and ethnicity. Prior to the
publication of Ethnic Groups and Boundaries (1969), ethnic groups were regarded as
‘culture-bearing units’. In his theory of ethnicity, Barth placed special emphasis on stan-
dards of evaluation with reference to which membership in an ethnic community can
be determined. Accordingly, all one needs for ethnic membership is competence in the
value standards set by an ethnic group. In this scholarly construct, ethnic boundaries
are basically permeable, and ethnicity is understood as an interest-based system of com-
munication. This interest takes two forms, corresponding to two different levels of
social integration and differentiation. At the group level, ethnic differentiation occurs
in the context of an ethnic division of labour that consists of complementary exchanges
between people inhabiting different ecological niches. At the individual level, the per-
meability of the boundaries of ethnic groups allows pragmatic choices in identification
– choices that may involve the crossing of boundaries, which nevertheless persist.
Barth’s theory of ethnicity has inspired a number of studies and produced fruit-
ful insights in some ethnographic regions. However, Barth’s approach does not allow
for the various forms that ethnic boundaries might take. Some ethnic groups define
their boundaries as permeable, but others do not. In this empirical and comparative
study, I have attempted to demonstrate the relevance and also to indicate the limita-
tions of Barth’s theory of ethnicity. Barth is at his best among the Nuer. In their con-
structivist imagination of an ethnic group, the Nuer even go beyond Barth. The
‘applicants’ to ethnic membership among the Nuer are actively assisted in acquiring
the value standards and developing competence in relevant cultural practices. Once
admitted, however, new members find Nuer identity to be ‘sticky’. For men, it is
objectified bodily – in the initiation mark, gar – as a form of lifelong commitment.
Thus, even the permeable Nuer ethnic boundary is only permeable in one direction.
The Nuer seem to have elaborated on the constructivist imagination of ethnic iden-
tity to an even greater degree than the scholarly advocates of constructivism.
212 Conclusion

Barth might feel uncomfortable among the Anywaa. There, the standards of eval-
uation and the criteria of ethnic group membership are based exclusively on biolog-
ical descent. In Anywaa identity discourse, ethnicity is something one is born into,
part of the ‘givens of social existence’ (Geertz 1963: 109). Here we find an example
of an emic primordialist conceptualization of ethnicity, where Clifford Geertz would
certainly feel at home. But, contrary to extreme formulations of analytical primor-
dialism, the conditions under which such identity discourses and practices emerge
need to be explained, rather than being assumed to be something ‘ineffable’, as in
Geertz’s formulation. In this book, I have offered explanations for the construction
of Anywaa primordialism by examining their relationship to their own traditions and
to their social experiences in the world they inhabit.
Thus, the ethnographic data presented in this book not only relativize Barth’s
message but also challenge the common definition of an ethnic group as a form of
collectivity based on a subjective belief in common origins. Not all ethnic groups
construct their identity on the basis of origins. We have seen, for example, how the
Nuer embrace the diversity of origins. By loosening the criteria of ethnic member-
ship, the Nuer make it easier to absorb and effectively incorporate outsiders.
These reflections on the models proposed by Barth and Geertz lead us to a choice
between two options: We may favour one definition of ethnicity over the other and con-
clude on this basis that either the Anywaa or the Nuer do not form a real ethnic group,
because they do not match our favoured definition; or we may adopt a more flexible def-
inition of ethnic groups. Clearly, I am suggesting that we opt for the latter. Drawing on
Wittgenstein’s idea of concept formation, I have argued that ethnic groups need not
conform to a single definition listing invariable characteristics but may, instead, differ
rather widely, while still bearing a ‘family resemblance’ to one another. When the
problem is formulated this way, it is possible to accommodate various and variously con-
stituted ethnic groups, without overlooking their structural comparability. What is
more, it becomes possible to analyse the social and historical conditions under which an
ethnic group with primordialist tendencies moves in the direction of assimilationism
and vice versa. The Anywaa and the Nuer have differed in the trajectories of their ethnic
identity concepts. Whereas the discourse of ethnic identity among the Anywaa evolved
from an earlier emphasis on assimilation to the contemporary emphasis on the purity of
origins, the Nuer discourse evolved in the opposite direction; and now, under changing
circumstances, both discourses seem to be subject to further transformations. In both
cases, however, the identity discourses have been constructed and reconfigured through
a combination of variable social, economic and political processes.
In the preceding chapters, I have followed the changing configurations of Nuer
ethnic identity in various socio-political contexts. What seems to have been an earlier
ideology of ethnic purity was transformed into an elaborate assimilationism in the
context of migration and conquest. Competition for followers among various would-
be leaders led the latter to augment their groups of followers by incorporating outsiders.
What is more, group size became an important factor in securing access to and control
of vital resources, and the intrumentalization of inter-ethnic marriages and friendship
networks served similar purposes. Beneath the surface of the Nuer moral discourses on
assimilationism, then, we find varieties of political and economic rationality.
 Conclusion 213

Most recently, new socio-political contexts have led to a new reconfiguration of

Nuer ethnic identity concepts. Their quest for modernity has corroded confidence in
local culture and induced changes in the status system, expressed, for instance, in the
decline of gar as a defining factor of intra-ethnic social status and as an ideological
means for inter-ethnic assimilation. The Nuer encounter with ‘unmeltable’ neigh-
bours has also made them realize the limits of assimilation. In fact, the Anywaa pri-
mordial reaction to Nuer assimilationism has introduced new primordial currents
among the Nuer, reflected in the way in which the contemporary Nuer relate to the
Anywaa in Gambella. Hutchinson (2000) observed a similar trend towards the pri-
mordialization of inter-ethnic relations among the Dinka and the Nuer in the
context of a protracted civil war and the militarization of society in southern Sudan.
In her analysis, the primordialist reaction of the Dinka to the ‘sticky grasp of the
Nuer’ has led the Nuer, in turn, to begin making the transition from a hitherto ‘per-
formative’ kind of ethnicity to a ‘more closed and fixed “primordialist” concept based
on procreative metaphors of shared human blood’ (Hutchinson 2000: 9). The post-
1991 political structure in Ethiopia, which involves institutionalized ethnopolitics
and group-based electoral politics, has undermined the assimilationist thrust of the
Nuer not only at the inter-ethnic level but also internally, especially with the frag-
mentation of previously integrative relations among diel, rul and jaang in local com-
munities. The emerging discourse of ‘cieng purity’ gives expression to new processes
contributing to the primordialist reconfiguration of Nuer identity concepts.
The Anywaa, like their Nilotic neighbours, were originally assimilationist in the
context of earlier migrations, upon which they embarked long ago with their Lwoo
cousins. Protracted intergroup warfare is often central in the production and repro-
duction of political identities; but in the new lands where they settled, along the
Sobat River and in Gambella, the Anywaa did not encounter competitive ethnic
groups that might serve as their relevant others. Thus, in the context of relative iso-
lation, i.e., of relatively infrequent interaction with other groups, the Anywaa devel-
oped their primordialist concept of ethnic identity with reference to a relevant
‘spiritual other’. The luo-jwok contrast is central to the construction of Anywaa
ethnic identity, as described in this book. The image of a belligerent supernatural
power, and of Anywaa resistance to it, significantly informs their cultural world.
Anywaa territoriality must be understood in its connection with Anywaa cosmology.
The sudden and massive encounter with new neighbours who encroached upon
Anywaa territories was experienced with reference to a particular image of Jwok,
which seem to have functioned as a dispositif, in Foucault’s sense, i.e., as a conceptual
toolkit or apparatus for the ‘grouping of heterogeneous elements into a common
network’ (Rabinow 2003: 51). In fact, the Anywaa associate the Nuer with Jwok.
That is, the Anywaa view of God is a metaphor of their relationship with the more
powerful Nuer, who have, seemingly, placed them under siege. Both Jwok and the
Nuer are resisted for their incessant encroachment upon Anywaa territories. The ter-
ritorial and cultural encroachments of the Ethiopian state have reinforced the
Anywaa belief in a conspiracy against them, as if god, neighbours and the state are
allied to challenge their existence. At the same time, the Nuer view their own benign
god as an ally who is opposed to the Anywaa as they, the Nuer, are themselves.
214 Conclusion

In articulating their identification strategies, the Anywaa and the Nuer have
played different games in various domains of social life. I have explored these diver-
gences in four areas: identity concepts; host/guest relations; images of the supernat-
ural; and the definition of the inter-ethnic conflict situation. The very terms
‘Anywaa’ and ‘Nuer’ refer to quite different ideas of ethnic identity. Ethnic identity
is ‘given’ for the Anywaa, whereas it is ‘acquired’ by the Nuer. The concept of
host/guest also has different connotations. For the Anywaa a guest has a permanent
status, while such status is temporary among the Nuer. The Anywaa and the Nuer
also differ in their images of divinity. Jwok is not worshiped by the Anywaa; rather,
it is avoided because of its destructiveness. Kuoth, on the other hand, is a father-
figure, and the Nuer frame their relation with divinity in kinship terms. Thus, Kuoth
is also an ‘ally’ of the Nuer in the inter-ethnic scene. Nor do the Anywaa and the
Nuer agree in the definition of the conflict situation. For the Anywaa, it is ‘the ethnic
conspiracy of the Nuer’ seeking to bring about their extinction that is the root cause
of the trouble between them. For the Nuer, the problem is the Anywaa’s ‘xenopho-
bic’ tendencies, which are part of their ‘nature’. These contrasting modes of ethnic
identity formation create a potential for inter-ethnic conflict, but, in the analysis of
real conflicts, they must be viewed in relation to other conflict generating variables
and processes.

Causes of Ethnic Conflict

Ethnic conflicts have often been explained in monocausal terms, such as competition
over scarce natural resources, elite competition over new state mediated resources, the
resurgence of ‘ancient hatreds’, hostility generated by deep primordial feeling, or the
‘clash of cultures’. The approach I advocated is novel in two respects. For one, ethnic
conflict is explained in multicausal terms. I have identified three variables that must be
taken into consideration in formulating an adequate causal explanation of conflict
between the Anywaa and the Nuer – variables that may also be applicable to other case
studies: the resource variable, the identity variable and the power variable. The resource
variable operates at two levels: among ordinary men and women competing for
resources in rural settings and among elites competing for state mediated resources such
as jobs and political offices in urban settings. Struggles over material resources often
help to explain the intensity and the passion of individuals who identify with a partic-
ular ethnic category. However, the extent to which material resources are implicated in
ethnic conflict needs to be specified in particular inter-ethnic contexts. This is so
because the conflict over access to and control of resources might well be fought within
ethnic groups, rather than between them, or in terms that have little or nothing to do
with ethnic identity. While determining the nexus between resource and ethnic con-
flicts, I have explored variations within ethnic groups as interest groups. A simultane-
ous engagement with resource conflicts at the inter- and intra-ethnic level helps us to
paint a more nuanced causal link between the resource and identity variables.
The second causal variable in explanations of ethnic conflict is identity. As I have
argued at length, the encounter between groups with contrasting modes of ethnic
identity formation, such as the Anywaa and the Nuer, may result in conflict. The
intervening variable in the causal link between identity and ethnic conflict is the dif-
 Conclusion 215

ferent ‘language games’ generated by the contrast and the misunderstandings related
to them. This is expressed, to cite one very important example, in the different value
that land has for the Anywaa and the Nuer, respectively, and in the inevitable mis-
understandings that emanates from struggles over land. For the Nuer, land has pri-
marily an economic value; and this, coupled with their relative marginality in the
distribution of vital natural resources, drives the territorial expansion of the Nuer.
For the Anywaa, on the other hand, the land is not only an object of negotiations in
inter-ethnic relations; it also fulfils fundamentally important symbolic functions in
the construction of subethnic identities. Therefore, in the conflict between the
Anywaa and Nuer, one can observe a ‘clash of cultures’ of a fundamental type.
The third cause of ethnic conflict may be understood in terms of the power variable
or, more specifically, the different patterns of incorporation of ethnic groups into ethni-
cally stratified states. Conflict between the Anywaa and the Nuer cannot be explained
without referring to their mode of interaction with the Ethiopian and Sudanese states
into which they are, generally, unfavourably integrated but within which they have been
variously positioned. Anywaa–Nuer conflicts and the dynamic (re)configuration of their
respective ethnic identities are generated through state-related political processes. In
Ethiopia, this reconfiguration is related to the failures of the Ethiopian state to create an
inclusive national identity. Instead, it has created a competitive political space within
which a ‘zero sum’ game is played. The state’s strategic cooption of the Anywaa or the
Nuer at various times has created fluctuating inter-ethnic power relations, instead of
helping them to articulate regional interests or a wider national identity. In fact, the his-
torical chapters of this book have shown how the Anywaa and the Nuer have been busy
in contesting or subverting each other’s power. The creation of this competitive political
space is as much by default as by design. The political tenure of the Ethiopian state in
border regions such as Gambella is tenuous, partly because of the greater cultural dis-
tance between the peoples of the borderlands and the dominant ethnic groups of the
Ethiopian polity. Instead of investing politically in fostering an Ethiopian national iden-
tity and corresponding feelings of belonging in the borderlands, successive Ethiopian
governments have exploited and fanned Anywaa–Nuer conflict to enhance their politi-
cal control over the Gambella region. More ominously, they have sought to connect
with the Highlanders to neutralize actual and potential ownership claims of the Anywaa,
the Nuer, or both in regional politics. The changing but consistently ambivalent rela-
tions among the Anywaa, the Nuer and a series of Ethiopian governments amply sub-
stantiates the truism that the conflict between the Anywaa and the Nuer is not bipolar.
The EPRDF seemed initially to have departed from this political tradition by introduc-
ing a new ideological spin to the age-old centre-periphery relations, i.e., between the
Ethiopian state and its minorities. Unfortunately, the EPRDF could not sustain what
might have been genuinely reformist aspects of its programme because of its drive
towards total control over post-1991 political processes and its pragmatic approach to
the volatile geopolitics of the Horn of Africa.
The conflict between the Anywaa and the Nuer has also been generated by the
protracted Sudanese civil wars. The Anywaa and the Nuer have been variously posi-
tioned in the politics of liberation in southern Sudan. The cattle wealth of the Nuer
and their larger demographic presence in the Sudan have made them more attractive
216 Conclusion

than the Anywaa to political actors seeking allies in the civil wars. Actors in Nuer
local communities have instrumentalized this appeal and constructed political and
military power, which they have exploited in determining the outcome of local
resource conflicts. The Anywaa, on the other hand, have largely been on the receiv-
ing end of the adverse impact of the Sudanese civil wars. This is expressed, above all,
in the changing demographic situation, which has reduced them to the status of an
ethnic minority in the region that they regard as their homeland. It has led to the loss
of further Anywaa territories to the Nuer. These factors have contributed to the
development, among the Anywaa, of a siege mentality and a tendency to subscribe
to conspiracy theories – a development that has undermined their capacity to forge
dynamic alliances with various political actors in regional politics in their own terms
and in the spirit of realpolitik. The seemingly desperate situation of the Anywaa only
increases the potential for further violence in Gambella.
 Glossary of Local Terms
Anywaa Terms
Abudho String of beads
Agelgalcher Agriculture
Agem Village rebellion
Agwaga A village anthem
Ajuwa Ritual expert
Amulo Calabash without neck
Anger Cob antelope
Awido Long, narrow–necked calabash
Bell Sorghum; leopard
Bilo Exchange marriage
Bura Local militia
Chirawiya A symbolic act denoting the right to kill
Ci-Jwok Wives of Jwok
Ciro Anywaa who live in the Akobo region
Dho Anywaa Anywaa language
Dho-oto Clan
Dibuoc Gilo Followers of Gilo
Digwi Watnaadhi Lizard of the Watnaadhi people
Dimui Blue glass beads used as bridewealth
Gaala Anywaa name for the Highlanders
Girgir The turmoil (specifically, during regime change in 1991)
Goro Gar (see ‘Nuer terms’)
Gungi The ceremonial low-bowing posture
Ji-kwaaro People of the headmen
Ji-nyiya People of the nobles
Jobur First settlers of a village
Jur Foreigner
Jwok A Supernatural Being
Jwok nyingalabuo The benevolent Jwok
Jwok nyidungu The malevolent Jwok
Kac Hunger
Kap Prostitution
Ker Calabash
Kew Boundary
Köro An organized rebellion against Jwok
Ku Theft
Kuac Skin
Kuruach Thief
Kwaari Headmen
218 Glossary of Local Terms

Kwaaro Headman
Kwac Begging
Kwar nyigilo Descendants of Gilo
Kwec gel Salaried government officials
Laegnmajid The wars of Majid
Lango Slave
Lull Anywaa who live in the forested areas
Luo Pure (as in ‘ethnic purity’)
Math Clan salutation
Medho The Dog which saved the Anywaa
Naak Dental evulsion
Naam duong A large river
Ngom Soil
Nyibur Deputy of Kwaaro
Nyikugu Adviser to Kwaaro
Nyilwinyjwok The chosen nation
Nyinya The nyiya who holds the royal emblems
Nyipem Prince
Nyiya Noble
Nyiye Nobles
Nyooyu Honorific salutation of the Jowatnaadhi
Nyuwei Iron
Odolo The village drum
Paak Honorific title
Padhano Nonhuman
Peath Witch
Piny Land
Pinykwara Ancestors
Pö A spontaneous ritual of resistance against the wrath of
Jwok
Ramo A spiritual transplanting of bone by the witch to the
victim
Safara Resettled Highlanders
Tung Lineage
Tung Akwei The Murle and Nuer captives who are attached to the
lineage of Nyiya Akwei
Tung goc Royal lineage based in the Adongo region
Tung udola Royal lineage based in the Abobo region
Uchuok The royal necklace
Ucuudho Divine ancestor of the nobles
Ugaala A person from mixed parents
Utak Beating of a husband by in-laws for failing to take care
of his wife
Walo The royal stool
Warakata Paper
 Glossary of Local Terms 219

Wat-ngomi Earth priest

Welo Stranger
Wora Ariat President Mengistu
Wudo Ceremonial respect

Nuer Terms
Bär People without cattle/Anywaa
Bell Sorghum
Bi jile duoth Accepting the leftover
Biem Cooling-off period after a feud
Boum Power
Bunam Total mobilization of the youth
Buny The Highlanders
Buny cie turuk ‘Highlanders are not modern’
Buth Agnatic kins
Caa naath Becoming Nuer
Camun naak Right of conquest
Chan Poor
Ciang Culture
Cie buny michar ‘There are no black Ethiopians’
Ciek joka Ghost marriage
Cieng Local community
Colwic God of thunder
Darchieng Original home of the Nuer
Deeth Ironsmith’s tools
Dhol Uninitiated boy
Diel Dominant/aristocratic lineage
Dil Member of a dominant/aristocratic lineage
Dwac Dwelling
Gaatnyiet Sons of daughters of diel
Gaattutni Sons of diel
Gam Half
Gar Male initiation mark
Gee Ancestor of the ‘original’ Nuer
Guan yier Master of the water
Guk Prophet
Jaang A non-Nuer immigrant/slave
Jal tang Long-distance guest
Jalab Arabs
Jenubi South Sudan
Jeop Axe which Latjor hid in his hair
Jikany An eastern Nuer tribe
Jing The Dinka foster uncle of Kir
Kal A man without cattle
Kir Ancestor of the Jikany Nuer
220 Glossary of Local Terms

Kom Chair on which Kir sat in the guard

Koat Lieh Place where rual was first practiced
Kuaar muon Leopard skin chief
Kuar Leader
Kume Government; state power
Kuoth The Nuer Supreme Being/sacrifice
Kur Tribal wars
Larcieng Betrothal
Lät Subhuman creature
Latjor Person who left the Jikany migration to the east
Luak Cattle byre
Luuch naath Murderous people, a Nuer derogatory name for the
Anywaa
Madh Intimate friendship
Mibor White (European)
Michar Black
Milwal Red (Highlander)
Mut A consummated marriage
Mut pini duong Tik’s spear name
Mut Wiu Divinity spear of Jikany
Naath cieng Homeland
Naath Mibor White people
Naath Michar Black people
Naath Milwal Red people
Nei ti naadth Distinct people
Ngok The blue heron bird
Ngut Wedding
Nok Prophet who helped Latjor mobilize followers
Nyal bany Airplane
Nyam duar Hinterland
Nyayou Honorific title of the Jikany tribes
Pac Cattle raiding
Piny Land
Ran mi jaang Person who is not original Nuer
Ran mi ran Real person/Nuer
Ric Age set
Rual A ritual of cutting a cow into two halves – i.e., forbid-
ding marriage or sexual intercourse between members
Ruic Spokesman
Rul A Nuer immigrant
Tele buny michar ‘There are no black Ethiopians’
Ter Intra-tribal feud
Thok dwiel Lineage
Thok nueri Nuer language
Tik Ancestor of the Cieng Reng
 Glossary of Local Terms 221

Toiche Green pasture during the dry season

Tuach kuach Leopard skin
Turbiel Car
Turuk mi thil kade Civilization without ‘salt’
Tut Bull (leader)
Tut kernyang Black-and-white brindled bull that Ngundeng gave to
the Highlanders
Tuut dhoali Uninitiated adults (bull-boys)
Was-ngok Place where the Jikany crossed the river during their
expansion to the east
Wut Initiated adult
Ye thiang Right of conquest
Yien Tied; attached to the dil

Amharic Terms
Abun Amharic Patriarch of the Orthodox Church
Awraja District
Balabats Local imperial officials
Bariya Slave
Biher Nation
Bihereseb Nationality
Birr The national currency
Chiqona Exploitation
Degegna Highlanders
Dejazmach Commander of the Gate
Derg Military regime that ruled Ethiopia from 1975–91
Ehadeg Ethiopian Peoples Revolutionary Democratic Front
Enjera Ethiopian pancake
Fangay Slave raiders
Fitawrari Commander of the Vanguard
Galla Derogatory term for the Oromo
Habesha Highlanders’ ethnonym
Ityopia tikdem Ethiopia first
Kebele Peasant association
Kilil Regional state
Kokeb Medal
Lemma Highlanders’ derogatory name for the Anywaa and the
Nuer
Liqemember Chairman of a peasant association
Lualawinet Sovereignty
Maqinat Civilizing mission
Metaweqiya Identity card
Niqatehilina False consciousness
Qenyazmach Commander of the Right
Qey Red
222 Glossary of Local Terms

Safari Resettled Highlanders

Shifta Bandit
Silitane Civilization
Tewelaj Native
Tikur Black
Tiqim Benefit
Wastina Collateral
Wereda Subdistrict
Wonbede Pejorative term for rebel
Yetesasate False
Zemach University students and high school teachers who par-
ticipated in the cultural revolution
Zemecha Cultural revolution

Oromo Terms
Dabi garba Slave quarter
Daga garba Stones on which the slaves sat
Guracha Oromo Black Oromo
Soso A type of grass put around slaves’ necks to make them
look healthier

Shilluk Terms
Cie nwar to bie tuokto An unruly child
Yi ba Onwa When someone is not behaving well
 References
Published Sources
Addis, H. 1975. Ethiopia: From Autocracy to Revolution. London: Merlin Press.
Anuak Justice Council 2006. ‘The Anuak Massacre of 2003: The Ethiopian
Government Attacks an Ethnic Group Listed by Cultural Survival in 1984 as
Endangered’, Testimony to the United States House Subcommittee on Africa.
Retrieved from http://www.anuakjustice.org/doc_index.htm.
Austin, M.H.H. 1901. ‘Survey of the Sobat Region’, Geographical Journal 17(5):
495–512.
Ayalew, G. 1997. ‘Conflict over Land Use: Pastoralism, Commercial and State
Agriculture: The Case of Afar’, in K. Fukui et al. (eds), Ethiopia in Broader
Perspective: Papers of the 13th International Conference of Ethiopian Studies, Vol.
2. Kyoto: Shokado, pp. 358–77.
Bahru, Z. 1987. ‘An Overview and Assessment of Gambella Trade (1904–1935)’,
The International Journal of African Historical Studies 20(1): 75–94.
–––. 2001. A History of Modern Ethiopia 1855–1991. Oxford: James Currey.
Barth, F. 1969. ‘Introduction’, in F. Barth (ed.), Ethnic Groups and Boundaries: The
Social Organization of Culture Difference. Bergen, Oslo: Universitetsforlaget, pp.
9–38.
–––. 2000 [1994]. ‘Enduring and Emerging Issues in the Analysis of Ethnicity’, in
H. Vermeulen and C. Govers (eds), The Anthropology of Ethnicity: Beyond Ethnic
Groups and Boundaries. Amsterdam: Het Spinhuis, pp. 11–32.
Bender, M. et al. 1976. Language in Ethiopia. London: Oxford University Press.
Brass, P. 1999. Ethnicity and Nationalism: Theory and Comparison, 9th ed. New
Delhi: Sage Publication.
Brewer, M. and N. Miller. 1996. Intergroup Relations. Buckingham: Open University Press.
Brown, D. 2007. ‘Ethnic Conflict and Civic Nationalism: A Model’, in J. Peacock,
P. Thornton and P. Inman (eds), Identity Matters: Ethnic Conflict and Sectarian
Conflict. Oxford: Berghahn, pp. 15–33.
Brown, R. 1995. Prejudice: its Social Psychology. Oxford: Blackwell.
Carlson, H. 1937. ‘Loan-Words from German’, American Speech 12(3): 232–34.
Central Statistics Authority. 1995. The 1994 Population and Housing Census of
Ethiopia. Results for Gambella Region. Addis Ababa: Central Statistics Authority.
Chabal, P. and J. Daloz. 1999. Africa Works: the Political Instrumentalization of
Disorder. Bloomington: Currey.
Clapham, C. 2002. ‘Controlling Space in Ethiopia’, in W. James et al. (eds),
Remapping Ethiopia: Socialism and After. Oxford: Currey, pp. 9–32.
Collins, R. 1971. Land beyond the Rivers: The Southern Sudan, 1898–1918. New
Haven: Yale University Press.
–––. 1983. Shadows in the Grass: Britain in the Southern Sudan, 1918–1956. New
Haven: Yale University Press.
224 References

Comaroff, J. 1996. ‘Ethnicity, Nationalism and the Politics of Difference in an Age

of Revolution’, in E. Wilmsen and P. McAllister (eds), The Politics of Difference:
Ethnic Premises in a World of Power. Chicago: The University of Chicago Press,
pp. 162–84.
Connor, W. 1994. Ethnonationalism: The Quest for Understanding. Princeton:
Princeton University Press.
Conversi, D. 2002. Ethnonationalism in the Contemporary World: Walker Connor and
the Study of Nationalism. London: Routledge.
Crazzolara, J.P. 1950. The Lwoo, Part I, Lwoo Migrations. Verona: Missioni Africane.
–––. 1951. The Lwoo, Part II, Lwoo Traditions. Verona: Nigrizia.
–––. 1954. The Lwoo, Part III, Clans. Verona: Nigrizia.
Deng, F. 1995. War of Visions: Conflicting Identities in the Sudan. Washington:
Brookings Institution Press.
Dereje, F. 2005. ‘Land and the Politics of Identity: the Case of Anywaa-Nuer
Relations in the Gambella Region’, in S. Evers and M. Spierenburg, H. Wels
(eds), Competing Jurisdictions: Settling Land Claims in Africa. Leiden: Brill, pp.
203–22.
–––. 2006a. ‘The Experience of the Gambella Regional State’, in D. Turton (ed.),
Ethnic Federalism: the Ethiopian Experience in a Comparative Perspective. Oxford:
Currey, pp. 208–30.
–––. 2006b. ‘Decentralization as Ethnic Closure, with Special Reference to a
Declining Negotiated Access to Natural Resources in Western Ethiopia’, in J.
Ribot and R. Oyono (eds), Decentralisation and Livelihood in Africa. Dakar:
CODESRIA, pp. 243–60.
de Waal, A. 1997. Famine Crimes: Politics & the Disaster Relief Industry in Africa.
London: African Rights and the International African Institute.
––– (ed.). 2004. Islamism and Its Enemies in the Horn of Africa. London: Hurst.
Donham, D. 1999. Marxist Modern: An Ethnographic History of the Ethiopian
Revolution. Berkeley: University of California.
–––. 2002. ‘Looking Back on Projects of the Socialist State, 1974–91’, in W. James
et al. (eds), Remapping Ethiopia: Socialism and After. Oxford: Currey, pp. 33–36.
––– and W. James (eds). 1986. The Southern Marches of Imperial Ethiopia: Essays on
History and Social Anthropology. New York: Cambridge University Press.
Epstein, A. 1978. Ethos and Identity. London: Tavistock.
Eriksen, T. 1991. ‘The Cultural Contexts of Ethnic Differences’, Man (NS) 26:
127–44.
Evans-Pritchard, E. 1940a. The Nuer: A Description of the Modes of Livelihood and
Political Institutions of a Nilotic People. Oxford: Oxford University Press.
–––. 1940b. The Political System of the Anuak of the Anglo-Egyptian Sudan. New York:
AMS Press.
–––. 1947. ‘Further Observations on the Political System of the Anuak’, Sudan Notes
and Records 28: 62–97.
–––. 1951a. Kinship and Marriage among the Nuer. Oxford: The Clarendon Press.
–––. 1951b. ‘Some Features of Nuer Religion’, Journal of the Royal Anthropological
Institute of Great Britain and Ireland 811(2): 1–13.
 References 225

–––. 1953. ‘Nuer Spear Symbolism’, Anthropological Quarterly 28(1): 1–19.

–––. 1956. Nuer Religion. Oxford: Clarendon Press.
Fenton, S. 2003. Ethnicity. Cambridge: Polity Press.
Fergusson. J. 1990. The Anti-Politics Machine: ‘Depolitisation’, and Democratic Power
in Lesotho. Minneapolis: University of Minnesota Press.
–––. 1999. Expectations of Modernity: Myths and Meanings of Urban Life on the
Zambian Copper-Belt. Berkeley: University of California Press.
Forbes, H. 2004. ‘Ethnic Conflict and the Contact Hypothesis’, in Y. Lee (ed.), The
Psychology of Ethnic and Cultural Conflicts. Westport: Praeger, pp. 69–88.
Foster, R.J. 1991. ‘Making National Cultures in the Global Ecumene’, Annual
Review of Anthropology 20: 235–60.
Fukui, K. and J. Markakis (eds). 1994. Ethnicity and Conflict in the Horn of Africa.
London: Currey.
Gebru, T. 1991. Ethiopia: Power and Protest; Peasant Revolts in the Twentieth Century.
Cambridge: Cambridge University Press.
Geertz, C. 1963. ‘The Integrative Revolution’, in C. Geertz (ed.), Old Societies and
New States. New York: Free Press, pp. 105–57.
–––. 1993. ‘Ethnic Conflict: Three Alternative Terms’, Common Knowledge 2:
54–65.
Getachew, K. 2001. Among the Pastoral Afar in Ethiopia: Tradition, Continuity and
Socio-economic Change. Utrecht: International Books.
Gil-White, F. 1999. ‘How Thick Is Blood? The Plot Thickens: If Ethnic Actors Are
Primordialists, What Remains of the Circumstantialist/Primordialist
Controversy?’, Ethnic and Racial Studies 22(5): 789–820.
–––. 2001. ‘Are Ethnic Groups Biological “Species” to the Human Brain?
Essentialism in Our Cognition of Some Social Categories’, Current Anthropology
42(4): 515–54.
Gough, K. 1971. ‘Nuer Kinship: A Re-examination’, in T. Beidelman (ed.), The
Translation of Culture. London: Tavistock Publications, pp. 79–121.
Haaland, G. 1969. ‘Economic Determinants in Ethnic Process’, in F. Barth (ed.),
Ethnic Groups and Boundaries. Bergen, Oslo: Universitetsforlaget, pp. 58–73.
Hagmann, T. and M. Khalifa. 2006. ‘State and Politics in Ethiopia’s Somali Region
Since 1991’, An International Journal of Somali Studies 6: 25–49.
Harris, M. 2001. The Rise of Anthropological Theory: A History of Theories of Culture.
New York: Rowman & Littlefield Publishers.
Headland, Thomas N., Kenneth L. Pike and Marvin Harris (eds). 1990. Emics and
Etics: the insider/outsider debate. Newbury Park, CA: Sage Publications.
Hechter, M., and S. Kanazawa. 1997. ‘Sociological Rational Choice Theory’, Annual
Review of Sociology 23: 191–214.
Holy, L. 1979. Segmentary Lineage Systems Reconsidered. Belfast: Belfast Queens
University Press.
Horowitz, D. 1985. Ethnic Groups in Conflict. Berkeley: University of California Press.
–––. 1998. Structure and Strategy in Ethnic Conflict. Paper Prepared for the Annual
World Bank Conference on Development Economics, Washington, D.C.,
20–21 April 1998.
226 References

–––. 2002. ‘The Primordialists’, in D. Conversi (ed.), Ethnonationalism in the

Contemporary World: Walker Connor and the Study of Nationalism. London:
Routledge, pp. 72–82.
Horstmann, A. and G. Schlee. 2001. Integration durch Verschiedenheit: Lokale und
Globale Formen Interkultureller Kommunikation. Bielefeld: Transcript.
Howell, P. 1954. A Manual of Nuer Law. London: Oxford University Press.
Hultin, J. 1996. ‘Perceiving Oromo: “Galla” in the Great Narrative of Ethiopia’, in
P. Baxter and J. Hultin (eds), Being and Becoming Oromo: Historical and
Anthropological Enquiries. Uppsala: Nordiska Afrikainstitutet, pp. 81–91.
Human Rights Watch. 2005. ‘Targeting the Anuak: Human Right Violations in
Ethiopia’s Gambella Region’. Retrieved from www.hrw.org/reports/2005/
Ethiopia 0305/.
Hutchinson, S. 1985. ‘Changing Concepts of Incest among the Nuer’, American
Ethnologist 12(4): 625–41.
–––. 1994. ‘On the Nuer Conquest’, Current Anthropology 35(5): 643–51.
–––. 1996. Nuer Dilemmas: Coping with Money, War and the State. Berkeley:
University of California Press.
–––. 2000. ‘Nuer Ethnicity Militarized’, Anthropology Today 16(3): 6–13.
–––. 2001. ‘A Curse from God? Religious and Political Dimensions of the Post-1991
Rise of Ethnic Violence in Southern Sudan’, The Journal of Modern African
Studies 39(2): 307–31.
Hyden, G. 1980. Beyond Ujamaa in Tanzania: Underdevelopment and an Uncaptured
Peasantry. Berkeley: University of California Press.
Issacs, H. 1975. Idols of the Tribe. New York: Harper and Row.
Jackson, H. 1923. The Nuer of the Upper Nile Province. Khartoum: El Hadra Printing Press.
James, W. et al. (eds). 2002. Remapping Ethiopia: Socialism and After. Oxford: Currey.
Jenkins, R. 2001. ‘Ethnicity: Anthropological Aspect’, in S. Neil and B. Paul (eds),
International Encyclopaedia of the Social and Behavioural Sciences. Oxford:
Elsevier, pp. 4824–4828.
Jessen, H. 1905. ‘South Western Abyssinia’, The Geographical Journal 25.
Johnson, D. 1982. ‘Tribal Boundaries and Border Wars: Nuer-Dinka Relations in
the Sobat and Zaraf Valleys’, Journal of African History 23: 183–201.
–––. 1986. ‘On the Nilotic Frontier: Imperial Ethiopia in the Southern Sudan,
1898–1936’, in D. Donham and W. James (eds), The Southern Marches of Imperial
Ethiopia. Essays in History and Social Anthropology. Oxford: Currey, pp. 219–45.
–––. 1989. ‘Political Ecology in the Upper Nile: the Twentieth Century Expansion
of the Pastoral “Common Economy”’, The Journal of African History 30(3):
463–86.
–––. 1994. Nuer Prophets: A History of Prophecy from the Upper Nile in the Nineteenth
and Twentieth Century. Oxford: Oxford University Press.
–––. 2003. Root Causes of Sudan’s Civil Wars. Oxford: Currey.
–––. 2009. ‘The Nuer Civil Wars’, in G. Schlee and E.E. Watson (eds), Changing
Identifications and Alliances in North-East Africa, Vol.II: Sudan, Uganda and the
Ethiopia-Sudan Borderlands. Oxford: Berghahn, pp. 31–47.
Jok, M.J. and S. Hutchinson. 1999. ‘Sudan’s Prolonged Second Civil War and the
 References 227

Militarization of Nuer and Dinka Ethnic Identities’, African Studies Review

42(2): 125–45.
Joseph, G. and D. Nugent (eds). 1994. Everyday Forms of State Formation: Revolution
and the Negotiation of Rule in Modern Mexico. Durham: Duke University Press.
Kelly, R. 1985. The Nuer Conquest: The Structure and Development of an Expansionist
System. Ann Arbor: University of Michigan Press.
Kinfe, A. 2001. Ethiopia: From Empire to Federation. Uppsala: NINA Press.
Kuper, A. 1982. ‘Lineage Theory: A Critical Retrospect’, Annual Review of
Anthropology 11: 71–95.
Kurimoto, E. 1992. ‘Natives and Outsiders: The Historical Experience of the
Anywaa of Western Ethiopia’, Journal of Asian and African Studies 43: 1–43.
–––. 1994. ‘Inter-ethnic Relations of the Anywaa (Anuak) in Western Ethiopia, with
special reference to the Majangir’, in H. Marcus (ed.), Ethiopia 94: Papers of the
12th International Conference of Ethiopian Studies. Michigan: Michigan State
University, pp. 899–912.
–––. 1996. ‘People of the River: Subsistence Economy of the Anywaa of Western
Ethiopia’, in S. Sato and E. Kurimoto (eds), Essays in Northeast African Studies.
Osaka: National Museum of Ethnology, pp. 29–58.
–––. 1997. ‘Politicization of Ethnicity in Gambella Region’, in K. Fukui, E.
Kurimoto and M. Shigeta (eds), Ethiopia in Broader Perspective: Papers of the
13th International Conference of Ethiopian Studies Vol. 2. Kyoto: Shokado, pp.
798–815.
–––. 2001. ‘Capturing Modernity among the Anywaa of Western Ethiopia’, in E.
Kurimoto (ed.), Rewriting Africa: Toward Renaissance or Collapse. Osaka: Japan
Centre for Area Studies, pp. 263–80.
–––. 2005. ‘Multidimensional Impact of Refugees and Settlers in the Gambella
Region, Western Ethiopia’, in Itaru Ohta and Yntiso Gebre (eds), Displacement
Risks in Africa. Kyoto: Kyoto University Press, pp. 338–58.
––– and S. Simonise. 1998. Conflict, Age and Power in North East Africa: Age Systems
in Transition. Oxford: Currey.
Kymlicka, W. 2006. ‘Emerging Western Models of Multination Federalism: Are
They Relevant to Africa?’, in D. Turton (ed.), Ethnic Federalism: The Ethiopian
Experience in Comparative Perspective. Oxford: Currey, pp. 32–64.
Leach, E. 1954. Political Systems of Highland Burma: A Study of Kachin Social
Structure. London: Athlone Press.
Lentz, C. 2000. Ethnicity in Ghana: The Limits of Inventionism. Hampshire: Macmillan.
Lester, A. 2002. ‘Colonial Settlers and the Metropole: Racial Discourse in the early
19th-Century Cape Colony, Australia and New Zealand’, Landscape Research
27(1): 39–41.
Lienhardt, G. 1957. ‘Anuak Village Headmen I’, Africa 27: 341–55.
–––. 1958. ‘Anuak Village Headmen II’, Africa 28: 23–36.
–––. 1962. ‘The Situation of Death and Anuak Philosophy’, Anthropological
Quarterly 35(2), East and Central Africa: 74–85.
Llewelyn, J. 1968. ‘Family Resemblance’, The Philosophical Quarterly 18(73):
344–46.
228 References

Markakis, J. 1987. National and Class Conflict in the Horn of Africa. Cambridge:
Cambridge University Press.
McKinnon, S. 2000. ‘Domestic Exceptions: Evans-Pritchard and the Creation of
Nuer Patrilineality and Equality’, Current Anthropology 15(1): 35–83.
McLaughlin, J. 1967. ‘Tentative Time Depths for Nuer, Dinka and Anuak’, Journal
of Ethiopian Studies 5: 13–27.
Medhane, T. 2007. ‘Gambella: The Impact of Local Conflict on Regional Security’,
in Researching Local Conflicts and Regional Security. Institute for Security.
Retrieved from http://www.cprdhorn.org/Gambella%20Report.pdf.
Meeker, M. 1989. The Pastoral Son and the Spirit of Patriarchy: Religion, Society and
Person among East Africa Stock Keepers. Wisconsin: University of Wisconsin Press.
Merera, G. 2003. Ethiopia: Competing Ethnic Nationalism and the Quest for
Democracy, 1960–2000. N.P.: Shaker Pub.
Merton, R. 1968. Social Theory and Social Structure. New York: The Free Press.
National Election Board of Ethiopia. 1995. ‘Election Law of Ethiopia 1995’
(Proclamation No. 111/1995). Retrieved from http://www.electionsethiopia.org.
Negash, T. and K. Tronvoll. 2000. Brothers at War: Making Sense of the Eritrean-
Ethiopian War. Oxford: Currey.
Nyaba, P. 2001. Politics of Liberation in South Sudan: an Insider’s View. Kampala:
Fountain Publishers.
Office of the Population and Housing Census Commission. 1984. ‘The Census
Result of the Illubabor Province’, in The 1984 Population and Housing Census.
Addis Ababa: Peoples Democratic Republic of Ethiopia, pp. 23–29.
Ogot, B. 1967. History of the Southern Luo. Vol. 1 Migration and Settlement
1500–1900. Nairobi: East African Publishing House.
Ojoch, J. 2003. ‘The Anywaa Need Peace and Happiness’. Retrieved from http://
www.gurtong.com.
Pal, T. 2006. ‘The Eastern Jikany Migration and the Evolution of Their Political
System’. Journal of the Maiut and Gambella Educational and Research Foundation
3: 12–23.
Pankhurst, A. 2002. ‘Surviving Resettlement in Wellega: the Qeto Experience’, in W.
James et al. (eds), Remapping Ethiopia: Socialism and After. Oxford: Currey, pp.
133–50.
Patrick, E. 2004. ‘The US Refugee Resettlement Program’, Migration Policy
Institute. Retrieved from http://www.migrationinformation.org/Feature.
Perner, C. 1994. The Anyuak – Living on Earth in the Sky. Vol. I. The Sphere of
Spirituality. Basel: Helbing and Lichtenhahn.
–––. 1997. The Anyuak – Living on Earth in the Sky. Vol. II. The Human Territory.
Basel: Helbing and Lichtenhahn.
Population Department. 2007. Ethiopian Population Images 2006. Addis Ababa:
Ministry of Finance and Economic Development.
Rabinow, P. 2003. Anthropos Today: Reflections on Modern Equipment. Princeton:
Princeton Press.
Redfield, R. 1955. ‘The Social Organization of Tradition’, The Far Eastern Quarterly
15(1): 13–21.
 References 229

Reh, M. 1996. Anywaa Language: Description and Internal Reconstructions. Köln:

Köppe Verlag.
Riang, Y. 2005. ‘The Gaat-Jak Nuer: One Nation, Two States’, Maiwut and
Gambella Education and Research Foundation 1(1).
Roosens, E. 1994. ‘The Primordial Nature of Origins in Migrant Ethnicity’, in H.
Vermeulen and C. Govers (eds), The Anthropology of Ethnicity: Beyond ‘Ethnic
Groups and Boundaries’. Amsterdam: Het Spinhuis, pp. 81–104.
Sahlins, M. 1961. ‘The Segmentary Lineage: An Organisation of Predatory
Expansion’, American Anthropologist 63(2): 322–45.
–––. 1963. ‘Poor Man, Rich Man, Big Man, Chief: Political Types in Melanesia and
Polynesia’, Comparative Studies in Society and History 5: 285–303.
–––. 1965. ‘On the Ideology and Composition of Descent Groups’, Man 65: 104–7.
–––. 1968. Tribesmen. Englewood Cliffs: Prentice-Hall.
Salamone, F. A. 1974. Gods and Goods in Africa: Persistence and Change in Ethnic and
Religious Identity in Yauri Emirate, North-Western State, Nigeria. New Haven:
Human Relations Area Files, Inc.
Schlee, G. 1994. ‘Ethnicity Emblems, Diacritical Features, Identity Markers: Some
East African Examples’, in D. Brokensha (ed.), A River of Blessings: Essays in
Honour of Paul Baxter. Syracuse: Syracuse University, pp. 129–43.
–––. 2002. Imagined Differences: Hatred and the Construction of Identity. Münster: Lit
Verlag.
–––. 2003. ‘Introduction: Identification in Violent Settings and Situations of Rapid
Change’, Africa 73(3): 333–42.
–––. 2004. ‘Taking Sides and Constructing Identities: Reflections of Conflict
Theory’, Journal of the Royal Anthropological Institute (NS) 10: 135–56.
–––. 2008. How Enemies Are Made: Towards a Theory of Ethnic and Religious
Conflicts. Oxford: Berghahn.
Scott, J. 1998. Seeing Like a State: How Certain Schemes to Improve the Human
Condition Have Failed. New Haven: Yale University Press.
Shils, E. 1957. ‘Primordial, Personal, Sacred and Civil Ties: Some Particular
Observations on the Relationships of Sociological Research and Theory’, British
Journal of Sociology 8(2): 130–45.
–––. 1995. ‘Nation, Nationality and Civil Society’, Nations and Nationalism 1(1):
93–118.
Shumet, S. 1986. ‘The Economic of Basis of Conflict among the Nuer and Anuak
Communities’, in The Proceedings of the 3rd Annual Seminar of the Department
of History. Addis Ababa: Addis Ababa University, pp. 34–49.
Soanes, C. and S. Hawker. 2005. ‘Statism’, Compact Oxford English Dictionary of
Current English, 3rd ed. Oxford: Oxford University Press.
Southall, A. 1976. ‘Nuer and Dinka Are People: Ecology, Ethnicity and Logical
Possibility’, Man (New Series) 11(4): 463–91.
–––. 1986. ‘The Illusion of Naath Agnation’, Ethnology 25(1): 1–20.
Stauder, J. 1971. The Majangir: Ecology and Society of a Southwest Ethiopian People.
Cambridge: Cambridge University Press.
230 References

Tadesse, B. 2002. ‘Modernist Dreams and Human Suffering: Villagization among

the Guji Oromo’, in W. James et al. (eds), Remapping Ethiopia: Socialism and
After. Oxford: Currey, pp. 116–32.
Tesema, T. 2006. The Political Economy of an African Society in Transformation: The
Case of Macca Oromo (Ethiopia). Asien und Afrika Studien 25 der Humboldt
Universität zu Berlin. Harrassowitz Verlag.
Teshale, T. 1995. The Making of Modern Ethiopia 1896–1974. Lawrenceville: The
Red Sea Press.
The Federal Democratic Republic of Ethiopia. 1995. ‘Constitution of the Federal
Democratic Republic of Ethiopia’. Retrieved from http: //www.elections ethiopia.org.
Tippet, A. 1970. People of Southwest Ethiopia. California: William Carey Library.
Vail, L. 1991. The Creation of Tribalism in Southern Africa. Berkeley: University of
California Press.
van den Berghe, P. 1981. The Ethnic Phenomenon. New York: Elsevier.
Varshney, A. 2002. Ethnic Conflict and Civic Life: Hindus and Muslims in India. New
Haven and London: Yale University Press.
Verdery, K. 1994. ‘Ethnicity, Nationalism and State-making’, in H. Vermeulen and
C. Govers (eds), Anthropology of Ethnicity: Beyond ‘Ethnic Groups and
Boundaries’. Amsterdam: Het Spinhuis, pp. 33–58.
Verdon, M. 1982. ‘When Have All the Lineages Gone? Cattle and Descent among
the Nuer’, American Anthropologist 84: 566–79.
Vestal, T. 1994. ‘Deficits of Democracy in the Transitional Government of Ethiopia
since 1991’, in H. Marcus (ed.), International Conference of Ethiopian Studies
Vol.2. Lawrencehill: Red Sea Press, pp. 188–204.
Wal, D. 1992. ‘The Nuer concept of Covenant and Convenantal Way of Life’,
Publius 22(4): 67–89.
Waters, A. 1997. ‘Conspiracy Theories as Ethno-Sociologies: Explanation and
Intention in African American Culture’, Journal of Black Studies 28(1): 112–15.
Wimmer, A. 2008. ‘The Making and Unmaking of Ethnic Boundaries: A
Multiethnic Process Theory’, American Journal of Sociology 113(49): 970–1022.
Wittgenstein, L. 1983. Philosophical Investigations, trans. G.E.M. Anscombe.
Oxford: Basil Blackwell.
Yeros, P. (ed.). 1999. Ethnicity and Nationalism in Africa: Constructivist Reflections
and Contemporary Politics. New York: St-Martin’s.
Young, C. (ed.). 1993. The Rising Tide of Cultural Pluralism: The Nation-State at Bay?
Madison: University of Wisconsin.
Young, J. 1999. ‘Along Ethiopia’s Western Frontier: Gambella and Benishangul in
Transition’, Journal of Modern African Studies 37(2): 321–46.
–––. 2003. ‘Sudan: Liberation Movements, Regional Armies, Ethnic Militias and
Peace’, Review of African Political Economy 30(97): 423–34.
–––. 2007a. ‘Local Conflict, Marginalisation, and the Threat to Regional Security’,
in Researching Local Conflicts and Regional Security. Institute for Security
Studies. Retrieved from http: //www.cprdhorn.org/Beja report/.
–––. 2007b. Armed Groups along Sudan’s Eastern Frontier: An Overview and Analysis.
Geneva: Small Arms Survey.
 References 231

Unpublished Sources
Abraham, S. 2002. ‘Breaking the Cycle of Conflict in Gambella Region. A Report of
an Assessment Mission’, Addis Ababa: UN Emergencies Unit for Ethiopia
(UNDP-EUE).
Agency for Cooperation and Research in Development – Ethiopia. 1999. ‘Pastoralist
Livelihood Programme, Livestock Research Report.’ Gambella: ACORD.
Bahru, Z. 1976. ‘Relations between Ethiopia and the Sudan on the Western Frontier,
1898–1935’, Ph.D. dissertation. London: University of London.
Birhanu, D. 1973. ‘A Historical Study of Trade in North-Eastern Illubabor and
South-Western Wellega (Ca. 1900–1935)’, B.A. dissertation. Addis Ababa:
Addis Ababa University.
Bureau of Planning and Economic Development. 2000. ‘Conservation Strategy of
the Gambella Region’. Gambella: Gambella Peoples National Regional State.
–––. ‘Situational Analysis for Gambella People’s National Regional State’. Gambella:
UNICEF.
Dereje, F. 2006c. ‘Framing Local Interests in Global Terms: The Identity Politics of
the Anywaa Diaspora in North America’, Paper given at the Martin Luther
University Halle-Wittenberg, Halle (Saale), Germany, 12 June 2006.
–––. 2007. ‘The Interface between National and Regional Level Decisions and Local
Conflicts in the Gambella Region’, Research Report Submitted to Pact/USAID,
Addis Ababa.
Duany, W. 1992. ‘Neither Places nor Prisons: The Constitution of Order among the
Nuer’, Ph.D. dissertation. Indiana: Indiana University.
–––. 1996. ‘Customary Law and Ways of Life in Transition among the Nuer of
South Sudan’, Workshop in Political Theory and Policy Analysis. Indiana: Indiana
University.
Ellman, A. 1972. ‘An Agricultural and Socio-Economic Survey of South Sudan
Refugee Settlements and Surrounding Areas in Gambella Awraja, Ethiopia’.
Addis Ababa: UNHCR.
Falge, C. 2006. ‘The Global Nuer: Modes of Transanational Livelihoods’, Ph.D. dis-
sertation. Halle (Saale), Germany: Martin Luther University Halle-Wittenberg.
Gambella Regional State, Bureau of Planning and Economic Development. 2000.
Jal, G. 1987. ‘The History of the Jikany Nuer before 1920’, Ph.D. dissertation.
London: University of London, History Department, School of Oriental and
African Studies.
Klinteberg, M. 1977. ‘Gambella Refugee: a Socio-economic Survey of Nilotes along
the Baro River in Western Ethiopia, with Special Emphasis on South Sudanese
Refugees’. Addis Ababa: UNHCR.
Kurimoto, E. 2000. ‘What Is Your Clan? The Significance of Patrilineal Descent in
Changing Anywaa (Anuak) Society’, 14th International Conference of Ethiopian
Studies. Addis Ababa.
Ojullu, O. 1987. ‘The Anywaa and Their Neighbours to c. 1970: An Essay on Local
Politics’, B.A. thesis. Addis Ababa: Addis Ababa University.
232 References

The Petroleum Operation Department. 2006. ‘Petroleum Exploration Potential of

Ethiopia’. Addis Ababa: Ministry of Mines and Energy.
Shandy, D. 2001. ‘Perils and Possibilities of Nuer Refugee Migration to the United
States’, Ph.D. dissertation. Columbia University.
‘The Nuer Invasion and Colonization of the Anuak Lands’. Anonymous and no date.
Vaughan, S. 2003. ‘Ethnicity and Power in Ethiopia’, Ph.D. dissertation. Edinburgh:
The University of Edinburgh.
Wolde Selassie Abute. 2002. ‘Gumuz and Highland Re-settlers: Differing Strategies
of Livelihood and Ethnic Relations in Metekel, North-western Ethiopia’, Ph.D.
dissertation. Göttingen, Germany: University of Göttingen.
Zerai, B. 1971. ‘Qegnazmach Majid’s, “The Legal Settlement of South-Western
Ethiopian Boundary (Gambella)”: Translated and Annotated’, B.A. thesis.
Addis Ababa: Addis Ababa University: Department of History.

Archival Materials
Gambella People National Regional State. 2002. (E.C). ‘The Inhuman Killing and
Destruction of Properties on 30 October 1994 E.C. in Itang Town and its
Surrounding Eight Kebeles’. Gambella Archive.
Gambella District Administration. 1973 (E.C). ‘Letter Written to Illubabor Province
Administrator Regarding the Security Situation in Jor and Gog Sub-district’.
Metu Archive.
–––. 1979 (E.C). ‘Letter Written to Illubabor Province Administrator Regarding the
SPLA Military Forces and Their Activities’. Metu Archive.
Governor of Gambella. 1951. ‘Letter Written by Qegnazmach Asfaw Abeje to
Captain H. R. Dibble, District Commissioner of Gambella’. Gore Archive.
Imperial Government of Ethiopia. 1957 (E.C). ‘A Mission Report on the Assessment
of Security Problems’. Addis Ababa: Ministry of Interior, 1–37. Metu Archive.
Ministry of Interior. 1964. ‘Letter Written by the Ministry of Interior to the Embassy of the
Democratic Republic of the Sudan’. Addis Ababa. September 1964. Metu Archive.
 Index

A Bär Jingmir, 70, 99–100, 136

abudho, 42 be hulet bila yibelalu, 202
Adongo, 24, 39, 41–43, 102, 122–23, BGPNRS, 4n5, 145
133n15, 178, 207 bia lac, 114
agem, 44, 56 bilo, 91
agnation, 61–64 Bilpham, 198
ajuwa, 46 Blood Exchange Pact, 116
Akobo, xi, 1, 3, 24, 39, 49, 70, 80–83, bouth, 134
86–87, 92, 96–98, 101, 103–4, British, the, 4, 43, 50, 87, 111,
116, 121–26, 134, 141, 152, 168, 121–29, 153, 162, 164–65, 171,
176, 179, 180n5, 181, 187, 173, 175, 180, 197, 206
195–96, 198, 202, 205, 206n15, buny, 6, 130–31, 170, 172–74, 176,
207–8 182, 201
alternative citizenship, 142, 153, 163 buny cie turuk, 168
American Presbyterian Church, the,
129, 164n9, 196 C
Amhara, 2, 4n5, 6nn6, 7, 106, 126, caa naath, 66
129, 134–35, 140, 145–46, 149, camun nak, 87
151, 166, 168–70 chirawiya, 109–11, 113, 190
amulo, 34 Cieng Reng, 25–27, 57, 59–60, 72, 86,
Anglo-Egyptian government, the, 197 88–91, 93, 107n8, 109, 141,
Anyanya, 194–96, 201, 204 172–75, 180, 185, 207
Anyanya-II, 198, 200 ci-Jwok, 46
Anywaa Ciro Anywaa, 86–87, 96
project of containment, 100, 107, confidential letter, the, 113, 116–18
147–60, 178, 190–91 conspiracy theory, 21, 112, 170
territoriality, 18, 21, 38–43, constructivism, 11–14, 16–17, 53, 62,
47–50, 59, 70, 78, 80–83, 74, 211
85–89, 92, 96, 113, 115–16, Cuai, 34
120, 125n8, 128 Jowatcuai, 34, 36, 39–41, 102–3
APDO, 159 cultural Darwinism, 207
awido, 34 cultural revolution, 50, 130–31, 134,
136, 140, 197
B
Banubdak, 84 D
bariya, 125, 127 Derg, 4, 130–42, 145, 149–50, 163,
Barth, Fredrik, 12–17, 95, 99, 119, 164n10, 179–80, 185n7, 197–
211–12 201, 204
bär, xi, 114, 176, 205 dhol, 69–70
234 Index

Dhol Koryom, 108 ethnicity, xii–iv, 4, 11–15, 17–22, 27,

dho-oto, 36 31, 50, 77, 99, 138, 141, 145, 153,
diaspora, 8, 23, 105n6, 115n12, 117, 160–66, 189–90, 205–6, 211–13
159, 164n9, 169, 202 ethnogenesis, 16, 31, 47–48, 50, 54,
dibuoc gilo, 32 150, 174
dil, 59, 65–66, 68–69, 73–74, 92, 106, EU, 122n2, 135, 173
112
Dimo, 32, 34 F
dimui, 37–38, 42, 97–98, 106, 109, family resemblances, 14, 31
126, 131–34, 163 France, 173
Dinka, 15n1, 16, 24, 32, 35, 37, 47,
54–57, 59–60, 66–68, 71–72, 74, G
78–80, 88n15, 92, 99–100, gaala, 6, 130–32
105–6, 111, 114, 116, 170, 172, Gaatgankir clan, 57
174–75, 183, 198, 200, 204–6, Gaat-Guang, 57, 72–73, 84, 86, 98,
209, 213 125n8
discursive modernity, 162 Gaat-Jak, 24–25, 57, 72–73, 84–86,
doctrine of mutual intervention, the, 88, 113–15, 122n3, 123–24,
193 125n8, 127, 129n11, 141–42,
Drang nach Osten, 81 148, 151, 171, 182, 185,
199–201, 203, 206
E Gaat-Jak/SPLA War, 200–1
ELF, 196, 199 Gaatnaar, 63
emic constructivism, 13, 17 gaatnyiet, 65–66, 69, 73
emic primordialism, 13, 31 gaattutni, 65
EPLF, 140, 146, 164n10, 199 gaatwac, 63
EPRDF, 4–6, 140, 142, 145–47, gar, 69–70, 80, 93, 95–99, 101, 103,
149–50, 154–59, 166–67, 106, 113, 169–71, 211, 213
183–87, 189, 199, 204, 209–10, Gee, 55–57, 59–60
215 Gaat-Gee, 56, 59
Ethiopian State, the, xi–iv, 3–4, 6–7, Geertz, C., 11–14, 31, 120, 129, 212
16, 18, 20–22, 31, 47, 49–50, 84, General Lemma, 126, 196
109, 115, 120–143, 145, 153–56, Germany, 23, 81n7, 173
159–60, 166, 168, 174–78, 182, Ghaak, 55–57, 60
189–90, 194, 196–98, 201, Gilo, 3, 32, 34–35, 39–40, 58, 80–81,
203–5, 209, 213, 215 86, 102–3, 121, 124, 125n8, 129,
ethnic 137, 178
boundary, xiv, 12–15, 36, 38, 48, GPDC, 157, 178n2
109, 211 GPDF, 157–58
conversion, xiii, 17, 24–25, 93, GPDM, 159
96–100, 103, 112, 118 GPLF, 159
federalism, xiii–iv, 4–6, 16, 21, GPNRS, 4–5, 83, 91, 145–49, 153–54,
50–51, 72, 89, 94, 143, 165–67, 169, 181, 185, 188,
145–91, 204 197n7, 204, 206, 209
 Index 235

great traditions, 103 Jwok, 16, 41, 44–48, 107–8, 161,

gungi, 43, 131 213–14

H K
Haile Selassie, 175–76, 186, 196 kew, 41, 101
hidden agenda, 101, 113–18, 153, 169, Kir, 40, 55–60, 72, 80, 174–75
189 Komo, 1–3, 6, 47, 82, 147–48, 176
Highlanders, xii, 2–3, 5–7, 22–23, Kong Diu, 25, 59, 88, 90, 173–74,
25–26, 47–50, 82, 100, 103, 113, 181, 185–86
119–20, 124–26, 129–31, 133, kume, xi, 171–74, 181–82
135n17, 137–39, 146–48, Kuok, 72
155–57, 159–61, 163, 166, 168, Kuoth, xi, 58, 93–94, 107–8, 177, 181,
171, 175–77, 182–84, 189, 195, 214
202, 215
L
I language games, 15, 18, 20, 215
Itang, 1, 3, 22–27, 40–41, 58–59, latent functions, xv, 25, 93
81–86, 88–92, 94, 107–9, 124, Latjor Dingyian, 59
126, 131, 141, 149–50, 158, lineage system, 62–63, 78
168–69, 174, 177, 180, 182–83, Lou Nuer, 70, 78, 87, 175, 207–8
187–88, 195, 198–99, 202 Lull, 39–40, 177
Luo, 31n1, 32, 84, 87, 97, 104–5
J luo, 44–45, 48, 91, 127, 213
Jalab, 173 Jur-Luo, 34, 72
Jal tang, 68 luuch naath, 100, 110
jang, 65–66, 99 Lwoo, 16, 31–33, 35, 44, 47–48, 57,
Jikany Nuer, 34–35, 56, 58, 60–61, 104–6, 150, 213
71–73, 78, 108, 122–23, 151, Lwoo migration, 32–33, 35, 44, 48
170, 174, 206–7
Jikaw, xi, 1, 3, 23–24, 67, 81–83, 86, M
94, 100n3, 104, 106, 108, 110, Majangir, 1–3, 6, 47, 82, 146, 148–51,
115, 123, 128, 131, 133, 135, 157, 159
136n18, 140n27, 141, 142n33, Makot village, 23–27, 58–59, 89–90,
150–52, 159, 168, 171, 179, 181, 107, 109, 141, 174, 181, 185
183, 185, 187, 195, 200–2, manifest functions, 93
206n15, 207 massacre on 13 December, 156–57,
ji-kwaari, 38 159, 166, 184, 210
ji-nyiye, 42 math, 36
Jingmir, 24, 70, 96–101, 103, 136 Medho, 44
jinubni, 183 Mekaneyesus Church, 24, 70
jiop naath, 183 Merton, Robert, 93, 113, 169
jobur, 39–40, 112 mission modernity, 163
jur, 36, 106 modernity, 160–64, 168–69, 189–91,
Jur-Luo, 34, 72 213
236 Index

MPDO, 159 P
Mut Wiu, 56, 58, 108, 175 paak, 36
padhano, 45–46
N Padiel Gakgak, 79
naak, 46, 114, 131 patrilineal descent, 36, 54, 61–64,
Naath, 35, 55, 63, 113 68–69, 72
caa naath, 66 piny kwari, 39
luuch naath, 100, 110 Pochalla, 23–24, 102, 115n12, 133,
naath cieng, 55 140n28, 159, 199n5, 209
Nasser, xi, 96, 104, 116, 150, 178, 204 postcolonial Sudan, 193
Naturalizers, 11, 13 primordialism, 11–14, 16–17, 31, 35,
nei ti naath, 60 41, 46–51, 72, 74, 212
NGO, 103, 153, 165, 194, 201–2 primordial modernity, 162
Ngok, 56, 58, 114 protonationalism, 160–66
ngok, 80
Ngok Dinka, 56, 68 R
Nilotes, 16, 31, 32n2, 47–48, 100, ran mi jaang, 59
104, 168, 177 ran mi ran, 55, 59–60
Nok, 80 red people, 100, 113, 115, 156, 183
NPDO, 73, 159 refugization, 201
Nuer cultural hegemony, 100–13 resettlement programme, the, 100, 130,
Nuer expansion(ism), xiv, 20–22, 137–39, 141, 164, 202
77–79, 86, 93, 104–6, 109, Riek Machar, 155, 170, 204, 206–7,
112–14, 118, 141, 152, 169, 207 209
nyamdoar, 85 rul, 65–68, 71, 73, 100, 110, 213
nyidungu, 44
Nyikang, 32, 34 S
nyingalabuo, 44–45 second class citizens, 154, 204
nyinya, 42–43, 102, 160, 178n2 shifta, 122n5, 159, 195–96
nyiya, 24, 38, 42–43, 47, 87, 102–3, Shilluk, 32, 34, 47, 104–7
123–24, 133, 178n2, 197n6, 207 small traditions, 103
SNNPRS, 4n5, 145
O sociobiologists, 11, 17
Ochom village, 71, 92, 111, 115, 172 sovereignty, 4, 103, 122, 129, 145,
OLF, 140, 146, 199 153–54, 185, 187, 197, 204
Openo, 39–40, 80, 158 SPLA, 87, 105, 140–41, 153, 170,
Openo Anywaa, 24, 39–41, 49, 183, 198–201, 204–6, 208–10
84, 88, 114, 125, 151–52, 177 SSIM, 87, 204–5, 208
Opo, 1–3, 6, 47, 70, 82, 104, 146, 148 statism, 160, 164–66
Oromo, 2, 6n6, 32n2, 47, 100n2, 106, Sudanese Anywaa, 28, 104–5, 107,
126–27, 129, 135, 140, 146, 149, 131, 133, 187
151, 166–69, 199 Sudanese civil wars, xiv, 2, 7–8, 74, 87,
Othieno, 34, 45n10 194, 215–16
 Index 237

Sudanese Nuer, 20, 128, 154–55, 169, W

196–97 Walo, 85
Sudanese state, the, 167, 193–94, 204, walo, 42
209 Watnaadhi, 34
wat-ngomi, 39
T Wechdeng, 23–24, 27, 66–69, 73–74,
thok dwiel, 62, 64–66, 68 84–85
Tiek, 56–57 welo, 39–40, 66, 110, 112
Tier Agak, 94, 113–15, 123, 152 Wittgenstein, 14–15, 18, 31, 212
Tigreans, 2, 103, 145–46, 154, 157, wut, 69, 71, 170–71
160, 166, 168
TPLF, 6n7, 140, 149, 157, 160, 199 X
tuach kuach, 56 xenophobic, 50, 111, 214
tung, 36, 42–43
Turks, 168 Y
turuk mi thil kade, 168 yien, 65
tut kernyang, 175, 177
Z
U zemach, 131
uchuok, 38n6, 42
Ucuudho, 102–3
Ukuna massacre, 156
UNHCR, 8, 140n27, 164, 197, 201–3
utak, 98, 131