. 1
( 11)


This page intentionally left blank
Cultural Contestation in Ethnic Con¬‚ict

Ethnic con¬‚ict often focuses on culturally charged symbols and rituals that
evoke strong emotions from all sides. Marc Howard Ross examines battles
over diverse cultural expressions and enactments, including Islamic head-
scarves in France, parades in Northern Ireland, holy sites in Jerusalem and
Confederate ¬‚ags in the American South, to propose a psychocultural fra-
mework for understanding ethnic con¬‚ict, as well as barriers to, and oppor-
tunities for, its mitigation. His analysis explores how culture frames interests,
structures demand-making and shapes how opponents can ¬nd common
ground to produce constructive outcomes to long-term disputes. He focuses
on participants™ accounts of con¬‚ict to identify emotionally signi¬cant issues,
and the power of cultural expressions to link individuals to larger identities
and shape action. Ross shows that, contrary to popular belief, culture does not
necessarily exacerbate con¬‚ict; rather, the constructed nature of psycho-
cultural narratives can facilitate successful con¬‚ict mitigation through the
development of more inclusive narratives and identities.

Marc Howard Ross is William Rand Kenan Jr. Professor of Political Science
at Bryn Mawr College where he has taught since 1968. He has had a long
term interest in social science theories of con¬‚ict management and has done
research in East Africa, France, Northern Ireland, the Middle East, Spain,
South Africa, and the United States. Professor Ross has written or edited six
books including The Culture of Con¬‚ict (1993) and The Management of Con¬‚ict
Cambridge Studies in Comparative Politics
General Editor
Margaret Levi, University of Washington, Seattle

Assistant General Editor
Stephen Hanson, University of Washington, Seattle

Associate Editors
Robert H. Bates, Harvard University
Helen Milner, Princeton University
Frances Rosenbluth, Yale University
Susan Stokes, Yale University
Sidney Tarrow, Cornell University
Kathleen Thelen, Northwestern University
Erik Wibbels, University of Washington, Seattle

Other Books in the Series
Lisa Baldez, Why Women Protest? Women™s Movements in Chile
Stefano Bartolini, The Political Mobilization of the European Left,
1860“1980: The Class Cleavage
Mark Beissinger, Nationalist Mobilization and the Collapse of the Soviet
Nancy Bermeo, ed., Unemployment in the New Europe
Carles Boix, Democracy and Redistribution
Carles Boix, Political Parties, Growth, and Equality: Conservation
and Social Democratic Economic Strategies in the World Economy
Catherine Boone, Merchant Capital and the Roots of State Power
in Senegal, 1930“1985
Catherine Boone, Political Topographies of the African State: Territorial
Authority and Institutional Change
Michael Bratton and Nicolas van de Walle, Democratic Experiments
in Africa: Regime Transitions in Comparative Perspective

(Continues after the index)
Cultural Contestation in
Ethnic Con¬‚ict

Bryn Mawr College
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
Information on this title: www.cambridge.org/9780521870139

© Marc Howard Ross 2007

This publication is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published in print format 2007

eBook (EBL)
ISBN-13 978-0-511-28467-0
ISBN-10 0-511-28467-5 eBook (EBL)

ISBN-13 978-0-521-87013-9
ISBN-10 0-521-87013-5

ISBN-13 978-0-521-69032-4
ISBN-10 0-521-69032-3

Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
To Katherine with love
for all the joys we share including
Katherine, Thomas
and the future members of their generation

page x
List of ¬gures

1. Introduction: easy questions and hard answers, what are
they ¬ghting about? 1
2. The political psychology of competing narratives 30
3. Narratives and performance: ritual enactment and
psychocultural dramas in ethnic con¬‚ict 63
4. Loyalist parades in Northern Ireland as recurring
psychocultural dramas 88
5. Where is Barcelona? Imagining the nation without a state 127
6. Digging up the past to contest the present:
politics and archeology in Jerusalem™s Old City 154
7. Dressed to express: Islamic headscarves in French schools 191
8. The politics of memory and memorialization in
post-apartheid South Africa 224
9. Enlarging South Africa™s symbolic landscape 251
10. Flags, heroes, and statues: inclusive versus exclusive
identity markers in the American South 280
11. Culture™s central role in ethnic con¬‚ict 312


4.1 Map of the United Kingdom and Ireland (www.
lib.utexas.edu/maps/europe/united_kingdom.html) page 93
4.2 Selected divergence in Protestant and Catholic
narratives in Northern Ireland 100
4.3 Two views of 1916. Photograph by author 102
4.4 Timeline of Political events and Parading in
Northern Ireland 105
5.1 Spanish nationalities and regions. From
˜The Federalization of Spain™ by Luis Moreno,
Copyright ª 2001 Frank Cass Publishers.
Reproduced by permission of Taylor and
Francis Books UK 137
6.1 Haram al Sharif and Western Wall. Photograph by
author 155
6.2 Map of Jerusalem™s Old City 157
6.3 Timeline of the Israeli“Palestinian con¬‚ict 171
6.4 The Third Temple astride the Temple Mount in
Jerusalem 185
Headscarf, a threat to the Republic? ª Thierry
Orban / Corbis 193
8.1 South Africa with trek routes 238
8.2 Voortrekker Monument, Pretoria.
Photograph by author 242
8.3 The Voortrecker Monument, Frieze. Photograph
by author 243
9.1 Hector Pieterson Museum. Photograph Picture
provided courtesy of www.soweto.co.za 253

9.2 Blood River/Ncome. Photographs by author 262
9.3 Cairn at Robben Island. Photograph by author 266
9.4 Street signs, District Six Museum. Photograph
by author 272
9.5 Map, District Six Museum. Photograph by author 273
9.6 Memory Cloth, District Six Museum. Photograph
by author 274
10.1 Square and rectangular Confederate battle ¬‚ags.
Square ¬‚ag provided courtesy of the Minnesota
Historical Society and rectangular ¬‚ag provided
courtesy of the National Park Service,
www.nps.gov/vick/visctr/sitebltn/bnnrglry.htm. 287
10.2 Of¬cial Confederate ¬‚ags. Provided courtesy of
www.netstate.com. 288
10.3 Georgia State ¬‚ags. Provided courtesy of
www.netstate.com. 290
10.4 Statue of Abraham Lincoln in Richmond, Virginia.
Photograph by author 302

Every effort has been made to trace all the copyright holders, but if any
have been inadvertently overlooked, Cambridge University Press will be
pleased to make the necessary acknowledgments at the ¬rst opportunity.


A decade ago I began puzzling about why, and how, what to some people
are innocent cultural expressions are to others provocative, aggressive,
politically signi¬cant acts. At one level, I had known this was the case for a
long time from personal observations and experiences as well as the
analyses of scholars such as Murray Edelman and Abner Cohen, both of
whom, in different ways, focused my attention on the political uses of
culture. My own investigation of culture and politics started thirty years
ago when I began a cross-cultural study of con¬‚ict that examined dif-
ferences between high- and low-con¬‚ict societies. This project led me to
articulate ideas concerning the complementary roles of structural and
psychocultural mechanisms that create societal dispositions toward par-
ticular forms and levels of con¬‚ict and violence. Next I utilized the same
framework to explore how any given theory of con¬‚ict has crucial
implications for the theory and practice of con¬‚ict management. For
example, if a con¬‚ict is viewed as one over resource competition, people
trying to end it will seek to negotiate an agreement to divide the resources
in a manner that all sides can accept, while those who attribute the same
con¬‚ict to incompatible identities will make bridging these differences
central to their con¬‚ict management efforts.
As part of my work on con¬‚ict management, I asked why some con-
¬‚icts are managed more successfully than others, and a case I investigated
in depth was the 1989 con¬‚ict in France that arose when three Muslim
junior high students were expelled for wearing headscarves in school
(Ross 1993b). I was living in France with my family at the time and was
astounded at how quickly the con¬‚ict expanded, at the intensity of
emotion it generated, and at how inadequate the outcome was as it failed


to address the deep identity needs of those involved. Several years later
I began studying Protestant Loyal Order parades in Northern Ireland
where sometimes thousands of police and army troops are mobilized to
separate Protestant marchers and Catholic protesters.
Once I began looking at, and talking about, intense con¬‚ict involving
expressive culture, more and more examples that seemed especially rele-
vant in long-term ethnic con¬‚icts became apparent, or were recounted to
me. In these con¬‚icts, while there are always substantive, tangible issues
dividing the parties, it is also the case that cultural assertion of exclusive
identities invariably contributes signi¬cantly to heightened tensions,
intransigent positions and, sometimes, to violence. As a result, con¬‚ict
often persisted over what seems to outsiders, but not to the parties
themselves, to be the most trivial of differences.
At some point, I realized how much I didn™t understand about the
psychocultural dynamics of con¬‚ict expansion in enduring con¬‚icts, and
why and how small incidents engage so many people so passionately, why
these con¬‚icts are so resistant to resolution, and decided that there was a
need for concepts to better analyze them, as well as for hypotheses about
why some cultural con¬‚icts take a more constructive turn than others.
The result is this book, a long answer to a short question that political
scientists have struggled to explain: why are many ethnic con¬‚icts so
intense and so hard to settle? The most common answer to this question
is that clashing interests over tangible resources such as land, jobs, or
control of the state, create zero-sum con¬‚icts that endure when the
parties believe losing them can have disastrous consequences. As a result,
institutions are unable to provide adequate security guarantees and the
parties are unable to overcome commitment problems to ¬nd ways to
share power and deescalate con¬‚ict. While such answers are useful, I also
¬nd them incomplete. What they lack is thoughtful consideration of
where interests come from in the ¬rst place, how interests get de¬ned in
speci¬c cultural contests, and the ways that culture structures appropriate
ways to pursue them. The theories and language of interest and institu-
tional analyses that are the bases for most political analyses make too little
space for identity and culture and pay too little attention to how they
directly affect con¬‚ict.
My starting point is different “ namely that rich analyses of the politics
of identity and culture provide an explanatory power that cannot simply
be incorporated into a rationalist framework. The analysis of identity
politics requires other tools. We need to investigate the power of culture

to order priorities, de¬ne enemies and allies, and offer meaning to large
numbers of people under stress, threat, and uncertainty. In this book I
emphasize the importance of psychocultural narratives and dramas and
the power of cultural expressions and enactments to shape how people
understand their group and its interests, to promote particular actions,
and to create a society™s symbolic landscape. Widely shared narratives
matter because they offer emotionally meaningful accounts of the world,
de¬ning groups and explaining their motives and actions. Because they
frame the world and shape action, powerful narratives must be central to
the analysis of ethnic con¬‚ict and steps taken to mitigate it. Finally,
politically relevant psychocultural analysis not only examines group nar-
ratives, but also considers the many ways in which they are enacted in
daily life and in a community™s sacred rituals. Enactment matters because
participation af¬rms core elements of a narrative and strengthens
attachment to it and to the group while linking present con¬‚icts and
people across time and space.
I emphasize identities and psychocultural dynamics because political
scientists have paid too little attention to them; I believe they are central
to analyzing core political questions about con¬‚ict, authority, and com-
munity. Our theories of con¬‚ict, and in particular its management, would
be richer if we expanded our understanding of con¬‚ict and articulated
more effectively how interests and identities interact and shape each other
rather than the position that one is necessarily determinant of the other.
Many con¬‚ict resolution practitioners understand this complementary
relationship intuitively, and have integrated this into their practice. For
them this book may be less relevant in terms of making explicit the role of
culture in ethnic con¬‚ict than in spelling out the theoretical basis for what
they already do, and providing some comparative examples.
This project involved ¬eld research in six countries and inspiration
during visits to at least ¬ve more. As the project evolved, I sharpened and
limited the cases I would examine and what I needed to know about each.
Having engaged primarily in quantitative comparative research in the
past, the methodological issues of what constituted not only evidence but
comparable data in the qualitative case studies was not always easy to
conceptualize let alone collect.
Two of the cases I included pressed me to separate my preconceptions
from my analysis, and to be particularly rigorous in applying the tools the
book offers. The ¬rst was the case of the Confederate ¬‚ag controversies in
the United States. When I began the research I didn™t plan to include race

in the United States as one of my cases but I changed my mind for three
reasons. One was a comment from political psychologist Dan Bar-On
who pushed me to focus attention on US racial con¬‚ict and not just think
about ethnic con¬‚ict in other societies. Bringing the tools of my analysis
back home would, he argued, perhaps provide important insights
and would engage me differently than the study of other societies had.
Second, my wife Katherine not only agreed with Bar-On™s points but
added that including my own society (and eventually my own city) along
with others in my project would be the best way to communicate to
readers that the problems of ethnic and racial con¬‚icts are not just found
in far-away lands. Third, for some time I had followed the Confederate
¬‚ag con¬‚icts in the South and realized that these were quite comparable
to the cultural contestation I had been examining elsewhere.
The second case that challenged my objectivity and my capacity to
separate my own preconceptions and emotions from the analysis was the
work I did in France, where I have lived a great deal in the past thirty-¬ve
years. To me, the argument that French culture is at risk from head-
scarves in schools did not seem plausible. I saw no threat in ten, twenty, or
even a thousand young girls wearing a scarf on their head in school, and
I was impatient with the lack of respect that many in power have
demonstrated, and the refusal of many French of¬cials to think through
their positions in constructive ways. However, as with the other cases, the
point is not which side is ˜˜right™™ and which one is ˜˜wrong.™™ My reaction
to the French narratives was no better than telling someone their feelings
are wrong or than an analyst telling a patient they had a stupid dream.
Dismissal of passionately held positions does not help us understand why
people feel as they do and what the signi¬cance of these strong feelings is
for political and social action; neither does it suggest useful steps to
address the con¬‚ict constructively.
What I found especially dif¬cult in studying con¬‚icts in the societies
I know best was listening actively to all the parties “ a crucial ¬rst step in
my methodological approach, which is based on my belief that you don™t
have to accept someone™s position, but you do need to be clear what it is
and then ask why they hold it so passionately. This is clear to me in
theory, but, for example, given my own upbringing and experiences, when
the object was the Confederate battle ¬‚ag, it was not easy for me to see it
as anything but a symbol of racial oppression. Explanations that it
represented heritage and alternative constitutional principles (at least for
some people) ¬rst struck me as false consciousness, if not deception. To

do the research and analysis effectively, I had to set aside these prior, and
deeply held, personal positions to take seriously the question of why
attachment to the ¬‚ag is so strong, and to consider ways in which heritage
primarily concerns the present rather than the past. Because this con¬‚ict
was close to home and so connected to my life experience and values I had
to be rigorous in applying the method that had been so much easier to use
in more unfamiliar cultures in order to understand that for blacks as well
as whites the heritage issues related to the Confederate battle ¬‚ag are
about recognition and loss as well as race, and that I owed them the same
acknowledgment, not just rejection, that I urge on theorists and practi-
tioners working on con¬‚ict in other cultures. My experience with the
French case was similar, if less intense.
Let me close by saying that ¬nishing this book has been very dif¬cult
because none of the cases I have examined sit still. Since the completion of
early drafts, one of the cases has heated up considerably, while some that
had been volatile have become more contained. I note this to make the
point that I am not in the business of making predictions but rather of
trying to understand both escalating con¬‚ict and the potential for con¬‚ict
mitigation, and of identifying tools that help theorists to analyze ethnic
con¬‚ict and practitioners to move toward ˜˜good enough™™ solutions.
What I hope readers will take away from this book is a sense that even
when ethnic con¬‚ict is intense, or when groups that appeared to have
found a way to coexist return to violence, we should not simply wring our
hands and believe that ancient hatreds make ongoing violent con¬‚ict
Undertaking this research and writing required help from more than a
few people and several grants that paid my research expenses, providing
time to read, think and write as I explored new theoretical problems and
literatures. A grant from the United States Institute for Peace got me
started and I hope that the people who awarded it to me will recognize the
core questions I promised to examine in this book despite the many
changes in cases and concepts. A generous grant from the Mellon
Foundation™s New Directions Program and regular sabbatical and
research support from Bryn Mawr College provided the time and
resources needed to do ¬eld work in six different countries and the time
needed for writing. Students in my Culture and Ethnic Con¬‚ict Man-
agement class that I have taught since 1999 continually obliged me to
sharpen my ideas and to make explicit the connection between the case
studies and my larger argument.

I bene¬ted from comments I received at professional meetings where I
presented parts of this work and at seminar presentations at Harvard,
Hebrew University, INCORE, Ohio State™s Mershon Center, Syracuse,
George Mason, the University of Pennsylvania, Complutense University
of Madrid, the Peace and Con¬‚ict Studies at the University of Granada,
and at the University of the Western Cape in South Africa.
I am deeply grateful to Kevin Avruch, Stuart Kaufman, and Lou
Kriesberg who commented on the entire manuscript, making a number of
very detailed and helpful suggestions as I struggled through the problem
of making connections among the cases in a theoretically useful way.
Many other people in many places have talked and corresponded with me
about parts of this project and I want to thank them for their time and
comments: Miguel Rodrigo Alsina, Eileen Babbitt, Gabriel Barkay, Dan
Bar-On, Yaacov Bar Siman-Tov, Dani Bar-Tal, Zvi Beckerman, Meir
Ben-Dov, Philip Bonner, Dominic Bryan, David Bunn, Ariane Chebel
d™Apollonia, John Coski, John Darby, Lionel Davis, Fannie du Toit, Roy
Eidelson, Mari Fitzduff, Tanya Gallagher, Mathias Gardet, Harvey
Glickman, Gershom Gorenberg, Deborah Harrold, Ron Hassner, Neil
Jarman, Riva Kastoryano, Herb Kelman, Cynthia Kros, Cecelia Kruger,
Yehezekial Landau, Ned Lebow, Ed Linenthal, Ian Lustick, Jannie
Malan, Charles Malcolm, Sabine Marschall, Clark McCauley, Siobhan
McEvoy-Levy, Hlengiwe Mkhize, Ifat Moaz, Rob Mortimer, Bob Mulvihill,
Dorothy Noyes, Brendan O™Leary, Gert Operman, Ciraj Rassool, Nadim
Rouhanna, Paul Rozin, Hal Saunders, Stuart Saunders, Astrid Schwenke,
Sandra Scham, Lee Smithey, Robin Wagner-Paci¬ci, Catherine Withol
de Wenden, Leslie Witz, and Alan Zuckerman. Many others made useful
suggestions for readings and gave me ideas about directions to pursue.
Lastly, I want to thank my family, and especially Katherine, for all the
help and support they provided me while I worked on this project, lis-
tening to my stories, reactions to accounts of what I was reading, and
always pushing me to think more clearly about what I was doing. Kim and
Warren talked to me about Southern culture, the Civil War and
reenactors; Aaron kept giving me new websites where I could ¬nd
material about the latest con¬‚ict from a non-American perspective;
Kristin told me that if I wanted just one more case it should be Rwanda,
preferably when she was working there; and Ethan listened, nodded, and
would ask questions that really made me think hard about what I was
doing. Katherine visited each of the countries in which I did research, and
read and commented on more drafts of more chapters than I can

remember having written. What is more incredible to me is that she never
stopped giving me her reactions, telling me what she thought needed
changing, highlighting, recasting and always doing it with care and love.
No one can be luckier than I have been to have had such a terri¬c partner
with whom to share a project. I can never tell her enough how meaningful
this has been to me, but at some deep level I think she knows. For that I
am incredibly grateful.


Introduction: easy questions and hard answers,
what are they ¬ghting about?

What are long-term ethnic con¬‚icts about? How do they develop? Why
are they so intense and hard to settle? Why do opposing sides view and
describe what are ostensibly the same events so differently? How does
identity shape why and how ethnic con¬‚ict is waged? What do good
settlements look like?
Over the past thirty years, political analyses have offered very diverse
answers to these apparently straightforward questions. In general, poli-
tical scientists approaching ethnic con¬‚ict have focused on the interests
motivating contending groups and the strategies by which these interests
are pursued. Some answers from this perspective are interesting and non-
obvious. On the whole, however, they are partial, and fail to address some
important issues, thereby limiting our understanding of ethnic con¬‚ict and
its management. For example, most existing work has little to say about
how interests are developed and de¬ned in different societies. In addition,
there is little effort to deal with the puzzles that arise when what are
apparently the same competing interests in two different settings result in
intense con¬‚ict and violence in one but not the other. Often, interest-
based accounts cannot explain why hypothesized preconditions for
intense con¬‚ict, such as ethnic group inequalities, produce high con¬‚ict
in some places but very little in others. Nor do they help us understand
why some societies with relatively little intergroup inequality, such as
Northern Ireland, have a great deal of con¬‚ict and violence, while others
with high inequalities, such as post-apartheid South Africa, have far less
intergroup con¬‚ict than many expected.
What is missing from many rationalist political analyses is attention to
group identity and the role it plays in ethnic con¬‚ict. Group identity is a
collective process that connects individuals to groups and de¬nes shared

Cultural Contestation in Ethnic Conflict

worldviews and interests (Northrup 1989). It is tied to culture and cul-
tural expressions that mark groups as distinct from each other. Identities
are frequently articulated through, and contested around, collective
memories and mundane, everyday cultural practices such as parades, ¬‚ag
displays, language, clothing, religious practices, and public monuments
that symbolically connect the past and present and are visible in a region™s
symbolic landscape. Mundane practices that represent one group to its
members become polarizing when their expression is felt as a threat by a
second group, and/or when attempts to limit the practices are perceived as
a threat by the group performing them.
Before I go further, a brief mention of how I am using a few key terms
is in order. By culture I refer to the shared system of meaning that people
use to make sense of the world (Geertz 1973a; Ross 1997; 2002). Culture
is expressed in a wide variety of symbolic forms, some highly formalized
(e.g., religious and national rituals), others less formal but widespread
(e.g. language, clothing, food, games). Sometimes culture is expressed in
physical forms that de¬ne the symbolic landscape such as monuments,
murals, or banners or at sacred sites; some of these are natural like rivers
or mountains; other forms are human constructions such as holy places or
battleground memorials. Attention to symbols, rituals, and the narratives
that members of a group use to make sense of the world is key to
understanding how culture shapes their lives and their collective beha-
viors. I have sought a single phrase that encompasses the many different
forms of cultural con¬‚ict of interest here and often the term ˜˜cultural
expression™™ is appropriate but sometimes I use the terms ˜˜cultural per-
formance™™ and ˜˜cultural enactment.™™ What they share is that each refers
to contextually signi¬cant activities, objects, and/or symbols that have
strong emotional meaning and become focal points of intergroup con¬‚ict.
Analyzing the dynamics of these con¬‚icts and their settlement can
provide us with useful insights about the roots of ethnic con¬‚ict and its
To examine contestation over cultural expressions and performance,
I extend my earlier work to develop and utilize the concepts of psycho-
cultural narratives and dramas; those help offer a more complete under-
standing of ethnic con¬‚ict than an interest-base approach alone can
provide. The analysis links collective psychological and social processes,
placing identity issues and their cultural enactments at the center of
ethnic con¬‚ict. It examines cultural expressions in ethnic con¬‚ict as
markers of divisive identity and mutually exclusive positions. At the same

time, the socially and contextually constructed nature of cultural
expressions and identities, and the rede¬nition of, or changes in, the
meaning of cultural narratives, offer opportunities for con¬‚ict mitigation
between former opponents, and allow them to develop a greater sense of
interdependence and mutually bene¬cial cooperative relations.
Cultural expressions are not just surface phenomena. They are
re¬‚ectors of groups™ worldviews and on-going con¬‚ict that can help us
better comprehend what a group™s deepest hopes and fears are, how it
understands an opponent™s actions and motives, and what a good enough
agreement with an adversary would provide. Cultural expressions play a
causal role in con¬‚ict, when they make certain action possibilities more
plausible, and therefore more probable, than others as they direct col-
lective understandings of the motives, interests, and behaviors of the in-
group and of opponents. In addition, cultural expressions serve as
exacerbaters or inhibiters of con¬‚ict. Cultural expressions and the narratives
associated with them communicate a worldview that ranges from highly
exclusive to highly inclusive. The more that exclusivity and mutual
incompatibility are expressed, the harder it is for opponents to alter their
relationship; conversely, the more that cultural expressions are, or
become, inclusive, the more likely it is that the parties can deal success-
fully with differences.
Cultural identities, from this perspective, are both barriers to, and
opportunities for, the mitigation of ethnic con¬‚ict. The argument devel-
oped here is that movement toward constructive con¬‚ict management in
long-term intergroup con¬‚icts is facilitated through the development of
inclusive narratives, symbols, rituals, and other cultural expressions in
contexts where mutually exclusive claims previously predominated. Signed
agreements between long-standing opponents, such as Protestants and
Catholics in Northern Ireland, are only one step in a peace process.
A cultural perspective obliges us to go beyond formal agreements to
recognize ritual and symbol as crucial to the implementation of agree-
ments for peacemaking and peacebuilding. Before opposing parties can
come to the table to renegotiate their incompatible interests and change
their behaviors and relationship, there often needs to be bridging in the
form of inclusive cultural expressions that link formerly opposing com-
munities or rede¬ne older rituals to be less threatening and exclusive.
Cultural expressions that become the focal points in ethnic con¬‚icts
take many forms; the chapters that follow offer extended cases that
include contested issues of parades, festivals, language, archeology, and
Cultural Contestation in Ethnic Conflict

holy sites, ¬‚ags, monuments, museums, and clothing from Northern
´bec, Jerusalem, India, the US, South
Ireland, England, Catalonia, Que
Africa, and France. The cases range from ones such as the Israeli“
Palestinian con¬‚ict in which violence has been high to those such as
Quebec or France where it is low; in addition, they vary in the extent
to which the con¬‚ict is currently intense and bitter, such as Northern
Ireland, to those in which it has moved in a more constructive direction,
such as Catalonia.
The roots of this inquiry lie in an earlier study I conducted on cross-
cultural differences in con¬‚ict and con¬‚ict management in 90 pre-
industrial societies (Ross 1993a; 1993b). That analysis showed, ¬rst, how
both structurally rooted interests and psychoculturally based identities
independently explain a society™s level and targets of con¬‚ict and violence,
and second, that both also matter in con¬‚ict mitigation. Despite being a
political scientist I became particularly interested in the psychocultural
side of con¬‚ict and its management and argued in my conclusions that
interpretations are central to con¬‚ict behavior because con¬‚ict evokes
deep-seated emotions in situations that are highly ambiguous and often
unstructured. The combination of emotion and ambiguity readily pro-
duces psychic threat, leading to regression with a return to earlier
experiences, and shapes how participants react to a con¬‚ict. Such inter-
pretations are cultural, not just personal, when they are nurtured and
socially reinforced, linking individuals in a collective process (1993b: 192).
I hypothesized that especially in long-term intractable con¬‚icts a pre-
requisite to constructive con¬‚ict management is modifying competing
psychocultural interpretations or narratives so that the parties in con¬‚ict
come to believe that there are people on the other side with whom they
can negotiate, and issues that are negotiable. After completing the cross-
cultural study, I began asking myself why in so many ethnic con¬‚icts
expressive practices and sacred places produce intense disputes that out-
siders quickly dismiss as irrational, and how a better understanding of this
phenomenon could help us manage these con¬‚icts more effectively. This
volume brings together my answers to these questions, placing at center
stage the competing accounts of participants in con¬‚ict. I ask little about
whether or not they are ˜˜true™™ and a lot about how they shape beliefs and
behaviors about one™s own side, an opponent, and what constitutes
appropriate and inappropriate action.
The goal of this book is to offer an alternative way to think
about successful and unsuccessful ethnic con¬‚ict mitigation to enrich

more well-known structural and interest-based approaches. It is intended
to help people studying ethnic con¬‚ict make better sense of it and to aid
participants and third parties seeking constructive outcomes. The key
points I emphasize include: taking seriously participants™ own accounts to
identify emotionally signi¬cant elements that must be part of any set-
tlement; making sense of why and how the narratives are emotionally
powerful; examining how the narratives shape beliefs that facilitate the
choice of some actions over others; analyzing the power of collective
memories in linking individuals to larger social and political identities;
emphasizing the widespread use of imaginative and politically effective
culturally grounded expressions and enactments to make claims, build
commitment, and mobilize action; considering the ways that political
actions shape identity, culture, and interests; identifying psychocultural
narratives about peoplehood; and recognizing how the constructed nature
of narratives makes possible successful con¬‚ict mitigation. To begin, the
next section of this chapter introduces two con¬‚icts that are treated in
greater length later: parade con¬‚icts in Northern Ireland, and language
con¬‚ict between Catalonia and Spain. The subsequent section outlines
core questions in the psychocultural analysis of ethnic con¬‚ict and speci¬-
cation of some limits to the cultural analysis of con¬‚ict; and the ¬nal
section explains my use of the concepts of culture, identity, and ethnic
con¬‚ict in this book.

Getting started: Northern Ireland and Catalonia

Northern Ireland
After generations of violent con¬‚ict, both Catholic and Protestant para-
military groups have since 1994 (mostly) observed a cease¬re in Northern
Ireland. In 1998 the major political parties, along with the British and
Irish governments, reached a negotiated agreement, variously called the
Good Friday Agreement (by most Catholics) or the Belfast Agreement (by
most Protestants). It called for a return to self-rule, and was rati¬ed
through a referendum by a majority of citizens in each group. However,
implementation of the agreement has been slow and incomplete. Con¬‚ict
did not disappear overnight, although it has taken new, less violent forms
with particular focus on cultural expressions such as parades and of¬cial
insignias. Listening to the parties describe what is at stake in these con-
¬‚icts quickly reveals the deepest fears and insecurities that still divide the
Cultural Contestation in Ethnic Conflict

people of the region, and the divergent narratives about Protestant Loyal
Order1 parades and other cultural expressions.2
Following time-honored traditions, throughout the ˜˜marching season™™
in Northern Ireland Protestant men in dark suits and bowler hats
assemble at local lodges, attend church services, and hold parades to mark
various sacred days, with a particular emphasis on two dates in the ¬rst
half of July: July 1, the anniversary of the Battle of the Somme in 1916, in
which many soldiers from Northern Ireland died; and July 12, the day
when, in 1690, William of Orange™s Protestant forces defeated Catholic
King James™ troops at the Battle of the Boyne. Protestant accounts of the
parades stress their solemn, religious nature and the occasions they mark
(Lucy and McClure 1997). Banners celebrate key events in Protestant
history and highlight important religious themes, symbols, and persons.
Bands accompany the marchers playing familiar music, and at signi¬cant
parades important politicians address the crowd (Bryan 1997; Fraser
2000a; Jarman 1997).
Catholic narratives about the parades emphasize their celebration of
the Protestant triumphalism and the oppression that marked centuries of
British rule and Protestant domination in the region. The aggressive
music of the ˜˜blood and thunder™™ bands made up of young men often
clad in paramilitary symbols, and the viciously anti-Catholic lyrics of
some of their songs are further evidence to Catholics that the parades are
acts of aggression. Catholics in many parts of Northern Ireland have
organized and demanded changes in the parades, and especially in parade
routes passing through Catholic neighborhoods.
Parading identity in Northern Ireland has a long history and often
raises sectarian tensions (Fraser 2000a). As a result, any parade in
Northern Ireland can easily become an emotionally charged, exclusive
political expression. The Loyal Order parades have been going on for
generations, with periods of greater and lesser vigor, and more and less
strife. In 1997, the British government took steps which led to the crea-
tion of a Parades Commission, charged with developing procedures
for overseeing parades generally and especially for contentious ones.

The term Loyal Order indicates the parades™ expression of support for the continuing link
between Northern Ireland and Britain and their ongoing allegiance to the British crown.
A brief note on terms: Protestants are generally referred to as either Unionists because
they favor the continued union of Northern Ireland and Great Britain or Loyalists because
they are loyal to the crown. Catholics are variously called Nationalists or Republicans and
favor reuniting the island into one political unit.


Even though there are only a little over 1.5 million people in the region,
the Parades Commission reported that in 2003 there were over 3100
parades of which about 7 percent to 8 percent are ˜˜contentious,™™ i.e. that
they were brought before the Parades Commission, which holds a hearing
before deciding whether they can proceed as planned or require mod-
i¬cation.3 The vast majority of parades are exclusively Protestant (70%)
or Catholic (4%) affairs that mark, celebrate, or commemorate events of
signi¬cance to each community,4 are typically celebrations of in-group
solidarity, and are widely perceived as statements of domination or
resistance (Bryan 1997; Jarman 1997). While Catholics living on or near
the Loyal Order parade routes strongly resent the parades, and the often-
aggressive behavior of the participants, Protestants contend that restric-
tions on parading along ˜˜traditional™™ routes are an infringement of their
religious and political rights. In recent years there have been confronta-
tions with police, and violence and death associated with parades, espe-
cially those in South Belfast and Portadown.
The annual parade in Portadown in County Armagh, long a site of
sectarian violence, has been especially contentious for quite some time. In
the 1980s, the police rerouted it away from a Catholic area near the start
of the route after several years of severe violence (Bryan 1997). From 1995
through 1997, following the IRA and other paramilitary cease¬res, the
police announced a ban on the last part of the parade down the Garvaghy
Road through another Catholic neighborhood; and each year they then
reversed their position and allowed the parade to proceed. Angry
Catholics attacked the police. In 1998, the ¬rst year of its existence, the
new Parades Commission addressed the Portadown situation, insisting on
dialogue between the Orange Order and the Catholic Residents™ Asso-
ciation; when the Orange Order refused to negotiate, the Commission
prohibited the marchers from returning to their Lodge along the
Garvaghy Road. Protestant violence in Portadown and deaths in other
parts of the province followed. Though the violence has since decreased,

http://cain.ulst.ac.uk/issues/parade/pc/pc231204ar.pdf. The Parades Commission also
reported that in 2002 they rerouted forty-six loyalist parades and imposed various
restrictions on others. In addition, there were twenty-six disorderly incidents involving
Loyalist parades, most typically involving clashes between marchers and their supporters
and nationalist protesters. Jarman (1997: 118“19) and Fraser (2000:4) report data from the
mid 1990s on the number of parades in Northern Ireland that are quite similar.
The remaining 407 parades are not classi¬ed as either Loyalist (Protestant) or Nationalist
(Catholic) and include such events as May Day parades held by trade unions and Salvation
Army parades. More details on parades in Northern Ireland are provided in Chapter 4.

Cultural Contestation in Ethnic Conflict

the annual confrontation has produced a yet unresolved stalemate and
seven years later each July in Portadown there are still large numbers of
security forces, including army units who ¬‚ood the ¬elds beneath
Drumcree church and string barbed wire to prohibit the marchers from
completing their circuit.
The Portadown District Orange Order Lodge insists the issue is a
matter of free speech and the ˜˜right to walk the King™s highway.™™ For
years, their ˜˜civil rights website™™ has proclaimed support for equality,
justice, tolerance, and respect; it prominently displays the words of
Martin Luther King, ˜˜Somewhere I read of the freedom of assembly.
Somewhere I read of the freedom of the press. Somewhere I read that the
greatness of America is the right to protest for right. And so just as I say,
we aren™t going to let any injunction turn us around. We are going on.™™5
To dramatize their position, in 1998 a small group of Orangemen camped
out in the ¬eld next to Drumcree church vowing not to leave until they
were allowed to walk down the Garvaghy Road to return to their lodge.
˜˜Here we stand. We can do no other,™™ proclaims the sign next to the
church where they have since been camped symbolically.
Catholic opposition to the present parade route emphasizes their
experiences of long-term victimization, and their right to be free from
intimidation, and asks why the parade route isn™t simply changed as many
have been in other towns and cities, including some in the early 1980s in
Portadown (Bryan 1997). Catholic residents™ associations in Portadown
and elsewhere (which Protestants often dismiss as Sinn Fein fronts)6 have
demanded that parade organizers negotiate contentious questions sur-
rounding parades with them and have declared, ˜˜No consent, no parade.™™
For Catholics (and for the Parades Commission), dialogue and negotia-
tions are the proper mechanism for managing differences over parades;
they point to other places in Northern Ireland where negotiations have
produced agreements around parades that have allowed them to proceed in
ways that each side has accepted. (An important example of this
con¬‚ict mitigation process, found in Londonderry/Derry, is discussed in
Chapter 4.) In contrast, many Orangemen, such as those in Portadown,
reject the idea that their basic, traditional right to parade needs to
be negotiated, let alone with Sinn Fein and former IRA members.

Sinn Fein, currently the largest Catholic political party in the region, is the political wing
of the Irish Republican Army (IRA).


The Parades Commission has pushed the Orange Order to enter into
negotiations and asserts that it considers the essence of engagement to be
attempts at genuine communication between protagonists to a particular
parading dispute.
This on-going con¬‚ict around parades is so polarized and so stuck
because con¬‚icts about parades in Northern Ireland are not fundamen-
tally about freedom of speech or religion or protection from intimidation,
but about the threatened identities of people in the region. ˜˜Put simply,
the parades issue goes to the heart of the deeply fractured society that,
sadly, Northern Ireland represents™™ (North 1997: 41). The importance of
social identity and its expression in ethnic con¬‚icts, how it is symbolized
and communicated, and how it affects political behavior and beliefs is
central to the analysis that follows (Ross 1997). Focusing on identity is
especially useful in explaining ethnic con¬‚ict™s intensity, and how the
content and salience of identity variously resists and yields to change
(Ross 1993a; 1997; 2001a). Understanding identity directs our attention
to the deepest fears that drive ethnic con¬‚ict, as for example when an
Orange Order website proclaims, ˜˜If Orange parades continue to be
stopped, then over the years, Protestant culture will be slowly strangled.™™7

Catalonia is a region of 6 million people in northeast Spain that includes
Barcelona. There is a long history of tension between the Spanish state
and the region which was once independent politically and linguistically.
Today Catalans are bilingual, speaking Catalan, a romance language
closely related to Provencal, and Castilian (Spanish). Catalonia was
incorporated into the Spanish state over several centuries8 and Castilian
came to be spoken more and more, especially among the region™s elites
(Laitin 1989). Industrialization, in-migration, and a Catalan linguistic and
cultural revival marked the nineteenth century as Catalans pressed for
regional autonomy. However, in the 1920s Spanish dictator Primo de
Rivera restricted Catalan cultural expression and political liberty, both of
which were revived brie¬‚y during the Republic in the early 1930s. The
region was a site of intense ¬ghting and brutal violence during Spain™s
Catalans solemnly mark September 11 as the day when they ¬nally lost their autonomy
through their defeat in the Battle of Malplaquet in which they were on the losing side in
one of the many wars of Spanish Succession.

Cultural Contestation in Ethnic Conflict

civil war (1936“39) and under Francisco Franco (1939“75) Catalan
autonomy and cultural expressions were severely curtailed. His regime
banned the use of the language in education and government, prohibited
publication in Catalan, and burned books written in the language.9
There was a slow, unof¬cial revival of Catalan as a spoken language in
the later years of Franco™s rule, and when he died in 1975 Catalans were
poised to demand political and linguistic autonomy for the region, and
the return to the use of Catalan in schools. At the time, many observers
expected that the strong emotions surrounding these issues would spark
con¬‚ict and violence between the Spanish state and hard-line Catalan
nationalists. Although there was a good deal of tension and tough talk
exchanged between the parties, there was no signi¬cant violence on either
side. Catalan cultural expression is often strident and even aggressive but
never violent. There have been repeated requests for direct representation
in European Community organizations, for example, that annoy, and
even exasperate, the Madrid government, but the dueling visions of Spain
have not clashed violently.
Two competing narratives have long existed here, but since 1975 the
opposing sides have found ways to bridge them to avoid stalemate. Since a
good theory of ethnic con¬‚ict needs to account for the absence of violence
and con¬‚ict as well as its presence, we have as much of an obligation to
explain the success of the Spanish government and Catalan nationalists in
developing a mutually acceptable solution without resort to violence as we
have to explain the failure to do so in Northern Ireland.
There is a long history of tension between the Spanish state and the
demands for linguistic and cultural autonomy from the historical
nationalities in Spain,10 and there have been periodic violence and wars
pitting the regions against the center, which is located, geographically and
politically, in Madrid. At various times, including during Franco™s rule,
the center™s wishes dominated. However, following Franco™s death, the
new Spanish government, including many who had served under Franco,
realized that change was needed both within the country and in Spain™s
relationship with its Common Market neighbors. Democratization and

The regime moved quickly against those expressing Catalan nationalist sentiments. For
example, in 1960 the future Catalan leader Jordi Pujol was sentenced to seven years in
prison and served two and a half for having led a Catalan nationalist campaign that
included singing the Catalan anthem in Barcelona™s Opera House at a time when Franco
was visiting Barcelona.
The term refers to the Catalans, Basques, and Galicians.


liberalization were seen as essential, and the new constitution provided for
signi¬cant regional autonomy including linguistic rights in ways that
recognized multiple identities “ Catalan nationality and Spanish citizen-
ship. King Juan Carlos, Franco™s chosen successor, strongly supported
these efforts and the country™s transition to democracy (Linz, Stephan,
and Gunther 1995).
Catalan quickly returned to the schools and media in Catalonia, while
Castilian remained the of¬cial language of Spain. The result was of¬cial
bilingualism in the region that the Spanish state accepted. Making this
work took some time given that even among those who spoke Catalan
during the Franco years, most did not know how to read or write it.
Three decades later, however, this has changed; both languages are now
used in schools and universities, the media, and government. What is
even more striking is that many Catalan speakers today are either
immigrants from other parts of Spain (and especially Andalusia) or their
descendants, who have been willing to learn and use Catalan. In great
part this is because Catalans have emphasized an inclusive, non-racial
de¬nition of what it means to be Catalan. As the Catalan nationalist
leader, Jordi Pujol has said, ˜˜A Catalan is someone who lives and works
in Catalonia.™™11
The negotiated relationship between Madrid and Catalonia produced a
mutually acceptable arrangement in contrast to the situation between the
Spanish state and the Basques, although the Basques have even more
autonomy than the Catalans.12 All but the most hardline Catalans are
more or less satis¬ed with the arrangements and despite tensions at sev-
eral points in the past twenty-¬ve years, there has been no violence.
Barcelona has prospered both economically and culturally and it is
referred to as one of the four motors of southern Europe. Nonetheless,
since 1980 the relationship between the center-right Catalan government
and Madrid has sometimes been complicated. At times, both the right and
the left in Madrid have found the Catalans insuf¬ciently attentive to the
needs of Spain as a whole and complain that Catalans look north toward
Europe too much and south too little.

Many add that the unstated premise here is that the person speaks Catalan. Of course, in
practice the designations are more nuanced.
A number of authors have sought to explain differences in the two cases. Most focus on
the very different nature of Catalan and Basque identity “ the former as more cultural and
inclusive and the latter as more genealogical and exclusive “ and differences in the social
organization of the two communities (see for example, Conversi 1997; Laitin 1995).

Cultural Contestation in Ethnic Conflict

Many of these tensions came to a head symbolically when in the mid-
1980s Barcelona was awarded the 1992 Summer Olympic Games. Putting
on the games required complex cooperation between the city, the region,
and the Spanish state and raised a number of identity issues. Perhaps the
most striking illustration of this was a two-page advertisement that
appeared in major North American and European papers a few months
prior to the games, asking on the left-hand page ˜˜Where is Barcelona?™™
On the right-hand page came the response, ˜˜In Catalonia.™™ Needless to
say, the Spanish government, which was paying the most money to host
the games, as well as many non-Catalans in Spain, was not pleased.
There were, however, pressures that limited the con¬‚ict to symbolic
expressions. The Catalans wanted to show they could put on such a visible
world-class event without any problems and to present Barcelona as a
dynamic world-class city, while the Spanish hoped the games would
showcase their country™s economic growth and political accomplishments
since Franco. In the run-up to the games, con¬‚icts arose continually
concerning the designation of of¬cial languages, the ¬‚ags to be displayed,
and the anthems to be played. These issues came to a head prior to the
games at the 1989 inauguration of the Olympic stadium “ one that had
originally been built to hold an alternative games to Hitler™s Berlin
Olympics in 1936 “ when the Catalans jeered long and hard as King Juan
Carlos entered. It was clear that symbolic issues threatened how the world
would view Spain and Catalonia.
In the six months leading up to the July 1992 games there were
extensive Catalan symbolic displays and a marked rise in tension that
never became violent. In the end, Castilian and Catalan were given equal
status among the of¬cial languages and both ¬‚ags and anthems were
featured in a way that all sides felt was balanced. During the opening
ceremonies, the king entered the stadium while the Catalan anthem was
playing “ to inhibit jeers from the crowd “ and then pronounced the
games open in Catalan and drew a huge warm response from the crowd.
He was visible throughout the games and served as a symbol of the
Spanish state but also through his presence and the nature of his invol-
vement offered a clear legitimation of the expression of Catalan identity.
Catalan ¬‚ags and cultural expressions were ever present during the games,
but there were no violent incidents. In the end, when the games were
over, both Catalans and people throughout the rest of Spain believed the
Olympics had been successful and expressed pride in the achievement
(Hargreaves 2000). There was increased support for the idea that people

could be both Catalan and Spanish and that these two identities were not
mutually exclusive.
When hearing about long-standing con¬‚icts in Northern Ireland, such
as those involving the parades, there is a tendency for outsiders to simply
dismiss them as unavoidable and irrational and to symbolically seal off the
region as ˜˜a place apart.™™ In contrast, reactions to the accommodation
between Catalan nationalism and the Spanish state are sometimes
explained in a different way “ as examples of the pragmatism and
rationality of the Catalans and the reasonableness of the post-Franco
Spanish state. Simply to accept this explanation is also a mistake for there
was no inevitability to the outcome that was reached; many contemporary
observers expected considerable disarray and violence. Despite a reputa-
tion for reasonableness, Catalan resistance to the Spanish state has
sometimes been strong, and the violence in Barcelona around social issues
in the nineteenth and early in the twentieth centuries has been far from
simply pragmatic on many occasions. The fact that the con¬‚ict evolved in
a deescalatory direction didn™t mean that this resolution, or others
described in this book, was inevitable or that initial differences between
the parties were not great.
Neither of these common responses to con¬‚ict or its absence offers any
real explanation for why and how differences on some cultural issues turn
bitter and violent and others are managed constructively. Nor do they
recognize that two decades ago few observers would have predicted either
that in Northern Ireland cultural questions such as parades would move
to the center of the con¬‚ict, or that the post-Franco state would effec-
tively negotiate extensive linguistic and cultural rights and meaningful
levels of political autonomy for the country™s historical nationalities.

Cultural expressions and ethnic con¬‚ict: initial questions

What are ethnic con¬‚icts about?
It would be foolish to suggest there is a single short answer to this
question, for ethnic con¬‚icts are about competing interests, constitutional
arrangements, and political power, as well as about incompatible iden-
tities. Evaluating competing theories of ethnic con¬‚ict is not a goal here
as I have treated this question in prior publications emphasizing the
differences between structural theories that focus on interests and psy-
chocultural theories that stress incompatible identities (Ross 1993a;
Cultural Contestation in Ethnic Conflict

2000a; 2001a). Interests and identities are often highly interconnected
(Bates, Rui J. P. de Figueiredo, and Weingast 1998).13
Identity issues are not just peripheral concerns; they are at the core of
long-term con¬‚icts such as in Northern Ireland and Catalonia. In these
intense con¬‚icts, the contending parties present mutually exclusive
positions in which if one side wins the other necessarily loses. In con¬‚icts
like these each side deeply fears that recognizing the claims of the other
invalidates their own. As a result there is an escalation of demands and
actions and an increase in claims that leave little room for compromise or
mutual recognition.
In Northern Ireland, opposing identities continue to be de¬ned in
mutually exclusive terms. Each side continues to fear the other and sees
them as seeking total victory. For Catholics this means fears of a return to
unfettered Protestant rule in which Northern Ireland was ˜˜a Protestant
state for a Protestant people,™™ as Northern Ireland™s ¬rst Prime Minister
James Craig put it. For Protestants, the fear is that if Ireland is reunited as
most Catholics would like, they will become a small minority whose
rights will be ignored. To date, despite the signed agreement that insures
protections against both of these outcomes without majority consent,
many on each side remain distrustful and the self-government that the
1998 agreement promised has functioned for only a fraction of the eight
years since the agreement was signed. The problem, it seems, lies not in
the arrangements which constitutional scholars have saluted for their
ingeniousness, but in the mutual distrust that makes the present-day
stalemate preferable to the risks of cooperation.
During Franco™s reign, con¬‚ict between the central government in
Madrid and the Catalans was framed in mutually exclusive terms and
Franco refused to recognize that Catalan or Basque identity could be
compatible with a Spanish one. In contrast, the ability of post-Franco

The distinction between the two is analytic, and people caught up in con¬‚icts intuitively
understand their empirical linkage. For example, it is easy to see how the achievement of
certain interest goals, such as gaining a political of¬ce or improved job opportunities, can,
at the same time, address a group™s identity and recognition concerns. To the extent that
interest claims are ˜˜tests™™ of a group™s acceptance as a legitimate political player, then
achievement of the interest claims also addresses concerns about identity. However, there
are times when a group may be ready to drop or alter an interest claim if identity needs
can be met in another manner, especially when identity-based fears of exclusion diminish.
Understanding intense ethnic con¬‚ict as involving both interests and identities thus
increases not only our analytical understanding but also our options for constructive
con¬‚ict management (Ross 1993a; 1993b).


Spain to recognize its citizens™ multiple identities and to develop political
arrangements for their expression provided the space for the rede¬nition
of the con¬‚ict in more inclusive directions and isolated the remaining hard
core on either side “ those who sought either Catalan independence or
Franco-like centralization “ as is explained at greater length in Chapter 6.

Why the intensity?
A striking feature of many identity-based ethnic con¬‚icts is the emotional
investment parties make in what to outsiders often seem unimportant
matters. The fact is, however, that any matter invested with emotional
signi¬cance is no longer trivial to those involved, and intransigent
intergroup disputes quickly become characterized by perceived threats to
group self-esteem and legitimation (Ross 1995). In this, dynamic identity
issues and threats to the group are at the core of peoples™ concerns.
Furthermore, the con¬‚ict becomes such a central part of their identity
that giving it up is giving up a part of oneself (Kaufman 2001; Kelman
1999; Northrup 1989). Such emotion-laden con¬‚icts are especially dif¬-
cult to settle. Often the fact that each side feels the same intense emotions
makes it dif¬cult to recognize their common experiences and shared parts
of their narratives (Nic Craith 2002), and promotes a ˜˜double minority™™
view of the con¬‚ict in which each side feels vulnerable. For example,
Protestants and Catholics in Northern Ireland each see themselves as
threatened and have trouble acknowledging the other™s parallel percep-
tions. This is because each party™s emotional concerns make it very dif¬-
cult to hear, let alone acknowledge, the other™s account, especially when
their own actions may be the root cause of an adversary™s feelings and
behavior. Battles over cultural expressions and the symbolic landscape are
central to group recognition and identity; it should not be surprising that
con¬‚icts around cultural questions are intense when they raise basic issues
concerning a group™s legitimacy and deep fears about threats to its exis-
tence. Central to understanding the intensity of some identity con¬‚icts is
uncovering the existential fears group narratives evoke as they try to
explain the social and political world.

Why does each party offer such different accounts of the same con¬‚ict?
How can people who have been in intense interaction and con¬‚ict with
each other for so long have such different ideas about what the con¬‚ict is
Cultural Contestation in Ethnic Conflict

about, about signi¬cant events, and about key motives for themselves and
opponents? These strikingly different narratives are often dismissed as
rhetorical devices for asserting the priority of their claims. But this is
misleading. While competing narratives serve rhetorical purposes, they
are much more than political posturing. They are also ˜˜obviously true™™ to
group members and offer plausible explanations about the con¬‚ict, its
causes, the motives of the parties, and what appropriate behavior follows
from them.
Examples of clashing narratives are accounts of the American civil war
that emphasize the importance of race and slavery versus those that focus
on states™ rights and constitutional principles; Israeli descriptions of the
1948 war as the war for independence and Palestinian descriptions of
the same event as El Nakba (the catastrophe); Protestant descriptions
of Catholic hunger strikers in 1981 as murderers while Catholics call the
same men martyrs; and the often cited different descriptions of perpe-
trators of political violence as terrorists versus freedom ¬ghters. As
worldviews, contrasting accounts are not just different; in addition, key
elements in one narrative can directly clash with those in others while the
emotional attachment to each narrative makes it dif¬cult, if not impos-
sible, to challenge them directly. The diametrically opposed descriptions
reveal alternative worldviews but in addition we need to recognize that
for those on each side even partial acknowledgment of an opponent™s
perspective is emotionally a threat to one™s own core beliefs.

Why should we study contested cultural expressions?
While it is certainly the case that ˜˜parades con¬‚icts aren™t really about
parades,™™ there are many good reasons to investigate such disputes. First,
cultural disputes offer a window through which we can better understand
the multiple layers and issues in long-standing intractable ethnic con¬‚icts
in which these disputes are embedded. As ˜˜hot spots™™ they reveal the most
basic identity dimensions, needs, and intense emotions of the larger
con¬‚ict in which they are embedded, revealing deeper differences
between groups (Zolberg and Woon 1999). They ˜˜become invested with
strong emotions because of past or current political, military, or historical
conditions™™ (Volkan 1997: 211). Second, contested cultural expressions
can focus attention on the emotional dimensions of a con¬‚ict and can
complement more instrumental bargaining, leading to negotiated agree-
ments and constitutional changes. Third, because cultural con¬‚icts are

constructed they can also be reframed and reconstructed in directions that
facilitate deescalation and settlement. Cultural understandings can, and
do, change, over time. In some cases, culturally de¬ned con¬‚icts that seem
completely intransigent at one point in time are rede¬ned in more
inclusive terms as part of a constructive con¬‚ict management process.
As compelling dramas, cultural con¬‚icts reveal powerful competing
accounts of the issues at stake to the parties themselves, to potential
dispute settlement specialists, and to scholars. Understanding the content
of contrasting narratives, and the mutual denial of the relevance of
opposing accounts, can build a deeper appreciation of why in these
con¬‚icts it is especially dif¬cult for opponents to view each others™ actions
in a benevolent light even when each side has very similar motives and
fears.14 As dramas, these con¬‚icts are staged for in-group and out-group
audiences. However, it is important to point out that when these dramas
¬rst develop they are rarely planned in advance. Rather, different players
act on the basis of their well-established assumptions about themselves
and their opponents, as well as their deep emotional commitments to
engage in long-term, impassioned, and sustained pursuit of their group™s
existential goals.

Culture, identity, and ethnic con¬‚ict
Culture and identity are core concepts for the study of ethnic con¬‚ict.
While I have written a good deal about them elsewhere, before pro-
ceeding to develop my larger argument it would be helpful to say how
studying them is relevant here (Ross 1993a; 1993b; 1995; 1997; 2001;
2002). While a ¬rst take is that culture and identity might pose nothing
but problems in ethnic con¬‚ict, this simple conclusion is premature; the
analysis here also explores ways in which they may lead us to solutions
that permit opponents to constructively manage con¬‚ict.

Culture is a shared system of meaning people use to make sense of the
world, ˜˜an historically transmitted pattern of meaning embodied in

In South Africa, for example, this is especially striking in the similarity between narratives
recounted in decades-old Afrikaner memorial sites and those in new memorials and
museums built in post-apartheid South Africa. Both emphasize themes of struggle,
liberation, and political independence as well as a common opponent “ British

Cultural Contestation in Ethnic Conflict

symbols, a system of inherited conceptions expressed in symbolic forms by
means of which men [sic] communicate, perpetuate, and develop their
knowledge about and attitudes toward life™™ (Geertz 1973a: 89).15 Culture
can be examined through the narratives people recount to explain their
own and others™ actions as well as the institutions and practices found in a
society (Ross 1997; 2002). This de¬nition of culture emphasizes public,
shared meanings. Behaviors, values, institutions, and social structure are
understood not as culture itself but as culturally constituted phenomena
(Spiro 1984). These behaviors include emotionally signi¬cant ritual
actions that link people across time and space while distinguishing
between in-group and out-group members. Within-group connections are
preserved in group memory through oral, written, and visual accounts and
reenacted through social action and ritual performance. Culture from this
perspective is a worldview that includes cognitive and affective beliefs
about social reality and assumptions about when, where, why, and how
people in one™s culture and those in other cultures are likely to act in
particular circumstances (Avruch 1998; Chabal and Daloz 2006; Ross
1997). In sum, culture is a framework for interpreting the world that marks
˜˜a distinctive way of life™™ characterized in the subjective we-feelings
among group members, and expressed though speci¬c behaviors including
customs and rituals “ both sacred and mundane “ that mark the daily,
yearly, and lifecycle rhythms that connect people across time and space.
Cultures and cultural differences do not themselves cause con¬‚ict
(Eller 1999; Posner 2004) but are the lenses through which the causes of
con¬‚ict are refracted (Avruch and Black 1993: 133“34). People begin
con¬‚icts, often for what they believe are only economic and political
reasons; but it is important to understand how leaders and groups evoke
cultural meanings and the deep feelings they evoke in organizing col-
lective action. Eller (1999: 48) describes culture as a code for authentic
and alternative groupness, and the basis of context-speci¬c political
claims. Cultural meanings and the emotions associated with them are
not invariant and inter-group interaction and contestation affect them.
For example, in the Middle East and South Asia there are sacred sites
that different religious communities have at times shared; at other
times as con¬‚ict between the groups has intensi¬ed, exclusive claims for
D™Andrade (1984: 88) points out the radical shift in the social sciences after the 1950s
from the view of culture as behavior that could be understood within a stimulus-response
framework to culture as a system of meaning. For a more complete discussion of culture
as meanings and symbols see Schweder and LeVine (1984).


their use has increased. Cultural con¬‚icts are not just out there ready to
happen as Kaplan™s (1993) ˜˜ancient hatreds™™ argument would have us
believe. Rather, as con¬‚icts evolve the intensity of emotions surrounding
cultural expressions and enactments as well as their meanings often shift
(Weeden 2002).
Many people have an image of culture as uni¬ed and monolithic. This is
a serious mistake. The de¬nition of culture I am using emphasizes only that
people within a given cultural group understand, but not that they always
agree with, one another (Avruch 1998; 2003). Understanding shared
meanings is not the same as agreement in values or engagement in the same
behaviors (Chabal and Daloz 2006; Cohen 1991). As LeVine argues,
˜˜Culture represents a consensus on a wide variety of meanings among
members of an interacting community approximating that of the consensus
on language among members of a speech-community™™ (1984: 68). This
does not mean there is unity in thoughts, feelings, behavior and even
conceptions of the social order. This point is crucial for my analysis for it
helps us understand that within the cultural communities discussed in the
following chapters there is frequent intragroup con¬‚ict and competition
such as one ¬nds between religious and secular Jews in Israel, federalists
and supporters of independence in Quebec, Republicans and Nationalists
among Catholics in Northern Ireland, and integrationists and pluralists in
France (Eller 1999). Culture, and its narratives, offer in-group agreement
about meaning but not necessarily about substance. As a result con¬‚ict and
its management is a two (at least) level game, with one level focusing on
within group competition and the second on what occurs between groups.
Finally, we need to recognize that what people who share a group identity
believe is shared is often greater than what is actually shared, and is over-
stated in the interest of presenting a uni¬ed, strong group to outsiders.
In a similar fashion, we often imply that cultures and ethnic groups are
bounded, uni¬ed, and purposive entities, when in fact as Brubaker says
˜˜˜Groupness™ is a variable™™ (Brubaker 2004: 4). Ethnic and national
groups are not organizations that charge dues and issue membership
cards, and who is included and who is not can be murky and contested.
Rather, they are af¬nity groups whose boundedness, unity and coordi-
nation vary across time and space (Barth 1969; Eller 1999). Our language
gets in the way here however since we readily employ collective nouns to
talk about large groups that are internally differentiated and often have
more trouble acting collectively than the term ˜˜group™™ implies. We write
˜˜Israelis think . . . ™™ when it is the case that what we mean is ˜˜a good
Cultural Contestation in Ethnic Conflict

number of Israelis, perhaps, an overwhelming majority of Israelis
think . . . ™™ But if we put in all the qualifying language to capture internal
variation in every sentence a manuscript would quickly be unreadable.
The point is that even though culture and ethnic identity are not bounded
entities, they are nonetheless important concepts in understanding
many social and political con¬‚icts, and studying them has its own
methodological complexity.
Identity-af¬rming cultural activities and institutions can play a central
role in the pursuit of political goals, as anthropologist Abner Cohen
(1969) observed over three decades ago. He pointed out that in many
settings where explicit political organization is either not possible or not
likely to be effective, cultural organizations build or maintain in-group
solidarity and organize collective action. Cohen realized that intensive
participation in shared cultural activities and organizations increased
feelings of group distinctiveness, facilitated within-group communication,
enhanced group decisionmaking, increased within-group authority, spread
shared ideology, and promoted within-group discipline (1969: 201“11);
these all facilitate ethnic groups behaving like interest groups.
Cohen™s emphasis on culture is quite different from Samuel Huntington™s
(1993) often cited clash of civilizations. Whereas Huntington emphasizes
political divisions built on broad-based long-standing cultural differences,
Cohen is more interested in the shifting nature of culturally de¬ned groups
and cultural expression in response to changing contexts. For example, in
Ibadan, a city in western Nigeria, he analyzes the rise of an Islamic broth-
erhood as a mechanism for political and economic coordination among
Northern Nigerians whose numbers were too small to form an effective
political party. In a more recent study, he uses the contested nature of the
Notting Hill Carnival in London to examine the changing nature of the
relationship between West Indians and native British in London (Cohen
Cohen™s studies explore culture as a resource that groups, not always
consciously, use strategically, although he also recognized culture™s
expressive power.16 In his study of the Carnival in London, Cohen
examines changing cultural meanings, contestation over the ˜˜ownership™™

While David Laitin (1986) contrasts this strategic use of culture with that of Clifford
Geertz, his reading of Cohen is in my view too limited. Cohen, like Geertz, is also
interested in how people construct meaning.


of cultural production (music and the instruments on which it is played),
and con¬‚ict over cultural expression as mechanisms for the management
of within- and between-group relationships. His analysis differs sharply
from Huntington in that Cohen emphasizes the fact that many con¬‚icts
Huntington calls cultural or civilizational are, in fact, political, even
though they are often framed in cultural terms and cultural issues play a
central role in them (Eller 1999: 141, Ross 2002). While there is often an
important linkage between culture and politics, as Huntington suggests,
cultural differences themselves are not at the core of these con¬‚icts.
Culture is neither the root cause of ethnic con¬‚ict nor an epipheno-
menon. Rather, con¬‚ict is about both material interests and collective
cultural identities (Ross 1993a, 1997; Bates, Figueiredo, and Weingast
1998), and understanding a con¬‚ict™s cultural frames is a central challenge
to the analysis and constructive management of them. Rejecting as too
simple the hypothesis that con¬‚icts are about cultural differences is not
incompatible with the proposition that culture plays a signi¬cant role in
con¬‚ict and that cultural issues often encapsulate and symbolize the dif-
ferences between the contending parties as cultural performance and
memory become hotly contested and intertwined with strongly held
political positions (Ross 1997).
Cultural enactment and performance refer to behavioral expressions that
evoke central meanings, images, and metaphors rooted in collective
memories that are emotionally signi¬cant for a group and its members.
Obvious examples include celebration of national holidays, display of
emblems such as ¬‚ags, sacred music, funerals, and religious ceremonies in
sacred locations. Cultural enactment is powerful because it is linked to
cultural memory, and serves to renew memories across generations by
expressing a group™s most basic hopes and fears. Cultural enactment
renews links among members while emphasizing distinctions between
group members and outsiders. Performance and memory create emo-
tional realities, making accompanying narratives seem truthful to group
members though they often puzzle outsiders. The need for these tools for
weaker or minority communities is obvious, but we must understand
that dominant groups use culture as well to de¬ne the playing ¬eld and
the rules of the game, to articulate and assert political claims, to mobilize
supporters, justify their dominant position, and to control minorities, as
for example in the development and implementation of state language
policy (Laitin 1998). Dominant and subordinate groups that feel threa-
tened can use performance to communicate identities that de¬ne
Cultural Contestation in Ethnic Conflict

boundaries between themselves and outsiders and can make it more or
less dif¬cult to become a group member or hold multiple identities.

Con¬‚ict and identity
Cultural identities, such as ethnicity, connect individuals through per-
ceived common past experiences and expectations of shared future ones.
In linking people across time and space, identity both de¬nes and rein-
forces social categories that organize a good deal of behavior. People
sharing a group identity possess, to a greater or lesser degree, a sense of
common fate including expectations of common treatment, joint fears of
survival/extinction, and beliefs about group worth, dignity, and recogni-
tion. Identity involves group judgments and judgments about groups and
their motives. For example, Horowitz (1985: 147“92) discusses the power
of assigning the labels ˜˜backward™™ and ˜˜advanced™™ to ethnic groups in
colonial and post-colonial settings and claims of entitlement that groups
do or do not make as a consequence of such a designation.
Humans have an evolved predisposition for sociality and a well-
developed capacity to form cohesive social groups (Howell and Willis
1989), and in-group identity provides the basis for a fundamental paradox
of human existence. It facilitates physical and emotional survival and
within-group cooperation; at the same time, strong in-group solidarity
promotes out-group competition and con¬‚ict (LeVine and Campbell
1972).17 Social identity, including attachment to a group, begins to
develop at the earliest stages of the lifecycle and its intensity is crucial in
explaining why people are willing to make great personal sacri¬ces in its
name (Stern 1995).18 People with the same identity share targets of
externalization (Volkan 1988) and high emotional salience is attached to

Campbell (1975) proposes that cultural institutions such as religion have been important
in extending cooperative behavior beyond small related kin groups to far larger social
aggregates. One could compile a long list of cultural practices which have effectively built
upon this propensity to form groups even among individuals who have no prior ¬rst-hand
knowledge of one another.
Modern psychoanalytic writing is particularly helpful for understanding identity
development and the relationship between individual and ethnic identity (Ross 1995).
Unlike older, drive-based theories of psychodynamic functioning, contemporary object
relations theory with its emphasis on linking a person™s inner and outer worlds focuses on
the social development of attachment (Bowlby 1969; Greenberg and Mitchell 1982). This
work sees early social relationships as providing a template for ones which develop later in
life, and it is especially concerned with the parts of the outer world brought inside and
with inner parts projected outward (Stern 1985; Volkan 1988).


group differences that are emphasized through in-group symbolic and
ritual behaviors even when the differences are objectively small (Eller
1999).19 Normal development involves developing progressively wider
attachments to those in one™s interpersonal world and building connec-
tions to larger, more abstract entities such as the nation or ethnic group
(Ross 1995).20
While there is widespread recognition that attachment to social groups
is a powerful dynamic, there are signi¬cant differences among the ways
social scientists theorize this attachment. A key difference occurs around
the extent to which identities are seen as mutable and contextual versus
enduring and even primordial. My own view is that both extremes are
caricatures in their denial of the other position. As Smith (1991) argues,
while group de¬nition is more socially constructed than popular images
hold, it is not as easily altered in the short run as some constructivist
accounts suggest (also see Horowitz 2002).21 A second signi¬cant dif-
ference is the power of rational interest maximization versus affective
Freud™s expression ˜˜the narcissism of minor differences™™ is very apt here.
Normal development, facilitated by what Winnicott (1965) calls the good-enough
mother, encourages both the attachment of the individual to others and separation-
individuation, as a person builds both a sense of self and connections to a progressively
wider circle of attachment (Mahler, Pine, and Bergman 1975). Winnicott (1958) describes
the importance of transitional objects “ teddy bears, soft towels, and other treasured
objects “ that link a child™s inner and outer worlds, and are infused with high emotional
signi¬cance. It is easy to extend this linkage process to social, cultural, and political
objects “ signi¬cant symbols and rituals which are ¬rst encountered in safe, within-group
contexts, often in childhood, revisited in adolescence when peer groups and wider social
attachments are especially salient emotionally, and embedded in daily practices and their
culturally speci¬c sights, smells, and sounds. Finally, it is important to recognize that
identity is not ¬xed early in life and unchanging. Quite the contrary, since identity is
complicated by the fact that there is great range in the form and content of groups
humans create. Identity involves the capacity to distinguish between people who are like
oneself and those who are different in speci¬c settings; and depending on the context the
same people may be variously classi¬ed as alike or different as people manage multiple
A related question is how to explain the striking contrast between the mutable, contextual,
and constructivist character of ethnicity “ widely documented in recent social research
(e.g., Cohen 1969; Eller and Coughlan 1993; Waters 1990) “ and popular political
discourse and ritual behavior that see ethnic groups as ¬xed, unchanging, and often
biological entities which ¬ght over ˜˜ancient hatreds™™ (e.g., Kaplan 1993). Why is it that

. 1
( 11)