Citation
Opposition, Politicisation and Simplification:
Social and Psychological Mechanisms of Elite-Led Mobilisation
by
Marie-Eve Desrosiers
A thesis submitted in conformity with the requirements
for the degree of Doctor of Philosophy (Ph.D.)
Graduate Department of Political Science
University of Toronto
© Copyright by Marie-Eve Desrosiers (2008)
Opposition, Politicisation and Simplification:
Social and Psychological Mechanisms of Elite-Led Mobilisation
Marie-Eve Desrosiers
Doctor of Philosophy
Graduate Department of Political Science
University of Toronto
2008
ABSTRACT
Drawing on insights from social psychological literature on identity formation, and on social
movement and contentious politics literature, this research focuses on elite strategies to
gain from or survive a crisis. The research specifically looks at strategies to foster popular
support and mobilisation. It explores the use of divisive and ethno-centric discourses and
policies aimed at mobilising supporters in times of instability or crisis.
More specifically, it studies why some elite mobilising appeals have traction. To do so, the
research examines social and psychological mechanisms behind group solidarity. A
heightened sense of group solidarity is what leads individuals to think in terms of the
group, a necessary step for mobilisation. From there, they can be made to feel appeals for
collective action are warranted.
Three mechanisms in particular are discussed: opposition, politicisation and simplification.
Opposing entails enhancing feelings of attachment by creating a sense of antagonistic
relations with another group. Politicising consists in ascribing to group identities a political
nature, more conducive to contentious relations. The final strategy is simplification. It
ii
amounts to simplifying interpretations of the situation and environment so as to make them
more readily internalisable.
This framework is applied to contemporary Rwanda and to the lead-up to the wars in
Yugoslavia. In the Rwandese case, cultural and historical references were repeatedly used
by ruling regimes to foster a Hutu uprising against the Tutsi population.
This tactic
eventually played a fundamental role in triggering the 1994 genocide. In the former
Yugoslavia, Croatian and Serbian elites antagonised group relations by agitating
nationalist rhetoric. Though this was a strategy to stay in power or gain support, it also led
to the break-up of Yugoslavia and to wars in Slovenia, Croatia and Bosnia-Herzegovina.
iii
To C., M. and E., may you find peace despite your haunting memories.
iv
There is no monopoly of common sense.
—Sting, Russians
A nation is a group of persons united by a common error about their ancestry and a
common dislike of their neighbours.
—Karl Deutsch, Nationalism and its Alternatives
It is not what is, but what people believe is that has behavioral consequences.
—Walker Connor, “A Nation is a Nation, is a State, is an Ethnic Group is a …”
Quand vint finalement la paix, elle avait l’odeur de celle qui s’abat sur les prisons et les
cimetières, linceul de silence et de honte qui pourrit l’âme et ne s’en va jamais. Aucune
main n’était innocente, aucun regard n’était pur. Nous tous sans exception, qui avons
assisté à cela, nous en garderons le secret jusqu’à la mort.
—Carlos Ruis Zafón, L’ombre du vent
v
ACKNOWLEDGEMENTS
Several family members, friends and colleagues have been instrumental in the writing of
this dissertation. While they are too many to name, my warmest thanks go to them.
Special thanks are due to Thomas Homer-Dixon, my supervisor, to Robert Matthews and
Paul Kingston, members of my committee, and to David Welch, for their support and
patience, and for all I have learnt from them. I also wish to thank the Department of
Political Science, the Faculty of Graduate Studies and the Munk Centre for financial
support. This research could not have been possible without funding from the Social
Sciences and Humanities Research Council of Canada and from the Department of
National Defence’s Security and Defence Forum. Additional thanks are due to Stéphane
Roussel and Alex Macleod at the Université du Québec à Montréal, who ‘housed’ me for
four years. Their encouragements and faith in me were invaluable. François Audet from
CARE Canada, the staff of CARE Rwanda, and numerous others deserve my thanks for
making my fieldwork possible, but also making it an extraordinarily enriching experience.
My gratitude goes to a number of colleagues who read and commented on parts of this
dissertation, in particular Cédric Jourde at the University of Ottawa. Finally, I wish to thank
my parents, Louise and Berthold, my brother and sister-in-law, Christian and June, my
family and friends for the understanding and support they showed me throughout the
endeavour, in good, inspired times and bad, neurotic ones. Most importantly, I wish to
thank Philippe for the grounding, encouragement and love he provided, but especially for
his patience.
The thoughts and opinions expressed in this work are my own. Any errors found herein are
mine alone.
vi
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ....................................................................................................................... vi
MAP OF RWANDA...................................................................................................................................x
MAP OF THE FORMER YUGOSLAVIA ................................................................................................. xi
ABBREVIATIONS .................................................................................................................................. xii
GLOSSARY........................................................................................................................................... xiv
PREFACE ............................................................................................................................................ xvii
INTRODUCTION......................................................................................................................................1
Leaders and Followers: Social Dynamics....................................................................................... 3
Collective Backgrounds and Identity: Psychological Dynamics...................................................... 4
When elites mould collective identity .............................................................................................. 8
Outline of Chapters ......................................................................................................................... 9
CHAPTER 1: SOCIAL CONSTRUCTIVISM AND IDENTITY: A PREMISE TO THE ARGUMENT ......12
Social Constructivism in International Relations........................................................................... 14
Social Constructivist Insights for the Study of Identity.................................................................. 17
CHAPTER 2: CROWD PSYCHOLOGY: PRECURSOR TO MODERN STUDIES ON COLLECTIVE
BEHAVIOUR ..........................................................................................................................................26
Scipio Sighele and the Suggestibility of Crowds .......................................................................... 28
Gustave Le Bon and the Mental Unity of Crowds......................................................................... 30
Gabriel Tarde and Imitation .......................................................................................................... 35
CHAPTER 3: SOCIAL PSYCHOLOGICAL LITERATURE ON IDENTITY FORMATION AND
GROUP DYNAMICS ..............................................................................................................................40
Muzafer Sherif’s Experiments ....................................................................................................... 41
Henri Tajfel and Social Identity Theory......................................................................................... 42
Self-categorisation Theory ............................................................................................................ 46
CHAPTER 4: SOCIAL MOVEMENT THEORY: FROM ITS PRECURSORS TO CONTEMPORARY
PERSPECTIVES....................................................................................................................................53
Important precursors..................................................................................................................... 54
The Political Opportunities, Mobilising Structures and Framing Approach .................................. 60
Framing ......................................................................................................................................... 63
Dynamics of Contention................................................................................................................ 69
CHAPTER 5: STRATEGIES FOR ELITE-LED MOBILISATION: OPPOSITION, POLITICISATION,
SIMPLIFICATION...................................................................................................................................75
Synthesising the sequence of elite-led mobilisation ..................................................................... 77
A New Social and Psychological Take on Framing Collective Appeals for Mobilisation.............. 85
REVISITING IDENTITY ................................................................................................................... 87
OPPOSITION ............................................................................................................................... 94
POLITICISATION......................................................................................................................... 101
SIMPLIFICATION ........................................................................................................................ 107
GENERAL APPROACH ................................................................................................................ 113
CHAPTER 6: METHODOLOGY AND CASE STUDY RESEARCH DESIGN......................................116
Complexity and Contingency: Their Implications for Research .................................................. 117
Case Studies and Field Research Design .................................................................................. 128
CASE SELECTION ...................................................................................................................... 128
FIELD RESEARCH ...................................................................................................................... 131
vii
CHAPTER 7: RWANDA: THE DEVELOPMENT OF AN ETHNO-CENTRIC AND DIVISIVE
IDEOLOGICAL BACKGROUND IN RWANDA ....................................................................................137
Pre Colonial Rwanda .................................................................................................................. 137
Colonial Rwanda......................................................................................................................... 142
First Republic .............................................................................................................................. 155
Second Republic......................................................................................................................... 162
Impact of the Two Republics on the Population ......................................................................... 174
CHAPTER 8: RWANDA: THE CIVIL WAR AND THE GENOCIDE (1990-1994) ................................183
Background Variables................................................................................................................. 183
The Instrumentalisation of the Civil War: A Sufficient Explanation?........................................... 194
Early Days................................................................................................................................... 198
Spiraling Down Towards the Genocide ...................................................................................... 207
The Arusha Peace Process ........................................................................................................ 215
Impact on the Population ............................................................................................................ 223
CHAPTER 9: RWANDA FROM THE PERSPECTIVE OF FRAMES FOR MOBILISATION ...............236
The Classic Framing Argument: Resonance, Severity/Urgency, Efficacy/Propriety .................. 239
Waking Up Hutu Solidarity: Opposing, Politicising, Simplifying.................................................. 250
MAKING HUTU SOLIDARITY PART OF THE REPERTOIRE OF STRATEGIES ....................................... 251
OPPOSITION ............................................................................................................................. 257
POLITICISATION......................................................................................................................... 261
SIMPLIFICATION ........................................................................................................................ 266
CHAPTER 10: THE BREAK-UP OF YUGOSLAVIA AND ENSUING WARS ......................................274
Origins......................................................................................................................................... 274
Contemporary History: The Twentieth Century .......................................................................... 283
The Spiral towards Break-Up and War ....................................................................................... 293
Campaigning for Nationalism, Winning War ............................................................................... 311
CHAPTER 11: COMMUNICATION STRATEGIES IN THE FORMER YUGOSLAVIA ........................318
Impact of Ethno-Centric Rhetoric in Yugoslavia: Did Populations Believe?............................... 323
Opposition, Politicisation and Simplification in the Former Yugoslavia ...................................... 335
SERBIAN NATIONALIST REVIVAL THROUGH THE LENSES OF SOLIDARITY BUILDING ........................ 338
CROATIAN RHETORIC TO OPPOSE, POLITICISE AND SIMPLIFY ...................................................... 352
CONCLUSION .....................................................................................................................................365
Implications of the Research ...................................................................................................... 365
Identifying Gaps .......................................................................................................................... 370
Directions for Future Research: Grounds for Generalisability .................................................... 373
APPENDIX ...........................................................................................................................................377
Interview Questionnaires ............................................................................................................ 377
ENGLISH ................................................................................................................................... 377
FRENCH ................................................................................................................................... 381
Responses to the Role of Beliefs in the Genocide ..................................................................... 385
BIBLIOGRAPHY...................................................................................................................................387
viii
TABLE OF FIGURES AND TABLES
FIGURE 4.1: THE OPPORTUNITY, MOBILISING STRUCTURES, FRAMING FRAMEWORK.......62
TABLE 4.1: FACTORS AT PLAY IN FRAME RESONANCE............................................................68
FIGURE 4.2: A DYNAMIC, INTERACTIVE FRAMEWORK FOR ANALYSING MOBILISATION IN
CONTENTIOUS POLITICS...............................................................................................................71
FIGURE 5.1: A PROBABLE SEQUENCE OF ELITE-LED MOBILISATION.....................................83
FIGURE 5.2: NETWORKS AND ONTOLOGIES...............................................................................88
FIGURE 5.3: A NETWORK OF IDENTITIES....................................................................................90
FIGURE 5.4: SOCIAL AND PSYCHOLOGICAL DYNAMICS OF POPULAR MOBILISATION......113
TABLE APPENDIX.1: DISTRIBUTION OF ANSWERS OF BELIEFS.............................................384
ix
MAP OF RWANDA
Source: Department of Peacekeeping Operations, United Nations Organisation.
x
MAP OF THE FORMER YUGOSLAVIA
Source: Department of Peacekeeping Operations, United Nations Organisation.
xi
ABBREVIATIONS
AMASASU
Alliance de militaires agacés par les séculaires actes sournois des
Unaristes
APROSOMA
Association pour la promotion sociale de la masse
BBTG
Broad Based Transitional Government
CDR
Coalition pour la défense de la République
CND
Conseil national de développement
CSP
Conseil supérieur du pays
DEMOS
United Slovenian Opposition
EC
European Community
FAR
Forces armées rwandaises
GDP
Gross Domestic Product
HDZ
Croatian Democratic Union
IMF
International Monetary Fund
JNA
Yugoslav Federal Army
LCY
League of Communists of Yugoslavia
MDR
Mouvement démocratique républicain
MRND
Mouvement révolutionnaire national pour le développement
MRND(D)
Mouvement révolutionnaire national pour le développement et la
démocratie
NATO
North Atlantic Treaty Organisation
NMOG
Neutral Military Observer Group
OAU
Organisation of African Unity
Parmehutu
Parti du mouvement de l’émancipation hutu
PDC
Parti démocrate-chrétien
PL
Parti libéral
PSD
Parti social démocrate
RADER
Rassemblement démocratique rwandais
RANU
Rwandese Alliance for National Unity
RPF
Rwandese Patriotic Front
RTLM
Radio télévision libre des mille collines
xii
SANU
Serbian Academy of Arts and Sciences
SAP
Structural Adjustment Program
SCT
Self-Categorisation Theory
SDA
Party for Democratic Action
SDS
Serbian Democratic Party
SFRY
Socialist Federative Republic of Yugoslavia
SIT
Social Identity Theory
SMO
Social Movement Organisation
SPS
Socialist Party of Serbia
TNA
Transitional National Assembly
UN
United Nations
UNAMIR
United Nations Assistance Mission to Rwanda
UNAR
Union nationale rwandaise
VMRO-DPMNE
Internal Macedonian Revolutionary Organization-Democratic Party
for Macedonian National Unity
xiii
GLOSSARY
Akazu
‘The little house,’ referring to the privileged group of
individuals around Habyarimana and his wife
Animations
A communal gathering overseen by local authorities to praise
the state party during the Second Republic
Antemurale Christianitatis
Border of the Christian world
Astridiens
Group of educated colonial-era Tutsi named after the city of
Astrida (now Butare), where the Groupe scolaire d’Astrida,
one of the best schools in the country for elite natives, was
located
Banus
Traditional Croatian Chief Executive
Benesabahinzi
‘Sons of cultivators,’ a reference to the Hutu
Četniks
Serbian nationalist paramilitary force
Clan de Madame
Referring to the privileged group of individuals around
Habyarimana’s wife
Corvées
Forced labour imposed on Rwandese during colonisation
Gucupira
Name given to the social fall of a Tutsi to the status of Hutu
Kosovo Polje
Field of the Blackbirds (Kosovo)
Kuna
Croatian currency used during the Middle Ages and the
Second World War
Ibikingi
Land conceded by royal order
Ibyitso
Accomplice, term used during the civil war to refer to Tutsi
suspected of having ties with Rwandese Patriotic Front
Impuzamugambi
‘Those with the same goal’, name of a militia during the civil
war initially tied to the Coalition pour la défense de la
République
Interahamwe
‘Those who work together’, name of a militia during the civil
war
initially tied to the Mouvement révolutionnaire national pour
le développement
xiv
Inyenzi
‘Cockroach’, name given to the Tutsi exiles of the 1960s who
attacked Rwanda at night; used again during the civil war to
refer to the Rwandese Patriotic Front
Kwihutura
Name given to the social rise of a Hutu to the status of a Tutsi
Maspok
‘Croatian Spring’, Croatian nationalist revival in 1971-1972
Muhutu (Bahutu or Abahutu) Hutu in the singular (Hutu in the plural)
Mututsi (Batutsi or Abatutsi) Tutsi in the singular (Tutsi in the plural)
Mutwa (Batwa or Abatwa)
Twa in the singular (Twa in the plural)
Muyaga
Word signifying ‘windy’ or ‘hurricane’, used to describe the
events of 1959, before they became known as the 1959
Social Revolution
Mwami
Title or name given to the king
Načertanije
Serbian national liberation programme developed in the
nineteenth century by Ilija Garašanin.
Narod
Nation
Narodnost
Nationality
Pacta Conventa
Convention signed in 1102 to institute Hungarian control over
Croatian territory
Petit clergé
Lower echelons of the clergy
Rubanda Nyamwinshi
‘The majority people’, referring to the Hutu
Sabor
Traditional Croatian People’s Assembly
Šahovnica
Croatian emblem
Séminaristes
Group of Hutu who were educated in seminaries during the
colonial era
Ubuhake
Contractual subordination in traditional Rwanda where an
individual tied himself to a patron in exchange for the loan of
cattle
Uburetwa
Form of mandatory work owed to patrons of the land
developed around the nineteenth century
xv
Ubwoko
Word signifying ‘type’ or ‘kind’ generally taken to mean ‘clan’
Umuganda
Mandatory labour imposed on the population under the
Second Republic
Ustaša
Political and military Croat movement at the head of the
Independent Croatian State during the Second World War
*Some of the translations from Kinyarwanda are based on Gérard Prunier, The Rwanda Crisis:
History of a Genocide (New York: Columbia University Press, 1995), pp. 366-373.
xvi
PREFACE
A trip to a war museum or reading the retelling of a riot, crisis or war has always been a
disturbing experience in my eyes. Every time, I feel I am confronted with my own sense of
self and with a troubling question: had I been caught in such an event, how would I have
reacted?
These encounters, in museums, books and the media, with violent
circumstances raise the question of the role we each could and can play in similar
instances: victim, bystander, supporter, resistant, leader? In the end, my own reflection
always leads to me to reject a ‘holier-than-thou’ answer to this question. Comforting as this
answer would be, I refuse to believe I was born different from any of those caught in
unsettling circumstances, including the atrocities of war, whether leader or follower,
opponent or supporter. And I therefore can not pretend to be immune to reacting the way
some do, such as many, if not most, Germans under Hitler, or members of the Hutu
population in Rwanda in 1994. My dissertation project stems from this reflection. It is
fundamentally about trying to understand what moves us, what drive us to believe, to
stand, to join, as well as to mobilise, even for the unthinkable.
Between ‘normality’ and crisis, however, we often build a dividing line. But if one
accepts this distance, especially between us and the victims and between us and the
killers, what remains as explanations for crises and civil violence? Is it that these
individuals partake in these events purely because they are barbarians or because they
have reverted back to a state of premodernity? Or can entire populations lose all capacity
for rational judgment as they cross into insanity? Though they have been upheld, such
explanations prove unsatisfactory.
Individuals caught in such instances, whether the
civilian, supporter, soldier, etc., are not any different than the average human being. They
share the same biology, the same cognitive capabilities, and are born with the same
capacity for judgment.
They are, however, in many, if not most cases, people
xvii
unfortunately caught in different circumstances, political, economic, institutional, social,
cultural, etc. I do not believe anything intrinsically or cognitively sets us apart from the
people mobilised in instances of crisis or conflict except for dramatically different
contextual circumstances.
More importantly, we certainly are not immune to some of the manifestations of
discontent, resentment and anger ‘they’ display. Some of the sentiments expressed on the
ground during political or civil crises also find their way into the rhetoric and opinions of
people in stable industrial nations. 1 ‘We’ are not exceptions to feeling like ‘they’ do; we
‘think’, ‘feel’ and ‘react’ in much the same way.
In a relatively peaceful and ordered
society, types of feelings such as fear, resentment and anger, including sentiments
directed against an ‘other,’ can still form the basis for some of the reasoning in stable
societies. And these can be the polarised perceptions that mobilise a group to demand
change, demonstrate, riot and even act violently against others.
One need not look at the most extreme cases of communal violence to find
examples of mobilisation. My society, the Québécois society, can illustrate this. Québec is
not an example of extreme polarisation leading to violence. Not many incidents of intergroup violence have occurred in Québec in the last decades. Opinions and perceptions
have, however, at times been quite polarised.
One such instance was the 1995
referendum on independence. 2 Although much of the fighting took place among political
elites, during the time leading up to the vote, many Québécois did experience conflict on a
day-to-day basis among inner circles of coworkers, friends and family. The rift between
political camps seemed reproduced on the ground, disturbing social milieus.
1
There have also been tragic examples of extreme group based violence in the West. The Holocaust is
certainly one of the worst examples of this.
2
Although the 1995 referendum in Québec can illustrate how collective identities become polarised or can be
intensified, it is important to remember that the context itself in 1995 played an important role in this
phenomenon. The referendum called for an either/or positioning on the part of citizens. Contexts rarely so
clearly demand to choose from antithetical positions.
xviii
Political identities, or whether one was for or against sovereignty, had following the
1980 referendum only surfaced occasionally and innocuously for most. During the months
leading up to the 1995 vote, however, these political convictions often became the
predominant identity marker in social encounters. It seemed as though throughout the
province social ties were strained by the emergence of two opposite camps of fervent
advocates. Each camp embodied political positions built around what had been heard or
read, reproducing the official positions, platforms and rhetoric.
What ended being
reproduced among the general population was the equivalent of two crystallized as well as
irreconcilable political positions. Gray areas along the federalist-sovereignist spectrum as
well as more neutral convictions had all but disappeared.
It seemed that only two
plausible positions remained, for or against independence, implying that each individual
was left to be categorised as ally or opponent. It was a fascinating but frightening thing to
observe: within months, political identities innocuously carried for years had become the
most prominent form of identification and had deeply divided families and friends, not to
mention an entire population.
Some might disagree with my depiction of events around the 1995 referendum. It
is, however, my perception of what occurred, my interpretation. And, it is what, at the time,
moved me into one of the camps. Circumstances, rhetoric, actors, and how I interpreted
all of these, convinced me to mobilise. If this says anything, in my mind, it is that changes
in perceptions, more potent than many would acknowledge, can drive people.
If there is an underlying message to be found in this research, it is that outcomes,
i.e. the degree of violence, the number of deaths, the amount of destruction, matter. But
these are not all there is to crisis, conflict and violence. There are also people, people who
change their minds and perceptions and mobilise. And, unleashed, they can become, in
and of themselves, in extreme cases, weapons of mass destruction, as events around the
world remind us. Understanding how populations reach that point should matter as much
xix
as the study of other factors and actors. It is a key to understand the outcomes
themselves. Understanding these changes is what I try to do in this dissertation.
xx
INTRODUCTION
Words are powerful. In the hands of cunning leaders, they are tools to sway and convince.
From Aristotle’s Rhetoric to Machiavelli’s Prince, from modern times’ crowd psychology to
the psychology of leaders, accounts abound of how words can be made the way to the
‘hearts and minds’ of populations and can serve as recipes to successfully achieve their
mobilisation. According to Charles Tilly, Aristotle devised four fundamental rules to use
rhetoric to win over supporters: “(1) make the audience well-disposed towards yourself and
ill-disposed towards your opponent; (2) magnify or minimize the leading facts; (3) excite
the required state of emotion in your hearers; (4) refresh their memories.” 1 Even if
predating modern social sciences and psychology by millennia, Aristotle’s advice is eerily
too familiar in a world of populist leaders and authoritarian demagogues: know and
understand your public, and play on its character and emotional state. Drawing on modern
literature on leaders, populations and mobilisation, this research develops its own take on
rhetorical strategies employed by elites to win popular support.
This research examines popular mobilisation during crises and conflicts in plural or
divided societies, specifically in instances of elite-led mobilisation. 2 It is an effort to shed
1
Charles Tilly, “Rhetoric, Social History, and Contentious Politics: Reply to Critics,” International Review of
Social History, vol. 49, no. 1, 2004, pp. 132-133.
2
Although the term ‘mobilisation’ is very common in social science literature, particularly in studies on conflicts
and social movements, very few authors provide a clear definition of it. Webster’s Ninth NewCollegiate
Dictionary defines mobilisation as “the act of mobilizing” or “the state of being mobilized.” Mobilising, in turn, in
relation to conflict, is defined as “to assemble and make ready for war duty” or “to marshal (as resources) for
action.” Webster’s Ninth New Collegiate Dictionary (Springfield, Mass.: Merriam-Webster Publishers 1988), p.
762. One of the few definitions of mobilisation provided by a social scientist is Eduardo Canel’s. He defines
mobilisation as “the process by which a group assembles resources (material and non-material) and places
them under collective control for the explicit purpose of pursuing the group’s interest through collective action.
But mobilization is more than resource accumulation; for mobilization to take place, these resources must be
placed under collective control and must be employed for the purpose of pursuing group goals.” Eduardo
Canel, “New Social Movement Theory and Resource Mobilization Theory: The Need for Integration,” in Michael
Kaufman and Haroldo Dilla Alfonso, eds., Community Power and Grassroots Democracy: The Transformation
of Social Life (Ottawa: IDRC/Zed, 1997), p. 207. In the context of this research, mobilisation will therefore be
understood as the process of assembling collectively to make ready for action in the pursuit of collective goals.
The terminology most often employed will be, however, ‘popular mobilisation’ to differentiate the mobilisation of
a population in the midst of a crisis, the central focus of the analysis, from the mobilisation of professional or
conscripted soldiers in war. The term ‘elite-led’ is borrowed from Stuart J. Kaufman, Modern Hatreds: The
Symbolic Politics of Ethnic War (London: Cornell University Press, 2001), p. 12. For more on mobilisation and
demobilisation, see Charles Tilly and Sidney Tarrow, Contentious Politics (Boulder, Colo.: Paradigm
Publishers, 2007). The concept of collective identity, on the other hand, as employed in this analysis, draws on
1
new light on one of the most puzzling questions surrounding mobilisation: Why do publics
answer their leaders’ calls and agree to follow them, especially against another group?
This seemingly simple question is a perplexing one, if not one of the most perplexing
participation or collective action problems.
Beyond the issue of personal egotism,
participation in such instances–at which time leaders can make extraordinary demands on
their citizens such as partaking in violence–often brings quite uncertain or diffuse benefits
to individuals. Instead, it frequently comes at great costs and risks for them, their family,
their entourage, and their community writ large.
In order to address the issue of elite-driven popular mobilisation, this research
focuses on collective identity. Collective identity is at the heart of the matter because it is
shared by leaders and followers alike: it is the basic bond that cements the group. When
made prominent enough, collective identity can prove to be a convincing rationale to
mobilise individuals in the name of the group. Accordingly, two separate but related issues
are at the heart of this study: elites-population relations, or social dynamics behind crises
or conflict, and collective identity, or the social psychological bond between members of a
group.
The focus on these two issues comes as a response to existing literature on
political and violent crises.
Analyses on conflicts have, rightly, explored background
the definition developed by Francesca Polletta and James M. Jasper. They “have defined collective identity
as an individual’s cognitive, moral, and emotional connection with a broader community, category, practice, or
institution. It is a perception of a shared status or relation, which may be imagined rather than experienced
directly, and it is distinct from personal identities, although it may form part of a personal identity. A collective
identity may have been first constructed by outsiders [...], who may still enforce it, but it depends on some
acceptance by those to whom it is applied. Collective identities are expressed in cultural materials–names,
narratives, symbols, verbal styles, rituals, clothing, and so on–but not all cultural materials express collective
identities. Francesca Polletta and James M. Jasper, “Collective Identity and Social Movements,” Annual
Review of Sociology, vol. 27, 2001, p. 285. A last concept to define since it constitutes one of the predominant
collective identities called upon by elites is ethnicity. For James Fearon and David Laitin, ethnicity is “defined
mainly by descent rules of group membership and content typically composed of cultural attributes, such as
religion, language, customs and shared historical myths.” James D. Fearon and David D. Laitin, “Violence and
the Social Construction of Ethnic Identity,” International Organization, vol. 54, no. 4, 2000, p. 848. In his
definition, Anthony Smith insists on solidarity that exists between members of a particular ethnic identity that
serves to unite them in what he calls the ‘ethnie.’ This solidarity is fundamental to the constitution ethnic ties: it
is the common acknowledgement that specific shared cultural attributes are the basis for a common identity.
See Anthony D. Smith, “The Ethnic Sources of Nationalism,” Survival, vol. 35, no. 1, 1993, p. 49.
2
variables and determinants of conflict and violence. Many have pointed to the role played
by political, economic, societal, environmental, and even at times psychological, factors in
triggering a crisis. 3 But if these are necessary conditions, they are not, in turn, sufficient.
Not all instances of dramatic change in these factors lead to crisis or conflict. Crises and
conflicts require engagement on the part of people in affected societies. People need to
feel these factors are reason enough to stand up and/or fight back. Crisis and conflict are
first born in the minds of those who live these conditions and who then choose to act upon
them. Social and psychological factors are as much part of the crisis equation as other
determinants. They are the filter that catalyses response to other variables.
Leaders and Followers: Social Dynamics
This research focuses on the actions of elites. Many authors studying instances of conflict
have underlined the role played by elites in triggering the violence.
In situations of
instability, elites “can produce rapid and profound polarization within a multi-ethnic
society.” 4 The Carnegie Commission on Preventing Deadly Conflict asserted that war and
violence are often the result of “deliberate calculations and decisions.” 5
It is these
calculations and decisions, it is elite strategies, that this research explores.
Without falling prey to an overly simplistic account of what elites have to gain from
instability and crisis, it is important to remember that the politics of leadership is a game in
which few can play. It remains very much a hierarchicased world in which few positions
exist and in which resources are sparsely distributed. More importantly, it remains a world
3
For a review of factors behind conflict, see for example Michael E. Brown ed. International Dimensions of
Internal Conflict (Cambridge: MIT Press, 1996), or Stephen Van Evera, “Hypotheses on Nationalism and War,”
in Sean M. Lynn-Jones and Steven E. Miller eds., Global Dangers: Changing Dimensions of International
Security (Cambridge: MIT Press, 1995). Incidentally, these authors also discuss to the role of elites and social
factors in mobilisation.
4
David A. Lake and Donald Rothschild, “Containing Fear: The Origins and Management of Ethnic Conflict,”
International Security, vol. 21, no. 2, 1996, p. 53.
5
Carnegie Commission on Preventing Deadly Conflict, Preventing Deadly Conflict (Washington, D.C.:
Carnegie Corporation of New York, 1997), p. 3
3
in which power is the prerogative of a select few. Times of turmoil amount to a shuffle of
the cards for this hierarchy. They tend to reconfigure the established balance of power as
well as change the range of opportunities available to elites. Instability, political changes
and transitions are turning points, and as such become moments of heightened
competition for elites to assert their existing hold on power or to benefit from new
opportunities. Many strategies exist to prevail in such a competition. Most tend to involve
a show of strength and resources. Popular support is the source of both: it is strength in
terms of numbers and in terms of access to resources, both material and human.
Supporters are, just as importantly, the source of influence and power. Keenly aware of
this, elites often make these moments of instability a competition for the allegiance of
supporters. 6
Collective Backgrounds and Identity: Psychological Dynamics
In determining ‘why followers follow,’ however, many have looked at the manipulation of
collective identity on the part of elites. 7 Various authors across disciplines have noted that
collective identity, “[a]s a political resource [...] is a potent rallying cry for politicians [and
elites more generally]–far more potent than, for example, class.” 8 Elites during times of
crisis transform themselves into what some call ‘ethnic activists,’ 9 ‘political,’ 10 ‘cultural,’ 11 or
‘ethnic entrepreneurs,’ 12 and even ‘social and political archeologists’ because of their
strategic “rediscovery and reinterpretation of the ethnic past [and symbols] and through it
6
V. P. Gagnon Jr., “Ethnic Nationalism and International Conflict: The Case of Serbia,” International Security,
vol. 19, no. 3, 1994/1995, p. 140. See also James D. Fearon and David D. Laitin, “Violence and the Social
Construction of Ethnic Identity,” p. 846.
7
Donald L. Horowitz, Ethnic Groups in Conflict (Berkeley, Ca.: University of California Press, 1985), p. 140.
8
Parentheses added. David Welsh, “Domestic Politics and Ethnic Conflict,” Survival, vol. 35, no. 1, 1993, p.
66. See also, for example, Anthony D. Smith, “The Ethnic Sources of Nationalism,” pp. 48-62.
9
David A. Lake and Donald Rothschild, “Containing Fear,” p. 41
10
Ibid.
11
Crawford Young, The Politics of Cultural Pluralism (Madison, Wisc.: The University of Wisconsin Press,
1976), p. 45.
12
Steven Majstorovic, “Ancient Hatreds or Elite Manipulation? Memory and Politics in the Former Yugoslavia,”
World Affairs, vol. 159, no. 4, 1997, p. 170.
4
the regeneration of their national community,” in their attempts to mobilise their
population. 13
Why this potency of collective identity? By drawing on this existing bond between
themselves and the population, elites are building on strong bases. Their membership in
the group means they are already integrally part of the collective project the ‘we’ forms and
have a better understanding than outsiders of what this ‘we’ represents. Furthermore,
their inclusion in the group means that they are working with an implicit pre-qualification:
relationships of commonality and trust between members of the group. By belonging to the
group, as insiders, it is easier for their constituents to identify with them and to accept them
as legitimate representatives. In all appearances, not only are they speaking for the group
but with the group.
The nature of collective identity, and particularly the cognitive mechanisms that
underlie it, also make it a potent resource for elites in attempts to win over supporters.
Categorisation, specifically, is a fundamental mechanism behind identity. A recurrent
pattern for humans, this and other cognitive mechanisms are like a toolbox of strategies
humans equip themselves with in order to cope with their social universe. Cognitive
mechanisms organise thoughts, perceptions and in turn responses. In a complex social
world, they are shortcuts for processing all of the data individuals are constantly
bombarded with. They are akin to roadmaps to the social world, roadmaps for which
categories play the role of signposts and markers.
Social boundaries and social categories are an important part of these cognitive
shortcuts. They order human interactions by defining one’s place in the social universe
along with everyone else’s place in it. To save people from the complex interactions an
undifferentiated social world implies, categories constitute a base of information about the
13
Referring to Anthony Smith, Steven Majstorovic, “Ancient Hatreds and Elite Manipulation?,” p. 172.
5
different people individuals encounter and in turn suggest certain appropriate attitudes and
behaviours.
Although people generally grow up in social environments which contain categories
set by others before, the process of categorisation nonetheless remains at the heart of
constructing new or evolving collective identities.
Imagining an undifferentiated social
place as the starting point for human interactions illustrates this. What would occur? How
would we know who we are? Or even what we are? If individuals come into the social
world devoid of any innate referents, they are not equipped to find these answers within
themselves. 14 Without internal reference points, humans only have what is beyond them
as a source of answers. In the case of social understandings and behaviour, it is others
that serve as references to build our understandings of the world. To know who we are,
we look at who they are. Individuals observe, notice similarities and differences, through
these develop schemas for comparison, compare, and through comparison differentiate.
We determine the extension of this differentiation: how it extends to individuals, who are
14
Some research in cognitive anthropology points to the existence of forms of innate patterns of reasoning.
Folkbiology, for example, “has revealed some apparently pancultural features of human thought about the
animate realm [patterns of reasoning about the living world]. Although classifications of living things are
culturally specific, the form of these classifications is the same everywhere. Organisms are classified
hierarchically, with five distinctive taxonomic levels: folk kingdom (e.g. plant, animal); life form (e.g. tree,
mammal), generic species (e.g. oak, dog); folk specific (e.g. white oak, poodle); folk varietal (e.g. spotted white
oak, toy poodle).” In Paul E. Griffiths, “What is Innateness,” The Monist, vol. 85, no. 1, 2002, p. 76. For more
on such cognitive anthropological research, see for example Scott Atran, Cognitive Foundations of Natural
History: Towards an Anthropology of Science (Cambridge, Mass.: Cambridge University Press, 1990) or
Douglas L. Medin and Scott Atran, eds., Folkbiology (Cambridge, Mass.: MIT Press, 1999). Starting from a
completely different point, the mind/body debate in philosophy, a strand of phenomenology asserts the
centrality of the body in cognisance. Maurice Merleau-Ponty, for example, sees the body as the conduit for all
more complex understandings. Beyond thought, embodied experience of the world is the source of meaning
more basic and primordial than socially determined understandings. Merleau-Ponty referred to this as ‘sense
experience’, le sentir. As Merleau-Ponty explained, “[v]ision is already inhabited by a meaning (sens) which
gives it a function in the spectacle of the world and in our existence. The pure quale would be given to us only
if the world were a spectacle and one’s own body a mechanism with which some impartial mind made itself
acquainted. Sense experience, on the other hand, invests the quality with vital value, grasping it first in its
meaning for us, for that heavy mass which is our body, whence it comes about that it always involves a
reference to the body. [...] Sense experience is that vital communication with the world which makes it present
as a familiar setting of our life.” (pp. 60-61) As he further developed, “[t]he sensible configuration of an object
or a gesture [...] is not grasped in some inexpressible coincidence, it is ‘understood’ through a sort of act of
appropriation which we all experience when we say that we have ‘found’ the rabbit in the foliage of a puzzle, or
that we have ‘caught’ a slight gesture. Once the prejudice of sensation has been banished, a face, a signature,
a form of behaviour cease to be mere ‘visual data’ whose psychological meaning is to be sought in our inner
experience, and the mental life of others becomes an immediate object, a whole charged with immanent
meaning.” (p. 67). Maurice Merleau-Ponty, translated by Colin Smith, Phenomenology of Perception (New
York: Routledge, 2003).
6
the referents for the classes of individuals we develop. We thus delimit boundaries both for
inclusion into and exclusion from the ‘we’. Without a relational form of categorisation,
there is no ‘this’ or ‘that’, ‘us’ or ‘them’.
To take collective categories and identities as emerging from this social interactive
process has implications.
As artifacts of social interactions, they derive from social
interpretations. To think of oneself as Canadian, Chinese or Congolese, for example, is a
social convention resting on a tacit habit of socially dividing our political sphere into entities
called countries rather than a ‘natural’ fact. 15 While all humans innately categorise, they do
not categorise the same way. Social upbringing and social contacts, as well as cultural
models and education, inconspicuously consolidate these categories into social identities:
they deepen collective understandings by layering their boundaries with meaning, by
providing their content, both in terms of non-normative characteristics, descriptive or
factual statements about identities (for example, physical traits such as short, blond, dark
skinned, etc.), and of normative connotations, value judgments about groups (such as
hardworking, trustworthy, etc.). Once they have been set, their content continues to
change as a result of exogenous input, i.e. through social encounters with others, and also
endogenously as the group itself changes.
We need not attribute to all social categories and identities the same importance.
Indeed, with the number of social identities individuals possess, it would seem impossible
to think each of them is equally important at all times. How people feel about their
identities, or prioritise their social attachments, changes. The intensity of one’s feelings of
belonging to groups varies. At times, certain identities may come to predominate, calling
up strong emotional ties, but at others, they might recede or be superseded by other
15
One of the most prominent proponents of this argument is undoubtedly Benedict Anderson. See Benedict
Anderson, Imagined Communities: Reflection on the Origin and the Spread of Nationalism (New York: Verso
Press, 1996). This is not to deny the existence of biological determinants of identity. Nonetheless, social
interpretations graft themselves onto biological identities and often supersede their biological determinacy, for
instance, the sex/gender difference.
7
collective attachments.
And, though not always the case, shifts towards stronger
emotional bonds often open the door to stronger, possibly more emotive, responses to
one’s collective identity.
When elites mould collective identity
In light of their nature, collective identities and attachments are malleable, even reenforceable, structures. As a result, they can be the medium through which elites
communicate to the population their need for support and the need for mobilisation. As
identity perceptions emerge from slow processes of internalisation, elites can generally
only hope to build upon the internalised conceptions of identity existing within their groups.
In that case, their success at mobilising people depends on how adept they are at
recycling existing shared cultural referents and using them in appeals to garner support
from their constituents. Cunning elites, with occasional success, can also work to inject
new directions or senses into existing social understandings so that these better serve
their plans. They recast existing collective referents, allowing these referents to legitimate
their plans.
Whichever approach they adopt, there is no doubt that both are powerful
strategies. In the name of the group, in the name of what it represents, appeals can be
made extremely compelling.
A common culture, a shared understanding of collective
history, perceptions of what the norms are for membership in the group or exclusion from
the group, of relations with other groups, can all be used as the building blocks to translate
popular sentiments and elite political projects into justifications for collective action. When
filtered through common identity referents, the language, the images and ideas they call
upon are so close to what individuals know, so consistent with their social reality that they
can seem commonsensical, justifiable and fitting. They are potent material to recall group
solidarity, a force that can trigger collective action.
8
Some leaders succeed in mobilising a population, others try and fail. As varied
outcomes for mobilisation efforts indicate, there is nothing automatic about mobilisation.
Much rests on elites’ ability to convince a population to join in the collective effort. To
better understand what it is about some mobilisation appeals that make them successful,
this research delves further into the social and psychological dynamics of elite-led
mobilisation. It is on some of the mechanisms behind the construction of these appeals,
particularly mechanisms to (re)create a sense of group solidarity, namely on opposition,
politicisation and simplification, that this dissertation focuses.
Outline of Chapters
This research deals with identity, culture and social backgrounds. It studies how they can
be rediscovered and recycled–changed to a certain extent–to serve the aims of
malintentioned leaders. If it suffers from any conscious bias, it is that it unabashedly
adopts a social constructivist epistemology for its stance on identity, culture and social
backgrounds. As the starting point of this dissertation, social constructivism is discussed in
chapter 1. This sets the epistemological underpinning of the analysis, as well as its
conception of identity. The rest of the research logically follows from this premise. At the
heart of this project lies the effort to integrate psychological and sociological insights into a
synthetic understanding of elite-led mobilisation. To attain this objective, a number of
theoretical bodies were surveyed. Three, in particular, are reviewed: crowd psychology,
the precursor to contemporary perspectives on collective behaviour (chapter 2), social
psychology (chapter 3), and social movements theory and contentious politics (chapter 4).
These various theoretical approaches are the building blocks for the framework
expounded in chapter 5. Chapter 5 begins by developing an original theoretical sequence
of elite-led mobilisation. Inspired from existing literature, particularly social movement
theory and contentious politics, this sequence serves as the foundation for the theoretical
9
framework developed for this dissertation. This sequence established the research turns to
an understudied phenomenon at the core of elite-led mobilisation processes: elite
strategies to draw support and supporters from a mixed population. Three in particular,
opposition, politicisation and simplification, are developed and discussed at length in
chapter 5.
In order to assess the performance of this theoretical framework, the rest of the
research consists in case studies. Following a chapter on methodology and case study
research design (chapter 6), three chapters are devoted to contemporary Rwanda. First,
chapters 7 and 8 recount two recent periods of Rwandese history, 1959-1990, from the
first steps toward Hutu control and independence to the unraveling of the Second Hutu
Republic, and 1990-1994, from the civil war with the Rwandese Patriotic Front to the
genocide. Beyond familiarising the reader with the case, these serve three purposes: (a) to
establish what existed in terms of beliefs and understandings of groups and group
relations in Rwanda at the time (chapter 7); (b) to establish how these were used by
political entrepreneurs as the basis for the mobilisation, and, more importantly, to
ascertain–which is almost never done in studies of this kind–whether the population
effectively accepted and internalised these appeals and the view of intergroup relations
they conveyed (chapter 8); and (c) to illustrate the sequence of mobilisation as developed
in the theoretical framework (chapter 8). A final chapter on Rwanda is devoted to
illustrating how the three mobilisation strategies studied, opposition, politicisation and
simplification, played out in the context of the 1959 Hutu Revolution and its consolidation
in following years, as well as in the lead up to the 1994 genocide (chapter 9). To further
check on the logic of the theoretical framework and to suggest the plausibility of
generalising this framework, chapters 10 and 11 turn to another case, the break-up of
Yugoslavia and ensuing wars. Chapter 10 summarises the trends and events which led up
to the disintegration of the former Yugoslavia and to conflicts in Slovenia, Croatia and
10
Bosnia-Herzegovina. Chapter 11 pertains to the impact of Yugoslav elites’ communication
strategies by (a) discussing whether the Yugoslav population internalised appeals by
political entrepreneurs or not and (b) studying more specifically the strategies employed by
ethno-centric entrepreneurs in the former Yugoslavia. The dissertation concludes with a
summary, a discussion of its findings and an assessment of future applications of its
framework.
11
CHAPTER 1: SOCIAL CONSTRUCTIVISM AND IDENTITY: A PREMISE TO THE
ARGUMENT
Neta Crawford insightfully stated that “no one theoretical approach will likely be able to
account for the complex relationships between experience, perception, cognition, culture,
and biology.” 1 Studying the social world is daunting task. It is rife with phenomena that are
the result of interlocking, interacting factors and variables that take on various meanings or
are interpreted differently across contexts and time. That being said, in the absence of a
perfect theory of social phenomena, some approaches have been developed with these
intricacies in mind and better reflect this complexity.
One of these is social constructivism, which focuses on the roles of ideational
structures and material factors in the formation of the social world. No novelty in the social
sciences, social constructivist thought can, to a certain extent, be traced back to the
philosophical idealism of Berkeley and Kant. 2
Some of its sources can also be found in
early work in sociology, which adopted a qualified form of idealism. 3
With its tenets
developed across disciplines and over time, social constructivism therefore became an
approach more than a bounded paradigm, but a powerful approach, nonetheless, for the
study of social phenomena. Thus, not quite a theoretical revolution, social constructivism
represents belated turn to sociological approaches on the part of various social science
disciplines.
1
Neta Crawford quoted in Jennifer Sterling-Folker, “Realism and the Constructivist Challenge: Rejecting,
Reconstruction, or Rereading,” International Studies Review, vol. 4, no. 1, 2002, p. 97.
2
Gerard Delanty, Social Science: Beyond Constructivism and Realism (Minneapolis: University of Minnesota
Press, 1997), p. 113. Neither Berkeley, nor Kant, referred to social constructivism. Some of these authors’
ideas, however, inspired later thinkers who worked out the principles of social constructivism.
3
In a famous quote, Max Weber states: “[n]ot ideas, but material and ideal interests, directly govern men's
conduct. Yet very frequently the ‘world images’ that have been created by ‘ideas’ have, like switchmen,
determined the tracks along which action has been pushed by the dynamic of interest.” See Max Weber, The
Protestant Ethic and the Spirit of Capitalism (New York: Scribner's, 1958 [1904-1905]) p. 280. Reviewing
Emile Durkheim’s work, Jack Snyder writes: “Emile Durkheim believed that “social life, in all its aspects and in
every period of its history, is made possible only by a vast symbolism.” However, he also believed that the form
these symbols take depends in part on aspects of the material setting, reflected in the division of labour in
society, and that the ‘collective consciousness’ of a group mirrors its social relationships.” See Jack Snyder,
“Anarchy and Culture: Insights from the Anthropology of War,” International Organization, vol. 56, no. 1, 2002,
p. 32.
12
Belated, indeed, but also timely. Unlike some of its theoretical contenders, social
constructivism adopts an open ontology that “does not include or exclude any particular
variables as meaningful.” 4 It also constitutes an accommodating epistemology. 5 Social
constructivists are found across the positivist-post positivist divide. Furthermore, its flexible
methodological approach does not impose overly rigid methodological rules, but on the
contrary it accommodates more traditional research methods, as well as alternative ones. 6
The following section focuses on social constructivism as developed in
International Relations and then turns to how it has been used to study collective identity
and its role in intergroup violence.
It should be remembered, however, that social
constructivist approaches vary greatly and are the source of a number of disagreements. 7
4
Ted Hopf, “The Promise of Constructivism in International Relations Theory,” International Security, vol. 23,
no. 1, 1998, p. 194.
5
Since it is more of an approach than a clearly bounded paradigm, social constructivism does not impose an
epistemology per se. As a result, social constructivist proponents fall on both sides of the positivist-postpositivist divide. A positivist epistemology, adopted by most mainstream IR authors, posits the existence of an
observable empirical reality (truth as correspondence), the methodological unity of science, and the value-free
nature of scientific knowledge. See Mark Neufeld, The Restructuring of International Relations Theory
(Cambridge, Mass.: Cambridge University Press, 1995), pp. 33-38. Social constructivists with post-positivist
allegiances generally employ interpretive approaches that claim that the social realm cannot be studied exactly
like hard science. Human consciousness, hardly amenable to first degree observation, needs to be factored in
to analyses. As Martin Hollis and Steve Smith explain: “[t]o locate the idea in an interpretative or hermeneutic
setting, we need to specify that there is meaning both in ‘the behaviour of others’ and in the ‘account’ which the
acting individual takes of it. That leads to the central hermeneutic [or interpretive] theme that action must
always be understood from within.” Emphasis in original. Martin Hollis and Steve Smith, Explaining and
Understanding International Relations (Oxford: Clarendon Paperbacks, 1990), p. 72.
6
Jeffrey Checkel in his assessment of the ‘constructivist turn’ in International Relations Theory states that: “…
constructivists do not reject science or causal explanations.” Jeffrey T. Checkel, “The Constructivist Turn in
International Relations Theory,” World Politics, vol. 50, 1998, p. 327. There does exist factions among
proponents of this approach: some contend that they can develop causal explanations through social
constructivist frameworks, while others of a more critical bent contradict this and believe that only constitutive
frameworks are possible. On the difference between causal and constitutive effects see fn. 29. Alex Macleod
differentiates, for example, between what he calls ‘dominant constructivism’ and ‘critical constructivism’.
Proponents of the first have tried to bridge the divide between positivists and post-positivists by explaining the
effects of ideational factors in international relations. They therefore believe they can develop causal
explanations of the social world. Critical constructivists are skeptical of this move on the part of dominant
constructivism which they judge to be overly conservative, and seek to break with scientific or causal
approaches. They focus instead on studying constitutive processes behind ideational structures. See Alex
Macleod, “Les études de sécurité: du constructivisme dominant au constructivisme critique,” Cultures et
conflits, no. 54, 2004, pp. 13-51. For a different categorisation of strands of social constructivism in
International Relations, see also Richard Price and Christian Reus-Smit, “Dangerous Liaisons? Critical
International Theory and Constructivism,” European Journal of International Relations, vol. 4, no. 3, 1998, p.
259-294.
7
The main divide is among mainstream social constructivists intent on scientifically studying social
constructions and critical social constructivists striving to bring attention to reified and obscured power
relations.This is only one form of categorisation of social constructivist proponents in IR. It therefore
represents more of an ideal-type than a true reflection of groups or trends. Other authors have proposed other
categorisation schemes on the basis of different criteria. See for example Alex Macleod, “Les études de
13
Instead of introducing each stance separately or taking a position in favour of one or the
other, the following review adopts a pragmatic approach: while broad concepts and
notions shared by all constructivists are addressed, the intent is to focus on certain specific
notions that underpin this dissertation.
Social Constructivism in International Relations
When it was introduced in International Relations Theory, social constructivism challenged
the materialism and structuralism of dominant approaches. Rejecting the narrow focus on
material factors as determinants of international relations, and positioning themselves on
the side of idealism, proponents of social constructivism called for ‘bringing ideas back in’
to analyses of the international sphere and the social world. 8 For social constructivists,
ideas or ideational structures are fundamental to our understanding of social relations and
the social world. They are not epiphenomenal, but have causal and/or constitutive effects. 9
Steering clear of absolute idealism, however, most constructivist work opts for a qualified
idealism in which ideational structures remain embedded in the material world.” 10 For
sécurité,” pp. 13-51, Richard Price and Christian Reus-Smit, “Dangerous Liaisons?” or John Gerard Ruggie,
“What Makes the World Hang Together? Neo-Utilitarianism and the Social Constructivist Challenge,”
International Organization, vol. 52, vol. 4, 1998, pp. 855-885.
8
Alexander Wendt, Social Theory of International Politics (New York: Cambridge University Press, 1999), p. 1.
9
Social constructivists differ on whether they treat ideational factors as having causal or constitutive effects or
both. Alexander Wendt described the difference between the two by stating that: “[i]n a causal relationship an
antecedent condition X generates an effect Y. This assumes that X is temporally prior to and thus exists
independently of Y. In a constitutive relationship X is what it is in virtue of its relation to Y. X presupposes Y,
and as such there is no temporal disjunction; their relationship is necessary rather than contingent.” See
Alexander Wendt, Social Theory of International Relations, p. 25 and pp. 77-88. John Gerard Ruggie further
described constitutive rules as defining “the set of practices that make up a particular class of consciously
organized social activity─that is to say, they specify what counts as that activity.” Emphasis in original. John
Gerard Ruggie, “What Makes the World Hang Together?,” p. 871. Ruggie used the example of chess, provided
by John Searle, to illustrate this: “the rules of chess create the very possibility of playing chess. The rules are
constitutive of chess in the sense that playing chess is constituted in part by acting in accord with the rules.
Ibid. On ideas as causes, see, for example, Craig Parson, “Showing Ideas as Causes: The Origins of the
European Union,” International Organization, vol. 56, no. 1, 2002, pp. 47-84.
10
Paul Kowert and Jeffrey Legro, “Norms, Identity, and their Limits: A Theoretical Reprise,” in Peter J.
Katzenstein, ed., The Culture of National Security: Norms and Identity in World Politics (New York: Columbia
University Press, 1996), p. 491
14
social constructivists, ideas do not exist in a ‘material vacuum’ since material factors also
have a determinant and constraining effect on ideational structures. 11
Social constructivism also asserts that identities, interests and other ideational
structures are not given, but on the contrary are, to paraphrase Alexander Wendt, ‘what
people make of them,’ or constructed. 12 As social phenomena, such constitutive ideas and
ideational structures are not arbitrary. As Karen Fierke argued, “[w]e are not free to give
objects or actions any meaning we like [as individuals].
[...] Instead, meaning is
intersubjective and social.” 13 Although individuals carry or embody these ideational
structures as human agents, it is because they collectively agree on their meaning and
collectively share them that they exist as social structures. 14
Because they are intersubjective, ideational structures are more enduring than
individual thoughts and opinions. Emanuel Adler explained that “they are reified
structures.” 15 Once established, ideational structures form the collectively agreed upon
background of individuals and groups. The latter, in turn, behave according to their culture
and society, unconsciously reproducing the structures in place. The process obscures
11
Social constructivism should not be conflated with postmodernism on this issue. Social constructivism, or
social constructionism in sociology, maintains the existence of two realities: the social reality and the physical
reality. Because the former is the result of human action, it cannot be known it the same manner the physical
world can: it must be understood through the meanings individuals ascribe to it. Both the physical and the
social reality are, however, very ‘real’ for social constructivists. Postmodernism, on the other hand, blurs the
lines between these two realities. The physical realm can never be known outside of the social realm. Mostly
focused on knowledge, postmodernism maintains that “knowledges are fully constitutive social processes
rather than dependent reflections of an independent real... Like other social processes, knowledges differ from
each other in the ways in which they are constituted and in their social effects, but they cannot be ranked
hierarchically on the basis of their closeness to or distance from a singular objective or unchanging ‘reality’.”
Graham quoted in David Byrne, Complexity Theory and the Social Sciences: An Introduction (New York:
Routledge, 1998), p. 43.
12
Alexander Wendt, “Anarchy Is What States Make of It: The Social Construction of Power Politics,”
International Organization, vol. 46, no. 2, 1992, pp. 391-425.
13
Emphasis in original. Karen M. Fierke, “Links across the Abyss: Language and Logic in International
Relations, International Studies Quarterly, vol. 46, 2002, p. 346. See also Stefano Guzzini, “A Reconstruction
of Constructivism in International Relations,” European Journal of International Relations, vol. 6, no. 2, 2000, p.
164.
14
As John Gerard Ruggie explained, this intersubjectivity gives rise to “social facts, or facts that, in the words
of the linguistic philosopher John Searle, depend on human agreement that they exist and typically require
human institutions for their existence. Social facts include money, property rights, sovereignty, marriage,
football, and Valentine’s Day, in contrast to such brute observational facts as rivers, mountains, population
size, bombs, bullets, and gravity, which exist whether or not there is agreement that they do.” See John Gerard
Ruggie, “What Makes the World Hang Together?”, p. 856.
15
Emanuel Adler, “Seizing the Middle Ground: Constructivism in World Politics,” European Journal of
International Relations, vol. 3, no. 3, 1997, p. 322.
15
their social origin. 16 These structures are come to be seen as preordained and become
relatively stable components of the social world.
Social constructivism also reacted to the prominence of structuralism. Like
structuralists, social constructivists believe that structures, ideational ones in their case,
are a fundamental variable of international relations.
These ideational structures,
however, not only constrain and guide behaviour in the international sphere, but play an
essential role in the constitution of actors and the social world. As Theo Farrell argued,
“ideas operate ‘all the way down’ to actually shape actors and actions in world politics.” 17
Ideational structures give meaning to the world. By so doing, they organise and order it.
They define the world and the shape it takes. Yet they also shape actors and how they
behave in this environment. To account for change in the system or in structures, however,
social constructivism leaves room for agency at the micro level. It allows for the conscious
choices and actions of individuals to play out.
Individuals in social constructivist
frameworks are not passive reproducers of structures. On the contrary, their actions can
be transformative: “if desires and beliefs are explained (wholly or partly) by facts about
social structures, [social structures] in turn have explanations […] they are the residues or
precipitates of many past intentional actions and are maintained or transformed by
action.” 18
Accordingly, the social constructivist approach does not give primacy to
structure or agency. Rather, it assumes them to be ‘mutually’ or ‘co-constituted’. This dual
process is the pillar of Anthony Giddens’ structuration theory. 19 As John Gerard Ruggie
explained, Giddens’ structuration theory reflects “the ‘duality’ of structure: at once
16
Jutta Weldes, Mark Laffey, Hugh Gusterson, and Raymond Duvall, “Introduction: Constructing Insecurity,” in
Jutta Weldes, Mark Laffey, Hugh Gusterson, and Raymond Duvall, eds., Cultures of Insecurity: States,
Communities, and the Production of Danger (Minneapolis: University of Minnesota Press, 1999), p. 17.
17
Theo Farrell, “Constructivist Security Studies: Portrait of a Research Program,” International Studies Review,
vol. 4, no. 1, 2002, p. 50. Social constructivists debate among themselves just how ‘all the way down’ ideas
operate. For many, a statement like ‘all the way down’ amounts to neglecting the role of material forces.
18
Michael Taylor, “Structure, Culture and Action in the Explanation of Social Change,” Politics and Society, vol.
17, no. 2, 1989, p. 117.
19
On the link between Alexander Wendt’s social constructivism and Anthony Giddens’ structuration theory,
see, for example, Bill Mc Sweeney, Security, Identity and Interests: A Sociology of International Relations (New
York: Cambridge University Press, 1999), p. 124.
16
constraining human action but also being (re)created by it.” 20 Constructivists are,
therefore, structurationists, not structuralists.
It is these premises of social constructivism that bring together “projects that aim
at displaying or analyzing actual, historically situated, social interactions or causal [or
constitutive] routes that led to, or were involved in, the coming into being or establishing of
some present entity or fact.” 21 Social constructivist analyses focus on highlighting and
historicising the processes behind the emergence, resilience and transformation of social
structures, as well as on the role agents play in these. This allows social constructivism to
move beyond assessing the effects of established structures in the system (or why a
structure has an effect) to instead also shed light on how such structures came to be. 22
Social Constructivist Insights for the Study of Identity
It is for its ability to study the emergence and transformation of structures and actors that
social constructivism has become a popular approach for the study of identities. Theorists
are increasingly aware of the need to distance themselves from approaches that take for
granted constitutive processes or that treat identities as epiphenomenal, exogenously
given or innate. With the growing recognition of social constructivism, many theorists
seized the opportunity to rethink identity after years of its ‘undertheorisation’ at the hands
of primordialist and rational choice or instrumental approaches, two prevalent approaches
for the study of identity related phenomena.
Social constructivism rejects the primordialist or rational choice premise that
identities are innate or exogenously given. 23 Identities are, on the contrary, another form
20
John Gerard Ruggie, “What Makes the World Hang Together,” p. 875.
Ian Hacking, The Social Construction of What? (Cambridge, Mass.: Harvard University Press, 1999), p. 48.
22
Richard Price and Christian Reus-Smit differentiated between interpretive ‘how’ questions which address
constitutive processes and ‘why’ questions that are of a causal nature. See Richard Price and Christian ReusSmit, “Dangerous Liaisons?,” p. 277.
23
The underlying premise of primordialism is that cultural identities, particularly ethnic attachments, are a
given. Primordialist authors generally assume that collective identity is the product of ‘objective’ cultural and
21
17
of constructed ideational structures. Accordingly, identities do not have a predetermined
nature or meaning. Instead, identities are fluid, a result of processes of interpretation by
their bearers. 24 Identity is “never a finished product; it is always in the process of being
constructed and reconstructed.” 25 This constant process obscures its constructed nature.
As James Fearon and David Laitin explained, “social identities are produced and
reproduced through the everyday actions of ordinary folk. […] Individuals think of
themselves in terms of a particular set of social categories, which lead them to act in ways
that collectively confirm, reinforce, and propagate these identities.” 26 This leads to an “on
the ground” or “everyday form of primordialism.” 27 Identities become so much part of
individuals’ perceptions of themselves in society and of their group that they are seen as a
normal, if not natural, part of their self and collective definitions, as an objective fact.
Most identities, especially collective identities, are not free-floating individual
perceptions, however.
Like other ideational structures, they are intersubjective.
As
Stephen Cornell argued, collective identities emerge from “a gradual layering on and
connecting of events and meanings,” that foster a common understanding of what links
social factors. For more on primordialism, see Jack David Eller and Reed M. Coughlan, “The Poverty of
Primordialism: the Demystification of Ethnic Attachments,” Ethnic and Racial Studies, vol. 16, no. 2, 1993, pp.
183-201. The main premises of rational choice are that: 1) human beings are rational; 2) that this rationality
takes the form of mean-ends calculations, or of a cost-benefit analysis; 3) they are based on exogenously
given interests translating into preferences; and 4) at the heart of interests is the desire to maximise outcomes
or benefits despite constraints; 5) which leads actors to choose their most preferred option for attaining their
goals.For more on rational choice or instrumentalism, see Randall Calvert, “Identity, Expression, and Rational
Choice Theory,” revised version of a paper presented at the Conference on the State of the Discipline,
Washington, D.C., December 8-9, 2000. While strict rationalism takes identities as given, recent hybrid
approaches have begun incorporating social or psychological insights to account for identity formation. One of
the first attempts to bridge the rational choice and psychology divide came from Richard C. Snyder, H.W.
Bruck and Burton Sapin in their Decision-Making as an Approach to the Study of International Politics
(Princeton: Princeton University Press, 1954). On other attempts to bridge rational choice and social theory,
see for example Randall Calvert, “Rationality, Identity and Expression,” in Ira Katznelson and Helen V. Milner,
eds., Political Science: The State of Discipline, 3rd Edition (New York: W.W. Norton & Co., 2002).
24
Bill McSweeney, Security, Identity and Interest, p. 162.
25
Roxanne Lynn Doty, “Sovereignty and the Nation: Constructing the Boundaries of National Identity,” in
Thomas J. Biersteker and Cynthia Weber, State Sovereignty as a Social Construct (New York: Cambridge
University Press, 1996), p. 123.
26
James D. Fearon and David D. Laitin, “Violence and the Social Construction of Ethnic Identity,” International
Organization, vol. 54, no. 4, 2000, pp. 855-856.
27
Ibid., p. 848.
18
members to the group. 28 Although there never is complete agreement on what symbols,
traits or events belong to the group or on what it means to be part of a specific group, the
process of forming a collective identity nevertheless entails creating a sense of
commonality, a narrative around the most common experiences that appeal to group
members. 29 Collective identity, therefore, forms a master narrative common enough and
agreed upon enough to represent all members of the group.
Many social constructivists studying identity and nationalism have looked at what
constitutes these collective identities. For Manuel Castells, “[t]he construction of identities
uses building materials from history, from geography, from biology, from productive and
reproductive institutions, from collective memory and from personal fantasies, from power
apparatuses and religious revelations.” 30 Powerful material to foster a sense of belonging
is found in a group’s past, from which can be woven this master narrative about a common
descent, origin or ancestors: “a storyline concerning their origins, the critical events that
define them as a people, and some broad argument over where they should be headed.” 31
Collective historical narratives are rarely a neutral reading of history. They are akin to
memories and myths that become, according to Anthony Smith,
sources of moral inspiration to [group] members, selective traditions (including
legends) about their past handed from earlier generations, in which certain events
and personages are remembered and others forgotten. This history becomes a
potent and malleable resource for ethnic communities, embellishing kernels of
historical ‘fact’ with exempla virtutis to create a sense of common history and
destiny. 32
28
Stephen Cornell in Kevin C. Dunn, “Imagining Mobutu’s Zaire: The Production and Consumption of Identity
in International Relations,” Millennium: Journal of International Studies, vol. 30, no. 2, 2001, p. 241.
29
Paul R. Brass, Ethnicity and Nationalism: Theory and Comparison (London: Sage Publications, 1991), p. 86.
30
Manuel Castells, The Power of Identity. The Information Age: Economy, Society and Culture. Volume II,
Second Edition (Malden, Ma.: Blackwell Publishing, 2004), p. 7.
31
Michael Barnett, “Culture, Strategy and Foreign Policy Change: Israel’s Road to Oslo,” European Journal of
International Relations, vol. 5, no. 1, 1999, p. 12.
32
Anthony D. Smith, “The Ethnic Sources of Nationalism,” Survival, vol. 35, no. 1, 1993, p. 50.
19
Collective identities might also emerge from shared expectations, hopes and dreams
about a common future, with or without being grounded in a common past. 33
What
matters is that the definition of this identity be broad enough and inclusive enough to
represent the community without too much inconsistency.
Inconsistencies and contestation can undermine the legitimacy of collective
identities. For this reason, part of the process behind the formation of identities consists in
fixing their meaning. Identities cannot be left fluid, or left to the whims of constant
reinterpretation, if they are to be shared by all. They must be made sufficiently stable to
be incorporated into the social background of both individuals and the group. The role of a
master narrative, of a stable interpretation of identity, is, therefore, not only to serve as the
uniting structure for the group, but also to represent, eventually, an accepted guiding
principle for the group. Consequently, master narratives limit the ability of group members
to think of themselves in different terms. They also constitute a norm for behavior. They
provide the socially and culturally condoned rules of the group, according to which actors
behave. 34 This process is what eventually fixes the meaning and understandings around a
specific identity. 35
33
Paul Ricoeur, for example, talked of imaginaires of reaffirmation (the past) and imaginaires of rupture (the
future). As he explained, “[e]very society [...] possesses, or is part of, a socio-political imaginaire, that is, an
ensemble of symbolic discourses. This imaginaire can function as a rupture or a reaffirmation. As
reaffirmation, the imaginaire operates as an ‘ideology’ which can positively repeat and represent the founding
discourse of a society, what I call its ‘foundational’ symbols, thus preserving its sense of identity. After all,
cultures create themselves by telling stories of their own past. The danger is of course that this reaffirmation
can be perverted, usually by monopolistic elites, into a mystifactory discourse which serves to uncritically
vindicate or glorify the established political powers. In such instances, the symbols of a community become
fixed and fetishized; they serve as lies. Over against this, there exists the imaginaire of rupture, a discourse of
utopia which remains critical of the powers that be out of fidelity to an ‘elsewhere’, to a society that is ‘not yet’.
But this utopian discourse is not always positive either. For besides the authentic utopia of critical rupture
there can also exist the dangerously schizophrenic utopian discourse which projects a static future without ever
producing the conditions of its realization.” Paul Ricoeur, “Dialogue with R. Kearney,” in Richard Kearney,
Dialogues with Contemporary Continental Thinkers (Manchester: Manchester University Press, 1984), pp. 2930.
34
Janice Bially Mattern, “The Power Politics of Identity,” European Journal of International Relations, vol. 7, no.
3, 2001, p. 364.
35
For a description of Alexander Wendt’s conception of processes behind identity formation, Maja Zehfuss,
“Constructivism and Identity: A Dangerous Liaison,” European Journal of International Relations, vol. 7, no. 3,
2001, p. 320.
20
These fixes, though given a semblance of innateness, remain precarious. 36 They
are, after all, the result of processes and choices made within the community at a
particular time. Identities are contingent, and, importantly, contextual. For authors that
refute the innateness of collective attachments, such as Eric Hobsbawm and Ernest
Gellner, identities “are feelings of community and solidarity which have evolved in social
processes and are therefore context-dependent.” 37 Some authors prefer to call them
‘situational.’ 38 They are tied to the historical trajectory of the group, and to its specific
evolution within its environment. As the environment, material and ideational, of groups
evolve, so do their perceptions and, concurrently, so does their identity.
These dynamics help explain the structural aspects behind the constitution of
identity. But agency is also central to identity. The importance of agency is most flagrant
when we turn to another of identity’s characteristics: the fact that it is relational. Individuals
must relate to others to develop a sense of who they are. Identities, therefore, come about
through a process of social interaction and, in turn, of comparison and construction.
Alexander Wendt illustrated this point. In a seminal piece entitled “Anarchy is What States
Make of It: The Social Construction of Power Politics,” he described two individuals, Alter
and Ego, encountering each other for the first time. 39
Without any pre-existing data to
draw from, one can only base himself/herself on this contact with the other to begin
developing a sense of one’s position and of how to relate with the other. According to
Jean-François Thibault,
[f]or Wendt, the configuration which in the future this relationship between ‘alter’
and ‘ego,’ between an agent A and an agent B, will take rests largely on their
interactions, i.e. on what each actor will do, on the actions that each will take. [...]
It is these various processes (signalling, interpreting or responding) that as
36
Jutta Weldes, “The Cultural Production of Crises: U.S. Identity and Missiles in Cuba,” in Jutta Weldes, Mark
Laffey, Hugh Gusterson, and Raymond Duvall, eds., Cultures of Insecurity: States, Communities, and the
Production of Danger (Minneapolis: University of Minnesota Press, 1999), p. 59.
37
John Martinussen, Society, State and Market: A Guide to Competing Theories of Development (London: Zed
Books, 1997), p. 324.
38
Anthony D. Smith, “The Ethnic Sources of Nationalism,” p. 49.
39
Alexander Wendt, “Anarchy is What States Make of It,” pp. 404-405.
21
actions, lead to the ”original set up” of an intersubjective space which grounds the
rules and structural resources with which agents will function. 40
The individual and the choices he/she initially makes are, therefore, at the center of the
process that leads to the emergence of intersubjective rules and meaningful, normative
structures. Individual choices and actions, taken together, form the necessary shared
understandings that serve as the basis for collective perceptions and identities.
Also implicit in this relational nature of identity is the role played by difference in the
constitution of identities. It is through comparison with others, through a relationship with
difference, that categories of what we are and are not emerge. 41 To define oneself in a
certain way means that someone else is inherently defined differently. 42
He/she
represents what we, through comparison, have judged not to form part of what we
represent. Thus, collective identity is intrinsically linked to interpretation of difference. In
this process of differentiation, however, there are no necessary pre-determined normative
connotations attributed to the self and other. Difference merely serves to establish
boundaries between categories and groups at first. With time, though, as meaning is
ascribed to categories through recurrent interactions, these categories “not only define a
boundary but also a locally variable set of relations across that boundary.” 43 Categories
come to form distinct identities.
40
Personal translation. Jean-François Thibault, “Représenter et connaître les Relations Internationales:
Alexander Wendt et le paradigme constructiviste,” Note de recherche no. 7, Centre d’études des politiques
étrangères et de sécurité, Université du Québec à Montréal, no. 7, janvier, 1997. Charles Tilly also described
the process in these terms: “[w]hen members of two previously separate or only indirectly linked networks
enter the same social space and begin interacting, they commonly form a social boundary at their point of
contact. [...] As interaction intensifies between clusters of previously unlinked or indirectly linked social sites,
boundaries between them become more salient.” Charles Tilly, “Social Boundary Mechanisms,” Philosophy of
the Social Sciences, vol. 34, no. 2, 2004, p. 218. In the same article, Tilly defines social boundaries as: “any
contiguous zone of contrasting density, rapid transition, or separation between internally connected clusters of
population and/or activity.” Ibid., p. 214.
41
Crawford Young, “Patterns of Social Conflict: State, Class and Ethnicity,” Daedalus, vol. 111, no. 2, 1982, p.
74.
42
Ronnie D. Lipschutz, “Negotiating the Boundaries of Difference and Security at Millennium’s End,” in Ronnie
D. Lipschutz, ed., On Security (New York: Columbia University Press, 1995), p. 214.
43
Doug McAdam, Sidney Tarrow and Charles Tilly, Dynamics of Contention (New York: Cambridge University
Press, 2001), p. 29.
22
For some, a link between identity and violence is found precisely in this relation to
difference. 44
For a number of authors, this differentiation process implies a binary and
antagonistic relation to the other. When groups constitute themselves as a ‘we,’ they
inherently define who ‘they’, the others, are. It establishes clear criteria and boundaries for
inclusion and exclusion in the group. Differentiation and the delimitation of boundaries
guarantee the coherence and unity of the group, the integration of the self. 45 This is done,
however, by simultaneously projecting outward everything excluded from the self. For
Michael Shapiro, the emergence of a coherent unified identity necessitates “confrontation
and radical separation from alterity.” 46 Other groups embody everything the ‘collective
self’ has chosen to eliminate from its own definition, shifting “diversity, disruptions, and
dangers to the outside.” 47
Because all negative or excluded categories and attributes are projected onto
others, intolerance, even fear, of these others can develop. 48 The danger lies precisely
with diversity and difference: the others have the potential to upset the order of the
ingroup, to disrupt and threaten just by being ‘other.’ 49 By representing alternatives modes
of identity, they can challenge the group’s sense of collective identity. As David Campbell
argued,
[t]he mere existence of an alternative mode of being, the presence of which
exemplifies that different identities are possible and thus denaturalizes the claim of a
44
Although some of the authors reviewed in this section of social constructivism and the link between identity
and violence would not categorise themselves as social constructivists but would probably fall in critical,
postmodern or post-structuralist schools, they nevertheless all adopt the view that identity is not innate but
socially constructed.
45
In his article, Rodney Bruce Hall reviewed the work of Iver B. Neumann. Rodney Bruce Hall, “Applying the
‘Self/Other’ Nexus in International Relations, International Studies Review, vol. 3, no. 1, 2001, p. 104.
46
Michael J. Shapiro, “Warring Bodies and Bodies Politic: Tribal Warriors Versus State Soldiers,” in Michael J.
Shapiro and Hayward R. Alker, eds., Challenging Boundaries: Global Flows, Territorial Identities (Minneapolis:
University of Minnesota Press, 1996), p. 455.
47
Roxanne Lynn Doty, “Immigration and National Identity: Constructing the Nation,” Review of International
Studies, vol. 22, 1996, p. 239.
48
Michael J. Shapiro, “Warring Bodies and Bodies Politic,” p. 477.
49
Iver B. Neumann, Uses of the Other: ‘The East’ in European Identity Formation (Minneapolis: University of
Minnesota Press, 1999), p. 17.
23
particular identity to be the true identity, is sometimes enough to produce the
understanding of a threat. 50
Anything beyond the realm of the ingroup is inherently a threat to the coherence of the
group. Campbell concluded that “inscribing the boundaries that make the installation of
the nationalist [or group] imaginary possible requires the expulsion from the resultant
‘domestic’ space of all that comes to be regarded as alien, foreign and dangerous. The
nationalist [or group] imaginary thus demands a violent relationship with the other.” 51
This conceptualisation of a link between identity and violence through a self-other
nexus prompted much criticism.
Many argued that this self-other dichotomy create
deterministic, if not essential, categories: the self and the other. 52 The argument, then,
becomes teleological: the existence of binary categories explains the relation between
bearers of these identities.
Because they are categorised as different, they should
inherently be antagonistically different. While collective identities are relational and set
boundaries between groups, we should not assume the nature of the emerging identities.
Differentiation does not mean antagonisation. Assuming that antagonistic identities
emerge from differentiation amounts to confusing identity formation mechanisms with the
contingent meanings attributed to an identity or a set of relations.
Another point of criticism addressed at this specific social constructivist literature is
that it leaves very little room for agency. For social constructivists linking insecurity to a
self-other nexus, the possibilities for action and relations, particularly of an alternate
nature, are extinguished by the set, antagonistic meaning inherent to binary identities. In
this view, “people are continually made and remade by discourses that are the essential
properties of […] groups.” 53
Individuals are left with no room to contest or change
50
David Campbell, Writing Security: United States Foreign Policy and the Politics of Identity (Minneapolis:
University of Minnesota Press, 1998), p. 3.
51
David Campbell, National Deconstruction: Violence, Identity, and Justice in Bosnia (Minneapolis: University
of Minnesota Press, 1998), p. 13.
52
Iver B. Neumann, Uses of the Other, p. 208.
53
James D. Fearon and David D. Laitin, “Violence and the Social Construction of Ethnic Identity,” p. 861.
24
antagonistic dynamics. On the contrary, human action appears limited to a passive role of
reproducing an antagonistic reading of intergroup dynamics. The clearest example of how
small a role agency plays in such analyses is in the role attributed to the ‘other,’ or other
groups. As Iver Neumann argued, “the other disappears as a subject.” 54 The other only
serves as a category in relation to which definitions of the collective self are developed.
With so little agency, this type of analysis borders on structuralism. With this overemphasis
on structure, and particularly on inherently antagonistic identity structures, these analyses
are incapable of explaining variations in group relations. Myriads of groups exist across the
world, but not all exhibit antagonistic tendencies in their relations with others. To
understand this, it is necessary to understand the role agents play in ascribing meaning,
including antagonism, to identity structures. Conscious choices and actions on the part of
volitional agents play a large part in the reinterpreting structures.
When steering clear of more radical takes on difference and violence, social
constructivism writ large is a vital approach for the study of social artefacts and relations.
It is recognised as a powerful epistemological stance. Insights on the impact of both
environmental factors and agency on identity structures, on the importance of comparison
and differentiation, and on the relative fluidity of meaning ascribed to specific identities, to
name a few, help shed light on identity dynamics. This makes it a strong base to address
the subject of this dissertation. This epistemological stance runs, in fact, as an
undercurrent to the research, but also to research in various fields. Its insights have been
validated, complemented and extended by various bodies of literature on determinants of
social perceptions and collective behaviour, both psychological and social. A number of
these are addressed in the following chapters.
54
Iver B. Neumann, Uses of the Other, p. 221.
25
CHAPTER 2: CROWD PSYCHOLOGY: PRECURSOR TO MODERN STUDIES ON
COLLECTIVE BEHAVIOUR
Developed at the turn of the 19th century, crowd psychology–also referred to as collective
psychology or group psychology–emerged in continental Europe, mainly in Italy and
France. 1 It paralleled the development in Germany of Völkerpsychologie. 2
Crowd
psychology had for focus ‘mass movements’, ‘mass related phenomena’. 3 In many cases,
these studies hid a deeper political agenda. Some authors, aristocrats and conservative
bourgeois, used crowd psychology as a means to express concern about the growing
power of the masses. They regarded events such as a wave of uprisings in France, the
emergence of socialism, and the growing popularity of syndicalism, as a threat, as contrary
to progressive evolution. These events and trends were a “reaction to sudden and wideranging processes of change such as industrialization and urbanization” 4 While advances
such as industrialisation and urbanisation, were signs of the establishment of a modern
structured social order, movements opposed to these represented stagnation, regression
or even irrational behaviour.
The sweeping changes of the modern era also brought with them a danger of a
different kind, the modern demagogue. Industrialisation and urbanisation displaced
traditional arrangements and affiliations, and led according to Durkheim to anomie or a
sense of individual loss amidst new structures. This, in turn, offered new leaders the
opportunity to play on people’s sentiments and achieve great power. As John McClelland
explained, according to crowd psychologists,
1
For an extensive review of this literature, see Ralph H. Turner and Lewis M. Killian, Collective Behavior,
Fourth Edition (Englewood Cliffs, N.J.: Prentice-Hall, 1993).
2
Developed in Germanic countries, Völkerpsychologie contended that individuals’ beliefs and behaviour are
dictated by character traits specific to their community or folk.
3
Martti Muukkonen, From Deviant Phenomenon to Collective Identity: Paradigm Shifts in Social Movement
Studies, master’s thesis, University of Joensuu, Joensuu, Finland, 1999, p. 36.
4
Antonio Famiglietti, “For a Realist Definition of Social Movements,” paper presented at the Debating
Realism(s) conference, Roskilde University, Denmark, 17-19 August 2001, p. 1 and Antimo L. Farro, Les
mouvements sociaux: diversité, action collective et globalisation (Montréal: Presses de l’Université de
Montréal, 2000), p. 36.
26
[l]iberal individualism meant that the masses were faced with all kinds of lifechoices which they never had to face before, and a certain economic ruthlessness
in liberal societies meant that life-choices would often have to be faced in conditions
of considerable uncertainty and insecurity. This was a demagogue’s dream. All he
had to do was to answer as many of life’s questions as possible with as few and as
simple answers as possible, and the masses would rush to put power into his
hands. 5
In the modern era, leaders, benevolent or malevolent, could learn to harness the
sentiments of crowds to serve their ends.
Sensitised to the rise of mass movements and fearful of the dangers of
manipulable crowds, crowd psychologists sought to understand why individuals–whom
they took in isolation to be reasonable beings–chose to partake in movements, and, when
they did, behaved in an abnormal manner. Individuals taking part in collective movements
were thought, “to succumb inevitably to an atavistic regression to unconscious primal
impulses.
Once immersed in a crowd, it was claimed, one experienced a loss of
differentiation and individuality and merged with the larger amorphous mass of an
impulsive, irrational group.” 6
Nowadays, crowd psychology is an often forgotten or dismissed approach.
Although little esteemed, it represents nonetheless an insightful, intuitive take on leaderscrowds psychology, as well as individual-group interactions, one which can contribute to
contemporary analyses of mobilisation. This chapter reviews this stance through the work
of three of its principal authors: Scipio Sighele in Italy, and Gustave Le Bon and Gabriel
Tarde in France. 7
5
John McClelland, “The Place of Elias Canetti’s Crowds and Power in the History of Western Social and
Political Thought,” Thesis Eleven, no. 45, 1996, p. 20.
6
Christine Poggi, “Folla/Follia: Futurism and the Crowd,” Critical Inquiry, vol. 28, 2002, pp. 711-712.
7
Some authors also include Sigmund Freud among crowd psychologists for some of this work on ‘masses.’
For one of Sigmund Freud’s main analyses on the issue, see Sigmund Freud, Group Psychology and the
Analysis of the Ego (New York: Norton, 1990 [1921]).
27
Scipio Sighele and the Suggestibility of Crowds
Scipio Sighele, an Italian anthropologist and criminologist, published in 1893 La coppia
criminale: Studio di psicologia morbosa (in English The Criminal Couple). Laying some of
the foundations for the analysis of collective phenomena, this book made him, for many,
the founding father of crowd psychology. 8 Sighele’s interest in crowds stemmed from his
study of ‘criminal mobs’, those responsible for riots, vandalism and acts of violence. He
held an extremely conservative view of the crowd, which he described as “capable of
every excess, possibly capable of only excesses, admirable at times for its abnegation,
frequently frightening in its ferocity, never or almost never even and measured in its
sentiments.” 9 Sighele took collective movements “to be qualitatively inferior to the
individuals comprising the crowd.” 10 This inferiority stemmed mainly from the fact that, in
crowds, individuals tended to forego reasoned judgment and to fall prey to more instinctive
forms of behaviours, including criminal types of behaviour.
While this conservative stance on the part of the Italian criminologist is the better
known part of his work, much less is known, outside of Italy, about his research on
suggestion, the fundamental concept he developed. Sighele’s focus on crowds and
suggestion developed in reaction to the views of his mentor, Cesare Lombroso. Lombroso
and social thinkers at the time of Unification in Italy tended to focus on dangerous
leadership. Leaders belonging to this group were described as capable of great control
over masses, including swaying them, like hypnotists.
This made them a danger to
society, because such a leader could be “capable of inflicting crimes on unsuspecting
victims by paralyzing their resistance [...], and more indirectly he could turn his victims into
8
Gustave Le Bon published his own work on crowd psychology in France in 1895. The few years separating
the publication of Sighele and Le Bon’s work led these two authors to each claim to have founded the theory of
mass psychology. Since Le Bon’s work gained more notoriety, as Suzanne Stewart-Steinberg argued, “[i]n the
annals of intellectual history, Le Bon appears to have won the debate.” Suzanne R. Stewart-Steinberg, “The
Secret Power of Suggestion: Scipio Sighele and the Postliberal Subject,” Diacritics, vol. 33, no. 1, 2003, p. 77.
8
Christine Poggi, “Folla/Follia,” p. 729.
9
Quoted in Christine Poggi, “Folla/Follia,” p. 718.
10
Personal translation. Antimo L. Farro, Les mouvements sociaux, p. 37.
28
automata with his powers of suggestion and thereby coerce the victim him/herself to
commit a crime.” 11
By attributing to criminal leaders the power to suggest deviant behaviour, authors
of the time blurred the lines of criminality. Criminal acts could not be construed anymore as
the result of the innate tendencies of a few deviants, but could result from suggestion to
normal unsuspecting individuals. The key to understanding deviant or criminal behaviour
did not, therefore, solely lie in pathology, but also in relations between individuals. This
intrigued Sighele. Thus, although like many of his predecessors, Sighele came to be
“obsessed with the drama of the influence of man over man,” unlike them, however, “what
interested Sighele more than the political leader were his followers,” and particularly their
susceptibility to suggestion. 12
For Sighele, suggestion was ubiquitous in all social relations and led to two
archetypes of interactions between individuals: absorption, in which one person dominates
another, leading the latter to erase him/herself in front of the stronger one, or fusion, in
which two equal partners exercise mutual fascination on one another, with both erasing
themselves somewhat in their fascination for the other. These types of relations produced
two distinct social positions or roles: the incubus, the head or thinker, and the succubus,
the arm or acting figure. 13 These were roles found throughout society, divided between
leaders and followers. Authorities, in particular, played the role of leaders.
According to Sighele, however, the modern era changed social relations.
As
Suzanne Stewart-Steinberg explained, “[m]odern society [...] no longer has an incubus, or
at least not one that is visible.” 14
In the past, authoritarian figures such as absolute
monarchs played the role of the incubus. Modern society replaced the absolute monarch
with a constitutional one, embodiment not of unquestioned divine authority but of
11
Suzanne R. Stewart-Steinberg, “The Secret Power of Suggestion,” p. 68.
Ibid., p. 66.
13
Emphasis in original. Ibid., p. 69.
14
Emphasis in original. Ibid., p. 73.
12
29
contractual ties, norms and rules with all members of society. The incubus is now “an
invisible crowd composed of [...] ancestors, [...] compatriots, [...] educators.” 15
The
predicament the modern person faces then is that while remaining in the position of the
succubus, the subordinate, he/she is now under the influence of this invisible, evasive
incubus. He/she remains incapable of free will, as the succubus, but can fall prey to the
auto-delusions and auto-suggestion of the headless crowd now in the leading position.
But, according to Sighele, in this absence of visible leadership, the modern individual is
also susceptible to ‘hypnosis’ by strong, criminal incubus personalities.
Gustave Le Bon and the Mental Unity of Crowds
Gustave Le Bon, Sighele’s French counterpart, shared his fears of the growing strength of
crowds and their susceptibility to manipulation. In discussing the political developments of
his era, Le Bon stated that
While all our ancient beliefs are waning and disappearing, and the old pillars of
societies are crumbling one after the other, the power of crowds is the only force
that nothing threatens and its prestige keeps on growing. The age we are about to
enter will in truth be the ERA OF THE CROWDS. 16
A contemporary of Sighele, Le Bon is certainly one of the better known crowd
psychologists, and one of the most conservative ones. He saw in the rise of popular
classes a sweeping, unstoppable destructive force of which the French Revolution
constituted a telling example. 17 Steve Reicher, for example, described The Crowd, Le
15
Part of the citation is from Scipio Sighele in Suzanne R. Stewart-Steinberg, “The Secret Power of
Suggestion,” p. 73.
16
Emphasis in original. Personal translation. Gustave Le Bon, Psychologie des foules (Paris: Félix Alcan
1905), electronic version, p. 11. Available online at
http://classiques.uqac.ca/classiques/le_bon_gustave/psychologie_des_foules_Alcan/foules_alcan.html.
17
In a telling excerpt, he explained that: “[t]he crowds only have the power to destroy. Their domination always
represents a phase of barbarism.” Personal translation. Gustave Le Bon, Psychologie des foules, p. 13.
Interestingly, though, although Le Bon discussed the destructive power of crowds and seemed to entertain a
fear of them, he also expressed a certain reverence for their brute strength and acknowledged that they are
also capable of positive actions, such as heroism. As he explained, “[c]rowds, undoubtedly, are also
unconscious but this unconsciousness might be one of the secrets of their strength. In nature, beings only
driven by instinct accomplish actions so complex that they amaze us.” Personal translation. Gustave Le Bon,
Psychologie des foules, p. 9.
30
Bon’s better known work, “as politically biased, reflecting bourgeois fears of the lower
classes.” 18 For Christian Borch, Le Bon’s conservative nature transpired not only in the
ideas he articulated, but also in the goals behind his work. As he explained, “The Crowd is
not merely concerned with theoretically understanding the phenomenon of crowds. It is
just as much conceived as a manual for statesmen on how to efficiently manage this
threat–a clear Machiavellian feature, as has often been observed.” 19 Le Bon barely hid his
political goals when he claimed that knowledge of the psychology of the masses had
become the last resource for politicians seeking to control the masses. 20
Much of Le Bon’s earlier work reflected the influence of Völkerpsychologie. In his
treatment of groups, he stressed the importance of hereditary, mainly racial, factors. His
turn to crowd psychology incorporated much of his earlier reflections. In The Crowd he
asserted that: “[o]ur conscious actions derive from an unconscious substratum created
mainly by hereditary influences.
This substratum contains the numerous ancestral
residues that constitute the spirit of the race.” 21 He noted, however, that in crowds such
character traits generally made way for new psychological characteristics, sometimes
differing radically from this original ancestral character.
A fundamental aspect of Le Bon’s work rested on the study of this displacement of
personal consciousness by collective traits shared by members of the crowd. He stated
that
[u]nder certain circumstances, and only under those circumstances, an
agglomeration of men presents new characteristics very different from those of the
individuals composing it. Conscious personality disappears, the sentiments and
ideas of all those in the group take one and the same direction. A collective mind is
formed, doubtless transitory, but presenting very clearly defined characteristics. [...
18
Quoted in Stephen Vider, “Rethinking Crowd Violence: Self-Categorization Theory and the Woodstock 1999
Riot,” Journal for the Theory of Social Behaviour, vol. 34, no. 2, 2004, p. 143.
19
Christian Borch, “Crowds and Economic,” p. 5.
20
Gustave Le Bon, Psychologie des foules, p. 14. Le Bon’s work effectively became a reference for politicians
and military leaders, of which the more famous are undoubtedly Adolf Hitler and Benito Mussolini.
21
Personal translation. Gustave Le Bon, Psychologie des foules, p. 20.
31
The group] comes to form one being submitted to the law of the mental unity of the
crowds. 22
The result, for Le Bon, is that with the erasure of the conscious personality in crowds, not
only does the individual begin thinking and behaving in ways guided by the new collective
unconscious, even if contrary to his normal state, but that he becomes an automaton
devoid of the capacity to reason on his own. 23
In crowds, “man regresses of many
degrees on the scale of civilisation.” 24
This explains why Le Bon saw crowds as inherently inferior to individuals in
isolation. 25 They are guided by their emotions, their primitive instincts, and not by reason.
As Elwis Potier explained: “[e]motional, evocative, unlimitedly gullible, strict and radical,
the crowd is a [...] collective being possessing an archaic psyche and spirituality.” 26 The
crowd adopts the characteristics of inferior beings, for which Le Bon provides the
examples of women, savages and children. 27
Le Bon studied three mechanisms he identified in crowd dynamics: the feeling of
invincibility derived from being in a group, contagion and suggestion. One of the sources
of the crowd’s irrational behaviour stems from the feeling of invincible power individuals
get when integrated to a group. Once part of the group, individuals can give in to all of
their instincts as a result of the anonymity the crowd affords them. They can act without
fear of consequences. The crowd, since it is anonymous, is an irresponsible entity, or at
22
Emphasis in original. Personal translation. Gustave Le Bon, Psychologie des foules, pp. 17-18.
As was already seen in the work of Scipio Sighele, the figure of the automaton is an important one in crowd
psychology. Akin to a hypnotic trance, the behaviour of individuals in crowds is also compared to, according to
Clara Gallini, “a machine, or more accurately, a man-machine, who imitates man in looks and functions, but is
unable to be autonomous because he moves on the basis of a mechanism created by man.” Quoted in
Suzanne R. Stewart-Steinberg, “The Secret Power of Suggestion,” p. 65. Such a choice of imagery could be
the result of parallels made with industrialization and modernisation. Recall similar images suggested by the
film Modern Times, featuring Charlie Chaplin.
24
Ibid., p. 22.
25
Personal translation. Gustave Le Bon, Aphorismes du temps présent (Paris: Les amis de G. Le Bon, 1978
[1913]), p. 244.
26
Elwis Potier, “Imaginaire du contrôle des foules dans l’armée de terre française,” Cultures & Conflits, no. 56,
2004, pp. 35-49. Available online at: http://www.conflits.org/document1609.html.
27
Personal translation. Gustave Le Bon, Psychologie des foules, p. 25. Many references to the femininity of
the crowd abound in crowd psychology literature of the time. As Christine Poggi explained, the crowd was
perceived as feminine “in its malleability, its incapacity to reason, its susceptibility to flattery and hysteria, and
its secret desire to be seduced and dominated.” Christine Poggi, “Folla/Follia,” p. 713.
23
32
least responsibility cannot be attributed to it. 28
As a result, liberated from all
responsibilities, the individual, unrestrained, can succumb to his instincts, which leads him
to feel powerful.
A second psychological factor at play in crowds is contagion. In Le Bon’s writings,
contagion plays a key role in cementing the crowd. As Christian Borch argued, “[i]t is this
contagious trait–clearly pointing to the crowds’ apparently pathological character–which
accounts for the, however ephemeral, unity of the crowd.” 29 Contagion was a sweeping
force for Le Bon. In groups, he believed “every sentiment, every action” to be “contagious,
and contagious to the point where the individual easily sacrifices his personal interests for
collective ones.” 30
Reminiscent of Sighele’s work, the third mechanism Le Bon focused on, and to
which he attributed great importance, is the suggestibility of crowds. 31
Inspired from
discoveries in psychology and from the refinement of hypnosis techniques, he believed
that individuals in a crowd are in a state akin to a hypnotic trance. 32 As Christine Poggi
asserted, “[a]ccording to Le Bon, once immersed in a crowd, individuals would soon find
28
Gustave Le Bon, Psychologie des foules, p. 21.
Christian Borch, “Crowds and Economic,” p. 5.
30
Personal translation. Gustave Le Bon, Psychologie des foules, p. 21.
31
According to many, Le Bon inspired himself heavily from the writings of Scipio Sighele. Le Bon never
acknowledged it in his work, however.
32
Many crowd psychologists of the time were inspired by advances made around hypnotism and by their
implications. As Suzanne R. Stewart-Steinberg argued “James Braid’s discovery in 1842 that hypnotic sleep
could be induced by the simple fixation on a luminous object had revolutionized magnetic theory and practice.
Braid’s was to prove an important discovery, for it made possible the shift from the putative supernatural
powers of the hypnotist–based on the theory of the transmission of magnetic fluids or electric currents–to an
analysis of characteristics thought to inhere in the hypnotized subject him-or herself. The real cause of
hypnosis, its very effectivity, was to be found [...] not in the magnetizer but in the subject. A rethinking, within
the late-nineteenth century literature on hypnotism, of what drives the subject, of what makes him act appears
predicated then on a very peculiar form of interiority, of self-causation within the subject himself. At the end of
its historical trajectory this rethinking would first result in the idea of auto-hypnosis or auto-suggestion, and
then, by the beginning of the new century, it would found the psychoanalytic subject, traversed as he is by
conflicting desires and topographically split between ego and superego. In both instances, however,
obedience is exacted as a result of an internalized command and–and here lies the peculiarity and utter
novelty of this subject–for that reason the subject ceases to be master of his own [domain].” Emphasis in
original. Suzanne R. Stewart-Steinberg, “The Secret Power of Suggestion,” p. 61. For example, for Enrico
Morselli, hypnotism involved “the deactivation of the inhibitory function of the brain and the consequent free
reign of the automatic reflex activity of the spinal cord. Without this inhibitory function in place, the subject
demonstrates the two most important psychological features of hypnotism: automatism defined as a lack of
spontaneity, and suggestion or the capacity to receive an infinite number of external stimuli and feelings.” Ibid,
p. 63.
29
33
themselves in a ‘special state, which much resembles the state of fascination in which the
hypnotised individual finds himself in the hands of the hypnotiser’.” 33 The crowd renders
individuals docile and readily amenable to suggestion; they can be swayed easily and
made to behave in ways they would not otherwise follow. With individuals susceptible to
suggestion in crowds, so is the crowd as a whole. As a result, although crowds are a
dangerous force to reckon with, mainly deviant and criminal in Le Bon’s mind, they can
nevertheless be swayed and made to follow suggested directions.
Therein lays the possibility for leaders to manage or manipulate groups. Crowds
naturally call for leadership and guidance. Like in the animal kingdom, “[t]he crowd is a
herd that needs its shepherd.” 34 In effect, history, according to Le Bon, has been shaped
by those he called les bâtisseurs de croyances, builders of beliefs: “[f]rom their graves,
grand gnostics (hallucinés) still bend [the wills] of millions of men with the enchantment of
their dreams.” 35 In the ‘era of the crowd,’ leaders need to harness the fury of groups to
manage their destructiveness and lead them in the right direction. As Poggi argued, “[t]his
required a leader who, like the hypnotist, would hold sway over his subjects through the
persuasive use of rhetorical images.” 36 For the French crowd psychologist, considering
the irrational nature of the crowd, images and emotions, more than reason and logical
arguments on the part of leaders, ensured their influence over the masses.
33
Partly quoting Gustave Le Bon, in Christine Poggi, “Folla/Follia,” p. 729.
Personal translation. Antimo L. Farro, Les mouvements sociaux, p. 40.
35
Personal translation. Gustave Le Bon, Aphorismes du temps présent, p. 209. In his original text, Le Bon
referred to ‘hallucinés’, which is translated here as ‘gnostics’ for lack of an equivalent in English. The term
gnostic is drawn from Eric Voegelin’s The New Science of Politics. In his preface to the 1987 re-edition of
Voegelin’s volume, Dante Germino explained that “[t]he Gnostic creed-movement gives its followers a sense of
superiority over the uninitiated, and its sage typically believes that he has become one with the godhead and
has achieved liberation from the world of ordinary human beings.” Eric Voegelin, The New Science of Politics:
An Introduction (Chicago: The University of Chicago Press, 1987 [1952]), p. vi. The grand gnostics are,
therefore, fevered idealists, visionaries, revolutionaries, etc.
36
Emphasis added. Christine Poggi, “Folla/Follia,” p. 729. Sigmund Freud proposed a similar conception in
Group Psychology and the Analysis of the Ego. Stephen Vider indicated that “Freud (1921/1959) argues that a
group is formed when an individual ‘gives up his ego ideal’–his moral conscience–‘and substitutes for it the
group ideal as embodied by the leader.’ From that point on, the leader (and the ideas/ideals he represents)
guides the behavior of the group as a hypnotist guides a patient. Even those group members who have not
idealized this leader will quickly fall in line, ‘carried away with the rest by ‘suggestion,’ that is to say, by means
of identification’.” Stephen Vider, “Rethinking Crowd Violence,” p. 151.
34
34
While Le Bon’s writings suggested important concepts for the study of collective
phenomena, many criticised his lack of objectivity and scientific reasoning. For many of
his contemporaries, the main problem with Le Bon’s work lay in its lack of clarity,
especially around the concept of crowd. Beyond differentiating between homogenous and
heterogeneous crowds, Le Bon amalgamated all forms of collective action, which led him
to conflate such things as parliaments, juries, riots and brawls, not noting the difference
between ‘natural’ crowds, gathered spontaneously, and ‘artificial’ crowds, consisting of
structured, organised and enduring groups. 37 This was Gabriel Tarde’s, one of Le Bon’s
contemporaries, main point of criticism. 38
Gabriel Tarde and Imitation
Although he shared Le Bon’s fear of crowds, Tarde, a French criminologist, and later
judge, nevertheless saw crowds as a ‘perfect and absolute’ sociality: “as soon as a good
idea arose in one mind it would be instantaneously transmitted to all minds.” 39 While
crowds posed a potential threat to society, they were also a positive force with their
extreme sociality.
This was the result, according to him, of imitation, a concept he
elaborated on and which became his better known contribution to crowd psychology. 40
For Tarde, imitation shared affinities with Sighele and Le Bon’s concept of
‘suggestion’. As Poggi explained, “[f]or the most part, such imitation occurred through
unconscious or automatic means.” 41 Unconscious imitation, for Tarde, did not necessarily
have the negative connotations Sighele and Le Bon ascribed to suggestion, however.
Tarde tried instead to develop this concept in a neutral manner. In his book Les lois de
37
Ibid.
Elwis Potier, “Imaginaire du contrôle des foules dans l’armée de terre française.”
39
Christian Borch, “Crowds and Economic,” p. 17.
40
Incidentally, because of this, Gabriel Tarde has recently been rediscovered as one of the forefathers of the
field of memetics. See for example, Paul Marsden, “Forefathers of Memetics: Gabriel Tarde and the Laws of
Imitation,” Journal of Memetics, vol. 4, 2000.
41
Christine Poggi, “Folla/Follia,” p. 727.
38
35
l’imitiation, first published in 1890, he explained how social interactions draw on imitation:
individuals function by integrating images or ideas of others and using them to develop
their own. 42 Individuals are therefore suggestible in the sense that they are directed and
guided by what they draw from this imitation process. While sensitive to the influence of
others, Tarde’s social individual is not Le Bon’s manipulable or gullible crowdgoers.
Imitation and suggestibility varied in degrees and according to circumstances.
What result from the overall process of imitation are the social resemblances
observable between individuals in society. These in a sense constitute society. Society
and its specificities, institutions, doctrines, norms, etc., are the result of this imitation
process. What intrigued Tarde, then, were the factors behind choices made as to what to
imitate and reproduce, and as to what to reject. As he explained, "[o]ur problem is to learn
why, given one hundred different innovations conceived of at the same time–innovations in
the form of words, in mythical ideas, in industrial processes etc.–ten will spread abroad,
while ninety will be forgotten." 43 Tarde set out to develop general principles of the
reproduction of imitations. These laws, which explained two types of reproduction, logical
and extra-logical, address some of the mechanisms that would be developed nearly a
century later in literature on framing processes. Tarde’s logical laws of imitation are as
follows:
(1) The origination of an invention involves the recombination of existing imitations,
and this origination will be influenced by the social context and abilities of those
involved with the recombination; (2) The success of an imitation in spreading
geometrically from its point of origination will be a function of its fit, that is,
compatibility, with the environment of existing imitations; (3) The selection, that is,
adoption of an imitation occurs either through "substitution" involving a "logical dual"
and "struggle" between two alternatives, or through "accumulation", a process
entailing a logical union of imitations. 44
42
Gabriel Tarde, Les lois de l’imitation (Paris: Kimé Éditeur, 1993 [1890]).
Quoted in Paul Marsden, “Forefathers of Memetics.”
44
Emphasis added. Ibid.
43
36
The logical laws of imitation stress the constraining role of previously accepted ideas and
ideational structures for the selection and acceptance of new ones. His extra-logical laws
are that:
(1) The reproduction of ends generally precedes the reproduction of the means to
those ends. In other words, in Tarde's view, goals tend to be imitated before the
actions that serve to attain them are adopted. For example, the imitation of a goal,
say to become rich, will generally antecede the adoption of imitations employed to
further this goal; (2) Imitations tend to propagate through a process of stratified
diffusion from those perceived as superior to those who perceive them as superior.
For example, traits originally associated with celebrities and otherwise privileged
tend to trickle down to those who associate them with such traits. 45
In this last law, Tarde pointed to the importance of prestige and privileged status of
discourse promoters for the diffusion and acceptance of elements to be imitated, a point
raised by modern social psychology.
Discourse promoters also play a role in logical
diffusion as Tarde stressed the importance of their abilities in the recombination of existing
material for the acceptance of imitations. Thus, although insisting on the importance of
pre-existing ideational structures for his extremely social process of imitation, Tarde, like
his contemporaries, ascribed a fundamental role to leadership in his approach.
Hence, while crowd psychologists acquired their name for their focus on mass
phenomena, the interplay between leaders and crowds was central to their analyses. So
was, for more conservative crowd psychologists, the role or characteristics of leaders, as
magnetisers of the masses.
There thus existed a fundamental tension in their work
between the individual in the crowd, falling prey to irrationality, and its antithesis, the
leader, representation of super rationality, as the puppeteer of the crowds. As Michael
Mack explained, “[m]ass psychology [...] opens up a divide between leaders and masses.
From Le Bon onwards, the leader represented rationality. [...] the leader had to
superimpose order and discipline on the irrationality and anarchism of the drive-ridden
45
Ibid.
37
masses.” 46 As a result, for many critics of crowd psychology, it inherently grew “sooner or
later, into an explicit Führer-theory.” 47
Overall, with the exception of Tarde’s contribution, crowd psychology is
overwhelmingly perceived nowadays as a pre-scientific, conservative approach. As a
result of this type of criticism, crowd psychology fell into disrepute. It became a straw man
for more collective perspectives on collective behaviour. This dismissal, however, was so
radical that some of the contributions crowd psychology can make to modern social
sciences have been ignored. While some of its bases are questionable, it presents a
number of intuitive ideas.
Crowd psychologists often referred to the exhilaration that
comes from being in a group. Though few today would claim that there is a real shift in
cognitive patterns in crowds or that individuals revert back to a primitive state, individuals
often describe being swept ‘as if by a wave’ when in a group. 48 This notion of how
individuals ‘let go’ in a crowd also led crowd psychologists to sense the ‘vacuum’ of
responsibility and authority this creates, one that can be exploited by cunning leaders. The
more sophisticated take on this relationship between crowd and leader is Tarde’s. Unlike
his peers who emphasised the hypnotic qualities of leaders, Tarde explored what lies
behind these qualities. In his work, he moved beyond analogies with the automaton, the
individual that acts on command without reflecting, to a focus on the interactive processes
of imitation. Human responses to others around them and to leadership are based on
communication, persuasiveness and a degree of conscious or unconscious choice.
46
Michael Mack, “How Literature Questions the Dogmas of Modernity,” Thesis Eleven, no. 82, 2005, p. 115.
Endre Kiss, “Does Mass Psychology Renaturalize Political Theory? On the Methodological Originality of
‘Crowds and Power’,” The European Legacy, vol. 9, no. 6, p. 729.
48
Nobel Prize winner Elias Canetti described his experience of being swept by the crowd in his book The
Torch in My Ear: He explained that “[h]e descends into the streets, joins the procession, indeed ‘dissolves in it’
and feels not ‘the slightest resistance to what the crowd was doing.’ At the same time, he says ‘[I was] amazed
that despite my frame of mind, I was able to grasp all the concrete individual scenes taking place before my
eyes.” Quoted in Eugene Goodheart, “The Power of Elias Canetti,” Partisan Review, vol. 67, no. 4, 2000, p.
614. See also Francesca Polletta, “’It Was Like a Fever...’ Narrative and Identity in Social Protest,” Social
Problems, vol. 45, no. 2, 1998, pp. 137-150.
47
38
Like all straw men and precursor literature, crowd psychology was what later
scholars reacted against. The points crowd psychology initially raised became points of
contention–and, in a sense, the point of emergence–for newcomers. Two issues raised by
crowd psychology, in particular, became the subject of contemporary fields: psychological
dynamics tying individuals to the group, studied by social psychology, and the specific
social dynamics of groups, including relations between leadership and members, at the
core of modern collective behaviour approaches and social movement theory. Both are
reviewed in the following chapters.
39
CHAPTER 3: SOCIAL PSYCHOLOGICAL LITERATURE ON IDENTITY FORMATION
AND GROUP DYNAMICS
Social psychology developed in reaction to crowd psychology and to individualistic
psychological approaches, such as Floyd Allport’s. 1 It focused instead on the individual, or
the self, in relation to the group. More specifically, it addressed the psychological
processes of group behaviour. To circumscribe its field, it concentrated on the
psychological group: “one that is psychologically significant for the members, to which they
relate themselves subjectively for social comparison and the acquisition of norms and
values [...] that they privately accept membership in, and which influences their attitudes
and behaviour.” 2
The concept of psychological group calls forth dynamics between the individual
and the collective, tied into the significance of and identification with the group on the part
of its members. On a political level, Oakes identified a number of these dynamics in recent
literature: leadership, nationalism and national identity, persuasion and influence on the
part of leaders, the development of consensus between members, collective behaviour
and social protest, cooperation, impression formation, stereotyping, prejudice and racism,
and public opinion. 3 To study these issues from the standpoint of the individual and social
dynamics, psychology needed an interactionist approach, which is what social psychology
sought to be. At the heart of this program was the work of Muzafer Sherif and later Henri
Tajfel’s Social Identity Theory (SIT), and Tajfel and John Turner’s Self-categorisation
1
In terms of group dynamics, as discussed in the previous chapter, crowd psychology evacuated such
fundamentals as processes of identification, shared interests and norms formation, or group emergence:
people were assumed to switch off rationality and melt into the collective, invariant of the crowd or group
involved. Individualistic psychology, on the other hand, followed the basic principle of modern psychology:
“psychological processes reside only in individuals–in the most literal sense.” John C. Turner, with Michael A.
Hogg, Penelope J. Oakes, Stephen D. Reicher and Margaret S. Wetherell, Rediscovering the Social Group: A
Self-Categorization Theory (Oxford: Basil Blackwell, 1987), p. 4. This trend was epitomised by the work of
Floyd Allport. He claimed that “the group is a nominal fallacy, a convenient fiction for summarizing the actions
of individuals–the individual is the sole psychological reality and there is nothing in the group not already
present in the individual member. Strictly speaking, in fact, groups do not exist–there are only individuals.”
Ibid., p. 10.
2
Ibid., pp. 1-2.
3
Penelope Oakes, “Psychological Groups and Political Psychology: A Response to Huddy’s ‘Critical
Examination of Social Identity Theory’,” Political Psychology, vol. 23, no. 4, 2002, p. 810.
40
Theory (SCT).
These provided the necessary “link between the psychology of the
individual–the representation of the self–and the structure and process of social groups
within which the self is embedded.” 4
Muzafer Sherif’s Experiments
Sherif exemplified the interactionist social psychologist. Giving precedence neither to the
individual nor to the collective, he believed that the two had to be understood in an
interactive manner.
He argued that “the properties of any part are determined by its
membership in the total functional system.” 5 In a group, the self remained, but acquired
new characteristics tributary to the group, which influenced individual behaviour.
From experiments he conducted in the 1940s and early 1950s, Sherif devised a set
of contentions. The first of these pertained to how specific ingroup processes lead to
intergroup competition or hostility. For Sherif, “groups naturally develop status structures
and group cultures, and then establish boundaries, providing for the opportunity of
intergroup conflict, particularly where resources need to be shared.” 6 Sherif believed, in
turn, that competition or hostility could be resolved–that group ego or ethnocentrism could
be superseded–if rivals were placed in a situation where they had to attain superodinate
goals, goals which neither group could achieve on their own, but for which they required
the help of the others. 7
To verify these hypotheses, Sherif conducted a new experiment, the ‘Robbers
Cave’ study. He selected twenty boys whom he divided into two groups, the Eagles and
the Rattlers. The boys were taken to an environment replicating a summer camp. The
4
Marilynn B. Brewer, “The Many Faces of Social Identity: Implications for Political Psychology,” Political
Psychology, vol. 22, no. 1, 2001, p. 115.
5
Muzafer Sherif, quoted in John C. Turner, with Michael A. Hogg, Penelope J. Oakes, Stephen D. Reicher and
Margaret S. Wetherell, Rediscovering the Social Group, p. 12.
6
In Gary Alan Fine, “Forgotten Classic: The Robbers Cave Experiment,” Sociological Forum, vol. 19, no. 4,
2004, p. 664.
7
Ibid.
41
investigators in the experiment posed as camp councillors while observing the evolution of
interactions between Eagles and Rattlers over the course of three weeks. In the first
stages of the study, “[t]hrough strategically planned activities the experimenters were able
to instigate polarization in the camp, which led to escalating intergroup antagonism and
conflict.” 8 Both Eagles and Rattlers strove to help group members as a means of ensuring
that their group would come out ahead. These results lent credence to Sherif’s contention
that under specifically competitive conditions, intergroup tensions arise. Following this,
Sherif modified the parameters of the experiment. He assigned the groups tasks requiring
intergroup collaboration for their completion. Confirming Sherif’s second hypothesis, the
boys overcame their competitive behaviour to achieve these superordinate goals
together. 9
Criticised today for taking place outside of a controlled environment, this
experiment remains one of the most telling studies on intergroup dynamics. 10 Sherif’s
experiment substantiated the existence of an interactive process between intergroup
comparison and group attachment, and, in certain settings, between ingroup attachment
and out-group animosity. 11
Henri Tajfel and Social Identity Theory
Sherif’s ‘Robbers Cave’ experiment launched a broad research agenda on intergroup
behaviour within social psychology.
conclusions in his own experiments.
Decades later, Henri Tajfel built on Sherif’s
These experiments, ‘minimal group’ studies,
produced fascinating results that shed light on the processes of social identification and
differentiation.
8
Ralph H. Turner, “Some Contributions of Muzafer Sherif to Sociology,” Social Psychology Quarterly, vol. 53,
no. 4, 1990, pp. 288-289.
9
Gary Alan Fine, “Forgotten Classic,” p. 664.
10
Ralph H. Turner, “Some Contributions of Muzafer Sherif to Sociology,” p. 289.
11
Ibid.
42
In these experiments, individuals were assigned to groups with little (minimal) or no
justification behind the assignment. In so doing, Tajfel sought to limit possible ingroup
attachment and affinities. Tajfel initially wanted to demonstrate that ingroup favouritism,
and to a larger extent ethnocentrism, was not inherent to group interactions. With little or
no natural affiliation to the group, subjects had been expected to act without discernment
towards ingroup and outgroup members.
These experiments produced unanticipated
results. Instead of the predicted behaviour, the test subjects consistently discriminated
against the other group. 12 This took the form of a constant inclination to favour the ingroup
in the allocation of rewards, even if it proved detrimental in terms of personal gains. Above
all, participants favoured a differentiation between ingroup and outgroup. 13 Similar studies
were conducted in various countries and consistently yielded comparable results,
indicating that cultural factors did not affect the validity of these findings. 14
Commenting on these results, Tajfel and his collaborators indicated that these
experiments: “show[ed] how easy it is to create discriminatory behaviour and to modify the
ingroup-outgroup perceptions of the subjects by placing them even in short-lived
competitive intergroup situations.” 15 The results of this research showed that there need
not be deep-seated intergroup differences to foster competitive behaviour among groups
or, to a larger extent, ethnocentrism. Tajfel and Turner further concluded that the key to
these results was to be found in “some factor or process inherent in the intergroup
situation itself.” 16 For these researchers, this key lay not in categorisation between the
groups, but in the dynamics of group identification and intergroup perceptions.
12
Jacob M. Rabbie, Jan C. Schot and Lieuwe Visser, “Social Identity Theory: A Conceptual and Empirical
Critique from the Perspective of a Behavioural Interaction Model,” European Journal of Social Psychology, vol.
19, 1989, p. 173.
13
Jenna Slotin, unpublished manuscript, University of Toronto, Toronto, Canada, 2003, p. 6. See also John C.
Turner, “Social Comparison and the Social Identity: Some Prospects for Intergroup Behaviour,” European
Journal of Social Psychology, vol. 5, no. 1, 1975, p. 13.
14
Jonathan Mercer, “Anarchy and Identity,” International Organization, vol. 49, no. 2, 1995, p. 240.
15
Henri Tajfel, M.G. Billig, R. P. Bundy, and Claude Flament, “Social Categorization and Intergroup
Behaviour,” European Journal of Social Psychology, vol. 1, 1971, p. 151.
16
Henri Tajfel and John C. Turner, quoted in Jonathan Mercer, “Anarchy and Identity,” p. 240.
43
Existing social psychological literature offered general insights into categorisation. 17
It had already been established that individuals use cognitive strategies and judgmental
shortcuts “in order to ease the burdens of information processing.” 18
Mechanisms of
categorisation, and related mechanisms of differentiation, are among these strategies.
Jacob Rabbie and collaborators asserted that categorisation helps individuals cope with
the uncertainty in their environment; 19 while for Jonathan Mercer categories are simpler
representations of the world, making it more readily apprehensible. 20 Tajfel et al. explained
that categorisation between groups serves a similar purpose: “the articulation of an
individual’s social world in terms of its categorization into groups becomes a guide for his
conduct in situations to which some criteria of intergroup division can be meaningfully
applied.” 21
But as Turner and Tajfel’s earlier comments indicated, categorisation was only part
of the process. To play a role in fostering antagonism, categories must be internalised and
be meaningful for individuals. 22 For Tajfel, the importance of internalisation pointed to the
role of the self in this process. As Oakes explained, in categorisation, “[t]he self could
become a stimulus in this process, such that sense making through categorization could
17
Alberto Voci defined categorisation, and by extension differentiation as “a process that operates on stimuli
present in the environment, modifying and reconstructing them. Through this process, otherwise disparate and
unorganized objects become meaningful, assimilated to some stimuli and, at the same time, differentiated and
contrasted from others.” Alberto Voci, “Relevance of Social Categories, Depersonalization and Group
Processes: Two Field Tests of Self-categorization Theory,” European Journal of Social Psychology, vol. 36,
2006, p. 73.
18
Joseph Lepgold and Alan C. Lamborn, “Locating Bridges: Connecting Research Agendas on Cognition and
Strategic Choice,” International Studies Review, vol. 3, no. 3, 2001, p. 18.
19
Jacob M. Rabbie, Jan C. Schot and Lieuwe Visser, “Social Identity Theory,” p. 177.
20
Jonathan Mercer, “Anarchy and Identity,” p. 241.
21
Henri Tajfel, M.G. Billig, R. P. Bundy, and Claude Flament, “Social Categorization and Intergroup
Behaviour,” p. 153.
22
In minimal group experiments, however, there seemed to be little on which to build meaningful categories.
For Tajfel, meaning in this case lay in group distinction itself. As he explained,” [m]eaning was found by
[participants] in the adoption of a strategy for action based on the establishment, through action, of a
distinctiveness between their own ‘group’ and the other, between the two social categories in a truly minimal
‘social system.” Henri Tajfel, quoted in Penelope Oakes, “Psychological Groups and Political Psychology,” p.
812. In a competitive situation calling for clear distinction between groups, the competitive situation itself could
provide a meaningful base for distinction.
44
include making sense of the self, developing a ‘self-definition in a social context’.” 23 This
insight launched Social Identity Theory (SIT), an elaborate conceptualisation of the
consequences of categorisation, differentiation and identification as uncovered by the
minimal group experiments.
SIT posits that the discrimination observed in these experiments is tied to
individuals’ basic need for a positive social identity. 24 According to this theory, positive
self-esteem is tied to a positive view of the group one belongs to. To gain this positive
view, comparison with other groups works to set them apart. 25 Individuals tend to seek a
clear differentiation between their group and outgroup members, based on criteria that
favour the ingroup and, therefore, enhance their perception of their group’s standing in
relation to others. 26 It is this drive to differentiate positively one’s group from others that
leads to competition with and discrimination against outgroup members. 27 This only occurs
under certain circumstances, however. Far from being a linear theory, “SIT predicts a
positive correlation between in-group identification and in-group bias, not as a main effect
but as an interactive outcome of several factors.” 28 Competition or discrimination towards
outgroups is particularly tied to situations that stress intense comparison, such as “where
23
Partly quoting Henri Tajfel, in Penelope Oakes, “Psychological Groups and Political Psychology,” p. 813.
Itesh Sachdev and Richard Y. Bourhis, “Social Categorization and Power Differentials in Group Relations,”
European Journal of Social Psychology, vol. 15, 1985, p. 416.
25
Gérard Lemaine, “Social Differentiation and Social Originality,” European Journal of Social Psychology, vol.
4, no. 1, 1974, p. 25.
26
Penelope J. Oakes and John C. Turner, “Social Categorization and Intergroup Behaviour: Does Minimal
Intergroup Discrimination Make Social Identity More Positive,” European Journal of Social Psychology, vol. 10,
1980, p. 295.
27
The social psychological reasoning behind SIT has been echoed in some analyses in the field of conflict
studies. In a seminal piece on ethnic conflict, Donald Horowitz saw ‘the affirmation of self worth’ as a
determining factor for human behaviour and conflict. For Horowitz: “self-esteem is in large measure a function
of the esteem accorded to groups of which one is a member […] the assessment of collective merit […]
proceeds by comparison, [and in unranked systems, by definition, relative group worth] remains enduringly
uncertain [… hence] the sources of ethnic conflict reside, above all, in the struggle for relative group worth.”
Quoted in Larry Diamond, “Ethnicity and Ethnic Conflict,” The Journal of Modern African Studies, vol. 25, no. 1,
1987, pp. 120-121. In attempts to valorise one’s group, others’ attributes are devalued, a dynamic that can lead
to ethnocentrism and, if exacerbated, to conflict among groups.
28
Craig McGarty, “Social Identity Theory Does Not Maintain that Identification Produces Bias, and Selfcategorization Theory Does Not Maintain that Salience is Identification: Two Comments on Mummendey, Klink
and Brown,” British Journal of Social Psychology, vol. 40, 2001, p. 174. These factors include the
internalisation of group categorisations; the existence of relevant and relational aspects for intergroup
comparison; and the relevance of the outgroup as an object of comparison. Ibid.
24
45
the status relations between identity-relevant groups are perceived to be illegitimate,
perhaps unstable, and the boundaries between groups impermeable.” 29
SIT focuses on certain dynamics of intergroup behaviour. Yet it fails to provide a
full understanding of group dynamics. SIT takes for granted the existence of groups and
their related identities and categories as experienced by individuals. Accordingly, SIT
does not address how individuals are able to act as a group. 30 Nor does it explain which
categories matter to individuals. For that, an explanation of ingroup dynamics is
necessary.
Self-categorisation Theory
Self-categorisation Theory (SCT), developed by John Turner and his collaborators,
addressed these questions. 31 The core assumptions behind SCT form an elaborate
understanding of an individual’s identity, or the self. According to SCT, the self is a
cognitive system that serves to process information. Included in this system are selfconcepts, images or representations individuals develop of themselves. 32 Individuals thus
possess a number of self-concepts. Each of these is an integrated part of the self, but
remains fundamentally different from the others and functions independently from them.
Because they function independently, self-concepts can vary in terms of their expression.
They can come to predominate or be made salient above others, depending on a given
situation or context. As Turner explained, “[a]ny particular self-concept (of those belonging
to any given individual) tends to become salient (activated, cognitively prepotent,
29
Penelope Oakes, “Psychological Groups and Political Psychology,” p. 813.
John C. Turner, with Michael A. Hogg, Penelope J. Oakes, Stephen D. Reicher and Margaret S. Wetherell,
Rediscovering the Social Group, p. 42.
31
SCT is sometimes also referred to as the Social identity theory of the group.
32
Ibid., p. 44.
30
46
operative) as a function of an interaction between the characteristics of the perceiver and
the situation.” 33
It is these self-concepts that form categories. These self-categorisations are
ordered hierarchically. They fall along a range of abstraction based on their degree of
inclusiveness. For example, wooden chairs and leather chairs belong to the same basic
category, while “’[c]hairs’ and ‘tables’ are categories at an intermediate level of
inclusiveness, which are themselves members of the higher order category of ‘furniture’.” 34
Turner identified three main levels of abstraction for self-categorisations. At the higher
level, or superordinate level, are representations of the self as a human being, or elements
that define one as a member of the human species. At the intermediate level are ingroupoutgroup categorisations that differentiate between groups of humans. Some of the selfcategorisations at this level include nationality, gender or occupation, for example. Finally,
at the last, or subordinate, level, are categorisations pertaining exclusively to the
individual, for example one’s specific personality. As Turner summarised, “[t]hese levels
can be said to define one’s ‘human’, ‘social’ and ‘personal’ identity respectively, based on
inter-species, intergroup (i.e. intra-species) and interpersonal (i.e. intragroup) comparisons
between oneself and others.” 35
Categories become relevant and operative when they are internalised by
individuals. SCT proponents did not, however, describe the internalisation process at
length. They did suggest that two primary modes of internalisation are “(1) simply as a
result of persuasive communications from credible, prestigious, or attractive others (or in
the terms of the present theory, from others with whom they identify), and (2) on the basis
33
Ibid.
Ibid., p. 45.
35
Ibid.
34
47
of public behaviour as group members leading to private self-attitude change.” 36 Both
modes recall Gabriel Tarde’s laws of imitation.
Once internalised, categories come to play a role in an individual’s identification
process. They do so varyingly, however. None of these forms of categorisations are
predominant, but they can be independently activated by contextual factors. Identification
is a situational and relational process. Situations act as cues that call up different
categories if relevant self-identifications exist. 37 More importantly, what SCT claims is that
the activation of a self-categorisation is always done with reference to a higher level of
categorisation. 38
For apples to be perceived as apples, for instance, they must be
compared with other objects. According to this principle, for the comparison to make
sense or be relevant, it must be made according to a higher level which includes all
compared objects. Apples are perceived as apples, or to share traits of ‘appleness’ in
comparison to oranges, which share the essence of being oranges, because both apples
and oranges belong to the family of fruits. It is relevant to distinguish among fruits in this
36
John C. Turner, with Michael A. Hogg, Penelope J. Oakes, Stephen D. Reicher and Margaret S. Wetherell,
Rediscovering the Social Group, p. 53.
37
According to Turner and his collaborators, “the salience of some ingroup-outgroup categorization in a
specific situation is a function of an interaction between the ‘relative accessibility’ of that categorization for the
perceiver and the ‘fit’ between the stimulus input and category specifications.” John C. Turner, with Michael A.
Hogg, Penelope J. Oakes, Stephen D. Reicher and Margaret S. Wetherell, Rediscovering the Social Group, p.
54. It should be noted that later iterations of SCT employ ‘perceiver readiness’ instead of ‘relative
accessibility’. Do categories exist and are they easily identifiable or accessible in terms of a frame of
reference? If so, are they pertinent in the given situation, do they fit the given situation? Haslam et al.
differentiated between fit with comparative and normative aspects of reality. As they explained, “[c]omparative
fit follows the principle of meta-contrast so that a given category (self- or otherwise will become salient to the
extent that it minimizes the difference perceived between people in the same category relative to differences
between category. This contextualizes categorization by tying it to judgements of relative differences in a
specific context. [...] Normative fit refers to the content-related aspects of this comparative match, such that a
given category will become salient where the observed similarities and differences are consistent with the
perceiver’s expectations about category meaning.” S. Alexander Haslam, John C. Turner, Penelope J. Oakes,
Katherine Reynolds, Rachael A. Eggins, Mark Nolan and Janet Tweedie, “When Do Stereotypes Become
Really Consensual? Investigating the Group-Based Dynamics of the Consensualization Process,” European
Journal of Social Psychology, vol. 28, 1998, p. 758. For both types of fits, the authors provided examples. As it
pertains to comparative fit, they explained that “for example, a number of Europeans are more likely to
categorize themselves in terms of a shared identity as ‘European’ (rather than as ‘French’ and ‘German’, say)
in a situation where Americans and Asians are also present rather than just different Europeans.” As for
normative fit, they explained that “[f]or example, in order to categorize oneself as Australian, Australians must
not only be observed to differ (in attitudes, actions, etc.) from a comparison group more than from each other
(comparative fit), but they must also do so in the right direction on relevant content dimensions of comparison
(perhaps if the comparison group were the Chinese, Australians should oppose nuclear testing and support
free speech).” Ibid., p. 758.
38
Ibid., p. 46.
48
case–the higher level of abstraction that bounds apples and oranges–, but not between
apples and chairs, or apples and cars, for example.
With respect to human self-
categorisations, Turner explained
[t]hat personal self-categorizations are based upon comparisons between self and
ingroup members (that interpersonal are intragroup comparisons), ingroup-outgroup
categorizations upon comparisons with other human beings (that intergroup are
intra-human comparisons) and human self-categorization are based upon
comparisons with other species in terms of some higher level identity. 39
Since comparisons are established in relation to the more inclusive level, it is at the
next lower level that salience is expressed. In other words, if fruits are compared, it is the
identity of apple or orange, for example, that is activated. In this process, then, various
categories can be activated. Individuals may, at times, think of themselves more in terms
of their personal, social or human identities. On average, however, people tend to situate
themselves at a mid-point of this range. As Turner and his collaborators explained, “[a]t
the midpoint of this continuum (where self-perception is likely to be located most of the
time) the individual will tend to define him- or herself as moderately different from ingroup
members, who in turn will be perceived as moderately different from outgroup members.” 40
Two principles stem from this. First, once a self-categorisation is activated, there is
a “perceptual accentuation of intra-class similarities and inter-class differences between
people as their characteristics are inferred from the defining identity of their class
membership.” 41 When a category becomes prominent, people are more aware of what
they share with other members of that category, but also more aware of differences with
outsiders to the group. Second, when one category is salient, it is salient above others. 42
While many have taken this to mean that SCT implies that identities and categorisations
per se are exclusive, it is not in terms of the categories that an antagonistic relationship
39
Ibid., p. 48.
Ibid., pp. 49-50.
41
Ibid., p. 49.
42
Ibid.
40
49
exists but in terms of their salience. SCT proponents do not conceive of identity as an ‘allor-nothing phenomenon’–quite the contrary. Their insistence on treating it as a continuum
stresses that self-categorisation “never fully embodies any one level but arises from a
dynamic, fluid process of conflict and compromise.” 43 In a subsequent article, Turner et.
al. further explained that “[m]ost of the time there are probably psychological and objective
factors making for the salience of multiple self-categories that may reinforce or conflict with
each other.” 44
What SCT claims then is that there exists an inverse relationship of salience
between levels, particularly between the salience of personal level self-categories and the
salience of the social level. In other words, thinking of oneself as unique and different from
others tends to make one perceive the group as less pertinent for comparison, blurring the
boundaries differentiating the group from others. The ingroup becomes less of a referent,
just as all others, groups or individuals, regardless of their affiliation, are simply perceived
as different from oneself and irrelevant for any form of comparison. The inverse is also
true. What stresses one’s affiliation to a group tends to displace more personal forms of
identity. As Jonathan Mercer stated: “[b]y identifying with a group, we de-emphasize our
personal identity and emphasize our group identity. We see ourselves as an exemplar of
the group rather than as an individual.” 45 Turner and his collaborators referred to this
process as depersonalisation. 46
It is this process of depersonalisation which is at the core of group phenomena
according to SCT theorists. As they argued, “group behaviour is assumed to express a
change in the level of abstraction of self-categorization in the direction which represents a
43
Quoting John Turner and Penelope Oakes, Penelope Oakes, “Psychological Groups and Political
Psychology,” p. 819.
44
Quoting John Turner et. al., Penelope Oakes, “Psychological Groups and Political Psychology,” p. 819.
45
Jonathan Mercer, “Anarchy and Identity,” pp. 244-245.
46
Turner et. al. defined depersonalisation as “the process of ‘self-stereotyping’ whereby people come to
perceive themselves more as the interchangeable exemplars of a social category than as unique personalities
defined by their individual differences from others.” John C. Turner, with Michael A. Hogg, Penelope J. Oakes,
Stephen D. Reicher and Margaret S. Wetherell, Rediscovering the Social Group, p. 50.
50
depersonalization of self-perception.” 47 Yet it should not be taken to mean the
disappearance of the self in the group or a form of regression, as crowd psychologists
argued. As Alberto Voci explained, “[t]his process is not considered to be negative, as it
leads individuals to join others, to assume socially relevant, shared traits, and to adopt a
consensual, and thus reassuring, cultural worldview.” 48
Theoretical frameworks such as Social Identity Theory and Self-categorisation
Theory are not flawless. SIT was scrutinised for the implications it drew from minimal
group experiments. For a large number of critics, problems exist with minimal group
experiments’ results.
For many, they are the direct consequence of the type of
environment in which subjects were placed. In experiments, groups always found
themselves in a situation in which the schemata through which they compared and
competed was oppositional, win or lose situations. They were in a zero-sum game. The
minimal group environment is, however, far removed from real world situations where
outcomes are rarely as definite and interactions between groups more complex. Hence,
the results are not as generalisable as first believed, or at least are valid only under very
specific conditions. So far, they have been proven only in instances in which a zero-sum
game environment is recreated. This is now recognised both by SIT and SCT theorists. 49
More broadly, SIT and SCT were criticised for not incorporating into their analyses
insights on meaning-work individuals do around identities–how they ascribe specific
significance and meaning to categories and identities. SIT and SCT concentrate instead
on mechanisms and dynamics of intergroup and ingroup dynamics. In a sense, these
approaches are about the procedural side of identity, and not its normative side. SCT does
47
Ibid.
Alberto Voci, “Relevance of Social Categories, Depersonalization and Group Processes,” p. 74.
49
Newer takes on SIT and SCT insist on the importance of situational factors in changing group dynamics. SIT
researchers increasingly studied factors behind competitive situations and pointed to the role played by the
status of groups and the permeability of categories. Penelope Oakes, “Psychological Groups and Political
Psychology,” p. 813. In a similar vein, SCT theorists were careful in raising the issue of context. Salience of
self-identification is situation-specific. Context is a fundamental factor in the shape self-group dynamics and
intergroup dynamics take.
48
51
not shed light on how individuals and groups choose among a multiplicity of identity traits,
interpret these or internalise them. More fundamentally, it does not explain how specific
schemas for comparison are adopted when differentiating between groups. SCT also
offers little on this issue; its analysis of category and group formation remains focused on
mechanisms and not meaning work. 50 While its insights on levels of identification and their
inverse salience proves insightful, it does not shed light on how individuals and groups
choose among a multiplicity of identity traits, interpret these or internalise them. Without
normative meaning, however, it becomes difficult to account for variations in the intensity
of individual attachment to the group and of intergroup rivalry. Meanings can affect the
intensity of sentiments an individual feels towards different identities at the personal,
collective and human level. Similarly, factors such as normative or cultural meanings and
their interpretation help explain why under similar circumstances groups sometimes fight
and under others they do not.
These limitations do not warrant rejecting SIT and SCT’s insights all together. On
the whole, social psychology can contribute to our understanding of collective behaviour.
In particular, SCT’s conceptualisation of identity as a continuum ranging from individual
forms of identification to more collective ones is critical for explaining identity shifts: how
individuals move from personal forms of identity to more collective ones, the source of
collective behaviour.
In terms of intergroup dynamics, the antagonistic dynamics
uncovered by minimal group experiments also offer an important insight for understanding
conflict. What this literature served to highlight is that a zero-sum type of environment, an
absolute form of opposition along one frame of reference or of comparison, more than
intergroup interactions per se, plays a part in competitive group behaviour.
50
Jack Snyder, “Anarchy and Culture: Insights from the Anthropology of War,” International Organization, vol.
56, no. 1, 2002, p. 27.
52
CHAPTER 4: SOCIAL MOVEMENT THEORY: FROM ITS PRECURSORS TO
CONTEMPORARY PERSPECTIVES
While social psychology made the psychological determinants of individual-group relations
and intergroup relations its focus, social movement theory studies collective action. It
examines why people choose, or are compelled, to join in collective enterprises and
mobilise in the name of collective goods or collective interests. At the heart of the study of
social movements, then, lies the problem of collective action. 1 Why would individuals
invest their time, efforts and resources in collective endeavours when they are not required
to do so, when others might do it for them, when they might not necessarily clearly gain
from it, but actually might even incur some risks? 2 Social movement theory, then, is
concerned with how the potential for action of individuals as a group is unleashed and, at
later stages sustained. This chapter focuses on the different trends in the study of social
movements. It looks at some of the precursors of social movement theory, as well as
social movement theory itself and a recent revision of it, the contentious politics
perspective.
1
At its most basic, the concept of collective action is intended to distinguish between activities undertaken by
individuals in isolation and collective enterprises. It therefore refers to the interactions of a number of
individuals, to the group. Although there exist many different definitions of collective action across the social
sciences, Ruth Meinzen-Dick, Monica Di Gregorio and Nancy McCarthy synthesised the most common
elements found in these. According to them, “[w]hat most definitions have in common is that collective action
requires the involvement of a group of people, it requires a shared interest within the group and it involves
some kind of common action which works in pursuit of that shared interest. Although not often mentioned, this
action should be voluntary, to distinguish collective action from hired or corvée labor.” Emphasis in original.
Ruth Meinzen-Dick, Monica Di Gregorio and Nancy McCarthy, “Methods for Studying Collective Action in Rural
Development,” CAPRi Working Paper no. 33, International Food Policy Research Institute, July 2004, pp. 4-5.
Available online at: http://www.capri.cgiar.org/pdf/capriwp33.pdf. Though this provides a generic definition of
what constitutes collective action, it is applied to varied types of activities. As Tarrow contended, “[c]ollective
action takes many forms–brief or sustained, institutionalised or disruptive, humdrum or dramatic.” Sidney
Tarrow, Power in Movement: Social Movement, Collective Action and Politics (Cambridge, Mass.: Cambridge
University Press, 1998), p. 2. He noted, however, that one important difference among the numerous types of
group activities is to be found between regular and contentious action. For Tarrow, contentious collective
action forms the basis of all social movements and revolutions. According to him, collective action “becomes
contention when it is used by people who lack regular access to institutions, act in the name of new or
unaccepted claims and behave in ways that fundamentally challenge others. It produces social movements
when social actors concert their actions around common claims in sustained sequences of interaction with
opponents or authorities.” Ibid. At the heart of contentious collective action, therefore, lie contestation and the
impulse for change.
2
Sidney Tarrow, Power in Movement, p. 6.
53
Important precursors
Modern approaches to collective behaviour, precursors to social movement theory, were
developed in the mid-twentieth century. Inspired by the Chicago school of sociology and
structural functionalism, they sought to devise scientifically consistent frameworks in
response to the previous eclecticism and lack of rigor around the study of collective
behaviour. 3 Important reflections at the time included those of Herber Blumer with his
symbolic interactionism, Neil Smelser, and Ted Robert Gurr and his Relative Deprivation
perspective. 4
Blumer coined the term ‘symbolic interactionism,’ which later developed into a
sophisticated theoretical perspective. 5
As its name underlines, symbolic interactionism
holds that individuals interact with each other through symbols (including language) which
represent meanings learnt from social sources. 6 Much like social constructivist thought, it
therefore maintains that humans are shaped by social structures and acquire their sense
and symbols from it, but that they are also shapers by continuing to interact with each
other and to change senses and meanings. 7 Blumer’s symbolic interactionist approach
emphasised mainly the agency or shaper component of the process. 8
As Blumer
explained, while individuals necessarily play out certain culturally and socially determined
3
For more on the Chicago school of sociology, see Anthony J. Cortese, “The Rise, Hegemony, and Decline of
the Chicago School of Sociology, 1892-1945,” The Social Science Journal, vol. 32, no. 3, 1995, p. 235-254;
Howard S. Becker, “The Chicago School, So-Called,” Qualitative Sociology, vol. 22, no. 1, 1999, pp. 3-12;
Daniel Breslau, “The Scientific Appropriation of Social Research: Robert Park’s Human Ecology and American
Sociology,” Theory and Society, vol. 19, 1990, p. 417-446; and Richard C. Helmes-Hayes, “Everett Hughes:
Theorist of the Second Chicago School,” International Journal of Politics, Culture and Society, vol. 11, no. 4,
1998, p. 621-673. For more on structural functionalism, see Talcott Parsons, The Social System (Glencoe, Ill.:
Free Press, 1951); Talcott Parsons, “An Outline of the Social System,” in Craig Calhoun et. al., eds., Classical
Sociological Theory (London: Blackwell Publishers, 2002); and A. Javier Treviño, “Parsons’ Action-System
Requisite Model and Weber’s Elective Affinity: A Convergence of Convenience,” Journal of Classical
Sociology, vol. 5, no. 3, 2005, p. 319-348.
4
In reality, Gurr’s work was published later than the the work of Blumer and Smelser.
5
According to Jean-Michel Chapoulie, although the term was first used by Blumer in 1937, it took twenty years
for it to become a central concept in sociology. Jean-Michel Chapoulie, “Everett Hughes and the Chicago
Tradition,” Sociological Theory, vol. 14, no. 1, 1996, p. 12, infra.
6
Robert M. Carrothers and Denzel E. Benson, “Symbolic Interactionism in Introductory Text-books: Coverage
and Pedagogical Implication,” Teaching Sociology, vol. 31, 2003, p. 163.
7
Ibid.
8
While the Chicago school focused on the agency part of the process, the Iowa school, and particularly its
most influential proponent, Manford Kuhn, focused on the ‘me’ component. Robert M. Carrothers and Denzel
E. Benson, “Symbolic Interactionism in Introductory Text-books,” p. 163.
54
roles, conventions, norms, etc., when they interact, they are also “directing, checking,
bending, and transforming their lines of action in the light of what they encounter in the
actions of others.” 9
Blumer applied this perspective to both individual and collective actors, such as
unions, corporations, gangs, or collective movements.
As he argued with regard to
movements, “a movement has to be constructed and has to carve out a career in what is
practically always an opposed, resistant, or at least indifferent world.” 10 This notion of
construction beyond emergence, of dynamic shaping beyond occurrence, guided the work
undertaken by symbolic interactionists. Blumer argued that “the methodological position of
symbolic interaction is that social action must be studied in terms of how it is formed [...] its
formation is a very different matter from the antecedent conditions that are taken as the
‘causes’ of social action.” 11
Symbolic interactionism remained influential in studies of collective behaviour,
particularly because it provided an alternative to structural functionalism, a perspective
very much wedded to the study of structural determinants of collective action and social
order. 12
Among the proponents of structural functionalism, Neil Smelser made collective
phenomena and social movements a prime topic of analysis. For Smelser, four
fundamental components are at the heart of collective action: ”(a) values, or general
sources of legitimacy; (b) norms, or regulatory standards for interaction; (c) mobilization of
individual motivation for organized action in roles and collectivities; and, (d) situational
9
Emphasis added. Quoted in Stanford M. Lyman, “Symbolic Interactionism and Macrosociology,” Sociological
Forum, vol. 3, no. 2, 1988, p. 300.
10
Martti Muukkonen, From Deviant Phenomenon to Collective Identity: Paradigm Shifts in Social Movement
Studies, master’s thesis, University of Joensuu, Joensuu, Finland, 1999, p. 37.
11
Quoted in Stanford M. Lyman, “Symbolic Interactionism and Macrosociology,” p. 300. Kevin Mihata
described symbolic interactionism as a perspective that “incorporates both self and other, both individual and
society, both agency and structure, and does so in a framework that is multi-dimensional, multilevel, and
dynamic.” Kevin Mihata, “Emergence and Complexity in Interactionism,” Symbolic Interaction, vol. 25, no. 4,
2002, p. 572.
12
Anselm Strauss, “Blumer on Industrialization and Social Change,” Contemporary Sociology, vol. 20, no. 2,
1991, p. 171, Stanford M. Lyman, “Symbolic Interactionism and Macrosociology,” p. 299.
55
facilities, or information, skills, tools, and obstacles in the pursuit of concrete goals.” 13
From these components, Smelser developed a typology of collective behaviour.
He
distinguished between value-oriented movements, “mobilized in the name of a generalized
belief envisioning a reconstitution of values”; norm-oriented movements, “mobilized in the
name of a generalized belief envisioning a reconstruction of norms”; hostile outbursts,
“mobilized on the basis of a generalized belief assigning responsibility for an undesirable
state of affairs to some agent”; and crazes or panics, “based on a generalized redefinition
of situational facilities”. 14
The fundamental trigger of collective behaviour, for Smelser, was structural strain,
disharmony in the social system. 15 Social movements, in this sense, are the result of
social systemic tensions. According to Smelser, social action followed a sequence: the
development of a structural conduciveness to action in the system; the occurrence of
structural strain; the growth and spread of a generalised belief; the interplay of
precipitating factors; the mobilisation of actors; and the failure of the means for social
control. 16
While an important contribution to the study of social movements, Smelser’s
approach remained wedded to a structural understanding of collective behaviour. This
structural emphasis eventually became a point of contention for a number of analysts of
collective behaviour, many seeing it as a neglect of other factors at play in collective
action.
Among these neglected factors were biological and/or psychological ones.
In
response to this, Ted Robert Gurr developed an approach to collective behaviour at the
13
Neil Smelser, Theory of Collective Behaviour (New York: Free Press of Glencoe, 1963), p. 9.
Ibid.
15
Stan Weeber and Daniel Rodeheaver defined structural strain as “the impairment of the relations among,
and consequent inadequate function of, the components of social action.” Stan C. Weeber and Daniel G.
Rodeheaver, “Militias at the Millennium: A Test of Smelser’s Theory of Collective Behavior,” The Sociological
Quarterly, vol. 44, no. 2, 2003, p. 184.
16
Ibid., p. 268.
14
56
cross-roads of structural and psychological perspectives. In 1970, Gurr published Why
Men Rebel, in which he developed his theory of Relative Deprivation. 17 The book
concentrated on disruptive, violent collective behaviour. Gurr believed that events such as
riots, rebellions and revolutions were the result of “perceived conditions of deprivation and
of the feelings of frustration associated with these perceptions.” 18 Negative perceptions of
objective conditions, in the form of a material or symbolic deprivation, led individuals to
rebel to redress this state. 19
To develop his model of Relative Deprivation, Gurr drew on perspectives from
across fields. In particular, he relied on psychology and psychologistic approaches for his
work on human motivations for violence. According to Gurr, the sources of motivation are
values “attributable or derived from basic needs”, of which he identified three types:
welfare values, power values and interpersonal values. 20 Welfare values are related to the
well-being of individuals, whether physically or emotionally, such as safety or selfrealisation. Power values refer to the influence people have on others or on institutions, or
to their ability to exercise their own actions outside of the influence of others. 21 Finally,
17
Ted Robert Gurr, Why Men Rebel (Princeton, N.J.: Princeton University Press, 1970). Often attributed to
Gurr, the notion of Relative Deprivation was not coined by him. It emerged in the United States where it began
being systematically employed around the 1940s. It is to be associated, in particular, with the publication of
the book The American Soldier. S. A. Stouffer, Edward A. Suchman, Leland C. DeVinney, Shirley A. Star and
Robin M. Williams, Jr., The American Soldier. Adjustment During Army Life (vol. 1) (Princeton, N.J.: Princeton
University Press, 1949). The book noted the discontent of a portion of American soldiers with regard to their
condition despite the fact that they received a good remuneration. As a result, it examined the role perceptions
play in the assessment of one’s situation. Authors such as Robert Merton, Alice S. Kitts and W.C. Runciman
reflected on similar phenomena to conclude that “those who felt most deprived were not the ones that
objectively were most deprived. The feeling of deprivation depends on whom someone is comparing himself
[to ... or on the] individual’s own past, some abstract ideal, or the standards articulated by his/her reference
group.” Martti Muukkonen, From Deviant Phenomenon to Collective Identity, p. 34. Relative Deprivation, thus,
reflects feelings of discontent borne out of a discrepancy between perceived expectations and objective
conditions.
18
Eduardo Canel, Eduardo Canel, “New Social Movement Theory and Resource Mobilization Theory: The
Need for Integration,” in Michael Kaufman and Haroldo Dilla Alfonso, eds., Community Power and Grassroots
Democracy: The Transformation of Social Life (Ottawa: IDRC/Zed, 1997), p. 191.
19
Patrick M. Regan and Daniel Norton, “Greed, Grievance, and Mobilization in Civil Wars,” Journal of Conflict
Resolution, vol. 49, no. 3, 2005, p. 320.
20
Ted Robert Gurr, Why Men Rebel, p. 25
21
As Gurr explained, “[p]ower values especially salient for political violence include the desire to participate in
collective decision-making–to vote, to take part in political competition, to become a member of the political
elite–and the related desires for self-determination and security, for example freedom from oppressive political
regulation or from disorder.” Ibid., pp. 25-26.
57
interpersonal values are tied to the social nature of human beings, to the satisfaction they
derive from their interactions with other individuals or groups. Based on these three sets
of values, individuals and collectivities develop expectations as to what they are entitled to,
both in the present and future. As Gurr argued: “[m]en ordinarily expect to keep what they
have; they also generally have a set of expectations and demands about what they should
have in the future, which is usually as much or more than what they have at present.” 22
On the other hand, Gurr understood aggression as a natural biological response to
frustration. Gurr’s frustration-aggression approach proposed that anger functions as a
drive. Basing himself on the work of Leonard Berkowitz, he asserted that “the perception
of frustration is said to arouse anger.” Aggression only occurs, however, if frustration can
be attributed to a source, an object or person. 23 A special case of frustration-aggression is
what Gurr referred to as the threat-aggression sequence.
Aggression, or violent
behaviour, can also be triggered in response to life threats, since “[a] threat to life is an
anticipated frustration. 24
With unfilled expectations tied to values (relative deprivation) as a source of
frustration and frustration as a source of aggression, it follows that conditions that increase
the expectations but not the capabilities (aspiration deprivation); that decrease the
capabilities but not the expectations (decrimental deprivation); or, finally, that increase
values but simultaneously decrease capabilities (progressive deprivation), intensify
discontent. 25 What determines if violent, disruptive collective behaviour erupts is the scope
and intensity of relative deprivation. 26
Amorphous or diffused frustration in society
develops into discontent, and, if widespread enough, into collective discontent. The
22
Ibid., p. 27.
Ibid., p. 34.
24
The last part is quoted from Leonard Berkowitz. Ted Robert Gurr, Why Men Rebel, p. 35.
25
Ibid., p. 57. Gurr defined Relative Deprivation as “a perceived discrepancy between men’s value
expectations and their value capabilities. Value expectations are the goods and conditions of life to which
people believe they are rightfully entitled. Value capabilities are the goods and conditions they think they are
capable of attaining or maintaining, given the social means available to them.” Ibid., p. 13.
26
Ibid., p. 15.
23
58
situation intensifies if collective discontent is politicised and blame is attributed. Frustration
transforms into collective action when violent action is directed against specific targets of
resentment, political objects and actors to whom blame has been attributed. 27
The
potential for collective political violence is, therefore, the result of the interplay of societal
variables affecting the distribution of resources, of perceptual factors involved in
interpretations of the situation, particularly the pattern of distribution, and of how these
perceptions are recuperated for action.
To this day, Gurr’s frustration-aggression model is still referred to in studies of
conflict and violence. Many of the concepts Gurr developed also influenced social
movement researchers. 28
His model identified opportunities and grievances as
fundamental factors of collective action, as well as to the translation of grievances into
action by blame attribution. 29 These remain at the heart of North American social
movement theory. Similarly, Blumer’s symbolic interactionism influenced the work of
authors such as Erving Goffman, Ralph Turner, Lewis Killian, and, more recently, David
27
Ibid., pp. 12-13.
As Walker and Smith argued, “By the 1980s, [Relative Deprivation] fell into disfavour and disrepute, partly
because of devastating reviews by McPhail (1971) and Gurney and Tierney (1982). Subsequent social
movement research relied almost exclusively on concepts such as resource mobilization to explain when and
why people engage in collective behaviour. The 1990s, though, saw the rediscovery of RD and its integration
into theories of collective behaviour. The ways in which people interpret grievance–central to RD–are now
recognized as essential to a full understanding of social movement participation.” Quoted in Yanick Noiseux,
Comment penser les mouvements sociaux dans les Amériques?, Figure 5. For some of the newer
perspectives on Relative Deprivation, see for example, Bert Klandermans, The Social Psychology of Protest
(Oxford: Blackwell Publishers, 1997); or Iain Walker and Heather J. Smith, eds., Relative Deprivation Theory:
Specification, Development, and Integration (New York: Cambridge University Press, 2002).
29
Opportunities are the societal factors that contribute to the emergence of sentiments of discontent, to their
articulation or the articulation of blame, or, in the end, to collective violence. Gurr highlighted a number of
these in his analysis, including “the extent of cultural and subcultural sanctions for overt aggression, the extent
and degree of success of past political violence, the articulation and dissemination of symbolic appeals
justifying violence, the legitimacy of the political system, and the kinds of responses it makes and has made to
relative deprivation,” such as coercive control or institutional support. Ibid., p. 13. A grievance, a term often
mistaken for dissatisfaction, discontent or deprivation, is defined by Robin Williams Jr. as “’a wrong considered
as grounds for complaint’–that is, a claim is made that an injustice has occurred. Just as a claim is a socially
legitimated demand, not just as desire or want, so a grievance is a normative protest, not just dissatisfaction.”
The first part of the quotation is taken from The Dictionary of the English Language published by Random
House. Robin M. Williams, Jr., “The Sociology of Ethnic Conflict: Comparative International Perspectives,”
Annual Review of Sociology, vol. 20, 1994, p. 58. He further indicated that, based on an analysis by Gurr
published in 1993, “grievances are created by poverty and political and economic differentials among
communal groups, and that restricted political access and a history of lost autonomy are important conditions
for separatist demands and rebellion.” Ibid., p. 62. Grievances, therefore, take many forms, such as economic
or political, for example.
28
59
Snow. 30 His focus on symbols communicated by volitional actors paved the way for work
on discourses aimed at triggering action, and more recently work on framing. The next
section addresses contemporary North American social movement theory and its political
opportunities, mobilising structures and framing approach.
The Political Opportunities, Mobilising Structures and Framing Approach
North American social movement theory established itself largely with the publication of
Sidney Tarrow’s Power in Movement and Doug McAdam, John McCarthy and Mayer
Zald’s Comparative Perspectives on Social Movements. 31 Incorporating insights from
sociology, political science and history, as well as working across the rationalist-culturalist
divide, McAdam, McCarthy and Zald, the former previously a political process theorist and
the latter two both ex-proponents of the resource mobilisation perspective, moved the
study of collective behaviour forward towards a new integrated approach. 32 Their approach
constituted a hybrid perspective combining both a synthesis of the greater strengths of
resource mobilisation and the political process model, and some insights on movements’
expressive dynamics–identity and cultural dynamics–found in New Social Movements
30
Robert M. Carrothers and Denzel E. Benson, “Symbolic Interactionism in Introductory Text-books,” p. 163
and Kevin Mihata, “Emergence and Complexity in Interactionism,” p. 571.
31
Sidney Tarrow, Power in Movement. Doug McAdam, John D. McCarthy and Mayer N. Zald, eds.,
Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural
Framing (New York: Cambridge University Press, 1996). The spark for much of the reflection on the topic in
North America at the time came, as a matter of fact, from the publication of an influential analysis on collective
behaviour that adopted a strictly individualistic, utilitarian logic: Mancur Olson’s The Logic of Collective Action.
Mancur Olson, The Logic of Collective Action (Cambridge, Mass.: Harvard University Press, 1971). Resource
mobilisation, sometimes also referred to as the organisation-entrepreneurial approach, is a middle-range
perspective developed in the nineteen-sixties by John McCarthy and Mayer Zald. The term ‘political process’
became popular after the publication, in 1982, of a study by Doug McAdam on the black protest movement in
the United States, inspired from an article by Rule and Tilly published in 1975. Many social movement
theorists, such as Charles Tilly, William Gamson and Anthony Oberschall, developed a framework around this
concept. At the heart of it are the efforts of movements to achieve integration into the political system. Thus,
much more than a focus on movement’s internal dynamics, political process theory centers on the goals of
those excluded from the polity, referred to as challengers, to gain recognition as legitimate political players in
the polity and, therefore, on their interactions with the political system.
32
Doug McAdam, John D. McCarthy, and Mayer N. Zald, “Introduction: Opportunities, Mobilizing Structures,
and Framing Processes–Towards a Synthetic, Comparative Perspectives on Social Movements,” in Doug
McAdam, John D. McCarthy and Mayer N. Zald, eds., Comparative Perspectives on Social Movements:
Political Opportunities, Mobilizing Structures, and Cultural Framing (New York: Cambridge University Press,
1996), p. 2.
60
theory. 33 In this synthesis, they identified three factors responsible for the emergence and
sustained action of social movements: “(1) the structure of political opportunities and
constraints confronting the movement; (2) the forms of organization (informal as well as
formal), available to insurgents; and (3) the collective processes of interpretation,
attribution, and social construction that mediate between opportunity and action.” 34 These
came to be better known as political opportunities, mobilising structures and framing
processes.
McAdam, McCarthy and Zald built on the existing concepts of political opportunities
and organisation found in works such as Gurr’s or in resource mobilisation and the political
process perspective. In their view, the emergence of social movements resulted from
changing opportunities. Movements are formed in response to changes in their
environment which affect their prospects and afford them new opportunities. 35 Sidney
Tarrow, on the other hand, developed the concept of ‘cycles of protest’, periods of
turbulence and re-alignment. 36 The occurrence of turbulence and the actions of earlyrisers, like a snow ball effect, facilitate the opening up of new spaces of contention for
other actors. They also create opportunities. Accordingly, preliminary change is selfreinforcing. It follows a logic of positive feedback and can lead to a sweeping cycle of
protest. More than stand alone structural factors, however, opportunities also have a
perceptual component. Much like Gurr’s sense of relative deprivation, it is the perceptions
33
New Social Movement theories (NSM), the mainly European stance on social movements is not reviewed in
this dissertation. Summarily, New Social Movement theory is a perspective that developed in continental
Europe based, among others, on the work of Alain Touraine. New Social Movement theories tends to stress
the identitary and cultural aspects of movements and their expressive action in the creation of new meanings.
One of the main proponents of NSM is Alberto Melucci. Ulf Hjelmar described Melucci’s work as “seeing
collective phenomena as the outcomes of various processes through which actors produce meanings,
negotiate, and make decisions. This constructivist view of collective action places emphasis on the creative
power of individuals acting collectively. Collective behaviour, in Melucci’s view, is the result of the ability of
individuals to define the content of the political struggle and organize their common behavior.” Emphasis in
original. Ulf Hjelmar, “Constructivist Analysis and Movement Organizations: Conceptual Clarifications,” Acta
Sociologica, vol. 39, 1996, pp. 172-173.
34
Doug McAdam, John D. McCarthy, and Mayer N. Zald, “Introduction,”, p. 2.
35
Sidney Tarrow, Power in Movement, p. 17.
36
Ibid., p. 18.
61
and interpretations of actors which construe changes in the system as opportunities or
not. 37
FIGURE 4.1: THE OPPORTUNITY, MOBILISING STRUCTURES, FRAMING FRAMEWORK
SOCIAL CHANGE
MOBILIZING
STRUCTURES
OPPORTUNITY
AND
THREAT
FRAMING
PROCESSES
REPERTOIRES
OF
CONTENTION
CONTENTIOUS
INTERACTION
Figure reproduced from Doug McAdam, Sidney Tarrow and Charles Tilly, Dynamics of Contention (New York:
Cambridge University Press, 2001), p. 17.
37
Jeff Goodwin and James M. Jasper, “Caught in a Winding, Snarling Vine: The Structural Bias of Political
Process Theory,” Sociological Forum, vol. 14, no. 1, 1999, p. 33.
62
Beyond opportunities for movement emergence, contemporary social movement
theorists also focused on movement structures. Previous perspectives primarily studied
well structured organisations (or more precisely the structured rational entity at the head of
the movement, the social movement organisation or SMOs, in the case of resource
mobilisation theory) and their dynamics. The political process model, on the other hand,
underlined the importance of pre-existing unstructured social networks for collective action.
In the new synthesis proposed by McAdam, McCarty, Zald and Tarrow, greater importance
was therefore attributed to informal networks as both the launching platform for more
formal structures, as well as vehicles to sustain collective action. Such pre-existing social
ties serve as preliminary outlets through which individuals can express and structure
dissent. Once the movement begins toform, these structures continue to play a strategic
role. Their integration within the formal movement structure ensures a strong horizontal
connectedness at the base, while also allowing for deeper attachment to the movement as
a result of these pre-existing links. 38
While McAdam, McCarthy and Zald’s conceptualisation of political opportunities
and mobilising structures drew on previous perspectives such as resource mobilisation
theory and the political process perspective, framing, their third core concept, was the
more novel addition to North American frameworks on social movements. Framing, the
strategic formulation of messages on the part of individuals and movements, is the subject
of the next section.
Framing
The concept of ‘frame’ was originally coined in 1974 by Erving Goffman in his frame
analytic perspective. Goffman used the term to convey dynamics of representation,
38
Sidney Tarrow, Power in Movement, p. 136.
63
interpretation and meaning. 39 In the field of social movements, Erving Goffman’s insights
were first recognised by William Gamson and David Snow et. al. 40 For Snow and his
collaborators, frames served as tools to interpret and represent meanings and emotions.
Drawing on Goffman, Snow and Robert Benford defined frames as “interpretive schemata
that simplifies and condenses the ‘world out there’ by selectively punctuating and encoding
objects, situations, events, experiences, and sequences of action within one’s past or
present environment.” 41 More than a depiction of objects, actions and events, however,
collective action frames are aimed at motivating action. 42 More specifically, framing efforts
are directed at interpreting, translating and guiding the emotions and perceptions of
individuals towards convergence with movement goals and activities.
Snow et. al. identified three functions played by frames: diagnosis, prognosis and
motivation. 43 Diagnosis framing pertains to how the situation is evaluated and conveyed.
As Benford and Snow argued, it relates to how “adherents negotiate a shared
understanding of some problematic condition or situation they define as in need of
change.” 44
Oftentimes, diagnosis framing takes the form of what Gamson called an
39
Erving Goffman, Frame Analysis: An Essay on the Organization of Experience (Cambridge, Mass.: Harvard
University Press, 1974).
40
Jeff Goodwin and James M. Jasper, “Caught in a Winding, Snarling Vine,” p. 47.
41
David A. Snow and Robert D. Benford, “Master Frames and Cycles of Protest,” in Aldon D. Morris and Carol
McClurg Mueller, eds., Frontiers in Social Movement Theory (New Haven, Conn.: Yale University Press, 1992),
p. 137. Although Snow and Benford defined frames as ‘schemata’ in their seminal 1992 piece and in previous
ones, they also insisted in their later work on the fact that schemata should not be confused with ‘schemas.’
Schema is a concept employed in cognitive psychology and represent “‘participants’ expectations about
people, objects, events, and settings in the world [and should be] distinguished from alignments being
negotiated in particular interaction,’ which is what frames do.” Partly quoting Tannen and Wallat. Infra, Robert
D. Benford and David A. Snow, “Framing Processes and Social Movements: An Overview and Assessment,”
Annual Review of Sociology, vol. 26, 2000, p. 614.
42
Doug McAdam, John D. McCarthy, and Mayer N. Zald, “Introduction,” p. 6. Benford and Snow, in 2000,
described the strategic version of framing processes employed by social movements as opposed to discursive
and contested approaches. As they explained, “[b]y strategic processes, we refer to framing processes that
are deliberative, utilitarian, and goal directed: Frames are developed and deployed to achieve a specific
purpose–to recruit new members, to mobilize adherents, to acquire resources, and so forth.” Robert D.
Benford and David A. Snow, “Framing Processes and Social Movements,” p. 624. Marc W. Steinberg, “Tilting
the Frame: Considerations on Collective Action Framing from a Discursive Turn,” Theory and Society, vol. 27,
1998, pp. 845-846.
43
While Snow et. al. subdivided core framing tasks into these three categories, specific frames and frame
packages can serve all three functions or a combination of them.
44
Robert D. Benford and David A. Snow, “Framing Processes and Social Movements,” p. 615.
64
‘injustice frame’. 45 Diagnosis does not end at naming grievances and injustice, however.
Nor does this suffice. 46
Grievances and injustice need further direction to serve as a
launching platform for collective action. 47 Diagnosis framing also entails, then, channelling
discontent and resentment, and focusing the blame on a specific object or actor. 48
Prognosis framing is the articulation of solutions to identified problems or injustices.
Hence, a prognosis frame entails the identification of reasonable solutions and “proposing
specific strategies, tactics, and objectives by which these solutions may be achieved.” 49
The final core function, motivation, is designed to move adherents from consensus
mobilisation, or agreement on the problem, its sources and ways of addressing the
situation, to active mobilisation. According to Benford, specific vocabularies motivate more
effectively. These include vocabularies of severity, urgency, efficacy, and propriety. 50
Severity and urgency underscore the need for collective action (the situation is grave, for
example, in the case of severity; and it really requires something to be urgently done, for
example in the case of urgency). Efficacy and propriety, on the other hand, provide the
rationale for collective endeavours (together, we can succeed; collective action affords us
more chances of success, for example, in the case of efficacy; and leaguing ourselves in
this endeavour is the appropriate, even dutiful, course of action, for example, in the case of
propriety).
Frames are rarely produced in the absence of pre-existing referents. Shared
backgrounds, cultural referents, social understandings provide an existing repertoire from
45
Francesca Polletta, ’It Was Like a Fever...’ Narrative and Identity in Social Protest,” Social Problems, vol. 45,
no. 2, 1998, p. 139. Robert Benford and David Snow, in support of Gamson’s injustice frame concept, indicate
that “[a] plethora of studies call attention to the ways in which movements identify the ‘victims’ of a given
injustice and amplify their victimisation.” Robert D. Benford and David A. Snow, “Framing Processes and
Social Movements,” p. 615.
46
Robert D. Benford and David A. Snow, “Framing Processes and Social Movements,” pp. 615-616.
47
Ibid., p. 616.
48
David A. Snow, E. Burke Rochford, Jr., Steven K. Worden, Robert D. Benford, “Frame Alignment Processes,
Micromobilization, and Movement Participation,” American Sociological Review, vol. 51, 1986, p. 474.
49
Kimberly Fisher, “Locating Frames in the Discursive Universe,” Sociological Research Online, vol. 2, no. 3,
1997, p. 6.
50
See Robert D. Benford, “‘You Could be the Hundredth Monkey’: Collective Action Frames and Vocabularies
of Motive within the Nuclear Disarmament Movement,” Sociological Quarterly, vol. 34, 1993, pp. 195-216.
65
which they are built. Movements are, therefore, ‘consumers’ of these cultural meanings
and ‘producers’ of new ones. 51 They draw on “a cultural toolkit of possible symbols [and ...]
choose those that they hope will mediate among the cultural underpinnings of the groups
they appeal to, the sources of official culture and the militants of their movements–and still
reflect their own beliefs and aspirations.” 52 They reinterpret cultural or social referents, as
to align them to their goals and interests. 53
Although cultural backgrounds are a resource to build appeals and claims, they
also tend to constrain the latitude of movement leaders in their choice of referents. 54
Stuart Kaufman explained that a collective background “is malleable to some degree, [but]
typically has root in history and culture that cannot be easily overridden [...] politicians’
ability to use [these] instrumentally is, therefore, limited by the cultural context in which
they operate.” 55 Movement entrepreneurs cannot continuously invent novel meanings and
means, since they also are part and parcel of the social environment, and necessarily
socialised to it. 56
This is true of publics too. Individuals are not absolute ‘free thinkers’, but are
bound by their cultural and social backgrounds. 57 These backgrounds are also their pool of
plausible resources: it constitutes the repertoire of what has been accepted, internalised
and makes sense. To make sense, then, frames must resonate with elements the
51
Sidney Tarrow, Power in Movement, p. 123.
Ibid., p. 122.
53
Snow et. al. call this frame alignment, “the linkage of individual and SMO interpretations, such that some set
of individual interests, values, and beliefs and SMO activities, goals, and ideology are congruent and
complementary.” David A. Snow, E. Burke Rochford, Jr., Steven K. Worden, Robert D. Benford, “Frame
Alignment Processes, Micromobilization, and Movement Participation,” p. 464. Snow et. al. subdivided frame
alignment into four different processes: frame bridging, frame amplification, frame extension, and frame
transformation. For a description of these, see Ibid., pp. 467-476.
54
Quoted in Jeff Goodwin and James M. Jasper, “Caught in a Winding, Snarling Vine,” p. 48.
55
Stuart J. Kaufman, “Symbolic Politics or Rational Choice: Testing Theories of Extreme Ethnic Violence,”
International Security, vol. 30, no. 4, 2006, pp. 50-51.
56
Steven Majstorovic explained, for example, that “ethnic entrepreneurs have to confront the reality of shared
historical experience, and the emotional symbols available for political manipulation cannot just be invented out
of thin air but must have socio-historical resonance.” Steven Majstorovic, “Ancient Hatreds or Elite
Manipulation? Memory and Politics in the Former Yugoslavia,” World Affairs, vol. 159, no. 4, 1997, p. 172.
57
Emphasis in original. Serge Moscovici, quoted in Kimberly Fisher, “Locating Frames in the Discursive
Universe,” p. 12.
52
66
population is familiar with, what has been previously established as common referents and
norms. 58
As Michael Barnett explained, “to legitimate and to make plausible [...] policies
requires demonstrating how they are consistent with the cultural terrain” shared by
individuals. 59 Frame resonance is analogous to literature on cognitive consistency in
psychology which asserts that actors are more likely to accept new ideas if the latter are
shown to conform with already established ones. 60
Two factors determine the degree to which a frame resonates: credibility and
relative salience. 61 Benford and Snow identified three forms of credibility: first, frame
consistency, “the congruency between an SMO’s articulated beliefs, claims and actions,”
i.e. the frame articulated should constitute a coherent ensemble, from diagnosis to
proposed solutions; second, empirical credibility, which “refers to the apparent fit between
the framings and events in the world”, a factor closely related to cognitive consistency; and
finally, “the perceived of credibility of frame articulators.” 62 Like Tarde, Benford and Snow
argued that factors such as status or knowledge lend more credibility to communicators,
which in turn make them more persuasive. 63
The other factor at play in resonance is a frame’s salience. Benford and Snow
identified three dimensions of salience: centrality, experiential commensurability, and
narrative fidelity. The first, centrality, pertains to “how essential the beliefs, values, and
ideas
58
Rodger Payne, “Persuasion, Frames and Norm Construction,” European Journal of International Relations,
vol. 7, no. 1, 2001, p. 38.
59
Michael Barnett, “Culture, Strategy and Foreign Policy Change: Israel’s Road to Oslo,” European Journal of
International Relations, vol. 5, no. 1, 1999, p. 15.
60
Sidney Tarrow, Power in Movement, p. 43.
61
Robert D. Benford and David A. Snow, “Framing Processes and Social Movements,” p. 619.
62
Ibid., pp. 619-621.
63
Ibid., pp. 620-621. The communicator is effectively taken to be an important factor in successful
communication in research on communication. Persuasive communicators are those perceived as trustworthy
and experts in the field. See Melvin L. Defleur and Shearon A. Lowery, Milestones in Mass Communication
Research, 3rd Edition (White Plains, N.Y.: Longman, 1994), pp. 165-188.
67
TABLE 4.1: FACTORS AT PLAY IN FRAME RESONANCE
FACTORS
AFFECTING
FRAME
CREDIBILITY
FACTORS
AFFECTING
FRAME
RELATIVE
SALIENCE
FRAME
CONSISTENCY
CENTRALITY
EMPIRICAL
CREDIBILITY
EXPERIENTIAL
COMMENSURABIL
ITY
CREDIBILITY OF
FRAME
ARTICULATORS
NARRATIVE
FIDELITY
associated with movement frames are to the lives of the targets of mobilization.” 64 Since
beliefs and values are ordered hierarchically, much as Maslow ordered human needs to
understand motivation, frames addressing issues that are judged important or central have
more traction with publics than those of lesser interest or value. 65 Experiential
commensurability is related to a frame’s congruence with individuals’ lived experiences.
More than simply an ideational fit between movements and individuals, resonance also
relates to whether frames are compatible with what people see, feel, and sense on a
practical level. 66 The final factor determining salience is narrative fidelity. Narrative fidelity
pertains to the degree to which frames correspond to predominant, and thus widely held or
accepted, ideological views. 67
While an important improvement on previous collective behaviour approaches,
many have criticised framing, as adopted by McAdam, McCarty, and Zald for its linearity.
As Clay Calvert contended, “many today in communication-science tradition reject the
64
Robert D. Benford and David A. Snow, “Framing Processes and Social Movements,” p. 621.
For Abraham Maslow’s hierarchy of needs, see Abraham H. Maslow, “A Theory of Human Motivation,”
Psychological Review, vol. 50, 1943, pp. 370-396.
66
David A. Snow, E. Burke Rochford, Jr., Steven K. Worden, Robert D. Benford, “Frame Alignment Processes,
Micromobilization, and Movement Participation,” p. 477.
67
Marc W. Steinberg, “Tilting the Frame,” p. 846.
65
68
magic bullet theory of direct, powerful, and uniform changes in attitudes and behaviors
caused by messages.” 68 Not only is there usually more than one group appealing to a
population simultaneously, which generally implies a variety of competing interpretations to
choose from, but framing efforts can never be expected to be perfectly effective and
‘resonant.’ Conscious of this criticism, some social movement analysts turned their efforts
to devising a more contingent and dynamic approach.
Dynamics of Contention
Drawing on insights from both social movement theory and revolution studies, Doug
McAdam, Sidney Tarrow and Charles Tilly set out to study contentious politics, “episodic,
public, collective interaction among makers of claims and their objects when (a) at least
one government is a claimant, an object of claims, or a party to the claims and (b) the
claims would, if realized, affect the interests of at least one of the claimants.” 69 Their joint
enterprise came as a response to the overly static nature of the traditional political
opportunities, mobilising structures and framing model that two of them had helped
developed. Assuming that the highly contingent nature of contentious politics prevents the
elaboration of sweeping generalisations, McAdam, Tarrow and Tilly concentrated on
comparatively assessing various forms of contention through “the analysis of smaller-scale
68
Clay Calvert, “Hate Speech and Its Harms: A Communication Theory Perspective,” Journal of
Communication, vol. 47, no. 1, 1997, p. 9.
69
Doug McAdam, Sidney Tarrow and Charles Tilly, Dynamics of Contention (New York: Cambridge University
Press, 2001), p. 5. Researchers in revolution studies have developed concepts and frameworks similar to the
ones found in social movement literature. Describing the recent state of revolution studies, Jack A. Goldstone
stated that “[s]cholars called for greater attention to conscious agency, to the role of ideology and culture in
shaping revolutionary mobilization and objectives and to contingency in the course and outcome of
revolutions.” Jack A. Goldstone, “Toward a Fourth Generation of Revolutionary Theory,” Annual Review of
Political Science, vol. 4, 2001, p. 141. In particular, revolution studies have begun analysing the role played by
ideologies and ideational structures in fostering identities for mobilisation. Knowledgeable actors have been
described as drawing and ‘operating’ on larger ideological systems in order to frame their call for mobilisation.
For an example of this literature, see William H. Sewell, Jr., “Ideologies and Social Revolutions: Reflections on
the French Case,” The Journal of Modern History, vol. 57, no. 1, 1985, pp. 57-85. John Foran, “Theories of
Revolution Revisited: Toward a Fourth Generation,” Sociological Theory, vol. 11, no. 1, 1993, pp. 1-20.
69
causal mechanisms that recur in different combinations with different aggregate
consequences in varying historical settings.” 70
Their contentious politics agenda improved on the ‘classical’ social movement
approach in a number of ways. Overall, contentious actions are conceptualised as more
path-dependent processes than they had been previously. Once the protest process
starts, it can take a life of its own and evolve towards more polarised, radicalised or
conflictual perceptions. Adopting the path of protest, therefore, tends to invite a spiral of
more extreme forms of contention, involving a growing number of actors, allies and
contenders, while making a return to benign contention more difficult. They also revised
core concepts of the ‘classical’ social movement approach. Opportunities, in their
framework, are understood as much as a perceptual factor as a material one. They
become a negotiated assessment of perceived changes. Mobilising structures are also
more dynamic. Existing structures or emerging ones are located within competitive
dynamics opposing movements and their competitors. Furthermore, more than fixed
entities, organisations are conceptualised as shifting networks affected by struggles and
reinforcements on the part of adherents and competitors.
McAdam, Tilly and Tarrow also chose to adopt a more contingent take on framing.
They treat framing as a process during which actors strive to construct effective appeals,
despite environmental and structural constraints on their actions, as well as the competing
counter-framing of competitors and other actors. 71 This requires, on the part of leaders
and movement entrepreneurs, constant reassessments and readjustments of frames
strategies from beginning to end. These readjustments are required to weather
70
Doug McAdam, Sidney Tarrow and Charles Tilly, Dynamics of Contention, p. 24. The key concepts in their
approach are, therefore, mechanisms, processes and episodes. They define mechanisms as “a delimited
class of events that alter relations among specified sets of elements in identical or closely similar way over a
variety of situations”; processes as “regular sequences of such mechanisms that produce similar (generally
more complex and contingent) transformations of those elements”; and episodes as “continuous streams of
contention including collective claims making that bears on other parties’ interests.” Ibid., p. 24. It is within
these episodes that the authors locate mechanisms and processes.
71
Ibid., p. 48.
70
environmental changes and the efforts of competitors. Framing is, thus, not the finite
production of ‘a frame’, but an evolving effort ‘to frame’–not an instance but a process.
FIGURE 4.2: A DYNAMIC, INTERACTIVE FRAMEWORK FOR ANALYSING MOBILISATION IN
CONTENTIOUS POLITICS
MEMBER
ATTRIBUTION OF
THREAT/
OPPORTUNITY
ORGANISATIONAL
APPROPRIATION
BROAD CHANGE
PROCESSES
CHALLENGER
INNOVATIVE
COLLECTIVE
ACTION
ESCALATION
OF
PERCEIVED
UNCERTAINTY
ATTRIBUTION OF
THREAT/
OPPORTUNITY
SOCIAL
APPROPRIATION
INNOVATIVE
COLLECTIVE
ACTION
Figure reproduced from Doug McAdam, Sidney Tarrow and Charles Tilly, Dynamics of Contention (New York:
Cambridge University Press, 2001), p. 45.
The improvements effected by McAdam, Tarrow and Tilly on the classic social
movement agenda strive to make it a more dynamic tool for the study of contentious
71
politics. While a much needed and commendable effort, it remains according to a number
of authors a limited one.
While McAdam, Tarrow and Tilly devote more attention to
processes and interactions than was done previously, their approach remains, however,
overwhelmingly focused on those taking place between actors. Fundamental interactions
between structures and agents are mentioned, but never sufficiently examined to give a
full sense of the dynamics involved in contention. Whether in terms of building contention,
engaging with constituents or challengers, or framing, much of the ideational elements
remain overshadowed by this focus on agents. For a dynamic approach, McAdam, Tarrow
and Tilly’s surprisingly skirt the deeper interactive linkages between actors and language,
ideology and culture. 72
On the one hand, contentious politics analysts misunderstand the nature of
expressive structures. Discursive and ideational resources are not as malleable as
material resources. 73 They are not replaceable, interchangeable or fungible. They are not
the possession of an individual, but are intersubjective and intergenerational. They belong
to specific historical and cultural contexts–an insight found in original framing literature, but
understated in the contentious politics program. With communication and framing so
fundamentally tied to collective backgrounds, it is not a free marketplace of ideas that
72
In terms of what could be considered the ‘expressive’ side of contention, they mostly focused on repertoires
of contention and on identity. Repertoires of contention–a concept previously coined by Tilly–are shared
understandings around forms and meanings of protest on which actors draw, build and innovate to construct
their own claims and actions. McAdam, Tarrow and Tilly revised their view of repertoires of contention to
include the role played by actors outside of the movement in terms of extension of repertoires, be they the
audience, members of the polity, the media, or rivals or enemies. As a result, repertoires evolve not only out of
innovation on the part of movement participants, but also through struggles in which they engage with other
agents around actions, meanings and ideas. Repertoires, therefore, emerge of the negotiation process
between contenders, which tends to further limit movement entrepreneurs’ leeway in strategically invoking and
innovating on previous repertoires. McAdam, Tarrow and Tilly also revisited the concept of identity. For these
authors, “[i]dentities in general consist of social relations and their representations, as seen from the
perspective of one actor or another.” Doug McAdam, Sidney Tarrow and Charles Tilly, Dynamics of
Contention, p. 133. Defined as social relations developed over contexts of social negotiation and struggle,
identities, have a political side, whether actualised or latent, since they raise issues of entitlement and the
common good. They are, as a result, a platform for claims-making. While McAdam, Tarrow and Tilly limit their
understanding of political or politicised identities to those calling forth claims and interactions with
governments, this implies that they are a central component to contentious politics.
73
Rhys H. Williams, “Constructing the Public Good: Social Movements and Cultural Resources,” Social
Problems, vol. 42, no. 1, 1995, p. 127.
72
exists for movement entrepreneurs, but an imposing, constrained, conflictual, and
selective one, whether they are aware of it or not. The capacity to frame en masse is
more elusive than the classic social movement framework, or the contentious politics
approach, take it to be. More than a simple stratagem, framing should be understood as a
‘battle for symbolic encoding,’ with all the difficulties battling entails. 74
Despite these shortcomings, social movement theory, and particularly its recent
reformulation in the form of the contentious politics perspective, constitutes one of the only
synthetic approaches to the study of collective behaviour. In a context sensitive manner–
represented by opportunities–, it sheds light on movement emergence and endurance by
focusing on the actions of individuals. It examines both how individuals form networks
through interactions and the dynamic evolution of these networks, including the evolution
toward more structured forms of organisation. To a certain extent, social movement theory
and contentious politics also focus on meaning-work behind movements in their own way.
This transpires from their attention to framing processes. Framing is an important insight,
one that complements social psychological literature on sources of collective attachment.
While social psychology describes some of the cognitive and emotive reasons behind
collective attachment, framing highlights the social and strategic dynamics behind it. There
is undeniably a preliminary ‘recruitment’ stage, at which the individual adjudicates between
belonging to the group or not. While on the recruit’s side, much rests on comparing oneself
to the group (Am I like them? Do they represent my interests, my values?, etc.), group
leaders and groups themselves are rarely passive entities simply submitting themselves to
comparison. They, on the contrary, engage, construct rationales for their coming together,
build grounds for commonality, and frame these to align with potential members to
stimulate adherence to the group. Group dynamics are as much about social interactions,
negotiations and framing battles between leaders, potential group members, and even
74
Ann Swidler in Kimberly Fisher, “Locating Frames in the Discursive Universe,” p. 20.
73
challengers, as it is about social-psychological individual-group dynamics as described in
the previous chapter. The strength of social movement theory and contentious politics lies
therefore in their ability to account for these more social and strategic interactions.
74
CHAPTER 5: STRATEGIES FOR
POLITICISATION, SIMPLIFICATION
ELITE-LED
MOBILISATION:
OPPOSITION,
If any common thread can be drawn from social constructivist work on identity, social
psychology and newer takes on social movement theory, it is that group behaviour and
group relations are dynamic, interactive processes at the confluence of structural, social
and psychological factors. This makes them highly complex social phenomena in which
structures play a determining role, as do actors–whether the individual as a unit of the
group or the group as an agent in and of itself–by acting on these environmental structures
and on each other.
In the case of elite-led mobilisation, one of these actions is the communication–
more specifically the framing–on the part of elites, of popular discontent, collective goals,
rationales for solidarity and justifications to gain the support of a population. This is one of
the necessary first steps in elite-led collective action. It is the recruitment and engagement
stage. Successfully communicating mobilising appeals is, however, a complex and
uncertain process. There always exists intervening variables, attenuating factors and
unforeseen system effects that affect the communication of appeals, their reception, and
the overall outcome of communicative acts.
As a result, elite strategies to mobilise
supporters, particularly through references made to collective identity, have varied
empirical effects.
In the way many authors treat elite appeals to foster mobilisation,
however, these are assumed to be effective. In these analyses, if messages are framed,
elites successfully mobilise their target population.
As a corrective, this framework posits that it is ‘properly’ framed appeals that
successfully mobilise. In other words, this research examines the potent strategies that
turn a population’s attitudes, collective identity and background into appeals that plausibly
justify action, even violence, against another group. Efficacy of appeals in this case is
75
defined as the acceptance and internalisation of the message by populations. 1 Efficacious
appeals are the ones that are accepted and internalised by individuals and that lead them,
as a group, to a state of mind conducive to mobilisation–or what could be called ‘felt
mobilisation.’
Accordingly, this framework is not about the outcomes of mobilisation per se–the
type of action undertaken or, in cases of conflict, the degree or form of violence resulting
from mobilisation, for example. Nor is it about extant or manifest mobilisation, although
extant mobilisation is the outcome of the processes studied. For all of these to occur,
people in contentious circumstances must first decide whether or not to participate. They
must take the decision to mobilise or not, to follow leaders in their plans or not, in response
to cues received from leaders. This is what makes up ‘felt mobilisation’, which eventually
can lead to manifest mobilisation and the resulting action, fighting, violence, etc.
The originality of this research is to look at how processes or patterns for
communicating appeals interact with social-psychological individual group and intergroup
dynamics. To achieve this, the framework disaggregates communication efforts on the part
of elites into some of their probable mechanisms, or linkage rules, aimed at aligning
individuals to the group and the group to elite goals. It looks at the strategies used to
stimulate collective solidarity among a mixed population, or felt mobilisation in the name of
the group. Although not an exhaustive list of existing mechanisms, this analysis identifies
and develops three primary strategies in the construction of mobilising appeals: opposition
(of categories and identities so as to create an antagonistic portrayal of groups and
intergroup relations), politicisation (or the creation of a politicised version of group
identities), and simplification (of constructed identity narratives).
1
As Henri Tajfel and his collaborators indicated, without the internalisation of categories and frames of
reference of a particular intergroup situation, intergroup competitive and/or conflictual and ingroup group
solidarity dynamics do not arise. See Henri Tajfel, M.G. Billig, R. P. Bundy, and Claude Flament, “Social
Categorization and Intergroup Behaviour,” European Journal of Social Psychology, vol. 1, 1971, pp. 149-178.
76
Drawing on the literature surveyed, particularly social movement theory and
contentious politics, this chapter first proposes an original probable sequence of elite-led
mobilisation through collective identity and backgrounds. Once the probable sequence of
events and actions established, the chapter revisits collective identity, to turn finally to the
study of opposition, politicisation and simplification, communication strategies to foster
collective solidarity.
Synthesising the sequence of elite-led mobilisation
If one developed a synthesis of the chain of events and processes behind elite-led
mobilisation, it would likely start with an assessment of the environment, or what social
movement theorists called opportunities. Structural factors and opportunities are essential
to the mobilisation process. They are what determines the openness or resistance of the
system to contention. 2 These directly influence elites and populations’ latitude for action.
This is why environmental changes, and specifically those that translate into changes in
opportunities, are background conditions to popular mobilisation writ large. They trigger
contentious human action, inasmuch as they are attributed importance by individuals and
groups.
2
This point was initially studied by the political process approach. It continues to be analysed. Overall, studies
have indicated that while the relative openness or even vulnerability of the system might facilitate access to it
by contenders, a more permeable system might also serve to co-opt the action of challengers and to
demobilise movement adherents. Oded Haklai, “Linking Ideas and Opportunities in Contentious Politics: The
Israeli Nonparliamentary Opposition to the Peace Process,” Canadian Journal of Political Science, vol. 36, no.
4, 2003, p. 795. Similarly, although authoritarian or repressive systems impede the formation of movements,
passed a certain threshold of repression, they might instead have the reverse effect. Kriesi et al. effectively
noted that “the state can invite action by facilitating access, but it can also provoke action by producing
unwanted policies and political threats, thereby raising the costs of inaction.” Quoted in David S. Meyer,
“Protest and Political Opportunities,” Annual Review of Sociology, vol. 30, 2004, p. 131. As a matter of fact,
Charles Tilly argued that protest and the political system followed a curvilinear relationship. As he
conceptualised it, “protest occurs when there is a space of toleration by a polity and when claimants are neither
sufficiently advantaged to obviate the need to use dramatic means to express their interests nor so completely
repressed to prevent them from trying to get what they want.” Quoted in David S. Meyer, “Protest and Political
Opportunities,” p. 128. Nonetheless, authoritarian and repressive regimes under duress, in which ‘cracks in the
system’ open up, can as a matter of fact stimulate movement action by allowing for the emergence of an
activist civil society when previously there existed no tolerance for it. System vulnerability and instability can,
thus, open opportunities for those generally excluded under normal circumstances, and give them access to
decision-making spheres and resources. Jeffrey M. Ayres, “Political Process and Popular Protests: The
Mobilization against Free Trade in Canada,” The American Journal of Economics and Sociology, vol. 55, no. 4,
1996, p. 473.
77
Structural variables are the backdrop of social dynamics behind mobilisation.
Before mobilisation even becomes an option, there often needs to be changes in the
environment that trigger a crisis. Crises result from a variety of complex background
conditions, such as political, economic, social or environmental factors. They confront
individuals with changes in their environment, variations they must react to or absorb. As a
result of the unfolding crisis, communities feel sentiments which can range from
detachment, fear, discontent, to outright anger. Individuals can be destabilised, fearful, and
made vulnerable by factors they can not control and often can not counteract without the
help or guidance of elites, an organised form of power or the state. Fear resulting from
crises can make a population more dependent on leadership and consequently more
receptive to manipulation by elites, giving the latter more leeway in terms of actions and
courses they can take. 3 Fear in situations in which a sense of threat exists opens the door
to dangerous forms of manipulation. Furthermore, fear can lead to extreme reactions.4
According to what Gurr referred to as the threat-aggression sequence, people in situations
of extreme fear might be more susceptible to accept the recourse to violence, even
extreme aggression. 5
But people might also feel aggrieved, frustrated or even angry.
3
Jenna Slotin, unpublished manuscript, University of Toronto, Toronto, Canada, 2003, p. 17. In a related
argument, Regan and Norton contended that it is the need for protection that drives popular choices and
actions towards mobilisation during instances of intense conflict. As they explained, “[b]ecause initial
mobilisation may carry minimal cost, grievances may well be sufficient to motivate the disaffected to participate
in protest activities. [...] However, the costs associated with full-scale revolt or civil war are such that only the
provision of selected benefits that outweigh the costs imposed by the state for participation can spur an
individual to remain committed to the cause, and protection may be the most important side payment.” Patrick
M. Regan and Daniel Norton, “Greed, Grievance, and Mobilization in Civil Wars,” Journal of Conflict
Resolution, vol. 49, no. 3, 2005, p. 324.
4
Fear does call forth its own dynamics. Stuart Kaufman, for example, indicated that fear leads people to
prioritise security over other interests. Stuart J. Kaufman, “Symbolic Politics or Rational Choice: Testing
Theories of Extreme Ethnic Violence,” International Security, vol. 30, no. 4, 2006, p. 51.
5
Ted Robert Gurr, Why Men Rebel (Princeton, N.J.: Princeton University Press, 1970), pp. 35-36. The
psychology literature Gurr drew on when describing the threat-aggression sequence upholds that frustration
induced aggressive behaviour, which the threat-aggression sequence is a form of, is “part of man’s biological
makeup; there is a biologically inherent tendency, in men and animals, to attack the frustrating agent.” Ibid., p.
33. While the present framework contends that fear is a determining factor in the development of ‘felt
mobilisation’ among populations, it does not however acknowledge this purported natural human predisposition
to aggression. Drawing on a study of the violence in the former Yugoslavia, Franke Wilmer contended that
“under conditions of economic and political instability, psychic stress is more likely to induce cognitive and
emotional changes that make scapegoating, projection, and ultimately violent behavior more likely.” See
78
Gurr also pointed to the dangers of such sentiments when shared by large segments of a
population. 6
They are especially noxious when given direction and targeted against
something or someone.
Although crises can generate feelings of fear and/or discontent in a population,
without the sense that these are collectively shared, they remain experienced individually.
They do not move a population or only translate into diffuse frustration or anomic
violence. 7 For these sentiments to serve as a platform for collective action, they must
acquire a social meaning. They need to translate into collective grievances or a project
that rallies people around a common plea and goals. 8 This is where elites can play a role.
As Sidney Tarrow underlined, “without leadership exercised through organizations,
rebellion remains ‘primitive’ and soon disintegrates.” 9 Gustave Le Bon, the French crowd
psychologist, already alluded to this role on the part of leaders in his writings. 10 In times of
turmoil, elites are in a privileged position to institute themselves as the voice of the many,
giving direction to sentiments expressed on the ground. They can harness popular
discontent to their benefit.
Franke Wilmer, The Social Construction of Man, the State, and War: Identity, Conflict, and Violence in the
Former Yugoslavia (New York: Routledge, 2002), p. 262.
6
Ibid., p. 319. While many authors ascribe to discontent and emotions a pivotal role in mobilisation and
conflict, another stance contends that greed is at the heart of these processes. While economic gains can be a
justification for action, it is the case most often then not for elites and organised groups, such as rebel
organisations, more than for the general population. On the greed approach to conflict, see, for example, Mats
R. Berdal and David M. Malone, Greed and Grievance: Economic Agendas in Civil War (Boulder, Co.: Lynne
Rienner Publisher, 2000). Economic factors do eventually come to play a role, however, since, as Daniel
Masters argued, “[t]he gist is that people cannot participate without the means to do so.” Economic resources
and benefits can serve as an encouraging and sustaining factor for collective action. “Support and Nonsupport
for Nationalist Rebellion: A Prospect Theory Approach,” Political Psychology, vol. 25, no. 5, 2004, p. 711.
7
Émile Durkheim developed the notion of anomie. Bernard Brown explains the concept in “Revolution and
Anomie.” He states that “[o]nce in a condition of anomie, the individual may react in altogether unforeseeable
ways. One tendency, we have noted, is toward renunciation, withdrawal, loss of zest for life, or suicide. But,
as is stressed by Durkheim in a less well-known part of his analysis, anomie gives birth to a state of
exasperation and irritated lassitude, “which can, depending upon circumstances, turn the individual against
himself or against others.” Bernard E. Brown, “Revolution and Anomie,” in Bernard E. Brown, ed., Comparative
Politics: Notes and Readings, Ninth Edition (Orlando, Fla.: Harcourt College Publishers, 2000), pp. 119-120.
Because anomie is first and foremost experienced individually, when it leads to violence, the latter is often
diffuse and disorganised.
8
Daniel Masters, “Support and Nonsupport for Nationalist Rebellion,” p. 710.
9
Sidney Tarrow, Power in Movement: Social Movement, Collective Action and Politics (Cambridge, Mass.:
Cambridge University Press, 1998), p. 135.
10
Christine Poggi, “Folla/Follia: Futurism and the Crowd,” Critical Inquiry, vol. 28, 2002, p. 729.
79
It is to their advantage to do so. By choosing to make themselves the
representatives of a group’s or a population’s interests they can gain their support, a nonnegligible resource, and especially non-negligible amidst crises. This is not to say that
elites are necessarily maximising, calculating egoistic actors. First and foremost, they are
not immune to sentiments such as those experienced on the ground. They too can feel
fear, discontent, resentment and anger. They certainly feel destabilised. Crises tend to
undermine and possibly even reconfigure the established balance of power, as well as
change the range of opportunities available to elites. McAdam, Tarrow and Tilly indicated
that, for elites, “these episodic moments of contention, [create] uncertainty [...]; [reveal]
fault lines, hence possible realignments in the body politic; [threaten] and [encourage]
challengers to take further contentious actions; [and force] elites to reconsider their
commitments and allegiances.” 11 Their rationality or interest-driven behaviour comes as a
result of the rules of the game they are part of: the political game. 12 Political competitions
are construed as a struggle in which outcomes are often binary. It is much like a zero-sum
game in which gains and losses are relative: to keep one’s post or not, to have influence
and power or not, to be in the winning party/group or not. Paul Pierson referred to this as
the ‘lumpy’ or ‘winner-take-all’ quality of some political goals: “politicians seeking
reelection, coup plotters, and lobbyists either win or lose; legislation either passes or is
rejected [...] Unlike economic markets, in which there usually is room for many firms, in
politics finishing second may not count for much.” 13 As a result, the parameters of the
game suggest instrumental behaviour in order to strategise ways to come out ahead of the
11
Doug McAdam, Sidney Tarrow and Charles Tilly, Dynamics of Contention (New York: Cambridge University
Press, 2001), p. 9.
12
It should also be remembered that in certain parts of the world, there is more to holding office than political
power. As Timothy Longman explained about Africa, for example, “[a]s a number of scholars have pointed out,
economic power in Africa tends to be derived from political power, rather than the reverse. Political officials
have not only enriched themselves but also built patrimonial networks of support by distributing the largesse of
the state and by placing clients in offices where they have access to resources.” Timothy P. Longman,
“Empowering the Weak and Protecting the Powerful: The Contradictory Nature of Christian Churches in
Rwanda, Burundi, and the Democratic Republic of Congo,” African Studies Review, vol. 41, no. 1, 1998, p. 55.
13
Paul Pierson, “Increasing Returns, Path Dependence, and the Study of Politics,” American Political Science
Review, vol. 94, no. 2, 2000, p. 258.
80
competition, ways not to lose. On the whole, then, whether elites are formally organised or
not, it is possible to extrapolate this as their generic behaviour under this form of
competition. They pursue goals and interests and defend ideas and platforms they judge
important or beneficial to their standing. They work for what they seek to achieve,
continued leadership at a minimum. If F.G. Bailey is correct in seeing “leadership [as] an
enterprise,” elites then can be referred to as political entrepreneurs, or when specifically
working on collective identity referents, ethno-centric entrepreneurs. 14
With instability, political changes and transitions as turning points, political
entrepreneurs may seek to benefit from the situation, either by fighting for the status quo
or by seizing new opportunities, depending on their position in the political system. 15 In a
political game, this often rests on securing the allegiance of supporters. Supporters matter.
They can be providers of resources, food or machinery, for example, or even themselves
as combatants. Popular support is the source of material and human resources. But
popular support is also the source of immaterial resources, influence and power. The more
people support a leader, the more popular legitimacy a leader gains, the more influence
and power the said leader has. Influence and power are about the ability to compel
individuals to do one’s bidding in the absence of coercion. 16 Power is also one of the more
fundamental sources of security. 17 Accruing power is a means of ensuring one’s safety
and standing. As a result, political entrepreneurs often make these moments of instability a
competition for supporters.
If this is the tactic adopted by political entrepreneurs, they must find the key to
convincing people to support them. They must find a way to ‘recruit’ followers, to make the
14
F. G. Bailey, Stratagems and Spoils: A Social Anthropology of Politics (Toronto: Copp Clark Publishing
Company, 1969), p. 36.
15
For example, the political process model differentiates between incumbents and challengers depending on
whether actors are part of the polity or excluded from it and between the actions deployed by both groups.
16
Hannah Arendt, On Violence (Orlando: Fla., Harvest Books, 1970).
17
This point was raised by Barry Posen in his discussion of the security dilemma in ethnic conflits. See Barry
Posen, “The Security Dilemma and Ethnic Conflict,” Survival, vol. 35, no. 1, 1993, pp. 27-47.
81
population recognise it has a vested interest in supporting its leaders. Sidney Tarrow
contended that “[t]he most effective organization of collective action draws on social
networks in which people normally live and work, because their mutual trust and
interdependence can easily be turned into solidarity.” 18
For a common purpose, one
needs a common denominator, and pre-existing ties are an already internalised platform
on which to build it. Culture, common identity and collective backgrounds serve as the
basis for some of these pre-existing social networks. Whether understood as collective
identity, ethnicity, culture, bonds of commonality, whatever form they take, collective
attachments are the source of categories, identities, norms and relations between ingroup
members, among themselves or with others. Onto these are woven common normative
references in the form of historical narratives, myths, knowledge, beliefs, values, attitudes,
aspirations, images, and symbols. These form what Stuart Kaufman referred to as the
‘myth-symbol complex’: “the combination of myths, memories, values, and symbols that
defines not only who is a member of the group but what it means to be a member.” 19
Playing the ‘ethnic, religious or communal card’ is a potent medium through which
political entrepreneurs can address a population. They are an existing platform of
relationships of commonality and trust between members of the group. 20 By calling on
communitarian ties, ethno-centric entrepreneurs build on an implicit pre-qualification. Their
membership in the group means they are already integrally part of the collective project
the ‘we’ forms. By belonging to the group, as insiders, it is easier for their constituents to
18
Sidney Tarrow, Power in Movement, p. 136.
Stuart J. Kaufman, Modern Hatreds, p. 25.
20
Brewer, for example, defines groups as “bounded communities of mutual trust and obligation that delimit
mutual interdependence and cooperation. Marilynn B. Brewer, “The Psychology of Prejudice: Ingroup Love or
Outgroup Hate,” Journal of Social Issues, vol. 55, no. 3, 1999, p. 430. As she further explained, “[a]n important
aspect of this mutual trust is that it is depersonalized, extended to any member of the ingroup whether
personally related or not.” Ibid, p. 433.
19
82
identify with them and to accept them as leaders. 21 In all appearances, they are not only
speaking for the group but with the group.
As for ethno-centric entrepreneurs’ communication with the group, attention to how
ethno-centric messages are constructed matters. While elites have ideas, wishes and
projects, if their end goal is mobilisation, they must make sure that they effectively
convince their constituents of the need for collective action. Much rests on how appeals
are articulated. To gain the support of a population, the language ethno-centric
entrepreneurs use to harness popular discontent–in their attribution of blame and in their
prescriptions for solutions–should be motivating and persuasive. Employing vocabularies
of severity, urgency, efficacy, and propriety in frames for collective enterprises creates a
sense of the need for collective action and of its potency 22 .
Another important factor behind effective reception is resonance. Resonance is
very much tied to the fit of ideas promoted by ethno-centric entrepreneurs and the
background, sentiments and experiences felt by their audience. As both socialcategorisation theory and framing literature pointed to, fit or the commensurability of
appeals–and their portrayal of identities, relations and grievances–with the context as
perceived by the population plays a role in how resonant and credible messages are taken
to be. 23 Myth-symbol complexes are constraints on what is judged to be ‘fitting’.
By
becoming historically and contextually accepted and internalised by groups, myths,
21
From a social psychological point of view, as Bernd Simon and Bert Klandermans explained, “[t]his should
be so because, by virtue of the shared group membership, the attitudes and actions of the leader are rendered
normative for ordinary group members who should feel compelled to follow the leader’s example to verify their
collective identity.” Bernd Simon and Bert Klandermans, “Politicized Collective Identity: A Social Psychological
Analysis,” American Psychologist, vol. 56, no. 4, 2001, p. 326. Social psychological findings also indicate that
individuals tend to prefer for what they judge to be familiar over the unfamiliar. For Brewer, this is noticeable in
ingroup-outgroup interactions and results in “social interactions within the group [being] more predictable and
understood than intergroup interactions.” Marilynn B. Brewer, “The Psychology of Prejudice,” p. 435.
22
For example, the prospect of collective action is enhanced when presented as the way to ensure success in
terms of gains or beneficial change, or, as Bert Klandermans explained, when it is supported by “conviction
that it is possible to change the situation via collective action at reasonable costs.” Bert Klandermans, “How
Group Identification Helps to Overcome the Dilemma of Collective Action,” American Behavioral Scientist, vol.
45, no. 5, 2002, p. 888.
23
As Turner et. al. indicated about social identity, for example, it “is not just a matter of social construction but
must fit with existing knowledge about ourselves and other people.” Quoted in Bert Klandermans, “How Group
Identification Helps to Overcome the Dilemma of Collective Action,” p. 894.
83
symbols, shared referents become what makes sense. They become the norm against
which new ideas and discourses are compared.
FIGURE 5.1: A PROBABLE SEQUENCE OF ELITE-LED MOBILISATION
(A)
CHANGES IN
THE
ENVIRONMENT
BACKGROUND
CONDITIONS
(A)
(B)
EMERGENCE OF
GENERAL
DISCONTENT,
FRUSTRATION,
ANGER AND/OR
FEAR IN
POPULATION
HEIGHTENED
COMPETITION
BETWEEN
POLITICAL OR
ETHNO-CENTRIC
ENTREPRENEURS
IF DONE APPROPRIATELY
(C)
(D)
ETHNO-CENTRIC
ENTREPRENEURS
FELT
FRAMING
MOBILISATION/
DISCONTENT AND
SUPPORT
APPEALS IN
‘URGENT BUT
EFFECTIVE’
TERMS AND
ACCORDING TO
RESONANT
COMMUNAL
BACKGROUND
X
INDEPENDENT VARIABLE
PROCESSUAL
(B) THROUGH (C)
Y
DEPENDENT VARIABLE
PHENOMENOLOGICAL
(D)
Resonance is therefore not only about giving people what they want to hear but it is also
about giving them what they understand, in light of internalised social structures. For this
reason, a common identity or culture, a shared understanding of collective history,
perceptions of what the norms are for membership in the group or exclusion from the
group, of relations with other groups, are potent material through which political projects
can be translated into justifications for collective action.
If we try to work out a probable sequence of elite-led communication of mobilisation
it should resemble Figure 5.1. For political entrepreneurs to begin a process of
mobilisation, conditions must be right. The system must open up to possible contention.
Changes in the environment must affect populations and heighten competition between
elites. It must make elites need and want to gain the support of populations to help them
84
pursue their goals. In order to mobilise people towards action, however, political
entrepreneurs must communicate the need for collective action. People must be recruited,
convinced by the aim. Working with pre-existing ties, calling on people in the name of a
pre-existing group, is a common tactic. In framing ethno-centric calls for mobilisation,
motivating people for action can be better achieved (1) by reverting to vocabularies of
severity, urgency, efficacy and propriety; and (2) by speaking their language, that is
choosing frames which resonate with what people know.
A New Social and Psychological Take on Framing Collective Appeals for Mobilisation
Considering the complexity of communicative processes and in the absence of a ‘magic
bullet theory of perfectly persuasive communication’, much can be gained by delving
further into the communication of mobilisation appeals. 24 Beyond generic framing
processes and the general frame alignment and resonance patterns described by this
sequence, other factors account for the variability of appeals’ effectiveness. 25 While use of
specific vocabularies and resonance explain why followers follow leaders around a goal,
they do not fully explain why individuals come to recognise themselves as potential
followers. This dissertation hypothesises the importance of this other side of
communicating mobilisation, recalling social-psychological sources of attachment to the
group.
Looking at this takes us beyond traditional framing models, beyond frame
alignment and resonance patterns to look at how, in the mobilisation process, individuals
24
As Clay Calvert contended, “many today in communication-science tradition reject the magic bullet theory of
direct, powerful, and uniform changes in attitudes and behaviors caused by messages.”Clay Calvert, “Hate
Speech and Its Harms: A Communication Theory Perspective,” Journal of Communication, vol. 47, no. 1, 1997,
p. 9.
25
What is referred to as frame alignment resonance patterns is the classic framing model addressing both
actions by political entrepreneurs to construct frames that tie their political projects to experiences of their
audience and the importance of resonance of frames with understandings and experiences of the public to
ensure their reception. Frame alignment resonance patterns are the interactive (emitters-receptors) approach
to framing as developed by most frame analysts thus far.
85
are also aligned to the group. Tajfel and his collaborators insisted on the importance of
acceptance and internalisation of frames of references, including group definitions, for
solidaristic group behaviour and competition with other groups to occur. The ability of
political and/or ethno-centric entrepreneurs to make a group buy into their mobilisation
appeals is tied to their parallel ability to make individuals buy into the group. This implies
looking at how an individual is made to believe that the group matters and that it is to
him/her as a member of the group that appeals are addressed.
Without feeling that
appeals are addressed to them, to them as a member of the group, how can people feel
appeals are justified and resonant? It is through an achieved, activated sense of we-ness
that appeals can be persuasive and resonant for people and achieve solidaristic ‘felt
mobilisation.’ Thus, buying into the group, at the heart of which lie social-psychological
group dynamics, is as much part of effective appeals for action as developing a sense of
severity, urgency, efficacy and propriety, as well as resonance through framing.
This should not be taken to mean that one comes prior to the other. Buying into
the group is not antecedent or subsequent to frame alignment and resonance. Both are
necessary steps in communicating mobilisation. They form a dialectical pair. Without
salient feelings of we-ness, there is little sense of necessity or of the efficacy of group
action and little ‘accessibility’ in terms of the categories or identity on which to build
resonance. It can be by building appeals aimed at persuading and resonating, however,
that the referents for we-ness are recalled. It is this dialectical process which can be
disaggregated into mechanisms. And while persuasion and resonance mechanisms have
already been studied at length by frame analysts, it is specifically on mechanisms for
fostering we-ness that this research focuses.
86
REVISITING IDENTITY
To understand these mechanisms and how they operate, it is necessary to return to the
nature of identity. 26 Identity structures are complex. They are fuzzy ensembles of ideas,
notions, conceptions, with relationships and patterns tying these different elements
together. 27 Identities can be treated as a network.
As Thomas Homer-Dixon explained, networks are “sets of nodes and the links
connecting those nodes.” 28
Nodes are units within the network (nodes/units can be
people, objects, concepts, for example, depending on the network) and links are the ties or
relationships weaved between these units. Simple networks are unordered structures.
The word ‘unordered’ should not be taken to mean they are disorganised, however. The
relationships between nodes are, in and of themselves, a form of organisation, be it one
that may not follow systematic or common rules for organisation. More complex networks
can have specific structures. One particular form of network, ontologies, has been
developed in computer science and informational science to map ‘knowledge networks’ or
26
It should be noted that social movement literature and social psychology do not define collective or social
identity exactly in the same manner. While social movement literature tends to focus on the collective and its
characteristics for its members’ identity, social psychology centers its definition of social identity on the
individual. To illustrate this, Klandermans gave the examples of Taylor and Whittier who defined collective
identity as “shared definitions of a group that derive from member’s common interests and solidarity”, and of
Tajfel, who defined social identity as “that part of an individual’s self-concept which derives from his knowledge
of his membership of a social group (or groups), together with the value and emotional significance attached to
that membership.” Bert Klandermans, “How Group Identification Helps to Overcome the Dilemma of Collective
Action,” p. 889. Despite this difference, both bodies of literature can complement one another.
27
In a similar vein, Serge Moscovici spoke of social representations. For Moscovici, [social] representations
‘conventionalise’ people’s knowledge through two ‘mechanisms’: 1) ‘objectification’-which identifies
relationships between images and ideas people store in their minds and objects they encounter in the physical
world; and 2) ‘anchoring’-which groups objectified relationships into broadly encompassing categories, and
tethers, or ‘anchors’, these categories together to form patterns of general knowledge. Consequently,
representations provide a structure, ‘environment’, or frame, into which people can fit new information, and
associate their information with a known category, and thereby ‘make something unfamiliar, or unfamiliarity
itself, familiar’. Quoted in Kimberly Fisher, “Locating Frames in the Discursive Universe,” Sociological
Research Online, vol. 2, no. 3, 1997, pp. 11-12. For more on Moscovici’s work on social representations, see
Serge Moscovici, “The Phenomenon of Social Representations,” in Robert M. Farr and Serge Moscovici, eds.,
Social Representations (London: Cambridge University Press, 1984).
28
Thomas Homer-Dixon, The Upside of Down: Catastrophe, Creativity, and the Renewal of Civilization
(Toronto: Knopf Canada, 2006), p. 113. In the social sciences, the concept of ‘network’ has paved the way for
an ambitious research program. What many now call ‘social network analysis’ is devoted to the study of
various forms of social organisations in which individuals are treated as nodes within the network linked by
relationships Social network analysis tends to focus predominantly on the ties between individuals and the
patterns created within the system. In other fields, such as in computer science, for example, the term
‘network’ is applied to a much broader array of structures, in which nodes can be individuals, objects, ideas,
etc.
87
knowledge structures. Ontologies–not to be confused with the metaphysical concept
developed in Western philosophy–are a way to schematically map knowledge, or units of
knowledge, and their relationships, while specifying some of the core constraints among
these relations. 29
Somewhat like a logic tree, an ontology is composed of concepts, also called
classes (the nodes of a regular network); units or individuals within these classes, also
called instances; relations between these units; and of restrictions that regulate these
relations. For example, the class pets can be divided into the subclasses dogs, cats, etc.,
which in turn are comprised of individuals or instances, such as bulldogs, golden retrievers
and schnauzers for dogs, and Siamese, Sphynx and Himalayan for cats. The most basic
relations in this ontology are ‘is-a’ or ‘is a-subclass-of’ or ‘is-a-superclass of’. Pets is a
superclass of dogs, and dogs is a subclass of pets. Fido is a dog, and dog is a superclass
of Fido. More complex relationships can be represented in an ontology such as ‘hates’ or
‘loves’, for example. In this instance, there could be a relationship between the classes
dogs and cats stipulating that dogs hate cats.
Finally, restrictions or formal axioms
constrain the interpretation and well-formed use of terms and relationships. These axioms
or restrictions are normative statements to describe relationships between elements. For
example, such axioms could be that ‘all dogs hate cats,’ or that ‘to hate is the opposite of
to love’, etc.
29
Thomas Gruber, for example, explained ontologies in these terms: “[a] body of formally represented
knowledge is based on a conceptualization: the objects, concepts, and other entities that are assumed to exist
in some area of interest and the relationships that hold among them. A conceptualization is an abstract,
simplified view of the world that we wish to represent for some purpose. [...] An ontology is an explicit
specification of a conceptualization. When the knowledge of a domain is represented in a declarative
formalism, the set of objects that can be represented is called the universe of discourse. [...] In such an
ontology, definitions associate the names of entities in the universe of discourse (e.g. classes, relations,
functions or other objects) [...], and formal axioms that constrain the interpretation and well-formed use of these
terms.” Emphasis in original. Thomas R. Gruber, “A Translation Approach to Portable Ontology Specifications,”
Knowledge Acquisition, vol. 5, no. 2, 1993, p. 199. A similar idea is conveyed by Self-categorisation Theory’s
levels with regard to classes or levels of representation: human, social and personal. In the case of SCT,
however, classes are ordered hierarchically, which is not necessarily the case in ontologies.
88
Networks and ontologies, as formal conceptualisations, serve the purpose of
making relations between things explicit. They help represent and understand
relationships between
FIGURE 5.2: NETWORKS AND ONTOLOGIES
b-An Example of an Ontology
a-An Example of a Common Network
nodes, and in the case of ontologies, between classes (nodes in a regular network),
subclasses, and instances, or amongst classes, subclasses and instances. Most network
analysts tend to focus more on these links or relationships than on instances or individuals
per se. It is these relationships that come to form specific patterns which generate the
shape, the organisation, a specific network or ontology.
If construed as a network, identity backgrounds resemble much more an ontology
than a generic network. Some restrictions in relationships among classes of identity exist.
For example, the class ‘mother’ can only be subsumed by the superclass ‘female’, and not
‘male’. Considering the importance of relations, this means that, when it comes to identity
structures, the sum is greater than the parts. 30 Identities are the shifting patterns, the
30
Émile Durkheim insisted on this characteristic of social phenomena: individual facts were to be distinguished
from social facts which acquired properties distinct from those of individuals constituting the social unit.
Interestingly, many working on social networks take Durkheim to be a precursor to Social Network Analysis.
See for example, Sandro Segre, “A Durkheimian Network Theory,” Journal of Classical Sociology, vol. 4, no. 2,
2004, pp. 215-235. The new properties of the larger social unit are also called ‘emergent properties’. Basing
himself on C.D. Broad, David Newman explained that “an emergent quality is roughly a quality which belongs
to a complex as a whole and not to it parts”. Broad’s account of emergence characterizes it as an alternative
to strict reductive materialism in which the behavior of a physical system is explained by appeal to the structure
89
ensemble of relations between nodes, and not the sheer amalgamation of nodes and
referents. Identity must be understood as the patterns that emerge from connections
between referents in a web of amorphous definitional elements, referents, and different
conceptions, that all coexist to constitute people’s understandings of who and what they
are, of their relations to their group and others, and of their place in the world (see Figure
5.3). 31
While connections between nodes, or referents, create an identity pattern,
relationships between existing nodes, even some of the same nodes, create different
identities. For example, the referent ‘has graduated from the faculty of medicine’ might be
tied to the pattern/identity ‘alumni of University X’ or through another combination might be
tied to the pattern/identity ‘member of class X in society’. From a cognitive and social
background emerges a multiplicity of identities: political, social, economic, biological, as
well as individual and collective identities.
This multiplicity of co-existing and interlocking of identities raises the issue of their
expression.
Should cognitive and social backgrounds be a regular network, these
patterns/identities could manifest simultaneously at all times. Because they resemble an
ontology more than a general network, however, relationships between nodes,
relationships that form identity patterns, are guided by restrictions affecting the
manifestation of identities. Some of these restrictions imply the existence of
multidirectional relations, of bilateral ones, or of unidirectional ones. 32 As a result,
of that system but not to the existence of particular components within that system. In addition, Broad thought
that the behavior of a system with an emergent property cannot be deduced from complete knowledge of the
properties and behavior of the components of the system.30
31
Social psychologists believe collective identities, in particular, serve five functions related to basic
psychological needs: belongingness, distinctiveness, respect, understanding (or meaning), and agency. As
Bernd Simon and Bert Klandermans explained, “collective identity confirms that one belongs to a particular
place in the social world. At the same time, it also affords distinctiveness from those other social places (or
people) to which one does not belong. It further signals that one is like other people, though not necessarily
like all other people, so that one can expect respect, at least from these similar others [...]. Moreover,
collective identity provides a meaningful perspective on the social world from which this world can be
interpreted and understood. Finally, collective identity signals that one is not alone but can count on the social
support and solidarity of other in-group members so that, as a group, one is a much more efficacious social
agent.” Bernd Simon and Bert Klandermans, “Politicized Collective Identity,” p. 321.
32
Some of these are also positive (reinforcing) and negative (weakening) feedback loops.
90
identities–according to the relationships between the nodes that comprise them–can be
complementary, some competing, and some others even mutually
FIGURE 5.3: A NETWORK OF IDENTITIES
IDENTITY-A
IDENTITY-B
IDENTITY-C
For the sake of clarity, this network has been represented as a simple one instead of an ontology.
As an ontology, it would be composed of subclasses, each comprised of instances, different types
of relationships and restrictions which could be attached to each individual node (or class).
exclusive. For example, a combination of some of the same nodes and different ones can
produce the identities ‘botanist’ and ‘environmentalist’, which can be thought of as
complementary.
Self-categorisation theorists see, on the other hand, a functional
antagonism between the expression of personal identities and collective ones. SCT
authors effectively believe that when collective identities become salient, they do so by
91
displacing individual ones. 33 While some identities can be expressed simultaneously,
others are, in terms of their expression, antagonistic.
What this points to–especially the existence of exclusive identities–is that not all
identities are expressed at once. There is variance in their manifestation. 34 An identity’s
status can change–an identity can become salient or recede– as the environment
changes, new situations develop, interactions with new actors arise, etc. 35 This should not
be taken to mean that the salience of identities is an either/or process, however. The
manifestation of identities is a range of possible expressions, latent or salient being only
the extreme of this range. Treating identities as patterns in a cognitive and social network
helps to understand this. With networks’ focus on relationships between nodes, salience
can be construed as changes with regard to these relationships. What is made salient
then is not so much a series of nodes.
What occurs is much more that certain
relationships between nodes become salient. This, in turn, is what allows for an identity,
or a pattern, to become prominent. It is somewhat like choosing to route electrical current
to specific sectors in a grid of street lamps in a city.
The fact that only certain
neighbourhoods have light while others do not results from choosing to direct electrical
33
See, John C. Turner, with Michael A. Hogg, Penelope J. Oakes, Stephen D. Reicher and Margaret S.
Wetherell, Rediscovering the Social Group: A Self-Categorization Theory (Oxford: Basil Blackwell, 1987), p.
49.
34
Highly directly connected nodes in a network are believed to be more central. They tend to be the ones that
connect a greater number of other nodes. These central nodes are called hubs. An identity hub could be ‘sex’
for example. It can be imagined to be tied to a great number of identities, such as mother, sister, daughter,
gender, status in society, certain political identities, etc. As for preferred patterns, another insight from network
approaches is what is called ‘network reach’. Paths that are more direct, that is involve more direct
connections between nodes and as a result imply less steps to tie nodes, are considered to be the ‘faster’ ones
and often the preferred ones. In a sense this can be related to the SCT insight on ‘relative accessibility’: the
use of a referent is tied to how easily it can be recalled by its user. Both of these can, in turn, be tied to the
general contention of social psychology on humans’ use of cognitive shortcuts, mechanisms to simplify and
quicken the treatment of the overwhelming amount of information they must face from interactions with their
environment.
35
As McAdam, Tarrow and Tilly explained, “[t]he same person who bears the identity mother in one context
easily adopts the identities manager, customer, alumna, and sister in others. Doug McAdam, Sidney Tarrow
and Charles Tilly, Dynamics of Contention, p. 133. Social psychologists, on their side, explain this contextual
nature of the salience of collective identity in these terms, “[s]alient ingroup membership and, therefore,
cohesion have a situational or contextual basis, varying with the psychological frame of reference (the pool of
relevant comparable other) and the available meta-contrasts between people on relevant dimensions of
comparison.” John C. Turner, with Michael A. Hogg, Penelope J. Oakes, Stephen D. Reicher and Margaret S.
Wetherell, Rediscovering the Social Group, p. 65.
92
current to only those sectors. Light is, thus, not an effect of the existence of lamp posts in
those sectors. Those lampposts exist, but so do the ones that do not receive electrical
power. Lit neighbourhoods are the result of how the electrical current is routed. The same
logic applies to identity. Salience of an identity does not mean others have disappeared.
They and their referents continue to exist. Patterns of thought at that time are routed
through other referents or identities, or certain patterns are modified in terms of relations
weaved between the nodes that comprise them. What this implies is that since they are
the result of relationships–which are dynamic elements unlike static nodes, patterns and
networks can be reconfigured. Obviously, this reconfiguration occurs within the bounds of
the existing components and in accordance to stipulated relationship restrictions.
The issue of salience is central to mobilisation. It is the key to understanding how
individuals shift their interest and focus towards the group, come to consider the collective
as predominant, and are put in a frame of mind to accept to join in collective efforts. To
gain support, ethno-centric entrepreneurs must heighten the importance of the group in
people’s minds or, in social psychological language, the salience of group identity. 36 It is
this group awareness that gives traction to framed calls for mobilisation. It is by being
filtered through a sense of collectiveness that frames can be read as motivating and
persuasive. In other words, the key to group solidarity is salience of collective identity, and
the key to effective appeals is a message that is made plausible through general framing
processes, such as vocabularies of severity, urgency, efficacy and propriety and
resonance around the salient group identity.
How is this to be done? What are the mechanisms or linkage rules in this process?
The play on perceptions of the situation, of the group, and of intergroup relations using
references to shared backgrounds, as can be achieved with framing, can affect the
36
Although he did not develop the idea further, Sidney Tarrow did allude to this in Power in Contention. As he
stated, “leaders can only create a social movement when they tap more deep-rooted feelings of solidarity or
identity.” Sidney Tarrow, Power in Movement, p. 5.
93
salience of specific identifications and foster group cohesiveness in a manner that is
beneficial to ethno-centric entrepreneurs.
By calling on certain ideas, referents and
concepts, and suggesting new or revised interpretations of circumstances and groups’
roles in them, shifts in salience towards the predominance of a mobilising, activated
identity can be achieved. This is what each of the mechanisms developed by the analysis
seeks to facilitate in its own way. Opposition, politicisation and simplification are framing
strategies aimed at creating perceptions of a salient collective identity, at weaving a
narrative to foster a heightened sense of we-ness.
OPPOSITION
A number of social psychologists maintained that differentiation and competition between
groups breed discrimination against outsiders and favouritism towards the ingroup. 37
Increasingly, however, this assumed ‘negative reciprocity,’ the inevitability of outgroup
discrimination and ingroup favouritism, has come up for question. Already, in Muzafer
Sherif’s experiments, changing parameters and putting groups in a situation in which they
needed to cooperate to achieve superordinate goals led to overcoming intergroup
antagonism. As for Henri Tajfel’s minimal group experiments on which Social Identity
Theory was built, their results and significance were judged problematic because the
control environment in which they took place forced groups to compare and compete along
a single set of criteria, a situation clearly unrepresentative of real world situations. Most
37
As Marilynn Brewer recalled, “[f]or Sumner, [this] proposition derived from his structural-functional theory of
the origins of groups in the context of conflict over scarce natural resources. In an environment of scarcity,
individuals needed to band together in groups to compete successfully with other groups for survival.” Marilynn
B. Brewer, “The Psychology of Prejudice,” p. 431. One of the contentions behind Muzafer Sherif’s Robber’s
Cave experiment was that specific ingroup processes lead to intergroup competition or hostility, or as Fine
summarised that “groups naturally develop status structures and group cultures, and then establish
boundaries, providing for the opportunity of intergroup conflict, particularly where resources need to be
shared.” Gary Alan Fine, “Forgotten Classic: The Robbers Cave Experiment,” Sociological Forum, vol. 19, no.
4, 2004, p. 664. With the emergence of Social Identity Theory, this contention held sway. As Brewer
explained, “the assumption of negative reciprocity between ingroup and outgroup was retained at the
psychological level. The same bipolar assumption was also applied to judgments of similarity-dissimilarity
such that increasing perceived similarity within groups is associated with increasing dissimilarity between
groups.” Marilynn B. Brewer, “The Psychology of Prejudice,” p. 431.
94
social psychologists today reject simple ‘negative reciprocity’. 38 Although ‘others’ serve the
original function of providing demarcation between groups, few today take ingroup and
outgroup dynamics to be necessarily antagonistic. 39
If one lesson can be drawn from these recent social psychological takes on
intergroup dynamics, however, it is that contextual factors have a great role to play in
intergroup dynamics and the form they take. Outgroup discrimination and ingroup
favouritism–or to a larger extent ethnocentrism–are not the essence of intergroup
relations, but can be under certain conditions. 40 Specifically, a highly competitive
environment is what leads to the linkage of ingroup dynamics to intergroup relations and to
the resulting outgroup discrimination and ingroup favouritism. Highly competitive
environments are situations in which groups are placed in a clear antagonistic or binary
opposition, instead of ambiguous, complex oppositions. 41 Such situations are zero-sum
environments.
Though it may seem tautological or circular to argue that highly competitive
environments, zero-sum environments, breed conflictual intergroup dynamics and ingroup
favouritism, it is necessary to stress it. Firstly, as critics have pointed out, it should be
made clear that conflictual intergroup dynamics are context-based and not an invariable
38
Bert Klandermans, “How Group Identification Helps to Overcome the Dilemma of Collective Action,” p. 892.
See also, Marilynn B. Brewer, “The Psychology of Prejudice,” p. 431.
39
Gordon W. Allport, quoted in Marilynn B. Brewer, “The Psychology of Prejudice,” pp. 429-430. According to
Brewer, this range of attitudes toward the outgroup includes, among others, mild positivity, indifference,
disdain, or hatred (p. 430).
40
Ethnocentrism can be defined as “positive sentiments toward the ingroup: pride, loyalty, and perceived
superiority.” This definition of ethnocentrism represents the views of Gordon W. Allport and William Graham
Sumner. Marilynn B. Brewer, “The Psychology of Prejudice,” p. 430.
41
Minimal group experiments, for example, illustrated the logic of binary opposition. Over the course of the
experiment, individuals were asked to compare their group, to which they had randomly been assigned to, to
another randomly formed group. Individuals consistently favoured the group they had been randomly
assigned, even at the cost of lesser personal gains. What many raised as a point of criticism for the minimal
group experiments was that the control environment in which they took place biased or determined their
results. It effectively forced the groups to compare and compete for interdependent gains or losses. It was an
environment in which, as Mummendey and Schreiber explained, “a limited number of points [are] split up
between ingroup and outgroup [...] it is impossible for both groups to perform well at the same time. At best,
both groups can score equal result,” or, in the case of some of Tajfel’s experiments, in which “[i]ngroup and
outgroup are rated on an identical dimension.” Emphasis in original. Amélie Mummendey and Hans-Joachim
Schreiber, “Better or Just Different: Positive Social Identity by Discrimination Against, or by Differentiation from
Outgroups,” European Journal of Social Psychology, vol. 13, no. 4, 1983, p. 390.
95
state of affairs. They are fundamentally tied to specific environments and/or factors.
Second, it is by isolating the specifics of environments in which such trends occur that we
get a better sense of what fosters intergroup animosity, and simultaneously how it leads to
ingroup solidarity.
The competition stemming from zero-sum environments is effectively extremely
intense due to the fact that gains and losses are then interdependent. 42 This occurs
because the opposition takes place along a unique dimension for comparison with a
unique scale and limited resources for gains and losses. There is therefore no possibility
for multiple ‘winners’ in different domains (one could achieve a higher score or standing in
domain A, force for example, and the other in domain B, riches for example) since only
one dimension for comparison exists and fixed gains imply that whatever one wins is
exactly balanced by equal losses for the other. Games such as chess, which only allow
for one winner, are illustrative of a zero-sum.
The clearest example of such type of
competition, however, is a life or death battle between two individuals. In that case, the
dimension for competition is survival and the scale or resource is life. For groups, zerosum environments demand outgroup discrimination and ingroup solidarity. Much like a
zero-sum competition between individuals, at the group level, whatever one group
42
Two psychologists, Amélie Mummendey and Hans-Joachim Schreiber illustrated the importance of
dimensions of comparison for competitive behaviour in a study they conducted in the 1980s. In this study,
sixty-six male students were divided into groups. Participants were then made, as a group, to compete with
another group on a task. They were asked to rate the performance of the two groups using three measuring
methods: one in which the same dimension and measuring scale was used for the assessment of the groups
(for example, a scale of a hundred points was divided between the two groups; if one achieved 60, the other
automatically achieved 40); one in which the same dimension was used for group assessment but for which
separate measuring scales were employed (for example, a separate scale of a hundred points would be used
for each group, each could therefore independently achieve all possible scores on the scale), and, finally, one
in which both ingroup and outgroup could be assessed on different dimensions. For each method employed,
the resulting degree of ingroup bias (ingroup favouritism and parallel outgroup discrimination) was evaluated.
Over the course of the study, the highest ingroup bias, or favouritism, was obtained when the same dimension
and scale were used for comparison between both groups. Overall, Mummendey and Schreiber’s study clearly
points to the high conflictual nature of group competition along a single dimension for comparison and a single
scale. See Amélie Mummendey and Hans-Joachim Schreiber. “Better or Just Different,” pp. 389-397. Drawing
implications for the real world, Marilynn Brewer indicated that: “[a] direct relationship between intense ingroup
favoritism and outgroup antagonism might also be expected in highly segmented societies that are
differentiated along a single primary categorization, such as ethnicity or religion. And this would be especially
true if the categorization is dichotomous, dividing the society into two significant subgroups.” Marilynn B.
Brewer, “The Psychology of Prejudice,” p. 439.
96
achieves, it does so at the expense of the other. Groups are therefore in an either/or
situation: they either win or lose to the others. Discrimination against the others and
support for one’s group ensures success.
In a laboratory setting, conditions can be set up to recreate a zero-sum opposition.
In the social world, however, situations are far too complex to ever truly translate into a
true zero-sum environment. Reasons for group antagonism are fuzzier, do not necessarily
operate along a single dimension of comparison, and gains or rewards are uncertain. Yet
environments need not truly be zero-sum for negative reciprocity dynamics to operate, as
long as individuals are made to believe they are in one. If individuals are made to believe
they are in a binary opposition, they behave accordingly. Interpretations and meanings
ascribed to a situation can serve to create a sense that there exists this type of opposition
between groups. It is, in turn, perceptions of intergroup relations that shape the form and
intensity of competition. The attribution of antagonistic meanings to intergroup relations
constitutes “ways in which the effects of social differentiation provide a fertile ground for
conflict and hate.” 43 Accordingly, with ingroup solidarity and outgroup discrimination being
correlated under competitive conditions, fostering ingroup favouritism can result from
heightening outgroup antagonism, an antagonism taking the form of a crystallised, binary
opposition, with collapsed stakes and dimensions for comparison between groups. 44 Two
strong forms of competitive reinterpretations of this kind are discourses of moral superiority
and the attribution of threatening intentions.
As groups grow and come to form communities, even societies, they develop
norms, precepts and values that constitute the structure of ‘how things should be’. They
constitute a moral order, or moral authority. On the whole, moral authority is regulative. It
sets bounds for beliefs and behaviour to keep the group functional. Moral authority
43
Marilynn B. Brewer, “The Psychology of Prejudice,” p. 435.
On this issue, see for example Thomas Homer-Dixon, They and We: An Empirical and Philosophical Study
of a Theory of Social Conflict, Ph. D. Dissertation, Massachusetts Institute of Technology, 1989.
44
97
adjudicates which beliefs and behaviours fall beyond the norm and what differences or
deviation from the norm the group or society is willing to tolerate. Moral authority can,
however, become intolerance and prejudice when it turns into doctrine. Absolute moral
orders consecrate one, and only one, morality. An absolute moral order is established as
the true and right, as the good. But absolute moral orders also mean the impossibility of
deviation or shades of grey. As Brewer contended, “[w]hen the moral order is seen as
absolute rather than relative, moral superiority is incompatible with tolerance for
difference.” 45 They are ideological much more than regulative. They delineate exclusionary
boundaries. 46 Those who do not subscribe to sanctified morality are beyond it. They are in
the wrong, their values are misguided. Those not abiding by prescribed morality are thus
immoral. Perceptions of moral superiority institute a Manichean opposition between
groups. Comparison is made on respect of moral prescriptions. It is an either or situation.
Groups, and their members, either fall on the right side by complying, or on the wrong
side, by straying. Such sentiments do not necessarily lead to outright aggression between
groups. They do, however, prove to be a justification for discrimination, oppression or
domination of those not following prescribed moral codes: the ‘outsiders’, ‘deviants’, the
‘lesser’. 47
More importantly, moral superiority works to heighten ingroup attachment. On the
one hand, because the opposition is binary, negative reciprocity operates. The ‘we’ tends
45
Marilynn B. Brewer, “The Psychology of Prejudice,” p. 435.
İsmail Beşikçi discussed this issue in strong words with regard to state morality. He stated that
“[s]ociological realities are denied by means of an official ideology [i.e., morality]. Official ideology is not just
any ideology. Official ideology implies legal sanction. Those who stray outside the boundaries of official
ideology are shown the way to prison.” Quoted in Joel S. Migdal, State in Society: Studying How States and
Societies Transform and Constitute One Another (Cambridge: Cambridge University Press, 2001), p. 22.
47
See, for example, Jim Sidanius, “The Psychology of Group Conflict and the Dynamics of Oppression: A
Social Dominance Perspective,” in S. Iyengar and W. McGuire, eds., Explorations in Political Psychology
(Durham, N.C.: Duke University Press, 1993). Strategies to drive in the difference between groups include
stereotyping and self-stereotyping, conformity processes, prejudice processes and discrimination. As Simon
and Klandermans explained, “stereotyping and self-stereotyping processes at the cognitive level and
conformity processes at the behavioral level accentuate intragroup similarities and intergroup differences. In
addition, prejudice processes at the affective level and discrimination processes at the behavioral level induce
group members to see their in-group in a positive light vis-à-vis relevant out-groups and to secure a privileged
position for their in-group.” Bernd Simon and Bert Klandermans, “Politicized Collective Identity,” p. 321.
46
98
to stand together in the face of those opposing what the ‘we’ stands for, its absolute,
supreme morality. French sociopsychologist Roger Mucchielli spoke of a similar pattern in
a handbook on recruitment and propaganda. For Mucchielli, the conscious sense that
there exists bases for a mass movement comes from “moulding a good consciousness
based on indignation toward an enemy perceived as a scapegoat [which can] legitimize
collective action based on widespread certainty of being on the side of the strongest and
the just.” 48 With perceptions of moral superiority, however, another factor plays into
heightening group solidarity: the sense of righteousness attached to moral superiority.
Social identity theory identified individuals’ basic need for a positive identity, their
‘affirmation of self-worth’. 49
For proponents of SIT, good personal self-esteem can be
attained through a positive evaluation of the ingroup. Put simply, being the member of a
group he/she esteems helps an individual feel better about himself/herself.
Moral
superiority is an absolute form of perceived group worth. As a result, a clear affirmation of
a group’s superior moral standing can justify and strengthen an individual’s attachment to
it. It makes individuals want to belong to the ‘good’, the moral group and stand up for it.
The attribution of threatening intentions also stimulates ingroup solidarity.
Attribution of blame is an important step in turning atomised frustration into directed
collective anger, a pattern identified by the frustration-aggression thesis. 50 A sense of
commonality, of having a common aim, stems from providing people with a target for their
discontent. As Simon and Klandermans recalled, “an external opponent or enemy, such as
a specific out-group, an authority, or ‘the system,’ must be blamed for the group’s
predicament,” if collective action is to take place. 51
48
Jean-Pierre Chrétien, “RTLM Propaganda: The Democratic Alibi,” in Allan Thompson, ed., The Media and
the Rwanda Genocide (London: Pluto Press/IDRC, 2007), p. 55.
49
Itesh Sachdev and Richard Y. Bourhis, “Social Categorization and Power Differentials in Group Relations,”
European Journal of Social Psychology, vol. 15, 1985, p. 416.
50
Ted Robert Gurr, Why Men Rebel, p. 33.
51
Bernd Simon and Bert Klandermans, “Politicized Collective Identity,” p. 325.
99
The attribution of threatening intentions–what framing literature also refers to as
adversarial framing–is a particular case of this. In the case of ascribing threatening intent,
not only are the culprits blamed for the ingroup’s predicament, they become the
predicament. They are made to embody danger–whether truly the case or not. Attribution
of threatening intent triggers a feeling of unavoidable confrontation and competition
between groups and those who belong to these groups. It demands reaction in the face of
a threat to what the group and its members hold dear and view as essential to their wellbeing. A threat is the source of frustration. It confronts a group with the potentiality of loss.
As Gurr pointed out, however, individuals and groups abhor loss. “Men ordinarily expect to
keep what they have; they also generally have a set of expectations and demands about
what they should have in the future, which is usually as much or more than what they have
at present.” 52 Others speak of loss-aversion. As prospect theory holds, for example,
human beings have “a general predisposition to avoid losses; hence, people will engage in
more risk, exert more effort, and persist over long periods of time to avoid losses than to
secure gains.” 53 No one wants to lose, but that is ultimately what a threat entails, a loss
which can take many shapes and forms, but, fundamentally, a loss.
In the case of threat, it creates a binary opposition. The stakes are collapsed
around loss (theft, injury, death, for example) or not. The loss comes with inaction.
Avoiding the loss demands reaction, reacting against them. 54 We lose to them or stand
against them. In other words, it is us (our loss) or them (our victory). Reaction is therefore
often portrayed as unavoidable if the group is to retain what they take to be theirs by right.
And engagement on the part of individual members is implied. Individuals fall back on their
group, lest they risk incurring, as members of the group, the losses underlying the threat.
52
Ted Robert Gurr, Why Men Rebel, p. 27.
Daniel Master, “Support and Nonsupport for Nationalist Rebellion,” p. 705.
54
It should be remembered that what is being discussed here is ‘felt mobilisation’, the frame of mind conducive
to mobilisation. When action and reaction are raised here, they mainly take the form of positioning oneself,
taking a stance against the source of the threat, not necessarily extant action per se.
53
100
Threat from a purported adversary, thus, demands rallying around the flag. A situation of
perceived threat demands that individuals put the interest of the group first, think of
themselves as members of this group above other affiliations and identities they have. The
attribution of threatening intents to a scapegoat group enhances “the salience of ‘us-them’
distinctions and thus the salience of collective identity.” 55
Overall, opposition as a mechanism to foster group solidarity is predicated on the
fact that groups in competition tend to fall back on discrimination against others and, in
parallel, to favour their own group. Creating a situation in which groups are made to feel
they are opposed, in particularly in a binary opposition, can favour group attachment. If the
situation is about ‘us’ versus ‘them,’ it tends to call forth a strong defensive fall back on the
group. This can stem from interpretations of groups’ relations built around discourses
about the moral superiority of the ingroup or the threatening nature of another. This fosters
a sense that the groups are in a zero-sum competition, one in which ‘we’ are the righteous
fighting the source of our misfortunes, the menacing others.
POLITICISATION
Politicisation, while different from opposition, taps into similar dynamics, and even
reinforces opposition. At the heart of politicisation lies the notion that while individuals
possess a number of collective identities, some of these are better vectors of collective
engagement and group solidarity. This is particularly the case with political and politicised
identities.
Political identities are by nature and function distinct from other forms of identity
such as cultural and economic ones. They are the source of our understandings and views
55
Bernd Simon and Bert Klandermans, “Politicized Collective Identity,” p. 325. As Brewer argued, “[w]hether
actual or imagined, the perception that an outgroup constitutes a threat to ingroup interests or survival creates
a circumstance in which identification and interdependence with the ingroup is directly associated with fear and
hostility toward the threatening outgroup and vice versa.” Marilynn B. Brewer, “The Psychology of Prejudice,”
pp. 435-436.
101
with regard to our station and groups’ station in the political sphere. In a narrow sense,
political identities developed with the birth of organised political entities, and remain today
tied to the formal institutions of the political realm, such as the state, its administration,
electoral competition, etc. 56 Political identities include civic or civil identities built around
one’s role as a citizen. They can also be forged in reaction to the state, as a contender or
challenger of state institutions and prerogatives. 57
Political identities also refer to a broader class of identities and relations, those
pertaining to the exercise of power relations defined broadly. The political realm
transcends formal political institutions to also encompass power relations. Such is the
reasoning behind the distinction between politics and the political. The former refers to the
formal political realm, while the latter pertains to relationships of influence, control and
struggle between actors. 58 André-J. Bélanger and Vincent Lemieux explained: “every
social relationship is political when it has this intervention component, whether on the
preferences or behaviour of an actor.” 59
The capacity for intervention–in the form of
influence or control– in the actions of others is, in a sense, an actor’s power. 60 On the
other hand, the capacity to evade others’ influence and control is also a form of power. 61
Political identities, understood in this larger sense emanate from power relations and
56
Mamdani, for example, defined political identities as resulting from the process of state formation. Mahmood
Mamdani, When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda (Princeton, N.J.:
Princeton University Press, 2001), p. 20; Doug McAdam, Sidney Tarrow and Charles Tilly, Dynamics of
Contention, pp. 134-135.
57
Mahmood Mamdani, When Victims Become Killers, pp. 22-23.
58
For example, Joel Migdal conceived of the state, society and state-society relations, what is implied in a
broader definition of the political realm, as a locus of regular struggles. He described these as “the ongoing
struggles among shifting coalitions over the rules for daily behavior [which] determine how societies and states
create and maintain distinct ways of structuring day-to-day life–the nature of the rules that govern people’s
behavior, whom they benefit and whom they disadvantage, which sorts of elements unite people and which
divide them, what shared meaning people hold about their relations with others and about their place in the
world.” Joel S. Migdal, State in Society, p. 11.
59
André-J. Bélanger and Vincent Lemieux, Introduction à l’analyse politique (Montréal: Presses de l’Université
de Montréal, 1996), p. 28.
60
Simon and Klandermans, for example, explained that “[f]rom a social psychological perspective, power is
generally viewed as a relational construct that describes a social relationship in which one party has, or is
perceived to have, the ability to impose its will on another to achieve desired outcomes [...] In short, someone
has power to the extent that he or she can control his or her own and other people’s outcomes. By the same
token, the power of a social group has typically been defined as the degree of control the group has over its
own fate and that of out-groups.” Bernd Simon and Bert Klandermans, “Politicized Collective Identity,” p. 322.
61
Ted Robert Gurr, Why Men Rebel, p. 27.
102
struggles. 62 They are tied to perceptions, beliefs and claims about the exercise of and
independence from influence, control and power more generally.
While political identities are circumscribed to this category, this does not imply that
other identities are necessarily apolitical. On the contrary, as Doug McAdam, Sidney
Tarrow and Charles Tilly indicated, “[s]een as social relations and their representations, all
identities have a political side, actual or potential.” 63 Various forms of identity, including
cultural ones, can be politicised when tied to political claims, grievances and power
relations. As Simon and Klandermans explained, “[p]oliticized collective identity is not an
all-or-nothing or on-off phenomenon. Instead, politicization of collective identity and the
underlying power struggle unfold as a sequence of politicizing events that gradually
transform the group’s relationship to its social environment.” 64
Through events and
actions, a collective identity “acquires political relevance.” 65 But the politicisation of identity
is also a process to which interpretation and shifts in perceptions contribute. Politicized
identities are the result of the specificities of the groups’ environment, but also,
fundamentally, of groups’ reading of the environment, of intergroup relations and of their
position in these.
Why do political and politicised identities matter for group solidarity? On the one
hand, political or politicised identities fundamentally recall strong sources for group
attachment. Tied to power–the exercise of power and independence from the power of
others–political and politicised identities call up values of security and participation. Both
62
Manuel Castells, The Power of Identity. The Information Age: Economy, Society and Culture. Volume II,
Second Edition (Malden, Ma.: Blackwell Publishing, 2004), p. 7. Castells differentiated between three types of
political identities: legitimising identity, “introduced by the dominant institutions of society to extend and
rationalize their domination vis à vis social actors”, resistance identity, “generated by those actors who are in
positions/conditions devalued and/or stigmatized by the logic of domination, thus building trenches of
resistance and survival on the basis of principles different from, or opposed to, those permeating the
institutions of society,” and project identity, “when social actors, on the basis of whatever cultural materials are
available to them, build a new identity that refines their position in society and, by so doing, seek the
transformation of overall social structure”. Ibid., p. 8.
63
Doug McAdam, Sidney Tarrow and Charles Tilly, Dynamics of Contention, p. 134.
64
Ibid., p. 324.
65
Bert Klandermans, “How Group Identification Helps to Overcome the Dilemma of Collective Action,” p. 890.
103
these categories of values can be fulfilled as part of a group. 66 Individuals have security
needs. They long for protection for themselves, security for their close ones. They also
need stability and a certain degree of freedom. Human beings’ sociality is partly born of
this need for security. Groups are a means to attain security, which the individual cannot
achieve alone. 67 By providing a certain degree of stability and security, groups afford
individuals a degree of independence from the influence and the control of others, a
certain freedom. Political and politicised identities also call up values of participation.
Individuals also have needs for accomplishment. But accomplishment often exists only in
so far as there is a degree of recognition on the part of others. More importantly,
individuals have needs for belongingness. Individuals need to feel they belong. Beyond
families and circles of friend, this is provided by the group. It stems from the sense of
solidarity between individuals sharing a common background, a common language or a
common culture. It stems, however, from the sense that one matters in this, that one is
made to have the privilege, the right, the importance, to belong to the group. In that sense
also, political and politicised identities ‘raise collective consciousness.’ 68 Participation then
is on a grand scale, it is about taking part in this commonality, in the ‘we’ project, and in
the future or fate of the community, nation, state, etc. Participation is therefore also about
feeling that one exercises influence in the groups one belongs to.
Political and politicised identities tap into these fundamental needs and values,
security (avoiding conflict, the control of others) and participation (belongingness,
exercising some influence), which call forth ties to the group. Because they are tied to
primary human needs and values, they belong to the category of identities that are more
prominent, more readily expressed, and more central. They belong to the category of
66
Ted Robert Gurr, Why Men Rebel, p. 27.
Social psychologists identified five functions played by collective identity. One is agency. As Simon and
Klandermans explained, “collective identity signals that one is not alone but can count on the social support
and solidarity of other in-group members so that, as a group, one is a much more efficacious social agent.”
Bernd Simon and Bert Klandermans, “Politicized Collective Identity,” p. 321.
68
Bernd Simon and Bert Klandermans, “Politicized Collective Identity,” p. 327.
67
104
identities more amenable to displacing others. More importantly, by being so intrinsically
linked to group dynamics, they are strong forms of collective attachment more amenable to
the displacement of personal forms of identity. They are identities more amenable to
making people think in terms of the group first. Displacement of personal forms of identity,
or depersonalisation, “the inclination of a person not to act as a unique individual but as a
member of a group,” is a fundamental part of heightening attachment to the group, and
consequently of collective action. 69
On the other hand, political and politicised identities are contentious–they are about
claims and struggles tied to power relations and central human values. What makes them
contentious is that influence, control and power are rarely equally distributed: “intergroup
power relations, like all power relations, are rarely symmetrical.” 70 There often exists a
reason to express grievances about the degree to which one has influence, control, and
power, or is submitted to those of others. Political and politicised are, as a result, strong
material from which to build a sense of frustration, of injustice, or of resentment. As Simon
and Klandermans claimed, “[i]n many respects, politicized collective identity is therefore
intensified collective identity with quantitatively stronger effects than its nonpoliticized
counterpart.” 71
Political or politicised identities become particularly potent at times of crisis or in
times of perceived competition between groups. Built around a struggle for influence,
control and power or the avoidance of someone else’s, political struggles come to be
about redressing power differentials benefiting others or ensuring a group’s standing.
Furthermore, as Simon and Klandermans explained, “when the power structure is unclear
or unstable, the struggle for power and the ensuing conflict may be particularly fierce
because each group is tempted to secure for itself the lion’s share or at least to prevent
69
Bert Klandermans, “How Group Identification Helps to Overcome the Dilemma of Collective Action,” p. 891.
Bernd Simon and Bert Klandermans, “Politicized Collective Identity,” p. 322.
71
Ibid. p. 327.
70
105
the other group from getting it.” 72 Intense situations of the sort can give rise to polarised
political identities. At its most basic, then, politics becomes what Carl Schmitt took it to be:
“[t]he specific political distinction to which political actions and motives can be reduced is
that between friend and enemy.” 73
Polarised identities give rise to a kind of political
difference where you must either be one or the other. You cannot partake of both. 74 As
with opposition, with such clear stakes, with sources and reasons for group attachment
clarified, these identities lend themselves to foster ingroup solidarity. 75
Reasserting a political identity or politicising identity can be conveyed through
discourse. Examples of this include injustice frames and discourses of victimisation.
Injustice frames take an unfortunate situation and read it, not as a misfortune, but as an
injustice. 76 They therefore provide a source for claims, for frustration and resentment.
Injustice frames diagnose a situation. The situation is re-articulated from immutable to
mutable, from tolerable to intolerable, and from a problem experienced individually to a
problem engaging the group. It is re-articulated as a situation demanding change. A similar
type of rhetoric are discourses of victimisation. As Benford and Snow indicated, “[a]
plethora of studies call attention to the ways in which movements identify the ‘victims’ of a
given injustice and amplify their victimisation.” 77 Examples of these types of rhetoric
include instances where group histories, myths and legends are used to claim the
existence of a secular discrimination or discourses alleging social, economic or political
discrimination at the hand of a political regime or another group.
72
Ibid.
Carl Schmitt wrote with the relations between states in mind. However, his perspective has relevance for
domestic dynamics, especially when considering times of political turmoil during which there exists competition
between political entrepreneurs and their followers. Carl Schmitt, translated by George Schwab, The Concept
of the Political (Chicago: The University of Chicago Press, 1996), p. 26.
74
Ibid., p. 23.
75
Marilynn Brewer, for example, indicated that “politicization, an important mechanism of social change, can
be added to the factors that may contribute to a positive correlation between ingroup love and outgroup hate.”
Marilynn B. Brewer, “The Psychology of Prejudice,” p. 438.
76
David A. Snow, E. Burke Rochford, Jr., Steven K. Worden, Robert D. Benford, “Frame Alignment Processes,
Micromobilization, and Movement Participation,” American Sociological Review, vol. 51, 1986, p. 466.
77
Robert D. Benford and David A. Snow, “Framing Processes and Social Movements: An Overview and
Assessment,” Annual Review of Sociology, vol. 26, 2000, p. 615.
73
106
In summary, political and politicised identities are tied to understandings and views
of a group’s station in the political and social realms. They are, however, tied to power and
group aspirations. As a result, political identities are strong forms of collective identity,
which can be a powerful lever to compel individuals to think in terms of the group before
thinking of themselves. Ties of a political nature, by fostering feelings of attachment to the
collective fate, depersonalise. Calling on political or politicised identities is a mechanism
that reinforces opposition. Winner-takes-all competitions in the political realm determine
which group has power, including ‘power over.’ They therefore raise the issue of the fate of
the group and provide further basis for strong group attachment.
SIMPLIFICATION
Communicating mobilisation is complex. It is about conveying the need for collective
action, but doing so in an environment already rife with meanings, beliefs and reasons for
action or not–reasons for inaction, for demobilisation, always exist. On the individual level,
one is always caught between different identities, different collective allegiances, some of
which risk interfering with the decision to mobilise or not. 78 Alternatives, complexity,
fuzziness confuse, they dissipate, and rarely make for persuasive communication. Calls for
group solidarity are not an exception. Simplicity is one of the keys to persuasion.
When addressing people, complex communication, complex words and complex
ideas risk having little traction. Especially amongst a group of diverse individuals, with
varying backgrounds and different views, complex communication entails the risk of not
reaching the target audience. Members of the target audience might not have the
knowledge to understand what is being conveyed. Complex communication also means
that ideas being expressed might not be received well. Because of their complexity, they
78
Marilynn B. Brewer, “The Psychology of Prejudice,” p. 439.
107
might be misinterpreted, half understood, or their meaning diverted. All of this can affect
the impact a communication has.
Simplicity remains a golden rule for effective communication. “What is presented
should not be elaborate: on the contrary, it is better if it is simple, even simplistic. Grand
theories do not have great leverage.” 79 The simpler the message the less chances of
miscommunications and misunderstandings. The simpler the story, the easier it is to
process, to compare and relate to what one knows, and, if fitting with one’s reality and
beliefs. The simpler things are, then, the easier they are to internalise and for them to have
an impact. Just as party strategy during elections includes promoting a platform, a
synthesis or summary of the party’s policy, succinct and simple formulations are easily
apprehensible, easier to engage with. This is even more so in difficult situations. In times
of stress, of duress, people even look for simpler explanations, when trying to come to
terms with the situation they face. 80
The principles apply to calls attempting to foster group solidarity. In this case,
however, another factor complicates the process. Individuals have different senses of self,
different identities. Multiple identities dilute attachment to one specific group. Complex
identities, interlocked into other diffuse affiliations and peripheral identities, do not call up
attachment and sentiments conducive to action. A strong salient identity does. In a sense,
this is like concentrating water current in one river instead of through a number of smaller
cascading rivers: the sheer hydraulic power obtained from the current of a single river is
79
Interviewee no. 12, Butare, Rwanda, 23 March 2004. Early in the twentieth century, Gustave Le Bon already
alluded to this in his work. According to Christine Poggi, in his suggestions on the form of rhetoric to adopt to
sway crowds, he “maintained that “the laws of logic have no action on crowds.” An idea could only exert
influence on a crowd when it “entered the domain of the unconscious. [...] The language of the unconscious
lay not in reasoned discourse but in images, for “crowds, being only capable of thinking in images, are only [...]
impressed by images. It is only images that terrify or attract them and become motives for action.” [...] These
images, however, should avoid the cumbersome trappings of cause and effect, always based on logic.”
Christine Poggi, “Folla/Follia,” p. 744.
80
Wilmer pointed to this in her study of conflicts in the former Yugoslavia. As she explained, “[i]n a context
where material and security needs are so threatened, individuals are more likely to abandon ambiguity in favor
of the sense of control that simple explanations, such as scapegoating, seem to offer. Said differently, when
psychically traumatised, we look for easy answers.” Franke Wilmer, The Social Construction of Man, the State,
and War, p. 155.
108
much stronger than the diluted power obtained from water running through a number of
parallel rivers. But individuals never lose identities. Strong, salient identities are about
activation. They are a reaction to environmental cues and perceptions, which call to the
fore the sense of self adapted to the situation.
On the one hand, then, simplification is about clarifying identity perceptions to allow
for this strong, salient identity to come through. This comes as a result of clarifying identity
boundaries. Identities can have accommodative, porous and inclusive boundaries. Lines
for differentiation and belonging are at that point blurry. But identities, specifically their
boundaries, can also be clarified, they can be hardened. 81 The hardening of boundaries
occurs when clear, exclusionary categories are created from previously open identities.
The impact of hardened boundaries is what Brewer referred to as optimal identity:
unambiguous, simple Manichean identities conducive to clear sentiments of belongingness
and loyalty to the group. 82
Open or closed identity boundaries are more of a conceptual tool than a reality.
They can be perceived as such, however. Identity perceptions are simplified and
boundaries clarified by globalising categories, what could be called a process of
‘singularisation of the plural.’ 83
All group members are perceived as unequivocal
representatives of their group. ‘They’ are all the same; just like ‘we’ are all the same,
without exemptions or exceptions. Because there is no ambiguity about who they are, who
we are, and why this difference exists, the sense of belonging, of attachment to the group
is heightened. This also feeds back on itself: the more individuals are attached to a group,
81
Charles Tilly referred to this process as ‘social boundary changes’. Charles Tilly, “Social Boundary
Mechanisms.” See also François Delor, “Dynamique identitaire et quête de reconnaissance,” quoted in J.
Nizurugero, “Question des réfugiés : argumentaire des régimes successifs et leurs connotations identitaires,”
in Faustin Rutembesa, Josias Semujanga and Anastase Shyaka, eds., Rwanda: identité et citoyenneté
(Butare: Éditions de l’Université nationale du Rwanda, 2003), p. 163.
82
Marilynn B. Brewer, “The Psychology of Prejudice,” p. 436.
83
Interview with Dr. Cedric Jourde, Professor at the School of Political Studies, University of Ottawa, 20
December 2004.
109
the more important it becomes for them to maintain group boundaries clear. 84 As a result,
the more they tend to mark the difference between themselves and others, as well as
notice the similarities between themselves and fellow group members. 85 Stereotypical
group ascriptions are one of the more common tactics to convey the sense of hardened
boundaries. Stereotypes are simplistic depictions of groups, and by extension of their
members, generally favourable towards the ingroup and critical or unfavourable towards
the outgroup. 86 But stereotypes act to simplify identity environments by providing clear
identifiable depictions of group members that are, often wrongly but easily, applicable to all
group members. Stereotypes are tools to ‘singularise the plural’.
Situations are also complex because there always exists a multiplicity of visions of
groups. The social world is rife with competitors taking part in framing battles, whether
consciously or not. A large part of simplifying, therefore, resides in focusing a group’s
members and public at large on a specific, exclusive and political reinterpretation of groups
and group relations. At its most basic, this means encouraging them to ignore alternatives
or discrediting and silencing rivals altogether.
Imposing rhetoric as the main reading key of a situation is also a function of the
rhetoric’s own characteristics. It is not simply a question of eliminating competitors. Even if
there are no alternatives, it does not guarantee the adoption of a particular discourse by a
population. A key to this, though, is to create a narrative simple enough and generic
enough to accommodate all members of a group. Groups and populations are composed
of a multiplicity of diverse individuals. Depictions that are general, consistent and coherent,
84
Marilynn B. Brewer, “The Psychology of Prejudice,” p. 438.
An important study by Tajfel and Wilkes lent credence to this. Tajfel and A.L. Wilkes affirmed the existence
of a human tendency to minimise differences among units or representatives of an existing category.
Perceiving a common denominator between instances, objects or persons, leads individuals to perceive
themselves as inherently more alike. Being provided with a justification for their belonging to a group, with a
common denominator, can therefore catalyse perceptions of similarity between group members and as a result
of belonging and attachment. Bernadette Park, “Some Thoughts on the Goals of Social Psychological
Research on Stereotyping,” Psychological Inquiry, vol. 3, no. 2, 1992, p. 181. For Tajfel and Wilkes’ description
of the study and the results they obtained, see Henri Tajfel and A. L. Wilkes, “Classification and Quantitative
Judgment,” British Journal of Psychology, vol. 54, 1963, pp. 101-114.
86
Bernd Simon and Bert Klandermans, “Politicized Collective Identity,” p. 325.
85
110
but that retain a degree of flexibility and ambiguity can adapt to the different perspectives
or understandings people might hold in light of different experiences, different
backgrounds, different values, etc. They accommodate variations. They allow for ‘varied’
identification’ with them by diverse individuals, while serving as a uniting principle. 87
Generic and general renderings or narratives reflect this. They can form a simple
master frame common enough and agreed upon enough to represent all members of the
group. Addressing groups of diverse individuals and bringing them together necessitates
convincing them of the existence of a common denominator, narrating the sources of their
ties and solidarity. As Francesca Polletta explained, “[a] compelling story seems to speak
to the shared experience of a larger group or in its collective voice; but without
demonstrating its representativeness.” 88 This construction of a generic rendering further
reinforces depersonalisation, because the generic works by “effac[ing] the connections
between the particular and the general.” 89 It provides individuals with a storyline in which
they are made to realise their exemplarity, their fit, with the group. 90
To a certain extent, this is achieved through ideologies and myths. One of their
characteristics, and their strength, is their ambiguity. Although they paint reality with broad
strokes, they remain vague on details, in terms of their content. They omit certain facts
87
I wish to thank Dr. Cedric Jourde, of the University of Ottawa, for highlighting this point.
Francesca Polletta, “Contending Stories: Narratives in Social Movements,” Qualitative Sociology, vol. 21, no.
4, 1998, p. 425. In her work, Polletta focused on narratives that she equated to stories, tales, myths, anecdotes
and accounts. She also clearly distinguished between and narratives. As she explained, drawing on Fine and
on Benford, “[f]rames are ‘expressed and made concrete’, and ‘exemplified’ through narratives. However,
subsuming ‘narrative,’ under the broader category of ‘frame’ obscures differences between the two in how they
organize and represent reality, their relation to collective identities, how they engage audiences, and their
criteria of intelligibility. These differences are a function of narrative’s dependence on emplotment, point of
view, narrativity, and a limited fund of plot lines.” Ibid., p. 421. While such differences exist between frames
and narratives in the abstract, such a clear distinction is hard to fathom in reality, especially when discussing
such broad ideational structures as master frames. Frames build on narratives to have grounding in existing
backgrounds just like narratives are invented or re-invented through frames.
89
Quoting Ewick and Silbey. Francesca Polletta, “Contending Stories,” p. 425.
90
Bruce Kapferer, for example, explained that “[c]ritical in remythologization, and in its fetishistic and reified
quality, is the totalizing orientation to constrain complex and diverse experience within the dictate of the logic of
its form; put another way, to confine the paramount reality of everyday life within a finite subuniverse of
meaning.” Bruce Kapferer, “Remythologizations of Power and Identity: Nationalism and Violence in Sri Lanka,”
in Kumar Rupesinghe and Marcial Rubio C., eds., The Culture of Violence (Tokyo: United Nations University
Press, 1994), pp. 68-69.
88
111
while strongly highlighting others. The storyline, the narrative, constructed and
reconstructed around group histories, necessarily involves some degree of indeterminacy.
They are not finite events, but unfolding processes for which a precise begin and end are
hard to pinpoint. As Polletta explained claimed, “[a]ll stories explain and fail to explain”:
at the heart of all narratives is a fundamental indeterminacy, a key question that
cannot be answered or even formulated, a ‘complex word’ or concept whose
meaning remains ambiguous. That ambiguity is what sustains our attention, and
the fact that it remains at the story’s end is what calls for more stories. 91
This is best captured by traditional bedtime stories’ opening, ‘once upon a time’. It leaves
the setting, often also the location, of the story indeterminate without detracting from the
meaning or moral, if one exists, of the story. As a result, it maintains the story’s interest
and validity for a wide readership. Similarly, ideologies, myths and cultural references
used to construct identity narratives often remain superficial, a platform more than a policy,
and as such are adaptive without affecting the coherence of the group, its image or
understandings members hold of it.
Overall, simplification, as a discursive strategy, is meant to ensure a message is
heard ‘loud and clear’. It entails creating a simple narrative that is sufficiently broad and
generic enough so that people recognise themselves in it.
It clarifies the identity
environment. The creation of global categories for identity, playing on generalisations,
“attributing the traits of a few individuals to all of those belonging to the same group,” and a
certain degree of ambiguity so that everyone can be made to fit within a category, are
ways of simplifying group perceptions and encouraging group attachment. 92 It eliminates
‘shades of grey,’ hence eliminating sources of hesitation and of dilution of attachments.
The use of simple language, strong resonant images and stereotypes to characterise
groups are ways to effectively communicate simple, if not simplistic, perceptions of group
91
Francesca Polletta, “Contending Stories,” pp. 428, 440.
Personal translation. Josias Semujanga, “Formes et usages des préjugés dans le discours social du
Rwanda,” in Faustin Rutembesa, Josias Semujanga and Anastase Shyaka, eds., Rwanda: identité et
citoyenneté (Butare: Éditions de l’Université nationale du Rwanda, 2003), p. 15.
92
112
identities. It thus reinforces opposition and politicisation by creating a clear and simple
rendering of the situation, as well as of sources of group attachment and intergroup
relations.
GENERAL APPROACH
When positioning themselves as defenders of their constituency’s plight, naming and
translating their discontent, becoming public advocates of their frustrations and anger, but
all in the name of the defence of a shared bond, of higher communitarian ties, ethnocentric entrepreneurs can build appeals for mobilisation, even violence, that are both
motivating and resonate with what a population feels and knows. Appeals working with
popular fears and discontent, channelling them towards action in the name of a shared
background, can thus achieve greater leverage. It is argued, however, that this is the case
if political entrepreneurs succeed in fostering group solidarity. Convincing and resonant
appeals have traction if people are made to prioritise group attachment. Tactics to foster
group solidarity–opposition, politicisation and simplification–are as much part of the
communication of mobilisation appeals as necessity and resonance.
The framework to understand popular mobilisation and the actions taken by ethnocentric entrepreneurs to foster popular mobilisation is as follows:
(a) A certain set of background circumstances trigger a crisis, a necessary condition for
generating both discontent in a population and a change in the array of opportunities
available to incumbents or to emerging elites.
(b) Populations are rendered more vulnerable by this situation and are therefore more
receptive to elite appeals, in light of the fear and discontent produced by the situation. In
parallel, elites seek to reassert their hold on power or benefit from the newly available
opportunities. They benefit from the support of their population, from whom they can gain
both internal legitimacy for their actions and human as well as material resources. These
political entrepreneurs must thus convince the population to join them through appeals.
(c1) Appeals serve to translate atomised sentiments felt on the ground, to channel fear
and discontent, into a collective creed. Appeals to collective action and mobilisation
achieve more leverage when couched in terms of severity, urgency, efficacy and propriety.
113
FIGURE 5.4: SOCIAL AND PSYCHOLOGICAL DYNAMICS OF POPULAR MOBILISATION
STRATEGIES/
LINKAGE
RULES FOR
MOTIVATING
AND
PERSUASIVE
APPEALS AND
ACHIEVING
GROUP
SOLIDARITY
(A)
CHANGES IN
THE
ENVIRONMENT
(B)
EMERGENCE OF
GENERAL
DISCONTENT,
FRUSTRATION,
ANGER AND/OR
FEAR IN
POPULATION
HEIGHTENED
COMPETITION
BETWEEN
ETHNO-CENTRIC
ENTREPRENEURS
(C1)
ETHNO-CENTRIC
ENTREPRENEURS
FRAMING
DISCONTENT AND
APPEALS IN
‘URGENT BUT
EFFECTIVE’
TERMS AND
ACCORDING TO
RESONANT
COMMUNAL
BACKGROUND
(D)
FELT
MOBILISATION/
SUPPORT
(C2)
OPPOSING
PERCEPTIONS OF
COLLECTIVE
IDENTITIES,
POLITICISING
IDENTITIES AND
SIMPLIFYING
IDENTITY
NARRATIVES
BACKGROUND
CONDITIONS
(A)
X
INDEPENDENT VARIABLE
PROCESSUAL
(B) THROUGH (C1) + (C2)
SOCIAL TRANSLATION/FRAMING OF
STRUCTURAL CHANGES AND
DISCONTENT, FRUSTRATION,
ANGER, FEAR
(C1)+(C2)
114
Y
DEPENDENT VARIABLE
PHENOMENOLOGICAL
(D)
Furthermore, for an appeal to be successful, it should resonate with the population. It
should be grounded in perceptions they already have so that it is perceived as plausible
and fitting. Cultural backgrounds are effective in achieving this.
(c2) It is, however, through a salient sense of collective identity that individuals are made
to feel that appeals are persuasive and resonant ‘for them’. Ethno-centric entrepreneurs
must thus create perceptions of a salient collective identity. They must weave a narrative
around a heightened sense of we-ness to foster group solidarity. Certain strategies or
linkage rules can serve this purpose:
(c2-a) the fostering of oppositional intergroup relations, both procedurally by collapsing the
lines of competition so that a zero-sum environment is recreated and by developing binary
schemas of categorisation between groups;
(c2-b) the politicisation of collective identity to favour depersonalisation by tying the
individual to the fate of the group amidst power struggles;
(c2-c) and the simplification of interpretations of the situation and environment, by creating
a predominant generic narrative of stereotypical categories and identities while retaining a
certain degree of ambiguity to accommodate individual differences.
Even if described individually, these three strategies should not be taken as
essential, stand alone mechanisms. On the contrary, in practice, each tends to feed back
into the others. Politicisation provides meaning for opposition and simplification clarifies
the normative grounds of a politicised identity and distinguishes between categories, for
example.
These combine to stimulate group attachment and solidarity necessary for
individuals to engage and internalise political entrepreneurs’ appeals for mobilisation. The
shape they take are therefore as much a result of the recipes of individual political
entrepreneurs as they are of contexts.
To illustrate the theoretical contentions of this dissertation, it now turns to an indepth study of contemporary Rwanda and a complementary study of the lead-up to the
wars in Yugoslavia.
115
CHAPTER 6: METHODOLOGY AND CASE STUDY RESEARCH DESIGN
The object of this analysis is to study how elites mobilise populations, including for
violence, by using ethno-centric frames. It dissects elite action toward mobilisation to
uncover how their mobilising appeals can have traction. 1 From the outset, the dissertation
postulates that, in elite-led cases, a good part of why mobilisation occurs is precisely
because elite appeals have been effective. The question, then, is ‘what is it about certain
elite appeals that make them effective?’ This analysis focuses on studying the
mechanisms behind elite action to foster group solidarity. These are hypothesised to be as
much the foundation of successful communication strategies as general framing strategies
(blame attribution, the identification of solutions, vocabularies of emergency, efficacy, and
resonance).
Few would contend that there is an invariant path to popular mobilisation or that the
same means are used every time to bring it about. Multiple factors lead to mobilisation,
and it can be triggered by a wide array of elite (or popular) actions and discourse types. 2
The strategy adopted in this research is to study specific mechanisms, opposition,
politicisation and simplification. These three mechanisms are certainly not exclusive. They
are one hypothesised set of patterns intended to shed light on elite-led mobilisation among
others. They are, however, likely ones and as a result were chosen as the focus of the
analysis.
The analysis is intended to serve as what Alexander George and Andrew Bennett
called a plausibility probe. 3 It assesses the performance of the theoretical contentions
proposed in this dissertation. In other words, it is an exercise to validate the logic of the
1
It should be recalled that mobilisation and intergroup violence do not necessarily result from ‘felt mobilisation.’
Popular action might come as the result of other factors (coercion, economic interest, obedience), more than
as the result of the internalisation of elite messages. To confirm the role played by the internalisation of elites
messages, interviews were conducted among the Rwandese population, as described below.
2
Recall that collective action can be elite-led or mass-led.
3
Alexander L. George and Andrew Bennett, Case Studies and Theory Development in the Social Sciences
(Cambridge, Mass.: MIT University Press, 2004), pp. 75-76.
116
theoretical framework by surveying a variety of elite actions and determining if
hypothesised mechanisms–opposition, politicisation and simplification–were at play in
cases studied. If these mechanisms are at play, the analysis thus establishes the
plausibility of the generalisability of identified mechanisms, and, more broadly, the
plausibility of the proposed approach on elite mobilisation.
This chapter on methodology and research design outlines assumptions underlying
the analysis and their bearing on methodological choices, particularly the choice of
process tracing as primary method of analysis. It then turns to issues pertaining to case
studies, mainly case selection and, in the case of Rwanda, field research design.
Complexity and Contingency: Their Implications for Research
Understanding the methodological choices behind this research requires an appreciation
of the importance it attributes to the complexity and contingency of social phenomena.
Social phenomena are often complex. Complexity can be best understood by appreciating
what the term means in relation to systems or environments studied. 4 Complex systems–
structures, agents and their interrelations–are judged to be complex because they are
open systems, as opposed to closed or controlled environments. Due to this openness,
they are comprised of a large and changing number of variables, some known, others
unknown. Some relations between variables and their effects are also undetermined, but
when determined are generally non-linear.
They are often characterised by multiple
causes for one outcome; interaction effects, “where in the simplest three variable case, the
relationship between two variables is modified by the value of a third;” by feedback loops,
negative or positive, which imply varied “consequences of change in the system;” and by
4
See Thomas Homer-Dixon, “Strategies for Studying Causation in Complex Ecological-Political Systems,” The
Journal of Environment and Development, vol. 5, no. 1996, pp. 132-148, or David Byrne, Complexity Theory
and the Social Sciences: An Introduction (New York: Routledge, 1998).
117
the non-additivity of causes. 5 As David Byrne explained, this non-additivity entails that
“the combined effect [of causes] is not necessarily the sum of the separate effects.” 6 On a
related issue, holistic dynamics often characterise these systems. The principle behind
holism is that the whole is greater than the sum of the parts. The interaction of units or
variables can lead to a phenomenon or outcome with its own characteristics, dynamics
and impacts. It ‘behaves’ differently than the variables that produced it. This is often
referred to as emergent properties, “a quality which belongs to a complex as a whole and
not to it parts.” 7
Another element to consider is the contingency of social phenomena. The Oxford
Concise Dictionary defined ‘contingency’ as: “uncertainty of occurrence; chance of
occurrence; thing that may happen at a later time; thing dependent on an uncertain event;
thing incident to another; incidental expense etc.” 8 At its most basic, contingency means
uncertainty. There is no certainty of occurrence with contingent phenomena and, more
importantly, there is no certainty as to how they evolve. Contingency, or what could be
called sequential contingency, can also describe the dependence of a phenomenon on the
prior occurrence of another phenomenon or event. In such a figure, event Z only occurs if
event A has occurred; event Z is contingent on the occurrence of event A. In more
complex scenarios, however, an outcome can be dependent on a preceding chain of
events. The study of history or historical processes illustrates this: a change in a chain of
5
David Byrne, Complexity Theory and the Social Sciences, p. 20 and p. 172. Thomas Homer-Dixon
distinguished multiplicative interactive causal relations from synergistic ones. Thomas Homer-Dixon,
“Strategies for Studying Causation in Complex Ecological-Political Systems,” p. 132.
6
David Byrne, Complexity Theory, p. 20.
7
Quoting C.D. Broad. David V. Newman, “Emergence and Strange Attractors,” Philosophy of Science, vol. 63,
1996, p. 246. N. Katherine Hayles explained holism and its recurrent dismissal in the following terms: “[f]rom
the system’s point of view, there is only the totality that is its environment. So strong is our belief in analysis,
however, that we take the environment to be the artificial and the collection of factors to be the reality.” N.
Katherine Hayles, “Introduction: Complex Dynamics in Literature and Science,” in N. Katherine Hayles, ed.,
Chaos and Order: Complex Dynamics in Literature and Science (New Practices of Inquiry) (Chicago: The
University of Chicago Press, 1991), pp. 16-17.
8
The Oxford Concise Dictionary accounts for both interpretations of the word ‘contingency.’ It defines it as:
“uncertainty of occurrence; chance of occurrence; thing that may happen at a later time; thing dependent on an
uncertain event; thing incident to another; incidental expense etc.” Oxford Concise Dictionary of Current
English, Seventh Edition (Oxford: The Clarendon Press, 1983), p. 204.
118
events can change the course of history. For Stephen Jay Gould, contingency is a central
principle of all history. Indeed, in history, “[t]he final result is therefore dependent on,
contingent on, everything that came before–the unerasable and determining signature of
history.” 9
When talking about the study of complex systems, however, contingency mainly
refers to their sensitivity to small influences. Complex systems are highly sensitive to initial
conditions, to small perturbations in initial conditions, including prior factors and/or
events. 10 In a number of sciences and fields, such as climatology, this is referred to as the
‘butterfly effect.’
Derived from chaos theory, it contends that, in some instances,
“variations in initial conditions of the scale of the force of a butterfly’s wing beat can
produce vastly different weather outcomes over quite short time periods.” 11
Complexity and contingency are real, and the study of social phenomena should be
mindful of them. As Kevin Mihata explained, “[w]hatever the specifics, complex adaptive
systems differ fundamentally from simpler systems. Their complexity makes outcomes
contingent, path dependent, and highly sensitive to initial conditions, which in turn make
simple predictions or explanations of such outcomes difficult.” 12 Complexity gives rise to
complex causality. 13 It thus calls for adapted methodological strategies to address the
specificities of complex social phenomena. It calls for specific methods. Ontology, “the
assumptions that scholars make about the nature of the political world and especially
9
David Byrne, Complexity Theory, p. 40.
Thomas Homer-Dixon, “Strategies for Studying Causation in Complex Ecological-Political Systems,” p. 134.
11
David Byrne, Complexity Theory, p. 19.
12
Kevin Mihata, “Emergence and Complexity in Interactionism: Comments on David A. Snow’s ‘Extending and
Broadening Blumer’s Conceptualization of Symbolic Interactionism,” Symbolic Interaction, vol. 25, no. 4, 2002,
pp. 572-573. Mihata defined ‘complex adaptative systems’ as “systems composed of multiple, interdependent,
often ‘intelligent’ and adaptive units. The system might be as ‘large’ as a society or as ‘small’ as a brain.” Ibid.,
p. 572.
13
According to George and Bennett, “[t]he simplest form of [causal processes] is linear causality, a
straightforward, direct chain of events that characterizes simple phenomena. However, many or most
phenomena of interest in international relations and comparative politics are characterized by more complex
causality, for which the assumption of linearity is misplaced.” Complex causality involves then the
‘convergence of several conditions, independent variables, or causal chains;” “interacting causal variables that
are not independent of each other;” and path-dependency. Alexander L. George and Andrew Bennett, Case
Studies and Theory Development in the Social Sciences, p. 212.
10
119
about the causal relationships within it,” and methodology, “the means scholars use for
ensuring that their inferences about the political world are valid and reliable,” should ‘talk to
each other’; they should be in synch. 14 Consequently, traditional methodological concerns
are not suited to the type of ontology that predicates complex causality. In particular,
complex causility creates problems for falsification, traditional or more formal conceptions
of the necessity and sufficiency of effects, and raises–for some–concerns associated with
path dependence.
Falsification, Karl Popper’s approach to science, consists in evaluating the
plausibility of existing statements, hypotheses or theories by trying to disprove them. As
Gerard Delanty explained, “Popper shows that the logic of science is determined not by a
path to absolute verifiable knowledge but by attempts to falsify the results of other theories
[...].” 15 As a result, much rests on the falsifiability of statements, hypotheses and theories.
For science to progress, scientific statements must be “statements that are in principle
open to falsification [...] Tautological or metaphorical statements cannot be scientific since
they cannot be falsified.” 16
Complex phenomena create a problem for falsification.
At its simplest form,
falsification entails that different researchers try to reproduce the results obtained in a
previous experiment. Until disproved, the preliminary results are held as provisionally
robust. Considering the characteristics of complex systems, reproducing experiments is
extremely difficult. Edward Lorenz’s reflection on the ‘butterfly effect’ stemmed from this
insight. While working on weather systems, he used data “which were accurate to three
decimal places instead of to the six the computer used in internal calculations. [... T]his
produced very different outcomes because the measures differed in the fourth decimal
14
Peter A. Hall, “Aligning Ontology and Methodology in Comparative Politics,” Annual Meeting of the American
Political Science Association, Washington D.C., United States, 2000.
15
Gerard Delanty, Social Science: Beyond Constructivism and Realism (Minneapolis: University of Minnesota
Press, 1997), pp. 31-32.
16
Ibid., p. 32.
120
place.” 17 The problem is particularly acute in the social sciences and humanities. In many
disciplines, such as history, sociology or political science, researchers cannot create
experiments, let alone reproduce them. Falsification therefore comes from establishing the
robustness of claims across existing cases. Often, however, the data they produce are not
of the sort that are sufficiently regular and deterministic to serve as falsification tools, at
least in Popperian terms. Overall, complex systems are simply so complex as to make
them difficult to ‘manipulate’. The variability of social data and the open nature of social
systems render it nearly impossible to control for other effects. Researchers would have to
hold an unqualified mastery of the test case and of control cases.
Another methodological issue pertains to the ‘necessity’ or ‘sufficiency’ of effects.
Most discussions of causes and conditions center on the concept of ‘necessity’ and
‘sufficiency’. As Douglas Dion described these concepts, when explaining an outcome Y,
“[c]ondition X is necessary for Y if, for Y to occur, X must also occur. By contrast, X is
sufficient for Y if the occurrence of X implies the occurrence of Y.” 18 Accordingly, three
figures are possible: (a) a condition can be necessary; (b) it can be sufficient; or (c) it can
be necessary and sufficient. In other words, a condition is (a) necessary if it must occur for
a specified outcome to occur, but is not enough to absolutely bring about this outcome; is
(b) sufficient if, by itself, its occurrence guarantees (is sufficient for) the occurrence of an
outcome, but it is not absolutely required for the occurrence of a phenomenon; or (c) is
necessary and sufficient if it must occur, and its occurrence, on its own, suffices to bring
about a specified outcome. Examples of these are (a) “being a male is a necessary, but
not sufficient condition, for being a father”; (b) “being a father is a sufficient, but not a
17
David Byrne, Complexity Theory, p. 19.
Douglas Dion, “Evidence and Inference in the Comparative Case Study,” Comparative Politics, vol. 30, no. 2,
1998, pp. 127-128. According to the Cambridge Dictionary of Philosophy, the most common interpretation of
conditions is “in terms of subjunctive conditionals, in such a way that P is a sufficient condition of Q iff [if and
only if] P would not occur unless Q occurred, or: if P should occur, Q would; and P is a necessary condition of
Q iff [if and only if] Q would not occur unless P occurred, or: if Q should occur, Q would.” Ernest Sosa,
“Condition,” in Cambridge Dictionary of Philosophy, Second Edition (Cambridge, Cambridge University Press,
1999, p. 171.
18
121
necessary condition, for being a male”; (c) “being a father is both a necessary and
sufficient condition for being a male parent”. 19 The last figure, in a sense, solves the
puzzle; the stand-alone and necessary condition for the occurrence of a specified outcome
has been identified. Or, it might be the case if all conditions that might be individually
necessary and are jointly sufficient (only if each, A, B, C, D, all necessary individually but
sufficient together, occur can it guarantee the occurrence of outcome Z) are identified. At
that point, some, if not most, would argue that the cause has been identified. For many, a
cause is defined as “a necessary and sufficient preceding condition.” 20
This might apply to a closed system in which all variables are known. It does not
perform as neatly in complex systems. J. L. Mackie reflected on the kind of causality,
complex causality, found in complex systems. He gave the example of a fire in a house
attributed to an electrical short-circuit. While experts might make the claim that the fire was
caused by the short-circuit, Mackie noted that they are not speaking of a necessary and
sufficient form of condition. The short-circuit cannot be necessary, since other factors can
lead to a house fire without the occurrence of a short-circuit in the house. Nor can it be
sufficient, because not all short-circuits lead to fires and, when they do, it is generally the
result of a combination of other factors, such as the presence of inflammable material or
the absence of adequate sprinklers. 21 As a result, the house fire is more accurately
understood as the result of a combination of the short-circuit and others factors. The
complex condition made up of these different factors, including the short-circuit, is the
sufficient but not necessary reason for the house fire (other patterns can lead to a house
fire). As a fundamental part of this complex condition, the short-circuit is “an insufficient
but necessary part of a condition which is itself unnecessary but sufficient for the result,”
19
Taken from Norman Swartz, Department of Philosophy, Simon Fraser University, The Concepts of
Necessary Conditions and Sufficient Conditions, http://www.sfu.ca/philosophy/swartz/conditions1.htm.
20
J. L. Mackie, “Causes and Conditions,” American Philosophical Quarterly, vol. 2, no. 4, 1965, p. 245.
21
Ibid.
122
the fire. 22 This is what Mackie called an INUS condition, for the first letters of the words in
italics in the previous sentence. All insufficient but necessary factors coalesce into an
unnecessary but sufficient complex condition responsible for a certain outcome.
This take on causality better captures relationships found in complex phenomena.
This is the case with mobilisation. Factors and actions that underlie mobilisation rarely
follow linear or simple causal pathways. To begin with, mobilisation is the result of
unnecessary conditions. This is best exemplfied by the fact that we speak of elite-led
mobilisation and mass-led mobilisation. 23 This indicates that there exists two alternative
paths to mobilisation, elite action and mass led action. Neither is necessary but both are
sufficient–sufficient, however, as a catalyst to a number of structural and environmental
parameters. Within each of these processes are factors that are, with regard to that path,
necessary but insufficient. Such is the case, it is argued, with achieving group solidarity
for elite-led mobilisation. Individuals must necessarily be made to think of themselves as
members of the group if the actions of entrepreneurs are to resonate with individuals’
collective backgrounds and have traction. This is the function the three mechanisms–
opposing, politicising and simplifying–serve. Situating them in an INUS condition helps to
clarify the type of relationship they have with mobilisation. Opposing, politicising and
simplifying clearly do not have a linear simple causal relationship with mobilisation. Their
impact comes in combination with other factors, such as the openness of the system to
challenge, for contenders, or the number of competitors engaging in framing battles.
An issue related to non-linearity in terms of conditions is path dependence. Path
dependence stresses the importance of history and time. What preceded matters. History
matters. William Sewell, for example, defined path dependence as meaning “that what
happened at an earlier point in time will affect the possible outcomes of a sequence of
22
Emphasis in original. Ibid.
Stuart J. Kaufman, Modern Hatreds: The Symbolic Politics of Ethnic War (London: Cornell University Press,
2001), p. 6.
23
123
events occurring at a later point in time.” 24 A narrower definition of path dependence
centers on ‘increasing returns.’ According to Paul Pierson,
[i]n an increasing returns process, the probability of further steps along the same
path increases with each move down that path. That is because the relative
benefits of the current activity compared with other possible options increase over
time. To put it a different way, the costs of exit–of switching to some previously
plausible alternative–rise. 25
In other words, as a process unfolds and follows a certain path, this path tends to
constrain future options, making it harder to choose options outside of this path. Margaret
Levi drew on the analogy of a tree to describe this narrowing of options. A tree branches
out from its trunk. When climbing it, there are many branches to choose from. Once a
climber has chosen one and climbs higher, however, although it is possible to change
branches, s/he will tend to keep climbing the same branch. 26 In other words, as the
process unfolds there is a reinforcing ‘lock-in’. One is increasingly compelled to stay the
course.
From this initial idea, James Mahoney developed two types of path-dependent
sequences: self-reinforcing sequences and reactive sequences. 27
Self-reinforcing
structures reflect Pierson’s increasing returns dynamics. Once a path is chosen, it is
increasingly beneficial to continue on the same path and increasingly costly to change
course. Reactive sequences, on the other hand, are the result of a succession of “causally
connected events.” As explained by Mahoney, “[t]hese sequences are ‘reactive’ in the
24
William H. Sewell Jr., “Three Temporalities: Towards an Eventful Sociology,” in Terence J. MacDonald, ed.,
The Historic Turn in the Human Sciences (Ann Arbor: University of Michigan Press, 1996), pp. 262-263.
25
Emphasis in original. Paul Pierson, “Increasing Returns, Path Dependence, and the Study of Politics,”
American Political Science Review, vol. 94, no. 2, 2000, p. 252. Pierson developed his perspective on the
importance of temporality in the social sciences, including path dependence, in Paul Pierson, Politics in Time:
History, Institutions, and Social Analysis (Princeton, Princeton University Press, 2004). For comments on
Pierson’s work, see, for example, Robert Jervis, “Timing and Interaction in Politics: A Comment on Pierson,”
Studies in American Political Development, vol. 14, Spring, 2000, pp. 93-100; and Kathleen Thelen, “Timing
and Temporality in the Analysis of Institutional Evolution and Change,” Studies in American Political
Development, vol. 14, Spring, 2000, pp. 101-108.
26
Margaret Levi, “A Model, a Method, and a Map: Rational Choice in Comparative and Historical Analysis,” in
Mark I. Lichbach and Alan S. Zuckerman, eds., Comparative Politics: Rationality, Culture, and Structure
(Cambridge: Cambridge University Press, 1997), p. 28.
27
James Mahoney, “Path Dependence in Historical Sociology,” Theory and Society, vol. 29, no. 4, 2000, pp.
508-509.
124
sense that each event within the sequence is in part a reaction to temporally antecedent
events. Thus, each step in the chain is ‘dependent’ on prior steps.” 28 Hence, while selfreinforcing sequences speak of a force, a pattern, steering the course, reactive sequences
speak of a chain of events, in which each new event is necessary for the next event to
occur.
What both self-reinforcing and reactive sequences share is a contingent point of
departure. Both types of sequences are set into motion by an event that is disconnected
from previous sequences, a turning point or catalyst for the sequence. It is a “critical
juncture that triggers movements along a particular path.” 29 But if path-dependent
sequences are born of contingent events, and may be sensitive to them, they are not
determined by these initial conditions. Again, it is the sequence, the intermediate events,
that determines the outcome. 30
In path-dependent processes, this sequence can be
thought of as a dynamic equilibrium: a path more and more resistant to change emerges
from each new step in the chain of event and from positive feedbacks. 31
Path dependence is at odds with traditional methodological approaches. As
Pierson noted, conventional arguments and explanations generally assume ‘fitting’
sequences “which attribute ‘large’ outcomes to ‘large’ causes;” they assume recurrence
and predictability, and, as a result, “the irrelevance of timing and sequence.” 32 Yet this
does not mean that path-dependent explanations are flawed. They capture time in a
different manner. They capture it in a manner attuned to the nature of many social
phenomena. In reality, timing and sequences are often not as neat and as the linear
equations composed of uncorrelated independent variables found in many conventional
28
Ibid., p. 509.
Kathleen Thelen, “Timing and Temporality in the Analysis of Institutional Evolution and Change,” p. 103.
30
Jack Goldstone, “Initial Conditions, General Laws, Path Dependence, and Explanation in Historical
Sociology, American Journal of Sociology, vol. 104, no. 3, 1998, pp. 834.
31
Pierson refers to this characteristic as inertia. Paul Pierson, “Increasing Returns, Path Dependence, and the
Study of Politics,” p. 263.
32
Ibid., p. 251.
29
125
models. Some social phenomena follow more chaotic and/or more complex patterns, one
of which is path dependence. If anything, path-dependent explanations are simply tools to
grasp yet another important characteristic of social processes, their temporality.
As
Pierson explained, “[i]t is not the past per se but the unfolding of processes over time that
is theoretically relevant.” 33
A classic example of a path-dependent process is the development of
technologies. The QWERTY keyboard, for example, is known to be a suboptimal keyboard
for its users. 34
Originally found on one of the first typewriters to be manufactured, it
eventually became the predominant keyboard in use. The more predominant the keyboard
became, the more difficult it became to abandon it for a more efficient alternative,
highlighting the ‘lock in’ character of path dependence.
Many of the processes described by this research are also path-dependent,
including communicating mobilisation. Once elites choose a language and narrative, a
discursive path is set, making it increasingly costly to abandon that path. Politicians are
often criticised by their supporters if they are perceived to ‘betray the cause’, for instance.
Just as the familiarity of users made altering the QWERTY keyboard less likely, in the
case of elite rhetoric, the more it is disseminated, the more people have heard it, the more
leaders risk been held to it. The further along a path, the more the system elites created
takes on a life of its own. It becomes a Frankenstein's monster; elite actions come to be
constrained by their own rhetoric. 35 Once an ideology is well disseminated, elites are
33
Ibid., p. 264.
The story of the QWERTY keyboard has been instrumental in starting debates around path dependence. It
came as a result of the publication of an article on the topic by an economist, Paul David. See Paul A. David,
“Clio and the Economics of QWERTY,” American Economic Review, vol. 75, no. 2, 1985, pp. 332-337.
35
The idea of a ‘Frankenstein effect’ is Thomas Homer-Dixon’s. Discussion with Thomas Homer-Dixon,
Trudeau Centre for Peace and Conflict Studies, Toronto, Canada, 13 May 2005.
34
126
forced to keep referring to it and to keep functioning within the ideological system, further
reproducing it. 36
All of the issues discussed above–the type of causality, path dependence and,
overall, the complexity and contingency of the processes studied–call for an adapted
method of inquiry. The starting point of research into complex social phenomena should be
to ascertain their nature amidst complexity. To understand some complex outcomes, it is
necessary to find the path that led to them, to retrace the intermediary steps, like following
Ariadne’s thread in a maze of factors and events. This is what process tracing, the method
employed in this analysis, does.
As Peter Hall stated, “process tracing is a methodology well-suited to testing
theories in a world marked by multiple interaction effects, where it is difficult to explain
outcomes in terms of two or three independent variables–precisely the world that more
and more social scientists believe we confront.” 37 Process tracing is a tool to handle the
problem of equifinality. By reconstructing the chain of events, “it offers the possibility of
mapping out one or more potential causal paths that are consistent with the outcome and
the process-tracing evidence in a single case.” 38 It does so by exposing the causal
mechanisms behind processes. 39 Process tracing is thus a fitting method for this research
because it explicitly speaks to the goal of uncovering the mechanisms behind
communicating mobilisation. Furthermore, process tracing does not limit itself to
36
This phenomenon is also referred to as ‘network externalities’. As Stan J. Liebowitz and Stephen E. Margolis
explained, “Network externality has been defined as a change in the benefit, or surplus, that an agent derives
from a good when the number of other agents consuming the same kind of good changes. As fax machines
increase in popularity, for example, your fax machine becomes increasingly valuable since you will have
greater use for it.” Stan J. Liebowitz and Stephen E. Margolis, “Network Externality”, in Peter Newman, ed.,
The New Palgrave Dictionary of Economics and the Law (London: Palgrave MacMillan, 1998). See also, for
example, W. Brian Arthur, “Competing Technologies, Increasing Returns, and Lock-In by Historical Events,”
Economic Journal, vol. 99, 1989, pp. 116-131.
37
Peter A. Hall, “Aligning Ontology and Methodology in Comparative Politics,” p. 18.
38
Alexander L. George and Andrew Bennett, Case Studies and Theory Development in the Social Sciences, p.
207.
39
Andrew Bennett and Alexander L. George, “Process Tracing in Case Study Research,” Paper presented at
the MacArthur Foundation Workshop on Case Study Methods, Belfer Center for Science and International
Affairs (BCSIA), Harvard University, October 17-19, 1997, p. 2.
127
reconstructing a historical narrative. It can also be a theory-laden methodological tool. For
George, process tracing can be a vital part of an adequate ‘plausibility probe’, an
instrument to check on the logic of theoretical arguments. 40 The validity of hypotheses on
mechanisms can be assessed by tracing back how they operated in the given sequence of
a real case. This is the goal of this study. It is intended as a ‘building block’ analysis
proposing hypothetical mechanisms, opposing, politicising and simplifying, to be assessed
through process tracing exercises on case studies.
Case Studies and Field Research Design
CASE SELECTION
Case study research is increasingly recognised as an important research method capable,
in certain areas, of outperforming quantitative approaches. As Bennett explained, though
quantitative methods have their strengths, “[c]ase study methods are superior at process
tracing and identifying causal mechanisms, identifying omitted variables, and measuring
qualitative variables, and they also have the advantage in the genetic explanation of
individual cases and of path-dependent processes.” 41
For the purpose of this analysis, two case studies are undertaken: an in-depth
analysis of recent Rwandese history (involving fieldwork) and a complementary study of
the lead-up to the wars in Yugoslavia (not involving fieldwork, due to resource constraints).
Limiting the research to a few cases allows for an in-depth observation of dynamics and
mechanisms at play in each case. 42 Case studies and small-N approaches are not without
their critics, however. A number of scholars have raised the issue of selection bias in
40
Alexander L. George, “Case Studies and Theory and Theory Development: The Method of Structured,
Focused Comparison,” in Paul G. Lauren, ed., Diplomacy: New Approaches in History, Theory and Policy
(New York: Free Press, 1979), p. 52.
41
Andrew Bennett, “Causal Inference in Case Studies: From Mill’s Methods to Causal Mechanisms,” paper
presented at the 1999 Annual Meeting of the American Political Science Association, Atlanta, United States,
1999, p. 3.
42
Michael Coppedge, “Thinkening Thin Concepts and Theories: Combining Large N and Small in Comparative
Politics,” Comparative Politics, vol. 31, no. 4, 1999, p. 465.
128
small-N analyses. 43 Others, however, have defended the specific strength of this type of
research. According to David Collier and James Mahoney, small-N research does not
systemically yield biased results. 44 They can be a potent tool for ‘building block’ research.
Single case studies and small-N analyses entail selecting cases “that focus on certain
outcomes of exceptional interest, for example, revolutions, the onset of war, the
breakdown of democratic and authoritarian regimes, and high (or low) rates of economic
growth.” 45 Accordingly, they allow for thorough observation of how processes unfold. They
produce “nuanced, conditional, complex, and contextualized hypotheses” and they tend “to
develop ‘thick’ (complex or multidimensional) concepts and theories”. 46
The recourse to two cases constitutes a strategy in itself. On the one hand, a
single case study can never completely eliminate suspicions of uniqueness, that the case
is an exceptional one, with particular characteristics and dynamics. The analysis of two
cases is therefore a strategy to check against uniqueness. If anything, the analysis of two
cases in which similar patterns are identifiable confirms the plausibility of generalisability of
identified patterns. This reasoning guided the choice of Rwanda and Yugoslavia as case
studies for the analysis. In light of the scale and intensity of violence unleashed during the
genocide, Rwanda is often thought of as an exceptional case. Whether or not this is so,
this does not mean tactics employed by elites were themselves exceptional. The recourse
to a complementary case, Yugoslavia, helps check against uniqueness, however. Both
Rwanda and Yugoslavia were major, well documented conflicts in the early 1990s, with a
43
David Collier and James Mahoney indicated that a bulk of the research raising the issue of selection bias in
small-N samples comes from the fields of econometrics and evaluation research. For a listing of this type of
work see Alexander L. George and Andrew Bennett, Case Studies and Theory Development in the Social
Sciences, p. 84.
43
David Collier and James Mahoney, “Insights and Pitfall: Selection Bias in Qualitative Research,” World
Politics, vol. 49, no. 1, 1996, p. 56, infra 2.
44
Alexander L. George and Andrew Bennett, Case Studies and Theory Development in the Social Sciences, p.
84.
45
David Collier and James Mahoney, “Insights and Pitfall,” p. 57.
46
Michael Coppedge, “Thinkening Thin Concepts and Theories,” p. 465. For Michael Coppedge, both large-N
approaches of a quantitative nature and small-N approaches of a qualitative nature have strengths and
weaknesses.
129
number of shared characteristics (see below). There is, however, a key difference between
the two: their geographic location. One is located in Africa, the other in the Balkans. This
implies very different historical trajectories for the two cases and certainly also suggests
economic, social and cultural differences. In light of these important differences,
identifiable similarities between patterns found in both cases, certainly would suggest the
potential for generalisability of the analysis’ contentions. Leaders’ recourse to similar
tactics to mobilise their population in Rwanda and Yugoslavia would point to possible
trends worth investigating further in other cases.
Beyond these broad motives, the choice of the cases for this analysis, pregenocide Rwanda and the lead-up to the wars in Yugoslavia, also rested on a number of
specific criteria: groups involved had to have a history of interrelations; movements had to
be predominantly elite-led; and popular mobilisation had to have occurred. In line with the
first criterion, groups were required to have had a history of interaction in order to have
developed views of one another.
Since this analysis studies the use of communal
backgrounds and perceptions of other groups in mobilisation appeals, at a minimum, such
perceptions needed to exist. Second, though not all forms of mobilisation are the result of
elite action, this analysis specifically focuses on strategies employed by elites to foster
collective action, on elite-led mobilisation.
Finally, elite action had to have led to the
outcome of interest: mobilisation had to have occurred. Both the Rwandese and
Yugoslavian cases shared these characteristics.
The principal method of inquiry employed for these case studies is historical. Much
has been written about developments in Rwanda and the lead-up to the wars in
Yugoslavia. As a result, considering the sweeping nature of mobilisation efforts prior to
the genocide, many accounts of strategies deployed by elites in both countries and their
rhetoric are readily available.
130
FIELD RESEARCH
In the case of Rwanda, fieldwork was conducted. It served specific ends. It was intended
as a way to probe at first hand beliefs held by the Rwandese population with regard to
groups in society (Hutu and Tutsi). While much work focuses on hateful rhetoric
propagated during the 1990-1994 civil war, little exists on social backgrounds and ideology
in previous decades. The purpose was to gain a better understanding of Rwandese
identity, their cultural and social backgrounds, and of which of these aspects were used as
building blocks to foster antagonistic perceptions of events and group identities. Fieldwork
was also meant to gauge the impact of elite strategies among the population. Often,
analyses on mobilisation strategies assume the success of mobilisation efforts if
mobilisation does occur. Other factors, however, can be at play. Fieldwork therefore also
served to establish if the Rwandese population identified belief in Hutu extremists’ rhetoric
as a lever for mobilisation. Accordingly, fieldwork was not conducted on the research
question per se (what were political entrepreneurs’ strategies to mobilise the population),
but it was nonetheless necessary to get a sense, directly from the population, of what
Rwandese believed and the role these beliefs played in recent events.
Given the importance of popular beliefs, the principal data collection method used
during fieldwork was interviews. Following an ethics review process, open-ended
interviews were conducted directly with the population in Rwanda between February 2004
and April 2004. 47 Over the course of the fieldwork, thirty-one respondents from the cities of
Kigali and Butare were interviewed, some on more than one occasion to ensure
consistency in answers and to delve further into certain issues. Interviews lasted from one
47
Review #10909 by the Social Sciences and Humanities Research Ethics Board, University of Toronto.
131
to two hours. For each interview, verbal informed consent was obtained and confidentiality
and anonymity were guaranteed to the respondent. 48
Interviewees were asked to recall teachings and beliefs about Rwandese
communal groups under Hutu regimes, as well as the rhetoric overheard between 1990
and 1994 civil war. In parallel, these interviews were intended to assess the effects that
antagonistic rhetoric by ethno-centric entrepreneurs had on beliefs and actions among the
general population. Consequently, questions also pertained to intergroup relations prior to
the civil war to establish the starting base of inter-communal relations, to see if
respondents perceived a shift in perceptions of collective identity, to identify the actors or
factors responsible for this shift, and to shed light on respondents’ perceptions of the
impacts of such a shift (See annexed questionnaires, in English and French).
It was clearly established, at the start of the interview, that if a respondent felt at
any time uncomfortable about answering a question or about the entire interview process,
s/he could decline to answer the question or terminate the interview. Due to the sensitive
nature of the issues addressed by the research, the questionnaire was also developed to
avoid traumatising interviewees. A number of strategies were adopted. First, questions did
not raise the genocide directly. In an introductory statement read before interviews, it was
made clear that events that occurred after the start of the genocide would not be
48
At the start of each interview, a brief statement described the goals of the research and provided some
information on the researcher. Following this, participants were invited to take part in the research and then
clearly asked if they agree to be interviews. The statement by querying participants on questions or comments
they might have had about the interview process and research. While the terms for ethical research involving
humans set by the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (TCPS) show
a preference for written informed consent, in certain instances it is tolerated that oral informed consent be
obtained instead. The case was made to the Social Sciences and Humanities Research Ethics Board at the
University of Toronto that an oral informed consent was more suitable for this research. Many individuals
interviewed in preparation for field work (former expatriates in Rwanda and members of the Rwandese
diaspora in Canada) described Rwandese society one in which trust is at issue. The Rwandese are wary of
interactions with individuals they are not familiar with. The current political climate in Rwanda is also feeding
wariness. The government is proving increasingly more intolerant of dissenters and opponents, leading people
to feel that they should steer clear of activities that might raise suspicion, including international research in the
country. Approaching interviewees with a form to confirm their informed consent might have deterred
individuals from participating because of the written documents involved. With an oral informed consent
procedure, they could that they would not have signed anything, thus limiting the traces of their participation in
the interview process.
132
addressed. Raising issues tied to acts perpetrated during the hundred days of the
Rwandese genocide could have brought back difficult memories for participants or placed
them in a situation in which they would have been made to discuss incriminating or
traumatising acts. In either case, since this research focused on ‘felt mobilisation’, or
changes in group perceptions leading individuals to accept the necessity of participation
and not mobilisation per se, centering the questions on the pre-genocide period sufficed.
Furthermore, the questionnaire was built so as to make no direct reference to
respondents’ personal experiences. Questions focused on perceptions of group identities
and group interactions, and were always phrased to make reference to general patterns
and events. Questions were therefore cast in relation to the group, not the individual. This
strategy had other benefits. A number of researchers and journalists who conducted
interviews in Rwanda indicated that Rwandese respondents feel more comfortable
answering questions enquiring about the group as a whole more than about them in
particular. For example, Jean Hatzfeld who interviewed some of the génocidaires indicated
that they avoided his questions when they made references to specific aspects of their
personal lives but were inclined to talk when questions were phrased around the group or
community as a whole. 49
Finally, questions did not enquire as to the group affiliation of the respondent. As a
result of the 1994 genocide, it is now reproved in Rwanda to ask an individual to divulge
his or her communitarian identity. This has led to the removal of its mention on identity
cards (communitarian identity–Hutu, Tutsi or Twa–had been inscribed on identity cards
since the 1920s) and to a policy that all nationals be indiscriminately called ‘Rwandese.’
Furthermore, in post-genocide Rwanda, some identities have become–or remain–
politicised. Categories such as Tutsi survivor, Tutsi returnee, Hutu returnee, etc., have
come to be tied to normative, sometimes pejorative, meanings and to standing in
49
Jean Hatzfeld, Une saison de machettes (Paris: Éditions du seuil, 2003).
133
Rwandese society. These categories have accrued enough importance to, in some
instances, signify the right to speak or not. 50 As a result, genocide politics in Rwanda
continue to be a source of frustration or even injury. Employing Hutu and Tutsi labels
during interviews could have risked stigmatising individuals or possibly traumatising them
in light of the current trends in the country. In all interview questions, the issue of a
respondent’s background was not raised. References were always made to Hutu and Tutsi
simultaneously so that respondents never had to divulge their communitarian affiliation.
Working in Rwandese society and the sensitive nature of issues addressed also
required certain specific measures. Pre-fieldwork meetings with individuals having lived in
Rwanda (former expatriates in Rwanda and members of the Rwandese diaspora in
Canada) flagged that the Rwandese are generally reserved and not inclined to open up to
strangers. For this reason, extensive networking was required to establish contacts and
build trust. Potential participants were approached through intermediaries. These
intermediaries were met through an affiliation negotiated with CARE Rwanda, a non
governmental organisation operating out of Kigali, and through contacts made at a hostel
frequented by nationals.
Although much of the selection process relied on contacts, there were established
criteria for the selection of potential interviewees. They had to have lived in Rwanda
between 1990 and 1994, at a minimum, and had to have been, at the time, at an age to
have a good understanding of the events taking place. This criterion served to exclude
people who were younger than of high school age at the time of the civil war. Furthermore,
because the research sought to assess the impacts of strategies used to achieve popular
50
The current Rwandese Patriotic Front dominated government, in power since the end of the genocide, has
become increasingly intolerant of accounts straying from their interpretation of events. In essence, some
genocide experiences are encouraged to be voiced, mostly from Tutsi survivors, while others, such as the
stories of Hutu coming back from months in camps after the RPF take over, are silenced. On the impact of the
current government’s policies, see for example, Susan M. Thomson, “The Politics of Trauma in Post-Genocide
Rwanda: Treatment for Some at the Expense of the Many,” talk given at the Peacebuilding and Trauma
Recovery: Integrated Strategies in Post-War Reconstruction Conference, University of Denver, Denver, United
States, 25 February 2007.
134
mobilisation by elites, interviewees had to be representative of the general population at
the time of the civil war; they had to be representative of those submitted to the ethnocentric rhetoric and not of the elite group promoting it.
The sample of respondents proved somewhat biased. Interviewees were residents
of Kigali and Butare, the two largest cities in Rwanda. Respondents were urban nationals
when in reality the urban population accounts for only a quarter of the entire Rwandese
population. 51 Furthermore, all interviews were conducted in French or English instead of
in Kinyarwanda, the national language.
While French remains a common second
language in Rwanda, in light of the country’s history of Belgian colonisation, individuals
speaking a second language are generally more educated or work in better paying service
jobs, for international non-governmental organisations, or the government. In the general
population, however, the level of education beyond primary school is low. Around 40
percent of the population has completed primary education and 14 percent are enrolled in
secondary schools. 52 Overall, most respondents belonged to the smaller portion of
urbanised and educated population instead of the more typical profile, that is, rural and
with a low level of education. Phrasing questions around perceptions in communities in
general at the time of the civil war instead of around personal perceptions was intended to
solve this bias somewhat.
Due to the political climate in Rwanda, a second type of bias was to be expected.
In light of the current government’s intolerance for dissenting accounts, including about the
civil war and genocide, a number of individuals might have been afraid to take part in the
study. On the other hand, certain categories of individuals in Rwanda, Tutsi survivors for
example, are encouraged to speak out. In gathering background information on
respondents prior to interviewing this was a source of concern. Despite this, certain types
51
According to UNICEF, 22 percent of the population was urbanised in 2005. UNICEF, Rwanda–Statistics,
http://www.unicef.org/infobycountry/rwanda_statistics.html.
52
UNICEF, Rwanda–Statistics, http://www.unicef.org/infobycountry/rwanda_statistics.html.
135
of individuals, Tutsi survivors, for example, might be more represented in the sample.
Without asking information as to individuals’ communitarian backgrounds, this is
impossible to ascertain, however.
In light of possible biases, archives, recorded political rhetoric, media transcripts,
accounts by nationals, as well as academic material which included its own interviews,
were used to complete interview data, and also to corroborate it. With ten years passed
between events in Rwanda and interviews, the accuracy of interviewees’ recollections was
also an issue. Secondary material thus served to assess the veracity of interviewee
statements with records and written accounts dating from the time of the civil war and
genocide or soon after.
136
CHAPTER 7: RWANDA: THE DEVELOPMENT OF AN ETHNO-CENTRIC AND DIVISIVE
IDEOLOGICAL BACKGROUND IN RWANDA 1
Pre Colonial Rwanda
Little is certain about Rwandese history before the country was colonised by Germany in
the late nineteenth century. Much of what is known of pre-colonial Rwanda rests on oral
accounts, since written history only developed after the arrival of Europeans in the region.
The accuracy of historical accounts about this complex society is therefore difficult to
gauge. 2 Furthermore, all historical ‘truths’ were blurred in the last century, the result of
historical reinterpretations by successive regimes in Kigali. As one Rwandese historian
explained: “Rwanda has moved from myth to myth.” 3 Each new regime reinvented a
mythical history for the country and presented it as a historical reality to legitimise their
rule. 4 As Philip Gourevitch argued: “Rwandan history is dangerous. Like all of history, it is
a record of successive struggles for power, and to a very large extent power consists in
the ability to make others inhabit your story of their reality–even, as is so often the case,
when that story is written in their blood.” 5
In last fifty years, in particular, two competing interpretations of Rwandese history
dominated.
During the two ‘Hutu Republics’ preceding the genocide, pre-colonial
1
Ideology is defined as “the set of beliefs that are used to justify or challenge a given social-political order and
are used to interpret the political world.” As the argument focuses on the development of collective political
identities, looking at ideology instead of Rwandese culture as a whole is warranted. Ideology is, at any rate, a
subset of culture, which is a broader set of societal beliefs that do not necessarily pertain to the political world.
For this definition of ideology, see Mayer N. Zald, “Culture, Ideology and Strategic Framing,” in Doug McAdam,
John D. McCarthy and Mayer N. Zald, eds., Comparative Perspectives on Social Movements: Political
Opportunities, Mobilizing Structures, and Cultural Framing (New York: Cambridge University Press, 1996), p.
262.
2
Despite what is often believed, according to many contemporary historians, during the pre-colonial era the
country was not as centralised as is often upheld. The monarchy’s hold on the country was stronger in certain
regions while others managed to remain more autonomous. As David Newbury explained, “both the diversity of
micro-environments within, and the diverse continuities with areas outside of the region, belie the homogeneity
assumed in most accounts of social structure in this area.” David Newbury, “Precolonial Burundi and Rwanda:
Local Loyalties, Regional Royalties,” The International Journal of African Historical Studies, vol. 34, no. 2,
2001, p. 263. Certain populations also preferred to remain peripheral to centers of power. For example,
according to Alison Des Forges, “[t]he pastoralists known as Bugogwe, clustered in the northwest, and those
called Bahima, located in the northeast, sought to avoid state power rather than to share in it.” Human Rights
Watch, Leave None to Tell the Story: Genocide in Rwanda (New York: Human Rights Watch, 1999), p. 32.
3
Personal translation. Interviewee no. 8, Kigali, Rwanda, 16 March 2004.
4
On representations of history in Rwanda and their impacts and dangers, see for example Nigel Eltringham,
Accounting for Horror: Post-Genocide Debates in Rwanda (London: Pluto Press, 2004).
5
Philip Gourevitch, We Wish to Inform You that Tomorrow We Will Be Killed with Our Families: Stories from
Rwanda (New York: Picador, 1998), p. 48.
137
Rwandese history was described as a period which saw, around the seventeenth century,
the invasion of the country by Tutsi cattle herders from the North. Once on Rwandese
territory, these pastoralists developed an oppressive monarchical system that bonded the
Hutu masses to the Tutsi monarchy. Pre-colonial Rwanda was, according to this view, a
political entity marked by class-based oppression and a strong antagonism. More recent
accounts, appearing in the aftermath of the genocide, argue that pre-colonial Rwanda was
characterised by peaceful co-habitation and great unity among the three components of
Rwandese society, the Twa, Hutu and Tutsi, and that all were ruled by a benevolent Tutsi
monarch. 6 The divisions and ideology that festered and led to the 1994 events stemmed, it
is said, from the colonial administration which favoured one group, the Tutsi, and used
them as intermediaries in an indirect administrative system. 7
Although still the subject of much speculation, recent research seems to indicate
that a more accurate version of the facts lies somewhere in between these two contrasting
accounts. 8 Some of the writings by the first Europeans in Rwanda appear to confirm this.
According to the descriptions of one of the first Germans to arrive in the country, Count
Von Goetzen, tensions did exist between groups, especially in the form of class struggles
between Hutu and Tutsi. But tensions also existed inside groups, particularly between the
6
The Twa are the lesser known of the three Rwandese groups. Comprising about one percent of the
population, they have generally been marginalised throughout history by both the Hutu and Tutsi but have
given without consistency their allegiance to Tutsi or Hutu. During the genocide, for example, the Twa were
both among the victims and killers.
7
This view is defended by the post-genocide Rwandese regime. In a document produced by the National
Unity and Reconciliation Commission, a speech by President Paul Kagame, delivered in 23 November 2001, is
quoted. Kagame stated that “colonial masters and religious preachers capitalised on ‘dived and rule policy.” [...]
It’s from the colonial rule that sprouted antagonism between Hutu and Tutsi.” National Unity and Reconciliation
Commission, Report on the Evaluation of National Unity and Reconciliation, Republic of Rwanda, Kigali,
Rwanda, June 2002, p. 5. One of the better known documents supporting this is Office of the President of the
Republic, The Unity of Rwandans: Before the Colonial Period and Under the Colonial; Under the First
Republic, Republic of Rwanda, Kigali, Rwanda, August 1999.
8
Peter Uvin, “Reading the Rwandan Genocide,” International Studies Review, vol. 3, no. 3, 2001, p. 78. For
authors who support this view, see for example Danielle de Lame, Une colline entre milles ou le calme avant la
tempête. Transformations et blocages au Rwanda rural (Tervuren: Musée royal de l’Afrique centrale, 1996);
Catharine Newbury, The Cohesion of Oppression, Clientship and Ethnicity in Rwanda 1860-1960 (New York:
Columbia University Press, 1988); David Newbury, “Precolonial Burundi and Rwanda,” pp. 255-314; and Jan
Vansina, Antecedents to Modern Rwanda: The Nyiginya Kingdom (Madison, Wisc.: University of Wisconsin
Press, 2004).
138
petits and grands Tutsi. 9 The petits Tutsi, whose living conditions resembled those of the
Hutu labourers more than those of the Tutsi aristocracy, were closer to the former than the
latter.
Similarly, the Hutu benefited from a clientele system tying them to the Tutsi
aristocrats and they tended to support their patrons.
Beyond the uncertainty that exists
around the nature and depth of tensions in Rwandese society, the most contentious
aspect of its history relates to the settlement of the region. Archeologists, ethnologists and
historians disagree as to whether groups share a common ancestry or originated from
different regions. In the case of the different origins argument, scholars are also unsure of
the arrival in Rwanda of the different groups, though many believe groups arrived at
separate points in time.
What much of the recent literature does agree on, however, is that the categories
Twa, Hutu and Tutsi, long viewed as primordial, actually changed in nature over the
centuries. These are believed to have been at first socio-economic categories, a schema
to organise social relations within Rwandese society in the form of a crude division of
labour. The Twa were a group of hunters and potters, the Hutu dedicated themselves to
agriculture, and the Tutsi were pastoralists.
Relations between these groups began
changing in the eighteenth and nineteenth centuries due to land scarcity.
As control of land became an issue, and with grazing pastures slowly beginning to
encroach on agricultural land, conflicts developed, particularly between agriculturalists and
pastoralists. Furthermore, in a world in which cattle and land were the principal sources of
enrichment, the Tutsi pastoralists began asserting their power in certain parts of the
country. One clan in particular, the Nyiginya, began asserting their power. With centuries
spent spreading their dominance, the Nyiginya eventually became the ruling dynasty of
9
The petits Tutsi were the lower classes while the grands Tutsi were the aristocracy. Interview with Gérard
Prunier, Centre français des études éthiopiennes, Addis Abeba, Ethiopia, 25 April 2004.
139
Rwanda. 10 In parallel, two trends spread: a form of territorial organisation, domains called
Ibikingi, and a form of bond to powerful land owners, the Uburetwa. The Ibikingi were
domains that the Mwami, the king, granted to his chiefs or followers. Individuals living on
these lands had to pay a form of rent to their owners. Cattle owners gave a cow to have
access to the pastures. Agriculturalists fell under the Uburetwa system. To cultivate the
land, they gave part of their harvest to their patron or worked for them for two out of four
days. 11 More Ibikingi concessions were granted, with a parallel extension of the Uburetwa,
without concern for previous land tenure rights, thus exacerbating the growing conflict
between pastoralists and agriculturalists.
These conflicts were exacerbated by the institutionalisation of distinctions between
Tutsi and Hutu. The term Tutsi had long been used to refer to the small group of
pastoralists in the country. The distinction between Tutsi and Hutu deepened with the
creation of an army for the Mwami composed of three corps, one comprised of Tutsi
warriors, another, also Tutsi, taking care of the cattle, and a third serving the warriors,
composed mostly of agriculturalists.
As Vansina argued: “[a]s most noncombatants
happened to stem from lineages of farmers, the elite eventually began to call all farmers
‘Hutu’ and to oppose this to the word ‘Tutsi,’ now applied to all herders, whether they were
of Tutsi origin or not.” 12 Combined with the extension of the Ibikingi and Uburetwa, these
categories took on a class based connotation. The nature of these categories changed
again with the arrival of European colonisers, who saw them as race based, and with the
first and second Rwandese republican political systems, which treated them as ethnically
based political identities.
10
David Newbury, “Precolonial Burundi and Rwanda,” p. 290. For more on the Nyiginya Kingdom, see Jan
Vansina, Antecedents to Modern Rwanda.
11
Jean-Paul Kimonyo, La participation populaire au Rwanda: de la révolution au génocide (1959-1994),
doctoral thesis, Département de science politique, Université du Québec à Montréal, Montréal, Canada,
September 2002, p. 47.
12
Jan Vansina, Antecedents to Modern Rwanda, p. 135.
140
As these different shifts indicate, however, these categories were fluid. So were
their boundaries in pre-colonial Rwanda, particularly between Hutu and Tutsi. “Individuals
could and did move between the categories Hutu and Tutsi as their fortunes rose and
fell.” 13 Wealth, particularly in terms of owning cattle, could make one a Tutsi and the loss
of assets a Hutu. The Rwandese also shared a multiplicity of other collective identities
which overlay personal ones, such as place of birth, family lineages, etc. One of the most
important among these was clan affiliation which in most cases cut across Twa, Hutu and
Tutsi categories. 14 Most pre-colonial Rwandese defined themselves in terms of belonging
to one of the fourteen major clans or to minor ones.
Despite greater polarisation towards the mid-nineteenth century, there still existed
factors of integration among the Rwandese society. Certain traditions also helped unify
the Rwandese, most importantly a common language, a common culture and a common
religion.
They also shared a common foundational myth: they all descended from
Gihanga, translated as ‘creator’ whose son Kanyarwanda had had three children, Gatwa,
Gahutu and Gatutsi, who became themselves respectively the fathers of all Twa, Hutu and
Tutsi. 15 Furthermore, the Rwandese groups were well integrated on a community level,
living side by side on the hills, even frequently intermarrying. 16
Mahmood Mamdani
indicated that “all accounts [he] heard or read speak of considerable intermarriage:
anywhere from a significant minority to a majority of contemporary Rwandans are likely to
13
African Rights, Rwanda: Death, Despair and Defiance, Revised Edition (London: African Rights, 1995), p. 3.
Jean-Pierre Chrétien, “Mythes et Stratégies autour des Origines du Rwanda (XIXe-XXe siècles), in JeanPierre Chrétien and Jean-Louis Triaud, Histoire d’Afrique: Les Enjeux de Mémoire (Paris: Éditions Karthala,
1999), p. 289.
15
Many different versions of this myth exist as well as politicised reinterpretations of it, some insisting on the
commonalities shared by the three sons, others on their differences. The latter generally institute a hierarchy
between them, with Gatutsi as the most deserving of the three. On the myth of Gihanga, see for example,
Dominique Franche, Rwanda: généalogie d’un génocide (Paris: Éditions Les milles et une nuits, 1997) or
Mahmood Mamdani, When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda
(Princeton, N.J.: Princeton University Press, 2001), pp. 79-80.
16
Rwanda’s geography is characterised, in a good portion of the country, by its hilly nature. The country is
often referred to as the ‘Country of the thousand hills.’ Hills are an important reference point for the Rwandese,
who often speak of life ‘on the hills’ when talking about life outside of big cities.
14
141
be children of Hutu and Tutsi intermarriages over the centuries.” 17 According to David
Newbury, realistic estimates are that a minimum of 25 percent of Rwandese have at least
one great-grandparent of each group and numbers rise going further back in time. 18
European colonisation at the end of the nineteenth century, however, profoundly changed
Rwandese society.
Colonial Rwanda
The first official contact between Europeans and the Rwandese authorities took place in
1894 when the German explorer Count Von Goetzen was received at the court by King
Rwabugiri. This meeting followed an 1885 conference in Berlin where fifteen European
governments divided Africa amongst themselves. Germany obtained control of Rwanda.
With the impending German arrival in the country, Rwandese dignitaries were warned by
others from the region not to resist the foreign visitors. As a result, the Germans were
welcomed and the Rwandese helped them settle.
The Germans retained control of
Rwanda until 1916, when Belgian troops stationed in the Congo took the colony from
them.
Following the end of the First World War, Belgium was officially mandated to
administer Rwanda in 1919.
Colonisation ‘racialised’ relations among the three groups in Rwandese society.
This is believed to have been the result of a mistranslation by translators employed by the
colonisers, commonly Swahili speakers. In Kinyarwanda one word only, Ubwoko, was
used to describe categories of all forms; much like the word ‘kind’ in English can be used
to speak of types of apples (what kind of apple is it?), the make of a car (what kind of car is
it?), types of personalities (what kind of person is s/he?). Ubwoko, however, was mainly
17
18
Mahmood Mamdani, When Victims Become Killers, p. 54.
David Newbury, “Understanding Genocide,” African Studies Review, vol. 41, no. 1, 1998, p. 84.
142
used in reference to clans. 19
Upon their arrival in Rwanda, the Europeans assumed
however that Ubwoko referred to racial categories. 20
This misinterpretation would
probably not have occurred had there not been a strong current of racial theorising en
vogue in Europe at the time.
This trend, epitomised by the work of Count Arthur
Gobineau, posited the existence of different races among humans which could be
hierarchically ordered along a continuum ranging from inferior to superior. 21
This
European ‘evolutionism’ was in turn applied to groups and societies in the colonies at the
turn of the nineteenth century. In Rwanda, this conception mainly served to differentiate
the Twa, Hutu and Tutsi along racial lines; a belief colonisers tried to validate through
physical anthropology, taking measures of foreheads, noses, waists, height, etc. Judging
the Tutsi to be slender, taller and to have finer facial features, the Europeans naturally
assumed they were the superior race in the region. As Hanning Speke, one of the first
explorers of the region, argued, the Tutsi were taken as “more ‘European’ and ‘superior’ to
the Hutus and Twa.” 22
Perceived physical similarities with European features necessarily translated into
racial superiority. For many at the time, such physical superiority translated into superior
‘mental’ capabilities.
For Europeans, “the Tutsi were more intelligent, reliable,
hardworking−in short, more like themselves−than the Hutu.” 23 Apparently confirmed by the
high degree of integration and organisation of the Nyiginya kingdom, the Tutsi were also
thought to possess great leadership skills. The Hutu were instead perceived as hard
working, but clearly of lesser intelligence than the Tutsi. The Twa were inferior to the two
19
Interviewee no. 19, Kigali, Rwanda, 27 March 2004.
The European believed so strongly in a racial nature of group categories that they completely ignored other
important forms of categorisation in Rwanda, including clans. In books published during the period describing
the Rwandese population, clan affiliation is consistently ignored. Interviewee no. 12, Butare, Rwanda, 23
March 2004.
21
See, for example, Joseph Arthur Gobineau, Essai sur l’inégalité des races humaines (Paris, Éditions Pierre
Belfond, 1967 [1853-1855]).
22
Quoted in Christine L. Kellow and H. Leslie Steeves, “The Role of Radio in the Rwandan Genocide,” Journal
of Communication, vol. 48, no. 3, 1998, p. 113.
23
Peter Uvin, “Prejudice, Crisis and Genocide in Rwanda,” African Studies Review, vol. 40, no. 2, 1997, p. 95.
20
143
other groups. Twa were seen as pygmies, and as such, not quite fully human. 24 As David
Newbury summarised, “[i]n this vision, race, culture, and power were all interlocked.” 25
In order to explain racial differences among groups in Rwanda, but more
importantly the perceived superiority of the Tutsi, Europeans developed the ‘Hamitic myth.’
For many, it seemed inconceivable that a civilisation as sophisticated as the Rwandese
monarchical system could have developed on the ‘Dark Continent.’ These achievements
could only be “taken as evidence of a civilizing influence of an outsider race.” 26 The Tutsi
were therefore presumed to have originated from regions closer to Europe and migrated to
Rwanda late in the second millennium (around the 16th or 17th century). Many speculated
on the origins of the Tutsi, who were even thought by some to be Caucasians that had
migrated to Africa and been blackened by the sun. The most common myth, however,
was of the existence of a Hamitic tribe, descendants of Ham, the son of Noah, who “after
founding some civilizations and attempting to keep their blood pure, [...] had become
hopelessly mongrelized by the native and inferior blacks.” 27 As for the Twa and Hutu, the
former were identified as pygmies, originating from the Rwandese forests, and the latter as
Bantu from the Congo or Cameroon, or both simply taken as true Africans, negroes of
Bantu descent, despite all the confusion surrounding the term ‘Bantu’ which refers to a
linguistic group and not a race. 28
24
Gérard Prunier provided some of the colonialist depictions of the three groups. About the Twa, the
colonialists had to say that “[m]ember of a worn out and quickly disappearing race [...] the Mutwa presents a
number of well-defined somatic characteristics: he is small, chunky, muscular, and very hairy; particularly on
the chest. With a monkey-like flat face and a huge nose, he is quite similar to the apes whom he chases in the
forest.” About the Hutu, they indicated that they “display very typical Bantu features. [...] They are generally
short and thick-set with a big head, a jovial expression, a wide nose and enormous lips. They are extroverts
who like to laugh and lead a simple life.” And finally, the Tutsi were described in the following terms: “[t]he
Mututsi of good race has nothing of the Negro, apart from his colour. [...] Gifted with a vivacious intelligence,
the Tutsi displays a refinement of feelings which is rare among primitive people. He is a natural-born leader,
capable of extreme self-control and of calculated goodwill.” Gérard Prunier, The Rwanda Crisis: History of a
Genocide (London: Hurst and Company, 1995), p. 6.
25
David Newbury, “Precolonial Burundi and Rwanda,” p. 258.
26
Mahmood Mamdani, When Victims Become Killers, p. 79.
27
Ibid., p. 83.
28
Interviewee no. 12, Butare, Rwanda, 23 March 2004.
144
The Hamitic myth had important implications for how the groups in Rwandese
society were defined. It posited three fundamental differences between them: they had
different geographical origins; they were of different races; and they had arrived in Rwanda
at different periods–the Twa being the first inhabitants of the region, the Hutu arriving
second and the Tutsi finally migrating to Rwanda late after the first two to conquer them.
This last belief would be of utmost importance in following decades in a traditional society
where claims to land ownership are a function of the date at which a group settled on a
territory; the first inhabitants are generally assumed to be the rightful occupants. But the
Hamitic hypothesis also displaced identity conceptions that already existed. What had
previously served to foster Rwandese unity, myths of a common ancestry, like the myth of
Gihanga, an allegiance to the same king, the base of an embryonic national
consciousness, were all supplanted by this new European interpretation.
In order to institutionalise this racial ideological framework, the Belgians conducted
a census in the early 1930s to determine the background of individuals in Rwanda. The
census allowed the colonial administration to introduce identity cards which mentioned an
individual’s race. Many Rwandese claim that, since it was difficult to distinguish among
the three groups, the Belgians arbitrarily assigned the cards according to the number of
cows owned by an individual. Someone owning ten cows or more was identified as Tutsi. 29
Less fortunate Rwandese were categorised as Hutu or Twa, depending on their
appearance.
According to Mahmood Mamdani, this belief probably oversimplifies the
system used by the colonial administration to categorise the Rwandese. He believed they
likely based themselves on three criteria: “oral information, provided mainly by the Church,
physical measurements, and the ownership of large herds of cows.” 30 Most Rwandese at
the time failed to see the danger in this exercise and agreed to register with the colonial
29
30
Human Rights Watch, Leave None to Tell the Story, p. 37.
Mahmood Mamdani, When Victims Become Killers, p. 99.
145
authorities. These administrative changes proved, however, to have a long lasting and
disastrous impact. What had previously been fluid identities became fixed ones, legally
inscribed on official documents. Switching between categories, in the form of the
“Kwihutura (the social rise of a Hutu to the status of a Tutsi) or Gucupira (the social fall of a
Tutsi to the status of Hutu)” which had long been Rwandese social customs, was no longer
possible. 31
This division of Rwandese society along racial lines not only helped to differentiate
among individuals, but it also served to structure the entire Belgian administrative system.
As in many African countries, colonial administrators developed in Rwanda a system
resembling indirect administration where the management of the lower echelons of society
was left to native authorities.
There was, however, one difference with the system
developed in Rwanda. Instead of allowing community leaders administer their constituents,
the 1926 Voisin Reform gave most administrative positions to the ‘superior’ group, the
Tutsi, replacing all other native chiefs by Tutsi chiefs. 32 The Tutsi, especially the grands
Tutsi, therefore became the native auxiliaries of the colonial administration, and thus
needed to be taught to read and write to fulfill their tasks. This decision was supported by
the Catholic Church, which was generally responsible for schools, and by a strong
supporter of the idea of Tutsi superiority, Monseigneur Léon-Paul Classe. 33
31
Classe
Ibid., p. 101.
Until then the country had been administered through a system resting on the authority of three chiefs: the
chief of lands, generally a Hutu, in charge of administering the land to be cultivated and of collecting taxes for
the exploitation of agricultural tenures, a chief of grass, generally Tutsi, administering pastures, and a chief of
armies, also generally Tutsi, managing the cattle belonging to the army and militias. The Voisin Reform
abolished this traditional system and replaced it with a system based on the authority of a single chief in
charge of supervising lower administrators. Most chiefs in the new system were Tutsi. Jean-Paul Kimonyo, La
participation populaire au Rwanda, p. 62.
33
The Catholic Church played a very important role in Rwanda. As the Belgian colonial administration judged
the country to have fewer natural resources than the Congo, they concentrated their efforts in the latter, but left
the Church to handle many aspects of colonial development in Rwanda. Timothy Longman explained that
“[d]uring the colonial period, the Catholic leadership developed a cooperative working relationship with both
colonial administrators and the indigenous elites in [Rwanda, Burundi and Congo]. The Belgian administrators
regarded the Catholic missionaries, many of whom were themselves Belgian, as allies in the struggle to
establish and maintain control over the local populations.” Timothy P. Longman, “Empowering the Weak and
Protecting the Powerful: The Contradictory Nature of Christian Churches in Rwanda, Burundi, and the
Democratic Republic of Congo,” African Studies Review, vol. 41, no. 1, 1998, pp. 54-55. See also Alison Des
32
146
believed that doing otherwise was against the natural order and would have grave
consequences. 34
The Church also benefited from this strategy. The Vatican had
instructed local Church representatives to educate African elites as part of an effort to
convert African populations. 35
The colonisers, therefore, set out to develop a schooling system for natives in
Rwanda mostly managed by the Catholic Church. It was a dual system where future
auxiliaries, mainly the sons of the grands Tutsi, were trained in French to become future
administrators and, at a later point, seminaries were established so that a few Hutu could
be educated in Swahili. The French system gave future administrators a strong education
that permitted graduates to become a native elite. Those trained in this system were the
Astridiens, after the city of Astrida that hosted the largest institution of this system. 36 The
educational system for the Hutu was a second rate system giving them an education
intended to train them to teach, join the priesthood, or hold lower posts. It served to
“continue to evangelize the Hutu people, while preparing them for a lesser station in life.” 37
Because they had been mostly educated through these seminaries, the Hutu intellectuals
of the time came to be known as the Séminaristes.
The impacts of this dual schooling system were dreadful. Due to its exclusionary
practices, it ensured that upward mobility was limited to the grands Tutsi, crystallising
divisions within society and further reproducing class distinctions among the groups in
Rwandese society.
It also played a fundamental role in disseminating racist ideas.
Rwandese myths and traditions were forbidden in these schools, it was strictly forbidden to
teach ‘old Rwandese superstitions.’ As a matter of fact, even the history of the Ancient
Forges, “Kings without Crowns: The White Fathers in Ruanda,” in Daniel F. McCall, Norman R. Bennett and
Jeffrey Butler, eds., Eastern African History (New York: Praeger, 1969).
34
Interviewee no. 20, Kigali, Rwanda, 29 March 2004.
35
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 60.
36
The then city of Astrida has been renamed Butare.
37
African Rights, Rwanda, p. 10.
147
Rwandese kingdom was forbidden; European history was taught instead. 38 What was
being taught about Rwanda was the Hamitic myth. As Mahmood Mamdani argued, “the
end product [of this education curriculum] was to construct a far more comprehensive
ideology of the Tutsi as a race, the Hamites, both civilizing and alien.” 39 Over decades,
this racial interpretation of the origin of the groups was accepted by educated Rwandese. 40
The colonial period thus further exacerbated divisions that had begun developing in
the Rwandese society around the mid-eighteen-hundreds. The Tutsi, far from being the
passive actor many have depicted them to be, were keenly aware of the advantages
offered them by colonial favouritism. Many identified with the Hamitic myth. They saw it as
a justification for what they took as their natural right to administer the Hutu and Twa and
to impose on them labour and financial burdens, especially since their authority had been
legally extended throughout the country. 41 “Tutsi notables were quick to take advantage of
the authority given to them by the colonialists, claiming as ‘traditional’ many onerous duties
from the Hutu that were in fact not traditional at all.”
42
But beyond simply using these
myths to ground their authority, many also internalised them. 43 As Jean-Paul Kimonyo
indicated: “they had started to believe in the natural character of their indigenous power
[…] [and continued] to reproduce the authoritarianism of the colonial system.” 44 They thus
collaborated fully with the colonial policy of racial discrimination.
38
Interviewee no. 2, Kigali, Rwanda, 9 March 2004.
Mahmood Mamdani, When Victims Become Killers, p. 89.
40
According to Alison Des Forges, these beliefs “were accepted by intellectuals in the circles of the court, even
those without European-style schooling–and integrated into their oral histories. [...] But even the majority of
Hutu swallowed this distorted account of the past. [...] Thus people of both groups learned to think of the Tutsi
as winners and the Hutu as losers in every great contest of the Rwandan past.” Alison Des Forges, “The
Ideology of Genocide,” Issue: A Journal of Opinion, vol. 23, no. 2, 1995, p. 45.
41
For more on this, see for example Catharine Newbury, The Cohesion of Oppression, Clientship and Ethnicity
in Rwanda 1860-1960.
42
African Rights, Rwanda, p. 6.
43
As Timothy Longman argued, “[t]he Tutsi [...] participated in the development of a mythico-history that
portrayed them as natural rulers, with superior intelligence and morals. [...] This history became accepted by
Rwandans of all ethnicities.” Timothy Longman, “Christian Churches and Genocide in Rwanda,” paper
presented at the Conference on Genocide, Religion, and Modernity, United States Holocaust Memorial
Museum, Washington D.C., United States, 11-13 May 1997, p. 5.
44
Personal translation. Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 84.
39
148
As for the rest of the population, it felt and expressed an increasing resentment
towards the system. Popular resentment was rooted in a growing sense of injustice
regarding the hardships imposed on the general population through taxation and corvées.
The colonial period proved especially harsh for the lower echelons of society, now officially
Hutu and Twa.
As Filip Reyntjens argued, “the burdens on the Hutu masses were
undeniably worse under the Belgian colonial administration than at any other time in the
past.” 45 For many Rwandese, the dual burdens of both the colonial and native
administrators, particularly in terms of labour, proved so overwhelming that their only
option was to emigrate to find work, free of imposed obligations, in neighbouring
countries. 46 Towards the end of the colonial era, limited access to higher education, to
employment, as well as the clear discrimination against certain groups in society, became
the focus of this resentment, channeled mainly by the Hutu intellectuals, the Séminaristes.
According to Kimonyo,
[t]his strata of seminarians, teachers, employees and small businessmen, known as
the Hutu counter-elite, was the subject of deep grievances because of barriers to
their aspirations of social ascension, barriers resulting of the almost complete Tutsi
monopoly in terms of middle to superior positions available to the indigenous
population. 47
Having experienced the limits, if not outright exclusion, imposed on them by the system
despite their education, they chose to channel and voice growing popular resentment.
Due to the practice of indirect administration, this resentment was mostly directed
at the Tutsi. As the auxiliaries of the European colonisers, they were the visible
oppressors.
The Hutu counter-elite did correct this perception, anti-Tutsi anger being
easier to channel than the population’s diffuse awareness of the colonisers’ role in their
45
Quoted in Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 64.
According to Human Rights Watch, thousands of Rwandese emigrated in the early 1920s. Human Rights
Watch, Leave None to Tell the Story, p. 35. By the end of the 1920s, the situation had become so difficult in
terms of imposed labour that individuals had to perform for the authorities that, according to Jean-Paul
Kimonyo, approximately 50 000 Rwandese (one male adult out of every six) seasonally migrated to British
colonies, Uganda, Tanganyika and Kenya, to find unbound and paid work. See Jean-Paul Kimonyo, La
participation populaire au Rwanda, pp. 64-65.
47
Personal translation. Ibid., p. 69.
46
149
oppression. It was for them a simple explanation to frame popular grievances, one that
resonated with the masses. They chose to target the most visible source of hardships
endured by the population: the Tutsi aristocracy’s role in ordering the corvées, the forced
labour, but, importantly, also as the hand that held the whip. Targeting the Tutsi was also
a strategic choice on the part of the Hutu counter-elite.
By the 1950s, the colonial
administration had begun supporting the Hutu in their emancipation struggle. To continue
to benefit from Belgian support, it was therefore important to draw attention away from the
colonisers’ role in imposing such hardships on the masses.
This shift in colonial allegiance was the result of two factors. Firstly, changes had
taken place within the colonial administration, especially within the clergy.
A new
generation had replaced the first priests in Rwanda, known for their conservatism. These
new priests “were likely to come from le petit clergé [and be of] ‘relatively humble
origins’.” 48 Combined with the development of a new social sensitivity in certain
progressive Catholic circles in Europe, the newcomers tended to be more sensitive to the
plea of the masses, mainly Hutu, and to the class-based injustices they endured. 49
Secondly, towards the 1950s, the political climate had begun to change drastically in
Africa, as many African populations were demanding their independence from European
colonisers.
Sensing this change, the Tutsi auxiliaries also began considering
independence as an option.
The Hutu counter-elite, on the other hand, publicly took
position against independence, claiming that the country was not ready. Seen as the more
docile of the two groups, the Europeans judged the Hutu to be better allies in Rwanda and
thus decided to back them in their struggle against ‘Tutsi oppression.’
From that point on, the political situation developed at a quick pace. Following
repeated visits to Rwanda, United Nations representatives published a number of reports
48
Mahmood Mamdani, When Victims Become Killers, p. 113.
On this new group within the clergy, see for example Jean-Paul Kimonyo, La participation populaire au
Rwanda, p. 70.
49
150
decrying the lack of political development in the country. 50 In response to these reports,
the Belgian authorities opted for a slow democratisation of the political system. They
instituted elected councils at the different administrative levels, headed by a Conseil
supérieur du pays (CSP) under the king’s authority. With this newly developed political
arena, the Hutu elites’ resentment now had a medium through which it could be
expressed.
The Hutu focused on two issues: the question of resolving the injustices
suffered by the masses and ensuring greater representation of the Hutu in the system.
A turning point was the publication on 22 February 1957 of a document entitled
Mise au point by the CSP, still governed mainly at the time by the king and Tutsi
dignitaries. 51
This document, addressed to another UN mission visiting the country,
presented demands for a quick independence while minimising the social problems in
Rwanda, particularly the plight of the Hutu.
The publication of the Mise of point
commanded a strong reaction on the part of the Hutu counter-elite. Targeting the Tutsi,
this response took on racist undertones, inspired heavily by colonial ideology, but under
social cover.
It came principally in the form of a document published by nine Hutu
intellectuals, all ex-seminarians, entitled Notes on the Social Aspect of the Racial Native
Problem in Rwanda. 52 Better known as Le Manifeste des Bahutu, it claimed that “[t]he
problem is above all a problem of political monopoly which is held by one race, the Tutsi;
political monopoly which, given the totality of current structures becomes an economic and
social monopoly which, given the de facto discrimination in education, ends up being a
cultural monopoly.” 53 A few months later, the Hutu opposition further took form when the
50
Ibid., p. 71.
The composition of the CSP was supposed to reflect the councils at lower echelons of society. In the first
communal elections in 1953, despite the demographic weight of the Hutu, most seats were won by Tutsi. In
the following elections, however, in 1956, the Hutu won 66 percent of seats in communal councils. The CSP
remained mainly Tutsi nonetheless. See Emmanuel Nkunzumwami, La tragédie rwandaise: historique et
perspectives (Paris: Éditions L’Harmattan, 1996), pp. 43-44.
52
Gérard Prunier, The Rwanda Crisis, 1995), p. 45.
53
Quoted in Mahmood Mamdani, When Victims Become Killers, p. 116.
51
151
Movement social Muhutu was founded by Grégoire Kayibanda and in November 1957 the
Association pour la promotion sociale de la masse (APROSOMA).
The following months and years saw a flurry of reactions to the political situation
developing in the country. One of the strongest responses came in the form of public
letters written in May 1958 by fourteen Tutsi dignitaries. Drawing on the Hamitic myth,
they rejected the Hutu’s demands for greater participation in political affairs. As Mamdani
argued: “they evoked the tradition of conquest: equal rights were out of the question
‘because our kings conquered the land of the Hutu, killed their ‘little’ kings and thus
subjugated the Hutu: how can they now pretend to be our brothers?’.” 54 With tensions
between groups mounting, the situation in the country was quickly deteriorating.
To resolve the growing crisis, the Belgian administration and Rwandese monarchy
accepted demands for further democratisation.
authorising the creation of political parties.
In May 1959 they signed an order
By September 1959, three main political
parties had already been formed: the APROSOMA newly recognised as a political party,
mainly Hutu but with a social rather than racial platform; the Union nationale rwandaise
(UNAR), a nationalist and monarchist party comprised of some Hutu members but
predominantly Tutsi; and the Rassemblement démocratique rwandais (RADER), a
reformist party in favour of keeping ties with Belgium. 55 The RADER appealed mainly to
urban intellectuals and enjoyed little popular support. In the last days of September, a new
party was to join the ranks: the MDR-Parmehutu, founded by Grégoire Kayibanda.
Promoting a clearly racial platform, it claimed to represent: “the Hutu, the one that has
been outraged, humiliated and despised by the Tutsi invader […]. [w]e are here to give
back the country to its rightful owners, it is the country of the Bahutu.” 56
Despite
appearances, party competition did not uniquely fall along Hutu/Tutsi lines. Though it
54
Mahmood Mamdani, When Victims Become Killers, p. 118.
Ibid., p. 120.
56
Personal translation. Quoted in Jean-Paul Kimonyo, La participation populaire au Rwanda, pp. 87-88.
55
152
tended to be at the heart of party composition, platforms and aims were nonetheless
varied. Some parties, wanting to distance themselves from divisionism, adopted a
reconciliatory discourse.
Instead of defusing tensions, though, the institution of political parties worsened the
situation.
By November, partisans were taking to the streets, stoked by UNAR Tutsi
radicals, on the one hand, and Parmehutu Hutu radicals, on the other. 57
Tensions
exploded early in the month when a prominent Parmehutu leader, Dominique
Mbonyumutwa, was beaten by UNAR Tutsi partisans. A few days later, a Tutsi notable
was killed by protesters.
These incidents were the start of nationwide revolts where
countless acts of violence such as looting, pillaging and the burning of houses were
committed, targeting mainly Tutsi chiefs and dignitaries.
somewhat anarchic, however.
These popular revolts were
Many individuals taking part in them simply did not
understand the rationale behind the violence but acted at the request of their leaders or
out of economic opportunism. 58 Some even believed they were acting on orders from the
Mwami. As an anecdote from the times indicates, “so firmly convinced were some of
official sanction that they stopped at the [king’s] Residence to ask for petrol.” 59
To put an end to the violence and gain control of the country, dignitaries from the
monarchist circles around the king mobilised elements of the royal army. As the military
managed to control the mobs, the Belgian administration declared a national state of
emergency. Bringing in a Belgian military force, they took control of the situation and
stopped the monarchical repression. These actions on the part of Belgium sent a clear
message to the Rwandese population: the Belgian administration now indisputably
supported the Hutu.
57
Emmanuel Nkunzumwami, La tragédie rwandaise, p. 50.
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 89.
59
Quoted in African Rights, Rwanda, p. 11.
58
153
Belgian interference in favour of the Hutu movement remained a feature in the
events that followed. In the campaign for the communal elections held in June 1960, the
Belgian administrator, Colonel Guy Logiest, openly took position in favour of the
Parmehutu. Following UNAR’s decision to boycott the elections, the Parmehutu managed
a sweeping victory, winning seventy percent of the vote. Sensing it had popular support,
the Hutu movement quickly demanded that national legislative elections be held. This
demand was, however, opposed by the United Nations, intent on stabilising the country
before taking further steps.
Wishing to go ahead despite the UN’s opposition, Hutu
communal leaders, complicit with Logiest, assembled in the city of Gitarama on 28
January 1961. 60 There, they formally proclaimed the abolition of the monarchy, the
institution of a republic, and the formation a government. 61 A referendum was organised in
the Fall and validated this proclamation. The institution of a new republican regime in
Kigali quickly led to Rwanda’s full independence from Belgium in July 1962.
The political crisis that shook the country between 1959 and 1962 came to be
known as the 1959 Hutu Social Revolution and proved to be a victory on the part of the
more extreme elements in the Hutu counter-elite movement. 62 These events had lasting
effects on the country over the following decades, particularly for the Tutsi population. In a
prescient statement following the 1961 events in Gitarama, the United Nations
Commission for Ruanda-Urundi declared:
A racial dictatorship of one party has been set up in Rwanda, and the developments
of the last eighteen months have consisted in the transitions from one type of
oppressive regime to another. Extremism is rewarded and there is a danger that the
[Tutsi] minority may find itself defenseless in the face of abuse. 63
60
On Logiest participation in this, see Philip Gourevitch, We Wish to Inform You, p. 60.
These events will come to be known as the Coup d’État de Gitarama.
62
At the onset of the crisis, it was referred to as muyaga, or hurricane, instead of revolution. Personal
translation. Paul Rutayisire, “Les mécanismes de l’exclusion des Tutsi,” Revue africaine des livres, vol. 1, no.
2, 2005, p. 20.
63
Quoted in Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 94.
61
154
Also sensing the turn of events, tens of thousands of Tutsi began fleeing the country as
early as 1959. 64
First Republic
The 1959 Social Revolution instituted the first Rwandese Republic under the Parmehutu
and, starting in 1961, had Grégoire Kayibanda as President. The regime was openly antiTutsi and worked to develop an ideological structure promoting the demonisation of the
Tutsi and an extreme differentiation among the groups of Rwandese society, a national
ethno-centrism. 65 The ideology upheld by the regime was not original, however. On the
contrary, it found its inspiration in old racial theories en vogue during the colonial period. 66
The Parmehutu leaders reinterpreted colonial racism, turned it on its head, and forged an
ideological structure aimed at scapegoating the Tutsi. What was novel about the regime’s
approach, though, is that it systematically indoctrinated the population.
While racial
ideology had previously been confined to the educated, the new regime promoted its
ideology among Rwandese society from top to bottom, seeking to reach all the way down
to rural communities.
The rhetoric developed around this ideological structure was unambiguous. It was
articulated around the idea that the Parmehutu, but by extension all Hutu, had achieved
victory over the oppressors–not the Belgians, but the first colonisers, the Tutsi. Echoing
the Hamitic hypothesis, this view was the inspiration for depictions of Rwandese groups.
The Tutsi were described by the regime as foreigners, as strangers to the land only
recently arrived, not true Rwandese. These depictions were also built around references
to the Tutsi monarchy. The Tutsi, as descendants of oppressive monarchs, were described
as feudalists that gladly enslaved the Hutu for centuries to ground their power. During their
64
It should be noted that some Hutu also fled the country, mainly royalists or individuals with strong ties to their
former Tutsi patrons.
65
Interviewee no. 19, Kigali, Rwanda, 27 March 2004. The slip towards referring to ethnic groups instead of
races occurred between the First and Second Republics, somewhere between 1959 and 1973.
66
African Rights, Rwanda, p. 12,
155
reign, they had been abusive, ruthless oppressors, traits inherent to their nature.
And
while the reality often proved otherwise, the Hutu elites maintained that the Tutsi were
wealthy, enjoying a wealth obtained when they monopolised the resources of the country
and sought all power for themselves. 67
Overall, as a group, the Tutsi were scapegoated in the regime’s rhetoric. Images
were used to present them in a negative light. For example, they were called ‘snakes,’ a
reference to the Bible’s portrayal of this animal as a source of evil.
Paradoxically,
however, despite all the evils attributed to the Tutsi, the ideology developed did not
invalidate some of the ‘superior’ characteristics the colonisers had attributed to them. The
Parmehutu rhetoric continued to describe the Tutsi as extremely intelligent, witty and
cunning. Such traits only served to reinforce their scapegoating of the Tutsi: they were so
smart, but also so self-interested, that they were not to be trusted. A saying claimed that
[y]ou give shelter to a Tutsi in your living room, he chases you out of your bedroom.” 68
Despite all the scapegoating, the Tutsi were nonetheless tolerated by the Kayibanda
regime, but tolerated as would be foreigners, as second rate citizens and not as true
Rwandese.
True citizenship belonged to the Hutu, the anti-thesis of the Tutsi. They possessed
all positive attributes in the Manichean communal identities developed by the regime. With
most depictions drawing their inspiration from the myth of the 1959 Social Revolution, the
Hutu people were described as the agents of democracy, a successful popular force that
overthrew the oppressive Tutsi monarchy. In a speech given at the inauguration of the
legislative assembly, Kayibanda claimed that: “[t]his date, October 26th, will go down in the
History of the Rwandese people. Today, a representative Parliament and a representative
67
Interviewee no. 19, Kigali, Rwanda, 27 March 2004.
Jean-Marie Vianney Higiro, “Rwandan Private Print Media on the Eve of the Genocide,” in Allan Thompson,
ed., The Media and the Rwanda Genocide (London: Pluto Press/IDRC, 2007), p. 74.
68
156
government have been instituted. Democracy has defeated feudality.” 69 But the rhetoric
built around this conception of the Hutu as true democrats also drew on the idea of
Rubanda Nyamwinshi, le peuple majoritaire or national majority. Since the Hutu were the
most numerous group in Rwanda, the regime used this fact to justify their control of
political and administrative institutions. A system dominated by the Hutu could only be
democratic since it inherently represented the majority. 70
On this idea of Rubanda
Nyamwinshi was also grafted the idea of the Hutu as the ‘natural’ inhabitants of the
country, who arrived centuries before the Tutsi to develop it for agriculture. 71 They were
depicted as the children of the land that struggled to clear it for cultivation, succeeded in
reclaiming it, and therefore had the right to call it their own. As such, they were presented
as the true and legitimate citizens and representatives of Rwanda.
The regime went to great lengths to promote these depictions of the groups and
this extreme differentiation between them. In their efforts, they moved beyond employing
political rhetoric as the principal vehicle for their ideology. As Kimonyo argued:
[u]p until independence, the racist anti-Tutsi ideology that the Parmehutu sought to
popularise had been propagated mostly by party officials, leaders and
propagandists. With independence, the Parmehutu widened the array of institutions
used for its propaganda enterprise to include state structures, newspapers, the radio
and schools. 72
State structures and the media were the perfect medium to publicise important speeches
or to disseminate ideas at the heart of the party platform. As for schools, they replaced
European history imposed by the colonial administration with Rwandese history. What
was promoted as Rwandese history, however, was mainly the myth of the different origins
69
Speech by Grégoire Kayibanda, 26 October 1960, National Archives, Kigali, Rwanda.
As Gérard Prunier explained, Rubanda Nyamwinshi is a “coded political expression [...] with the connotation
that one must be Hutu to be allowed to rule and that whoever rules in the name of the ‘majority people’ is
ontologically democratic.” Gérard Prunier, The Rwanda Crisis, pp. 369-370.
71
Interestingly, according to the successive migration thesis, the Twa were the first inhabitants of the region.
Presenting the Hutu as the natural inhabitants of Rwanda contradicted this conception even though it was
based on it. This paradox seems to have never been addressed.
72
Personal translation. Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 98.
70
157
and different arrival dates of Rwandese groups. 73 Apart from these more official channels,
the regime also adopted more ‘populist’ strategies to spread its ideology. One of these
was to use songs infused with ideological rhetoric written by musicians hired by the
Parmehutu. These were then played on the radio to large numbers of people.
One
Rwandese recalled that these were songs of victory. They retold how the Hutu had
recovered their rights and could now rule the country–their country–forever. 74 Some of
these became instant hits.
Violence too became a means to ingrain this extreme differentiation between
groups. This violence, recurrently sweeping the country, condoned by the regime and
stoked by community leaders, targeted the Tutsi. The first years of the Kayibanda regime
in particular were the source of many of these repeated outbursts of anti-Tutsi violence.
Close to ten important ones took place between March 1961 and November 1966. 75 And,
though the violence generally began as attacks on Tutsi dignitaries, it tended to quickly
turn into reprisals against the general Tutsi population. Their houses and possessions
were pillaged or burnt, their cattle killed, and many Tutsi killed.
These outbursts were not random, however. They followed a pattern. Most came
in reaction to attempts by the Inyenzi, Tutsi exiled in previous massacres and nicknamed
cockroaches, to return to the country.
Among this group were some Tutsi notables,
monarchists and opponents of the Kayibanda regime that sought to return by force. While
it was this group in particular that the regime in Kigali feared, it was nevertheless the Tutsi
living inside Rwanda that suffered the brunt of these reprisals. Almost every time there
was an attack by forces outside the country, there was violence within Rwanda. As many
have described, the Tutsi inside the country were held hostage, as potential subversive
elements or accomplices of this external enemy, but also as leverage to use with the
73
Interviewee no. 10, Kigali, Rwanda, 19 March 2004.
Interviewee no. 20, Kigali, Rwanda, 29 March 2004.
75
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 96.
74
158
invading forces. In a public address to Rwandese immigrants and refugees, President
Kayibanda declared that the Tutsi in Rwanda would suffer if actions by Tutsi in
neighbouring countries continued:
The Tutsi that remained in the country are afraid of the popular rage triggered by
your incursions. [...] If we imagine the impossible, that you manage to take Kigali,
how would you measure the resulting chaos of which you would be the first victims?
You say it amongst yourselves: ‘it would be the complete and precipitated end of the
Tutsi as a race’. 76
The regime, therefore, laid the blame for the violence on the exiled Tutsi trying to return. 77
The massacres were justified as necessary. 78 In light of the unwarranted attacks from
external enemies, the country had to defend itself. It even institutionalised this strategy by
creating within administrative regions in the country, the communes, what it called
‘commune self-defence committees’ through which ordinary citizens were invited to take
part in actions against the Tutsi.
This sanctioned violence against the Rwandese Tutsi population highlighted the
double standard the Kayibanda regime sought to institute: the Tutsi could be treated
differently due to their status as second rate citizens. It was further reinforced by the fact
that the violence taken against the Tutsi remained unpunished. Committing violence
against the Tutsi with impunity confirmed their lower status, a notion that slowly took hold
of people’s minds. In an interview, a Rwandese woman explained that this even began to
translate into popular sayings. One of these stated: “to kill a Tutsi is not a sin.” 79
Despite the fact that the regime used a discourse of race or ethnic affiliation, Hutu
and Tutsi identities took on a political form during the First Republic.
The anti-Tutsi
ideology of the regime, as well as the labels Hutu and Tutsi themselves, served a political
purpose, grounding the legitimacy of a regime that had started on shaky grounds, by
76
Personal translation. Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 98.
An important difference between the Kayibanda era and what happened in Rwanda during the 1990-1994
period is that it tolerated the Tutsi living in Rwanda, but clearly held the Tutsi outside the country as enemies.
Between 1990 and 1994, this distinction finally made way to an undiscerning labeling of all Tutsi as enemies.
78
Interviewee no. 24, Kigali, Rwanda, 2 April 2004.
79
Interviewee no. 22, Kigali, Rwanda, 1 April 2004.
77
159
winning the population over with a very thin political platform, extreme even heinous
opposition to the Tutsi administrators. The Kayibanda regime was in a sense trapped to
reproduce the platform it had developed during the Revolution in order to retain its
legitimacy. But the scapegoating of the Tutsi promoted by the regime also served as a
perfect justification to replace Tutsi chiefs or political opponents that remained in the
country with Hutu supporters. Once in place, these leaders could further reproduce party
ideology to indoctrinate the masses but could also take control of the property, especially
the land, of their predecessors. This land, in a country where land scarcity was a constant
problem, could then be redistributed to create their own clientage structure. 80 The political
nature of Hutu and Tutsi identities was, however, especially clear in times of political
turmoil, when the regime began to ardently propagate antagonistic visions of intergroup
relations, at the core of popular support for the Parmehutu, to renew with their popular
bases. Thus, although communal tensions remained a constant during the first years of
the Kayibanda presidency, their intensity varied following the ups and downs of the
political situation. The more difficult the political conditions were for the regime, the more
the Tutsi were persecuted.
The regime’s legitimacy was increasingly questioned, however. During the late
1960s, in particular, the Kayibanda regime faced growing dissent. It had already been
struggling with economic problems, inherited from the colonial period. Rwanda lacked
basic equipment to develop, national production was meager, and the new Hutu elite often
were missing the administrative know-how to manage national institutions. 81 Rwanda was
economically stagnating. This situation led to mounting discontent, especially on the part
of intellectuals, disappointed not to see the better living conditions promised by the 1959
Social Revolution.
80
81
Interview with Alison Des Forges, Human Rights Watch, Kigali, Rwanda, 4 March 2004.
Jean-Paul Kimonyo, La participation populaire au Rwanda, pp. 100-101.
160
The regime also faced discontent from within the party. It came mainly in reaction
to a trend it had begun to develop, the regionalisation of its power base. Kayibanda had
started to narrow his privileged support base around a specific region, the Gitarama region
in the south of the country. Not only was this a source of frustration for many, but it also
constituted a dangerous strategy. Elites from the Northern part of the country, long
marginalised, were growing discontented. Furthermore, the Kayibanda regime was also
isolated on the international scene. It had very few allies; its sole supporters were a
limited number of international religious movements. Claudine Vidal paints a very bleak
picture of the situation in Rwanda by the early 70s:
Slowly the country turned into an island. The government feared its whole
environment. [...] The inhabitants were inward-looking and bore the country’s slow
shrinkage in silence. There were several forms of censorship: from a triumphant
Catholic church and from the government which was afraid both of possible
communist inspired social movements and of the traditional manifestations which
could be a reminder of the Tutsi imprint which it considered with something like
phobia. To the generalised lack of trust, rumour, secrecy, lack of breathing space: on
top of material deprivation–the country was one of the poorest in the world and
lacked almost everything–was added something like mental paralysis. 82
The political situation was becoming untenable for the Kayibanda regime.
A pretext to reignite communal tensions in the hope of regaining popular support
arose following events in 1972 in neighbouring Burundi. 83 The Tutsi dominated regime of
Burundese President Michel Micombero and Tutsi extremists within the army organised a
massacre of the Hutu population. 84
According to many, the regime in Kigali took
advantage of these events to overshadow the mounting political problems in Rwanda by
resorting once more to its tactic of instigating anti-Tutsi violence. Beyond the massacres
in Burundi, some of the party leaders also justified the need to take action against the
82
Claudine Vidal, quoted in Gérard Prunier, The Rwanda Crisis, pp. 59-60.
Burundi and Rwanda are mirror images of each other. Problems in one of these countries have often times
tended to spill over in the other. As Gérard Prunier argued: “Rwanda and Burundi have been, since
independence, the two opposite ends of a political seesaw. Their parallel–and at times common–past
histories, their comparable social structures, their constant and almost obsessive mutual scrutiny, fated them to
be natural mirrors of each other’s hopes, woes and transformations.” Gérard Prunier, The Rwanda Crisis, p.
198.
84
Emmanuel Nkunzumwami, La tragédie rwandaise, p. 72.
83
161
Rwandese Tutsi by claiming that they remained overly predominant in important sectors of
society, despite their effective growing marginalisation, a winning strategy in times of
economic turmoil. Groups were therefore organised to chase out the Tutsi from high
schools and universities, from the public sector, as well as from private businesses. This
last extreme outburst of anti-Tutsi violence on the part of the Kayibanda regime had a
clear target: the remaining Tutsi intellectuals in Rwanda. 85
Second Republic
Following this renewed violence and to restore order in the country, Major-General Juvénal
Habyarimana, the military chief of staff, with the help of Alexis Kanyarengwe, the director
of National Security, and nine other military officers, most originating from the North,
staged a coup d’État. 86 They overthrew the Kayibanda regime on 5 July 1973, quickly
dissolving the Assembly, dismantling the Parmehutu, as well as arresting and torturing
many of the political leaders from the South. Despite their imposition of a military regime,
in early August the perpetrators of the coup proclaimed their intention to follow democratic
and peaceful values. They thus created a new Comité pour la paix et l’unité nationale, the
Committee for national peace and unity, and tasked it to head the country.
Not surprisingly, after the brutality and abuses of the First Republic, it was almost
with public relief and support that Habyarimana took power and instituted this Second
Republic. The new president, with his conciliatory and pacifying speeches, was seen by
many as a liberator. As one Rwandese explained: “[w]e really danced in the streets when
Habyarimana took power.” 87 Another recalled that: “[d]uring the first years, I adored
85
The events of 1972 involved mainly the urban and educated strata of Rwandese society. As Prunier
explained, “[t]he most eager to carry out this ‘purification’ through the vigilante committees were educated
people who could expect to benefit from kicking the Tutsi out of their jobs. In the hills the peasants showed no
interest.” Gérard Prunier, The Rwanda Crisis, pp. 60-61.
86
Emmanuel Nkunzumwami, La tragédie rwandaise, p. 73.
87
Philip Gourevitch, We Wish to Inform You, p. 69.
162
Habyarimana. I kept on saying he was the man Rwanda needed.” 88 This point of view
was shared by many, even outside Rwanda.
To this day, the first years of the
Habyarimana regime are described as days of greater tolerance, even as peaceful times.
This impression was, however, more the reflection of Juvénal Habyarimana’s talent
as a politician and public speaker than of radical changes in governance. The second
Rwandese President was extremely astute. He knew how to sway his public, what
language needed to be used to get support. For the Rwandese population, he spoke of
the need to overcome the divisions that had afflicted the country in the previous decades.
As he stated during a tour of the country: “[e]verything that divides, whether ethnic
affiliation or clan, should be avoided. The pride felt for belonging to an ethnic group is
legitimate, as long as it does not hinder the needed unity and peaceful cohabitation.” 89 For
the inhabitants of Rwanda, he publicly wished for better relations among groups, for
national reconciliation, and for peace in the country after decades of conflict. President
Habyarimana went as far as to state that “during four centuries, the country knew division
and conflict which the 1959 Revolution put an end to when it gave power to the Hutu
majority. However, feelings of resentment did not dissipate, and up until 1973, the year in
which was instituted a new regime, the Second Republic, the Tutsi lived under threat.” 90
He, on the other hand, claimed to want a unified Rwandese nation.
Consequently, the anti-Tutsi divisive ideology was displaced to a certain extent by
a new conciliatory rhetoric concerning communal identities, particularly towards the Tutsi.
References to the Tutsi as foreigners fell out of favour. The regime preferred to refer to
them as a national ethnic minority. 91 This new terminology implied a rediscovery of the
Tutsi’s indigenousness. As Mamdani argued, the regime “began a discussion of the Tutsi
as an indigenous ethnic group as opposed to a nonindigenous race, and of Tutsi rights as
88
Personal translation. Interviewee no. 29, Kigali, Rwanda, 15 April 2004.
Personal translation. Quoted in Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 122.
90
National Archives, Kigali, Rwanda.
91
Interviewee no. 19, Kigali, Rwanda, 27 March 2004.
89
163
minority rights.” 92 Yet, however, under the Habyarimana regime, minority rights proved not
to be a means to ensure the protection of a demographic minority, but a reflection instead
of the Tutsi’s status as latecomers in Rwanda and therefore as deserving fewer rights.
Besides pacification and reunification, the other major objective of Habyarimana’s
regime was development.
Following a similar trend across the African continent,
development and progress became buzzwords for the President’s public appearances.
Development was such a predominant goal that Habyarimana, the Président-Fondateur,
chose to name the state party the Mouvement révolutionnaire national pour le
développement (MRND). 93 He also renamed the National Assembly Conseil national de
développement (CND) in 1981. In the mid-1970s, this focus on development appeased
the population who finally saw some signs of a long promised economic prosperity:
economic stability, better terms of trade, higher standards of living, etc. The development
rhetoric also pleased the international community. The Habyarimana regime even became
a favourite of international development assistance, with his stance on the country’s need
for change and progress proving popular among foreign donors. Rwanda received among
the largest amounts of aid per capita on the continent and had one of the largest
contingents of international aid workers on its territory. 94 It even came to be held as a
model of development. With this foreign support, Habyarimana was, in turn, able to further
legitimate his presidency. With so many international benefactors, no one could question
that the regime was a proficient one.
92
Mahmood Mamdani, When Victims Become Killers, pp. 16-17.
Habyarimana adopted this title after founding his political movement in 1975. He inspired himself from
General Mobutu Sese Seko who referred to himself as the Président-Fondateur of the Mouvement populaire
de la Révolution in Zaire. See Emmanuel Nkunzumwami, La tragédie rwandaise, p. 74. President
Habyarimana refused to consider the Mouvement révolutionnaire pour le développement a political party. He
described it as a social and political movement incorporating all Rwandese to promote peace and development
in the country. Interview in 1976 with M. E. Ugeux, National Archives, Kigali, Rwanda.
94
On the development efforts in Rwanda by the international community, see Peter Uvin, Aiding Violence: The
Development Enterprise in Rwanda (West Hartford, Conn.: Kumarian Press, 1998).
93
164
The rhetoric adopted by Habyarimana and the MRND, even their actions, allowed
for some respite particularly from the mass violence of the First Republic. There was a
relative pacification of relations in the country. Though some killings did occur, they were
on a smaller scale than what the country had known previously.
This calm was
misleading, however. Though breaking with the pattern of brutality of the previous years
as well as claiming to be distancing itself from the corruption, the Habyarimana regime
was very much the prolongation of Kayibanda’s.
Despite changes in the rhetoric adopted, it continued to stay true to the principles
of the 1959 Social Revolution. Habyarimana did not break with what he saw as the gains
(acquis) of 1959.
For him, the Revolution remained the founding moment of modern
independent Rwanda and, as such, needed to remain the foundation of subsequent
regimes, including his own. 95 For example, to mark the continuity between his regime and
1959, but to distance himself from the corruption and abuse of the Kayibanda years, he
after the fact called his coup the ‘Moral Revolution.’ Furthermore, the fundamental basis of
the ideological platform remained the same. Though not as explicitly as during the First
Republic, the ideological underpinnings of the regime remained very much the conception
inherited from the colonial era: the Hamitic myth, and its reinterpretation by the Parmehutu.
For example, in a public speech, President Habyarimana stated that: “[w]e know that for
more than four hundred years, the Tutsi subjugated the Hutu and prevented the Twa from
remembering they were human. Our revolution in 1959 was the first to overthrow the Tutsi
authority by giving back to the majority both its self esteem and power.” 96 With such beliefs
still at the heart of government’s policies, the Tutsi had to remain a group under
surveillance and control.
95
The idea of a founding moment is taken from an interview with Gérard Prunier. Interview with Gérard
Prunier, Centre français des études éthiopiennes, Addis Abeba, Ethiopia, 25 April 2004.
96
Quoted in Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 122.
165
Thus, although breaking with the violence of the First Republic, the Habyarimana
regime did not, in practice, break with oppressing the Tutsi. Instead, it chose to subtly
exercise its discrimination.
There was a clear difference between how the First and
Second Republics practiced their oppression in terms of the methods chosen for political
action. As Kimonyo explained, “[born out of violence], the First Republic concentrated on
legitimating its ethnic ostracism policy by institutionalising a dramatisation of ethnic
antagonisms, the leaders of the Second Republic, when dealing with ethnic issues,
preferred sobriety.” 97 Thus, under Kayibanda, the regime demonised the Tutsi, choosing
to brutally mark the difference between groups. Under Habyarimana, on the other hand,
the Tutsi was diminished, systematically excluded and left to a second rate status. With
extreme sophistication, in a technical, rational manner, this status came to be
institutionalised in the Second Republic through the development of mechanisms and
structures ingraining an ethno-centric and divisive ideology in the country, diffusing ethnic
based conceptions of groups and their dynamics, and more importantly marginalising one
of them, the Tutsi.
The regime began this effort by systematising a quota system that gave the Tutsi
limited access to institutions that would have allowed them to prosper. The impacts of
such measures were dreadful. Not only were the Tutsi constrained in terms of social
mobility, but as a group they were not given the capacity to develop intellectually. 98 This
systematic form of exclusion was not new in Rwanda.
A quota system had been
developed under the First Republic, but had not been systematically implemented. 99
Following the violence and purges of the last months of the Kayibanda regime,
Habyarimana applied his version of the system to control the number of Tutsi in specific
sectors, particularly in the public sector, the army and in the education system. This quota
97
Personal translation. Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 131.
Interview with Jean-Paul Kimonyo, Kigali, Rwanda, 30 March 2004.
99
As a matter of fact, the 1972 purges under the Kayibanda regime were intended to ensure the quotas were
respected. Gérard Prunier, The Rwanda Crisis, p. 60.
98
166
system prevented the Tutsi from having more postings or places in schools than the
proportion they represented among the Rwandese population, which the regime estimated
to be between nine and fourteen percent. According to Gérard Prunier, these quotas were
invariably violated as some Tutsi managed to pass for Hutu by forging their identity
cards. 100
Many Tutsi interviewed in Rwanda held the opposite view, however.
They
claimed that the quota system was indeed generally violated but to the advantage of the
Hutu who tended to be overrepresented in all sectors. Such an impression could stem
from the fact that, over time, positions in and admissions to schools became a resource
the authorities used for patronage. 101 Ironically, the regime called the system ‘ethnic and
regional balance’ and claimed that the policy was instituted in order to redress previous
inequalities, mainly injustices against the Hutu that had been committed during the colonial
period, but also the regional favouritism that had prevailed during the Kayibanda years.
Quotas were also applied on employment.
The system served to regulate
appointments in the public sector, for example. Few Tutsi could be hired as civil servants.
Furthermore, as the state controlled many sectors, including the health, education and
research sectors, it also limited their access to certain liberal professions. As an example,
a Rwandese woman retold in an interview how it had proved to be an ordeal for her Tutsi
husband to find a job after he had completed his studies. According to her, “he went to
look for employment at the Department of public service. He was told that ‘there is an
opening for a statistician. But you will have to find nine others so there can be balance.’
For him to get the job there had to be nine Hutu and one Tutsi.
openly.” 102
He was told that
After much negotiating, he eventually managed to get the post.
Many
interviewees recounted similar stories of being refused because quotas had been reached.
With so little opportunities for employment in these sectors, many Tutsi turned to private
100
Interview with Gérard Prunier, Centre français des études éthiopiennes, Addis Abeba, Ethiopia, 25 April
2004.
101
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 125.
102
Interviewee no. 23, Kigali, Rwanda, 1 April 2004.
167
sector entrepreneurship to make a living.
There were also constraints in this sector,
however, since, in many cases, in order to be successful, they had to gain the support of
Hutu patrons.
Quotas also existed for postings in the military, especially at the officer level. Strict
rules were applied for admittance in the military academy, which tended also to exclude
Hutu from the South or of mixed descent. Military officers were also generally forbidden to
marry Tutsi women. All marriage licenses were approved by the minister of Defence to
ensure the rule was enforced. 103
The most pervasive application of quotas was in the education system, however.
While elementary school was accessible to all, earning a higher education proved difficult
for Rwandese Tutsi. They were allowed to complete a first cycle of high school but further
access was controlled by quotas. Admissions to university were even more restricted. 104
To this day, many Rwandese still retell their personal stories of having been excluded from
the education system or having had to cross, at a very young age, to neighbouring
countries to complete their schooling.
As an example, an interviewee remembered a
conversation he had with the director of his high school upon his return to school in
September 1973, following purges in the last year of the Kayibanda regime. He recalled
the director telling him that:
We have received instructions from the Ministry of Education stating that in high
school there should be a maximum of ten percent of Tutsi. In this school, we have
150 students, so normally I should not have more than 15 Tutsi [...] I have counted
recently and I have 25 Tutsi. I wrote to the minister [...] and asked him what I should
do. The minister replied in a letter that I should keep the situation as it is, that I
should not accept any other Tutsi [...] that things would balance out eventually [...] in
the following years. [...] So you, young man, I am sorry, I can not accept you, you
must go home. 105
103
Interviewee no. 15, Kigali, Rwanda, 24 March 2004.
Some private schools were created in the 1980s by individuals wishing to bypass this restrictive quota
system. They were only given the right to deliver official diplomas in the late 80s however. Interviewee no. 15,
Kigali, Rwanda, 24 March 2004.
105
Personal translation. Interviewee no. 20, Kigali, Rwanda, 29 March 2004.
104
168
He eventually left for Burundi, where he completed his studies. Furthermore, according to
many, the few available places in schools were often attributed to children of richer
families who bought their children’s access to education. Such was the case for some
Hutu as well. It was sometimes difficult for Hutu of humble origins to gain access to higher
education, especially to university, as priority was often given to children of families within
or with ties to the state apparatus or who came from the regions favoured by the regime:
Ruhengeri and Gisenyi.
The limited education the Rwandese Tutsi had access to, however, proved to be
one of the regime’s preferred tools to disseminate its ethno-centric and divisive ideology.
Education was a medium to teach differences between groups, to indoctrinate and to
influence impressionable young minds. 106 This was done firstly through informal dynamics
in class which emphasised group affiliation. The Rwandese who were educated under the
Habyarimana regime often remember how the teachers insisted on openly marking group
differences between children in class. Some would be asked to bring their father’s identity
card at the beginning of their schooling so their background could be known. 107 The Tutsi
children could therefore be singled out by the teachers. 108 Oftentimes, at the beginning of
each term, the children were asked to raise their hand to identify to which community they
belonged and separated into groups accordingly.
One of the Rwandese interviewees
recalled that the very young children sometimes would not know which group to join. 109
But once aware of the distinction, many Tutsi children felt shame during this exercise of
having to admit their affiliation and were teased by their classmates.
The shame and teasing resulted from the fact that stereotypical distinctions and
connotations were ascribed to Hutu and Tutsi identities, particularly in civic education and
106
For more on the role of education in Rwanda, see for example Elisabeth King, “Educating for Conflict or
Peace: Challenges and Dilemmas in Post-Conflict Rwanda,” International Journal, vol. 60, no. 4, 2005, pp.
904-918.
107
In Rwanda, group affiliation is patrilineal.
108
Interviewee no. 10, Kigali, Rwanda, 19 March 2004.
109
Interviewee no. 29, Kigali, Rwanda, 15 April 2004.
169
history courses, though the teachings in both often overlapped. Civic education courses
were developed to produce good citizens. And for the Rwandese regime of the time, good
citizens amounted to the ones that recognised the difference between Hutu and Tutsi. 110
At the heart of civic education classes were the ideological conceptions underpinning the
regime’s conception of democracy and political rights. The Hutu were portrayed as the
Rubanda Nyamwinshi, the majority and natural inhabitants of the land, which justified their
hold on power and that the Presidency along with higher political postings belong to the
Hutu.
This civic education was complemented by what was taught in history classes, the
Hamitic myth, with special attention paid to the different origins of groups in Rwandese
society, thus different racial or ethnic backgrounds, and their different physical appearance
and traits of character. As was the case in previous decades, it was taught that the Tutsi
had been late comers in the country, had arrived as invaders, but had finally been
successfully ousted from power by the Revolution.
History courses therefore built a
version of history rooted in Manichean perceptions of Rwandese communities. The Hutu
and Tutsi were presented in an antithetical manner, one the positive pole, the other the
negative.
While students learnt that the Hutu had freed the country as true Republicans,
the Tutsi were depicted as oppressors and monarchists. The latter were often stigmatised
as a whole, traits ascribed to the Tutsi of previous eras undifferentiatedly being attributed
to their contemporaries. Clichés often depicted the Tutsi in general terms, as cunning and
power-driven. Anecdotes were used to illustrate these and shape children’s minds. One
quite often retold in interviews is the story of Kanjogera, an evil Tutsi queen of the turn of
the nineteenth century, presented as the supreme representative of the oppressive
monarchy. Kanjogera, the children learnt, was a bloodthirsty woman who would plant
spears in the back of Hutu slaves in order to help herself off her throne.
110
Interview with Jean-Paul Kimonyo, Kigali, Rwanda, 16 March 2004.
170
What both civic education and history courses achieved was the demonisation of
the monarchy.
It was portrayed as an absolute dictatorship, a system in which man
exploited man. The courses painted such a horrendous picture of this political system that
children puzzled over monarchical systems still in existence around the world. As an
interviewee recalled, “as children, we would hear about countries like Belgium or England
where there was still a monarchy.
We didn’t understand. We had learnt to hate the
monarchy and to demonise the ones that were in power before the Republic.” 111 The 1959
Revolution, moreover, was often compared to the French Revolution, the French
Revolution had banished French monarchy just as the Hutu had overthrown Rwandese
Tutsi feudalists. 112
These brutal images of the Rwandese royaume conveyed in class
made a lasting impression on school children. Many remembered times when, after these
lessons, some of their comrades were angered by what they had just learnt.
Some
expressed resentment towards their Tutsi companions: “[l]ook at what you did to our
ancestors. We will take revenge.” 113 For their part, the young Tutsi were made to feel
shame for belonging to a group that had for decades subjugated the Hutu. One father
remembered his child coming home from school one day saying: “Dad, we learnt in school
that the Tutsi were really bad, they mistreated the Hutu. I can’t be Tutsi.” 114 As one of the
principal means for intellectual development and the socialisation of the young Rwandese,
this biased education proved to be an extraordinary tool to transmit the regime’s ideology.
It was not the only mechanism employed for this enterprise, however. The MRND
itself was an elaborate organisational structure serving to propagate the ethno-centric and
divisive ideology of the regime. Reaching even the most remote hills in the country, this
authoritarian political structure exercised total control of the political sphere and limited all
other forms of association. There were no opposition parties, very few civil society
111
Personal translation. Interviewee no. 24, Kigali, Rwanda, 2 April 2004.
Interviewee no. 10, Kigali, Rwanda, 19 March 2004.
113
Interviewee no. 18, Kigali, Rwanda, 26 March 2004.
114
Personal translation. Interviewee no. 10, Kigali, Rwanda, 19 March 2004.
112
171
organisations, such as human rights associations, as well as few trade unions and coops. 115 The ones that existed were part of the MRND: “[w]omen’s organizations, youth
groups, labor unions, and farmer’s cooperatives were subsumed under the party
umbrella.” 116 The Mouvement révolutionnaire national pour le développement, of which all
Rwandese were automatically members by birth, was transparent about its controlling
nature. It demanded of all Rwandese unwavering support. A platform stated that:
The Mouvement is for the masses and requires uncompromised dedication, in other
words, the behaviour of the masses, of the entire society, has to be modeled around
a unique standard leading to a unified point of view, harmony, cohesion from cell to
the higher instances of the Movement, of the entire nation. No individual or group of
individuals can escape this structure of social control. 117
Control of the masses was ensured through an intricate system of mechanisms tying
individuals to this overarching organisation.
For example, each parish was given its
project within a national development initiative. The project was supervised by the local
administration, which was itself under the supervision of the regional administration, which
was in turn subordinated to higher administrative levels. All these connected echelons
ensured the local level remained tied to the regime.
Another system the MRND
developed to mobilise the Rwandese population was the Umuganda. Inspired from a
system previously in place during the colonial era, the Umuganda was a form of
mandatory community work each individual had to take part in for the good of the country
through activities such as roadwork, reforestation or the construction of public buildings. 118
Through their close involvement in the development efforts of the country, the Rwandese
were asked to demonstrate their attachment to the regime.
115
Peter Uvin, Aiding Violence, p. 22.
Timothy Longman, “Christian Churches and Genocide in Rwanda,” pp. 6-7.
117
Personal translation. Quoted in Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 134. Pregenocide administrative units in Rwanda were organised along a hierarchy going from cells (cellules), sectors
(secteurs), communes (communes) to prefectures (préfectures). A smaller informal level also existed: the Ten
houses.
118
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 137.
116
172
This was further reinforced by the fact that a good part of the Umuganda was
devoted to the animations held on Saturdays, a communal gathering overseen by local
authorities that were an occasion for the population to express its allegiance to the
Mouvement. On such occasions, citizens were required to praise the President and the
regime through songs and dances led by professional dancers and organisers hired by the
government. “Propaganda teams of singers and dancers vied for honors in regular
competitions, often dressed in fine costumes bought by party faithful.” 119 In addition, the
population was made to voice its support through litanies developed by the authorities.
Emmanuel Nkunzumwami provided an example of the slogans en vogue at the time that
the Rwandese were made to repeat publicly: “[m]ay the President-Founder live long, may
he prosper, may he be the greatest, may he always move forward without fear, because
we all support him unconditionally.” 120 But the repertoire of songs and slogans for the
animations also included songs to the triumph of Gahutu over Gatutsi, praising his struggle
and finally victory in freeing himself from the bonds of the monarchy.
Furthermore, as under the First Republic, the Habyarimana regime employed antiTutsi violence to mobilise and foster popular solidarity. Although violence was not as
systematic as under Kayibanda, small political crises occasioned the re-igniting of ‘ethnic’
tensions. As a direct continuation of his predecessor’s feats, Habyarimana resorted to
ethno-centric and divisive rhetoric when the political situation turned sour, channeling
popular support against a favourite scapegoat, the Tutsi. Violence against the Tutsi was
one of the preferred mobilising mechanisms when the regime faced its most challenging
crisis at the end of the nineteen-eighties.
119
120
Human Rights Watch, Leave None to Tell the Story, p. 43.
Emmanuel Nkunzumwami, La tragédie rwandaise, p. 96.
173
Impact of the Two Republics on the Population
The First and Second Republics had a dramatic impact on the country’s population and on
their perceptions of community and intergroup dynamics. Under Kayibanda, the Tutsi was
demonised. The regime brutally entrenched an extreme differentiation between groups.
Under Habyarimana, the Tutsi were tolerated but lessened, marginalised and
dehumanised; they were allowed to exist but given very few rights, forced to dwindle in an
environment in which discrimination was systematic. Both regimes worked, in their own
way, to ingrain among the masses an ethno-centric and divisive ideology.
Nonetheless, despite these actions on the part of authorities, on the ground, in
terms of horizontal relations between people, there remained relative peace during the
nineteen-eighties. Outside of recurring incidents of violence and relations with the state,
when not competing for social promotion, there was relative calm. Despite a few
exceptions–certain regions in the country were, for historical and demographic reasons,
the subject of more intense propaganda and repression on the part of the regime, and the
older generations who remembered more vividly purges against the Tutsi and reacted by
keeping their resentment alive–relations on the ground were good, even amicable,
especially among the peasants. 121 This was the case even for the Tutsi. As Prunier
argued: “[a]ll in all, life was difficult for the Tutsi who were victims of institutional
discrimination, but in everyday life it was quite tolerable.” 122 On a daily basis, the groups
cohabited peacefully.
During the First Republic and early days of the Second, they lived intermixed, at
church, at the market, at the cabaret sharing drinks, on the hills, generally
undifferentiatedly.
Often because of the hardships of rural life, they were required to
121
This assessment is based on a series of interviews conducted in Rwanda between March and April 2004.
Most respondents indicated that outside of political crises which were the occasion of political violence, the
groups co-habited peacefully. Some regional differences did exist however. The oppression of the Tutsi
population tended to be more extreme and perceptible, thus affecting social relations, in regions where the
Tutsi population was larger or in regions that for historical reasons tended to be more dissident.
122
Gérard Prunier, The Rwanda Crisis, p. 76.
174
collaborate, to help each other.
In many communities, there existed a culture of
amiability. 123 They would cultivate the land together and share resources when needed.
When a child was born, the tradition demanded that neighbours bring milk and sorghum to
the new mother. Or, some could lend a cow for a year or two to poor families so they
could have milk to raise their children. They also carried the sick together on the long road
to the clinic, as was done traditionally. This collaboration also extended to times when
regime-sanctioned violence recurred. During crises, many Hutu warned, helped and hid
their Tutsi neighbours.
Some interviewees even contended that there existed an old
protection pact among communities. 124
Solidarities were thus still very real among
communities on the hills during the two Republics: communitarian ties and identities were
strong in everyday life.
The workings of the regimes filtered through, however. The ethno-centric and
divisive ideology, and tensions associated with it, transmitted through a descending
vertical relationship from the authorities to the masses was taking hold. It did so in its own
way. The ‘ethnic’ rhetoric was undeniably developed and upheld primarily by the
authorities and intelligentsia at the time and not by the masses.
It was therefore in
relations with these authority figures that it transpired the most. And, although Kayibanda
and Habyarimana each had their preferred method of indoctrination, the two regimes
nevertheless achieved the same result. 125 They set up ethnic affiliation as a filter, a
referent, for specific dynamics involving relations to authority. This translated into tense
123
Interviewee no. 20, Kigali, Rwanda, 29 March 2004.
Interviewee no. 4, Kigali, Rwanda, 11 March 2004.
125
It would be difficult to assess the regimes’ intentions in terms of ideology. Was its development and
promotion under the two Republics uniquely conscious and strategic? Or was part of it the result of choices
made previously, during the 1959 Social Revolution for example, that ended up closing off options and
dictating the path to take? For situations like the latter, Thomas Homer-Dixon makes a parallel with Dr.
Frankenstein: elites create a system that then takes on a life of its own. Their actions thus become
constrained by what they have themselves created. In an ideological system, they therefore are forced to keep
referring to the system and functioning within the system, thus further reproducing it. Discussion with Thomas
Homer-Dixon, Trudeau Centre for Peace and Conflict Studies, Toronto, Canada, 13 May 2005. It would be
difficult to establish clearly the motivations of elites, at this point. Nonetheless, the outcome of their actions
remains the enshrinement of its ideology which they in turn did not hesitate to refer to in order to stoke the
masses and mobilise them.
124
175
and ‘ethnicised’ relations for the population in contacts with the state. Ethnic affiliation was
the predominant form of identification used to determine one’s place in the system.
Citizenship and political identities were inherently filtered through the prism of ‘ethnicity’
and implicitly antagonistic.
The ideology and categories were made unavoidable within the system and it was
overwhelming.
Every form of political authority upheld the official ideology. Whether
political authorities through speeches and public rhetoric, intellectuals and teachers
through research, publications and teaching, or simply because they were the educated
few and thus represented a thinking authority, political life upheld the regime’s ideology.
Even the church, the moral authority in the country, reproduced the ideology. 126 Each of
these groups succeeded in reaching the Rwandese in one sphere or another of their public
life, and ensured they were submerged by the ideology. The system was omnipresent,
particularly in the early days of the Second Republic. Little space remained in which one
could think freely. 127 Little space remained for contestation, for alternative stories to
emerge. It was made difficult to express dissidence. Everything emanated from the top,
from a single political project, from the regime’s platform.
126
As Timothy Longman explained, “[t]he missionaries and their Rwandan protegés played an essential role in
eliminating the flexibility in Rwanda’s ethnic structure, creating a rigidly defined system in which the Tutsi
monopolized power and benefits. The national history they constructed presented this new structure as an
historical artifact that became the basis for pride among the Tutsi, and, later, resentment among the Hutu. Just
as the churches had previously been the champions of the ‘noble Watusi’, at independence they became
champions of the humble Hutu.” (p. 6) Longman described at length the ties between churches and the regime.
As he stated, “[i]n general, [...] the churches ad the state cooperated closely. Both institutions, building on the
history of cooperation established during the colonial era, found that they could function more effectively by
working together on education, health care, and development. The interests of the church employees at all
levels were closer to the parallel state elite, with whom they may have attended school and who lived a similar
privileged lifestyle, than to the rest of society.” Timothy Longman, “Christian Churches and Genocide in
Rwanda,” p. 7.
127
As described, the MRND reached deep within society. It controlled associations and unions. It ensured its
representation at all administrative levels. For a description of the MRND’s organisation, see Human Rights
Watch, Leave None to Tell the Story, p. 41. Furthermore, freedom of the press remained elusive during the
period, whether under Kayinbanda or Habyarimana. The government continued to keep a tight control of the
press. See for example, Annie Bart, La presse au Rwanda. Production, diffusion et lecture depuis le début du
siècle, doctoral thesis, Université de Bordeaux III, Bordeaux, France, 1982; Bernard Mulinda, La liberté de la
presse au Rwanda (1974-1994), licence thesis, Faculté des lettres, Université nationale du Rwanda, Butare,
Rwanda, December 1999.
176
Furthermore, this system predominated for so long, the span of decades, that it
eventually became more than simply an ideology. It turned into socialisation.
It was
present from youth to adulthood and its responsibilities such as participating in sanctioned
ceremonies or programs, but also in cultural and social contacts, in the media, etc. This is
what one was raised in and lived in. People grew up in an environment permeated with
this ethnocentrism. The ideology infiltrated so many aspects of the Rwandese’s lives that
it became a social structure. 128 It was consolidated in the form of ideology as practice: not
an abstraction anymore but a presence in every aspect of public social existence. And as
such it became extremely compelling, much easier to buy into. It seeped through on
formal levels but also in the informal and the subconscious. Groups could not remain
unaffected by this ideology and all groups were affected in their own way.
The Hutu population, despite the fact that it lived peacefully in daily life with its Tutsi
neighbours, allowed itself to be convinced of the existence of ‘essential’ differences
between groups. The Hutu increasingly adopted the regime’s ideology. This was the case
especially among youth who had benefited from the 1959 Revolution in terms of social
mobility, particularly in terms of educational opportunities and who were given the
opportunity to constitute the new elite of the country. Having completed their schooling,
many reintegrated into their communities. Brought up with the ethno-centric ideology of
the time, but having also benefited from this system, they became the new carriers of the
ideology, and some of the strongest defenders of the Revolution and the ensuing Hutu
regimes. Having been molded by the system, sensitised and radicalised by the education
they received, they were active socially and politically. 129
128
Aware of the importance of
Interview with Jean-Paul Kimonyo, Kigali, Rwanda, 16 March 2004.
Many of this new intelligentsia unwaveringly supported and promoted the regime’s ideas because they
owed to the Revolution their privileged position, having gained through it access to school and job
opportunities when the favouritism towards the Tutsi was reversed. The fact that they benefited greatly from
this system facilitated their acceptance of it and, in turn, their indoctrination.
129
177
these activities for further upward social mobility, they reproduced and disseminated the
ideological structure among their own rural communities.
This rural intelligentsia served as the intermediaries of the message developed in
the capital, playing the role of organic intellectuals that mediated between the political
system and the base. 130 They constituted a critical mass reproducing these teachings
among their families, as well as within their social and professional circles, in
conversations and debates. But as educated individuals, they constituted the new elites in
these
communities,
often
taking
posts
as
provincial
representatives,
sector
representatives, members of the party at the local level, or community leaders, and thus
exercised a great influence in their respective communities, turning the MRND into a
powerful overarching mobilising system reaching all the way down to the smallest
administrative unit of Rwandese society, the Ten houses.
Their role as conveyors of the ideology was even more important in light of the fact
that for most peasants, the intellectualised ethno-centric discourse upheld in schools was
incongruous with their own cultural background, the traditional Rwandese myths, including
the myth of Gihanga. This new elite thus served as interpreters of these theories. Having
come from these communities and therefore cognizant of popular cultural referents, they
translated and simplified the ideology using references their communities could
understand. They tailored the ideology to render it accessible, compatible with the existing
background.
Many reinterpreted, for example, the myth of Gihanga by insisting on
differences between the three brothers instead of focusing on what united them. 131 They
130
Interviewee no. 14, Butare, Rwanda, 24 March 2004. The term ‘organic intellectuals’ was coined by
Antonio Gramsci. For Gramsci, organic intellectuals are formed by the class they belong to and come to serve
its specific interests. For example, the ones formed by the dominant system in turn help ensure its
reproduction. See Antonio Gramsci, Prison Notebooks: Selections, translated and edited by Q. Hoare and G.
N. Smith (New York: International Publishers, 1971).
131
It would be interesting to analyse the impact of this reinterpretation of traditional Rwandese myths. The
individual lost his cultural referents, his roots, and was made to doubt what he/she previously held as intrinsic
societal beliefs. Could this sense of destabilisation, of loss, of the traditional being confronted by the modern,
have made him/her more malleable to the strategies of elites, particularly that in this situation the new elite
178
also simplified the myths they promoted, steering clear of grand theories which could not
really have an impact on these populations. Many of the latter were uneducated and,
without certain historical bases, could not relate to some of the more abstract theories
about arrival dates and ethnic origins. The new elite adopted simple formulations with
clear contrasts, such as native/foreigner or oppressor/liberator. With such powerful yet
familiar representatives of the ideology, many in the communities lent credence to what
this new ‘voice of the system’ said.
But this new tailored version of the ideology also appealed because it resonated
with the population’s proximate environment. What they saw around them seemed to
confirm this version of the facts. People started noticing that the differences spoken of in
ideological precepts did translate into their reality.
The differentiated treatment of
communities within Rwandese society was turning into a permanent feature of the system.
The Hutu were in fact privileged. People started adjusting to the fact that the Hutu had
more rights, even accepting it. After all, the Hutu predominated in the political system, in
schools as well as in the economy. It was so blatant as to be an unquestionable reality. It
was a Hutu Republic. Under both Kayibanda and Habyarimana, Rwanda belonged to the
Hutu. And, in this reality, the Tutsi were second rate citizens, not only because they were
limited by various constraints, but also because of the impunity that often accompanied
injustices and acts of violence perpetrated against them. Persecution of the Tutsi was
rarely condemned. The fact that violence against the Tutsi often took place in all impunity
seemed to tacitly condone it, as if it was within the realm of the acceptable. For thirty
years, impunity reigned, normalising the injustice and violence.
It was, therefore, a
characteristic of the system but one projected unto the Tutsi. They had a lesser status.
They were obviously not deserving of full protection and full citizenship.
created the ‘internal’ cultural confusion, but simultaneously offered to replace it by its external certainty, offered
to become a guide in this destabilisation?
179
The long institutionalisation of this social norm of differentiation and discrimination
also had its impact on the Tutsi. While a good number of them left Rwanda during the
many purges, especially during the first years of the Kayibanda regime, many adapted.
They did not really voice their frustrations or oppose their unjust treatment. There was
little internal contestation on their part. This apathy was the result of fearing the
consequences of speaking out of suffering the consequences. But there also existed a
feeling of resignation among the community, resignation to the workings of the system and
to their role in it as the scapegoat. Many simply felt defeated, the victims of history. 132
They accepted to be less than others, to be renters, strangers, in their own country. They
resolved to remain silent in the face of the recurrent violence against their group. For the
many that had chosen to remain in Rwanda, they had agreed to live under the system, and
it thus meant accepting its workings. As an illustration, a Rwandese mother remembered
telling her child coming home from school after having been bullied for being Tutsi: “it’s like
that, you have to accept it. Us too, growing up, we were made to understand that the Hutu
were superior. You have to adapt.” 133 One interviewee even stated that he had reached
the conclusion that not being persecuted was simply abnormal. 134 He further stated that
they took “massacres as bad bouts of rain, hard times that had to be endured, and then
when it is all over life went back more or less to what it was before.” 135 Not only did they
resign themselves to this state of affairs, but many interviewees claimed they also began
internalising the government’s rhetoric and as a consequence developing an inferiority
complex, the victim’s syndrome, and feelings of guilt.
Some began feeling that the
treatment received was a justified punishment for past deeds committed by their
ancestors. 136 This feeling ran so deep in some that an analyst hypothesised that, in light
132
Interview with Jean-Paul Kimonyo, Kigali, Rwanda, 16 March 2004.
Personal translation. Interviewee no. 22, Kigali, Rwanda, 1 April 2004.
134
Personal translation. Interviewee no. 27, Kigali, Rwanda, 13 April 2004.
135
Ibid.
136
Interview with Jean-Paul Kimonyo, Kigali, Rwanda, 16 March 2004.
133
180
of this, many even reacted with complete apathy during massacres, letting themselves
passively be killed. 137
The ideology increasingly took hold of people’s minds in a number of regions
across the country. 138 Life under the two Republics became schizophrenic. Individuals
were split between two opposed modes of existence.
The Rwandese lived a
dichotomisation of their life, their communitarian and civic identities experienced in an
antithetical manner. On the one hand, the Rwandese had their communitarian identity, an
accommodating identity devoid of ethnic markers. This identity remained the prevalent
filter for amicable relations on the hills, for a parallel existence allowing them to escape the
ideological or sometimes physical brutality of the state apparatus. But their civic and
political identity remained, on the other hand, constructed by this authoritarian state
apparatus, demanding deference and an unqualified adherence to its ideological precepts.
Identities were thus read through the prism of ethnicity and discrimination, of the extreme
differentiation between groups. Both forms of identity, the communitarian/personal one and
the civic/political one, though antagonistic, coexisted simultaneously. Rwandese existence
proved to be quite erratic as this dichotomisation demanded that individuals slip constantly
from one to the other.
Among individuals and communities something was seeping through, however.
With every passing year, every new crisis, the ideology entrenched itself deeper.
disposition remained.
A
At the surface, there clearly remained the regime-sponsored
ideology, its effective mechanism for mobilisation and solidarity, a deeply ingrained but
137
Ibid.
Between 1959 and 1990, there existed regional variations in terms of the degree of internalisation of the
ideology. Much depended on efforts by the government in various regions at propagating the ideology. For
example, purges, violence and the ideological propaganda tended to be concentrated in regions where there
was a larger Tutsi community. Furthermore, the regimes tended to focus on the North and South of the
country. The East and West remained peripheral, outside of power struggles, but were also as a consequence
less affected by efforts to diffuse the ideology.
138
181
somewhat latent ideology, at least in its most noxious forms. 139 It was latent, however,
until its activation during crises, at which point it could be used to re-ignite violence.
Underneath the surface, this slow socialisation to the regime led to the travesty of values
and standards, in the form of an acceptance of discrimination, injustice and even violence.
In turn, it reconfigured the collective matrix of acceptable thinking and behaviour. 140 In light
of the extent of brutality in society and of impunity for acts against the Tutsi increasingly
segregation and even violence moved within the realm of normality. This trend was,
according to many, the start of a process that would culminate in the complete
displacement of communitarian ties in 1994. It had begun opening the door to genocidal
violence, sweeping violence that in light of this reconfigured matrix of behaviour was not
unthinkable anymore. Rwanda had to experience civil war, though, before the levies broke.
139
Bruce D. Jones, “Civil War, the Peace Process, and Genocide in Rwanda,” in Taisier M. Ali and Robert O.
Matthews, eds., Civil Wars in Africa: Roots and Resolution (Montreal: McGill-Queen’s University Press, 1999)
p. 70.
140
Interviewee no. 14, Butare, Rwanda, 24 March 2004.
182
CHAPTER 8: RWANDA: THE CIVIL WAR AND THE GENOCIDE (1990-1994)
Background Variables
On 1 October 1990, a civil war broke out in Rwanda following an attack by the Rwandese
Patriotic Front (RPF), composed mainly of Tutsi refugees in Uganda.
Though in the
following years efforts were made to quell the violence and stabilise the country, the worst
of the bloodshed was yet to come. Within years, the country would instead be swept by
gruesome violence: the 1994 Rwandese genocide.
In trying to make sense of the unthinkable, 850 000 killed in a hundred days, a vast
number of authors have turned to the well-documented planning and orchestration of the
genocide by a group of elites to explain this devastating explosion of violence. 1 Against
some of the more simplistic media accounts describing the events as tribal violence, this
instrumentalist camp contended that the planning of mass violence in Rwanda was a
reaction on the part of a group of extremists dominating the state apparatus who suddenly
found their privileged position threatened by an unfolding multifaceted crisis undermining
the regime’s legitimacy. 2 Their genocidal plan was in fact implemented in 1994 with dire
consequences for the Tutsi population.
There is no doubt that the simultaneous occurrence of a variety of factors led to
mounting stress and pressure on the regime in Kigali, leaving it with few options to resolve
the crisis and save what little legitimacy and popular support it still had. To paraphrase
Pierre Lizée, for a regime faced with a multiplicity of problems, compounded stress, and
with a range of opportunities and strategies narrowing rapidly, violence was becoming “the
1
The number of people killed during the genocide is still uncertain. Gérard Prunier estimates it to be 850 000,
a number he has based, among other things on data from the Red Cross. See Gérard Prunier, The Rwanda
Crisis: History of a Genocide (London: Hurst and Company, 1995), p. 265. Furthermore, this estimate only
takes into account deaths that occurred at the time of the genocide. Many will die at a later date, and many
continue to die, from being infected with HIV/AIDS during the killings.
2
Jean-Paul Kimonyo, La participation populaire au Rwanda: de la révolution au génocide (1959-1994),
doctoral thesis, Département de science politique, Université du Québec à Montréal, Montréal, Canada,
September 2002, p. 5.
183
only currency for power.” 3
Thus, to understand what happened in Rwanda, and to
understand the actions of elites behind the orchestration of the genocide, it is necessary to
look at the background variables that led the regime to face a deadlock by the early
nineties.
As these analysts contended, some of the roots of the 1994 genocide can be
traced to political trends that had begun developing in the 1980s. By the late eighties, the
regime was undeniably showing signs of exhaustion, facing problems on the domestic,
regional and international fronts.
Internally, the Habyarimana regime, like its predecessor, was confronted with
growing criticism for its regional bias. Its preference for the Northern parts of Rwanda had
already been apparent in its early days, a majority of the instigators of the 1973 coup
having originated from the North. This favouritism towards the North had continued to
develop throughout the late 1970s and early 1980s. This trend was particularly visible in
how the Habyarimana regime chose to apply its policy of ‘ethnic and regional balance.’
This tendency had gotten worse as years passed with more and more power positions
being attributed to Northerners. 4 This created discontent in Southern regions.
Increasingly, however, this regional focus came to be confined to a specific Northern
region, Gisenyi, birthplace of President Habyarimana and his wife. As a result, other
Northern regions were overlooked, including Ruhengeri, which had once been also
privileged by the regime.
In 1980, relations soured with the Ruhengeri faction in the
government when Alexis Kanyarengwe, one of the instigators of the 1973 coup, attempted
to overthrow Habyarimana.
As years passed, the rift between the old allied regions,
Gisenyi and Ruhengeri, worsened. It finally reached a point, following the RPF attack,
when the Gisenyi powerbase, the clique in the immediate entourage of the President,
3
Pierre Lizée, presentation during the ‘War to Peace Transitions’ panel, Eight Peacebuilding Consultations,
Ottawa, Canada, 19-20 January 2005.
4
Peter Uvin, “Tragedy in Rwanda: The Political Ecology of Conflict,” Environment, vol. 38, no. 3, 1996, p. 12.
184
became suspicious of their former allies. As Alison Des Forges explained: “they feared the
Ruhengeri Hutu were going to ally with the RPF and with the Hutu from the center.” 5
This narrowing of the privileged power base around the President was also partially
clan-based.
While Habyarimana came from a humble background, his wife Agathe
Kanzinga came from the Bushiru clan and belonged to a family that had once ruled one of
the independent Hutu principalities in the Gisenyi region. 6 Habyarimana was therefore
keen on entertaining favoured relations with his wife’s familial entourage, mainly along clan
lines, since he benefited both from its standing and networks.
This was done to the
detriment of relations with other Northern clans and groups, such as the Bakiga from
Ruhengeri.
Ties between the regime and the Kanzinga family had become so
conspicuous that few were surprised to see her brothers and cousin as prominent figures
within the regime. This powerful clique, first nicknamed le clan de Madame and at a later
point the Akazu, or little house, was by the late 1980s the main pillar upholding the regime.
Through the exclusion of larger and larger parts of Rwandese society, power had been
concentrated in the hands of the Akazu and its associates, along with much of the control
over the country’s resources. 7 Even among the Akazu dissent was surfacing. Much of the
fighting was about postings or opportunities offered by the regime, but some of it also
targeted the President. Prominent clique members sought to bend him to their will. As a
telling example, one of the Akazu’s influential members, Colonel Serubuga, had the
President’s protégé and anticipated successor, Colonel Mayuya, murdered in the late
1980s because his rise to power was displeasing the clique who hoped instead to see one
of its members replace Habyarimana. 8
5
Interview with Alison Des Forges, Human Rights Watch, Kigali, Rwanda, 4 March 2004.
Gérard Prunier, The Rwanda Crisis, p. 86. In ancient Rwanda, on top of larger entities such as the Nyiginya
kingdom, the territory also housed certain smaller Hutu kingdoms.
7
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 150.
8
Gérard Prunier, The Rwanda Crisis, p. 87.
6
185
Discontent was not only limited to the power base of the regime, though. It was
also growing among Hutu elites in general, including within the MRND. As Emmanuel
Nkunzumwami argued, this became particularly noticeable at the time of the 1988
elections: “[i]n 1988 were held the last presidential elections along with elections for the
CND in a climate of strong contestation and revolt against the political military
dictatorship.” 9 Many were beginning to sense the failings of the system, realising it was
unraveling, and were demanding changes to the party’s institutions. In August 1990, a
group of thirty-three intellectuals took position against the regime, criticising it publicly and
demanding democratisation. This emerging voice for change in Rwanda would come, in
time, to constitute a new opposition, especially with the birth of political parties early in the
nineteen-nineties. The regime remained wary of dissent, however, and reacted to this new
trend and morose political climate by falling back further on its narrow rank of supporters,
and by becoming extremely distrustful of competing personalities within the MRND. 10
Frustration with the system was trickling down to the population as well. After
years of authoritarianism, ‘cracks in the system’ were showing and the government’s
control over the country was slipping. The population was increasingly expressing open
resentment of stagnating or deteriorating living conditions, and of the glaring injustices and
failings of the regime. It was discontented with the considerable gap that existed between
their reality and the ideals of development and conciliatory rhetoric of the 1973 coup. 11
People saw the effects of an authoritarian system and were beginning to build the will to
free themselves of the ‘political and social imprisonment’ imposed on them by the
MRND. 12 This was in itself a telling sign of the regime’s loss of control over the country
and its estrangement from its popular bases. As Kimonyo argued: “[a]t the end of the
9
Personal translation. Emmanuel Nkunzumwami, La tragédie rwandaise: historique et perspectives (Paris:
Éditions L’Harmattan, 1996), p. 76.
10
Interviewee no. 10, Kigali, Rwanda, 19 March 2004.
11
Interviewee no. 14, Butare, Rwanda, 24 March 2004.
12
Interviewee no. 11, Kigali, Rwanda, 19 March 2004.
186
1980s, the Rwandese state failed to achieve total control of its peasant population. The
latter used all sorts of conducts, oblique, subversive, of avoidance, referred to in
Francophone africanist literature as conduites d’indocilité or politique par le bas, or as
‘disengagement’ in Anglophone literature.’ 13 These took various forms, from retreating
from the state, through finding comfort in one of the multiple religious sects sprouting up
throughout the country, to vandalism such as uprooting coffee trees and even outbursts of
anomic social violence. 14
Regional factors also poisoned the political climate in Rwanda at the end of the
eighties. Massacres in Burundi in 1988-1989 created tensions in Rwanda. Being the
mirror image of Rwanda in terms of communal composition, violent events in Burundi
always tended to affect the other and destabilised the region. As Prunier argued about the
two neighbouring states: “Their parallel–and at times common–past histories, their
comparable social structures, their constant and almost obsessive mutual scrutiny, fated
them to be natural mirrors of each other’s hopes, woes and transformations.” 15 It was
proving to be the case again.
Another predominant regional factor was the growing crisis around Rwandese
diasporas in neighbouring states.
The number of Rwandese, especially Tutsi, living
outside of Rwanda in the Great Lakes region, had been steadily rising in the previous
decades. Every outburst of anti-Tutsi violence sent new waves of refugees into exile.
With indicators pointing to greater stability in Rwanda under the Second Republic, some
entertained the idea of returning to the country.
Furthermore, a change in relations
between Rwanda and Uganda, first between people and then between leaders, brought
the question of return to the forefront.
Starting in the early 1980s, conflicts between
Ugandans and the Rwandese diaspora began developing. For a number of reasons,
13
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 25.
Interviewee no. 14, Butare, Rwanda, 24 March 2004.
15
Gérard Prunier, The Rwanda Crisis, p. 198.
14
187
including political and economic ones, the presence of Rwandese refugees in Uganda was
becoming problematic. As Prunier explained, the refugees “suddenly discovered that
people among whom they had lived for thirty years were treating them as hated and
despised foreigners.” 16 In reaction, many joined guerilla fighters led by Yoweri Museveni
who were seeking to change the regime in Kampala. Many of them even played an
important role in helping Museveni overthrow Milton Obote’s regime. Tensions did not
dissipate, however. Eventually, Museveni’s Rwandese allies became a burden. 17 As a
result, Yoweri Museveni began contemplating their return to Rwanda.
Habyarimana had always taken a strong stance against reintegrating the
diasporas, however. He had recurrently claimed that Rwanda could not accommodate a
large number of returnees. In a comparison, famous among Rwandese, he stated that
Rwanda was already like a glass of water, full to the rim, to which no more could be
added. 18 He firmly expressed his wish that they become nationals of the state in which
they had settled. The regime translated its position in practice as a great reluctance to
deal with the issue altogether. Every time an opportunity had arisen to negotiate a solution
with neighbouring countries, the regime in Kigali had stalled the process until it ground to a
halt. 19
The government tried to conceal the problem within Rwanda as well.
It
discouraged, if not outright forbade, contacts with the refugees, surveying exchanges with
16
Ibid., p. 70.
Emmanuel Nkunzumwami, La tragédie rwandaise, p. 77. Catharine Newbury, “Background to Genocide:
Rwanda,” Issue: A Journal of Opinion, vol. 23, no. 2, 1995, p. 13.
18
This comparison was cited by a good number of people interviewed in Rwanda and in other casual
conversations with the Rwandese population.
19
One such instance occurred in 1974. The President of Uganda at the time, Idi Amin Dada, reacting to the
change of regime in Kigali, approached Habyarimana about negotiating an agreement on returning the
refugees to Rwanda, insisting on his intention of letting the Kigali regime proceed at its own pace.
Habyarimana agreed to the proposal and even organised consultations with the refugees. He eventually sent
a delegation composed of prominent MRND members to Uganda to begin discussions with the Ugandan
government and the diaspora. The meeting was intended to work out a plan to repatriate the refugees. Once
the discussions concluded, the delegation sent a secret memo to Habyarimana explaining that they did not
sense the economic situation was ripe for the refugees’ return, that Rwanda was already overpopulated, and
that these refugees were in fact enemies of the state. As a result, the plan was never implemented. Interview
with Jean-Paul Kimonyo, Kigali, Rwanda, 16 March 2004.
17
188
them but also movements between states by confiscating passports upon one’s return to
the country. 20 The regime kept the issue so secret within Rwanda, especially demands by
the refugees for their right of return, that many Rwandese were unaware of the problem.
They were clueless as to the size of the diaspora and its discontent. Faced with the
constant rebuttals of Kigali, refugees from various countries began organising themselves
politically, the ones in Uganda creating the Rwandese Alliance for National Unity (RANU),
which renamed itself Rwandese Patriotic Front (RPF) in 1987. They began demanding the
right to return to Rwanda more forcibly and visibly in international public forums, casting a
shadow on the regime’ reputation.
On the international level, another trend had repercussions in Rwanda. Towards
the late 1980s, a wave of democratisation had begun sweeping the African continent, a
political development welcomed and encouraged by the international community. 21
Indeed, the international community was eager to see the Habyarimana regime
democratise, though the regime had long been reluctant to open the country to democratic
competition. Rwanda finally reached a turning point in June 1990 during a France-Africa
summit in La Baule, France.
There, Habyarimana was pressed by one of his most
prominent allies, French President François Mitterrand, to democratise the country.
According to Prunier, “President Mitterrand, who was then on a ‘liberal’ political course
concerning African affairs and who seemed to want to link economic aid with political
democratisation advised Habyarimana to introduce a multiparty system in Rwanda.” 22
Habyarimana’s initial reaction was not a warm one. In an article titled “Multipartyism is not
necessary for Rwanda to be a democracy,” he is quoted as saying “Africans should not
listen to imperialists defending multipartyism, source of divisions and, in Rwanda, source
20
Emmanuel Nkunzumwami, La tragédie rwandaise, p. 102.
On this trend, see for example Naomi Chazan, “The New Politics of Participation in Tropical Africa,”
Comparative Politics, January, 1982, pp. 169-189 or Bruce Baker, “The Class of 1990: How Have the
Autocratic Leaders of Sub-Saharan Africa Fared under Democratisation?,” Third World Quarterly, vol. 19, no.
1, 1998, pp. 115-127.
22
Gérard Prunier, The Rwanda Crisis, pp. 89-90.
21
189
of weakness for the ‘majority people’.” 23 Within weeks, urged to comply, Habyarimana
finally gave into pressures and agreed to begin democratising. In doing so, he also opened
Rwanda to intense political competition among elites at a time his own party was
weakened by the unfolding crisis.
On the economic front, things were unraveling for the regime in Kigali as well,
despite the fact that it had managed to fare relatively well for a few years. Rwanda had
experienced a degree of economic growth between 1974-1981, thanks to good weather
conditions, favorable terms of trade for its main exports, but mostly because of the
increase in international aid. Rwanda was receiving close to double the amount of other
African countries. 24 It ranked, for example, as the top recipient of Swiss public foreign
aid. 25
Good economic performance had, in turn, helped stabilise Rwanda and ease
communal tensions. 26 By the early 1980s, however, Rwanda’s economic situation rapidly
began deteriorating. The economy had reached its ceiling and the trend had reversed: the
country’s economy was plummeting.
One factor triggering the economic crisis was the growing corruption of the regime.
It was proving to be a strain on the economy. As the political situation deteriorated, the
cost of keeping supporters rose. As Prunier argued: “peace could only be […] maintained
through sufficient financial lubrification of the élite.” 27 As with many other corrupt regimes,
a small clique had managed to amass fortunes. The situation had reached a point where
land and real estate in Rwanda was concentrated in the hands of a few individuals within
or with close ties to the regime. 28
23
The article was printed in Kangura in June 1990. Quoted in Bernard Mulinda, La liberté de la presse au
Rwanda (1974-1994), licence thesis, Faculté des lettres, Université nationale du Rwanda, Butare, Rwanda,
December 1999, p. 66.
24
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 139.
25
Gérard Prunier, The Rwanda Crisis, p. 81.
26
In certain regions ‘ethnic’ tensions had started dissipating and assimilation between groups had even begun.
27
Gérard Prunier, The Rwanda Crisis, p. 82.
28
Emmanuel Nkunzumwami, La tragédie rwandaise, pp. 104-105.
190
While the few accumulated wealth and land, the situation for the Rwandese
population, of which 95 percent lived in rural areas and 90 percent relied on agriculture to
sustain themselves, became critical. 29 For a people dependent on agriculture an unfolding
land scarcity crisis was turning into a nightmare. 30
recurring problem in Rwandese history.
Land issues had always been a
Already in past centuries they had been the
source of quarrels between pastoralists and agriculturalists. The new crisis, however, was
triggered by a demographic boom in Rwanda. From a population estimated at one to two
million between 1900 and 1950, it had grown to close to 8 million by the 1980s. 31
Peasants had initially found ways to deal with the growing scarcity.
They converted
pastures, forests and wetlands into cropland and migrated to less populated areas. 32 By
the 1980s, however, little land remained to be reclaimed and the migration fluxes had led
to a balanced repartition of the population throughout the country. Rwanda had reached
one of the highest population densities in the world. As a consequence, the size of farm
holdings was rapidly declining. While an average plot for a family measured 2 hectares in
1960, it was reduced to an average of 0.7 by the 1990s. 33 The impact of this land scarcity
was dramatic. Most notably, birth rates declined sharply. According to Valerie Percival
and Thomas Homer-Dixon, this trend was the result of couples delaying marriage and
limiting their number of children. 34 The group most affected by this, though, was young
men. Many found themselves in the middle of intergenerational conflicts with their elders
29
For these statistics on rural areas in Rwanda, see Valerie Percival and Thomas Homer-Dixon,
Environmental Scarcity and Violent Conflict: The Case of Rwanda, Occasional Paper, Project on Environment,
Population and Security, American Association for the Advancement of Science and the University of Toronto,
June 1995. Available online at http://www.library.utoronto.ca/pcs/eps/rwanda/rwanda1.htm.
30
For more on the issue see Catherine André and Jean-Philippe Plateau, “Land Tenure under Unbearable
Stress: Rwanda Caught in The Malthusian Trap,” Journal of Economic Behaviour and Organisation, no. 38,
1998, pp. 1-47 and David Newbury, “Ecology and the Politics of Genocide: Rwanda 1994,” Cultural Survival
Quarterly, vol. 22, no. 4, 1999, pp. 32-35.
31
Peter Uvin, “Tragedy in Rwanda,” p. 10.
32
Ibid.
33
Ibid.
34
Valerie Percival and Thomas Homer-Dixon, Environmental Scarcity and Violent Conflict.
191
when trying to gain a plot. This resulted in climbing rates of intra-family violence. 35 A
large number, however, simply could not find land or employment on the hills, a situation
which fuelled resentment, with dreams of a stable life and marriage beyond reach.
The land crisis also affected water availability and food production. With wetlands
and watersheds reclaimed for cultivation, water resources were reduced. Furthermore,
with little land available, farmers cultivated their fields two to three times a year, rapidly
impoverishing the soil. 36 Combined with a number of important climatic shocks, these
factors led to endemic droughts and famines in the country. Peripheral regions were
particularly affected, with death tolls in certain areas rapidly mounting. But, with many of
these droughts and famines taking place outside of the regime’s privileged region, the
North, the government chose to ignore the problem. 37 The regime refused to discuss the
situation publicly and attempts were made to control foreign press coverage of the issue.
According to Prunier, during one of the famines, the Chief of Security in Rwanda told a
Belgian journalist: “[j]ournalists should not write articles which can irritate the highest
authorities.” 38 The indifference of the regime in the face of such hardships fuelled popular
discontent and further undermined the regime’s legitimacy with the population.
Regardless of the regime’s clear stance of not acknowledging the droughts and
famines, its ability to respond to such crises had greatly diminished as well. External
factors had shrunk the government’s financial resources. Much of the economic growth
the country had enjoyed in the late 1970s was dependent on generous international aid.
The situation changed. And, although aid flows to Rwanda continued, donors were
beginning to reduce the amounts transferred. 39 A harder blow came from a drop in the
terms of trade of Rwanda’s main exports, coffee and tea. This downturn affected the
35
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 145.
Valerie Percival and Thomas Homer-Dixon, Environmental Scarcity and Violent Conflict.
37
Interviewee no. 10, Kigali, Rwanda, 19 March 2004; and interviewee no. 8, Kigali, Rwanda, 16 March 2004.
38
Gérard Prunier, The Rwanda Crisis, pp. 87-88.
39
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 139.
36
192
government, reducing among other things its pool of resources for bribes and indulgences,
often drawn from coffee revenues.
It also had dramatic impacts on the population.
According to Peter Uvin, “[c]offee export receipts fell from $144 million in 1985 to $30
million in 1993. Aggregate GDP per capita decreased from $355 in 1983 to $260 in
1990.” 40 In addition, while a crisis following an earlier drop in the price of coffee in the late
1970s had been averted by compensating with revenues from tin mining, around the mideighties the terms of trade for coffee, tea and tin dropped simultaneously. 41
Facing
multiple stresses at once, Rwanda’s economy was in disarray.
In light of the unfolding economic crisis, the World Bank pressured Kigali to
introduce a Structural Adjustment Program (SAP). One was finally implemented in 19901991, imposing a sharp devaluation of the Rwandese franc. 42
It also came with the
destabilising effects these programs are known to have on national economies.
Furthermore, the SAP did not have the anticipated impact. For many analysts, it was
implemented at an inappropriate time, in the midst of a deep political crisis, and political
variables had not been taken into consideration in the design of the program. 43 To make
matters worse, according to Emmanuel Nkunzumwami, some of the aid granted in parallel
to the SAP, instead of being reinvested in the country’s economy, went to financing the
army, the intelligence sector and some party propaganda. 44
All these factors, political, economic, social and environmental, combined in the
early 1990s to create a generalised crisis for the system–a multifaceted crisis weakening a
regime that had already fallen into disrepute. The RPF had been waiting for just this kind
of opportunity. They had been following developments in Rwanda, mainly by obtaining
information from inside sources.
Two of these insiders, businessman Valens
40
Peter Uvin, “Tragedy in Rwanda,” p. 11.
Gérard Prunier, The Rwanda Crisis, p. 84.
42
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 195.
43
Peter Uvin, Aiding Violence, p. 58.
44
Emmanuel Nkunzumwami, La tragédie rwandaise, p. 104.
41
193
Kajeguhakwa and former civil servant Pasteur Bizimungu, had a clear message for the
RPF leadership: “the Rwandese political system was on the verge of collapse and any
strong push from outside would complete the process.” 45 Sensing the time was right, the
RPF launched an attack on Rwanda on 1 October 1990. As René Lemarchand explained,
“[t]he assumption, fed through rumor and self-induced optimism, was that the
Habyarimana regime was a pushover, and would quickly collapse in the wake of the
invasion.” 46
The ensuing civil war further accentuated the crisis, imposing on the country the
strain of war and displacing one tenth of the population, mostly from the North, pushing
them towards Kigali. 47 Ironically, however, it also provided the Habyarimana regime with a
solution for the deadlock it was facing: a rationale to rekindle popular solidarity in light of a
new war declared by a common enemy. The RPF attack could be used to renew popular
solidarity, and more specifically the ontological solidarity at the heart of the two Republics:
Hutu solidarity. The regime in Kigali quickly seized the opportunity and portrayed the
attack as: “a revenge of monarchist extremists from the 1960s UNAR party, driven by the
will to reinstate the monarchy and feudal system vanquished by the 1959 Social
Revolution.” 48
The Instrumentalisation of the Civil War: A Sufficient Explanation?
The instrumentalist thesis advanced by a large number of authors is helpful to understand
the 1994 Rwandese genocide. By highlighting certain broad trends that developed in the
1980s, they shed light on the reasoning that led to elite actions in the country at the time,
mainly the planning, orchestration and execution of a genocidal scheme developed to
45
Quoted in Gérard Prunier, The Rwanda Crisis, p. 90.
René Lemarchand, “Rwanda: The Rationality of Genocide,” Issue: A Journal of Opinion, vol. 23, no. 2, 1995,
p. 8.
47
On displaced populations, see Valerie Percival and Thomas Homer-Dixon, Environmental Scarcity and
Violent Conflict.
48
Emmanuel Nkunzumwami, La tragédie rwandaise, p. 107.
46
194
ensure all would take part in the killings, so that “[t]hat way the blood of the genocide
would stain everybody. There could be no going back for the Hutu population; Rwanda
would be a community of killers.” 49 As a matter of fact, background variables, political,
economic, social and environmental, did all collide to weaken the regime’s credibility,
legitimacy and hold on power. In the face of growing popular discontent, the situation
unfolding inherently threatened the regime and its benefactors. As Faustin Rutembesa
argued, to succeed in acquiring or maintaining a certain social power, be it limited,
incumbents have to be promoters of a group in the population, to have a certain popular
base. 50 On this front, the regime was in a critical position as it had been losing ground.
The situation had to be turned around. The crisis faced by the regime had reached such
proportions that resolving it could only be accomplished through drastic measures.
The RPF attack on 1 October 1990, and the looming civil war, provided an
opportunity to overshadow problems in the country and to channel popular resentment and
anger away from blaming the regime. With a war about to erupt, the government was in a
position to reassert itself as the legitimate authority to respond to it. The unfolding civil war
could be used as a tool to rekindle popular support. The regime in Kigali opted, however,
to work on an unwavering form of solidarity, blood solidarity, Hutu solidarity, by devising
and implementing a plan to further accentuate divisions among the Rwandese groups.
This plan would allow them to rally the most important segment of the population, the Hutu
community, around ‘historical’ bases for its solidarity, its age old struggle against the Tutsi.
Under cover of this claimed fight against a renewed Tutsi feudal plot, in a situation of war,
with a population already made vulnerable by the hardships brought about by the
multifaceted crisis, the regime also had an excellent opportunity to feed its constituency’s
49
African Rights. Rwanda: Death, Despair and Defiance, Revised Edition (London: African Rights, 1995), p.
XX.
50
Faustin Rutembesa, “Contraintes économiques coloniales et développement des conflits,” in Faustin
Rutembesa, Josias Semujanga and Anastase Shyaka, eds., Rwanda: Identité et citoyenneté (Butare: Éditions
de l’Université Nationale du Rwanda, 2003), p. 97.
195
fears, to instill a climate of alarm and mistrust, in which it would be given the latitude to
proceed with arbitrary arrests, denunciations, and even the elimination of disturbing
elements, who were often political detractors of the regime more than real military
enemies. All this was achieved through the construction of appeals for mobilisation based
on a credible threat, a scapegoat, one that could be easily tied to the RPF. They chose as
the main character the Tutsi feudal oppressor, the one that had been a central figure of the
Rwandese ideological repertoire for decades already.
The instrumentalist thesis tells the story of elites and extremists behind the state
apparatus and of how their actions between 1990 and 1994 served to catalyse the
violence of the genocide. In all the genocide’s complexity, however, it does not tell the
population’s story: why people gave into appeals. After all, as Mamdani argued, “[t]he
violence of the genocide was the result of both planning and participation.” 51 If we want to
understand the Rwandese genocide in its entirety, and more specifically its main
characteristics, the overwhelming participation by the masses in acts committed at the
time, the instrumentalist account is insufficient. To understand this phenomenon, it is
necessary to look at the popular motives in parallel to what elites presented them with to
determine if they believed what they heard and if that is the case, to understand what
made them receptive to appeals.
It is key to one’s understanding of the 1994 events to take into account that the
actions of certain extremist elites benefited from a ‘favorable environment’ for effective
mobilisation: a population sensitized to the ethno-centric and divisive ideology upheld by
successive regimes in Kigali for decades and passed on to the masses. The story of the
genocide does not start in 1994. It is not the result of a popular murderous frenzy. It also
seems unlikely that the story of this popular mobilisation began in 1990. Appeals for
51
Mahmood Mamdani, When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda
(Princeton, N.J.: Princeton University Press, 2001), p. 7.
196
mobilisation need to be convincing to be successful and a population can hardly be easily
convinced to join in a plan of mass murder. What seems more plausible is that, in 1990,
something already laid latent, ready to be worked on by elites: the well ingrained ideology
developed by authorities between 1959 and 1990, an ideology of demonisation and
dehumanisation of the Tutsi, of the normalisation of violence and impunity, of the travesty
of morals and values. 52
These years of indoctrination amounted to a long process of
socialisation that reconfigured the population’s ideological and normative background and
created a matrix of thoughts and behaviour in which violence could now fall within the
realm of the justifiable and even of the acceptable, in which the Tutsi could plausibly be
constructed as dangerous. This matrix, in turn, served in 1990 as a formidable resource
from which to build justified and resonant mobilising appeals.
This dual account, the interactive process between elite actions and popular
background, makes better sense of the 1994 genocide. As a matter of fact, elites did not
invent new narratives at the onset of the civil war. Nothing new was heard in the public
rhetoric surrounding the civil war and genocide. References were mostly made to old
concepts already in use in the preceding decades: the Tutsi depicted as feudalists,
monarchists, enemies of democracy, of the 1959 Revolution and, therefore, of the
Republic. The bases of the ideological discourse were the same, but a new sense of
threat was infused into the discourse.
The ethno-centric and divisive ideology was
reactivated for the occasion but through the prism of urgency and danger in the context of
the civil war. The war, with all the destabilisation and fear it brought, acted as the final
element to be included in an equation resulting in genocidal violence. As François Delor
explained:
[i]n situations of social threat, diverse forms of identity hardenings (raidissements),
offensive and defensive, develop. From that point on, there is an impermeability
between people and groups that imposes itself as the reading key to differentiate,
52
Interviewee no. 8, Kigali, Rwanda, 16 March 2004.
197
define, circumscribe and exclude. You then see identity mobilised to destroy the
other. 53
The new climate of war and fear thus served as the justification to reawaken ideological
references to an Other. First it was the RPF. But rapidly it became the Tutsi, treated for
decades as different, historically threatening but tolerable. All these old beliefs, for long
instilled as a latent undercurrent, were now reawakened in a new light. The Tutsi was an
enemy, menacing and in need of elimination. This shift in rhetoric creating a sense of
threat, attributing the danger to an entire group, the depictions in terms of a zero-sum
conflict (us or them), did not only render genocidal violence plausible, but defined it as
necessary.
Early Days
The regime and extremists chose to re-enact this macabre tale of ethnic confrontation. In
this enterprise, each new year between the attack in 1990 and the genocide in 1994 saw
new events and the development of new strategies to spread anti-Tutsi views and instill
them through propaganda and fear. The situation in Rwanda spiraled down until the start
of the massacres when this scheme achieved such levels of internal division and hatred
that the populations were coaxed into taking part in the work to be done, eliminating the
Tutsi. 54
From the start of this descent towards genocidal violence, beginning with the RPF
attack, political entrepreneurs were quick to start elaborating a tale of ethnic cleavages.
They used, as in the past, Manichean imagery to reinforce their message, careful in their
construction of the RPF as the menacing ‘other’ to catalyse popular antagonism and their
53
François Delor, quoted in J. Nizurugero, “Question des réfugiés: argumentaire des régimes successifs et
leur connotation identitaire,” in Faustin Rutembesa, Josias Semujanga and Anastase Shyaka, eds., Rwanda:
Identité et citoyenneté (Butare: Éditions de l’Université Nationale du Rwanda, 2003), p. 163.
54
The word ‘work’ had been associated with the action of killing in previous historical upsurges of violence.
See Christine L. Kellow and H. Leslie Steeves, “The Role of Radio in the Rwandan Genocide,” Journal of
Communication, vol. 48, no. 3, 1998 p. 120.
198
depiction of the stake: safeguarding the Hutu Republic. Within weeks of the attacks, the
regime claimed to have been the victim of an attack by the Ugandan army. 55 It depicted
the events as an invasion by an armed force intent on destabilising the country, but, more
importantly, by an armed force with support from strong allies, well trained during the
Uganda conflict to the ways of war, and, as such, not to be underestimated.
Many
accounts at the time insisted on the foreign nature of the RPF, presenting them as
mercenaries from as far as Malawi or Zimbabwe, but more often than not, insisting on the
Ugandan connection; these were Ugandans claiming to be Rwandese but in reality simply
posing as Rwandese in order to take over. However, another narrative imposed itself, one
that did not deny the RPF’s ties to Rwanda, particularly to the Rwandese in exile.
According to these other accounts, the Tutsi that had left the country in 1959 and the
following years had returned and attacked.
These were the same Inyenzi, the
cockroaches, a pejorative nickname that resonated with historical connotations. Inyenzi
had in effect been the name for groups of refugees staging recurrent attacks on Rwanda in
the 1960s because they tended to attack at night. 56
As the comparison further developed, RPF members were presented as the Tutsi
that had left the country at the time of the 1959 Social Revolution because of their
attachment to the monarchy or as descendants of these diehard monarchists. They were
therefore members of the aristocracy that had been in power before the Revolution and
had oppressed the Hutu masses. As such, and since they had reverted to the use of
force, their intentions could only be to reinstate the monarchy and their Tutsi domination
once more. What had been feared for decades, what the ‘courageous’ Hutu masses had
tried to prevent by keeping the memory of Tutsi oppression alive during all these years,
was coming true. By taking arms again, illegally attacking a democratic and free nation
55
56
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 155.
Gérard Prunier, The Rwanda Crisis, p. 368.
199
well on the road to development, they were a group seeking to destabilise “the peace we
had known [following the Revolution]; tranquility, social peace, peace between ethnic
groups. And now they come to stir trouble.” 57
In light of this, there lay the need to
mobilise against them. These invaders were threatening what had been achieved over the
decades, the Hutu struggle to free themselves from the bonds of monarchical servitude.
The population needed to stand strong against these invaders but also to stand behind the
force that could defend them, the democratic Hutu regime of Habyarimana, natural
defender of Hutu rights, ablest at opposing a group of Tutsi so obviously intent on
destroying the Hutu.
In the face of a Tutsi threat from the outside, links could, however, also easily be
made with the Tutsi population in Rwanda. There were filial ties between the Tutsi inside
and outside the country, some members of refugee families had remained in Rwanda
while others left for exile. But the regime also upheld the existence of stronger ties: the
natural allegiance of the Tutsi to their group, blood ties running deep. In the regime’s
rhetoric following the attack, insinuations of a greater Tutsi plot surfaced, the existence of
RPF accomplices living within the country, the principal reason for suspicion being
‘Tutsiness.’
With the previously ingrained ethno-centric and divisive ideology, such
insinuations could resonate with the population’s common normative background.
While at first, following the RPF attack, some Tutsi might have felt hopeful that this
armed force would come and make things better for them, many instead feared the
regime’s reaction, particularly upon hearing the rhetoric Rwandese authorities chose to
adopt. 58 These fears were confirmed on 4 October 1990. On that night, the regime began
its strategy to instill fear in the population and to foster a sense of threat by staging an
attack by the ‘invaders’ in the street of Kigali. Gunshots were fired by the Forces armées
57
An interviewee remembering the type of rhetoric overheard at the time. Interviewee no. 8, Kigali, Rwanda, 16
March 2004.
58
Interviewee no. 23, Kigali, Rwanda, 1 April 2004.
200
rwandaises (FAR) and were later framed as an attack by the RPF. This set up, in turn,
served to justify the arrest of thousands of alleged government opponents, mostly Tutsi.
The large number of Tutsi arrested on suspicion of complicity following the incidents were
imprisoned at the National stadium, some for months, in terrible conditions, with little or no
food and medication. During their imprisonment, many were tortured, some killed and
women raped.
Despite this important dramatisation and the civil war unfolding, the
rhetoric developed by the authorities had in these first months a limited impact on the
majority of peasants in the country. It mostly reached elites and urban populations. 59
These new developments, combined with the previously unresolved problems, put
increased pressure, however, on the government to initiate a process of democratisation in
Rwanda. This new multi-party system paved the way for a more sweeping, engulfing,
mobilisation enterprise that would finally reach the rural populations. As Kimonyo argued,
and from what the author learned in interviews conducted in Rwanda, “everywhere, most
people identified the emergence of political parties as the moment at which there was the
resurgence of an ethno-centric and ideological antagonism against the Tutsi.” 60 Initially,
however, none of the new parties necessarily claimed Hutu and Tutsi affiliation and the
Rwandese held hopes that a more democratic system would bring about positive changes
in the country.
On 10 June 1991, the regime presented the population with a new constitution
allowing for the existence of opposition parties. Political groups quickly organised and the
MRND planned its survival strategy in this new political game. As if to confirm its intention
of playing fairly, the MRND even renamed itself the Mouvement révolutionnaire national
pour le développement et la démocratie, or MRND(D). 61 New parties started to emerge.
The first instituted was the Mouvement démocratique républicain (MDR). Though taking
59
Interview with Jean-Paul Kimonyo, Kigali, Rwanda, 16 March 2004.
Personal translation. Ibid.
61
Gérard Prunier, The Rwanda Crisis, p. 126.
60
201
the name of the party originally behind the 1959 Social Revolution, the MDR-Parmehutu,
the new party abandoned the ‘Parmehutu’ acronym, claiming that, though true to the
legacy of the Revolution, unlike its predecessor, it sought reconciliation and peace. 62
Other parties soon followed, including the Parti social démocrate (PSD), the Party libéral
(PL) and the Parti démocrate chrétien (PDC). There also emerged, in parallel with these
larger ones, a flurry of smaller parties with various platforms, from environmental interests
to the representation of the Muslim Rwandese. As of July 1991, most parties had been
created.
The difficult process of including these parties within Rwandese political
institutions could begin.
On 13 October 1991, Habyarimana, without consultation, named a member of the
MRND(D), Sylvestre Nsanzimana, as Prime Minister. 63 This approach to democratisation
strongly displeased the new opposition parties.
The new parties organised
demonstrations for months to contest this nomination. Facing growing discontent, the
regime began negotiating a more concerted transition with the opposition parties, and on 2
April 1992 Habyarimana named Dismas Nsengiyaremye of the MDR as the head of a new
Transition Government. In his new position, Nsengiyaremye was tasked with instituting
the new government in consultation with all opposition parties. 64
Having been developed and instituted mainly by members of the old state party,
the new multi-party system remained inherently biased. 65 It continued to be dominated by
the MRND, a political movement still highly structured, particularly in comparison to
emerging parties.
The political arena was also becoming rife with intense political
competition between strong personalities. The MRND used this to blame the new parties
for being the source of problems and division in the country. 66
62
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 156.
Ibid.
64
Ibid., p. 158.
65
Interviewee no. 8, Kigali, Rwanda, 16 March 2004.
66
Ibid.
63
202
Furthermore, these
conflicts between political elites and parties increasingly turned towards sectarianism,
especially with the MRND so strong within the system and promoting divisive rhetoric in
parallel to the war. The regime and extremists in Rwanda had, in effect, gained a new
public platform to disseminate their propaganda. Ironically, by playing by the rules of the
new system they developed the ultimate tool to promote their views: by creating a new
party expressly for that purpose.
The creation of the Coalition pour la défense de la République (CDR) in March
1992 was a strategic move of the part of the MRND to resist the forces of democratisation
by agitating the threat of war and dissension in the country. It was “an openly racist party
[…], a limited political organisation but extremely violent. The CDR wanted the creation of
a common Hutu front that would fight by all means necessary all the Tutsi as well as Hutu
collaborating with the Tutsi.” 67 Many saw this as a scheme on the part of the clique of
extremists behind the Habyarimana regime to develop an alternate medium for public
hateful rhetoric, which the MRND, to maintain a semblance of international credibility,
needed to refrain from.
One Rwandese interviewee recalled campaigns by the CDR
where individuals aboard a motorised vehicle shouted in megaphones: “I swear on my
mother’s head, the CDR is the best party and there is only one enemy: it is the Tutsi.” 68
This arrival of the CDR on the political scene dramatically changed the language tolerated
within the new political system.
The result of the democratisation process in Rwanda, therefore, was the opening
up of a public space for opinions after decades of governmental control. Well aware of
this, the new parties seized the opportunity to express their views in public appearances.
The larger new parties also turned to other means for promoting their platforms. All of
67
68
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 161.
Interviewee no. 20, Kigali, Rwanda, 29 March 2004.
203
them equipped themselves with party newspapers, for example. 69 In the few months and
years following the start of the democratisation process, the population was bombarded
with party propaganda and overwhelmed by constant campaigning and rallies. People
were given tee-shirts and hats bearing the colours of new political movements. Personal
political opinions became hot topics of conversation in social gatherings, restaurants and
cabarets. The new space, however, was opened to a range of political ideas, a range in
which racist and divisive opinions also had their place. With this newly acquired right of
freedom of speech, corrosive ideas could also rightfully be expressed.
And, with the
arrival of the CDR, this is exactly what started to happen. Campaigns radicalised, leaving
citizens to feel that “things really started heating up.” 70
Competition for posts and party standing within the democratising political arena
was becoming fiercer. At the time, the President of the PSD, Félicien Gatabazi, described
the growing tensions and violence of the period in these words: “[e]ach time there are
some difficulties (in the democratic process) there is a flare-up of tribal violence instigated
by the regime, and threats of civil war are used to justify the status quo.” 71
For the
population, the new multi-party system was turning out to be a source of renewed ethnic
divisions instead of marking the beginning of a new democratic era. Some comments by
people interviewed by African Rights are telling. One stated that: “[m]ultipartyism brought
confusion and violence.” 72 Another explained that: “[t]here was no real peace after 1992,
because of the activities of political parties inciting hatred and general concern about the
RPF advance.” 73 A female interviewee stated that: “[t]hese troubles were caused by the
political parties that had just been formed. All they did was incite ethnic violence between
69
Gérard Prunier, The Rwanda Crisis, p. 131.
Interviewee no. 26, Kigali, Rwanda, 5 April 2004.
71
Quoted in Gérard Prunier, The Rwanda Crisis, p. 144.
72
Quoted in African Rights, Tribute to Courage (London: African Rights, 2002), p. 243.
73
Ibid., p. 245.
70
204
Hutus and Tutsis.” 74 Finally, a woman from the sector of Ngoma explained that: [t]hese
political parties created a lot of tension in Mbogo in 1992. They claimed that the Tutsi RPF
had killed all the Hutus in Byumba and so the Hutu should take revenge.” 75 Not all parties
held such trenchant views, but ethno-centric divisionism did make a remarkable entrance
on the political scene.
In parallel with the growing resurfacing of anti-Tutsi rhetoric on the national political
scene, other forms of authority were also mobilising in favour of this more extreme
tendency and beginning to make public appearances in support of it. Certainly one of the
most infamous examples of this trend is the incendiary speech given on 22 November
1992 by Léon Mugesera, a renowned linguist affiliated with the MRND. Reactualising the
old Hamitic myth, he invited the Hutu to send the Tutsi back to their homeland, Ethiopia,
via a shortcut, the Nyabarongo River which flows northwards and feeds the Nile. 76 At the
local level too, authorities joined in the effort, as the local administration remained very
close to the presidential party’s structure. They organised rallies and demonstrations,
even incited anti-Tutsi violence, to continue to play their role as intermediaries to the
masses for messages emanating from the capital. Even the Church played its part, though
not through its actions but instead lack thereof. As Timothy Longman stated, “[t]his silence
confirmed for most Rwandans the widely held assumption that church leaders were
themselves biased against Tutsi.” 77
With growing numbers joining the rank of propagandists and idéologues, the antiTutsi rhetoric was overwhelming.
The nature of messages started changing as well.
74
Ibid., p. 244.
Ibid., p. 231.
76
Samantha Power, “A Problem from Hell”: America and the Age of Genocide (New York: Harper Collins,
2002), pp. 339-340. Gérard Prunier reproduced in his book Mugesera’s speech: “[...] We have to take
responsibility in our own hands and wipe out these hoodlums. [...] The fatal mistake we made in 1959 was to
let them [the Tutsi] get out. [...] They belong in Ethiopia and we are going to find them a shortcut to get there by
throwing them into the Nyabarongo River. I must insist on this point. We have to act. Wipe them all out!”
Gérard Prunier, The Rwanda Crisis, p. 172.
77
Longman explained that higher echelons of both the Catholic and Protestant churches in Rwanda were held
by Hutu, even if “Tutsi were actually disproportionately represented among both Catholic and Protestant
clergy.” Timothy Longman, “Christian Churches and Genocide in Rwanda,” pp. 8-9.
75
205
Particularly with the arrival of the CDR on the political scene, the threat rhetoric, focused at
first on the RPF, was beginning to taking on a larger breadth. Through shifts in the
semantics used, the general Tutsi population had gone from suspected accomplices in an
anti-Hutu plot to enemies, ‘enemies within.’ 78 One of the most obvious examples of this
shift came from a report titled Definition and Identification of the Enemy published by a
Commission of highly ranked army officials. According to this document, the enemy was
“the Tutsi inside or outside the country, that are extremists or nostalgic for power, who
have never recognised... the realities of the Social Revolution of 1959 and who want to
take power in Rwanda by any means, including by force.” 79 The enemy was the one who
wanted to bring back the monarchy, the corvées, the oppression, and draw the Hutu back
into servitude.
More and more, the Tutsi were depicted as monarchists, feudalists,
threatening the majority in Rwanda, while the Hutu should stand united to defeat a
common enemy, eliminate the threat, to stay true to the nation’s republican values and to
their ancestors that had valiantly fought for the Revolution.
To build this characterisation of the group as the enemy, the extremists also
referred to vile characterisations and old stereotypes attributed to the Tutsi. 80 They were
described as evil, cunning, not trustworthy, especially in light of their powerful attachment
to their kin, which they inherently saw as superior. This last attribute was used to portray
them as haughty and spiteful with the Hutu. A poem of the time intended to represent the
words and thoughts of the Tutsi was even entitled ‘I hate the Hutu.’ 81 This purported
‘natural’ hatred of the Tutsi for the Hutu served to further ground their categorisation as
enemies. After all, it could be claimed that their inherent tendency to want to oppress
those they despised was simply reasserting itself.
78
Peter Uvin, “Prejudice, Crisis and Genocide in Rwanda,” African Studies Review, vol. 40, no. 2, 1997, p.
102.
79
Ministère de la défense nationale, quoted in African Rights, Rwanda, p. 39.
80
Interviewee no. 27, Kigali, Rwanda, 13 April 2004.
81
African Rights, Rwanda, p. 75.
206
An interesting feature of the reactualisation of these depictions of the Tutsi is that
physical stereotypes began to resurface, a fact not without significance considering the
importance physical traits would take in the selection of who got killed in the absence of
proper identification when crossing roadblocks during the genocide.
This physical
stereotyping often insisted on the strength of the Hutu as one of his finest physical features
in comparison to the slender, aristocratic Tutsi.
An interviewee even recalled that it
became an issue during the national competition for the selection of Miss Rwanda. During
the competition, many Hutu took a stand against the common view that Tutsi women were
more beautiful than their Hutu counterparts. According to her recollection, some said:
“why do we always look for Miss Rwanda among the Tutsi. Our girls are pretty too. They
have a nose with which they can breathe easily. They have strong legs.” 82
Spiraling Down Towards the Genocide
The picture painted of the Tutsi as a plausible enemy was, therefore, a complete one, from
his plans of destruction, to his evil nature, even to his physical denigration.
It was,
furthermore, supported by a powerful arrangement of the actors in play. According to
Kimonyo, it is probably around this time, towards the end of 1992, that plans for the
genocide were developed. 83 The following years, 1992 and 1993, were a period when
Rwandese society spiraled towards radicalisation. Numerous new schemes were
developed by extremists to stir up tensions. Particularly effective were the militias, the
infamous Interahamwe and Impuzamugambi.
82
The Interahamwe were created in April
Interviewee no. 22, Kigali, Rwanda, 1 April 2004. It is important to note that Tutsi women were often the
target of the extremist hateful rhetoric, starting with the infamous Ten Commandments published in Kangura, a
radical Rwandese newspaper, in December 1990. The first three of these commandments stated that: “1.
Every Hutu should know that a Tutsi woman, wherever she is, works for the interests of her Tutsi ethnic group.
As a result, we shall consider a traitor any Hutu who: marries a Tutsi woman, befriends a Tutsi woman,
employs a Tutsi woman as a secretary or concubine. 2. Every Hutu should know that our Hutu daughters are
more suitable and conscientious in their role as women, wife and mothers of the family. Are they not beautiful,
good secretaries and more honest? 3. Hutu women be vigilant and try to bring your husbands, brothers and
sons back to reason.” Reproduced in Samantha Power, “A Problem from Hell,” pp. 338-339. They were also
the target of very specific violence during the genocide, such as humiliating sexual tortures and mass rape.
83
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 164.
207
1992 by the MRND, a strategy quickly followed by the CDR who created its own
Impuzamugambi. 84 At the start, the Interahamwe were formed from the youth wings of the
MRND. These youth wings were composed of very devoted young men who served as an
activist force within the presidential party. In light of recurrent hostilities with the RPF and
of the insufficient number of troops in the Forces armées rwandaises, the regime claimed it
needed to compensate by organising these youths as militias. 85 With the help of French
troops, it provided them with training, weapons and colourful uniforms, uniforms first
ridiculed somewhat by the population, but soon feared by them. 86 As a matter of fact,
once trained, the militias were dispatched to communities where they worked to sensitise
the population to the ‘Tutsi threat,’ not always pacifically, at times “inciting the Hutu to
violence” or terrorising the inhabitants of these communities. 87
In parallel, other groups were developed as well. The existence of one of them, the
Zero Network, was revealed in October 1992 by Filip Reyntjens in a press conference.
This group, whose leaders for the most part also belonged to the Akazu, is described by
Prunier as “a death squad [developed] on the Latin American model. […] It was made up
of a mixture of off-duty soldiers and MRND(D) militiamen who were given weapons by the
army.” 88 It was reported to have played a part in some of the massacres that took place
between 1992 and 1994, as well as in political assassinations. Another group called the
Alliance de militaires agacés par les séculaires actes sournois des Unaristes (Alliance of
soldiers annoyed with the tricky persistent acts of the Unarists, or AMASASU, which in
Kinyarwanda signifies bullets) was a secret society within the army. On 20 January 1993,
it sent a public letter signed by a certain ‘Commandant Mike’ to Habyarimana insisting that
to attain peace, war was needed. The AMASASU declared that they could not be stopped
84
Organisation of African Unity, Rwanda: The Preventable Genocide, Report of the International Panel of
Eminent Personalities, July 2000, p. 56.
85
Interviewee no. 22, Kigali, Rwanda, 1 April 2004, Interviewee no. 24, Kigali, Rwanda, 2 April 2004.
86
On the French involvement, see Gérard Prunier, The Rwanda Crisis, p. 165.
87
African Rights, Tribute to Courage, p. 59
88
Gérard Prunier, The Rwanda Crisis, p. 168.
208
from punishing traitors living within Rwanda; they would strike hard when the RPF
attacked. 89 With the militias, Zero Network and the AMASASU, the regime had powerful
forces to terrorise the country.
The militarisation of Rwandese society was also pursued by other means thanks to
the dramatic increase in the military budget approved in January 1992. 90 Budget increases
were accompanied by a significant rise in French aid allotted to the Rwandese military, still
very much under MRND control. These new financial resources, but also weapons and
ammunition, “sent on direct orders from the highest circles at the Élysée,” allowed for an
increase in the number of troops in the army, which actually grew from the 5000 troops in
existence before the RPF attack to approximately 35 000 men. 91 Furthermore, the regime
distributed weapons to civilians and offered them military training. In September 1992 and
again in February 1993, guns were handed out in certain communes. 92
The trend
accelerated. According to Alison Des Forges, judging firearms to be too costly for mass
distribution, the extremist clique around Habyarimana “advocated arming most of the
young men with such weapons as machetes.
Businessmen close to Habyarimana
imported large numbers of machetes, enough to arm every third adult Hutu male.” 93
Between January 1993 and March 1994, close to 581 000 machetes found their way into
the country. 94
Another indication of the radicalisation of society was the growing number of
massacres, mainly of Tutsi, taking place across the country. These seemed to follow the
89
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 163.
Organisation of African Unity, Rwanda, p. 56.
91
Gérard Prunier, The Rwanda Crisis, p. 148; African Rights, Rwanda, p. 48
92
Organisation of African Unity, Rwanda, pp. 56-57.
93
Human Rights Watch, Leave None to Tell the Story: Genocide in Rwanda (New York: Human Rights Watch,
1999), p. 5
94
Samantha Power, “A Problem from Hell,” p. 337. Philip Verwimp, using data obtained from IBUKA, a
genocide survivor association, studied the probabilities of being killed by firearm or machete. According to
Verwimp, the odds of getting killed by a traditional weapon, a machete for example, depended on the location
of the victim, if the victim was killed in a grand scale massacre, the victim’s age, and the number of days after
the start of the genocide the victim was killed. See Peter Verwimp, “Machetes and Firearms: The Organization
of Massacre in Rwanda,” Journal of Peace Research, vol. 43, no. 1, 2006, pp. 5-22.
90
209
logic of previous massacres under the Kayibanda regime, which had traditionally taken
place after military actions by the refugees in neighbouring countries. The population was
targeted following actions by the RPF to discourage the RPF from undertaking military
operations or from continuing their advance on the country. One such massacre occurred
on 27 January 1991 in Kinigi, where 500 to 1000 Tutsi were killed apparently to retaliate
against the RPF who had taken control of Ruhengeri. 95 In the following days, the violence
spread to other communes and continued until mid-March. Despite an end to the mass
killings, the Bugogwe population, a Tutsi group living in the North-West, continued to be
persecuted and to be the target of killings in the following months. These violent upsurges
intensified around 1992 when massacres became more common, many beginning to be
reported by human rights organisations such as Human Rights Watch. An important one
took place around that time in the Bugesera region. In March 1992, many communities in
this region south of Kigali witnessed the massacre of hundreds of Tutsi and Hutu members
of the opposition. 96
By June 1993, Human Rights Watch was reporting that, since the beginning of the
year, instances of mass killings had occurred almost monthly across the country. Many of
these had been “carried out by civilians with government support.” 97 In certain cases, the
population was reported to have taken part in these violent acts encouraged by authorities
telling them to take revenge for atrocities committed by the RPF against the Hutu. In an
interview with African Rights, a woman described this tactic: “[her] commune was a scene
of killings in 1992. The Hutus killed the Tutsis, claiming the RPF was massacring Hutus in
Byumba.” 98 In retrospect, many interviewees believed these massacres conducted before
the 1994 genocide were trial runs for the extremists’ plan to annihilate the Tutsi.
According to Prunier, these massacres did share all of the characteristics of the killings
95
Gérard Prunier, The Rwanda Crisis, p. 136.
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 162.
97
Organisation of African Unity, Rwanda, p. 57.
98
African Rights, Tribute to Courage, p. 232.
96
210
that would take place during the genocide. Most would be preceded by a political meeting
organised and presided over by known local authorities. These meetings “seemed to have
been designed to put the local peasants ‘in the mood’, to drum into them that the people
they were soon to kill were Ibyitso, i.e. actual or potential collaborators of the RPF archenemy.” 99 These meetings were soon followed, a few days later, by an order emanating
from higher authorities to begin the ‘work,’ or what the bourgmestre usually called ‘bush
clearing,’ the two being synonymous with killing.
A large part of this process of radicalisation was also the result of the media
diffusing heinous propaganda, particularly with the creation of the private radio station
Radio Télévision libre des milles collines (RTLM) in June 1993, which joined the growing
group of hate media already active in Rwanda at the time. 100
The circles around
Habyarimana and the radical political factions had been quick to realise the utility of the
media for their anti-Tutsi campaign. This was particularly the case with radio in a country
where oral communication was predominant, but also in which radio was the prime source
of information people relied on, and even more so to keep informed of developments in the
civil war and resulting peace talks. 101
Newspapers also played a role, although more existed and a wider range of views
was expressed in print. Pro-MRND and CDR papers battled opposition party papers, each
side defending trenchant views. Pro-MRND and CDR papers portrayed opposition party
leaders as “RPF puppets, traitors and embezzlers of public funds, demagogues,
opportunists and idiots motivated by the desire to settle scores with President
99
Gérard Prunier, The Rwanda Crisis, pp. 137-138.
On the role of the media in Rwanda, see for example, Reporters sans frontières, Les médias de la haine
(Paris: La Découverte, 1995).
101
Referring to the use of radio by political officials, Christine Kellow and Leslie Steeves indicated that “[t]o the
extent that oral tradition remains strong and illiteracy is widespread, the radio may have great impact.
According to Hachten, ‘[l]isteners tend to conceive it as literally the government itself speaking’.” Christine L.
Kellow and H. Leslie Steeves, “The Role of Radio in the Rwandan Genocide,” p. 116.
100
211
Habyarimana.” 102 Papers in support of opposition parties, on the other hand, depicted
MRND leaders as “evil and corrupt: liars, idiots, animals, bloodthirsty murderers and
warmongers.” 103 Most of the media controlled by the government or extremists or created
by them were, however, a platform for the extremists’ rhetoric and paralleled the political
agenda of the génocidaires.
A flagrant example of the use of the media to propagate these ideas came in
December 1990, soon after the RPF attack. The radical newspaper Kangura (‘Wake him
up’ in Kinyarwanda) published an editorial entitled Appel à la conscience des Bahutu,
which included the infamous ‘Ten Commandments.’ 104
This editorial, an unambiguous
attempt to frame the October attack as an ethno-centric plot against the Hutu, already
raised the issue of general Tutsi complicity based on what was described as their age old
plan of domination over the Hutu masses, a plan passed on from generation to generation.
One key statement indicated that: “[t]he Muhutu is the Mututsi’s sole enemy; the latter
must therefore hate him and hunt him down wherever he is.” 105 The Kangura editorial
further called on all Hutu to rise up against this Tutsi ploy to reinstate their monarchy and
dominate Rwanda by fighting back, using all means necessary.
While other media sources were at first more careful about adopting such ethnically
and violently connoted rhetoric, publications and radio broadcasts propagating anti-Tutsi
messages often recalling ‘historical’ facts or old prejudices, became common in the
following years. The creation of RTLM in 1993 proved to be a powerful catalyst for this
trend. This private radio station, with reported close ties to the Akazu, was according to
many created to serve as an alternate private station–controlled by circles close to
Habyarimana–which could promote extremely hateful anti-Tutsi propaganda without being
102
Jean-Marie Vianney Higiro, “Rwandan Private Print Media on the Eve of the Genocide,” in Allan Thompson,
ed., The Media and the Rwanda Genocide (London: Pluto Press/IDRC, 2007), p. 74.
103
Ibid., p. 75.
104
Filip Reyntjens, L’Afrique des Grands Lacs en crise: Rwanda, Burundi 1988-1994 (Paris: Éditions Karthala,
1994), p. 93.
105
Kangura, 10 December1990.
212
publicly associated with the authorities. It was also intended to supplant the national radio
network, Radio-Rwanda, perceived as increasingly too close to the official opposition’s
stance and too critical of the regime.
As a new radio station, RTLM’s popularity rapidly soared. People were drawn to
the songs it played and its lively entertainment. According to Prunier, even RPF soldiers
were fascinated by RTLM and some “preferred listening to it than to Radio Muhabura
(‘Radio Beacon’), their own ‘politically correct’ and rather preachy station.” 106 The new
private station was a success. As Alison Des Forges indicated,
RTLM at one point said that they estimated they reached 75 percent of the
households of Rwanda at the height of their power. [...] I remember one witness
describing how in part of Rwanda, it was difficult to receive RTLM, and so he had to
climb up on the roof of his house in order to get a clear signal, and he would stand
up there on the roof of his house with his radio to his ear listening to it, and then
shouting out to the crowd what was being said. 107
But this popularity also ensured that RTLM’s message would be relayed among the
general population. So many listened to it that broadcasts often became one of the most
popular topics of conversations among colleagues at work, between acquaintances, or
during social gatherings.
Despite the entertaining format often adopted by programs, RTLM’s broadcasts
were far from innocent. Broadcasters reminded their listeners of the RPF’s advances,
accusing them of atrocities, insisting on the greater threat posed by all Tutsi, on their plan
to oppress if not annihilate the Hutu, and on what they would have to endure under Tutsi
control, a ‘long calvary.’ 108 The Hutu therefore had to stand strong for democracy and the
legacy of the Revolution. They had to fight back. As Jean-Pierre Chrétien explained,
if we take the subjects of the RTLM, given all the programs that they broadcast, we
can see that they’re based on a double register, that of racist passion against the
Tutsi, and also the feeling of legitimacy on behalf of the majority people. [...] The
106
Gérard Prunier, The Rwanda Crisis, p. 189.
Alison Des Forges in The Media and the Rwanda Genocide Symposium Report, School of Journalism and
Communication, Carleton University, Ottawa, 13 March 2004, p. 27. The complete report is available online at:
http://www.carleton.ca/mediagenocide/documents/transcript/index.html.
108
Interviewee no. 4, Kigali, Rwanda, 11 March 2004.
107
213
essential reference from the months preceding genocide and during the massacres
was therefore that of the majority people. The legitimacy of their self defense against
a clique, a feudal clique, so therefore the normalcy of the massacre by the majority
as an expression of anger, a democratic anger if you will. 109
This underlying message was constant in broadcasts to drive it into people’s minds. In the
mind of one interviewee, “the rhetoric at the time had been developed to really reach the
masses with the radio. Words said on the radio [were] like orders from the government,
everything was done to mobilise by sensitising to the idea of danger.” 110 To achieve this,
RTLM also interspersed the programs and songs it played with interviews with known
personalities and intellectuals.
These interviews were conducted in a serious tone,
contrasting with their regular comical content, to give credibility to their anti-Tutsi rhetoric.
One such interview was conducted with an important Rwandese historian,
Ferdinand Nahimana. During the program, he elaborated on the links between the RPF
and the Inyenzi, the refugees that had fled the country in 1959 and the 1960s, to point to
the existence of an age-old plot to reinstitute the Tutsi monarchy through violence, to drive
into people’s mind the clear continuity in Tutsi ploys. An interviewee also recounted how,
the day following the beginning of the genocide, one such interview was conducted with
the President of the Parti libéral and minister of commerce, industry and crafts, Justin
Mugenzi. According to the interviewee, during the broadcast, Mugenzi stated that: “[t]his is
the last war for Rwanda. It is the last war in the history of Rwanda. Hutu, you must win it.
If you lose [...] you will have to leave Rwanda. You are doomed for eternity. But if you
win, it will be eternal happiness. [...] Now it is clear, it is a war between Hutu and Tutsi.” 111
The climax of this media-based effort to turn the Hutu against the Tutsi came
following the plane crash that killed Habyarimana in April 1994, when radio stations not
109
Jean-Pierre Chrétien in The Media and the Rwanda Genocide Symposium Report, School of Journalism
and Communication, Carleton University, Ottawa, 13 March 2004, p. 20. The complete report is available
online at: http://www.carleton.ca/mediagenocide/documents/transcript/index.html.
110
Personal translation. Interviewee no. 8, Kigali, Rwanda, 16 March 2004.
111
In the words of the interviewee as he remembered the interview on RTLM. Interviewee no. 20, Kigali,
Rwanda, 29 March 2004.
214
only circulated lists of targets for killings but also fervently encouraged the Hutu population
to ‘get to work’ and cut the ‘tall grass’ or ‘trees,’ a reference to the Tutsi, often depicted as
taller than the Hutu. 112 The media rhetoric had by that time a deep impact on people’s
psyche.
In the words of Charity Kagwi-Ndungu, a trial attorney in the office of the
prosecutor during the media trials at the International Criminal Tribunal for Rwanda, “what
the Kangura Journal and the RTLM did was to stigmatize the Tutsi ethnic group within [...]
Rwanda, to stigmatize them to such a level that killing them not only became a matter of
national duty, but killing them was not like killing a human being.” 113
The Arusha Peace Process
While the media had an impact on its own, it nonetheless was used to parallel other
developments within Rwandese society. The creation of RTLM in June 1993, for example,
coincided with the last leg of the Arusha peace process set up to find a diplomatic solution
to the conflict between the RPF and the governmental forces. The process had been an
extremely intense political joust both between the parties to the conflict and among
Rwandese elites. It had fostered tensions within political spheres more than among the
general population which, felt mostly disconnected from negotiations taking place outside
of the country. The Arusha peace process had begun as a substitute for the defunct
N’Sele (the former Zaire, now the Democratic Republic of Congo) cease-fire agreement
signed on 29 March 1991 and the Gbadolite (in the former Zaire, now the Democratic
Republic of Congo) cease-fire agreement signed on 16 September 1991, both the result of
regional summits, but which had, for lack of will on the part of the two sides, been violated.
The renewal and intensification of hostilities had been hard on the belligerents, though:
“low intensity operations dragged on, without either side gaining any definite
112
Interviewee no. 21, Kigali, Rwanda, 31 March 2004.
Charity Kagwi-Ndungu in The Media and the Rwanda Genocide Symposium Report, School of Journalism
and Communication, Carleton University, Ottawa, 13 March 2004, p. 42. The complete report is available
online at: http://www.carleton.ca/mediagenocide/documents/transcript/index.html.
113
215
advantage.” 114 This situation brought the warring parties back to the negotiation table in
May and June 1992, during which time they agreed on a framework to work out a
settlement: regular talks involving key actors from both sides would be held in Arusha,
Tanzania. 115 The talks began on 10 July 1992 and resulted within days in an agreed
cease-fire between belligerents to be supervised by a small Neutral Military Observer
Group (NMOG) from the Organisation of African Unity (OAU).
As soon as the talks began, however, Rwandese political elites rapidly split into two
opposed camps, the pro and anti-Arusha factions, with the latter intent on framing the
negotiations as dealings with menacing Tutsi invaders, a betrayal of the Revolution and its
legacy. 116 As Prunier argued,
The announcement of the cease-fire [...] caused consternation among the supporters
of the extremist Hutu state. Within days the MRND(D) ministers were boycotting
cabinet meetings and demonstrations hostile to Prime Minister Dismas
Nsengiyaremye [heading the Rwandese negotiation team] had erupted in the
strongly conservative préfectures of Gisenyi and Ruhengeri. 117
Despite growing tensions, the negotiations progressed relatively smoothly and on 18
August 1992 the first protocol of the agreement was signed. This protocol on the Rule of
Law consisted of a jointly approved assessment of the conflict and drew up plans for
restoring peace in Rwanda through national unity, reconciliation, democracy, pluralism and
the respect for human rights. 118
The negotiations were not as trouble-free for the second protocol on power sharing
or for ensuing ones, however. As a matter of fact, as Kimonyo explained: “[t]he ideological
and political climate changed radically the day after the first protocol of the Accord was
signed. Following this, every period around the finalisation date for a protocol saw a flare
114
Gérard Prunier, The Rwanda Crisis, p. 135.
For an in-depth analysis of the Arusha Peace Process and its impacts, see Bruce D. Jones, “Civil War, the
Peace Process, and Genocide in Rwanda,” in Taisier M. Ali and Robert O. Matthews, eds., Civil Wars in Africa:
Roots and Resolution (Montreal: McGill-Queen’s University Press, 1999).
116
Interviewee no. 21, Kigali, Rwanda, 31 March 2004.
117
Gérard Prunier, The Rwanda Crisis, pp. 160-161.
118
For a transcript of this protocol and of the complete Arusha Peace Agreement see Emmanuel
Nkunzumwami, La tragédie rwandaise, pp. 391-461.
115
216
up of political tension and new massacres of Tutsi as a strategy by the regime to derail the
peace process.” 119 The second protocol, in particular, could only be the source of political
struggles within the country. Already the establishment of the Nsengiyaremye transition
government, composed of the five main parties, had been extremely difficult, but the
negotiations at Arusha meant that the ‘enemy’ now had to be included as an equal in a
Broad Based Transitional Government (BBTG) and in a Transitional National Assembly
(TNA), a situation hardly acceptable for the more extremist factions throughout Rwandese
political spheres. One main issue of contention was that the hardliners from the extremist
groups among the MRND and the CDR saw the team in charge of negotiations in Arusha,
under Prime Minister Nsengiyaremye of the MDR, as biased in favour of defending the
interests of opposition parties.
Their frustrations were also tied to the fact that they
resented that the talks on the allocation of ministerial portfolios were mostly conducted
without consultation of elites in Kigali. A final issue of contention was that the negotiators
had opted for a nominated Transitional National Assembly over an elected one, a
proposition that had been brought to the table by the RPF. 120 Tensions at the political
level were having an impact on the population. In Rwanda, supporters of the extremists
kept clashing with supporters of the opposition during demonstrations, engendering
renewed political violence in the country.
Habyarimana’s position during the negotiations seemed rather ambiguous. He was
increasingly perceived as too soft by the extremists, even within his party. By agreeing to
the Arusha peace process, he had alienated himself from them. He was also disregarded
by the opposition parties which had, from the start of the democratisation process, built
their platforms on opposing the President’s authoritarian state party. Caught between
these two forces, he found himself in a weakened political position, and saw hindering the
119
120
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 160.
Gérard Prunier, The Rwanda Crisis, p. 163.
217
peace process as his chance to buy time to strengthen his position. While he continued to
give his support to the negotiations, in parallel, he worked within Rwanda to undermine
them. At one point, he went as far as to call the July cease-fire agreement ‘a piece of
trash... which the government is not obliged to respect’.” 121 The comment only served to
exasperate the negotiating team in Arusha, frustrated with Habyarimana’s double stance.
The negotiations on the protocol were finally concluded, with a first section signed
on 30 October 1992 and the second on 9 January 1993. It stipulated that the Presidency
was to remain in the hands of the MRND, the Prime Minister’s post to be attributed to the
MDR and the Vice-Prime Ministerial position to the RPF. Ministerial portfolios were split
among the parties represented at the negotiations and 59 of the 70 seats in the
Transitional National Assembly were also attributed to them. 122
Soon after, however, the MRND and the CDR reacted by organising violent
protests in the streets. Between 21 and 25 January, hundreds of Tutsi were massacred in
Gisenyi and Kibuye. This renewed anti-Tutsi violence convinced the RPF to break the
cease-fire and launch an attack on 8 February 1993. Their attack brought them within 50
kilometers of Kigali, while displacing hundreds of thousands fleeing before their troops. 123
On 20 February, though only a small distance away from the capital, the RPF proclaimed it
was renewing its decision to respect the cease-fire. Many attributed this decision to the
RPF’s realisation of the negative impact their military advance had had on the civilian
population’s opinion. 124
Negotiations began again in Arusha.
The Hutu opposition parties delegated
representatives to Bujumbura, Burundi, to meet with the RPF on neutral ground.
Simultaneously, however, Habyarimana had organised a meeting in Kigali between the
121
He made the comment during a speech in Ruhengeri on 15 November 1992. Quoted in Gérard Prunier,
The Rwanda Crisis, p. 170.
122
Emmanuel Nkunzumwami, La tragédie rwandaise, pp. 157-160.
123
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 162.
124
Gérard Prunier, The Rwanda Crisis, p. 178.
218
MRND, the CDR and some of the more extremist representatives of the opposition parties.
The rift between the pro and anti-Arusha camp was growing, even within parties.
Despite the growing dissension among political elites, a third protocol on the
repatriation of refugees and resettlement of displaced persons was signed on 9 June
1993. Negotiations began on a thorny issue: the integration of armed forces. Considered
the most contentious question after power sharing, it involved determining the composition
of the new Rwandese army and the number of troops from each side to be included in it.
The regime initially proposed to the RPF to include 20 percent of its troops in the new
armed forces, a proportion the RPF found unacceptable.
They finally settled on
constituting an army composed of 40 percent of soldiers from the RPF, the remaining 60
coming from the FAR. Getting the negotiating parties to agree on the proportion for the
officer corps proved more difficult. An equal share of posts for each side was finally
agreed on with special care paid to developing its structure to ensure equal representation
of both the RPF and the FAR in the various components of the new army. They worked
on the principle that “no two hierarchically consecutive positions (i.e. commander and
second-in-command) in any given unit be held by the same side.” 125 The protocol was
finally signed on 3 August 1993 along with an additional protocol on miscellaneous matters
and final provisions.
The latter included: “the determination of the duration of the
transitional period; the timetable for the implementation of the peace agreement; the
relationship between the peace agreement and the National Constitution; and procedures
for the indictment of the President in case of violations of the peace agreement.” 126 The
next day, the Arusha Peace Agreement was officially signed by the parties at a ceremony
attended by the Heads of State of Rwanda, Tanzania, Uganda, Burundi and the Prime
Minister of Zaire.
125
Ibid., p. 193.
Ami R. Mpungwe, “Crises and Response in Rwanda: Reflections on the Arusha Peace Process,” in
Monograph No. 36: Whither Peacekeeping in Africa? (Pretoria: Institute for Security Studies, 1999).
126
219
The Arusha peace process successfully achieved a negotiated settlement. New
transitional institutions were to be set in place within 37 days of the finalisation of the
Accord and the implementation to be supervised by UN troops. Future success, at least in
its implementation, was far from guaranteed, however.
The peace process had also
served in parallel to crystallise the schism between political factions in Rwanda, positions
that had become so conflicted that a trouble-free transition seemed highly improbable.
This split, particularly flagrant among the opposition parties and encouraged by
cooptation and corruption on the part of the presidential clique, was the extremists’ final
major achievement before the genocide.
This political development, exacerbated by
tensions surrounding the Arusha negotiations and particularly the February war which had
raised doubts about the RPF’s peaceful intentions, split into moderate and extremist wings
all new parties with the exception of the PSD. The new radical front, composed of the
extremist wings of the MDR, PL and PDC, now posed as an alternative to the ‘softer’
moderate wings, as well as to the old state party. As Prunier argued, this extremist front
sought “to give the impression of a broad multi-party movement which would preach
‘common sense’ by giving a new ‘intrinsically democratic’ voice to the Rubanda
Nyamwinshi–the ‘majority people’, i.e. the Hutu.” 127 This new group, choosing to go by the
name of Hutu Power, rapidly became popular among the masses.
It is in this extremely polarised political environment that the UN supervision force,
the United Nations Assistance Mission to Rwanda (UNAMIR), and 600 RPF troops arrived
in Kigali to begin instituting the Broad Based Transition Government. Only a few weeks
after their arrival, polarisation within Rwandese society reached new levels. The CDR and
MRND joined the Hutu Power front following the assassination of President Melchior
Ndadaye of Burundi. The murder of the first Hutu President in Burundi by Tutsi elements
127
Gérard Prunier, The Rwanda Crisis, p. 183.
220
in the national army on 21 October 1993 was followed by massive civil unrest and the
exodus of some 300 000 Hutu, most crossing the Burundi-Rwanda border.
These events in neighbouring Burundi were to have a dramatic impact on politics in
Rwanda. With its blatant ethnic characteristics, the assassination seemed to lend
credence to the rhetoric of a ‘Tutsi threat’ in Rwanda developed since the start of the war.
The new extremist front could now exploit
the shock felt in Rwanda after Ndadaye’s death [...] to rally many simple or hesitant
people. They presented the situation in terms of almost biblical urgency. To the
fear of losing one’s privileges (rational level) they added the fear of losing one’s life
(visceral level) and the fear of losing control of one’s world (mythical level). 128
These events consecrated the complete ‘ethnicisation’ of political debates within
Rwandese society. 129 President Ndadaye’s assassination provided a powerful justification
for the Hutu Power’s anti-Arusha stance but more importantly for their anti-Tutsi campaign
among the general population.
The creation of a common front among all extremist factions played a fundamental
role in achieving the mobilisation of the Hutu masses. Without this alliance with the new
extremist wings of the opposition parties, the presidential party, discredited over the years
in certain regions, would not have been able to rally the masses across the country. As
Kimonyo explained,
After three years of multi-party politics, the ex-state
political control of many regions, which in turned
opposition parties, especially the most powerful of
regions, it is these opposition parties, the MDR in
agents of the popular mobilisation in the genocide. 130
128
party, the MRND, had lost the
had passed under control of
them: the MDR. […] In these
particular, that were the main
Ibid., p. 200.
Interviewee no. 8, Kigali, Rwanda, 16 March 2004.
130
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 28. According to Kimonyo what played a very
important role in this case is that in certain regions old sentiments of affiliation to the original MDR-Parmehutu
of the 1950s and 1960s resurfaced. There existed an ideological continuity among the masses between
support for the party between 1959-1963 and between 1991-1994. As he explained, “people decided to
affiliate with the MDR in the 1990s by fidelity to the MDR-Parmehutu,” p. 563. These were “family groups,
keepers of the historical legacy of the Parmehutu [:] generally [...] rural social elites [that] owed their upward
social mobility to the Revolution,” p. 569.
129
221
Without this collaboration among extremists, the genocide would have remained limited to
the strongholds of the MRND and the CDR. 131 The newly consolidated extremist forces
had the political impact to weaken and overshadow the more moderate factions, which had
lost some of their more popular politicians to the extremists and had now become a
constellation of eclectic parties facing a political behemoth, Hutu Power. There remained
few actors capable of contesting the latter’s hateful and violent agenda. The extremists
could work, almost unimpeded, to stall the implementation of the Arusha Peace
Agreement. Among some of their more important tactics were the MDR, PL and PDC’s
extremist wings’ demands that the lists of candidates for the BBTG include representatives
of the Hutu Power party wings and the CDR, despite the fact that the CDR had boycotted
negotiations around the new political institutions for months. 132
Habyarimana, still trying to find a way to regain his standing, also chose to resist
the implementation of the agreement, desperate to buy time as “a kind of survival
reflex.” 133 Time for his stalling tactics was running out, however, as international pressure
to institute the BBTG began mounting. He was ultimately confronted by his peers on the
issue on 6 April 1994 when taking part in a regional summit in Dar-es-Salaam on the crisis
in neighbouring Burundi. The talks quickly turned to the Rwandese situation, though, and
Habyarimana was strongly urged by the Heads of state of the Great Lakes region to
comply with the Arusha Peace Agreement.
History did not allow him to announce his intentions. The same evening, while
returning to Rwanda aboard his private plane, two missiles were launched from the
surroundings of the Kigali airport. The plane was shot down, killing all occupants on
board, including President Cyprien Ntaryamira of Burundi.
131
To this day, neither the
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 194.
For an insider’s view of the political tractations that went on during these last months before the genocide,
see Roméo Dallaire, Shake Hands with the Devil: The Failure of Humanity in Rwanda (Toronto: Random
House Canada, 2004).
133
Gérard Prunier, The Rwanda Crisis, p. 203.
132
222
perpetrators of the assassination, nor the interests behind it, have been officially identified,
though rumours abound. The event nonetheless went down as the trigger for one on the
most brutal genocides in history. Within hours of the plane crash, radio stations blamed
the RPF and the Tutsi for the incident and called on the Hutu population to ‘get to work.’
One interviewee recalled that the message was unambiguous: “[t]hey killed our father, now
we must also eliminate them.” 134 Within hours roadblocks had been set up by the military
and militias and the massacre began, its first victims turning out to be not only Tutsi but
also Hutu moderates opposed to the extremists. 135 The latter represented some of the last
that could have publicly stood in the way of the genocidal enterprise. In the following days,
as René Lemarchand described, “the diffuse anxiety of a return to Tutsi hegemony was
replaced by collective psychosis that the media amplified and channeled into paroxysmal
violence.
Ethnic hatred and fear lead to panic to murder.” 136
The Hutu population
massively began taking part in the genocide, whether actively participating in killings, or
controlling the fleeing Tutsi populations, hunting or denouncing the ones that were hiding,
or passively as silent observers refusing to help the Tutsi that begged for help. 137
Impact on the Population
At the heart of this research lay the difficult task of studying popular mobilisation, and, in
particular, elite strategies with respect to this process.
The construction of mobilising
appeals is, however, an intricate process, especially in complex cases like the Rwandese
134
Personal translation. Interviewee no. 21, Kigali, Rwanda, 31 March 2004.
Catharine Newbury and David Newbury, “A Catholic Mass in Kigali: Contested Views of the Genocide and
Ethnicity in Rwanda,” Canadian Journal of African Studies, vol. 33, no. 2, 1999, p. 295.
136
René Lemarchand, “Les génocides se suivent mais ne se ressemblent pas: l’Holocauste et le Rwanda,” in
L’Afrique des Grands Lacs. Annuaire 2001-2002 (Antwerp: L’Harmattan, 2002), p. 23.
137
Active participants in the genocide included the planners, high civil and military officials as well as the local
administrations, and the killers, the Presidential Guard, the militias, the Forces armées rwandaises (FAR) and
volunteers. In many regions, however, the general Hutu population, the peasants, also massively joined in the
killings. As Prunier stated, “the main agents of the genocide were the ordinary peasants themselves. This is a
terrible statement to make, but it is unfortunately borne out by the majority of the survivors’ stories. The degree
of compulsion exercised on them varied greatly from place to place but in some areas, the government version
of a spontaneous movement of the population to ‘kill the enemy Tutsi’ is true.” Gérard Prunier, The Rwanda
Crisis, p. 247.
135
223
one. Most importantly, a communicative act should not be treated as one-sided. It is an
interactive process that has both a sender and a receiver. In order to study a case such
as the strategised mobilisation in Rwanda, it was therefore necessary in the previous
section to study why and how appeals were instrumentally constructed by Rwandese
elites. But the analysis also requires an additional step, looking at the receiving end of the
process. This stands on its own as an extremely important question: how can individuals
agree with a political project or ideology that can lead to deep divisions among their
communities, even among neighbours, friends and families?
Rwanda proved an extreme case. As David Newbury stated, “the killers and victims
were often known to each other, as teachers and students, as neighbors and colleagues,
and even as family members.” 138 This raises the question of what motivate individual
action in the genocide. Not addressing these popular motives amounts to denying the
Rwandese population’s capacity for agency. While they lived in an authoritarian society,
there is no denying that the population, though constrained by the system, remained
capable of making individual political choices. These choices need to be taken into
account in any analysis of mobilisation, or of collective action around identity politics.
Failing to do so might lead to two pitfalls identified by Bruce Berman, Dickson Eyoh and
Will Kymlicka: “(i) the inclination to treat both ethnic communities and elites as
homogeneous and static; and (ii) the tendency to view the politicizations of ethnicity as
phenomena manufactured by corrupt elites and consumed by more or less gullible
masses.” 139
Taking this into account, successful appeals and, as a result, successful
mobilisation occurs when popular motives are made to converge with the goals of elites.
In the case of Rwanda, more specifically, to say that mobilisation during the genocide was
138
David Newbury, “Understanding Genocide,” African Studies Review, vol. 41, no. 1, 1998, p. 78.
Bruce Berman, Dickson Eyoh and Will Kymlicka, “Ethnicity and Democracy in Historical and Comparative
Perspective,” in Bruce Berman, Dickson Eyoh and Will Kymlicka, eds., Ethnicity and Democracy in Africa
(Oxford: James Currey, 2004), p. 3.
139
224
the result of elite manipulation of communitarian backgrounds, it is therefore necessary to
determine if the internalisation of appeals by the population did effectively take place, to
determine if the Rwandese population effectively bought into the rhetoric developed by the
regime and extremists.
Determining if that was the case is not an easy task, however, especially since, at
least in the Rwandese case, what it means to ‘believe authorities’ is not straightforward. In
interviews conducted among the Rwandese population, it became apparent that the act of
believing took on different meanings for different people. An important distinction had to
be made particularly between accepting what the authorities said, without questioning it,
thus acting upon it, and believing as in truly internalising their ideas, integrating them as
one’s convictions and acting upon them. It proved quite a conundrum in itself.
As a means to circumvent this problem, an alternative was to look at communal
and personal identities. As was discussed previously, even before the onset of the civil
war, during the 1959-1990 era, the Rwandese lived in segregation. The exception to this
was their communal and personal life which they sheltered from the oppression ubiquitous
in relations with the state. Among these community based and personal contacts, they
were able to disengage from such dynamics and pursue a more amicable, cooperative,
and undifferentiated existence.
To determine the impact of the reactivation of the decades-old ethno-centric and
divisive ideology on populations, it was possible to look at if and how those personal and
community ties were affected by the resurgence of strong and relentless ideological
propaganda following the RPF attack. Did the ideology and events of the period have an
impact on relations between community members, coworkers, neighbours and friends, or
among these inner circles? The hypothesis was that if changes were observed in the
personal and communal realms in the direction intended by the propaganda, people had
225
lent credence enough to political rhetoric to import it into their personal and communitarian
existence.
For most people interviewed, the 1990 Rwandese Patriotic Front attack did prove
to be a turning point in terms of communal relations. It marked the beginning of a change
within relationships that was accentuated by ensuing events, particularly the creation of
political parties and, at a latter date, of the Hutu Power front. While many, especially in
cities, had noticed that in previous years the tendency had been towards greater
integration among groups, according to them the trend began reversing again in 1990. 140
With the growing media and political propaganda, increasingly linking the Tutsi inside
Rwanda to the RPF, even within communities, the social climate became unhealthy.
Developments on the political scene were having their impact.
The first years of the 1990-1994 period were generally described as a time of doubt
and concern.
With little to go on to assess the situation but the media, government
rhetoric and numerous rumours of conspiracy, the Rwandese began to wonder if what they
heard was true. 141 People became wary and started to pay careful attention to what was
going on in their community to see if what they experienced on the hills did give credence
to what they heard. With growing tensions in the country, however, doubt eventually led to
distrust at the community level. 142 As an interviewee told African Rights: “[a]t first we
remained in good terms with the Hutus. But once they realized the scale of what was
happening, [...] we lived in a climate of mutual suspicion.” 143 Among communities, people
raised questions about possible ties between their Tutsi neighbours and the conflict,
asking them if they had been aware of the RPF’s intentions when they invaded or if,
because of their Tutsi background, they might be privy to some secret information. A Tutsi
140
Kimonyo came to the same conclusion in his analysis of pre-1990 Rwanda, and went even further. He
indicated that “by the end of the 1980s, the Tutsi component of the population had started to be assimilated by
the Hutu majority.” Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 106.
141
Interviewee no. 1, Kigali, Rwanda, 9 March 2004.
142
Interviewee no. 6, Kigali, Rwanda, 15 March 2004.
143
African Rights, Tribute to Courage, p. 60.
226
interviewee recalled, for example, being asked by her neighbours about the RPF. Judging
her to be in a better position to know about the group, they inquired about rumours they
had heard of RPF soldiers having tails and long ears. 144
With time, however, many described these inquiries as turning more and more
frequently into open confrontations: some individuals were beginning to express their
anger at the fact that the Tutsi in their community had not been neighbourly, failing to warn
them of an attack perpetrated by their kin, the RPF. With relations deteriorating following
growing tension in the country, distrust eventually made way for surveillance. 145 Rumours
not only circulated at the political level but increasingly spread among communities about
some of their more ‘suspicious’ members. A condemned killer, for his part, admitted that,
“over time, the atmosphere encouraged me to suspect the Tutsi. Even when they seemed
nice, I had to see them as menacing because of the dark rumours.” 146 Wealthy Tutsi, for
example, were suspected of contributing to the enemy’s, the RPF’s, military enterprise.
Or, if a group of Tutsi was seen together, people looked suspiciously at them, wondering if
they were conspiring, working on a secret plot against the Hutu. 147 One of the most
common rumours was that the Tutsi were leaving to join the maquis.
Community
members paid close attention to the whereabouts of their Tutsi neighbours or to the
goings-on of their family members.
For example, following the 1972-1973 purges in
schools and in light of the quotas in the education system, many Tutsi had children who
had studied or were studying abroad; rumours circulated about them joining the ranks of
the RPF because of ties they had developed abroad. Furthermore, anyone who had to
travel outside the country was suspect. In an interview, a Rwandese echoed what could at
times be heard upon their return to Rwanda: “he’s coming back from training, he’s coming
144
Rumours about the RPF abounded. They were at times described as devils, with tails and long ears.
Interviewee no. 22, Kigali, Rwanda, 1 April 2004.
145
Interviewee no. 15, Kigali, Rwanda, 24 March 2004.
146
Personal translation. Jean Hatzfeld, Une saison de machettes (Paris: Éditions du seuil, 2003), p. 257.
147
Interviewee no. 18, Kigali, Rwanda, 26 March 2004, interviewee no. 15, Kigali, Rwanda, 24 March 2004.
227
back from the RPF camps, he could constitute a threat, we will have to watch him.” 148
Such rumours even circulated about longtime friends.
As the political situation deteriorated in the country, paralleling what was happening
on the national scene, it became more common among communities that the Hutu accuse
“their Tutsi neighbours of sympathy with the rebels.” 149
People noticed that as a
consequence of this growing climate of suspicion, regular social interactions they had
within their community, village or hill, began changing. They were careful in public, or in
groups, in cabarets for example; starting to avoid discussing certain issues for fear of not
knowing who was listening. 150
Social dynamics also changed within circles of friends.
People kept their distances, sometimes avoiding contacts with neighbours or friends, even
though traditional customs had insisted on close contact, mutual assistance and the
sharing of resources. 151 The Rwandese were keeping to themselves, to their close family,
more and more, trying to get by without the help of others. Even among close friends,
many talked of a growing distance, some even fighting openly while relations had
previously been amicable. Some simply refused to talk to each other. 152 According to an
interviewee, there started to be the sense that it was safer to keep to one self because “a
public enemy had been identified, you [had] to do everything to avoid him, to avoid being
with him, if not, someone [could] denounce you and say ‘he’s cooperating with the
enemy’.” 153
Fear, taking on various forms–of an enemy living next door or of being denounced
as a collaborator, for example–increasingly became a feature of people’s lives. It was the
result of governmental rhetoric vilifying the RPF and the Tutsi, presenting them as
enemies posing a monstrous threat for all Hutu, seeking to reinstitute an oppressive
148
Interviewee no. 9, Kigali, Rwanda, 17 March 2004.
African Rights, Tribute to Courage, p. 243.
150
Interviewee no. 15, Kigali, Rwanda, 24 March 2004.
151
Interviewee no. 29, Kigali, Rwanda, 15 April 2004.
152
Interviewee no. 25, Kigali, Rwanda, 5 April 2004, interviewee no. 21, Kigali, Rwanda, 31 March 2004.
153
Personal translation. Interviewee no. 15, Kigali, Rwanda, 24 March 2004.
149
228
monarchical regime. 154 Rhetoric was not the only tool used to stoke up fear though. The
agitation of the RPF’s ‘atrocities,’ recurring massacres, militia actions, and political
assassinations heightened the paranoia.
So did smaller acts of violence among the
population which often served the purpose of driving into people’s minds that the country
remained in a state of war, a condition that demanded vigilance and unwavering support.
Betrayal, in particular, could not be tolerated. Punitive actions against people accused of
collaborating with the Tutsi exemplified this. A Rwandese told African Rights about such
actions taking place in his community around the time of the genocide. After arresting a
‘Hutu collaborator,’ “[t]he Interahamwe dragged him all around [the city], through every
street in town. They killed him in front of the Hutu residents. They wanted to show the
people that anyone who tried [...] to oppose them, would be killed.
The Hutu were
intimidated. Most threw out the Tutsi they had been hiding.” 155
Fear was also clearly a reality for the Tutsi living in Rwanda at the time. With what
was happening in the country, they were increasingly aware of the danger they faced as
targets of the rhetoric and regime actions. In their everyday life, they tried to hide the deep
restlessness they felt in order not to attract attention to themselves. According to most,
however, they lived in constant fear. An interviewee remembered that throughout the
period, although she and her husband continued to go about with their work, they would
everyday wear two layers of clothing knowing that if they were taken they at least had
spares. 156 Many also tried to negotiate arrangements with Hutu neighbours to hide their
belongings or ensure their children’s safety. Fear was the lot of their every day existence
in the last years before the genocide. According to a Rwandese that remembered the
times, because they constantly feared for their lives, “the Tutsi had fallen into a state of
154
Interviewee no. 10, Kigali, Rwanda, 19 March 2004.
African Rights, Tribute to Courage, p. 102.
156
Interviewee no. 23, Kigali, Rwanda, 1 April 2004.
155
229
survival.” 157 Their existence was limited to simply getting through the day, in a state of
expectancy, not knowing when the violence they sensed coming would finally be
unleashed.
Fear did not play a negligible role in the events that occurred in Rwanda. At one
level, it played an important part in leading people to isolate themselves, to keep to
themselves, and thus in turning their backs on what had been their traditional ways of
relying on their community for help even during previous periods of violence. At another
level, though, isolated from one another, they were now increasingly at the mercy of the
authorities. Fear rendered them “more easily manageable, manipulable.” 158 With fear
progressively taking hold of people’s minds, but with individuals increasingly alone to face
their fears, people were in need of guidance, which the authorities were eager to provide.
They were more receptive to the authorities. But with fear rooted in irrationality, it also
made them more receptive to the irrational. 159 And in the case of Rwanda, the irrational
took the form of turning away from lives of cooperation within communities to adopting
relations seen through the prism of division as prescribed by the authorities.
Trends described by various authors and interviewees seem to confirm the
hypothesis that the strategic actions of elites, their interpretation of events at the time, had
an impact on personal and communal relations. In the words of one Rwandese, “before,
[during crises], solidarity ties were never completely severed but that is exactly what the
genocide did.” 160 Ethno-centric and divisive perceptions of identities, political identities,
finally superseded personal and communal ones that had been sheltered and a shelter in
previous decades. When, in parallel to describing changes on the ground, interviewees
were asked if the masses bought into the rhetoric of the regime and extremists, most
answered affirmatively. As a matter a fact, a large proportion of respondents in interviews
157
Interviewee no. 26, Kigali, Rwanda, 5 April 2004.
Personal translation. Interviewee no. 6, Kigali, Rwanda, 15 March 2004.
159
Interviewee no. 12, Butare, Rwanda, 23 March 2004.
160
Interviewee no. 8, Kigali, Rwanda, 16 March 2004.
158
230
conducted in Rwanda between March and April 2004 answered that, in their view, people
did in fact believe what the authorities were saying about the RPF, and that this alone or in
combination with other factors led to mobilisation. (See Appendix). Similarly, for a large
number of Rwanda analysts, there was a progressive internalisation of the rhetoric on the
part of the Hutu population. 161 There were notable exceptions: some regions in which the
process proved slower and more limited and certain individuals that worked even
throughout the genocide to protect and save the Tutsi. 162 Convincing populations should
not be understood as an absolute act. There always exists degrees. For example, in his
analysis of the conflict in Bosnia, David Campbell recounted encounters with individuals
refusing to accept the ‘ethnicised’ narratives defended by ethno-centric leaders. As he
explained, “even in the most difficult circumstances individuals exercise the capacity for
ongoing and ever present judgment, evaluation, and discrimination that problematizes the
chauvinistic and xenophobic rationales offered as accounts for those extreme
situations.” 163 On a grand scale, however, appeals seemed to have reached a critical mass
which opened the door to a progressive acceptance of the idea of genocide. 164
Although this last section centered on the normative motives behind mobilisation, it
should be noted that some authors have pointed to complementary motivational factors
behind the massive popular participation in the Rwandese genocide. For example, Prunier
underlined the existence of a Rwandese political tradition which he described as “one of
161
A number of authors studying popular motives behind participation in the genocide have defended this point
of view. Human Rights Watch, for example, explained that “Hutu who killed Tutsi did so for many reasons, but
beneath the individual motivations lay a common fear rooted in firmly held but mistaken ideas of the Rwandan
past. Organizers of the genocide, who had themselves grown up with these distortions of history, skillfully
exploited misconceptions about who the Tutsi were, where they had come from, and what they had done in the
past. From these elements, they fueled the fear and hatred that made genocide imaginable.” Human Rights
Watch, Leave None to Tell the Story, p. 31. See also Jean-Paul Kimonyo, La participation populaire au
Rwanda; René Lemarchand, “Les génocides se suivent mais ne se ressemblent pas” ; and Gérard Prunier,
The Rwanda Crisis.
162
There are extraordinary examples of resistance during the Rwandese genocide. These include JeanBaptiste Habyarimana, the head of the Butare préfecture until two weeks into the genocide, who did not
comply with the regime’s orders to stoke up violence. Butare was spared from the killings until he was
replaced. They also include the numerous heroic Hutu who risked their life hiding Tutsi. Some of their stories
can be found in African Rights, Tribute to Courage.
163
David Campbell, National Deconstruction, p. 2.
164
Interviewee no. 8, Kigali, Rwanda, 16 April 2004.
231
systematic, centralised and unconditional obedience to authority.”
165
As he further
explained,
there had always been a strong tradition of unquestioning obedience to authority in
the pre-colonial kingdom of Rwanda. This tradition was of course reinforced by
both the German and the Belgian colonial administrations. And since independence
the country had lived under a well-organised tightly controlled state. When the
highest authorities in that state told you to do something, you did it, even if it
included killing. 166
A report by Amnesty International also stated that this obedience “characterized the
political culture of the independent Rwandese state.” 167 A number of interviewees also
raised the issue. One explained, for example, that figures of authority had a large impact
on peasants: “peasants always thought that each time they had to respect the directives
that came from the top. [...] They did not have the possibility to say no to a decision
emanating from the top.” 168 Bill Berkeley quoted a Rwandese businessman stating that
“[t]he popular masses in Rwanda are poorly educated. Every time the powers that be say
something, it’s an order. They believe someone in political authority. Whatever this person
demands, it’s as if God was demanding it.” 169
Some believe this cultural trait might have led the Rwandese to be inherently more
receptive to appeals on the part of incumbents at the time of the genocide. To illustrate
the importance of this respect for authority, many have looked at regional dynamics during
the genocide. In the commune of Butare, for example, the killings started much later than
in other parts of the country. The préfet in place, Jean-Baptiste Habyarimana, had from
the start of the genocide been opposed to the massacres. 170 He was eventually removed
and massacres also took place in this commune. That a figure of authority, by giving
165
Gérard Prunier, The Rwanda Crisis, p. 141
Ibid., p. 245.
167
Amnesty International, Gacaca: A Question of Justice, Rapport AFR 47/007/2002 (London: Amnesty
International, 2002), p. 4.
168
Personal translation. Interviewee no. 2, Kigali, Rwanda, 9 March 2004.
169
Bill Berkeley, “Sounds of Violence: Rwanda’s Killer Radio,” New Republic, vol. 211, no, 8-9, 1994, p. 19.
170
See Gérard Prunier, The Rwanda Crisis, p. 244.
166
232
orders opposed to a call to murder found in the rest of the country, delayed the killing, for
some authors, is a sign of the importance of deference in Rwandese society.
The obedience hypothesis is contested, however. The obedience argument is a
deterministic one where individuals are portrayed as puppets or robots. In his analysis of
factors behind popular mobilisation in Rwanda, Kimonyo opposed this notion of ‘blind
obedience.’
According to Anthony Giddens, quoted in Kimonyo’s work, “when an
individual or a collectivity is in a position of subordination in a power relation, this individual
still has nevertheless a certain leeway in the way he submits to order [...]. There is a
continuum of possible attitudes: the extremes being voluntary obedience or open
rebellion.” 171 The term ‘voluntary obedience’ still denotes a sense of volition and not a
quasi-hypnotic state. Examples throughout recent Rwandese history point to the fact that
the Rwandese do contest authorities in various ways. A good example of it is the
conduites d’indocilité, such as the uprooting of tea and coffee trees by peasants, that
occurred in the late eighties. Even under mundane conditions, in their daily lives,
Rwandese eluded authority: “evading taxes, smuggling, avoiding mandatory meetings,
escaping from community labor, engaging in petty crime, or illegally migrating.” 172 Overall,
Rwandese remained, despite heavy indoctrination, independent people capable of
individual judgment. 173
Another factor raised as a plausible individual motivation for taking part in the
genocide is the impact of economic conditions on individual motivations. Many killers
succeeded in enriching themselves through the genocide.
Jean Hatzfeld quoted a
convicted killer as saying: “[i]f the [RPF] had not conquered the country, chasing us out,
we would have killed one another upon the death of the last Tutsi, that is how hungry we
were to divide up the loot. We could not stop wielding our machete, we profited too much
171
Personal translation. Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 26.
Peter Uvin, “Reading the Rwandan Genocide,” International Studies Review, vol. 3, no. 3, 2001, p. 84.
173
Peter Uvin, Aiding Violence, pp. 67-68.
172
233
from it.” 174 By the 1990s, a significant portion of the Rwandese population was living in
dire economic conditions. The economic crisis that had started developing in the mideighties had left them extremely poor and vulnerable. The extremists throughout the years
leading up to the genocide reminded the masses of the economic benefits of attacking the
Tutsi, mainly by being able to secure their goods, cattle or land. Such a rationale proved
convincing, according to many. As one interviewee explained, “[t]hey were poor, incredibly
poor, [...] so during the purges, even if they gained only a meter of land, for a peasant it
was a lot.” 175 To further exemplify the importance of small gains in these terrible
conditions, the same interviewee retold the story of an old woman that rejoiced at the fact
that the massacres of Tutsi in her region had allowed her sons to seize a cow which they
had killed and eaten. She had been overjoyed to be able to eat meat sauce, after twenty
years of having gone without.
This factor seems also to have played an important role in convincing some of the
rural youth to mobilise against the Tutsi. This group had been particularly affected by the
economic crisis. Coupled with a demographic boom, they were often left unable to find
employment within their community and to provide for themselves.
Resentful of this
situation, many jumped on the opportunity to get paid, trained and to belong to a
sanctioned institution, when the regime and political parties began recruiting for their
militias.
Though this economic explanation of motives behind mobilisation is also contested,
many do see it as a reinforcing factor. Developing the argument further, Kimonyo, for
example, believed that massacres and purges in decades prior to the genocide had
sometimes allowed people to succeed in securing small economic benefits. With such
types of violence recurring throughout the years with almost complete impunity, it might
174
175
Personal translation. Jean Hatzfeld, Une saison de machettes, p. 106.
Interviewee no. 27, Kigali, Rwanda, 13 April 2004.
234
have developed the sense that collective violence was a plausible option to satisfy basic
economic needs during hardships. As Ted Robert Gurr argued,
[p]robably the most potent determinant of perceived utility of political violence is
people’s previous success in attaining their ends by such means. Psychological and
comparative evidence [...] suggests that people who obtain their demands through
aggression are likely to use it as a tactic in the future. Intermittent rewards for
aggression lead to the establishment of very persistent aggressive habits. 176
The importance of such economic justifications for mobilisation in the Rwandese
genocide warrant further research. Considering how messages conveyed to mobilise
the masses also included reminders of economic benefits gained from looting the
property of the deceased, it probably was a contributing factor. This said, it was hardly a
sufficient condition to achieve the levels of participation, to reach numbers that did to
cross beyond morality into genocide. As Alison Des Forges explained, “[authorities] also
offered attractive incentives to people who are very poor, giving license to loot and
promising them the land and businesses of the victims. [...] But even with the powerful
levers of threat and bribe, officials could not have succeeded so well had people not
been prepared to hate and fear the Tutsi.” 177
176
Ted Gurr, Why Men Rebel (Princeton, N.J.: Princeton University Press, 1970), p. 218.
Alison Des Forges, “The Ideology of Genocide,” Issue: A Journal of Opinion, vol. 23, no. 2, 1995, p. 44.
According to Des Forges, “[i]n some cases, local officials even decided ahead of time the disposition of the
most attractive items of movable property. Everyone knew who had a refrigerator, a plush sofa, a radio, and
assailants were guaranteed their rewards before attacking.” Ibid., p. 44.
177
235
CHAPTER 9: RWANDA
MOBILISATION
FROM
THE
PERSPECTIVE
OF
FRAMES
FOR
The previous chapters explained the history, events and factors that led to the 1994
Rwandese genocide. They also served to establish that elite machinations in the early
1990s effectively played an important part in the population’s widespread participation in
the genocide, that the population to a large extent bought into elite rhetoric and ‘felt
mobilisation.’ Too often, this link is simply assumed.
Yet, a mobilised mass has not
necessarily responded to calls for action. It might have responded to greed or fear, for
example, more than belief. In the case of Rwanda, a large proportion of respondents
interviewed for this research (See Appendix) and a number of specialists agree that
propaganda and the ensuing indoctrination were critical for the mobilisation of Hutu
masses during the genocide. The ethno-centric rhetoric disseminated for decades, and
more urgently propagated following the Rwandese Patriotic Front’s attack in October 1990,
did not fall on deaf ears. On the contrary, it reached growing numbers already converted
to the cause of recreating Hutu solidarity in the face of a menacing Tutsi ploy.
According to Peter Uvin, “[a]ll genocides in history have been instigated, organized,
and legitimized by the state.” 1 Genocide is a modern crime that requires a degree of
rationality and organisation to be implemented. It is fundamentally an elite-led
phenomenon. Having established that Rwanda was a successful case of elite-led
mobilisation, that a large segment of the population was brought to feel the need for
collective action, it is therefore essential to understand the strategies elites deployed to
mobilise their supporters in Rwanda.
Instrumentalist explanations of the Rwandese genocide are popular.
They are
defended in whole, or in combination with, other arguments by “most scholars from
1
Peter Uvin, “Reading the Rwandan Genocide,” International Studies Review, vol. 3, no. 3, 2001, p. 80. On the
inherently elite-led nature of genocides, see Wilhelmus P. Du Preez, Genocide: The Psychology of Mass
Murder (London: Boyars/Bowerdean, 1994). For an in-depth review of genocide and literature on the
phenomenon, see Helen Fein, Genocide: A Sociological Explanation (London: Sage Publications, 1993).
236
different disciplinary perspectives.” 2 There certainly was calculated rationality behind the
1994 Rwandese genocide. The gruesome violence of the early 1990s was brought about
by a multifaceted crisis that began developing in the mid-1980s. For the incumbent regime
and a number of its backers, internal and external contestation, the looming threat of a
liberalised political arena, a depleted economy and a shrinking source of bribes and
indulgences to buy support, suggested an end to their unqualified control of the country.
Faced with the possibility of having to share power or being replaced all together,
something had to be done. A plan was needed to ensure elites in and around the state
apparatus maintained their privileged position.
The first actions taken in the late 1980s were more reactive than part of a
coordinated strategy. Political opponents of the regime and vocal detractors were
eliminated. These assassinations, often disguised as car accidents, targeted “people seen
as too critical of the regime, such as the courageous editor of Kinyamateka, a widely-read
newspaper written in Kinyarwanda [...], and an outspoken and popular woman deputy.” 3
When the Rwandese Patriotic Front invaded the country across the Northern border with
Uganda and the civil war began, the situation changed. An opportunity opened up for the
Habyarimana regime and its extremist supporters. While it was an additional problem for
the regime, the war with the RPF also constituted an opportunity to deflect popular
opposition and channel it towards a new cause, the defence of the country from an
invasion. With the country in need of defence, incumbents could rationally justify keeping
their post–ensuring their political survival–as experienced leaders of the nation, who were
best suited to wage war against the enemy.
Instrumentalist analyses such as these tend to focus on background conditions to
the genocide: the economic crisis, the legitimacy crisis the government was facing, the
2
Peter Uvin, “Reading the Rwandan Genocide,” p. 80.
Catharine Newbury, “Background to Genocide: Rwanda,” Issue: A Journal of Opinion, vol. 23, no. 2, 1995, p.
13.
3
237
RPF invasion. The brunt of the argument focuses on the late 1980s and early 1990s.
However, the tactics deployed by extremists in Rwanda in the years prior to the genocide,
especially the narrative they used to frame the civil war, can not be understood without
referring to the First and Second Republics and even beyond, to the colonial and precolonial era.
The period between 1959-1990 is essential to understanding how the rhetoric of
the civil war against the RPF constituted a re-interpretation of the ideology disseminated
during the First and Second Hutu Republics. It could achieve resonance through a rereading of the previous decades’ ideology unto which a sense of severity/urgency and of
the efficacy/propriety of collective action were grafted starting in 1990. This constitutes a
classic framing argument. While informative–it gives an idea of the general framing
process used to motivate the Hutu–this line of argument does not shed light on a particular
detail of the process. Motivating vocabularies, severity/urgency and efficacy/propriety, as
well as resonance in terms of cultural/ideological references all relate to the group. It is
urgent because ‘we’ are threatened; together ‘we’ can succeed; remember ‘we’ were
this/’they’ were that. For these arguments to have traction, though, the ‘we’ must be
brought to the fore. Another reason, then, why modern Rwandese history matters is that it
illustrates the role of three mechanisms, opposing, politicising and simplifying, used to
(re)create popular support, to (re)create a sense of ‘we-ness.’ Not only were these used
during the civil war, but recourse to these mechanisms is seen throughout Rwandese
history since independence. Both Hutu Republics are cases in which ethno-centric rhetoric
was manipulated to foster Hutu solidarity in the face of uncertain popular support or
threatened legitimacy. Each new use of these strategies further inscribed them, and
legitimated them, as a credible repertoire of political action.
238
This chapter begins by applying the classic framing agenda to Rwanda. It then
turns to looking at the use of solidarity mechanisms both during the 1959-1990 period and
the lead-up to the 1994 genocide.
The Classic Framing Argument: Resonance, Severity/Urgency, Efficacy/Propriety
A classic framing argument is predicated on the fact that communication is two-sided. In
order to frame with success, messages must accommodate their recipients, they must be
tailored to them. In particular, to reach a target audience, a message should make sense
to this audience, it should seem sensible and credible. A message built around elements a
group is already familiar with is easier to accept. It resonates with what people know. In
addition, conveying a frame is done by employing specific vocabularies. While the
narrative, or script, is the essence of the message, its vocabulary, its words, are its vessel.
But more than simply a medium, different vocabularies read into a discourse a direction or
emotion. Emphasising or employing certain words, terms and connotations can change a
script’s tone. It changes the effect of the statement, giving it an additional sense or
meaning. Frame analysts have highlighted the importance of vocabularies of
severity/urgency and efficacy/propriety for calls for collective action. Vocabularies
emphasising the severity and urgency of a situation, an imminent threat, for example,
press the need for action: ‘these are exceptional times that call for exceptional action,
including collective action.’ Efficacy and propriety, on the other hand, confirm the
sensibility of action and condone it: ‘together, it can be achieved; it is the appropriate, even
dutiful, course of action.’ These are motivating vocabularies.
Applied to Rwanda, this type of argument shifts the focus of the analysis to the
ethno-centric ideology of the First and Second Republic. It can be argued that if there had
not been, over the course of decades, the development and dissemination of this ethnocentric ideology among the Rwandese population, the extremists’ schemes would not have
239
succeeded, or not to the extent they did. This same ethno-centric ideology, reactivated in
full force during the civil war, served as the material from which to build resonant and
motivating mobilising frames.
The First and Second Republic depicted the Tutsi as historical foreign oppressors,
as power hungry Ethiopian or Hamitic aristocrats devoted to the idea of a Tutsi monarchy.
They were not fully Rwandese, at least not as much as the ‘Sons of cultivators,’ the
Benesabahinzi, the Hutu, as children of the land could claim to be. 4 The Tutsi were not
deserving of the rights that came with full natural citizenship reserved for the Rubanda
Nyamwinshi, the demographic, and thus democratic, majority. They needed to be kept in
check, surveyed and controlled. Even while during the Second Republic the Tutsi were
recognised as an ethnic group within the country, they remained second rate citizens.
This ideology was spread among the population through education. Schools taught it,
particularly in their history and civic education classes. Once educated to the principles of
this ideology, people reproduced it within their communities. Informal education
propagated the teachings of the formal education system. It was also spread by political
rhetoric, from the local administrator’s orations to the Président-Fondateur’s speeches.
The omnipresence of the Mouvement révolutionnaire national pour le développement at all
levels of society ensured consistency and predominance of the ideology. This ethnocentric ideology also translated into practice, particularly under Habyarimana. The second
rate citizenship of the Tutsi was entrenched in a quota system. The Tutsi were
marginalised in, or excluded from, various facets of society, from higher education, public
employment and the military. This was further ingrained by recurrent violence against the
Tutsi community in the early 1960s and between 1972-1973. Treated with impunity, and
more often than not triggered by authorities, this violence further confirmed the Tutsi’s
4
Jean-Pierre Chrétien, “RTLM Propaganda: The Democratic Alibi,” in Allan Thompson, ed., The Media and the
Rwanda Genocide (London: Pluto Press/IDRC, 2007), p. 58.
240
lesser condition. Overall, ideology and practice were in synch for decades, making it a
rigid, overarching socialisation structure. It became a part of one’s background and
cultural, social and political referents. Keeping in mind that many of these ideas had
already been circulating for decades–the “faulty history had long been accepted by both
Hutu and Tutsi”–they were accepted as veritable depictions of groups and group
relations. 5
This was essential for the events of the early 1990s. Faced with a growing
legitimacy crisis, rendered particularly problematic by the prospect of power-sharing in a
democratised political arena, the regime decided to fall back on this ethno-centric ideology
to mobilise the support of the Hutu population. Under the circumstances, the ideology
proved particularly fitting. A link could be drawn between the Rwandese Patriotic Front’s
fighters and the Tutsi that had fled the country in previous decades. The RPF was mostly
composed of Tutsi refugees from the purges of decades past, or their descendants. These
same refugees had, for a good part of the 1960s, recurrently attacked the country to
overthrow the Kayibanda regime. The RPF could be tied to those ‘feudalists’ that had fled
the country following the abolition of the monarchy, to those monarchists who in the early
days of the young Republic were intent on destabilising it and taking power again. 6 As an
article in Kangura claimed,
[s]ince the revolution of 1959, the Batutsi have not for one moment relinquished the
notion of reconquering power in Rwanda, of exterminating intellectuals and of
dominating Bahutu farmers [...] The war declared against Rwanda in October 1990
is undoubtedly aimed at achieving what the Batutsi attempted to accomplish
through guerrilla warfare and terrorism, from 1962 to 1967. 7
References were made to the threat they posed for the Hutu Republic, as Tutsi foreign
invaders, which left the preferred role of the saviour of the Rubanda Nyamwinshi to the
5
Alison Des Forges, “The Ideology of Genocide,” Issue: A Journal of Opinion, vol. 23, no. 2, 1995, p. 44.
In reality, not all of the groups which attacked Rwanda in the sixties were monarchists. Some were, but other
espoused leftist allegiances. If anything, these military groups tended to disagree. See Gérard Prunier, The
Rwanda Crisis: History of a Genocide (London: Hurst and Company, 1995), p. 55.
7
In Marcel Kabanda, “Kangura: the Triumph of Propaganda Refined,” in Allan Thompson, ed., The Media and
the Rwanda Genocide (London: Pluto Press/IDRC, 2007), p. 64.
6
241
Habyarimana regime. Over time, Tutsi within the country also came to be seen as a
source of threat, as accomplices to the invaders. As Bruce Jones explained, ethno-centric
entrepreneurs played on “a second reverberation of the memory of earlier ethnic
domination. [...] the akazu used the fear of Tutsi domination as a tool to legitimate violence
against Tutsi in Rwanda and the Tutsi-dominated RPF.” 8
The story concocted by the extremists in the early 1990s to frame the RPF attack
was specifically built, as Uvin argued,
on longstanding myths and beliefs, expressed in stories, aphorisms, and proverbs;
on decades of political ideology transmitted through speeches, official histories
recited in school books, and exclusionary policies. [This was] widely shared by all
Rwandans–elites and masses, rich and poor, Hutu and Tutsi. Like every social
phenomenon, it [was] not equally internalized by all nor uniformly acted upon by all,
but it [was] a living reality for all. 9
As a consequence of the pervasiveness of these myths and ethno-centric ideology, when
the Rwandese Patriotic Front invaded Rwanda across the Ugandan border on 1 October
1990, much of the ensuing script had already been written–or at least, not to fall into the
trap of over-determinism, certain storylines seemed more probable than others. This
social, cultural and political background constituted an elaborate narrative complete with
well-defined characters, images and background. It was a narrative familiar to the
population that circumscribed how the war with the RPF might be scripted or framed. The
‘ideology was reality for all,’ and, as such, determined what made sense to the Rwandese.
As reality for all, it also determined elites’ belief system. 10 Elites were not outside of the
cultural, social and political environment the population had been exposed to. As a result,
8
Bruce D. Jones, “Civil War, the Peace Process, and Genocide in Rwanda,” in Taisier M. Ali and Robert O.
Matthews, eds., Civil Wars in Africa: Roots and Resolution (Montreal: McGill-Queen’s University Press, 1999)
p. 60.
9
Peter Uvin, “Reading the Rwandan Genocide,” p. 81.
10
In this, Uvin noted that many authors supporting the instrumentalist thesis have “a tendency to
underestimate, if not entirely neglect, how much the elite themselves believe in the ideological messages they
broadcast. The capacity to blame the Tutsi for all of society’s evils and to eventually consider eradicating them
like a cancer from society does not materialize from thin air.” Peter Uvin, “Reading the Rwandan Genocide,” p.
81.
242
the ideology imposed limits on just how much people could think outside of it, especially in
a collective manner.
With this ethno-centric ideology as a predominant normative structure, frames built
around it were resonant. According to frame analysts, frame resonance is a function of two
set of factors (see Table 4.1 in chapter 4). It is function of frame credibility, dependent on
frame consistency, empirical credibility and the credibility of articulators. The case could be
made that frames developed were consistent. They were credible, they coherently weaved
a narrative of blame around Tutsi feudalists of old and RPF invaders tied to solutions
aimed at controlling the Tutsi. They were also empirically credible. They were after all in
response to a military action provoked by Tutsi in Uganda. There were facts elites could
draw on to support their case. Finally, articulators were credible. These were known
politicians building on the help from known intellectuals and artists. Resonance is also a
function of a frame’s relative salience. Relative salience is dependent on centrality,
experiential commensurability, and narrative fidelity. Tied much more with senses and
experiences of an audience, relative salience pertain to how frames address issues,
interests and values of importance for an audience; whether they fit with what people see,
feel, and sense on a practical level; and if they correspond with existing beliefs, myths and
ideologies. During the civil war, the regime’s discourse focused on what was the
paramount issue in Rwandese political reality at the time. It centred on the fate of the
country, and the fate of one of its communities, the Hutu, in particular. Frames pertained to
an undeniably central issue. More importantly, by reproducing the storyline of Rwandese
‘faulty history’ and ethno-centric ideology, frames were ensured of their commensurability,
not only in terms of what people experienced–differences between groups in terms of
treatment they received by the state in practice–but also with cultural myths, education and
beliefs. They ensured both experiential commensurability and narrative fidelity. Elite calls
for mobilisation used familiar references, central to the ideological system of the time and
243
its matrix of behaviour and thoughts, accepted and internalised norms of proper actions
and beliefs. Frames employed by extremists, therefore, fell on familiar ground. Frames
developed over the course of the civil war possessed all characteristics necessary to make
them resonant.
This discourse resonated with the ideology disseminated during previous decades,
but frames were, as well, conveyed with a sense of threat, of severity and urgency. In the
midst of a war, under cover of a ‘Tutsi scheme’ to take over the country, the situation was
portrayed as severe. Times were grave. The early rhetoric focused on the RPF. This
armed force was depicted by the government as a group of expert, experienced
assailants. The regime’s agents insisted on RPF soldiers’ strong military training and the
years of experience they had gained in the Ugandan conflict and in various conflicts on the
African continent. They also drew strength from the military and financial support of allies
in the region and outside of the continent. The country was, therefore, facing a powerful
enemy, one who was not to be underestimated. Further stressing the danger, rumours
spread that RFP soldiers were not human. They had tails and pointed ears, like devils.
And like devils, they committed atrocities when they attacked. 11 In an article titled “A
cockroach can not give birth to a butterfly,” the journal Kangura stated that “[t]he
unimaginable crimes that today’s Inyenzi commit against citizens recall the ones
committed by their elders: killing, pillaging, raping young girls and women, etc.” 12
Increasingly, the discourse shifted its focus from the RPF to Tutsi in general. The
link between the RPF and local Tutsi was first made to point to the existence of Tutsi
accomplices in Rwanda, with those targeted often being opponents of the regime. Rapidly,
however, as Kimonyo argued, ethno-centric entrepreneurs and ideologues “dramatised the
11
According to a number of interviewees, some people believed these tales about the RPF. For others who did
not, these stories were a source of amusement, a joke, like depictions of RPF soldiers as Inyenzi,
cockroaches, was. Jean Hatzfeld, Une saison de machettes (Paris: Éditions du seuil, 2003), p. 276.
12
Personal translation. Quoted in Bernard Mulinda, La liberté de la presse au Rwanda (1974-1994), licence
thesis, Faculté des lettres, Université nationale du Rwanda, Butare, Rwanda, December 1999 p. 76.
244
fear of the RPF and associated all Tutsi to it.” 13 There existed a much larger Tutsi plot
which reached deep within Rwanda, making all Tutsi nationals suspect. With this type of
framing, a new degree of severity was reached. Not only was the enemy threatening at
the border, as an invading force, it was already inside the country. An article published in
Kangura in 1990, presented as a reprint of an article written in 1976, spoke of Tutsi
collaboration in these terms: the Tutsi with the help of Hutu and Twa traitors “[...] are
constantly communicating and co-operating, to eventually take revenge on the Rwandan
Republic [...] This is the two-headed dragon, one head outside Rwanda and the other
inside.” 14 Fear of an external attack supported by internal accomplices fostered paranoia:
‘they walk among us, even our land, our homes, are not safe now’.
The enemy was strong, numerous, and close. The enemy’s goals also rendered
the situation urgent, demanding immediate attention and immediate action. Not only were
their intentions to come back to the country, but they were to come back and reinstitute a
Tutsi monarchy, enslave the Hutu, and, if need be, to annihilate them. 15 According to
UNAMIR transcripts, Radio Rwanda attributed goals of ‘ethnic purification’ to the RPF. 16
Kangura published an editorial in 1991 exclaiming that “[the] enemy plans to exterminate
the majority people. [...] Today, they plan to exterminate us all.” 17 The Hutu were
confronted with their destruction as a group; the situation raised fears of extinction. An
interviewee recalled a radio interview, at the onset of the genocide, with Justin Mugenzi,
minister of commerce, industry and crafts. In it, Mugenzi raised, in words echoed in many
other speeches, the idea of a last battle between Hutu and Tutsi. According to the
interviewee, Mugenzi exclaimed that “[n]ow things are clear. [...] This is the last war in the
13
Personal translation. Jean-Paul Kimonyo, La participation populaire au Rwanda: de la révolution au
génocide (1959-1994), doctoral thesis, Département de science politique, Université du Québec à Montréal,
Montréal, Canada, September 2002, p. 572.
14
Quoted in Marcel Kabanda, “Kangura.,” p. 68.
15
Interviewee no. 10, Kigali, Rwanda, 19 March 2004.
16
Bill Berkeley, “Sounds of Violence: Rwanda’s Killer Radio,” New Republic, vol. 211, no, 8-9, 1994, p. 18.
17
Personal translation. In Bernard Mulinda, La liberté de la presse au Rwanda (1974-1994), p. 77.
245
history of Rwanda. Hutu, you must win it. If you lose [...] you are doomed for eternity.” 18
The cues were obvious: ‘it was now the time to act or never.’
Discourses also made reference to the propriety and efficacy of collective action, of
mobilising against the Tutsi. According to Jean-Pierre Chrétien, two predominant themes
characterised Radio télévision libre des milles collines’ broadcasts: one vitriolic hatred of
the Tutsi ‘cockroaches’ and the other pertaining “to the legitimacy of the elimination of
these ‘cockroaches’ by the ‘majority people.’ 19 Mobilisation was, on one level, justified as a
legitimate act of self-defence. Faced with the threat to the entire Hutu community posed
by the Tutsi, it was appropriate to take action against the Tutsi as a defensive stance in the
face of an aggression. Roger Mucchielli, a psychosociologist working on group dynamics
and propaganda, referred to this tactic as ‘mirror propaganda’ or ‘accusations in a mirror.’
It consists of attributing to others one’s intentions. 20 As a result, “propagandists can
persuade listeners and ‘honest people’ that they are being attacked and are justified in
taking whatever measures necessary ‘for legitimate [self]-defence.” 21 Examples of this
type of attribution abound in the rhetoric of the civil war. In an article published in Ijambo
intended to frighten Hutu by describing purported atrocities committed by Tutsi in Burundi,
the author claimed that
[t]he day a Hutu will learn to cut the flesh on the body of a Tutsi and make him eat
it–like Batutsi do it today to Bahutu–that will be the day the Tutsi lose their
arrogance. [...] The day the Bahutu will massacre the Tutsi by throwing them in the
18
Personal translation. Interviewee no. 20, Kigali, Rwanda, 29 March 2004.
Jean-Pierre Chrétien, “RTLM Propaganda: The Democratic Alibi,” p. 56.
20
According to Chrétien, all of Mucchieli’s work could be found at the National University in Rwanda. It inspired
a Rwandese intellectual who produced his own manual for propaganda to back the extremists’ plans in
Rwanda. Jean-Pierre Chrétien, “RTLM Propaganda: The Democratic Alibi,” p. 55. For more on Roger
Mucchielli’s work on propaganda, see Roger Mucchielli, Psychologie de la publicité et de la propagande:
Connaissance du problème, applications pratiques (Paris: ESF, 1972).
21
Taken from a document titled “Note relative à la propagande d’expansion et de recrutement,” written in
Butare prior to the genocide and inspired by Mucchielli’s work. The document, in all appearances, seemed a
brief on propaganda techniques to adopt to convince an audience. The author of the brief recommends using
“lies, exaggeration, ridicule, and innuendo” to discredit an opponent and attribute to him/her wrong intentions.
To do this, one can have recourse to ‘accusations in a mirror,’ for example, or to staging events that ‘lend
credence to propaganda.” In Human Rights Watch, Leave None to Tell the Story: Genocide in Rwanda (New
York: Human Rights Watch, 1999), p. 66.
19
246
Akanyaru river–like today in this Akanyaru River are found the bodies of Bahutu–
that will be when a mututsi learns that at muhutu is also a human being. 22
The idea of self-defence was repeatedly used during the years preceding the genocide.
According to Des Forges, in March 1992, a massacre in Bugesera occurred after Radio
Rwanda warned Hutu in the region that they had received information of an imminent Tutsi
attack on them, urging Hutu to strike first to protect themselves. 23 The following year,
groups of civilians recruited to help in the fight against the general Tutsi enemy, RPF and
Rwandese nationals, were constituted in a ‘self-defence force.’ The overall strategy
against the Tutsi was even referred to by Colonel Théoneste Bagosora, who played a
leading role in the coordination of the genocide, as a ‘national self-defence plan’. 24
On another level, action against the Tutsi was presented as the ‘only’ viable
response. As Des Forges explained, extremists expressed “the belief that previous
measures to end Tutsi control, killing some and driving away others–had failed and that
the only way to ensure they would never take power again was to eliminate them
completely.” 25 Hutu were reminded of the ruthlessness of their enemy. To escape
enslavement, or worst extermination, half measures would not do. The entire Tutsi
community had to be targeted. Removing the Tutsi threat meant eliminating the Tutsi
altogether. To drive in this point, references were often made to the mistake made in the
1960s, letting the Tutsi live. An article published in November 1993 stated that “if a Tutsi
punches you, take your machete right away and finish him off, if not he can strike back at
you. [...] The mistake that was made in 1959 is that the Hutu let the Tutsi escape.” 26 One
interviewee recalled a speech in which Hutu were asked to “rise up and remember that
22
Personal translation. Quoted in Bernard Mulinda, La liberté de la presse au Rwanda (1974-1994), pp. 80-81.
Alison Des Forges, “Call to Genocide: Radio in Rwanda, 1994,” in Allan Thompson, ed., The Media and the
Rwanda Genocide (London: Pluto Press/IDRC, 2007), p. 42.
24
Ibid, p. 43.
25
Alison Des Forges, “The Ideology of Genocide,” p. 46.
26
Personal translation. Quoted in Bernard Mulinda, La liberté de la presse au Rwanda (1974-1994), p. 83.
23
247
Kagame fled the country at three years old,” which, as he explained, stressed the need to
kill all Tutsi, including infants, to end it once and for all. 27
To further confirm sensibility of collective action against the Tutsi, the rhetoric
adopted during the civil war period stressed the likelihood of success. The press and
political speeches reminded the Hutu of their greatest strength, their numbers. Hutu could
rejoice in the fact that they possessed ‘demographic strength.’ For each Tutsi in the
country, there was a much larger number of Hutu. In a broadcast on Radio Rwanda
around March 1994, a Parti Libéral leader claimed that “[n]o-one can fight seven million
people. Forces drawn from ten percent of the population [the estimated number of Tutsi in
the country] cannot fight that many millions and win over them.” 28 Interviewees recalled
being told that “they are less numerous, they are one against seven. That means that you
can manage. That one person will shoot two people, not more, and five of your will
survive.” 29
Numerous radio broadcasts declared that victory was certain. The majority
people would prevail.
Overall, this resonating frame created during the civil war, calling on the Hutu
population to ‘rise up’ against the Tutsi, rested on an inherent tension between the severe
and urgent nature of the threat, and the propriety and chances of success of collective
action against this threat. Part of the tactic to create the sense the Tutsi represented a real
danger was to stress Tutsi wit and capacity to overcome obstacles in their way. As Des
Forges explained, discourses conveyed “the conviction that the Tutsi had been able in the
past to subjugate far larger numbers of Hutu because of superior intelligence or
27
Personal translation. Interviewee no. 20, Kigali, Rwanda, 29 March 2004.
African Rights, Rwanda: Death, Despair and Defiance, Revised Edition (London: African Rights, 1995), p.
79. In a similar vain, an interviewee recalled hearing a Hutu say “the Hutu are eight million and the Tutsi are
one million. So, eight old women can attack a single old woman, and eight young boys can attack one young
boy. [...] you will all disappear. You will not be saved, even if there were two million [RPF] soldiers.” Personal
translation. Interviewee no. 23, Kigali, Rwanda, 1 April 2004.
29
Personal translation. Interviewee no. 19, Kigali, Rwanda, 27 March 2004.
28
248
deviousness and still possessed the capacity to do so.” 30 At the same time, however,
rhetoric was built around the legitimacy, propriety and efficacy of Hutu collective action.
The rhetoric therefore rested on a tension between the fear of extinction, ‘they will destroy
us’ and hyper ethno-centrism, ‘we are righteous, strong and we will prevail.’ 31 Such a
tension did not inhibit mobilisation. Both frames, independently of the other, were built on
strong drives for action. A negative one, the confrontation with a dangerous enemy, called
for action against the threat, and a positive pole, inflated images of the group’s virtue and
strength, as well as beliefs in success, condoned action. Action, or reaction in an
aggressive form to a threat, real or perceived, rested on resolving the tension between
these two. It was, as Gurr explained, “‘a function of the individual’s perceived power to
control or hurt his frustrater relative to the frustrater’s power to control or harm him’.” 32
Extreme collective mobilisation occurred at the junction of these two powerful drives, fear
of extinction and ethno-centrism.
To understand this, however, requires moving beyond the classic framing agenda
and towards the social and psychological dynamics of groups. Frames developed during
the civil war fundamentally called forth perceptions of the ingroup in relation to perceptions
of the outgroup. They were about group dynamics. Hyper ethno-centrism and the fear of
extinction that underlay frames’ vocabularies of severity/urgency and propriety/efficacy are
fundamentally rooted in recognition of the group (I am one of them, therefore I am targeted
because we are targeted) and positive attachment to the group (we are great, strong,
righteous, etc.). But group solidarity is necessary for these feelings to operate.
To
understand what specifically targets Hutu solidarity within these broader resonant,
severe/urgent, appropriate/efficacious frames, it is necessary to focus on levers of
30
Alison Des Forges, “The Ideology of Genocide,” p. 46.
I wish to thank Prof. Thomas Homer-Dixon for pointing this tension out. Discussion with Thomas HomerDixon, Trudeau Centre for Peace and Conflict Studies, Toronto, Canada, 13 May 2005.
32
Quoting Leonard Berkowitz. Ted Robert Gurr, Why Men Rebel, p. 35.
31
249
solidarity, to look at opposing, politicising and simplifying, mechanisms for Hutu solidarity
found in the script developed by extremists during the Rwandese civil war.
Waking Up Hutu Solidarity: Opposing, Politicising, Simplifying
Looking at the framing mechanisms to foster Hutu solidarity prior to the Rwandese
genocide does not contradict the classic framing agenda, however. On the contrary, it
complements it. It holds the classic framing perspective as a general and generic process
which can be taken a step further by looking at some of the underlying mechanisms
behind framing, some of the linkage rules to make people buy into the group and feel, in
the name of the group, collective mobilisation.
Furthermore, the classic framing argument stressed the importance of Rwandese
history to understand the bases of frames built in the 1990s, to understand references to a
pre-existing ideology particularly entrenched by the Kayibanda and Habyarimana regime.
There is more to Rwandese history than this. The 1959-1990 period is not only about the
transmission of a belief system, of an overpowering ethno-centric ideology. It also marked
the constitution of a repertoire of actions for mobilisation, of framing strategies, stressing
Hutu solidarity and aimed at legitimising a young, untried regime. As Charles Tilly’s
concept of ‘repertoire’ implied, a recurrent use of tactics and strategies enshrines them as
reasonable means to attain goals. Tactics and strategies are not reinvented at every new
occasion. Those previously used and proven can come to form a repertoire, or tool box,
from which to draw when an opportunity arises. Innovations made by the Kayibanda
regime early on in terms of strategies to stoke up Hutu solidarity proved their potency. As a
result these strategies were repeatedly deployed in following years. They were eventually
recuperated by the Habyarimana regime in the early 1990s, a time at which these
strategies achieved an unparalleled success. Both the Kayibanda and Habyarimana
regimes, thus, employed ethno-centric rhetoric, strategies aimed at fostering Hutu
250
solidarity, to bolster the regime’s support in the face of uncertain popular backing or
threatened legitimacy. The rest of this chapter surveys the recourse to opposition,
policisation and simplification under Kayibanda and Habyarimana.
MAKING HUTU SOLIDARITY PART OF THE REPERTOIRE OF STRATEGIES
In the late 1950s, the Parmehutu platform inscribed the legitimacy of the nascent Hutu
regime in the virtue of the Rubanda Nyamwinshi. It inscribed the legitimacy of the new
Hutu Republic in the Hutu’s purported solidarity against the Tutsi monarchy, having
brought the majority people freedom from oppression. This served to advertise and create
solidarity where it had never existed. The Hutu had never been a unified group in previous
eras, whether as a result of being split in different kingdoms in ancient Rwanda, or, during
and following colonisation, split geographically, in different classes (labourers or
intellectuals) or because of different political affinities (in support of the monarchical
system or in support of the Revolution). 33 In other words, Kayibanda’s strategy aimed at
ensuring its political survival by inscribing it in a celebrated Hutu solidarity.
In the late 1980s and early 1990s, the Habyarimana regime used similar tactics. It
faced growing discontent at all levels. Confronted with waning legitimacy and the prospect
of sharing power in a liberalised political arena, the regime and a group of its backers
sought to stimulate support by tapping into a strong source of allegiance, Hutu solidarity.
This strategy was employed to recreate this solidarity, which had been diluted by quarrels
between Northern and Southern Hutu created by Habyarimana’s regional favouritism. 34 So
33
It should be recalled that a number of Hutu gave their allegiance to their former Tutsi patrons or to the Tutsi
royalty. Some even fled the country in the 1960s with Tutsi exiles.
34
The disunity of Hutu was sometimes attributed to the Tutsi. In an article published in Kangura in January
1991, it is argued that “The [RPF]’s objective is becoming clearer. To stop Rwanda from enjoying its
independence by creating a conflict between Hutu from Nduga [in the South-Centre of the country] and those
from Rukiga [in the North], a strategy deployed since 1967, when the refugees defeated militarily opted for
propaganda and lies. Because Hutu cannot see through rumours [...] Tutsi started crying crocodile tears while
claiming that the Bakiga [the Hutu from the North] were getting ready to exterminate the Banyenduga [the Hutu
from the South].” Personal translation. Reproduced in Bernard Mulinda, La liberté de la presse au Rwanda
(1974-1994), p. 76.
251
marked was the conflict between Hutu from the North and South that the regime even
feared, at the time of the RPF invasion, that Southern Hutu might align with the
attackers. 35 The regime sensed new attachments and allegiances were being created,
eating away its support base. One form of allegiance needed to be rekindled in their plan
to keep power, the unity of the Bahutu, which could potentially sweep away other
allegiances. In both cases, similar strategies were used for a similar political goal,
consolidating power.
In 1959 and early 1960s, the young Hutu political movement rested on the fragile
bases of a regime instituted in a coup or revolution, depending on how events are
interpreted. Either way, it was a new regime whose power base and legitimacy base
required solidification. In 1959, Kayibanda consolidated his bases by targeting Tutsi with
remaining ties to power after the ousting of the monarchy. Again, in 1961, following the
coup d’État de Gitarama, which allowed the Parmehutu to seize power at the national
level, Tutsi huts were pillaged and burned. The violence targeted opposition parties–
UNAR, RADER and Aprosoma–members. 36
Few Tutsi remained in official posts after
these events and a large number of them fled the country. In the years that followed, the
Tutsi refugees in bordering countries launched repeated attacks against Rwanda with the
aim of topping the Kayibanda regime. After every new attack, in retaliation for actions by
the refugees, the Rwandese authorities organised purges against Tutsi in Rwanda.
Kayibanda was clear about his intentions of holding the Tutsi population in Rwanda
hostage to refugees’ actions. The regime’s tactic was also a self-serving strategy. In early
days, new attacks by the Inyenzi, the ‘cockroaches’ operating at night, were opportunities
for the political authorities to rid themselves of opponents within Rwanda, on account of
ties between them and refugees.
35
René Lemarchand, “Rwanda: The Rationality of Genocide,” Issue: A Journal of Opinion, vol. 23, no. 2, 1995,
p. 9.
36
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 94.
252
One attack in particular, conducted by exiles in Burundi in December 1963,
resulted in the defeat of the refugees, but also the subsequent arrest and execution of
some of the few political opponents left in Rwanda, mainly senior officials and leaders of
the UNAR and RADER parties. 37 In the countryside, the regime also seized the
opportunity to build its popular support. National authorities ensured that local politicians
and administrative representatives fuelled anger against the Tutsi and organised the
population in self-defence groups targeting the Tutsi community. Local authorities drew
links between the attackers and Tutsi within the country, “whom they accused of aiding the
invaders.” 38 More than 10 000 Tutsi were massacred in the following days in what
Bertrand Russell called “the most horrible and systematic massacre since the
extermination of Jews by the Nazis.” 39 For political authorities, targeting the Tutsi allowed
them to consolidate their power. As Des Forges argued, these authorities benefited from
this strategy “both directly, by showing they were strong enough to get people killed, and
indirectly, by confiscating and then redistributing the assets of victims who had been killed
or driven away.” 40 Recently instituted as the politically favoured community, the Hutu
community was still resource poor.
Taking the fleeing or killed Tutsi’s assets gave
authorities the means to reward potential political supporters. André Nkeramugaba, a
politician of the area where the massacres took place, used these resources in more than
one way to win the following year’s election to the National Assembly. In the election he
chose as a slogan “[i]f I am not elected, charges may be brought against you; but if I am
elected I shall do my best to prevent all investigations.” 41 Overall, anger projected at the
Tutsi and parallel affirmation of Hutu solidarity served to provide grounding for the new
Hutu political authorities of the late 1950s and early 1960s.
37
Ibid., p. 97.
Alison Des Forges, “The Ideology of Genocide,” p. 45.
39
In an interview with Le Monde, quoted in Marcel Kabanda, “Kangura,” p. 69.
40
Alison Des Forges, “The Ideology of Genocide,” p. 45.
41
René Lemarchand, Rwanda and Burundi (New York: Praeger, 1970), p. 226.
38
253
The same tactic was used by the Habyarimana regime at the start of the civil war to
consolidate its power base, shaken by “another crisis of governance, a struggle over who
would control the state.” 42 The regime and its backers took the Tutsi population hostage in
retaliation for RPF attacks. 43 In the days following the initial RPF attack on 1 October,
more than 9 000 people were arrested. Among these were a large majority of Tutsi, but
also Hutu opponents and critics of the government. RPF actions became an excuse for
arbitrary denunciations, arrests and the suppression of political detractors of the regime.
As in the past, to succeed in this strategy, incumbents extended the blame to the
Rwandese Tutsi population, framing them as potential accomplices of the RPF within
Rwanda. Actions taken against Rwandese nationals were justified as part of the war on
the enemy, particularly the ‘enemy within.’ They also provided the government with a
known plausible scapegoat against which to redirect popular frustrations. This, in turn,
provided a rationale for a rekindled Hutu solidarity overshadowing quarrels between the
North and the South that had been growing as a result of Habyarimana’s favouritism
towards the North. It provided a justification for unity led by Hutu elites, in defence of the
community threatened by the Tutsi and their ploys to reinstitute their monarchy and
bondage system. The strategy was systematically applied over the next few years. At
each politically contentious moment, every time the regime and its backers faced a
renewed threat to their privileges–at important times in the democratisation process or
around the finalisation of important protocols of the Arusha Peace Process–anti-Tutsi
violence was encouraged. Félicien Gatabazi explained that “[e]ach time there are some
42
Catharine Newbury, “Ethnicity and the Politics of History in Rwanda,” Africa Today, vol. 45, no. 1, 1998, pp.
17-18.
43
Clear warnings were also sent to the RPF in the months prior to the genocide, warning them of the
consequences if they took up arms again. Hassan Ngeze, in an article published in February 1994, stated that
“[i]f war was to start again, an incredible number of Tutsi would die. [...] The Tutsi started the conflict, but it is
not only Hutu who died in it, like [the Tutsi] had wished it [...] Before starting another war, the [RPF] should start
by warning Tutsi to flee. If not, anything is possible.” Personal translation. Quoted in Bernard Mulinda, La
liberté de la presse au Rwanda (1974-1994), p. 83. In an article published in March 1994 in La médaille
Nyiramacibiri, the author explained that “if the [RPF] dared once more start their attacks [... i]t is at that moment
that the Tutsi would be exterminated [...] and it is possible that their ethnic group become extinguished.”
Personal translation. Quoted in Bernard Mulinda, La liberté de la presse au Rwanda (1974-1994), p. 78.
254
difficulties (in the democratic process) there is a flare-up of tribal violence instigated by the
regime, and threats of civil war are used to justify the status quo.” 44 So was the case with
the Arusha peace process, renewed violence against the Tutsi was used to disrupt
negotiations. 45
Commenting on the instrumentalisation of anti-Tutsi sentiments and violence,
Catharine Newbury explained that, between the actions taken by the Kayibanda regime
and those of the extremists in the 1990s, “one can see a recurrent pattern: the tendency
for a regime threatened by external attack to target an internal scapegoat and to
rationalize its behavior by propagating a corporate view of ethnicity.” 46 More than a
reaction to an external threat, however, it constituted the strategy of two regimes, one
inexperienced, the other discredited, to overcome their fragile legitimacy and secure their
hold on the country. For these two successive Hutu regimes in Rwanda it was a survival
tactic; it served to ground their authority in an alternative form of legitimacy, not democratic
but communal. An innovative strategy built on an existing theme, perceptions of Hutu-Tutsi
animosity developed in the last centuries of monarchical regime and during colonial
administration, tried and tested in a few crises involving refugees in the early 1960s, it
came to be inscribed as a means to attain political goals. It became part of the repertoire
for action available to elites. It was employed on different occasions, sometimes with
success, others without. Although Kayibanda agitated communal sentiments in 1972, he
was ousted by a coup the following year. Yet, it eventually achieved dreadful success in
the case of the Hutu support for the extremist agenda in the 1990s.
For the population, those outside of the political arena, this recurrent use of
violence over the course of decadess had a strong impact. With every new crisis, violence
44
Quoted in Gérard Prunier, The Rwanda Crisis, p. 144.
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 160.
46
Catharine Newbury, “Ethnicity and the Politics of History in Rwanda,” p. 16.
45
255
was increasingly forming a part of their normal existence. 47 Consequently, they were
progressively becoming desensitised to it. 48 A disposition remained. Social psychological
explanations speak of this as the “destruction of intergroup social capital and the
hardening of identity categories that are likely to follow” episodes of violence. 49 As one
interviewee recalled, “they fought and after remade the bonds, but it never fully eliminated
or erased what had happened.” 50 On the surface the regime-sponsored ideology was
disseminated, an effective mechanism for mobilisation and solidarity. It was also ingrained
through this slow acceptance of the regular recourse to violence. Underneath the surface,
the slow process of socialisation within this environment led to the travesty of values and
standards, in the form of an acceptance of discrimination, injustice and even violence. In
turn, it reconfigured the collective matrix of acceptable thinking and behaviour. 51
Increasingly, segregation and even violence moved within the realm of normality and
became an accepted part of life. They came to form part of the repertoire of actions taken
in the community–when authorities demanded it.
Violence was not automatic, however. The Hutu population did not, mob-like, throw
itself at the Tutsi. The massacres of the First and Second Republic were the result of
coaxing by authorities. To achieve Hutu solidarity, ethno-centric entrepreneurs had to sell
this solidarity. They had to convince the population to unite. For this, they had recourse to
certain mechanisms to achieve solidarity. In their actions, three in particular can be traced:
opposition, politicisation and simplification.
47
Human Rights Watch, Leave None to Tell the Story, p. 4.
Peter Uvin described over time a “dehumanization and increased emotional distance from the target of
violence,” on the part of communities engaging in violence. Peter Uvin, Aiding Violence: The Development
Enterprise in Rwanda (West Hartford, Conn.: Kumarian Press, 1998) p. 34. Jean Hatzfeld cited a condemned
killer as indeed saying that “[w]e did not see humans anymore when we caught Tutsi in the marshes. I mean
people like us, sharing similar thoughts and feelings. The hunt was wild, the hunters were wild, the prey was
wild, savagery took hold of the mind.” Personal translation. Jean Hatzfeld, Une saison de machettes, p. 57.
49
Peter Uvin, “Reading the Rwandan Genocide,” p. 85.
50
Personal translation. Interviewee no.12, Butare, Rwanda, 23 March 2004.
51
Interviewee no. 14, Butare, Rwanda, 24 March 2004.
48
256
OPPOSITION
In Rwanda, recourse to opposition tactics began prior to the 1959 Social Revolution. The
Tutsi during the colonial period encouraged myths of their difference to ground their right
to a superior status. On their side, prior to the 1959 Social Revolution, the Hutu used their
fight against the Tutsi administrative and political system as a platform for mobilisation.
This served to inscribe Hutu-Tutsi antagonism as a strong, but path-dependent, discourse
for reading future events. Particularly for Kayibanda’s Parmehutu, which built its struggle
around freeing the masses from Tutsi oppression, this rhetoric constrained what could be
said and done. In a sense, for Kayibanda and his party, adopting that platform–adopting it
in an uncompromising way by selling Hutu-Tutsi antagonism as a secular one–and
obtaining power in its name, made any dilution of the rhetoric a betrayal of the spirit of the
Parmehutu. Obviously, the impact of this ingrained perspective was not deterministic. It
needed not be employed in following years, but its adoption in early years made changing
course increasingly less likely. Antagonistic portrayals of relations between Hutu and Tutsi
continued to be a feature of the two Hutu Republics in the decades that followed. This
antagonism was not only used as an amorphous ideology, however. It was also turned into
a tool to stir the sentiments of the Hutu community on occasion. Two pivotal moments
were the early 1960s and the early 1990s.
Recourse to this opposition mechanism was straight forward in both instances.
Both situations were read through the prism of antagonistic group dynamics. In each
case, the situation was read as an opposition of two corporate Manichean identities. In
both cases, framing rested on the use of strong dualities, such as foreigner-autochthon,
minority-majority, aristocracy-mass, oppressor-victim, enemy of the state-good patriot, evil-
257
good, cunning-trustworthy, lazy-hardworking. 52 The opposition centered on two poles, a
negative and a positive one. One was the embodiment of the negativity in the relationship.
Malevolent intentions were indeed ascribed to the Tutsi, depicted as haughty, interested
only in their kind. The Hutu, on the other hand, were portrayed as the eternal target of
Tutsi ploys, courageously, rightfully defending themselves against the Tutsi. As a result,
positions were irreconcilable. To think otherwise was to gravely misunderstand these two
communities. To think otherwise was a betrayal of the nature of the groups. 53 In essence,
the two groups were, due to the actions of one, caught in an extreme competition. They
were in a secular, intractable zero-sum conflict. This was the pursuit of a 400 years old war
between Hamite and Bantu. The stakes did not allow for a middle-ground. These stakes
were too high. In 1959 and the early 1960s, these stakes were the freedom of the Hutu
people. In the early 1990s, they were life, the survival of the Hutu people. In both
instances, Hutu and Tutsi were fundamentally antithetical identities and the opposition was
zero-sum.
Opposition was conveyed by employing to rhetorical strategies, by attributing to the
Tutsi threatening intentions and by claiming Hutu moral superiority.
Underlying a
vocabulary of urgency/severity lay references to Tutsi’s culpability and the threat all Tutsi
posed. Rhetoric developed to describe the Tutsi in 1959 and the 1960s centered on their
role as past oppressors. This characterisation not only described their previous status; it
was extended to their nature. If the Tutsi had acted as ruthless tyrants in the past, it was
because they were in essence a power-hungry people. Even having been ousted from
power and scared into fleeing, their true nature as an oppressive creed could not be
changed. From the borders of Burundi, Congo, Uganda and Tanzania, they plotted their
return. In a speech on the 1959 Social Revolution, Kayibanda reminded the Hutu that “had
52
53
Interviewee no. 14, Butare, Rwanda, 24 March 2004.
Jean-Marie Vianney Higiro in Bernard Mulinda, La liberté de la presse au Rwanda (1974-1994), p. 79.
258
the Tutsi won, no Hutu would have been left alive save for the ones that would have
accepted ten times to be bound to them through the bond of [Ubuhake].” 54 Playing on the
“completeness of Tutsi control” and the threat of the reinstitution of the Tutsi monarchy,
Hutu authorities stirred up anger against the Tutsi and, as a result, they lifted their own
popularity as decriers of Tutsi ploys. 55 To ensure their freedom, the Hutu had better side
with their creed.
In the early 1990s, the rhetoric adopted by extremists played on the same theme
as in 1959: the danger posed by Tutsi refugees returning to enslave the Hutu anew. The
famous ‘Ten Hutu Commandments’ published early in the civil war reminded the Hutu to
be wary of the Tutsi whose sole aim “is the supremacy of his ethnic group.” 56 Questioning
this was treason, “[t]he Social Revolution of 1959, the Referendum of 1961, and the Hutu
Ideology, must be taught to every Hutu at every level. Any Hutu who persecutes his
brother Hutu for having read, spread and taught this ideology, is a traitor.” 57 It became
increasingly so as discourses shifted with definitions of the threat. In the 1960s, the Tutsi
were castigated as oppressors. By the 1990s, Tutsi were said to have genocidal
ambitions. A pamphlet produced in 1991 attributed to the RPF and its accomplices, on top
of recovering power in Rwanda, three other objectives: “Extermination of the HUTU
majority ethnic group; [t]he slaughter of political and administrative authorities; [the]
Slaughter of tutsi who refused to collaborate with the aggressors.” 58 A report produced by
a commission of military officers distributed in 1992 insisted that the Tutsi enemy would
never stop and would use all means to achieve its ends. 59 The situation was clear, there
54
Speech by Grégoire Kayibanda given in 1964, reprinted in the December 1991 issue of Kangura, quoted in
Jean-Paul Kimonyo, La participation populaire au Rwanda, p. 578.
55
Alison Des Forges, “The Ideology of Genocide,” p. 45.
56
Samantha Power, “A Problem from Hell”: America and the Age of Genocide (New York: Harper Collins,
2002), p. 339.
57
Ibid.
58
Emphasis in original. African Rights, Rwanda, pp. 69-70.
59
Human Rights Watch, Leave None to Tell the Story, p. 62.
259
was only one front: “[t]he war [was] between the Tutsi and the Hutus and the only solution
[was] public awakening.” 60
With stakes such as these, survival of the Hutu, the situation was polar and
required solidarity. One was either backing the Hutu cause or undermining it.
Disagreement, inaction or collaboration was ultimately dangerous for the Hutu. On this
point, David Newbury noted that “[f]or genocide, if ‘everyone’ from one group is involved,
then killing others becomes not only acceptable but necessary [...] dissent is often seen as
treason.” 61 The journal Kangura warned its readers that “when the majority people is
divided, [then] the minority becomes the majority.” 62 Léon Mugesera, one of the extremists’
ideologues, reminded the Hutu in a speech to “[r]emember that the person whose life you
save will certainly not save yours.” 63 In this either-or situation, threat overpowered previous
intergroup fraternities. A convicted killer explained that the Tutsi “had become a threat
superior to everything they had lived together before, that surpassed what they view of life
on the [hill]. That is how we reasoned and how we killed at the time.” 64
While the Tutsi were portrayed as a threat both in the 1960s and in the 1990s, the
Hutu were portrayed as the embodiment of righteousness. They were the morally superior
group. 1959 and the events of the early 1960s constituted an important shift for this type
of discourse. Prior to the 1959 Hutu Social Revolution, Hutu identity had been a reactive
one. It was positioned in opposition to Tutsi identity and the oppressive bondage system
the Tutsi imposed.
With the Social Revolution came the opportunity to develop the
positive pole in this antagonistic relationship between Hutu and Tutsi, to ascribe specific
positive meaning to the Hutu in this relationship. The Hutu could now be portrayed as the
bearers of freedom for the country, since they had liberated it from Tutsi oppression. They
60
Quoted in Marcel Kabanda, “Kangura,” p. 67.
David Newbury, “Understanding Genocide,” African Studies Review, vol. 41, no. 1, 1998, p. 77.
62
Ibid., p. 82.
63
Quoted in African Rights, Rwanda, p. 77.
64
Personal translation. Quoted in Jean Hatzfeld, Une saison de machettes, p. 145.
61
260
could be described as the ones who fought to redress injustices. Rhetoric in the early days
of the Kayibanda regime built on this. As Human Rights Watch explained, “they crafted the
myth of the Hutu Revolution as a long and courageous struggle against ruthless forces of
repression. For them, the battle had been legitimate as well as brave.” 65 The Hutu cause
was a moral one. Hutu solidarity, a stand against refugee raids, was a stand for the
defence of the Revolution and its higher principles.
Similarly, in the early 1990s, rhetoric adopted by the extremists promoted the
legitimacy of the Hutu cause. So self-righteous were the Hutu that, according to JeanPierre Chrétien, propagandists “invoked [...] God, the Holy Family and the Virgin Mary [...]
all mobilized for the sacred cause of the Hutu people.” 66 In light of the mortal threat posed
by the Tutsi, the idea of the righteous cause took on another meaning. Defence went
beyond defending one’s life. Extinction of the Hutu meant the extinction of the principles of
the Hutu Revolution. Hutu solidarity was, therefore, the defence of a chosen people. It was
the defence of a higher moral order.
POLITICISATION
Although they find their source in communities, social interactions and culture, it is
fundamentally contentious, politicised identities that were agitated at times of conflict in
recent Rwandese history. Both in the early days of the Kayibanda regime and in the last
days of Habyarimana’s, it is claims tied to rights, citizenship and the state that were used
to justify the need for Hutu solidarity in the face of a Manichean confrontation with the
Tutsi. This transpires from the fact that the principal reading key used to describe HutuTutsi antagonism in 1959, the early 1960s and in the early 1990s was the opposition
between feudalists/monarchists and democrats.
65
66
Human Rights Watch, Leave None to Tell the Story, p. 39.
Jean-Pierre Chrétien, “RTLM Propaganda,” p. 58.
261
In the midst of instituting a new regime, Kayibanda and the Parmehutu pushed their
political agenda. Building on their ideological bases of the mid-1950s, “’Hamitization’ [as]
a Tutsi plot to exclude [Hutu] from positions of responsibility” and the Hutu population from
upper echelons of society, they worked on the agenda of Hutu rights. 67 From the start
these were set as the terms of reference of the Hutu-Tutsi conflict. As the Manifeste des
Bahutu, a statement pivotal for the events of the late 1950s and early 1960s, explained,
“[t]he problem is above all a problem of political monopoly which is held by one race, the
Tutsi; political monopoly which, given the totality of current structures becomes an
economic and social monopoly which, given the de facto discrimination in education, ends
up being a cultural monopoly.” 68
The fight between Hutu and Tutsi was first and foremost sold as a defence of Hutu
rights. It transpired in how the political entrepreneurs of the time recouped the Hamitic
myth, the foreigner-native opposition. It served an important purpose, establishing groups’
rights. In many African cultures, rights to land are established by possession of the land.
The first group on a territory claims ownership of it. 69 Co-opting the Hamitic myth, the
regime could therefore claim that as the last to arrive in Rwanda, the Tutsi had taken
control of the land from its rightful owners, the Hutu. The Hutu were made to recall that
they had toiled the land, prepared it for agriculture, centuries before the Tutsi’s arrival.
The Hutu were the Benesabahinzi, the ‘sons of cultivators,’ the children of the land.
Rwanda belonged to the Hutu. As such, the Tutsi had to negotiate their presence on the
territory, not impose it. The paradox, of course, was that, according the same historical
accounts, the Hutu had arrived in Rwanda after the Twa. According to these histories, the
Twa had been the first inhabitants of the land. If the Hutu claimed to be the rightful owners
of the land, according to this logic, it was by ignoring the rights of the Twa.
67
René Lemarchand, “Rwanda,” p. 9.
Quoted in Mahmood Mamdani, When Victims Become Killers: Colonialism, Nativism, and the Genocide in
Rwanda (Princeton, N.J.: Princeton University Press, 2001), p. 116.
69
Interviewee no. 14, Butare, Rwanda, 24 March 2004.
68
262
This reaffirmation of Hutu rights shed new light on the Tutsi monarchical system. It
was necessarily illegitimate, having been imposed by usurpers. Dominated by the Tutsi
minority, it violated the rights of the Hutu. The fight was one of emancipation. The Hutu
political system emerging from the 1959 Revolution and 1961 Referendum redressed past
injustices. It constituted a fair democratic political system. It finally represented the
interests of the majority, the Rubanda Nyamwinshi, the populous Hutu. The choice of
words, ‘revolution’ and ‘referendum’–instead of coup–played on notions of demographic
and democratic majority, enshrining the legitimacy of the Kayibanda regime. As a result,
attacks on the country by refugees in neighbouring states could be castigated as affronts
to a rightful regime. Rwanda had finally achieved democracy and these attacks could only
be the spectre of feudalism trying to come back. In a speech in April 1964, the president of
the National Assembly described the actions by Hutu communities following the previous
December’s attack on the country by Tutsi refugees in these words:
[a]s soon as the Hutu became aware of the atrocities perpetrated [...] they
understood the great danger of returning to prior circumstances. They remembered
the abuse they endured under feudal rule. [...] They then felt a great anger and
vowed not to fall victim to the fate of losers. The anger was intensified by the fact
that the former servant had, for four years, experienced the flavourful treats of
democracy. 70
Hutu solidarity in the face of these attacks was a stand for Hutu rights and freedom.
It is in terms of this same political battle that the situation in the early 1990s was
framed. New ideologues chose to justify their actions by making the same claims. In what
Chrétien referred to as ‘racist propaganda wearing the mask of democracy’, the Rwandese
Patriotic Front’s attack was explained as: “it is myth attacking right, it is feudal nostalgia
pursuing mercilessly, and arming itself against, democracy.” 71 Despite the RPF’s claims to
the contrary, Habyarimana proclaimed that the RPF and its accomplices sought the
70
Quoted in Marcel Kabanda, “Kangura,” p. 70.
Jean-Pierre Chrétien, “RTLM Propaganda,” p. 60. Excerpts personally translated and quoted in Bernard
Mulinda, La liberté de la presse au Rwanda (1974-1994), p. 75.
71
263
reestablishment of monarchical rule and Tutsi supremacy. 72 As Jean-Marie Vianney Higiro
explained, the attack “was an attempt to roll back the social and economic progress made
by the Hutus since the social revolution of 1959.” 73 To reinforce the importance of this
claim, the rhetoric recalled the misery of Tutsi oppression. An article in Kangura, for
example, explained that during centuries under Tutsi rule, “never was the Hutu master of
his work. When he cultivated [the land], the Tutsi harvested. [...] The riches and all assets
that the Hutu gave the country when they arrived in Rwanda were confiscated by the Tutsi
monarchy; while [the Hutu] rotted in poverty, helpless.” 74 Rhetoric played on Hutu
victimisation at the hands of Tutsi monarchs.
The difficulty with making the monarchy-democracy opposition a reading key of the
civil war was that, by the early 1990s, most Rwandese had never lived under the Tutsi
monarchy. 75 Why would they have perceived it as so menacing? Historical legacy played
a part in these perceptions–a historical legacy of decades-old teachings about the
wrongdoings of the monarchical system. Indeed, the perceived threat posed by the return
of this oppressive regime might actually have been more convincing because most had not
experienced it. Indirect education could scapegoat the monarchy, but the masses left
without the resources to distinguish between fact and fiction were incapable of
discernment. In the opinion of a Rwandese intellectual, these individuals were actually
more dangerous because whatever judgment they made was one based on the opinions
72
Catharine Newbury, “Ethnicity and the Politics of History in Rwanda,” p. 15.
Jean-Marie Vianney Higiro, “Rwandan Private Print Media on the Eve of the Genocide,” in Allan Thompson,
ed., The Media and the Rwanda Genocide (London: Pluto Press/IDRC, 2007), p. 74.
74
Personal translation. Quoted in Bernard Mulinda, La liberté de la presse au Rwanda (1974-1994), p. 79.
75
Demographic statistics for the early 1990s are hard to obtain. Although there was a census conducted in
1992, data from it found in secondary literature is incomplete. According to recent data, life expectancy at birth
was
44,1
years
in
2005.
World
Bank,
Rwanda:
Quick
Facts,
http://web.worldbank.org/WBSITE/EXTERNAL/COUNTRIES/AFRICAEXT/RWANDAEXTN/0,,menuPK:368741
~pagePK:141132~piPK:141109~theSitePK:368651,00.html, (last accessed 31 May 2007). A population
pyramid of Rwanda in 2000 produced by the U.S. Census Bureau indicated that the country had the population
distribution of a developing country, the largest segment of the population was less than 20 years old. United
States Census Bureau, Population Pyramid Summary for Rwanda, http://www.census.gov/cgibin/ipc/idbpyrs.pl?cty=RW&out=s&ymax=250, (last accessed 31 May 2007). If these trends are taken to be
somewhat representative of Rwanda in the early 1990s, 30 to 35 years after the overthrow of the Tutsi
monarchy, the major part of the Rwandese population had not lived under Tutsi rule.
73
264
of others, based on teachings they had internalised. Someone who had lived under the
monarchy and known it first hand would have been capable, on the contrary, of a critical or
more moderate judgment. 76 With little to go on to contest the interpretation of authorities,
the monarchy could be presented as an absolute threat for the Hutu and its image be used
to nurture a growing sense of fear.
Ideologues did deploy efforts to adjust the portrayal of the threat to the reality at the
time, updating it to the reality of the early 1990s. Kangura and its editor, Hassan Ngeze,
claimed that while the Hutu had let their guard down, enjoying the peace and contentment
of the two Republics, the Tutsi had relentlessly plotted since the Revolution and “managed
to reverse the former position of the Hutu on the social, cultural and political fronts.” 77 This
had gone on enough to ensure that, insidiously, the Tutsi had, for all intents and purposes,
regained control of the country. These claims were reminiscent of the accusations brought
against Jewish financiers by the German Nazi regime. Statistics were provided to illustrate
the imbalance between the Tutsi’s demographic weight and their power. One article
published in 1991 stated that the Tutsi “hold prominent positions everywhere. However,
this ethnic group constitutes 10 percent of the population. National wealth, trade and
industry are in the hands of the Batutsi, who often use civil and military authorities as a
cover-up.” 78
Overall, then, attacking the country to reinstitute a monarchical system after years
of weakening Rwanda from inside, the Tutsi were ‘enemies of the state,’ ‘enemies of the
nation,’ and their Hutu accomplices–often Hutu moderates critical of the extremists–were,
therefore, ‘traitors.’ 79 Fighting them was fighting for Hutu freedom, Hutu democracy and
Hutu rights. This was a historical fight for the country. As a radio announcer claimed on
76
Interviewee no. 8, Kigali, Rwanda, 16 March 2004.
Marcel Kabanda, “Kangura,” p. 63.
78
Ibid., p. 64.
79
Catharine Newbury, “Ethnicity and the Politics of History in Rwanda,” p. 7; Alison Des Forges, “Call to
Genocide”; René Lemarchand, “Rwanda,” p. 11.
77
265
Radio télévision libre des milles collines, “I am convinced that we are in the middle of a
revolution, a revolution similar to that of 1959, one that I think is an ultimate revolution.” 80
Building on polarised political identities, the message sent was that “[t]his Rwanda is mine.
I am the majority. It is I, first and foremost who will decide, it is not you.” 81 Rwanda, its
political system, political representation and rights, were not objects that could be divided
between such opposed groups. As such, the Hutu had better chose the right camp, the
fight for the rights of the majority.
SIMPLIFICATION
Fostering solidarity in Rwanda, turning the many into one driven collective, also involved
simplification.
A number of means were used to this end under Kayibanda and
Habyarimana.
In the early days of Kayibanda’s Parmehutu movement, an initial simplifying
strategy was to eliminate political opponents, Tutsi notables of the prior regime and
members of the new parties.
While this obviously served to physically get rid of
opposition, it also served the Parmehutu’s cause on another level. All these opponents
also represented alternatives to the vision the Parmehutu defended. They represented
other interpretations of the situation, which served to provide nuance, to contrast the
Parmehutu platform. The UNAR party, for example, defended the Tutsi monarchy; the
RADER proposed a non-ethno-centric, accommodating alternative. Eliminating these
individuals also meant silencing competing views, limiting options. As Catharine Newbury
explained, “[a] major characteristic of the 1950s conflicts [...] was the destruction of the
80
Reproduced in Christine L. Kellow and H. Leslie Steeves, “The Role of Radio in the Rwandan Genocide,”
Journal of Communication, vol. 48, no. 3, 1998, p. 120.
81
Kantano Habimana, RTLM announcer quoted in Jean-Pierre Chrétien, “RTLM Propaganda,” p. 57.
266
political middle ground.” 82 Once alternatives were eliminated, the Parmehutu platform
could predominate. It could be made the principal reading key of the situation.
With alternatives gone, nuanced views eliminated (RADER, for example, proposed
accommodation between groups as its platform) and dissenting voices extinguished, the
Parmehutu was given the freedom to inflate the ‘Tutsi threat’. With few to contest,
categories could be globalised. With one voice left, it became easier to promote an
essential reading of groups. In particular, all Tutsi could be ascribed the same essential
characteristics. All Tutsi could be simplified, ‘singularised’, as the same character, each
unchangingly a representative of “one particular race, the Mututsi.” 83
It was possible to
speak of one unified category, ‘le Tutsi,’ of a corporate identity. 84 In a sense, then, it also
implied there were no exceptions in the population. All Tutsi could be made out to be
accomplices in a ploy to defend their race’s interest. Each had to be treated with suspicion.
In turn, as Alison Des Forges argues, if ‘they’ “[were] perceived as one, invariable, united,
ignoring divisions among themselves, so we need[ed] to ignore divisions among
ourselves.” 85
To help promote this clear and accessible reading of group dynamics, the regime
adopted ‘populist’ strategies. It employed simple resonating ideas, often couching their
depictions of groups in stereotypes and images the population was familiar with.
Agricultural and religious references ensured the rhetoric was grounded in the population’s
reality. Another strategy was using popular channels to communicate these views.
According to Kimonyo, a group of singers was even hired by the regime to propagate their
ideology. Their songs “played continually on the radio and some songs even became
82
Catharine Newbury, “Ethnicity and the Politics of History in Rwanda,” p. 15.
Le Manifeste des Bahutu, quoted in Marcel Kabanda, “Kangura,” p. 63.
84
See Catharine Newbury, “Ethnicity and the Politics of History in Rwanda,” pp. 7-24.
85
Interview with Alison Des Forges, Human Rights Watch, Kigali, Rwanda, 4 March 2004.
83
267
popular hits. Some had a sad tone, pitying the Hutu for his long servitude, others a
triumphant one, celebrating the revolution and the majority people’s victory.” 86
In the early 1990s, extremists in Rwanda adopted strikingly similar strategies. They
silenced their competitors, and strengthened their case against the RPF and its
accomplices by promoting corporate views of group identities. Throughout this time, tactics
were found to eliminate opposing views. At its most basic, this consisted in physically
eliminating opponents. Prior to the RPF attack on the country, incumbents had begun
using assassinations to rid themselves of dissenting individuals. The October 1990 attack
was another grand occasion to eliminate opponents. Soon after the RPF had crossed the
border, the Habyarimana regime imprisoned thousands of people, many of whom
threatened the government’s credibility much more than the country. Four years later, in
the early days of the genocide, the first victims were moderates and political opponents left
in Rwandese society. Throughout the civil war period, extremists also sought to silence
their opponents by discrediting them. During those years, a ‘war of characterisations’ took
place between supporters of the Habyarimana and extremist camp and defenders of
opposition parties. Both sides sought to damage the reputation of the other through vile
caricatures. Extremists strove, in particular, to discredit opposition leaders as “RPF
puppets, traitors and embezzlers of public funds, demagogues, opportunists and idiots
motivated by the desire to settle scores with President Habyarimana.” 87 In many instances,
ethno-centric entrepreneurs even presented political opponents as disturbing elements
intent on taking power and destroying the Hutu Republic, thus defining them as enemies of
the state.
Following a simplification logic, the regime also succeeded in constructing the
notion of danger around the entire Tutsi community through their ‘enemy within’ discourse.
86
87
Jean-Paul Kimonyo, La participation populaire au Rwanda, pp. 98-99.
Jean-Marie Vianney Higiro, “Rwanda Private Print Media on the Eve of the Genocide,” p. 74.
268
They globalised categories by reinterpreting the danger posed by the RPF and associating
it with internal accomplices participating in the Tutsi plot to take power. The regime’s
interpretation of who constituted a threat shifted over time. It went from links between
1960s refugees and the RPF, to links between the RPF and Rwandese nationals of Tutsi
affiliation, as RPF accomplices, to finally blurring all lines, making all Rwandese Tutsi
‘enemies within’. They achieved the ‘generalization of blame.” 88 All Rwandese Tutsi within
the country were suspected of taking part in a plan to defeat the Hutu. Tutsi within the
country could not help but serve their creed. Extremists insisted on what they saw as a
natural Tutsi solidarity. In a clear attempt to blame all Tutsi, a Kangura article called in
1992 on the Hutu to: “[k]now that a prideful and blood-thirsty minority walks among you to
dilute you, to divide you, to dominate you and massacre you. [...] The nation is artificial, but
ethnicity is natural.” 89
Paradoxically, this was at times even achieved by accusing the Tutsi themselves of
globalising categories, of seeing Hutu undifferentiatedly, which justified doing the same to
the Tutsi. In an article titled “The Sovereign Rwanda Needs Hutu Unity and Solidarity,”
Kangura claimed that “[a]ll Hutu should know, when the feudalists will arrive in Rwanda,
they will not make a distinction between Hutu from the North and Hutu from the South,
they should know that it will be the end of them all.” 90 This served to further drive in the
need for Hutu solidarity. There could not be any doubt of Tutsi intent, they all shared the
plan to enslave or eliminate the Hutu. There needed not be hesitation, there were no good
Tutsi.
Communicating this vision also centered on populist strategies. Language used by
extremists borrowed from people’s existence, with references to their personal, social and
work environments. Christine Kellow and Leslie Steeves explained that “[t]he frequent use
88
Catharine Newbury, “Ethnicity and the Politics of History in Rwanda,” p. 7.
Personal translation. Kangura quoted in Jean-Pierre Chrétien, “Rwanda: la propagande du génocide,” in
Reporters sans Frontières, Les médias de la haine (Paris: La Découverte, 1995), p. 49.
90
Personal translation. Quoted in Bernard Mulinda, La liberté de la presse au Rwanda (1974-1994), p. 77.
89
269
of popular culture, biblical references, and familiar history context strengthened the power
of the broadcasts.” 91 The rhetoric used by RTLM, for example, was so in synch with
people’s reality that many described as listening to friends conversing around a banana
beer. In many instances, ethno-centric discourse was framed as entertaining, which
facilitated engaging with it. The media promoted stereotypes, turning them into jokes.
Newspapers frequently used caricatures, while the radio adopted, outside of interviews
and news bulletin, a light tone. People laughed, discussed broadcasts and caricatures
with family members, friends and colleagues. Not only was it ‘racist propaganda wearing
the mask of democracy’, as Chrétien described it, but it was also racist propaganda
disguised as entertainment. 92 This form of entertainment complemented serious
broadcasts. While serious news bulletins, on RTLM for example, were intended to drive
fear into people’s hearts, light but ethno-centric or racist broadcasts brought clichés,
stereotypes, and even hatred, home.
Opposition, politicisation and simplification strategies to foster Hutu solidarity
served their purpose in the early 1990s. Hutu-Tutsi relations were cast as a binary
opposition between two arch-enemies, one the villain, blood-thirsty and power hungry,
intent of enslaving, even massacring, the Hutu. Hutu were portrayed as their polar
opposite. They were the righteous ones, good, hard-working, the first to make Rwanda
what it is and the first to defend it from ill-intentioned usurpers. They made this fight a
political one, one about Hutu rights, freedom and power, and more importantly about Hutu
independence from the bonds of oppression. This served to drive home the importance of
the collective fight. It was necessarily about the group. Beyond the self, it was about
defending what the Hutu were, defending the Hutu legacy. And efforts combined to create
a simplified, powerful rendering of the situation. Not only were fall back positions
91
92
Christine L. Kellow and H. Leslie Steeves, “The Role of Radio in the Rwandan Genocide,” p. 119.
Jean-Pierre Chrétien, “RTLM Propaganda,” p. 60.
270
eliminated, but the ones proposed were simple Manichean understandings of the situation,
vague enough to accommodate individual interpretations, but broad enough to apply to all
without too much inconsistency, especially to the Tutsi. There were no good Tutsi left to be
saved in extremists’ account. All were guilty and all needed to be eliminated before they
eliminated the Hutu and everything the sons of the land had achieved and given to
Rwanda. In this simplified environment, stakes were, therefore, black and white. On the
one hand, one needed to fight for the group; if not, one was a traitor. But on the other
hand, fighting for the group was the honorable cause, it was the fight for freedom. Hutu
were given the choice between being traitors or heroes, without in-between options. This
constituted a persuasive, powerful narrative to stoke up Hutu solidarity.
The story of these tactics does not begin with the genocide, however, nor with the
civil war. Discourse was rooted in a ‘faulty history’ of groups and intergroup relations
developed and slowly internalised over decades, which allowed the rhetoric to resonate
with what the population knew and felt. The tactic themselves, however, had already been
used in Rwanda. The Habyarimana regime had not been the only one to resort to Hutu
solidarity in the face of its growing lack of legitimacy. In the late 1950s and early 1960s,
Kayibanda’s young movement faced its own legitimacy problems. It had to create a power
base and a legitimacy base where none or little existed before. Under Kayibanda, agitating
anti-Tutsi sentiments and in parallel Hutu solidarity became a strategy to gain support and
legitimacy. This was conveyed by promoting antagonistic, policiticised and simple
renderings of group and group interactions. The fact that authorities recurrently and, more
often than not, successfully employed these strategies served over time to inscribe them in
the repertoire of potent action for political entrepreneurs, but also in a repertoire of
tolerated beliefs and behaviour for the population. It was from this repertoire again that
Habyarimana and the group of extremists around him chose their own strategies in the
early 1990s.
271
While strategies employed by the two regimes to ensure their political survival were
extremely similar, flagrant differences existed between the two periods, however. As
Catharine Newbury explained, these are “most notably in the scale of violence and loss of
life, and the extent of destruction of the country’s social fabric and its material
infrastructure.” 93
Levels
of
violence
attained
during
the
1994
genocide
were
unprecedented, even if the country had known its share of massacres in previous
decades. To understand this, it is important to remember that while strategies employed by
the two regimes were similar, the ends they advocated, with regard to the Tutsi, were not.
In the late 1950s and early 1960s, Tutsi were depicted as oppressors to be chased away
or attacked. In discourses, freedom was at stake. In the early 1990s, ideologues also
chose to agitate the fear of extermination as well as the threat to Hutu rights. As a result
the Hutu had to rise up and take action against this threat to themselves and to all that the
Hutu represented. Tactics were intended to get all Hutu’s support, get them all involved.
Extremists sought to achieve ultimate Hutu solidarity: a fight for survival. If all Hutu took
part in the massacre, with blood on their hands, none could defect afterwards. They would
all be solidaire as a nation of killers.
In the end, around 1994, ethno-centric entrepreneurs’ schemes reached their
target. To a large extent, Hutu solidarity had been rekindled. Without it, the genocidal
scheme of extremists would not have succeeded as it did. The MRND had been
discredited, Hutu had been split by quarrels across the North and the South. The
promotion of the Hutu cause increasingly supported by a clear, strong pro-Hutu camp,
particularly with the creation of the Hutu power in 1993, helped foster the cross-regional
Hutu solidarity needed as a platform to mobilise the Hutu population. 94 In doing this, not
only had the clique of extremists managed to reactivate antagonistic political identities, but
93
Catharine Newbury, “Ethnicity and the Politics of History in Rwanda,” p. 18.
According to Kimonyo, support for the MRDN was so weak in certain regions of the country that it is only its
association with the MDR, popular in regions where the MRND was not, that the Hutu cause achieved the
levels of popularity it did. See Jean-Paul Kimonyo, La participation populaire au Rwanda.
94
272
they had finally succeeded in displacing all other forms of collective identity. While in
everyday life, in previous decades the Rwandese had co-habited relatively amicably,
personal and community based identities subsuming ‘ethnic’ and political identities in calm
times, the 1994 genocide served to sever these ties. Hutu solidarity in the face of the Tutsi
threat had become the one salient form of identification. According to a condemned killer,
by 6 April 1994, in Rwanda, “Hutu of all sorts had suddenly become patriotic brothers
without any more political discord.” 95
95
Personal translation. Quoted in Jean Hatzfeld, Une saison de machettes, p. 21.
273
CHAPTER 10: THE BREAK-UP OF YUGOSLAVIA AND ENSUING WARS
Origins
Yugoslavia and Rwanda differ greatly. While in Rwanda the population seems more
homogenous, sharing important characteristics such as language and religion, Yugoslav
populations were more diverse. Yugoslavia brought together groups who spoke different
dialects or languages and who were divided by their Eastern Orthodox, Islamic and
Roman Catholic faiths. Furthermore, while Tutsi and Hutu live relatively intermixed on the
Rwandese territory, the dispersion of populations in Yugoslavia was more complicated. 1
Movements of populations on the territory created zones where one group tended to
predominate. When the SFRY was created, the federation was built with this distribution in
mind. Borders within the former SFRY delimited six republics (Bosnia-Herzegovina,
Croatia, Macedonia, Montenegro, Serbia and Slovenia) and two autonomous provinces
within the Republic of Serbia (Kosovo and Vojvodina). All, with the exception of BosniaHerzegovina, which is more heterogeneous, were inhabited by a strong communal
majority. Many, however, also had large minorities, often members of the majorities of
other Republics. In Croatia, Macedonia and Serbia, for example, the majority cohabited
with a substantial minority (between 20 to 30 per cent of the population). 2 Yugoslavia’s
population was composed of nations (narod), Slovenians, Croats, Bosnian Muslims
(recognised after the others), Serbs, Montenegrins and Macedonians, as well as of a
number of nationalities (narodnost), referring to large groups of non-Slavic populations,
such as the Albanians. 3
An important characteristic the Yugoslavian case shares with Rwanda, however, is
the disagreements on the origin of groups. Conflict between Yugoslav populations found
1
As indicated in previous chapters, communities in Rwanda are mixed, but the proportion of Tutsi per Hutu is
smaller in Northwestern regions.
2
Franke Wilmer, The Social Construction of Man, the State, and War: Identity, Conflict, and Violence in the
Former Yugoslavia (New York: Routledge, 2002), p. 43.
3
Michael Sells, “Crosses of Blood: Sacred Space, Religion, and Violence in Bosnia-Hercegovina,” Sociology of
Religion, vol. 64, no. 3, 2003, p. 310.
274
its source, among other things, in the different founding myths of each nations, which
insisted on the different characters, different origins, and, as a result, different rights to the
territory each could claim. Much like Rwanda, the Yugoslavian crisis thus saw its share of
antagonistic characterisations of the ‘other’, as well as quarrels for rights to the land,
based on claims of autochthony or of acquired rights whether by might or treaty. 4 In
actuality, the principal communities in Yugoslavia, the Croats, Serbs and Bosniaks, traced
their origin to the same Slavic tribes. 5 Distinction between them, their perceived cultural
diversity, resulted from their being at the ‘cultural crossroads’ of Western Roman
civilisation and Eastern Byzantine civilisation for millennia, not from distinct origins. 6
Serb and Croat histories are telling in this regard. Each population found itself on
one side of the divide between East and West, and followed very different historical
trajectories as a result. The Middle Ages constituted a golden age for Serbia. Under the
Nemanjić dynasty, the Serbs controlled an independent kingdom reaching well beyond the
borders of actual Serbia to incorporate Serbia, Macedonia and Herzegovina. 7 The
Nemanjas consolidated Serbian military and political power in the area. During this golden
age, Stefan Nemanja, the patriarch of the dynasty, abandoned Catholicism to adopt
Eastern Orthodoxy in 1196, a choice that his son Rastko, later canonised as Saint Sava,
reaffirmed and extended, making Eastern Orthodoxy the Church of all Serbs. 8 Serbian
glory was fleeting, however. Two hundred years later, in 1389, the Serbian kingdom
4
Obviously, in Rwanda, the different origin and arrival theses having been integrated to regime ideology, it was
rarely contested by the population and was therefore not the subject of quarrels between Tutsi and Hutu. It
served recurrently, however, as justification for the oppression of Tutsi under Hutu regimes, as well as for
justification of Hutu, presented as first arrived on the territory, control of Rwanda. These theses are to this day
a question for debate.
5
Michael Sells, “Crosses of Blood,” p. 311.
6
Ivo Banac, “Foreword,” in Sabrina P. Ramet, Balkan Babel: The Disintegration of Yugoslavia from the Death
of Tito to the Fall of Milosevic (Cambridge, Mass.: Westview Press, 2002), p. x.
7
Aleksandar Pavković, The Fragmentation of Yugoslavia: Nationalism and War in the Balkans (New York: St.
Martin’s Press, 2000), p. 8. Tracing much larger borders for the Serbian kingdom of the Middle Ages, Steven
Majstorovic explained that Serbian borders at the time reached “from the Adriatic to Western Bulgaria and to
most of Albania and some areas in northern Greece.” Steven Majstorovic, “Ancient Hatreds or Elite
Manipulation? Memory and Politics in the Former Yugoslavia,” World Affairs, vol. 159, no. 4, 1997, p. 174.
8
Tim Judah, “The Serbs: The Sweet and Rotten Smell of History,” Daedalus, vol. 126, no. 3, 1997, p. 25.
275
suffered a defeat at the hands of the Ottoman Empire marking the beginning of five
centuries of nearly uninterrupted Ottoman domination.
Not uncontested, the Ottoman domination of the region was sporadically shaken by
rebellions, encouraged by Austrian incursions into of the area. These led to migrations on
the part of Serbs to Austrian dominated areas across the Danube, particularly in the
Vojvodina (1690), and to the parallel migrations, encouraged by the Turks, of Albanians to
Kosovo. 9 What came to be known as the ‘Great Migration of the Serbs’ is noteworthy
because, as Aleksandar Pavković explained, “the Habsburg emperor Leopold I granted
these Serb refugees church self-government on condition that they enter imperial service
as farmer-soldiers on the Military Frontier [with the Ottoman Empire]”. 10 This granted
settlement on the military frontier–or Vojna Krajina in Serbian–would become the source of
Serb claims to rights on the region known as Krajina during the break-up of Yugoslavia. 11
Despite rebellions, despite episodic Austrian support, none of the early attempts to shake
the yoke of Ottoman domination succeeded. Serbia would only regain its autonomy–and
eventually decades later its independence– in the nineteenth century, following uprisings
in 1804 and 1815.
Croatia’s history is also one of domination, but at the hands of the Ottomans’ rivals
in the Balkans, the Austro-Hungarian Empire. Much like the Serbs, Croatians enjoyed free
rule for a time in the Middle Ages. The establishment of an independent Croatian polity is
generally associated with the recognition by the Roman Catholic Church, in the 10th
century, of King Tomislav as ruler of a region extending from Croatia to certain parts of
Bosnia and to the Dalmatian Coast on the Adriatic Sea. 12 Though parts of the Coast were
intermittently occupied by Venetians, the rest of Croatian territory finally lost its
9
Ibid., p. 28.
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 10.
11
On the Vojna Krajina, see for example Stuart J. Kaufman, Modern Hatreds: The Symbolic Politics of Ethnic
War (London: Cornell University Press, 2001), p. 168.
12
Franke Wilmer, The Social Construction of Man, the State, and War, p. 33.
10
276
independence following a defeat at the hands of the King of Hungary at Gvodz in 1097. 13
From 1102 on, Croatia fell under Hungarian authority. It passed under the Austrian
Habsburgs’ rule in the sixteenth century when they gained the Hungarian crown. In total,
for over eight centuries, Croats lived under Hungarian and then Austro-Hungarian control,
making them much more subject to European history, whether in terms of currents such as
the Reformation and Counter-reformation, Bonapartism, or the age of capitalism and
industrialisation, than their Serbian counterparts. 14
Despite such different historical trajectories, both Serbs and Croat opened the
nineteenth century under foreign domination, the Croats under the overstretched AustroHungarian Empire and the Serbs at the hands of the increasingly weakened Ottoman
Empire. Owing in part to the weakness of these empires, the nineteenth century thus
marked the birth of contemporary Serbian and Croatian nationalisms. 15 Drawing on
national historical myths, these nationalisms became the platform for liberation ideologies.
Nationalist intellectuals wrote the new histories of their nation, interpreting “the history of
their emerging nations as a mission of recovery of the freedom and independence these
nations possessed in the past.” 16 In these ‘rediscovered’ histories laid Serbian and
Croatian claims of autonomy or independence from the domination or influence of others.
The proof of historical existence, even of a previous grandeur, justified the right to stand
alone, independent, alongside powerful nations.
At the heart of Serbian nationalist imagery lies, for example, the myth of Prince
Lazar and the Battle at Kosovo Polje (Field of the Blackbirds). This myth tells of a battle
between Serbs and Turks, on St. Vitus’ day in 1389, believed to have taken place where
13
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 7.
Stephen Schwartz, “Beyond ‘Ancient Hatreds’: What Really Happened to Yugoslavia,” Policy Review, vol.
97, 1999, p. 43.
15
Franke Wilmer, The Social Construction of Man, the State, and War, p. 36.
16
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 6.
14
277
the Lab and Sitnica rivers meet, near Priština, in Kosovo. 17 According to many accounts,
the battle appears to have been a draw or to have even been won by the Serbs. 18 Yet the
Serbian myth holds that the Serbs were defeated that day and, thus, began their long
submission to the Ottomans. 19 Serbian poetry and mythology tells the story of Prince
Lazar Hrebeljanović and his men who bravely fought that day Sultan Murad’s Turkish
troops. While one of Lazar’s knights, Miloš Obilić, is said to have attacked Turkish
encampments and killed the Sultan, another Serb, Vuk Branković, betrayed Lazar. Thus,
on the day of the battle, the Sultan died, but so did Prince Lazar, and many heroic Serbian
noblemen. The tale is one of courage in battle, but it also holds a more mystical meaning.
According to legend, the night prior to the battle, Prince Lazar dreamt he was visited by an
angel (or a grey hawk or falcon according to other versions). 20 The angel offered Lazar a
choice. The Prince could choose to be victorious in the battle against the Turks, winning
an earthly kingdom. Or he could die a martyr for Christian faith and lose the battle. The
latter choice, however, also promised to award him a heavenly kingdom, his sacrifice
“sanctifying Serbia, celestial Serbia, for all time.” 21 The legend of the Battle of Kosovo
claims that Prince Lazar chose the heavenly kingdom.
The legend of Prince Lazar and Kosovo Polje is central to Serbian national
consciousness for a number of reasons. On the one hand, it speaks of the loss of Serbia’s
grandeur, Serbia’s extended territory, and most importantly Serbian independence at the
hands of foreigners. Kosovo, at the time the heart of the Serbian kingdom and home of
17
Olga Zirojević, “Kosovo in the Collective Memory,” in Nebojša Popov, ed., The Road to War in Serbia:
Trauma and Catharsis (Budapest: Central University Press, 2000), p. 189.
18
David Bruce MacDonald, Balkan Holocausts? Serbian and Croatian Victim-Centred Propaganda and the
War in Yugoslavia (Manchester: Manchester University Press, 2002), p. 69.
19
Tim Judah, “The Serbs,” p. 26. As Olga Zirojević explained, “[t]he greatest difficulty in ascertaining even the
most basic facts regarding this Serbian-Turkish clash does not lie in the scarcity of contemporary sources
(sources which are, at the same time, contradictory), but in the creation of the legend of Kosovo at a very early
date.” Olga Zirojević, “Kosovo in the Collective Memory,” pp. 189-190. Soon after the battle, a series of epic
poems were written to recount the events and more in ensuing centuries.
20
David Bruce MacDonald, Balkan Holocausts?, p. 69.
21
Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown: The Serbian National Awakening
as a ‘Revitalization Movement’,” Nationalities Paper, vol. 32, no. 4, 2004, p. 769.
278
some of the Serbs most important shrines, became where “the Serbian state perished and
its independence buried.” 22 But more than speaking of a defeat, the legend of the Battle of
Kosovo is also a source of great pride. It speaks of the Serbs special relationship with
God. When God tested him, Lazar chose martyrdom, which consecrated a covenant with
the divine. 23 The legend thus speaks of the rebirth, through martyrdom, of the Serbs as a
chosen people. The Battle of Kosovo alludes also to another form of rebirth, however. For
Serbs, parallels between the passion of the Christ and Prince Lazar’s story, a last supper
with disciples before the night of the battle, betrayal by one of them, and, finally,
martyrdom, a sacrifice for the heavenly kingdom, speak of Serbian resurrection. 24 The
underlying message of the legend is, according to Žarko Korać, professor of psychology at
the University of Belgrade, that Serbs “’are going to make a state again.’ Just as Jesus is
‘coming back,’ so is Lazar.” 25 Lazar’s story is about the resurrection of the Serbian state. It
is a legend that, according to many, allowed the Serbs to bear five centuries of Ottoman
domination and helped revive national consciousness against Ottomans in the nineteenth
century. 26 They endured with the hope of a coming kingdom.
Similarly, Croatians drew from their historical myths to ground their nationalist
revitalisation in the late eighteenth and nineteenth century. At the core of Croatian national
imaginary are their continued statehood and their role as the Antemurale Christianitatis.
The myth of a thousand years old Croat state is linked to the existence of an independent
Croatian kingdom in the first millennium. From this initial kingdom developed, in the
seventh century, two institutions, the Banus (the chief executive) and the Sabor (the
people’s assembly). 27 These two institutions survived the Croatian defeat at the hand of
the Hungarians in 1097. The Pacta Conventa signed in 1102 to institute Hungarian control
22
S. Ćirković, quoted in Olga Zirojević, “Kosovo in the Collective Memory,” p. 189.
Dusko Doder and Louise Branson, Milosevic (New York: Free Press, 1999), p. 52.
24
Michael Sells, “Crosses of Blood,” p. 312.
25
Quoted in Tim Judah, “The Serbs,” p. 26.
26
Tim Judah, “The Serbs,” p. 29.
27
David Bruce MacDonald, Balkan Holocausts?, p. 114.
23
279
over Croatian territory guaranteed the indirect rule of Croatia via a Ban, or representative
of Croatia at the Hungarian court, and thus allowed for the continued existence of the
Banus and Sabor. 28 Even when Croatia came under control of the Habsburg house,
historians claimed that the maintenance of the Pacta Conventa’s prerogatives, along with
guarantees obtained on the part of the Austro-Hungarian Empire, “gave clear proof of the
continuation of Croatia’s sovereignty.” 29
The continued statehood of Croatia is debatable on a number of levels, however.
The myth maintains that the Sabor granted Croats the right to elect kings other than the
Hungarian or Austro-Hungarian monarchs. In reality, this occurred only three times over
the eight centuries of foreign control. 30 Furthermore, while Croatia was given the right to
keep some of its institutions–not only the Banus and Sabor, but also the right to keep their
judicial body, their home guard, and to preserve their language–their autonomy remained
very much the prerogative of their conquerors who could appoint or remove the Croatian
Ban at will, who taxed 55 per cent of Croatian revenues, and who eventually seized
different parts of Croatian territory and incorporated them to the Austro-Hungarian
Empire. 31 Overall, then, the claim to an uninterrupted thousand years old Croatian
statehood is more myth than historical fact. Like many myths, however, it served a
function. The existence of thousand years old state institutions in Croatia, whose
continued existence eighteenth and nineteenth century intellectuals tried to trace, served
as evidence of the right of Croatia to independent statehood. 32 The continued existence of
Croatian institutions called for their recognition. It also comforted some in the belief of their
28
Ibid., pp. 114-115.
Ibid., p. 114.
30
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 7.
31
David Bruce MacDonald, Balkan Holocausts?, p. 116.
32
Steven Majstorovic, “Ancient Hatreds or Elite Manipulation?,” p. 175.
29
280
nation’s importance, maybe even superiority, since as MacDonald explained, “their nation
had outlasted most others in history–supposedly without being conquered.” 33
Another component of Croatian mythology was tied to the Antemurale
Christianitatis. In traditional myths, Croatia is believed to have been “the easternmost
rampart of Christian Europe, and was the sole defender of the West against the East.” 34
Much like Serbs from the Krajina, Croatians saw themselves as keepers of the Eastern
border against the Eastern realm, to which, in their minds, Serbs belonged. They traced
the establishment of a line between the West and East to Roman Emperor Theodosius’
decision in 393 A.D. to divide the Roman Empire. The Emperor drew the dividing line
across the territory occupied by the former Yugoslavia. 35 For Croatian nationalist
historians, the divide between West and East, and by extension, between themselves and
Serbs, was therefore civilisational. Croats belonged to the civilised world, to the
Mediterranean realm, to Europe, while Serbs belonged to the barbarians, on the other side
of the divide. The gulf separating the two groups was reinforced by the Serbs’ decision to
break with the Church of Rome and to adopt Eastern Orthodoxy, following the Great
Schism of 1054 A.D. in the Christian world. Finally, the Ottoman subjugation of the Serbs
accentuated differences between the Western and Eastern realm in the Balkans.
According to some myths, “[w]ithin these five years of Ottoman rule, Serbs supposedly
learnt their cruelty and despotism, while becoming further separated culturally from the
Croats.” 36
The Antemurale Christianitatis component of their mythology is important for Croat
national consciousness. It speaks of the Croatians’ role as defenders of the Catholic realm
against the dangers to the East. It therefore speaks of their sacrifice, their choice to serve
as martyrs in the defence of their faith. Weaved unto this vision is the image of Croats as a
33
David Bruce MacDonald, Balkan Holocausts?, p. 115.
Ibid., p. 116.
35
Stephen Schwartz, “Beyond ‘Ancient Hatreds’,” p. 42.
36
David Bruce MacDonald, Balkan Holocausts?, p. 117.
34
281
peace-loving and righteous people. Their role in this narrative is a defensive one. Croatian
mythology speaks of Croats as a spiritual people who, in the defence of their faith, chose
to forego aggression, though submitting themselves to the aggression of others. According
to legend, Croats never attacked beyond their borders. 37 This decision on the part of
Croats was to be the source of a Covenant with God and of their election as a divine
people. 38 Croat mythology holds that the Pope sent priests to Croatia to baptise the
population in the third century. Croats promised the Pope and God that they would never
invade foreign territory. In exchange, Croatians received the divine protection of God and
St. Peter against attackers. This pointed to Croatia’s blessed quality. Much like the Serbian
the myth of the Battle of Kosovo, the Croatian myths confirmed that they were a chosen
people.
By the end of the nineteenth century, Serbs and Croats revived their traditional
myths, developing them into contemporary nationalist platforms intended to drive political
action towards autonomy and the defence of the national territory. The Croatian state right
was revived in the eighteenth century to gain Austro-Hungarian support in reclaiming the
Dalmatian Coast from the Venetians. 39 The following century, however, it served as the
basis for claims for independence from Austro-Hungarians. Croatian mythology even
served, at the hands of nationalist Ante Starčević, to ground claims for the ‘recreation’ of
Greater Croatia, a mythical territory extending “from the Julian Alps to the river Timok on
the border of Serbia and Bulgaria.” 40 All living within this territory, including those claiming
to be Slovene or Serb, were, according to Starčević, in reality misguided.
In parallel to this nineteenth century Croat nationalist revival, Serbs experienced
their own nationalist revival. Following rebellions early in the nineteenth century, Serbs
continued to fight for autonomy and independence from the Ottomans. Revitalisation of
37
Ibid.
Ibid.
39
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 9.
40
Ibid.
38
282
Serbian mythology, particularly of legend of Prince Lazar and the Battle of Kosovo, stoked
animosity towards the Ottomans and dreams of the rebirth of the Serbian kingdom.
Serbian mythology inspired more ambitious dreams as well. A Serbian nationalist, Ilija
Garašanin, developed the first Serbian national liberation programme, the Načertanije,
which proposed to recover all lands said to have belonged to the great Nemanjić
dynasty. 41 The dream of Greater Serbia included Macedonia, Bosnia and Herzegovina, but
also parts of Croatia and Slavonia, which certainly conflicted with Croatian aspirations.
While competing nationalist platforms–at least in terms of territorial ambitions–were
developed in the nineteenth century, so was the idea of unifying Slavic populations living in
the Southern Balkans. For some, primarily Croat intellectuals, the road to unification lay in
the creation of a common Serbo-Croat language. 42 For others, it resided in the creation of
a cultural unity between South Slavs, Yugoslavs. 43 In the nineteenth century, most of
these efforts never reached beyond the small circle of intellectuals that promoted them.
South Slav populations would have to wait for the twentieth century to unite. But unity
would come from more pragmatic goals than common language or culture. Instead, it
would be freedom from existing or future imperial domination that would lead them to
combine their forces, but so would avoiding fighting among themselves, particularly for
disputed territories. 44
Contemporary History: The Twentieth Century
Ten years into the twentieth century, the final combat against foreign domination in the
Balkans began. The year 1912 marked the beginning of the Balkan wars. Sensing the
41
Tim Judah, “The Serbs,” p. 28.
Franke Wilmer, The Social Construction of Man, the State, and War, p. 36.
43
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 12.
44
It should be noted that the Principality of Serbia regained its independence following the Congress of Berlin
in 1878, which created independent states out of former Ottoman territory. During this Congress, it was
decided, however, that Bosnia-Herzegovina would be ceded to the Austro-Hungarian Empire. Serbs opposed
this decision, which represented, for them, the continued foreign domination of Serbian territory.
42
283
agony of the old Ottoman Empire, Balkan nations–Serbia, Montenegro, Greece and
Bulgaria–joined their efforts to expel the remaining Ottoman forces from their land. 45
Among the liberated territories were Kosovo-Metohija, South Serbia and Macedonia.
Serbs rejoiced at the recovery of territories they claimed as historically theirs. In this
victory, however, lay the seeds for the Second Balkan war, one that, soon after the defeat
of the Ottoman Empire, pitted Serbs against Bulgarians for control of Macedonia. Serbia
eventually defeated Bulgaria and regained Macedonia, a territory composed of large
populations of non Serb Christian Slavs, Muslims and Albanians. 46 For Serbs, this
constituted the recovery of another part of the ‘Greater Serbia’. The liberation of the rest of
the territory remained, however, to be achieved and would come with the expulsion of
Austro-Hungarians from the Balkans.
Balkan nations would not have long to wait long to see the Austro-Hungarians
leave. The following year, the First World War began, marking the beginning of the end of
Austro-Hungarian control of parts of the Balkans. It is in fact in the Balkans that the
descent towards war began, with the assassination, in Sarajevo, of Archduke Franz
Ferdinand of the Habsburgs on 28 June 1914. The murder of the Archduke at the hands of
Gavrilo Princip, a Serb, was intended to press a political agenda. Princip belonged to a
group, Young Bosnia, who agitated for the liberation of Bosnia-Herzegovina from AustroHungarian control and for its unification with Serbia. Princip’s attack followed another
attempt on the Archduke’s life by one of his comrades Nedeljko Čabrinović, who had
drawn on historical myths to explain his actions. Archduke Franz Ferdinand had chosen
St. Vitus day, the anniversary of the Battle of Kosovo, to travel to the Serbian land that his
crown controlled. 47 This affront, to as an oppressor step on Serbian land on such a
hallowed day, had infuriated Čabrinović. Following the assassination and ultimatums on
45
Tim Judah, “The Serbs,” p. 31.
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 17.
47
Tim Judah, “The Serbs,” p. 34.
46
284
the part of Austro-Hungarian authorities for an enquiry into the murder, the AustroHungarian Empire declared war on Serbia. This decision marked the beginning of the first
world-wide conflict.
While the specific events of the war itself were interesting, its consequences were
critical for the history of Yugoslavia. Early advances by a common front constituted of
Austro-Hungarians, Bulgarians and Germans in Serbian territory were defeated by Allied
forces, which marked the end of Austro-Hungarian control of the lands of South Slav
nations in the Balkans. Under Allied pressure, South Slav nations were asked to discuss
their future as a unified political entity. Discussions took place between representatives of
the Kingdom of Serbia, the polity previously under Ottoman domination, and the
Delegation of the National Council of Slovenes, Croats and Serbs, the legislative body of
the newly created State of Slovenes, Croats and Serbs, formed by the territories
previously under the Austro-Hungarian yoke. 48
The new unified ‘Kingdom of Serbs, Croats and Slovenes’, the first Yugoslavia, was
proclaimed on 1 December 1918. 49 The new kingdom did not survive long, however. This
multinational state was plagued by problems arising from the contentious relations of
nations that composed it, particularly between the Serbs and Croats. 50 Although
discussions during the creation of the state had intended to resolve power-sharing issues,
as Vesna Godina explained, “[b]ecause formal equality strategies for both nations were not
established, there was a constant drive for domination by each side in this officially
48
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 3.
According to Stuart Kaufman, “[f]aced with the choice of being treated as enemies by the Allies or joining
with Serbia in a new South Slavic state, the Croats and Slovenes chose union with the Serbs as the lesser
evil.” Stuart J. Kaufman, Modern Hatreds, p. 168.
50
According to Glenn Bowman, the problems of this first Yugoslav state stemmed from the different, and
somewhat incompatible, reasons for uniting of the different nations. “Slovenes wanted a state guaranteeing
political autonomy to Slovene-speakers formerly under the Austro-Hungarian empire; Croats wanted self
determination for Croat-speaking Catholics entailing independence from that same empire and protection from
the expansionist nationalism of their Hungarian neighbours; and Serbs wanted all Serbs [...] to enjoy union
under a single state.” Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary. Antagonism and
Defensive Solidarity in ‘Palestine’ and ‘Former Yugoslavia’,” Social Anthropology, vol. 11, no. 3, 2003, p. 328.
49
285
multinational state.” 51 For newly unified nations, Yugoslavia failed to provide a
superseding, supranational identity. 52 On the contrary, quarrels between nations ensured
nationalist platforms a continued relevance. Serbs, in particular, succeeded in becoming
the more powerful nation within the kingdom, which prompted the other nations to revive a
nationalist resistance against this ‘new oppressor’. Serbs on the contrary defended
themselves from accusations of abusive control and upheld that “the country was
decentralised, federal, and equal.” 53 In following years, the young kingdom’s existence
was “scored with assassinations, coups and the violences of nationalist movements
fighting to seize the state for their own respective peoples.” 54 The Yugoslav state,
nonetheless, survived implosion until the Second World War.
Yugoslavia experienced the Second World War as both an international conflict
and a civil war. 55 The war set South Slav nations against each other in a manner that
previous quarrels never had. Prior to the war, Croats and Serbs had never fought each
other in an open military confrontation. 56 They would fight each other, and brutally so, over
the course of World War II.
On 6 April 1941, Adolf Hitler ordered the invasion of Yugoslavia. This marked the
entry of Yugoslavia in the Second World War. Rapidly defeated, the Yugoslav kingdom not
only lost its independence but was also divided into smaller political entities by the Axis. To
the West, the Italian fascist regime took control of Slovenia and the Dalmatian coast. 57
Croatian and Serbian territory, on the other hand, were reorganised as two separate
political entities governed by collaborationist regimes. Created after the 1941 Nazi
invasion, the Independent Croatian State was governed by Ante Pavelić and his Ustaša
51
Vesna V. Godina, “The Outbreak of Nationalism on Former Yugoslav Territory: A Historical Perspective on
the Problem of Supranational Identity,” Nations and Nationalism, vol. 4, no. 3, 1998, p. 411.
52
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 23.
53
David Bruce MacDonald, Balkan Holocausts?, p. 89.
54
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 328.
55
Franke Wilmer, The Social Construction of Man, the State, and War, p. 45.
56
Tim Judah, “The Serbs,” p. 35.
57
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 328.
286
movement. The Ustaša movement had been created in 1929 with support from Italian
fascists. 58 The Ustaša quisling regime was fervently anti-Semitic, but also upheld an antiSerb ideology that blamed “all of Croatia’s misfortunes, including the failure to include all of
Bosnia-Herzegovina within its borders, on Serbs or their partners.” 59 The Croatian state’s
agreement with Nazi Germany also represented an opportunity to claim territories
considered to have historically belonged to Croatia. In exchange for their recognition of
Italian gains in Slovenia and the Dalmatian coast, the Ustaša-led Croatia gained BosniaHerzegovina, as well as Eastern Slavonia, a territory extending almost to Belgrade. To
consolidate their hold on their territory, old or newly acquired, the Ustaša regime
proceeded to purify it in ethnic terms, displacing or massacring thousands of Jews, Roma
and Serbs.
In neighbouring Serbia, the collaborationist regime instituted by the Nazis was on
shakier grounds. It faced rebellion, though rebel forces were far from united. The Serbian
puppet regime faced, on the one hand, nationalist and royalist forces led by Draža
Mihailović, who benefited from the armed support of the Četniks, a Serbian nationalist
paramilitary force developed at the beginning of the twentieth century. 60 Collaborators in
Belgrade faced, on the other hand, the Partisan movement, a communist inspired rebel
force. The predominantly Serb Partisan movement–the movement also included
approximately 20 per cent of Croats fighting the Ustaša regime–also fought Četnik
forces. 61 This tripartite conflict between collaborators, Četniks and Partisans turned the
war into an open civil conflict on Serbian territory.
The war ended with the victory of Partisan and Allied forces, but not before taking
hundreds of thousand of lives. The Ustaša purification program took its toll on Jewish,
58
Xavier Bougarel, “Yugoslav Wars: The ‘Revenge of the Countryside’: Between Sociological Reality and
Nationalist Myth,” East European Quarterly, vol. 33, no. 2, 1999, p. 171.
59
Quoting J. R. Lampe, Franke Wilmer, The Social Construction of Man, the State, and War, p. 39.
60
Xavier Bougarel, “Yugoslav Wars,” p. 172.
61
Franke Wilmer, “Identity, Culture, and Historicity: The Social Construction of Ethnicity in the Balkans,” World
Affairs, vol. 160, no. 1, 1997, p. 9.
287
Roma and Serbian populations. The better-known part of these massacres took place in a
death camp complex called Jasenovac, where an estimated 600 000 to 750 000 people
perished, most of them Serbs. 62 While Ustaša crimes are notorious, other actors of the
conflict were also responsible for large-scale killings. Četniks are reputed to have
massacred large numbers of Muslims, including civilians, and Partisans to have killed
thousands of Ustaša and Četnik fighters. 63 Furthermore, following their victory, Partisans
executed without trial a large number of people who had been attempting to flee the
country. Once caught, they were killed in forests around Bleiburg. 64 The ‘Bleiburg
Massacre’ took the lives of collaborators and perpetrators of war crimes, but also those of
innocent victims. Overall, a tenth of the Yugoslav population perished during the Second
World War, However, most of these victims dying not at the hands of foreigners, but at the
hands of fellow Yugoslavs. 65 According to Glenn Bowman, “[a]t least 1 014 000 of a prewar population of 17 186 000 were killed with eighty percent of deaths inflicted by
Yugoslavs,” making it much more of a civil war than an international one. 66
The Partisans movement’s victory meant that a communist regime would take
control of post-war Yugoslavia. Under the leadership of Marshal Josip Broz Tito, a
prominent figure of the anti-fascist resistance during the war, the new Federal People’s
Republic of Yugoslavia was proclaimed in January 1946. 67 Intent on putting the past
behind as quickly as possible to ensure the survival of the new state, Tito made
‘brotherhood and unity’–a revised version of the armed brotherhood and unity he made his
rallying cry during the war– the motto of the country. 68 Determined to make his vision of a
united Yugoslavia a reality, Tito left little space for mourning. For example, those who died
62
Ibid., p. 8.
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 328.
64
Franke Wilmer, The Social Construction of Man, the State, and War, p. 41.
65
Franke Wilmer, “Identity, Culture, and Historicity,” p. 8.
66
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 328.
67
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 47.
68
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 329.
63
288
in the Jasenovac genocide were left generally unacknowledged in historical accounts,
except as ‘victims of fascism’. 69
To limit disputes between nations and nationalities, including territorial disputes that
had in the past divided Slavs, Tito’s vision also emphasised the right of self-determination
for all nations belonging to Yugoslavia. 70 The five constitutive nations–Croats,
Macedonians, Montenegrins, Serbs and Slovenes–, were granted their own state within a
federal arrangement. 71 The Federal People’s Republic of Yugoslavia was therefore
composed of six republics, Bosnia-Herzegovina, Croatia, Macedonia, Montenegro, Serbia
and Slovenia, one republic, Serbia, also comprising two autonomous provinces, Kosovo
and Vojvodina. This arrangement kept the constituting units in check. Furthermore, while
socialist Yugoslavia adopted some Soviet inspired centralised bodies and political
arrangements, following a break with Stalin in 1948 and adjustments of federal state
controls, it adopted other institutions designed as a reaction against centralised Stalinist
communism. 72 As Gregory Hall explained, “[t]he Yugoslav model provided for devolution of
political and economic power from the State to the people through lower-level governing
and nonpolitical bodies–republican governments, communes, workers’ councils, and
others.” 73 The Second Yugoslav system also rested on a system of differentiated
appropriation and distribution of resources across republics. The system served to help
less developed regions and defuse potential resentment that might emerge as a result of
uneven economic development. 74 It is within these arrangements that the republics and
69
Tim Judah, “The Serbs,” p. 36.
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 47.
71
Following constitutional changes, a few decades later, Bosniaks were also considered a constitutive nation.
72
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals: The Case of Yugoslavia,” International Journal
of Politics, Culture and Society, vol. 11, no. 3, 1998, p. 387.
73
Gregory O. Hall, “The Politics of Autocracy: Serbia under Slobodan Milosevic,” East European Quarterly, vol.
33, no. 2, 1999, p. 234.
74
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 329.
70
289
the autonomous provinces found their autonomy. Yugoslavia was, therefore, a unique
political entity, not a mere a copy of the Soviet Union. 75
Overall, then, Tito developed a system aimed at keeping this Second Yugoslavia
together. Tito’s model glossed over the past and developed structures that guaranteed the
autonomy of constitutive nations. But at the heart of his system also lay an ideology that
socialised the population into accepting his political vision. His ideology rested on an
opposition to threats faced by the country and the creation of identities capable of
competing with national ones. For Tito, unity could stem, as in the past, from a struggle
against enemies threatening all Yugoslavs. A number of these enemies lay outside
Yugoslavia. At times, the enemy was presented as the Soviets, at other times it was the
capitalist bloc. In both cases, however, the aim was to present the interests of Yugoslavia
and its people as “threatened by the conspiracies of a labile set of enemies located outside
Yugoslavia’s territorial and ideological borders.” 76 But Tito’s ideological platform equally
relied on the identification of internal threats. Internal enemies could be found among
across the political board in any large group whose interests and beliefs questioned the
Yugoslav model. Like the external enemies identified, these included capitalists and
Stalinists and a number of others, including, according to Edvard Kardelj, an important
Yugoslav ideologue, the ultra-left and its anarchists, ‘spontanism’ and Trotskists, and the
right with petty bourgeois and Stalinism. 77 In Titoist ideology, however, another prominent
internal enemy of Yugoslavia was nationalism. With ‘brotherhood and unity’ as sanctified
precepts, nationalism represented a sweeping force capable of destroying both.
Nationalism and nationalists were “a malignity through which ‘one society aspires to
75
As a matter of fact, Yugoslavia always remained relatively independent from both the communist and
capitalist blocks and strove to develop a system of ‘self-management’.
76
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 329.
77
Vesna V. Godina, “The Outbreak of Nationalism on Former Yugoslav Territory,” p. 415.
290
dominate, exploit or despoil the others.” 78 To ensure people understood and internalised
this vision, Tito made his ideology a central component of school curricula. Much like in
Rwanda, the preferred vehicles for the ideology were civic education and history courses,
which remind pupils of the importance of ‘brotherhood and unity’, and of the dangers of its
enemies, particularly nationalism. 79
This socialisation to Tito’s precepts allowed for the development of another tool that
counteracted the centrifugal forces of nationalism: a supranational Yugoslav identity,
superseding national ones. Without denying the existence of national identities–on the
contrary, they were recognised and made the principle for the geographical organisation of
the country–but to diffuse them, Tito promoted a Yugoslav identity that was rooted in pride
for in the Yugoslav model, the country’s ability to stand strong before both the capitalist
and communist block, and in the precepts of ‘brotherhood and unity’. 80 The promulgation
of this Yugoslav identity was quite successful. Most adopted and lived this supranational
identity, even years after the disintegration of the SFRY. Identification with one’s nation
was further diluted by more limited identities, which developed from belonging to
communes or workers’ organisations, for example.
Despite all of Tito’s effort, nationalism recurrently disturbed political relations
between nations and nationalities in Yugoslavia. 81 In the late 1960s, Albanians protested
in Kosovo and Macedonia to demand that Kosovo become a republic and that they be
recognised as a nation instead of a nationality, as Muslims had been granted in May
1968. 82 The more important manifestation of nationalism during Tito’s era began in 1970.
At the time, Croatian elites began agitating for increased rights, including the right to use
78
Partly quoting Sabrina P. Ramet, Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p.
329.
79
Vesna V. Godina, “The Outbreak of Nationalism on Former Yugoslav Territory,” p. 415.
80
Tim Judah, “The Serbs,” p. 36.
81
See Franke Wilmer, The Social Construction of Man, the State, and War: Identity, Conflict, and Violence in
the Former Yugoslavia (New York: Routledge, 2002), p. 45. On this see also Jean-Pierre Derriennic, Les
guerres civiles (Paris: Presses de la Fondation nationale des sciences politiques, 2001), p. 81.
82
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 66.
291
their traditional state symbols, to use their own language (returning to the use of a
Croatian with no Serbian words in it) and to have their own armed force. Tito conceded to
these demands. In reaction to this, a number of Croatian leaders returned with new
demands, mainly centered on the recognition of Croatia’s right to secede and the reintegration of certain Bosnian and Herzegovinian regions to the Croatian territory. 83 Serbs
reacted negatively to these demands and began arming the Serb population in the Krajina.
Sensing the explosive nature of the crisis, Tito arrested the Croatian leaders who made
these demands and proceeded with massive purges of all elites within the different
republics who did not stand behind him. 84 These incidents were followed by increased
intolerance for nationalist rhetoric and demonstrations. Nationalism was banned from the
public realm and, according to Pavković, “confined nationalist polemics to the closely
watched realm of intellectual dissidence.” 85
For a number of analysts, however, this struggle to keep nationalism at bay, this
battle to make nationalism one of the Yugoslav Federation’s main enemies, had the
unanticipated effect of making it the alternative to the regime’s ideology. As Jovic
explained, “[b]y promoting a non-ethnic base for Yugoslav unity, the elite made nationalism
the main rhetorical antipode to the dominant ideology of the regime. [... And] by declaring
everyone who opposed the regime a nationalist, the regime [...] promoted such nationalism
as the main alternative.” 86 It the late 1970s and early 1980s, when Yugoslavia began
facing a multifaceted crisis shaking its economic and political foundations, the alternative,
nationalism, would come to the fore again.
83
Ibid., p. 68.
In the following years, Yugoslavia underwent constitutional changes resulting in a new constitution in 1974,
which further devolved powers to the republics, further enshrining each’s right to self-management.
85
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 85.
86
Dejan Jovic, “The Disintegration of Yugoslavia: A Critical Review of Explanatory Approaches,” European
Journal of Social Theory, vol. 4, no. 1, 2001, p. 105.
84
292
The Spiral towards Break-Up and War
Tito died on 4 May 1980, while still at the head of Yugoslavia. News of his death was
followed by widespread popular grief. For hours, people waited to see their leader in
Belgrade where he was laid to rest, confirming his undeniable popularity among the
Yugoslav population. 87 Tito had been their leader for decades. Yet he was also the symbol
of Yugoslav unity, the man who had brought the country together after the Second World
War and kept it together during all the following years. At the helm of the country, he had
repeatedly, and successfully, acted as mediator for all types of crises that had arisen,
particularly nationalist, political and economic ones. 88
The Second Yugoslavia had lost its father, and he was not to be replaced with
anyone with the same centrality in the system or the same power. At the head of the
country, many described Tito as having been “the only real decision-maker, the real
sovereign in Yugoslavia. [...] He was above the law and outside the law.” 89 Tito had further
enshrined his power in the 1974 constitution, which guaranteed him his position and power
for life. After his death he was to be replaced by the State Presidency, a political organ
composed of a representative from each republic and province, which elected each year a
president from among its members following a predetermined rotation. 90 The State
Presidency did not enjoy Tito’s overreaching powers and popularity. Tito’s death thus also
signified, for the Yugoslav system, the loss of one of the central forces keeping the country
united in times of crisis. As Jovic explained, [w]hen he died in May 1980, there was no one
to reconnect the broken bonds and to take decisions in the conflicts of interests within the
country.” 91
87
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 76.
Gregory O. Hall, “The Politics of Autocracy,” pp. 234-235.
89
Dejan Jovic, “The Disintegration of Yugoslavia,” p. 112.
90
Tim Judah, “The Serbs,” p. 36; Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 69.
91
Dejan Jovic, “The Disintegration of Yugoslavia,” p. 112.
88
293
His death occurred, however, at a time when Yugoslavia needed strong leadership
in order to face a looming, multifaceted crisis. The late 1970s and early 1980s were times
of growing economic crisis, which in turn fuelled massive popular discontent, particularly in
poorer Yugoslav republics. Simultaneously, a fringe of frustrated elites and intellectuals
threw the country into a political crisis, questioning the system’s logic and legitimacy. In
parallel to these two trends, Yugoslavia also began experiencing nationalist agitation,
particularly in Kosovo. 92 Numerous analysts and Yugoslavs sensed that these trends were
converging and could culminate in ever-greater problems for Yugoslavia. Accordingly,
Tito’s death was accompanied not only by grief, but also by widespread fear. 93
On the economic front, Yugoslavia had fared relatively well in preceding decades. It
had a relatively strong production and it benefited from a number of income sources other
Eastern European nations could not access. Because of Yugoslavia’s more opened
system, Yugoslav nationals were permitted to work in Western European nations as guest
workers. These workers, in turn, sent their revenues home, providing Yugoslavia with
welcomed foreign currencies. In addition, Yugoslavia allowed foreigners to travel within the
country. Yugoslavia thus enjoyed a booming tourist industry, particularly on its Dalmatian
Coast. Another source of revenue lay in the subsidies the country received from both the
capitalist and communist blocs in the hope of winning the Yugoslavs’ allegiances. As
Stephen Schwartz explained, “the Yugoslavs were paid by the Russians, in hard currency,
for construction and [other projects], while the U.S. subsidized the Yugoslav military on the
92
These can be listed as the background conditions of the crisis which triggered the break-up of Yugoslavia
and the ensuing wars in Slovenia, Croatia and Bosnia-Herzegovina. To these, however, a number of authors
also add international trends in the late 1980s, particularly the end of the Cold War. Susan Woodward, for
example, argued that “[i]n fact, Yugoslavia was held together not by Tito’s charisma, political dictatorship, or
repression of national sentiments, but by a complex balancing act in the international arena [between the East
and West] and a mixed economy and political system that provided governmental protections of social and
economic equality and of shared sovereignty among its nations. [...] But critical [to the Yugoslav system’s]
breakdown was change from the outside, in the foreign economic and strategic environment on which the
country’s stability had come to depend.” Susan L. Woodward, Balkan Tragedy: Chaos and Dissolution After the
Cold War (Washington, D.C.: The Brookings Institution, 1995), pp. 21-22.
93
Sabrina P. Ramet, Thinking about Yugoslavia: Scholarly Debates about the Yugoslav Breakup and the Wars
in Bosnia and Kosovo (Cambridge: Cambridge University Press, 2005), p. 148.
294
presumption that in a war between the Warsaw Pact and NATO, Yugoslavia would side
with the West.” 94
What was more, the SFRY discovered another source of funding in the 1970s.
Around 1973, Western banks and governments began granting massive amounts in loans
to Eastern European countries. 95 Yugoslavia followed wholeheartedly the trend and
borrowed large sums from the West. Many of these loans were, however, wasted in poor
investments.
Consequently, by the early 1980s, the country was saddled with an $18 billion USD foreign
debt, a sum that represented nearly half of Yugoslavia’s annual social product (the
equivalent of the GDP), making the repayment of this debt extremely difficult. 96 The
Yugoslav economy was undermined by a series of other problems as well, many of these
tied to the organisation and functioning of the system itself. According to Harold Lydall, the
combination of a socialist economy and a very decentrisalised federalism promoting each
federated unit’s self-management was a recipe for disaster:
[i]nvestment projects are duplicated, enterprises in one republic or province are
protected from competition from enterprises in other republics and provinces, there
is more trade with the outside world than with other republics or provinces within the
country, obstacles are put in the way of financial flows across republican and
provincial borders, and each republic and province tries to hold on to as much as
possible of the foreign exchange for ‘its’ exports. 97
The set-up of the Yugoslav system, particularly its oligarchic economy, also made it
vulnerable to dramatic changes in the international terms of trade. 98 Owing to all these
factors, at the turn of the decade, the Yugoslav economy was weakened and headed
towards collapse. Trends at the time were telling. The produced national income dropped,
agricultural production declined and industrial production was uneven. 99 The country
94
Stephen Schwartz, “Beyond ‘Ancient Hatreds’,” p. 44.
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 77.
96
Ibid.
97
Harold Lydall, Yugoslavia in Crisis (Oxford: Clarendon Press, 1989), p. 81.
98
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals,” p. 383.
99
Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown,” p. 766.
95
295
began suffering from rampant inflation and exploding unemployment rates. As a result of
these trends, the standards of living in Yugoslavia as well as the condition of public
services declined dramatically. 100
The economic crisis affected everybody to some degree–although admittedly some
more than others–, and reactions to it came quickly, both from the population and from
elites in the republics and provinces. The economic crisis hit the Yugoslav population hard.
At first, people found ways to cut down their expenses, by limiting their energy
consumption for example. But in light of growing economic hardships, frustration grew. 101
Street protests became a popular means of voicing discontent with the economy. One
group, semi-skilled labourers, was particularly aggrieved. Yugoslavia had undergone in
last decades a massive exodus from rural areas to cities. In 1948, 73 per cent of the
population lived in rural areas and farmed as a means of subsistence. By 1981, this
number had dropped to 27 per cent. 102 Within two generations, Yugoslavia had undergone
an accelerated urbanisation and modernisation. This process had left many maladapted or
badly integrated between the two worlds. This group was one of the hardest hit by the
economic crisis. As Devic explained, “[t]he numbers of unemployed and ‘temporarily
unemployed’ grew from 1,500,000 in 1961 to 6,400,000 in 1987, the majority of whom
where low and semi-skilled laborers (largely of rural origins, but residing in industrial
cities)”. 103 This group was the most frustrated, and it was eager to uphold its grievances in
popular protests. These riots cast a shadow on the regime, calling into question the federal
party leadership’s ability to run the country. 104
The economic crisis also had an impact on relations between republics and
provinces. The economic downturn affected the various federative units of the country in
100
Gregory O. Hall, “The Politics of Autocracy,” p. 235.
Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown,” p. 766.
102
Xavier Bougarel, “Yugoslav Wars,” p. 165.
103
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals,” p. 383.
104
Gregory O. Hall, “The Politics of Autocracy,” p. 235.
101
296
different ways, the poorest being hit the hardest, while the richest republics managing to
keep afloat. As noted, following the Second World War, Yugoslavia had developed a
subsidies system whereby richer republics and provinces contributed to the economic wellbeing of the country and of poorer republics and provinces. With growing economic
hardships facing the country, Slovenia and Croatia were therefore asked to help revive the
country’s economy. This led to mounting discontent on the part of Croatian and Slovenian
elites. They felt that that “they were being pressed to pick up the tab for most of the bad
debts accumulated by Yugoslavia.” 105 Inter-republican and provincial relations soured
further with the implementation of a plan to redress the failing economy. The plan,
developed with the help of the International Monetary Fund (IMF) built on two main
proposals, (1) to boost the country’s more competitive industrial and manufactured export
sectors through subsidies and tax incentives and (2) to recentralise the country’s market
and banking system, inefficiently split between republics and provinces. 106 The two
proposals were contentious.
On the one hand, this plan directed subsidies and tax
incentives to the industrial republics, Croatia, Slovenia, and to a certain extent Serbia,
while disregarding republics and provinces with large agricultural bases or industries
producing raw and semi-manufactured goods. On the other hand, the plan required that
republics and provinces abandon part of their autonomy, an autonomy many had fought to
guarantee through constitutional arrangements, in order to constitute a new unified market.
Beyond the loss of autonomy, for some intellectuals in richer republics, this signified
having their strong economy amalgamated to weaker ones, an option that greatly
displeased them, a point they insisted on advertising. Accordingly, economic grievances
were also an opportunity to question and criticise the Yugoslav system, its overbureaucratisation and ineffective management of the economy and federal relations.
105
106
Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown,” p. 766.
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals,” pp. 389-390.
297
Indeed, elites began targeting the federal system as the culprit for Yugoslavia’s problems.
As Sabrina Ramet explained, “the federal system itself was disparaged and demonized,
even by those republic elites who owed their power to the federal principle.” 107
Yugoslavia’ economy eventually emerged from the economic crisis, thanks in part
to the economic reforms imposed by Ante Marković’s government in the late 1980s. By
1990, the inflation rate dropped to a manageable level and productivity began increasing
again. Despite these encouraging trends, the collapse of the country was not prevented.
Poor economic conditions had had a dramatic, and lasting, impact on relations in
Yugoslavia. As Jovic argued, these economic trends “played an important role in causing
differentiation between different parts of Yugoslavia which, ultimately, resulted in growing
demands for changes both from within the political elite and from the population.” 108 This
differentiation and grievances, in turn, were seized by political entrepreneurs who turn
them to their advantage, delegitimising the federal principle that had held Yugoslavia
together for the past decades, and encouraging the reading of problems through republic
or provincial level lenses, or in many cases through nationalist lenses.
The first nationalist flare-up involved Kosovar Albanians. Albanians, the most
important non-Slav population in Yugoslavia, had at various points in their history resisted
Serbian domination and incorporation into any form of pan-Slavic political entity. It was
only by force against resisting Albanian groups that Tito had succeeded in bringing them
into the federation. 109 Following their integration into the federal system, Albanians had
been at the heart of some nationalist contention of the course of Yugoslavia’s history,
particularly in 1968, but had obtained greater autonomy for Kosovo–a province in which
they constituted the majority–in the 1974 constitution. Fuelled among others things by the
economic situation Kosovo found itself in as a result of the national economic crisis and
107
Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown,” p. 776.
Dejan Jovic, “The Disintegration of Yugoslavia,” p. 103.
109
Tim Judah, “The Serbs,” p. 37.
108
298
because of Albanians’ sense that as non-Slavs they did not benefit of the same
opportunities as others, particularly in terms of employment, Albanians took to the street
again in April 1981. As the more audacious had done in 1968, demonstrators adopted the
slogan ‘Kosovo – republic!’, demanding to be recognised as the seventh Yugoslavian
republic, and some going as far as demanding Kosovo’s unification with neighbouring
Albania. 110 The federal government reacted to this uprising by declaring a state of
emergency and sending in soldiers and the police to end the riots. Federal authorities also
purged the Kosovo state party. Additionally, 1000 Kosovars were imprisoned for having
taken part in the demonstrations.
This action on the part of the Albanian community in Kosovo had important
repercussions for the stability of the province and of the entire federation. Albanian
demonstrations were not well received by the Serbian population. Relations between the
two communities had been tense in the past, but they had worsened as a result of
changes in power distribution made by the 1974 constitution. The 1974 constitution
accorded more power to Kosovo and Vojvodina, as well as more independence from
Serbian control. For a number of Albanians, however, this did not go far enough. Though
they obtained more power, they remained within Serbia. 111 Serbs were not necessarily any
happier with the arrangement. On the one hand, as Gregory Hall explained, the devolution
of further power to Kosovo and Vojvodina “upset many Serbs, who perceived that Serbia’s
political status was being diminished relative to that of the other republics in the
federation.” 112 The central importance of Kosovo in Serbian historical myths contributed to
their discontent as well. Serbs felt frustrated at their loss of control over lands they claimed
to be historically theirs.
110
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 81; Michael Sells, “Crosses of Blood,” p. 311.
Tim Judah, “The Serbs,” p. 37.
112
Gregory O. Hall, “The Politics of Autocracy,” p. 236.
111
299
But if Serbs, elites in particular, were expressing concern about their loss of power
of Kosovo, another form of loss bothered the population, a loss of in the relative size of
their nation. The number of Kosovar Albanians had more than doubled between the 1940s
and 1980s. While it accounted for 63.7 per cent of the Kosovar population in 1948, it had
jumped to 77.4 per cent in 1981, and, by the demise of socialist Yugoslavia in the early
1990s, it represented approximately 90 per cent of the entire population in Kosovo. 113 This
demographic boom was due, in part, to high Albanian birth rates. Since the 1960s, this
rate had averaged around 27 births per 1000 inhabitants; the highest recorded in
Europe. 114 The increase in Albanians in Kosovo was also the result of a vast emigration of
Kosovar Serbs and Montenegrins. Numerous reasons are cited to explain this emigration,
including alleged discrimination towards Serbs exercised by the growing Albanian
population and the dire economic conditions in Kosovo. 115 Albanian riots in Kosovo did not
help the situation. An additional 20 000 Serbs are recorded to have left the province
between 1981 and 1987. 116
Tensions also grew between Kosovar Serbs and Albanians in the 1980s as a result
of Albanian nationalist riots and of the demographic situation. Stories of alleged
mistreatment of Serbs at the hands of Albanians began surfacing. An early example of this
is a call, in 1982, on the part of some Orthodox clergymen demanding, in light of the
situation with the Albanians, the protection of Serbs in Kosovo and of their Church’s holy
shrines. 117 With time, depictions of mistreatments worsened. Serb intellectuals blamed the
Serbian exodus from Kosovo on the fact that “Albanians were threatening Serbs into
leaving, and that the police and judiciary were not protecting Serbs against Albanian
violence” or stopping the desecration by Albanians of Orthodox churches and
113
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 81; Gregory O. Hall, “The Politics of Autocracy,”
p. 236.
114
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 81.
115
Gregory O. Hall, “The Politics of Autocracy,” p. 236.
116
Sabrina P. Ramet, Thinking about Yugoslavia, p. 19.
117
Ibid., p. 141.
300
monasteries. 118 Serbian media began reporting cases of brutal rape or beatings of Serbs
by Albanians. Most of these were exaggerations. In actual fact, instances of rape in
Kosovo remained below the rates in other regions in Yugoslavia. 119 Nonetheless, the
Serbian population was struck by stories of brutal rapes. One case in particular mobilised
media attention. A Serbian farmer named Djordje Martinović was admitted to a hospital in
Priština after having been allegedly raped with a bottle by a group of Albanians. Despite an
official investigation concluding that Martinović’s wounds were self-inflicted, the case
continued to spark Serb interest. A book published on the case sold out, despite the
50 000 copies printed, and 200 intellectuals from Belgrade signed a petition in which they
argued that: “[t]he case of Djordje Martinović has become that of the whole Serb nation in
Kosovo. Even among crimes it would be hard to find a crime like this.” 120
The path had been set for rhetorical inflation: the Kosovar Serbs’ plight was made
to sound ever more worse. The better-known example of this trend is the Serbian
Academy of Arts and Sciences’ (SANU, Srpska akademija nauka i umetnosti) draft
memorandum, leaked to the public in September 1986. 121 The Memorandum, produced by
intellectual authorities, left no doubts as to how the Serb emigration from Kosovo should
be interpreted. Serbs in Kosovo, noted the Memorandum, were submitted to “a physical,
moral, and psychological reign of terror,” which for all intents and purposes amounted to a
“physical, political, legal, and cultural genocide of the Serbian people in Kosovo.” 122 The
Memorandum also discussed the situation of Serbs in Croatia, indicating that their
situation had never been so grave, except under the Ustaša regime. The Memorandum
thus called for the Serbs to react to the situation: “[t]he Serbian people cannot stand idly by
118
Anthony Oberschall, “The Manipulation of Ethnicity: From Ethnic Cooperation to Violence and War in
Yugoslavia,” Ethnic and Racial Studies, vol. 23, no. 6, 2000, p. 990.
119
Ibid.
120
Quoted in Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 331.
121
Matthew Levinger and Paula Franklin Lytle, “Myth and Mobilisation: The Triadic Structure of Nationalist
Rhetoric,” Nations and Nationalism, vol. 7, no. 2, 2001, p. 182.
122
Quoted in Tim Judah, “The Serbs,” p. 38.
301
and wait for the future in such a state of uncertainty.” 123 The Memorandum set the tone for
a type of extreme, accusatory nationalist rhetoric that became a common feature in the
following years.
Hence, in the midst of an unfolding economic crisis and nationalist agitation on the
part of Serbian intellectuals, media and Orthodox Church in response to Albanian
expressions of nationalism in Kosovo, the 1980s were times of fear and resentment in
Yugoslavia, particularly in Serbia. The situation certainly did not necessarily need to lead
to open war at that point. To reach dangerous levels the situation required the intervention
of interested elites and political entrepreneurs, those willing to ‘orchestrate social
discontent’ for their own interest. 124 It would require, as Franke Wilmer argued, “a
conscious (and some might say irresponsible) choice [on the part of political leaders] to
rally support by appealing to grievances which had long been a subject of political
discourse, and which were constructed in terms of identity within both political and
historical narratives.” 125 The conditions in the 1980s created sufficient grievances and
frustrations to become a field on which political entrepreneurs could play. And Yugoslavia
had a number of these. Each republic had its share of ‘small power elites’ that had made a
name for themselves under Socialist Yugoslavia, and who, in the late eighties with “their
authoritarian spirit and inability to compromise [...] faced a general communist social crisis
and the danger of overthrow.” 126 Growing dissatisfaction with the traditional Yugoslav
system also opened space for new politicians willing to propose alternatives, including
highly nationalist ones.
A bit of both, having risen to a position of leadership through the weakened
socialist system, and as the first high-ranking politician to bring nationalism to the public
realm, Slobodan Milošević, more than any other Yugoslav politician, epitomised this new
123
Ibid., p. 39.
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals,” p. 391.
125
Emphasis in original. Franke Wilmer, The Social Construction of Man, the State, and War, p. 29.
126
Zlatko Isaković, Identity and Security in Former Yugoslavia (Burlington, Vt.: Ashgate, 2000), p. 231.
124
302
class of instrumental elites. The ‘slickest con man in the Balkans,’ according to former U.S.
Ambassador Warren Zimmerman, Milošević had great ambitions. 127 Rising rapidly through
the ranks of political elites within Serbia, his ultimate aim was to conquer Yugoslavia, to
rise to the position his hero, Marshal Tito, had occupied at the head of the entire country,
to occupy the “empty space of power vacated after [Tito’s] death”. 128 To achieve his rise to
the top, Milošević successfully manipulated different political rhetorics, putting forward
discourses rooted in pan-Serbian nationalist sentiments to appeal to the Serb population–
ironically destroying the nationalism-free space Tito had striven to develop–and promoting
an ‘anti-bureaucratic revolution’ to appeal to Serbian intellectuals. 129 His defence of all
things Serb, including sacred historical claims to Kosovo, won him the support of the
Orthodox Church. Milošević also quickly realised the importance of the media for his
success. As a result, he spent a great amount of energy ensuring his control of some of
the primary media in Serbia, whether audio-visual or print media. 130 By the late 1980s,
having secured the support of Serbian intellectuals, Serbian media and the Orthodox
Church, Milošević had built a powerful political machine. The man’s ambitions and the
allies he recruited in his campaign for power combined to make him one of the more
important actors of the Yugoslav crisis. 131
Prior to his ascension to the top of Serbia’s political elite, Milošević had served as
an executive in the Serbian banking sector and in public sector enterprises. At that time,
127
Stuart J. Kaufman, Modern Hatreds: The Symbolic Politics of Ethnic War (New York: Cornell University
Press, 2001), p. 5.
128
Dejan Jovic, “The Disintegration of Yugoslavia,” p. 112. According to Doder and Branson, increasingly,
Milošević modeled himself on Tito, whom he admired greatly. He adopted his manner of dress, of speech,
even exchanged cigarillos he smoked for the Havana-style cigars Tito preferred. Dusko Doder and Louise
Branson, Milosevic, p. 69. On Milošević’s quest for power, see also Stuart J. Kaufman, Modern Hatreds, p.
181.
129
Gregory O. Hall, “The Politics of Autocracy,” p. 237; Sabrina P. Ramet, Thinking about Yugoslavia, p. 153.
130
For more on Milošević’s campaign to control Serbian media, see Renaud de La Brosse, “Les médias
machines de guerre en ex-Yougoslavie,” Reporters sans Frontières, Les médias de la haine (Paris: La
Découverte, 1995).
131
A number of analysts focus on the actions of Milošević as the catalyst of the break-up of Yugoslavia and the
wars that followed in Slovenia, Croatia and Bosnia-Herzegovina. But the crisis in Yugoslavia is far too complex
to have resulted from the actions of one man. On this issue, see Gregory O. Hall, “The Politics of Autocracy,”
pp. 240-241 and Dejan Jovic, “The Disintegration of Yugoslavia,” p. 112.
303
he was known as a defender of the Yugoslav system, even condemning nationalist
expressions on occasion. In 1986, with the support of his mentor, Ivan Stambolić, then
President of Serbia, Milošević was named at the post of Chairman of the Serbian League
of Communists (the state party in Serbia). In the midst of the continuing crisis in Kosovo, at
the demand of both radical Kosovar Serbs and, in the hopes of defusing tension, of
Stambolić, Milošević met with angry Kosovar Serbs in April 1987. Milošević’s visit to
Kosovo, which had begun as a regular town hall meeting, developed into what would be
the turning point of his career, transforming him from a political apparatchik into a populist
leader. When he arrived at the meeting, Milošević found thousands of Kosovar Serbs
awaiting him. Kosovo police force, predominantly composed of Albanians, began violently
turning protestors away. Seeing this, Milošević sided with the Kosovar Serbs. In a speech
addressed to them, he condemned police actions: “[n]o one has the right to beat the
people!,” he exclaimed, and extolled them to stay in Kosovo, to endure the injustice,
humiliation and obstacles they faced there, because this was their country: “these are your
houses, your fields and gardens, your memories.” 132 He then spent the night at the
meeting hall listening to the grievances of Kosovar Serbs.
That day, Milošević broke the Titoist taboo of never raising nationalism, including
national pride, in the public realm. 133 And while his actions troubled the party
establishment, stunning among others his ally Stambolić, many within the Serbian media
were ecstatic. Milošević’s speech was aired in a loop. Journalists praised him in articles
and reports. According to one Serbian journalist, at the time, “[h]e instantly became the
leader of all Serbs.” 134 In light of the response he received, Milošević, for his part, must
132
Agneza Bozić-Roberson, “Words Before the War: Milošević’s Use of Mass Media and Rhetoric to Provoke
Ethnopolitical Conflict in Former Yugoslavia,” East European Quarterly, vol. 38, no. 4, 2005, p. 397; Tim Judah,
“The Serbs,” p. 40.
133
According to Aleksandar Pavković, “[n]o communist politician in Yugoslavia, before or after World War II,
openly appealed to the national traditions and national pride of the Serb and Montenegrin nations.” Aleksandar
Pavković, The Fragmentation of Yugoslavia, p. 104.
134
Quoted in Dusko Doder and Louise Branson, Milosevic, p. 44.
304
have sensed the political opportunity that lay in playing on Serbian pride and fears. 135
Hypernationalism was his platform for that point forward, a platform for gaining power.
Milošević began developing his strategy. He worked on building alliances within the
party, to ensure that when the time came he could successfully remove his mentor,
Serbian president Ivan Stambolić. His opportunity came not long after, when a deranged
Albanian soldier shot fellow soldiers, killing four of them, including a Serb. Milošević
secretly promoted a media campaign that stoked mass anger against Albanians through
references to a greater Albanian plot against Serbs. The reaction of moderate Serbian
politicians, trained in the Titoist tradition, was to condemn such nationalist agitation.
Milošević’s supporters within the media, however, discredited the moderates. Moderate
politicians, including Stambolić, were soon dismissed from office. 136 Those who had given
Milošević their help and support to engineer this coup were rewarded with high posts in the
party or in the media. 137
In the following months and years, Milošević continued to use populist strategies to
consolidate his power in Serbia. His strategy rested on well planned media campaigns and
the organisation of rallies, dubbed ‘happenings of the people’, during which he, or his
allies, would promote inflammatory ideas. 138 At the heart of this effort lay two main themes,
the “delegitimisation of Tito’s rule and constitutional foundations” and “united, single
Serbia”. 139 This last principle, a unified Serbia, entailed reconquering the autonomous
provinces, particularly Kosovo, the place of Serbian vanished glories. 140 For months in
1988 and in early 1989, large rallies were organised in Kosovar cities, during which
135
Agneza Bozić-Roberson, “Words Before the War,” p. 398.
For a detailed account of these events, see Dusko Doder and Louise Branson, Milosevic, pp. 45-50.
137
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 105.
138
According to Mueller, a large segment of the participants in these rallies were professional protesters or
were crowds which Milošević’s party paid “with free food, transportation, and liquor.” John Mueller, “The
Banality of ‘Ethnic War’,” International Security, vol. 25, no. 1, 2000, p. 46.
139
Ibid.
140
Dusko Doder and Louise Branson, Milosevic, p. 51; Anthony Oberschall, “The Manipulation of Ethnicity,” p.
992.
136
305
protestors demanded the dismissal of Albanian politicians from Kosovo’s institutions. In
February 1989, with popular backing, Milošević dismissed Albanian politicians and
arrested one of the more popular Albanian youth leaders, Azem Vlassi. Milošević replaced
them with a new Albanian leadership he controlled. 141 After this change of leadership in
Kosovo, he proceeded with his unification plan. In April 1989, Milošević’s government
adopted modifications to the Serbian constitution that revoked Kosovo and Vojvodina’s
autonomous status, reintegrating them to Serbia. 142 A few months later, additional
arrangements ensured that Serbia control the seats and votes of the provinces in federal
institutions. 143 These decisions affected the entire country. Kosovo and Vojvodina had
previously sent their national representatives to federal and republican institutions. Under
these new constitutional arrangements, Anthony Oberschall explained, “[t]he nationality
balance in Yugoslav politics was thus disturbed. Serbia gained control of over half the
votes in all federal bodies and institutions.” 144 With this new power, Milošević moved
towards his new goal, “the (re)establishment of a centralized Yugoslavia”. 145
This new dominant Serbia stoked up nationalist sentiments. While it gave rise to a
pan-Serbian national movement that mobilised various Serb organisations, including those
outside Serbia, it also sparked a reactive stance against Milošević and Serbian nationalism
on the part of other republics. The situation in Yugoslavia deteriorated rapidly. According
to Ramet, Milošević’s actions and policies “destroyed what remained of any consensus in
the [Yugoslav] system [...] In its place were four emerging national environments, which
claimed the primary loyalty of their citizens.” 146 These were Serbia (which in its nationalist
141
Dusko Doder and Louise Branson, Milosevic, p. 59.
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 105.
143
Olivera Milosavljević, “Yugoslavia as a Mistake,” in in Nebojša Popov, ed., The Road to War in Serbia:
Trauma and Catharsis (Budapest: Central University Press, 2000), p. 73.
144
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 992.
145
Gregory O. Hall, “The Politics of Autocracy,” p. 239.
146
Sabrina P. Ramet, Balkan Babel, p. 21.
142
306
movement was also joined by Montenegrins), Slovenia, Croatia, and Macedonia. 147
Macedonia, at the time, continued to defend, to a certain extent, ‘Yugoslavism’, but the
other three adopted positions antithetical to the status quo. Under Milošević, as just seen,
Serbia sought to consolidate its power within Yugoslavia. Slovenia and Croatia, however,
increasingly sought to distance themselves from the decaying federation.
Slovenia had begun its nationalist reawakening earlier in the decade at the hands
of intellectuals defending the republic’s culturally and linguistically distinct character. They
rooted their claims in efforts to standardise the Slovene language by their forefathers in the
nineteenth century. Pavković explained that, at the time, “current Slovene men of letters–
intellectuals–often considered themselves as successors to the creators of Slovene
nationhood, whose main task was to guard, vigilantly, its cultural and linguistic
distinctness.” 148 Nationalist agitation in Slovenia intensified and reached beyond circles of
intellectuals following a series of confrontational events. The first of these was the leak of
the Serbian Academy of Arts and Sciences’ Memorandum in 1986. The Memorandum
described the hardships Serbs endured in the Yugoslav federation, but also discussed a
“return to an ‘integrative’ Yugoslavia.” 149 For Slovene intellectuals, this project threatened
Slovenia’s autonomy, and, as a result, its ability to defend its culture and language. In part
in reaction to the SANU Memorandum, in 1987 Slovene intellectuals published
contributions to a Ljubljana journal, Nova Revija, a Slovenian National Programme,
presenting a series of argument in support of Slovenian independence. 150
Another pivotal moment came when the Yugoslav federal army (JNA)–
predominantly composed of Serbs–arrested four Slovene journalists who had been
criticising the military body and discussed a “military plan to mop up liberalism in
147
Bosnia-Herzegovina was internally divided among the different communities which composed it and, as a
result, did not defend a nationalist platform on the federal scene.
148
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 91.
149
Ibid.
150
Olivera Milosavljević, “Yugoslavia as a Mistake,” p. 73.
307
Slovenia”. 151 The JNA chose to prosecute the journalists in Serbo-Croatian, which
infuriated the Slovene population, who took to the streets in protest. The four were found
guilty. The events consolidated the nationalist movement in Slovenia, which interpreted the
JNA actions as “the breach of Slovene state sovereignty by a ‘foreign’ force while the use
of Serbo-Croatian symbolised the ‘foreign’ oppression of Slovene culture and a denial of
Slovene sovereignty.” 152 A final element comforting Slovene nationalism were Serbian
policies towards Albanians in Kosovo, particularly the brutal repression of strikes by the
Albanian population in Kosovo in reaction to the purges the Milošević government had
exacted on the Kosovar leadership. Slovenes elites sided with the Albanians in defence of
nations and nationalities’ rights in Yugoslavia, rights that Serbian authorities seemed very
willing to violate.
The Slovene nationalist revival of the early 1980s provided the leaders of the
Slovene Communist party with an opportunity to re-invent themselves as nationalists, a
tactic that aimed at keeping them in power. Party leader Milan Kučan repeatedly took a
strong stance in favour of Slovene culture and sovereignty at demonstrations. Kučan and
the old state party probably sensed the importance of this political move. In the months
following the JNA case against the four Slovene journalists, a flurry of new political parties
and movements were created, many of them with a nationalist platform. These new parties
and movement promised to erode support for the Communist Party. Kučan’s decision to
adopt a nationalist platform was therefore a strategic one, but one he adopted
wholeheartedly. In December 1988, he declared that Slovenia should be granted the right
to secede. Less than a year later, in September 1989, the Slovenian Assembly, following
151
152
Sabrina P. Ramet, Balkan Babel, p. 24.
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 110.
308
modifications to the republic’s constitution, unilaterally granted Slovenia the right to break
from the Yugoslavia federation. 153
The revival of nationalism in Croatia did not follow the path of its neighbours. Unlike
in Serbia and in Slovenia, the leaders of the Communist party remained stern defenders of
Tito’s approach. Early in the 1980s, they continued to maintain anti-nationalist policies,
fighting emergent nationalist movements. The rapidly changing Yugoslav environment
demanded they adapt, however. In light of the unfolding tension between Serbian and
Slovenian authorities, the Croatian traditional leaders were being forced to choose a side.
Their affinities lay with the Slovenes. Much like for the Slovene leadership, for Croatian
Communist leaders, “Milošević’s programme of recentralisation of Yugoslavia and his panSerb mobilisation [was perceived] as a direct threat to their hold on power.” 154 The core of
the nationalist revival effort came, however, from the nationalist dissidents whom
Communists had repressed. Drawing on the myth of uninterrupted Croatian statehood,
they upheld the right of Croatian sovereignty, whether through independence or within a
confederal arrangement.
The majority of Croatian nationalist favoured the confederal option, at least, as a
first step. Many judged that the international climate was not ripe for a complete separation
from Yugoslavia, that is, that the international community would not condone secession.
According to Pavković, however, another factor behind the popularity of the confederal
option was the large minority of Serbs living in Croatia, particularly in the Krajina and in
Eastern Slavonia. 155 Serbs in Croatia had been granted a special status in Croatia as a
result of the sufferings they had endured under the Ustaša regime. As a result, the
constitution recognised them as a constituent nation of Croatia–Croats being the sole
153
Sabrina P. Ramet, Balkan Babel, p. 24.
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 110. A former Croatian party secretary (purged
by Tito in 1971), expressed, for example, the view that Croatia “is gravely threatened by Milosevic, who is
trying to bring about a totalitarian revolution and achieve Greater Serbian hegemony. This threatens not only
Croatia but the other republics as well.” Quoted in Sabrina P. Ramet, Balkan Babel, p. 28.
155
Aleksandar Pavković, The Fragmentation of Yugoslavia, pp. 92-92.
154
309
other constitutive nation–and they were granted privileged access to positions in the
government. These Serbs risked opposing an outright independence, or worse, as a
nation, uphold their own right to self-determination, demanding the secession of the
Krajina region. A Croatian ideologue proposed his solution to the problem posed by the
Krajina Serbs. Franjo Tudjman, an increasingly popular nationalist, denied that the severity
of the massacres committed against Serbs by the Ustaša regime during the Second World
War, the source of Serbs special status. In Bespuća (Wastelands), a book published in
1989, Tudjman significantly revised estimates on the number of people killed in
Jasenovac, the Ustaša death camps complex. 156 His account infuriated Serbs across
Yugoslavia and started a war of words between Serbs and Croats, each exchanging
counts of who had lost the most during the Second World War. Alleged mass graves were
even dug up, the bodies exposed in the media, to support claims.
Around the same time, Croat catholic nationalism also surfaced around the
canonisation of Cardinal Stepinac. Stepinac had opposed communist infringement on the
Catholic Church’s affairs during the Tito era and as a result had become a religious hero
for Catholics, mainly Croats. 157 Stepinac was also a suspected Ustaša supporter, however.
He refused to defend those dying in the death camps or to condemn the actions of the
Ustaša regime. Efforts to promote his canonisation antagonised Serbs and the Serb
clergy.
The media coverage helped Tudjman become a public figure in Croatia and
propelled his political career. In February 1989, he founded the first opposition in the
republic, the Croatian Democratic Union (HDZ). At the head of the HDZ, Tudjman pursued
a nationalist campaign aimed at “the development of a Croat nationalist consciousness,
156
157
Sabrina P. Ramet, Thinking about Yugoslavia, p. 18.
Michael Sells, “Crosses of Blood,” p. 316.
310
unifying all its past diverse strands, and for a confederal reorganisation of Yugoslavia.” 158
But Tudjman also continued his campaign to rehabilitate the Independent Croatian State
and to condemn Četnik crimes during the Second World War, he claimed were committed
by Serbs to achieve Greater Serbia. 159 Although not alone, Tudjman certainly contributed
to the development of a Croatian nationalism that exhibited a “marked hostility toward
everything Serbian,” the reinvented Eastern barbarian standing in glaring opposition of
Western civilised Croats. 160
At the close of the 1980s, Yugoslavia, its political realm and society, had
undergone a thorough radicalisation, in the form of an ethnicisation. Two culprits in this
were the various nations’ religious authorities and intellectuals, which began agitating the
spectre of nationalism in the early 1980s. The trend accelerated to the point where,
according to Devic, Yugoslav academics and writers had in the majority become part of an
‘ethnicised intellectual community’, advocating ‘ethno-national rights’. 161 Increasingly,
however, as the decade wore on, these nationalist platforms were hijacked by politicians,
old or new, who sought a key for popular success. With the help of co-opted media,
nationalist entrepreneurs increasingly took control of nationalist platforms and of the
political scene. They would continue to do so in the early 1990s, with elections to be held
in all states, feeding the war of words between nations and preparing the grounds for open
war.
Campaigning for Nationalism, Winning War
With so many politicians having chosen to reinvent or affirm themselves as nationalists in
the defence of their nation’s rights, the tone and themes seemed set for political struggles,
and even dissolution, of the Yugoslav federation at the beginning of the 1990s. The coup
158
Aleksandar Pavković, The Fragmentation of Yugoslavia, pp. 112-113.
Stuart J. Kaufman, Modern Hatreds, p. 186; David Bruce MacDonald, Balkan Holocausts?, pp. 132-159.
160
Sabrina P. Ramet, Thinking about Yugoslavia, p. 22.
161
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals,” p. 402.
159
311
d’envoi of a new politicised nationalism stage in Yugoslavia soon came, in January 1990,
with the dissolution of the League of Communists of Yugoslavia (LCY, the federal state
party) 14th Congress. The Congress, set to bring together the leadership of the republics to
discuss the state of the country, unravelled over the issue of Serbia’s unilateral abolition of
the autonomous status of Kosovo and Vojvodina. Enraged by the Serbian leadership’s
unwillingness to bend on the issue, Bosnia, Croatian, Slovenian and Macedonian
delegations left the proceedings. For a number of others, this marked the “breakdown of
the Yugoslav Communist option”, at least as a moderator of disputes, and was, thus, “the
beginning of the end of Socialist Yugoslavia.” 162
Much like in the Rwandese case, a large part of the intensification of ethno-political
struggles was the result of republican-level elections throughout Yugoslavia. A good part
of the spring of 1990 was devoted to pre-campaigning on the part of the numerous newly
formed or recast political parties. But as many had ridden the nationalist wave to rise to the
forefront of the political scene, nationalist rhetoric continued to dominate the campaign. It
certainly became an unavoidable theme, and was even picked up by newcomers. As
Oberschall explained, “[e]very town and city experienced the founding of political parties,
often at a huge rally in a public building or a sports stadium, during which speaker after
speaker gave vent to exaggerated nationalist rhetoric and hostile pronouncements and
attacks against other nationalities.” 163
Across the country, the 1990 elections brought nationalist parties to power. In April,
Slovenes elected, with 55 percent of the vote, the United Slovenian Opposition (DEMOS),
while Croats elected Tudjman’s HDZ with 44 percent of the vote in elections for the
Croatian Diet and 41.5 percent of the vote for the Socio-Political Chamber of the Diet. In
November, Macedonia, which had begun to live its nationalist awakening, held its elections
162
163
Olivera Milosavljević, “Yugoslavia as a Mistake,” p. 74; Gregory O. Hall, “The Politics of Autocracy,” p. 240.
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 994.
312
in which the Internal Macedonian Revolutionary Organization-Democratic Party for the
Macedonian National Unity (VMRO-DPMNE) was elected, while elections in BosniaHerzegovina ended in a good performance on the part of the Bosnian Muslim party, the
Party for Democratic Action (SDA), (86 seats and 37,8 percent of the vote), the Serbian
Democratic Party (SDS) (72 seats and 26,5 percent of the vote) and the HDZ in BosniaHerzegovina (44 seats and 14,7 percent of the vote). Finally, in December 1990, the
Socialist Party of Serbia (SPS), Milošević’s party, won the elections with 45.8 percent of
the vote. 164
Nationalist parties succeeded during these elections because their candidates
appealed to the electorate not to ‘split the ethnic vote’ and to vote for their ethnic
representatives. 165 This had to be done, they argued, because other communities were
going to bloc vote, gain power, and therefore pose a threat to their divided community.
Nationalist thus appealed to their constituency, but the latter did not overwhelming support
their communal representatives. As a matter of fact, as John Mueller indicated, the
nationalists’ victory in all republics “was not all that deep”, and benefited from the ‘winnertake-all’ electoral system that exaggerated the number of seats obtained in comparison to
the percentage of votes received. 166
Once elected, however, the nationalist leaders moved to cement their hold on
power. They worked to eliminate moderate politicians from their ranks. In a number of
republics, nationalists once in power removed their opponents, particularly moderate ones,
from important party and state posts. 167 Expunging opponents sometimes took on more
violent forms. Some were intimidated, others reported killed. 168 Many also strove to
164
All preceding electoral results drawn from Olivera Milosavljević, “Yugoslavia as a Mistake,” p. 74 and from
Aleksandar Pavković, The Fragmentation of Yugoslavia.
165
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 995.
166
John Mueller, “The Banality of ‘Ethnic War’,” p. 45; Anthony Oberschall, “The Manipulation of Ethnicity,” p.
995.
167
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 996.
168
Ibid.
313
develop a coercive capability. Slovenian and Croatian leaders, for example, increasingly
moving towards a secessionist position and anticipating an altercation with the Yugoslav
federal army–the 1974 constitution stipulated that it was the JNA’s responsibility to
challenge any breach to the country’s territorial integrity–began making military
preparations. 169 Serbs also participated in this militarisation process, both within Serbia
and across its borders. According to Judah, “[t]his was done by the clandestine arming of
the Serbs of Croatia and Bosnia and the adaptation of Tito-era local defense units.” 170 By
mid-March 1991, for example, the Serb dominated JNA had provided Serb militias in
Bosnia-Herzegovina with approximately 52 000 firearms. 171
Events developed quickly thereafter. At the federal level, efforts were still being
deployed to keep all parties relatively content. Slovenia and Croatia, in particular, in a bid
to secure their sovereignty without destroying the Yugoslav federation, developed a joint
proposal for the creation of a new Yugoslav confederation. 172 The project was discussed,
but inconclusively in late 1990 and early 1991. While Macedonia and Bosnia-Herzegovina
finally adopted the confederal cause, Serbia and Montenegro remained firmly opposed to
it. But while solutions were being discussed at the federal level, centrifugal forces within
the republics were undermining the stability of the country. Following their election earlier
in 1990, Slovenian and Croatian nationalist leaders had proceeded with modifications of
their republic’s respective constitutions. These amendments affirmed the two republics’
sovereignty and gave them the prerogative to “counter moves by Yugoslav federal
organs.” 173 Serbs living in Croatia did not take well to these constitutional changes. Serbpopulated regions in Croatia were swept by protest and violence. In January 1991, four
169
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 125. Pavković, for example, indicated that “the
Slovenian government planned a military confrontation with the Yugoslav military army immediately following
its proclamation of independence.” Ibid., p. 136.
170
Tim Judah, “The Serbs,” p. 42.
171
Sabrina P. Ramet, Thinking about Yugoslavia, p. 23.
172
Sabrina P. Ramet, Balkan Babel, p. 48.
173
Aleksandar Pavković, The Fragmentation of Yugoslavia, pp. 125-126.
314
predominantly Serb regions in Croatia declared their creation of a Serb Autonomous
Region of Krajina. 174 The Croatian leadership reacted by sending Croatian police reserves.
Faced with the escalation of the situation, the Yugoslav Presidency and the Yugoslav
federal army demanded that Croatia call back and dismantle its police forces, which had
been illegally assembled outside of Yugoslav regular forces. Pressured, Croatia
demobilised part of its police reserves, but kept some posted in Krajina. The Yugoslav
state presidency reacted by sending in the JNA. 175
With no changes on the federal scene, in June 1991, Slovenian and Croatian
leaders implemented their plan of jointly declaring their independence. As Slovene
politicians and tacticians had anticipated, JNA troops stationed in Croatia crossed into
Slovene territory. The conflict between Slovenes and Serbs ended relatively quickly.
Within a month, a truce negotiated under supervision of the European Community (EC)
had been enforced. In Croatia, however, the situation was taking a turn for the worse. A
series of altercations took place between Croatian police forces and JNA troops stationed
in the Krajina. Tensions mounted and the crisis eventually escalated into an all out war in
Croatia in September 1991. Over the next few months, the conflict took the lives of more
than 10 000 people and displaced 700 000 others. 176 By November 1991, however, the
war in Croatia had reached a stalemate. Neither the Croatian forces, composed mostly of
volunteers and conscripts, nor the JNA, whose drafting efforts had not been met with much
success, had the capacity to defeat the other outside of the zones populated by their
group. 177 The belligerents eventually agreed to a United Nations sponsored peace plan
and to the deployment of a UN force. But, as Ramet argued, “[t]he truce set the stage for
174
Ibid., p. 129.
Ibid.
176
Samantha Power, “A Problem from Hell”: America and the Age of Genocide (New York: Harper Collins,
2002), p. 247.
177
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 146.
175
315
the expansion of war in Bosnia.” 178 A few months later, following a similar pattern, it was
Bosnia-Herzegovina’s turn to break into open conflict. On 29 February and 1 March 1992,
Bosnia-Herzegovina held a referendum on its proposed independence. Following a
boycott on the part of the Serbian population in Bosnia-Herzegovina, the referendum
passed in favour in independence. On 6 April, Bosnia Serbs opened hostilities. Initially
working from lists, “they began rounding up non-Serbs, savagely beating them, and often
executing them.” 179 They scared non-Serb populations into fleeing Serb populated area,
‘ethnically cleansing’ them. Each community, in turn, would proceed with its own ethnic
cleansing of ‘its zones’. The war in Bosnia-Herzegovina would last until 1995.
Due to their complexity, the break-up of Yugoslavia and the wars that followed in
Slovenia, Croatia and Bosnia-Herzegovina cannot be reduced to a simple explanation.
The debates that surround the events confirm this. 180 In these debates, each side tries to
identify specific factors that had the key role in the events that took place in the 1980s and
1990s. 181 Certain authors lean towards structural explanations, pointing to the role played
by the death of Tito, or the role of the economic crisis beginning in the late 1970s, or to
role of changes in the international system in the late 1980s, precipitating a crisis for a
system that had for decades benefited from playing the Eastern and Western blocs.
Others concentrate on agency arguments, focusing on the personality of certain leaders,
or on the actions of instrumental elites in the 1980s striving to secure power or their
position.
Neither explanation makes sense, however, without the other in the context of
Yugoslavia. Structural conditions were a determining factor in the conflict. But they were in
as much as they worked to provide a space for contestation in which grievances,
178
Sabrina P. Ramet, Balkan Babel, p. 245.
Samantha Power, “A Problem from Hell”, p. 249.
180
For more on these debates, see Sabrina P. Ramet, Thinking about Yugoslavia, in which Sabrina Ramet
conducted a review of 130 books pertaining to the Yugoslav case and discussed a large number of debates
surrounding the country’s break-up and the ensuing conflicts.
181
For a good review of the main arguments, see Dejan Jovic, “The Disintegration of Yugoslavia.”
179
316
frustrations and fear could emerge and in which political elites felt threatened enough to
kick competition into high gear. Structural factors played their role by setting the stage for
political entrepreneurs. In a newly opened competitive political environment, it is however
the actions of these ethno-centric entrepreneurs, building propaganda machines, co-opting
cultural and historical repertoires, “fanning the flames of hatred for their own purposes,”
becoming the predominant voice, which polarised Yugoslav societies from the top down. 182
They stoked up ethno-centric sentiments in their constituents in the hopes of securing their
hold on power, in so doing they ‘prepared their people’ to accept resentment of the ‘other’–
the Albanian, the Croat, the Serb, the Slovene–in the name of the nation. 183
In the end, ethno-centric tactics led to more than dissolution. They also led to war.
A part of the actors in the Yugoslav crisis did not necessarily seek war. They did not
necessarily anticipate what would happen. 184 But conflict did explode and engulf three
republics of the former Yugoslavia, although admittedly not with the same intensity and
consequences for the three. A few years after the end of the war in Bosnia-Herzegovina,
the path that had been set by ethno-centric entrepreneurs would make new victims.
Kosovo also descended into war.
182
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals,” p. 404; Steven Majstorovic, “Ancient Hatreds
or Elite Manipulation?,” p. 170.
183
Tim Judah, “The Serbs,” p. 43.
184
According to Miloš Vasić, Milošević “himself was surprised by the fact that the war of ethnic extermination
gained such a momentum as to make it a self-supporting suicidal machine.” Miloš Vasić, quoted in Sabrina P.
Ramet, “A Theory about the Causes of the Yugoslav Meltdown,” p. 771.
317
CHAPTER 11: COMMUNICATION STRATEGIES IN THE FORMER YUGOSLAVIA
As the previous chapter illustrated, the break-up of the former Yugoslavia and the ensuing
conflicts in Slovenia, Croatia and Bosnia-Herzegovina were complex processes, which,
unlike Rwanda, were spread across a number of federated political units and involved half
a dozen nations and nationalities. Notwithstanding this complexity, a pattern is identifiable.
A large number of analysts agree that what occurred in the former Yugoslavia in the 1980s
and 1990s was primarily brought about by the actions of nationalist politicians. They hold
that the Yugoslav disintegration was a top-down, elite-led process. It followed the common
sequence of elite generated conflict in which, responding to a crisis that changes
opportunities, elites seek popular support, and instrumentalise their constituents’
grievances, beliefs and identity to obtain it.
To be sure, in the case of the break-up of Yugoslavia, elites both reacted to, and
created, conditions of crisis.
An economic and political crisis shook up the country,
rendering the Yugoslav system, a system traditionally authoritarian and controlling of
dissent, vulnerable to contention. Aggravating elements, such as the steps taken toward
devolution during the drafting of a new constitution in 1974, further weakened federal
power relative to that of Yugoslavia’s constitutive units; the death of the country’s leader
and father, for some the embodiment of the state itself, which deprived Yugoslavia of its
unifying symbol and mediator of internal disputes par excellence; economic problems that
brought the country’s economy to the verge of collapse and that imposed hardships,
including unemployment, rising living costs and fewer social services, on the Yugoslav
population–all culminated in a legitimacy crisis that called the status quo into question. The
Yugoslav state, faced with this multifaceted crisis, was overwhelmed and unable to cope.
The situation was likely worsened by international developments. Like Humpty Dumpty
sitting on the wall, Yugoslavia had tried to negotiate precariously a position between the
Eastern and Western blocs. When the Cold War confrontation ended, one of Yugoslavia’s
318
most fundamental raison d’être, resistance to these two blocs, ceased to be a uniting
force.
All these factors are merely conditions with indeterminate impacts in the absence of
human agency. As Ramet argued, in Yugoslavia “[t]he descent to internecine conflict
involved a syndrome in which these sundry factors came into play, were assigned values
and operationalized.” 1 The Yugoslav population certainly had a role to play in the tragic
events of the 1980s and 1990s. When Yugoslavs were confronted with their country’s
economic and political crisis, they reacted with fear–fear of what the future held, fear of
their economic prospects, but also fear of the uncertainty, unsure of what the shake-up of
the political system might mean. As is often the case, however, this fear was accompanied
by frustration, because the crisis worsened people’s lives. Queues formed for basic
commodities. When commodities were readily available, the money people earned could
not buy as many goods as before. In many cases, high unemployment meant that there
was no money earned at all. Carried by this fear and frustration, Yugoslavs reacted; they
took to the streets.
The drive to turn these contentious sentiments into a more uniform movement
came, however, primarily from political elites, rather than the population. The actions of
ethno-centric entrepreneurs catalysed the events and popular sentiments. In Yugoslavia,
those who deployed efforts to ‘operationalise’ the crisis, as Ramet put it, were the elites, at
first intellectuals, the clergy, but increasingly and, with time, overwhelmingly politicians.
The 1980s were a turning point for Yugoslav elites. Those who benefited from
privileged positions when the crisis hit faced the possibility of a collapse of the system that
had brought them to power. The power arrangements and relations they had learnt to
navigate and that they had used to achieve their ascendancy were crumbling. If the
1
Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown: The Serbian National Awakening
as a ‘Revitalization Movement,” Nationalities Papers, vol. 32, no. 4, 2004, p. 775.
319
system fell, they too could tumble. Cracks in the system, moreover, created space for
newcomers. After years of an authoritarian mode of functioning, which had weeded out
those unwilling to conform to apparatchik rules of engagement, changes allowed for the
emergence of new elites. For these newcomers, crisis created opportunity.
Whether they were members of the old guard seeing their world withering away or
new ones sensing the novel opportunities, elites found themselves in a competition for
positions, standing and power. At the political level, this competition took place in a
liberalised environment, particularly during the first multi-party elections across republics in
1990, when there emerged numerous new parties. But the competition also took place in a
liberalised public sphere. Whereas Tito’s Yugoslavia had controlled what could be put in
the public sphere, ensuring consistency with the state party’s (or more fundamentally the
Marshal’s) doctrine, the 1980s brought with them an increasing breakdown of controls.
This resulted in the resurgence of ideas, currents and ideologies that had previously been
taboo, as exemplified by the wave of criticism addressed posthumously to Tito’s regime,
criticism that went as far as to raise doubts about Tito’s Partisans actions during the
Second World War. This liberalisation of the public realm also transpired in the rise of
nationalism, the expression of which had been forbidden under Tito’s regime. Thus, what
occurred in Yugoslavia starting in the 1980s, particularly in the late 1980s and 1990s, is
the parallel liberalisation of the political game–political entrepreneurs would now have to
compete for votes, for popular support–and the liberalisation of the means to play this
game–with controls on what could be thought or said waning, candidates were now free to
campaign along the lines they chose.
Under these new circumstances, a large segment of elites chose to reinvent or
invent themselves as populists. They chose to rely on popular support to vault themselves
to power. This was, admittedly, in part, support some of them paid for. Slobodan
Milošević’s party, for example, had become by the end of the 1980s expert in organising
320
‘happenings of the people’, rallies made up of professional protestors or people whose
participation had been secured through money, food or drinks. To stress how much these
rallies represented an expression of the party’s popular legitimacy, they were called ‘street
democracy’. 2 Such manufactured demonstrations were intended to signal to opponents
where popular support lay. They were also organised in the hopes of gaining popular
support by means of a ripple effect, wherein seemingly ordinary people taking part in street
rallies would convince other members of the population to join them.
When appealing to the people, many politicians also chose to address them in a
specific sense, in a nationalist sense; they spoke to their exclusive people. 3 A large
number of these transformed or emerging elites relied on ethno-centric or nationalist
means to ensure their popularity, drawing on the background they shared with their
constituents, on their common identity and myths, to give direction to popular grievances.
Ethnocentrism or nationalism were attractive avenues for elites in the Yugoslav
context. As a means to unite the country after the Second World War, Tito had declared,
among other things, the importance of the self-determination of all Yugoslav nations. He
had thus made the recognition and respect of constitutive nations and nationalities one of
the country’s core principles, one that could guarantee unity. According to Tito, because
the self-determination of each group had been achieved once and for all within the
Yugoslav system, through an attribution of autonomy and a political entity to all constitutive
nations, the problem of nationalism had been solved. As a result, nationalism did not need
to be expressed anymore. It could not be tolerated anymore. Tito’s regime had struggled
to quell nationalism, to keep it at bay. Ironically, however, as Dejan Jovic indicated, this
2
Aleksandar Pavković, The Fragmentation of Yugoslavia: Nationalism and War in the Balkans (New York: St.
Martin’s Press, 2000), p. 106.
3
In his speeches, Milošević often referred to the people, a word he used in the nationalist sense. During his
speech in front of Kosovo Serbs in 1987, he exclaimed “[n]o one has the right to beat the people!”. The same
reference can be found in a speech he gave in March 1989, when celebrating the 600th anniversary of the
Battle of Kosovo. Agneza Bozić-Roberson, “Words Before the War: Milošević’s Use of Mass Media and
Rhetoric to Provoke Ethnopolitical Conflict in Former Yugoslavia,” East European Quarterly, vol. 38, no. 4,
2005, pp. 397, 402.
321
had served instead to make it ‘the main rhetorical antipode’ to the regime’s ideology and
precepts. 4 Nationalism had kept being recalled, raised, and mentioned as the primary
enemy of ‘brotherhood and unity’, cardinal rules of Tito’s Yugoslavia. Nationalism had thus
continued to be an alternative to ‘brotherhood and unity’ in a matrix of identities within the
Yugoslav population.
But it had become an antithetical alternative, the antithetical
alternative. In the Yugoslav system, nationalism was an antagonist force, the one
diametrically opposed to all that the system stood for.
In the 1980s, with Tito dead and with increasing discontent being directed at the
failings of the Communist system, its inability to manage the problems the country faced,
nationalism was an opportune option. Nationalism was an option that allowed established
and rising political entrepreneurs to differentiate themselves in an antithetical fashion from
the old Communist system and apparatchiki. It was the option that allowed these
entrepreneurs to present themselves as the ultimate alternative to the system, but one that
was at the same time familiar to Yugoslav populations, that called on their roots. What was
more, nationalism was the alternative that had the potential to offer these political
entrepreneurs an unwavering constituency, at least if nationalist sentiments were
successfully recalled. If reawakened, nationalism promised attachment to the group and,
by extension, support of the group’s leaders, those who stood up for it.
Nationalism was the path a relatively large segment of politicians chose to take to
power in Yugoslavia in the 1980s. Many chose nationalism as the platform to sway their
exclusive people to their cause, to reawaken nationalism in the hopes of awakening
support in their favour. Ethno-centric entrepreneurs strove to stoke up sentiments of
frustration and to awake ethno-centrism in the respective populations, hoping to gain their
support in the process. But if this is what ethno-centric entrepreneurs sought to achieve,
4
Dejan Jovic, “The Disintegration of Yugoslavia: A Critical Review of Explanatory Approaches,” European
Journal of Social Theory, vol. 4, no. 1, 2001, p. 105.
322
how successful were they? Although political elites have wishes and goals, and develop
platforms to try to gear popular sentiments and beliefs towards their aims, there is no
certainty as to whether they will succeed. Much, if not all rests, on how populations will
react to these platforms, to their leaders’ calls for support and mobilisation. Mobilising
support is as much about the request for support, communicating a need for collective
action, as it is about the response to this request, whether the audience buys into calls for
support.
In the former Yugoslavia, political elites, with the support of intellectuals, the clergy
and the media deployed vast efforts to stoke ethno-centric sentiments in their
constituency, but did populations in fact believe what their leaders told them? Did they buy
into nationalist platforms? The remainder of this chapter assesses the effectiveness of
ethno-centric rhetoric in the former Yugoslavia and studies the strategies behind the
rhetoric. It first analyses the impact of the political rhetoric heard in the former Yugoslavia
in the 1980s and early 1990s. Did popular beliefs and behaviour, in effect, change in
reaction to the nationalist rhetoric promoted by a large number of politicians in Yugoslavia
at the time? It then turns to an analysis of nationalist platforms deployed by Serb and
Croat leaders. It specifically looks at tactics aimed at fostering Serb and Croatian solidarity
through the recourse to opposition, politicisation and simplification.
Impact of Ethno-Centric Rhetoric in Yugoslavia: Did Populations Believe?
There is no consensus in the literature on the degree to which ethnocentric/nationalist
rhetoric was effective in the former Yugoslavia, that is, on the degree to which such
rhetoric swayed the republics’ populations. While a large portion of authors assume that
populations supported their leaders’ platforms to a certain degree, others believe
nationalist sentiments among the population were superficial. John Mueller, for example,
remained unconvinced of Yugoslav populations’ support for their leaders’ nationalist
323
causes. As he explained, “support for militant nationalism in Yugoslavia was not all that
deep even at the time of its maximum notice and effect in the early 1990s.” 5 Mueller points
to three elements that should make us question the existence of fierce nationalist popular
sentiments: (1) results obtained by nationalist parties during elections at the republican
level in 1990 did not translate into a landslide victory (though the system exaggerated
winnings and seats obtained); (2) poll data indicated, at the time, that a good proportion of
Yugoslavs wished for Yugoslavia to remain unified; (3) once the violence broke out, it was
mainly perpetrated by bands of “opportunistic, sadistic, and often distinctly nonideological
marauders”, more than by die-hard nationalists. 6 Mueller’s claims are important and must
be addressed if a case for the effectiveness of elite appeals–for their impact on
populations–is to be made.
Mueller’s claims are accurate. The implications he infers from them, however, are
questionable. Mueller contended that fighting during the wars in Slovenia, Croatia and
Bosnia-Herzegovina was principally undertaken by armed bands of thugs and militias, and
not necessarily by convinced nationalists. The militarisation of Yugoslav societies in the
early 1990s lends credence to this contention. Massive arms transfers were reported, a
large part going to arm ‘defence groups’, militias, of the different nations. These militias
compensated for deserters and the poor results obtained when trying to draft regular
soldiers, for the confusion or disillusion of enlisted soldiers, who according to Mueller
“professed they did not know why they were fighting,” as well as for deserters. 7 Those who
did join militias were not necessarily nationalists, however. Much like was the case with
militias in Rwanda, strong nationalist convictions were not a prerequisite for belonging to a
militia. As a matter of fact, Miloš Vasić estimated that ‘semi-independent paramilitary and
5
John Mueller, “The Banality of ‘Ethnic War’,” International Security, vol. 25, no. 1, 2000, p. 45.
The poll data cited is from a survey conducted on the entire Yugoslav territory in summer and fall 1990.
According to Mueller, “[t]he question, ‘Do you agree that every (Yugoslav) nation should have a national state
of its own?’ elicited the following responses: completely agree, 16 percent; agree to some extent, 7 percent;
undecided, 10 percent; do not agree in part, 6 percent; and do not agree at all, 61 percent.” Ibid., pp. 43, 46.
7
Ibid., p. 48.
6
324
warlord units’ were comprised of a mere 20 percent ‘fanatical nationalists’. 8 The larger
segment of recruits seemed, according to Mueller, to have been attracted by greed or
bloodlust.
In and of itself, who perpetrated the violence, and whether they did it for nationalist,
instrumental or even sadistic reasons does not necessarily imply that other groups did not
have nationalist convictions. The fact that most of the violence was committed by militias
and armed bands does not mean that the Yugoslav populations did not have nationalist
sentiments. Mueller, however, insisted that a refusal to fight was indicative of a lack of
nationalist sentiments on the part of populations. But draft dodging was not unique to the
early 1990s in Yugoslavia. Xavier Bougarel explained that throughout modern history of
the region, armies repeatedly had a difficult time recruiting soldiers among their population
to go to war, and as a result turned to militias. 9 A reliance on militias, Bougarel explained,
was a common feature of fighting in the Balkans. All that can be ascertained from
recruitment problems faced by military institutions is that for an unclear reason, Balkan
populations were averse to conscription. Whether or not populations supported the
nationalist cause is unclear from this information alone.
In defence of his argument on the shallowness of nationalist support, Mueller relied
on two other contentions, the continued existence of attachment to Yugoslavia within
Yugoslav populations and the unimpressive electoral results obtained by nationalist parties
in the 1990 elections. Unlike his claim about the militias, these two contentions deal
specifically with popular beliefs. Mueller is not the only author to have identified a rise of
‘Yugoslavism’ in the 1980s and early 1990s. Jovic, for example, explained that two parallel
trends marked the last decade and a half before Yugoslav disintegration, the ethnicisation
8
Ibid., p. 50.
Bougarel indicated that “[i]n the former Balkan wars, as in the recent Yugoslav ones, militias and paramilitary
formations made up for the difficulties and inability the states had in sending their own population to the front.”
Xavier Bougarel, “Yugoslav Wars: The ‘Revenge of the Countryside’: Between Sociological Reality and
Nationalist Myth,” East European Quarterly, vol. 33, no. 2, 1999, p. 173.
9
325
of Yugoslavia and the re-emergence of declared belongingness to Yugoslavia. The
number of declared Yugoslavs had sharply risen from 1971 to 1981, from 273 000 to
1 219 000, and the trend is expected to have continued throughout the 1980s. 10 Ana Devic
described a similar trend occurring later, in the mid-1990s, and indicated that “sixty-two per
cent of Yugoslavs in mid-1990 believed that their ‘Yugoslav’ affiliation was very important
for them.” 11
All that said, continued sentiments favouring the unity of Yugoslavia do not
necessarily preclude the existence of ethno-centric sentiments, understood as heightened
attachment towards one’s group and resentment directed at other groups. One difficultly is
that Mueller conflates ethnocentrism and nationalism. In a sense, the mistake is a common
one with regard to the Yugoslav case because groups are called narod, translated as
nation, and narodnost, translated as nationality. Platforms agitated by elites in Yugoslavia
during those years are often referred to as nationalist, because they pertain to nations and
nationalities. But nationalism, or wishing to obtain recognition of sovereignty or a country
for one’s group, is not necessarily the same as ethnocentrism, which can be defined as
heightened positive sentiments for one’s group and, in tense situations, heightened
negative sentiments towards other groups. Mueller, in trying to disprove the existence of
hatred between groups pointed to opinion polls conducted in the early nineties in which the
various Yugoslavian communities voted against the break-up of Yugoslavia, against their
republic’s independence. He pointed to polls that measured nationalism, understood in the
traditional sense, and not ethno-centric sentiments.
Not all groups in Yugoslavia agitated for their own state. Kosovar Albanians did
agitate for independence recurrently during the Tito years and later in the 1980s. In the
mid-1980s, some Slovenian nationalists integrated independence into their nationalist
10
Dejan Jovic, “The Disintegration of Yugoslavia” p. 107.
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals: The Case of Yugoslavia,” International Journal
of Politics, Culture and Society, vol. 11, no. 3, 1998, pp. 392-393.
11
326
demands, although a number of Slovenian elites continued to advocate for confederation
as a solution to keep Yugoslavia together until late in 1990. But Serbs did not advocate
separating from Yugoslavia. On the contrary, their nationalist political plans implied the
continued existence of Yugoslavia, even a recentralised Yugoslavia, to ensure that all
Serbs belonged supranational entity. Serbs in the Krajina and Eastern Slavonia regions of
Croatia and in Bosnia-Herzegovina declared the independence of these Serb dominated
regions, but their declarations was more reactive than affirmative. These declarations
responded to Croatian moves towards secession (in 1991, which resulted in the formation
of Serb Autonomous Region of Krajina) and Bosnia-Herzegovina (in 1991, following the
discussion of a ‘Memorandum of Sovereignty’, Serb controlled areas of BosniaHerzegovina, East Herzegovina, Romanija, Northeast Bosnia and Bosanka Krajina, were
declared autonomous regions). 12 For Croats the autonomy of their republic was
sacrosanct and this was reaffirmed by nationalist leaders. But they did not seriously
manifest secessionist intentions until 1990, before which they continued to advocate in
favour of the confederal option.
Considering these trends, it is not surprising that a good proportion of Yugoslavs
voiced their disagreement with the dissolution of Yugoslavia or the granting of sovereignty
or independence to its constitutive units. Favourable views towards the dissolution of
Yugoslavia were not clearly advocated until late in 1990 (and then only by specific
communities), when the relations between groups grew tenser. At that point, Yugoslavia
became much more an entity in which the aspirations of some groups were being
constrained by the beliefs and actions of others at the federal level. Serbs and
Montenegrins recurrently blocked proposals by other republics, becoming an obstructionist
force. The federal system, at the mercy of this force, appeared to be a repressive tool
wielded by ‘foreigners’. Until this shift, Yugoslav identity was not necessarily in
12
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 159.
327
contradiction with national identities, including more ethno-centric interpretations of
national identities. The situation had not reached a dead-end in which the national
aspiration seemed better served outside of Yugoslavia. The existence of ‘Yugoslavism’ or
the absence of nationalism understood in the traditional sense, prior to 1990, is not
however a clear indication of the absence of sentiments of group superiority and parallel
resentment or even hatred of others. Ethno-centric sentiments could have developed in
Yugoslavia within existing federal political arrangements.
As for the resurgence of ‘Yugoslavianism’ in the mid-1990s, though more research
should be conducted on the subject, it could be attributed to ‘war fatigue’, a feeling
expressed by combatants and populations when war, which strains their existence,
exhausts them. Already six months after the start of the conflict in Croatia, for example,
morale on the two fighting sides was low, which, in part, led to the stalemate that allowed
for the international community’s intervention. 13 The conflict in Bosnia-Herzegovina,
exacting its toll on both combatants and populations in terms of lives, displacements and
destruction, could have led to nostalgia for the war-free past, an idealised Yugoslavia
during which quarrels between nations were kept at bay. Devic indicated that “[t]he ceasefire between the Serb, Croat and Bosnian armies, enforced by the Dayton Accord [1995],
was both preceded and followed by a decrease in expressions of ethnic hostilities among
the ‘masses’.” 14 Though much work remains to be done to understand this new wave of
Yugoslavism in the mid-1990s, it should not, however, necessarily be taken as an
anomaly, and certainly not as a sign of the absence of ethno-centric sentiments in prior
years.
Mueller’s final point pertains to the results of the 1990 republican elections. Most
agree with Mueller that these elections were not a sweeping victory on the part of
13
14
Ibid., p. 146.
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals,” p. 376.
328
nationalist parties. Despite the numbers of seats they won, vastly exaggerated by ‘winnertake-all’ electoral systems, “most nationalist parties did not win large majorities of the
vote.” 15 While it is certainly true that in the absolute, the results obtained by the larger
nationalist parties were not sweeping, doubts can be raised as to how this should be
interpreted.
It should be recalled that, as is often the trend in political systems newly opened to
multiparty competition, a flurry of new parties emerged in Yugoslavia prior to the 1990
elections. Competing in the 1990 elections were old communist parties reinvented as
nationalist ones, for example Milošević’s party in Serbia, the Socialist Party of Serbia, or
powerful new parties, such as Franjo Tudjman’s HDZ in Croatia. In all republics, however,
these larger formations also competed with smaller ones. Among these new political
formations, a number of them shared political goals with nationalist parties and certainly
played a role in eating away at the larger parties’ electoral results. As a result, there is a
strong likelihood that the nationalist vote was split across parties that proposed nationalist
or nationalist inspired platforms. More importantly, it should be remembered that most
republics were populated by large minorities. These certainly need to be taken into
account when analysing the results of the 1990 elections. 16 In Croatia, for example, the
HDZ’s claim to 44 percent of the vote must be understood in light of demographic realities.
Croatia was composed of 78 percent Croatians, 12 percent Serbs and 10 percent ‘Others’.
Considering the HDZ’s ethno-centric platform, it would have been highly unlikely that the
party manage to sway the votes of non-Croats. Had all Croats decided to vote for the HDZ,
it could have hoped to obtained, at the most, 78 percent of the vote. Similarly, with respect
to Serbia, while some analysts present the 45.8 percent obtained by Milošević’s SPS as
15
Anthony Oberschall, “The Manipulation of Ethnicity: From Ethnic Cooperation to Violence and War in
Yugoslavia,” Ethnic and Racial Studies, vol. 23, no. 6, 2000, p. 995. See also Ana Devic, “Ethnonationalism,
Politics, and the Intellectuals,” p. 393.
16
Numbers are taken from Franke Wilmer, The Social Construction of Man, the State, and War: Identity,
Conflict and Violence in the Former Yugoslavia (New York: Routledge, 2002), p. 54.
329
poor results, the fact that Serbia is populated by a large minority of Albanians and
Hungarians should indicate that sweeping results were not to be expected. Serbs
constituted 66 percent of the population of the Serbian republic at the time. Even if all had
voted in favour of the SPS, their votes would not have amounted to a landslide victory.
Furthermore, despite this proportion of Serbs in Serbia, Milošević’s secured 65.4 percent
of the popular vote in the election for the Serbian presidency in 1990, a fact that attests to
his popularity, even though he was one of the more vibrant nationalist in Yugoslavia at the
time. 17 The more telling numbers of the elections, however, are certainly those obtained in
the elections in Bosnia-Herzegovina. Election results in Bosnia-Herzegovina clearly
mirrored demographic realities. The SDA, the more nationalist Muslim party, was elected
with 37.8 percent of the vote. Muslims in Bosnia-Herzegovina represented 44 percent of
the population. The SDS, the Serbian party, obtained 26.5 percent of the vote, Serbs
representing 33 percent of the entire population; and Croats won with 14.7 percent of the
popular vote, Croats representing 17 percent of the entire population of BosniaHerzegovina. Overall, with demographics taken into account, the results of the 1990
elections do not make a clear case for the shallowness of nationalist support. On the
contrary, they bolster arguments that point to the role played by nationalist votes. It is
highly unlikely that Serbs in Croatia voted for the Croatian nationalist party, that Albanians
in Serbia voted for Milošević and his nationalist platform, and that people voted for the
other nation’s party in Bosnia-Herzegovina. Nationalist victories could not be sweeping
because different nationalisms competed.
None of these elements–who fought during the wars, the resurgence of
‘Yugoslavism’ and the results obtained during the 1990 elections–confirms or disproves
claims of popular support for ethnocentric rhetoric.
Trends observed among the
population are a more direct, though admittedly not perfect, indicator of the extent to which
17
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 121.
330
ethno-centric rhetoric took hold of Yugoslav societies. The degree to which people change
their behaviour from the ‘normal frame’, the degree to which they integrate ethno-centric
behaviour to their daily activities and relations, when exposed to ethno-centric
propaganda, can be indicative of the degree of absorption of this rhetoric. 18 Change
observed in more personal forms of contact, with friends and family, even among
colleagues at work, in school, or in church, are significant to the extent that in this personal
realm, behaviour is generally less guided by external norms and ideology, than by
personality and personal convictions. In the personal realm, individuals feel freer to
behave and express their personal beliefs over prescribed ones.
As a result, ethno-
centrism passing into the personal realm has a greater potential of being representative of
a change in attitude that changes observed in more public fora, which are easier to stage
or coerce. Changes observed in the personal realm are a better indicator of the
acceptance of ethno-centric rhetoric, a decision people thus make much more on their
own.
It is specifically this type of shift that was observed in the former Yugoslavia over
the course of the crisis that led to the 1990s wars. Yugoslav society went from a more
plural state (although an authoritarian state, a multiplicity of identities were tolerated,
though subsumed by a supranational Yugoslav identity) to a polarised state, a state in
which national identities became the more prominent ones, not for all Yugoslavs, but for a
good proportion of the population.
A number of authors agree that, at the start of the crisis in the 1980s, Yugoslavia
was a pluralist environment. As a result of Marshal Tito’s efforts to downplay nationalist
identities, individuals expressed a flurry of identities, from a supranational one,
‘Yugoslavism’, to national/republican, to civic and personal ones. As Devic indicated,
18
Anthony Oberschall also spoke of normal frame, by comparison with a crisis frames. He did not, however,
describe the process through which a crisis frame can displace a normal frame. Anthony Oberschall, “The
Manipulation of Ethnicity.”
331
“many studies [...] suggest that some civic identities had been developing in Yugoslavia for
almost five decades.” 19 Furthermore, these identities were not necessarily contentious. On
the contrary, prior to events in the 1980s and 1990s, they tended to be “broader, more fluid
and negotiable.” 20 Relations in communities seemed to mirror these more cosmopolitan
identities. Cheryl Bernard, who worked with Muslim refugees during the Bosnian conflict,
explained that all of those she spoke with described their relations with Serbs prior to the
war as amicable. Indeed, many work relationships and friendships cut across national
lines. 21 Even more telling is Anthony Oberschall’s account of interethnic relations during
the Tito years and in the 1980s. According to him, “ethnic relations were cooperative and
neighbourly. Colleagues and workers, schoolmates and teammates transacted routinely
across nationalities. Some did not even know or bother to know another’s nationality.
Intermarriage was accepted.” 22 In light of this congeniality in many communities, a number
of people did not think open war possible and were taken by surprise at how quickly the
situation deteriorated. 23
The crisis and rhetoric did change things, however. Analysts of the Yugoslav crisis
reported that, as occurred in Rwanda, difficult conditions began straining relations. Based
on interviews conducted in Yugoslavia, Oberschall explained that developments at the
political level impacted on interethnic friendships. People stopped discussing politics and
public affairs with friends who belonged to another group, or they cut ties with former
friends and “turned for discussion of such matters to a fellow ethnic with whom agreement
19
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals,” p. 376.
Franke Wilmer, The Social Construction of Man, the State, and War, p. 29.
21
John Mueller, “The Banality of ‘Ethnic War’,” p. 54.
22
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 989.
23
For example, with regard to the Muslim population in Bosnia Herzegovina, Samantha Power explained that
“[e]ven once it was clear that was consuming Bosnia and the radio brimmed with gruesome reports of
summary executions and rapes, Muslims continued to console themselves that the war would never infect their
neighborhoods. ‘That’s a long way off,’ they would say. ‘We have been living together for years’.” Emphasis in
original. Samantha Power, “A Problem from Hell”: America and the Age of Genocide (New York: Harper
Collins, 2002), p. 255. See also Anthony Oberschall, “The Manipulation of Ethnicity,” p. 988.
20
332
was likely.” 24 Within communities, people began avoiding each other. This led to a
situation where, by keeping to their own group, and on account of an increasingly biased,
ethnicised media coverage, people lost access to diverse accounts and sources of
information. As they progressively relied more on information stemming from their own
group, “each group bec[ame] encapsulated; dialogue and understanding cease[d].” 25 In
this impoverished environment, people increasingly began lending credence to what they
heard. This trend was reflected in the results of opinion polls conducted among the
Serbian population in 1992 by the Institute of Political Studies in Belgrade. In these polls,
over 60 percent of television viewers in Serbia indicated that they ‘did not doubt the truth of
what they heard on television’. 26 Overall, with ethno-centric political platforms upheld by a
large segment of Yugoslav politicians as well as coverage and inflation of ethno-centric
rhetoric by co-opted media, nationalist attitudes began passing into society. Increasingly,
according to Oberschall, “people echo[ed] the intellectuals’ and the media crisis
discourse.” 27
Due to this radicalisation of political discourses, Yugoslav society was polarised.
With mounting tensions and increasing ethno-centric rhetoric, identities, which had been
fluid and mostly amorphous, began crystallising. Slavenka Drakulić observed this
phenomenon occurring in Croatia and described it at length:
Along with millions of other Croats, I was pinned to the wall of nationhood [...]. That
is what the war is doing to us, reducing us to one dimension: the Nation. The
trouble with this nationhood, however, is that whereas before, I was defined by my
education, my job, my ideas, my character–and yes, my nationality, too–now I feel
stripped of all that. I am nobody because I am not a person any more. I am one of
4.5 million Croats. 28
24
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 993.
Ibid.
26
Sabrina P. Ramet, Thinking about Yugoslavia: Scholarly Debates about the Yugoslav Breakup and the Wars
in Bosnia and Kosovo (Cambridge: Cambridge University Press, 2005), p. 18.
27
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 991.
28
Quoted in Franke Wilmer, The Social Construction of Man, the State, and War, p. 96.
25
333
Conflict and rifts at the elite level began filtering into societies and changing perceptions of
identities, changing them particularly from the more innocuous to the less accommodating.
Political issues and conflicts infiltrated personal and communal realms. Private and public
realms converged. Societies mirrored camps described by the various groups’ leadership.
These camps became part of the political landscape and people were increasingly called
up to side with one. In an interview, an individual told Franke Wilmer that “in 1991, when
hatred between the Serbs and the Croats started to grow in Croatia, I began to feel
pressure in my immediate environment to take sides.” 29
If rifts were created within Yugoslav societies, the process was not uniform,
however. Believing is not an open/shut process. People do not necessarily go from nonbelief to absolute belief, particularly when it comes to issues that would radically alter an
initial frame of belief, such as ethno-centric beliefs in a predominantly pluralist society.
What this means then is that different people believed differently and to different degrees.
Every Serb, Croat, Bosniak, for example, reacted differently to nationalist rhetoric. Some
rejected it, some were intrigued by it, and others bought into it wholeheartedly. In the
former Yugoslavia, like in most cases of mobilisation and conflict, dissenting voices rose to
oppose the growing radicalisation of Yugoslav societies and eventually to oppose the
wars. Anti-radicalism and anti-war critiques stemmed from the world of politics, from
intellectual circles, in particular from a younger generation of ‘social scientists,
philosophers and writers’, and from moderate clergy. 30 But dissent was not solely the
prerogative of elites. Individuals within Yugoslav populations also expressed their
opposition to unfolding events. Often, they joined protests in defence of their stance.
Although a segment of the Yugoslav population did not buy into nationalist
platforms and ethno-centric rhetoric, others did however. Admittedly, not everyone
29
Franke Wilmer, The Social Construction of Man, the State, and War, p. 110.
Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown: The Serbian National Awakening
as a ‘Revitalization Movement’,” Nationalities Paper, vol. 32, no. 4, 2004, p. 772; Ana Devic,
“Ethnonationalism, Politics, and the Intellectuals,” p. 402.
30
334
belonged to the ultra-nationalist fringe, those indoctrinated and fervent enough to join a
militia–the 20 percent of ‘fanatical nationalists’ among militiamen. 31 Changes in social
relations and identity patterns were indicative of a trend within society, however. They
revealed that enough tensions passed into ordinary people’s lives to produce changes in
daily interpersonal exchanges and worldviews. Populations began echoing elite rhetoric.
Nationalist leaders were voted into office, which indicated these leaders had a base of
supporters. 32 Though the entire Yugoslav population did not believe in awakened
nationalism in the 1980s and 1990s, a segment did. And these supporters of nationalist
platforms gave ethno-centric entrepreneurs the power to operate. As Oberschall argued,
“[w]ithout the tacit, overt or confused support of the majority, the nationalist leaders could
not have escalated ethnic rivalry and conflict into massive collective violence.” 33
Opposition, Politicisation and Simplification in the Former Yugoslavia
Feiwel Kupferberg claimed in a paper on ethnic conflicts in Eastern Europe that
[f]ar from confirming the resilience of ethnic hatred and the strength of ethnic
identities in the former Yugoslavia, the level of violence suggests the opposite. It
was only because Yugoslavia had become an ethnic melting pot that the struggle to
separate what had grown together had to be so fierce. 34
It would be difficult to prove that violent hatred existed between Yugoslav populations in
the decades prior to the crisis and wars of the late 1980s and early 1990s. It would be
equally difficult to point to the existence of ancient hatreds, simmering just below the
surface, whether innate or collectively learnt from times immemorial spent ‘fighting’ the
others. It would be difficult because, as was just seen, life in Yugoslavia under Tito was
relatively free of intergroup disputes. Intergroup relations were relatively amicable. But a
shift did occur. A change took place. A process of persuasion and slow acceptance of an
31
John Mueller, “The Banality of ‘Ethnic War’,” p. 50.
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 998.
33
Ibid., p. 999.
34
Ana Devic, “Ethnonationalism, Politics, and the Intellectuals,” p. 406.
32
335
ethno-centric norm operated, wherein people went from peaceful to tense cohabition, to
talking to and visiting one another to avoiding each other, from friendship to distance. The
culprit behind this shift is not the re-activation of innate or collectively learnt violence. The
story of the Yugoslav break-up and wars is more complex.
If the violence is to be tied to the past, it should be tied to a re-envisioned past, to a
recreated image of what intergroup relations had been, and what this meant for what they
should be in the present. It is by exploiting national memories and myths, as well as by
playing on perceptions of collective identities born of these memories and myths, that
ethno-centric entrepreneurs called on their constituents for support. It is through their
rhetoric, built around these myths, national memories and identities, that they sought to
deemphasise the ties and identities that had been fostered by Tito’s Yugoslavia, spark
ethno-centric sentiments, and rekindle group solidarity to achieve their aims.
Yugoslavs had their myths, memories and identities. Tito’s Yugoslavs had their
‘Yugoslavism’ and their myth-ideology, the founding of their country for ‘brotherhood and
unity’, and their struggle in the face of East and West. But Yugoslavs also had ‘national’
myths and identities, Albanian, Croat, Muslim, Macedonian, Montenegrin, Muslim, Serb
and Slovene myths, protected by recognition to the national principle found in Yugoslavia’s
political arrangements. But myths and identities can be reinterpreted, rearranged. And it is
the meaning, and the signification in particular, of national myths and identities that
changed in Yugoslavia. By casting a different light on what people understood to be the
groundings of who they are, who others are, where they come from, where they are going,
and what they are entitled to–admittedly at a time where people felt aggrieved–senses and
sentiments shifted.
336
For that to occur, myths had to become nationalist teleologies. 35 The mythical past
and national histories were tied to the present, a direct line imagined between the two. The
present was interpreted through the lenses of the mythologised past. Leaders collapsed a
complex, unfolding present around simplified, almost caricatured myths of the past and of
nationalist histories that had been impoverished by centuries of repetitions. They
overshadowed the complexity and ambiguity of the present with the simpler renderings,
categories and labels of the imagined nation and myths. By so doing, they thus ‘mythified’
the present. This mythified present was comprehensible, readily apprehensible, to those
who had grown up with these myths, among their family and community. It resonated.
Politicians therefore achieved resonance by building their readings of the situation,
of people’s present, around the national myths and histories their constituents shared.
There was in fact little new material at the base of their appeals and of the myths and
schemas of identification they called up. They built on old foundational myths, such as the
Battle of Kosovo for Serbs and millennial state rights and defence against the East for
Croats. Serbian leaders, by drawing on common references of Serbian culture, “memories
of vanished glories [that] had been kept alive in legends and folk songs on which every
Serb child [...] has been reared for the past six centuries,” struck a responsive chord in
their constituents. 36 Similarly, by grounding their nationalism in a response to Serbian
nationalism, to Serbs as representatives of the East, the barbarians, Croats leaders reinvented themselves as their ancestors, standing to defend thousand year old Croat
institutions and the Catholic world, as the Antemurale Christianitatis. Yugoslav elites also
built on “the experiences and memories of the Balkan wars, the first and second world
wars–and other wars before that.” 37 But experiences and memories of wars were often
mythified themselves, to a great extent with the passage of time. Wars are such traumatic
35
David Bruce MacDonald, Balkan Holocausts? Serbian and Croatian Victim-Centred Propaganda and the
War in Yugoslavia (Manchester: Manchester University Press, 2002), p. 64.
36
Dusko Doder and Louise Branson, Milosevic, p. 51.
37
Anthony Oberschall, “The Manipulation of Ethnicity” p. 989.
337
events that they are remembered collectively less as careful historical accounts, but
instead as the story of heroes and sufferings. Nationalist elites built on these revised tales
of war. They also borrowed symbols from the past. Serbian examples of this included the
“rediscovery of epic songs and [...] a general ‘folklorization’ of political vocabulary and
symbols.” 38 Tudjman, for his part, reintroduced the kuna, a medieval Croatian currency
also reintroduced during the Ustaša regime, and the šahovnica, a red-and-white
checkerboard emblem found in the Croatian coat-of-arms. 39 Political entrepreneurs in the
former Yugoslavia “breathed new life into popular memories and myths”, rearticulating
them and reading them unto present problems and frustrations. 40 Couched in these
references and images, elites’ rhetoric “resonated at the grass roots”. 41
Resonance alone does not necessarily bring about support, however. Elite rhetoric
also needed to foster a heightened sense of group attachment to be operative, in order to
translate into support. For that, Yugoslav ethno-centric elites played on some mechanisms
to rekindle group solidarity. Serb and Croat leaderships built nationalist platforms on
opposed, politicised and simplified perceptions of group identities and events. These, more
than resonance, pressed the need for solidarity.
SERBIAN NATIONALIST REVIVAL THROUGH THE LENSES OF SOLIDARITY BUILDING
Serbian nationalist rhetoric during the 1980s and 1990s constructed a resonant narrative
on the unbroken link between the kingdom of the Nemanjas, the Serbian golden age, lost
and awaiting to be rebuilt, and the events of the late twentieth century. Though the Battle
of Kosovo and the themes attached to it became part of an elaborate nationalist platform
aimed at awakening Serbian nationalist consciousness, it is the manner in which it was
exploited that pressed the need for Serbian solidarity. Drawing on myths, memories and
38
Xavier Bougarel, “Yugoslav Wars,” p. 167.
Sabrina P. Ramet, Thinking about Yugoslavia, p. 6.
40
Ivelin Sardamov, “Identity’s Role in the Serbo-Croatian Conflict,” Peace Review, vol. 9, no. 4, 1997, p. 463.
41
Anthony Oberschall, “The Manipulation of Ethnicity” p. 999.
39
338
current events, the Serbian leadership constructed opposed perceptions of group identities
(it identified Serbia’s enemies and the sources of Serbian moral and religious superiority),
it politicised Serbian identity (it tied Serbian nationalist revival to the defence of Serbian
rights), and finally it simplified readings of the identities and events. By collapsing identities
and categories around simpler versions, these identities and categories could be tied to
roles found in Serbian mythology.
Serbian nationalist rhetoric created a sense of threat, a sense that Serbia faced
enemies. In the same way that Prince Lazar and Serbs had been threatened by the
Ottomans in their golden age, now Serbs were threatened again. Although not a concerted
effort, over time Serbian nationalist rhetoric developed an increasing sense of threat,
building on a growing group of foes surrounding the nation, in a wave of ‘Serbophobia’. 42
The initial push to revive Serbian nationalism, however, came with the perceived threat
Albanians posed in Kosovo. This was the threat to wake Serbs from their torpor. Albanians
threatened Serbs on their land; they threatened from within. A vicious campaign against
Kosovar Albanians began in the media. They were depicted as “irredentists, separatists
and terrorists.” 43 Kosovar Albanians were accused of destroying Serbian shrines, trying to
erase their presence in Kosovo. More importantly, Kosovar Albanians were accused of
erasing Serbs themselves from Kosovo, of orchestrating a genocide against Kosovo
Serbs. 44 Rhetorical artifices were used to stress the danger. One author spoke of the
‘demographic time bomb’, while an association of Serbian scientists denounced a
‘gynaecological conspiracy’, “expos[ing] a plot to make Albanian women more fertile, so
they could engender a ‘demographic explosion never before seen, the most potent in the
world’.” 45
42
David Bruce MacDonald, Balkan Holocausts?, p. 63.
Agneza Bozić-Roberson, “Words Before the War,” p. 402.
44
Stuart J. Kaufman, Modern Hatreds, p. 180.
45
David Bruce MacDonald, Balkan Holocausts?, p. 77.
43
339
While Kosovar Albanians constituted the initial threat, Serbian nationalist rhetoric
soon proclaimed the existence of a new plot against Serbia. The timing at the end of the
1980s seemed ideal. Serb control of Kosovo had been ensured through Milošević’s
annulment of its special status, but growing opposition was forming in other republics in
reaction to Milošević’s actions. Although, as Marković explained “Kosovo continued to be
the predominant issue” in 1988, nationalist rhetoric unearthed other conspiracies, other
threats, which promised to become the next menace to Serbia. 46 Yesterday’s Austrians
became today’s Croats with their Catholicism and Fascism, set in their genocidal ways like
their Ustaša grandparents and parents had been. And, with the looming conflict in BosniaHerzegovina, yesterday’s Ottomans became today’s Islamic Fundamentalists, Albanians
and Bosniaks. 47
Croats, in particular, were targeted by Serbian nationalist rhetoric. As David Bruce
MacDonald explained, “Serbian writers specifically targeted the Croats as the harbingers
of Serbophobia, viewing them as a truly Biblical antagonist, which had been operating
against the Serbs since the division of the Roman Empire.” 48 The campaign against the
Croatian threat began with the alleged mistreatments of Serbs in Croatia. In 1986, SANU
had claimed that, with the possible exception of the Ustaša regime, “the Serbs in Croatia
have never been as jeopardized as they are today.” 49 Increasingly, the media printed
stories of the ordeal endured by Serbs in Croatia, particularly following the Croatian Diet’s
decision to revoke the Serbs’ special status in the republic in the summer of 1989.
46
Zoran M. Marković, “The Nation: Victim and Vengeance,” in Nebojša Popov, ed., The Road to War in Serbia:
Trauma and Catharsis (Budapest: Central University Press, 2000), p. 593.
47
Renaud de La Brosse, “Les médias machines de guerre en ex-Yougoslavie,” Reporters sans Frontières, Les
médias de la haine (Paris: La Découverte, 1995), p. 127. With regard to Bosniaks, depictions centered
increasingly, once the conflict in Bosnia-Herzegovina started, on Bosnian Muslims as the reincarnation of the
Ottoman threat and on the larger, more modern, threat posed by Islamic fundamentalism. On the latter, Ramet
indicated that “[t]hroughout the war years, Serb propagandists never tired of accusing Bosnian president
Izetbegović of seeking to establish an Islamic fundamentalist state.” Sabrina P. Ramet, Thinking about
Yugoslavia, p. 14.
48
David Bruce MacDonald, Balkan Holocausts?, p. 64.
49
Tim Judah, “The Serbs,” p. 26.
340
Following this decision, journals published stories of vandalism and violence against Serbs
in Croatia, of discrimination against them, even of their assimilation. 50
More virulent forms of Croat scapegoating followed. The publication of Tudjman’s
book on the Second World War, revising the number of Serbs killed in the Jasenovac
complex, marked a new level of threat ascription. With memories of the Ustaša regime
revived, Croats suddenly became a genocidal enemy. Serbs in Croatia mounted a media
campaign around the sufferings they had endured at the hands of the Ustaša, taking
journalists and photographers to alleged mass graves of Serbs killed during the Second
World War. Bones were disinterred and commemorative ceremonies were held,
accompanied by what Michael Sells calls a ‘pornography of victimhood’, “glossy
magazines with graphic pictures of the torture and dismemberment of Serbs in WWII–
reviving, with the power of the repressed, the memories and bitterness that Tito had
decreed buried.” 51 This revival of memories of the Second World War could be
superimposed to the treatment reserved to Serbs in Croatia to drive in the threat posed by
Croatians. Like what the Ustaša had done to Serbs in the Second World War, Serbs in
Croatia risked genocide. 52 The threat was grave and thus concerned all Serbs. So great
was the portrayal of the Croat threat, amidst recounting the atrocities they had committed
in the past, the numbers of Serbs killed in Jasenovac, and the alleged atrocities against
Serbs in Croatia in the 1980s and early 1990s, that it drove in the need for Serbs to fall
back on their kind. As Bogdan Denitch explained, “[e]veryone was traumatised by all the
talk of world war two atrocities [...] even those who had seemed immuned to nationalism.
50
Zoran M. Marković, “The Nation: Victim and Vengeance,” pp. 597-599.
Michael Sells, “Crosses of Blood: Sacred Space, Religion, and Violence in Bosnia-Hercegovina,” Sociology
of Religion, vol. 64, no. 3, 2003, p. 313. See also, on media coverage, Renaud de La Brosse, “Les médias
machines de guerre en ex-Yougoslavie,” pp. 116-117.
52
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary. Antagonism and Defensive Solidarity
in ‘Palestine’ and ‘Former Yugoslavia’,” Social Anthropology, vol. 11, no. 3, 2003, p. 333.
51
341
Old personal ties and friendships crumbled as many intellectuals I knew, as well as friends
and family members, rallied to the defense of their own nation.” 53
To emphasise the extent of the Croat threat, anti-Croat propaganda focused on the
enduring nature of Croat hatred. The atrocities committed by Croats against Serbs in the
Second World War were not a unique event–except perhaps in terms of the degree of
violence. Croats, it was said, had hated Serbs for centuries. As a Catholic nation, Croats
had vowed to hate Serbs and Orthodox peoples, as part of a larger Catholic ploy “to
destroy all vestiges of Orthodoxy in the Balkans,” as part of a Catholic expansionist plot. 54
For a thousand years, since the Great Schism between Christians, Croats had lived to
hate the Serbs, secretly working to eliminate them. The genealogy of Croat hatred of
Serbs could be traced throughout history, one of the latest example of Croat-Catholic ploys
having been the massacres of Serbs under the Ustaša regime. Serbian nationalist rhetoric
tied the Catholic Church to the Ustaša regime. Serbian nationalists denounced clergymen,
such Cardinal Stepinac (who Croats sought to canonise) for their silence with regard to
Ustaša crimes. Individual clergymen were certainly responsible, but the entire Catholic
establishment had also played a part in the war. In an interview a Serbian ideologue
indicated that “[t]here is no doubt that the Ustasha were too few in number to carry out all
this if they had not been backed by a powerful organization such as the Catholic Church,
who instigated the ideology of a ‘religiously pure area’ in the battle for domination over the
Orthodox ‘schismatics’. 55
While Serbian nationalist rhetoric targeted a number of enemies in an attempt to
unite Serbs against them, it also sought to distance Serbs from these enemies. It did so by
presenting Serbs as both the defenceless victim and a chosen people. Discourses of
53
While Serbs revived memories of Croat atrocities during World War II, to press the genocidal nature of
Croats in the present, so did Croats insist of Četnik atrocities during the Second World War to stress their
genocidal intents in the 1980s and 1990s. Bogdan Denitch, Ethnic Nationalism: The Tragic Death of
Yugoslavia (Minneapolis: University of Minnesota Press, 1996), p. 81.
54
David Bruce MacDonald, Balkan Holocausts?, p. 84.
55
Zoran M. Marković, “The Nation: Victim and Vengeance,” p. 596.
342
victimhood served to place Serbs in the morally superior position of the victim over that of
its aggressor. Serb victimhood was a theme directly borrowed from Serb mythology. By
projecting the logic of the Battle of Kosovo on to current events, ideologues stressed the
Serbs’ lot as defenceless prey. Just like an alien force had defeated them on their lands
and subjugated them centuries ago, so Serbs now continued to be the innocent victims of
their alien enemies’ ploys. They were the victims of Albanian plans to dominate Kosovo.
They were the subject of Slovenian and Croatian Fascist conspiracies. They were also the
victims of the Vatican’s millennial schemes to extend Catholicism’s hold on the region.
According to Kaufman, Serbs built a ‘cult of victimhood’, in which they saw themselves as
having been “’overrun, tortured, killed and stolen from by the Turks, the Austrians, the
Bulgarians, the Germans, Tito [...]’; not to mention the Croats under the Ustasha regime.” 56
These recurrent torments inflicted on Serbs throughout their history were depicted in the
media as “an unending chronicle of ethnic martyrdom.” 57 To stress the martyrdom of Serbs
at the hands of others, Serb intellectuals drew parallels between what their people had
endured and the plight of the Jewish people. Orthodox clergymen, in a petition to
denounce the situation of Serbs in Kosovo at the beginning of the 1980s, explained that
the Jews had lived 2000 years of suffering and remembered themselves and their history
in all this, “[i]n a similar manner, the Serbian people have been fighting their battle at
Kosovo since 1389, in order to save the memory of its identity, to preserve the meaning of
their existence against all odds.” 58
Rhetoric around Serbian martyrdom allowed Serbs to rise above their enemies.
They were the victims, they held no responsibility for suffering what others had chosen to
do unto to them. They were blameless. 59 They thus stood on higher moral ground than
their enemies. As a result, this made them a chosen people. Just as Lazar had chosen to
56
Partly quoting a Serb journalist, Stuart J. Kaufman, Modern Hatreds, p. 177.
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 330.
58
David Bruce MacDonald, Balkan Holocausts?, p. 75.
59
Agneza Bozić-Roberson, “Words Before the War,” p. 404.
57
343
fall as a martyr to win the kingdom of heaven, so the martyrdom of Serbs throughout
history ensured their continued covenant with God. The sanctity of the Serb cause was
further confirmed by the fact that the Orthodox Church defended it. By siding with
Milošević and nationalists, Church representatives confirmed the spiritual underpinnings of
Serb nationalism. 60 Nationalist rhetoric presented the Serb cause and the Serbian people
as so great, that it led to exaggerated, even ludicrous claims, on the part of some Serbian
elites. The president of a Serbian political party, for example, exclaimed that “[e]very nook
of Serbian land and the Serbs themselves are a heavenly wonder, and an inspiration and
example to all other peoples and countries.” 61 Serbian author Olga Luković-Pejanović
claimed in her book, The Serbs: The Oldest Nation, that the Garden of Eden had been in
Serbia, that Serbs had invented writing and that Ovid, the Greek poet himself, had written
his work in Serbian. 62
Overall, Serbian rhetoric worked to developed two antithetical positions, one of
enemies of the Serbian people, bloodthirsty and forsaken, and another of a righteous,
chosen people, the Serbs. They created two Manichean positions, with a negative pole
threatening the righteous one. The situation justified standing for the Serbs, standing for
what they stood for, goodness because they were not attackers but innocent victims,
sanctity because they were the chosen ones, standing against enemies that threatened to
destroy this.
Beyond opposing group identities, Serbian nationalist rhetoric also worked to
politicise them. Serbian national revival was tied to injustices committed against Serbs and
to rights Serbs needed to recover. Much of this politicisation drew on the myth of the lost of
Great Serbia at the Battle of Kosovo. Serbs had known their golden age, when Serbian
territory had extended well beyond the rump borders it had been confined to. This territory
60
David Bruce MacDonald, Balkan Holocausts?, p. 75.
Quoted in David Bruce MacDonald, Balkan Holocausts?, p. 73.
62
Sabrina P. Ramet, Thinking about Yugoslavia, p. 16.
61
344
had been lost, but for a higher goal. Prince Lazar at the Battle of Kosovo had chosen to be
a martyr and won, for all Serbs, the heavenly kingdom. Serbian nationalist historically
interpreted this myth as also speaking of a second coming, of a rebirth for Serbs.
Particularly in the nineteenth century, Lazar’s tale was retold to inspire a Serbian rebellion
against the Ottomans. Lazar had given up the earthly kingdom, but according to revised
mythology, “[j]ust as Jesus is ‘coming back,’ so is Lazar,’” and Serbs would “make a state
again.”
63
Serbs had prevailed and endured Ottoman oppression, so they would be
rewarded. In fighting Ottoman oppression, Serbs would regain their land. The nineteenth
century version of the tale therefore spoke of a Serbian ‘fall’ from a ‘glorious past’ to a
‘degraded present’. 64 The degraded situation was redressed, however, when Serbs
managed to free themselves from Ottoman rule.
Twentieth century Serbian nationalism built on a very similar pattern. It also
developed the tale of an injustice committed against Serbs–robbery of their property and
denial of their rights, which needed to be redressed. According to Serbian ideologues and
politicians, although Serbs had freed themselves from Ottomans in the nineteenth century,
they had fallen prey to another form of oppression in the twentieth century, the one
imposed by Tito’s Yugoslavia. The Yugoslav system had been specifically designed to
discriminate against Serbs; it was a conspiracy to weaken them. Two arguments in
particular supported this contention, the fact that Serbia had been submitted an unequal
treatment in the Yugoslav federation and the fact that Serbia had been purposefully
handicapped, by having two autonomous provinces, Kosovo and Vojvodina, carved out of
its territory. According to the first argument, while other republics had been allowed to
flourish, particularly Slovenia and Croatia, Serbia had been kept in a state of
63
Quoted in Tim Judah, “The Serbs,” p. 27.
Matthew Levinger and Paula Franklin Lytle, “Myth and Mobilisation: The Triadic Structure of Nationalist
Rhetoric,” Nations and Nationalism, vol. 7, no. 2, 2001, p. 178
64
345
backwardness and been exploited economically. 65 In the late 1980s, Serbian media
published numerous accounts decrying the situation. With the help of statistics, journalists
painted the portrait of injustices committed against Serbia. One reporter concluded that, all
things considered, “it is completely obvious that over the past decades there has been a
systematic and systemic exploitation of Serbia.” 66 Among Serbian grievances towards the
Yugoslav system, the creation of autonomous provinces came first, however. The issue
was made particularly sensitive following the institution of the 1974 constitution, which
granted additional powers to the provinces, making them essentially as powerful the
republics. Serb nationalist recalled this in the 1980s, presenting the constitutional
arrangement as a “deliberate policy of destroying the Serbian national state.” 67 This ploy
on the part of the Yugoslav state had weakened Serbia in comparison to other republics;
no other republic had its hold on its territory upset by the creation of sub units. More
importantly, it had “stolen the Serbian homeland of Kosovo from Serbia.” 68
Kosovo was recovered through amendments to the Serbian constitution the
Milošević government made, annulling Kosovo and Vojvodina’s special status. With the
effective reintegration of Kosovo and Vojvodina to Serbia, Milošević had addressed one of
the more predominant Serbian grievances. Yet this did not curtail the politicisation of
Serbian nationalist rhetoric. With mounting tensions between Yugoslav republics and
amidst increasing discussion of separation in Slovenia and Croatia, Serb nationalist moved
to the defence of a new right. As Ramet explained, “Milošević, Ćosić, and other prominent
Serbs executed a dramatic escalation when they began to insist that all Serbs were
entitled to live in one state, in an expanded Serbia that would embrace all contiguous
65
Zoran M. Marković, “The Nation: Victim and Vengeance,” p. 594.
Milorad Vučelić in Duga, quoted in Zoran M. Marković, “The Nation: Victim and Vengeance,” p. 594.
67
Agneza Bozić-Roberson, “Words Before the War,” p. 396.
68
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 332.
66
346
lands inhabited by Serbs.” 69 Serbs, like others, had the legitimate right to live together in a
space whose borders were not unlike those of the Serbia of the past.
These new claims were initially aimed at Croatia, and later at Bosnia-Herzegovina,
the two republics that had large Serbian minorities. In the face of Croatian talk of
secession, Serbs reacted with historic claims to Krajina and Eastern Slavonia. A Serbian
geographer, Jovan Ilić set out to justify Serbs rights to the two regions. 70 With regard to the
Krajina, Serb claims could be traced back to Serb migration in the region, particularly in
the fifteenth century. Encouraged by the Austro-Hungarian Empire, Serbs had settled the
region as farmer-soldiers and defended this Vojna Krajina, this borderland, from the
Ottomans. In exchange, the Habsburg Emperor had granted these Serbs autonomy. 71 Ilić
defence of Serbian claims on the Krajina therefore rested on a legal argument: the region
had been ceded to Serbs by a charter in 1630. In actual fact, though, the AustroHungarians had eventually ceded the region again to Croatians at a later date. Yet the
Serb claim was also a moral one. Serbs had valiantly given their lives to defend the West
from invaders on these lands and thus “deserved the region as their reward.” 72 There were
no such historical myths to ground claims on Eastern Slavonia, however. Ilić built the case
for Serbian territorial rights on the fact that the region was predominantly Serb. Similar
arguments were eventually used to lay claim to certain parts of Bosnia-Herzegovina.
These claims turned the other republics’ secession efforts into an infringement of a
Serbian fundamental right, the right to be together, the right to their state. Others’
affirmation of their statehood thus was an injustice that violated Serbian rights. Read in this
light, this infringement of Serbian rights legitimised and justified Serbian actions against
the others’ independence. Serbs were defending their rights. They were fighting for what
was truly theirs. Whether the rhetoric centered on historic rights to Croatian or Bosnian69
Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown,” p. 771.
David Bruce MacDonald, Balkan Holocausts?, p. 80.
71
Aleksandar Pavković, The Fragmentation of Yugoslavia, p. 10.
72
David Bruce MacDonald, Balkan Holocausts?, p. 82.
70
347
Herzegovinian regions, or on the centrality of Kosovo for Serbs, all of it spoke of injustices
and rights and entitlements. They made the issues highly contentious, but at the same
time workable. Referring to rights and entitlements implied holding them up, defending
them, and redressing injustices committed unto the ‘us’. They speak of holding up ‘our’
rights, defending ‘our’ rights. They thus press the importance of solidarity.
Simplification tactics were also used by Serbian nationalists at the end of the 1980s
and in the early 1990s. These tactics typically took two forms, tactics to ensure the
predominance of nationalists’ rhetoric and strategies aimed at simplifying what was
conveyed to ensure its potency.
A key tactic to ensure the predominance of Serbian nationalists’ rhetoric was the
elimination of moderates. In the first stage of Serbian nationalist revival, opposing
moderates proved useful for the emergence of nationalist elites.
These moderate
nationalists or Titoist federalists were presented as the old guard who had to be dismissed.
Taking the antagonistic stance, adopting fervent nationalism, allowed new elites, or old
ones keen on accruing more power, to distance themselves from the old guard, to stand
out. These early moderates and anti-nationalists, such as Ivan Stambolić, President of
Serbia at the time of Milošević’s rise to power, were ousted through repeated references to
their failures and comparisons between their policies and the promises held by the new
rising guard. The coup to remove Stambolić was a media spectacle, accompanied by
mass demonstrations, marking the passage from the old to the new. The old guard were
said to be out of date and outmoded, of belonging to a former era, and more importantly,
as being as weak, not up to the task, and unable to solve Serbia’s pressing problems. The
strategy worked. Within months, Milošević proceed to purge all moderate Serbian
politicians from the party and in the Serbian government. 73 Thus, these initial moderates
and anti-nationalists, though opponents, played an important part in the emergence of
73
Agneza Bozić-Roberson, “Words Before the War,” p. 398.
348
nationalists. The moderates formed the antithesis that allowed the nationalists to shine
through. Their position could be turned into a weakness and justify their removal, and in
return comforted nationalists in their new found position of power.
Subsequent moderate groups who rose in reaction to the wave of Serbian
nationalism were not judged to be as useful. On the contrary, they promised to dissolve
nationalist support. By proposing alternatives and agitating identities that could compete
with the resurgence of ethnic identities, they threatened to eat away at the ethnic solidarity
nationalists were trying to awake. A number of these were disposed of by staging
demonstrations and encouraging inflammatory media coverage against them, a practice
that discredited them and permitted their swift removal from power or dismissal from office.
Other purges also took place alongside these mediatised discreditations. According to
Sells, during those years, “Milošević purged the army, secret police, and other institutions
and place nationalist extremists in key positions of power.” 74 The coup de grâce to
moderates came after the 1990 elections, in which Milošević won the Serbian presidency
and his party took 78 percent of the seats in parliament. In this position of power, Serbian
nationalists had nearly a free hand in eliminating remaining moderates, and non-Serbs
from key positions, as well as intimidating intellectuals opposed to the nationalists’ agenda.
Stressing how dangerous the times were for moderates, a Serb economist explained that,
“[i]t was safer to take sides that being for peace in the middle.” 75
Next to the purge of alternative, mainly moderate views, Serbian nationalists
ensured the predominance of their rhetoric through another means, the control of major
Serbian media. In Yugoslavia, state owned media were governed at the republican level.
Media were first and foremost Serb, Croat, Slovene, Macedonian, etc. 76 The division of the
media along ethnic lines allowed nationalists to take control of state media outlets and to
74
Michael Sells, “Crosses of Blood,” p. 314.
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 996.
76
Renaud de La Brosse, “Les médias machines de guerre en ex-Yougoslavie,” p. 115.
75
349
use them to advance their extremist agendas. Milošević began to take control of the media
in Serbia as early as his initial putsch against moderates in 1987. At the time, he
successfully placed his supporters in decision-making positions of the key state run
television stations and largest newspapers in Serbia, which included replacing the entire
direction of Radio Televizija Beograd (RTB) and taking control of Politika, an influential
newspaper. 77 The take-over of Politika was particularly important since, as Agneza BožićRoberson explained, “the paper had a great influence on Serbian public opinion.” 78 In
following years, Milošević continued his take over the media, ensuring the coverage of
events in Serbia and the greater Yugoslavia as to support the Serbian cause and its
leaders. Co-opted media outlet continued throughout the crisis and the ensuing wars to
propose biased coverage and inflammatory rhetoric, leading Tadeusz Mazowiecki, former
Polish Prime Minister in charge of producing a report on the media for the UN Commission
on Human Rights, that “[t]he media in the former Yugoslavia were one of the most
important tools to contribute to the propagation of the military conflict in the region.” 79 The
media certainly contributed to flooding Serbian society with nationalist propaganda. This
take over proved particularly dangerous because people tended to read only their national,
their group’s, newspapers. 80 Consequently, people were regularly exposed to increasingly
ethno-centric political rhetoric, particularly following the purges of moderates. Their news
was ethnicised thanks to the control of the major media outlets by nationalists. Even
popular culture, including popular fiction, sporting events and rock bands–baptised ‘turbo
folk’ music–began signing a nationalist repertoire. 81 These tactics ensured the
77
Ibid., pp. 115-116.
Agneza Bozić-Roberson, “Words Before the War,” p. 400.
79
Renaud de La Brosse, “Les médias machines de guerre en ex-Yougoslavie,” p. 114.
80
Sabrina P. Ramet, Balkan Babel: The Disintegration of Yugoslavia from the Death of Tito to the Fall of
Milosevic (Cambridge, Mass.: Westview Press, 2002), p. 29.
81
For an account of the ethnicisation of culture in the former Yugoslavia, particularly the press and musical
scene, see Sabrina P. Ramet, Balkan Babel, pp. 57-104.
78
350
predominance of one message above all others; they ensure the predominance of the
extremist agenda.
Just as the erasure of alternative voices from Serbian society contributed to
simplify the ideological and political environment in Serbia, the language and rhetorical
strategies chosen also clarified existing options. Serbian nationalist proposed a platform
that was simple to better ensure its potency. It often played on known references, such as
references to Serbian myths, and on strong images, images of victimhood (atrocities
against Serbs), injustice (‘theft’ of Serbian land), but also of survival (continued resistance
of Serbs in other republics) and of triumph (celebration at Gazimestan of the reconquering
of Kosovo). These references, images and certain key words were repeated, reprinted ad
nauseam, the subject of innumerable broadcasts, in an attempt to give them a semblance
of truth, to make them so common that be taken for granted, unquestionable. 82
Among the more prominent of these images and words were simplified renderings
of groups’ identities. Perceptions of identities were collapsed around Manichean views of
‘us’ and ‘them’, around stereotypes. As Franke Wilmer explained, “[a]s the political climate
of hostility worsened, categories of identity were reduced to stereotypes and prejudices.
Perhaps worst of all, people more and more lost the ability to define for themselves what it
meant to be Serb, Croat, or Muslim.” 83 Group identities turned to caricatured versions of
what they had been: Serbs as civilised, sanctified, righteous, while others were
bloodthirsty, forsaken, misguided or blinded. In the rhetoric employed, for example, Croat
and Ustaša were regularly associated and at times presented, in certain accounts, as
synonymous. Bosniaks, for their part, were increasingly ‘turkified’ by these simplified
nationalist teleologies. As Sells argued, “it became possible to categorize them as Turks,”
ascribing to Bosniaks, by the same token, all traits ascribed to Turks, foreign, occupiers
82
83
Renaud de La Brosse, “Les médias machines de guerre en ex-Yougoslavie,” p. 121.
Franke Wilmer, The Social Construction of Man, the State, and War, p. 29.
351
(rediscovered as occupiers of Serbian territory in Bosnia-Herzegovina), infidels, etc. 84
Serbian rhetoric insisted that these were corporate, globalised categories. There was no
space in between. All individuals, whether Albanian, Bosnian Muslim, Croats and
Slovenes, were simply representations of their ethnic group. Individuals were subsumed
by their nation, erased before it. The identity of their nation completely absorbed theirs.
With nationality as the predominant marker for all, as Sardamov explained, “collective,
rather than individual actors, [became] the true bearers of rights, responsibilities, and even
guilt,” and in an oppositional setting made “individuals representing these communities [...]
innate and thus incorrigible.” 85 In Serbian political rhetoric, none could escape their
nationality, stereotypes applied to all. Issues were clarified, camps were clear. So were,
therefore, the reasons for siding with one’s group.
CROATIAN RHETORIC TO OPPOSE, POLITICISE AND SIMPLIFY
The revival of Croat nationalism in the late 1980s and early 1990s came, in large part, as a
reaction to events in Serbia, as a reaction to Serbian nationalism. Much of the Croatian
nationalist rhetoric was a mirror of what was presented by Serbian nationalists. Similar
themes were employed, and so were similar strategies. Like Serbian nationalist rhetoric,
and particularly because it was reacting to it, Croatian nationalism centered on Serb-Croat
antagonism. It concentrated on opposing these two identities. Although this constituted a
central feature of Croatian nationalist rhetoric, it also developed a rights-based discourse
around Croatia’s entitlement to statehood. Both these strategies were deployed within an
ideological and identity environment increasingly dominated by nationalist rhetoric
promoting stereotypes of Yugoslav nations and nationalities.
84
85
Michael Sells, “Crosses of Blood,”, p. 312.
Ivelin Sardamov, “Identity’s Role in the Serbo-Croatian Conflict,” p. 465.
352
Croats mythology and historical memories can be the source of a positive inward
focused nationalism. But in the late 1980s and particularly the 1990s, the nationalist revival
in Croatia was essentially a reactive one. The sources of its resurgence were external.
Croat nationalism was, in large part, a reaction to the rise of Serbian nationalism. Not
surprisingly, the principal instigators of the 1980s Croatian nationalist revival built their
platform around the idea of standing up to the Serbian menace. Tudjman, whose actions
played a significant part in carrying this revived nationalism, articulated his party, the HDZ,
platform “almost exclusively in anti-Serbian terms.” 86 Nationalist rhetoric thus centered
largely on this opposition with the Serbs.
For the most part, Croats nationalists reacted to what they saw as another episode
of Serb expansionism. Milošević–particularly in light of how he had conducted himself with
regards to Kosovo and Vojvodina, robbing them of their autonomy, and in light of his
increasing threats to claim Krajina, Eastern Slavonia, and parts of Bosnia-Herzegovina–
seemed undeniably intent on expanding Serbian borders and menacing neighbouring
republics. Elites built on this notion of Serbian expansionism, drawing on the past, to
inflate the threat into a plot to remake Greater Serbia. As MacDonald explained, “a long
and horrific history of Serbian imperialism and danger was created, so that the
contemporary crop of Serbian nationalists would not be seen as a cabal of power-hungry
opportunists, but rather as representatives of a typical, age-old Serbian strategy of
expansion and repression.” 87 Croat nationalists depicted Serbs as an unrelenting
threatening ‘other’. Serbs had always represented an other, from across the East-West
divide that split the territory that would eventually form Yugoslavia. They had been the
Eastern counterparts, backward, barbarian, “power-hungry, aggressive,” looking to expand
86
87
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 333.
David Bruce MacDonald, Balkan Holocausts?, p. 105.
353
their possession at all cost. 88 This had been the character of Serbs before the Ottomans
arrived, and centuries of subjugation to the Ottomans had turned these traits into a
penchant for cruelty and despotism. 89 In the nineteenth century, Serbian imperialism had
awoken again, to free Serbs from the Ottoman yoke, but also to recreate Greater Serbia,
the Serbia of the Golden Age that spanned a good portion of the Southern Balkans.
Nineteenth century Croatian authors, alarmed at nationalist fervour of their Serbian
neighbours, concluded that Serbs would turn to war and massacres to recover Greater
Serbia. Though it was never completed, this nineteenth century plan had never ceased to
animate Serbs.
They had periodically continued to oppress, even kill, neighbouring
populations throughout the rest of the nineteenth century and in the early twentieth
century. 90 The Serbian nationalist revival of the 1980s, in turn, promised to be another
revival of the plan to recover Greater Serbia and eliminate other Yugoslav populations in
the region. Serbs had, in all these centuries, never changed. Twentieth century Croatian
nationalists aptly constructed revised teleologies of the region’s history “to demonstrate
that, very much like a form of Biblical evil, Serbian expansionism and bloodlust was
timeless and had no sell-by date”. 91
If Croat nationalists played on the notion of the Serbian threat, they also played on
the twin theme of their own victimisation in this. 92 If Serbs threatened the region, they
threatened Croats first and foremost, as they had always done so in the past. Painted as
the eternal victims of Serbian ploys, Croats gained the higher moral ground. They were the
positive pole in this antagonistic couple. While the Serbs posed as the threatening other,
88
Xavier Bougarel, “Yugoslav Wars,” p. 159.
David Bruce MacDonald, Balkan Holocausts?, p. 117.
90
Branko Miletić, a Croatian historian, asserted, for example, that “Greater Serbianism has cost the lives of
some 600,000 Croatians, 400,000 Muslims, 100,000 Albanians, and countless others this century, not to
mention non-conformist Serbs, and even people not from the Balkans. It has ethnically cleansed some five
million inhabitants since 1900, wounded, maimed and imprisoned over two million, and caused hundreds of
billions of dollars worth of material damage.” Quoted in David Bruce MacDonald, Balkan Holocausts?, p. 210.
91
Ibid., p. 110.
92
Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown,” p. 773.
89
354
the oppressor, the Croats were the irreproachable victims. They were above Serbian
barbarianism, their bloodlust, and quest for plunder and land. They were innocent and
peace-loving and had been for time immemorial. To build this image, Croat nationalists
drew on Croatian mythology, particularly on the myth of Croatia as the Antemurale
Christianitatis.
The Croatian myth the Antemurale Christianitatis, of having for centuries sacrificed
themselves for the defence of the Easternmost rampart of the civilised, European,
Christian (but more importantly, for Croats, Catholic) realm, was made to fit the events of
the late twentieth century. Once more, a threat was rising from the East, or, in the words of
a Croatian geographer, “an attack by the last of the Barbarians coming from their darkness
to the lights of the Mediterranean, to Rome.” 93 Croats would have to defend themselves,
their life and land, against this force. As martyrs, moreover, they would be defending their
faith. Just as the myth of the Battle of Kosovo for Serbs promises to reward Serbian
martyrdom with divine election, so did the Croatian myth the Antemurale Christianitatis.
Martyrdom in the name of their faith renewed the covenant Croats had with God, the
Covenant they had won by renouncing aggression and sacrificing themselves for the
defence of the Christian world. The revival of the Antemurale Christianitatis infused a
spiritual element into the opposition to Serbian projects.
This spiritual element of the Croat-Serb antagonism of the late 1980s and early
1990s was exploited by Croat nationalists using the myth of Medjugorje. Medjugorje, a
small town in Herzegovina, drew media attention in 1981 following the start of alleged
apparitions by the Virgin Mary. Though the initial message of the apparition took on an
anti-communist connotation, over the years, it became increasingly a message focused on
prayer, peace and love for humanity. 94 Croats nationalists exploited the myth of
93
94
David Bruce MacDonald, Balkan Holocausts?, p. 119.
Michael Sells, “Crosses of Blood,” p. 319.
355
Medjugorje for their cause, a powerful tool in light of Croats’ predominantly Catholic
denomination. The Virgin’s message was fitting with Croat mythology. Much like the Virgin
Mary had appeared to preach peace, Croats had always been a peace-loving people.
Medjugorje signalled the righteousness of the Croatian cause. Croatian ideologues built on
the symbolism of the apparition to mark distinction between Croats and Serbs. Croats
were an enlightened, softer, ‘mother-centered’ culture, peaceful, nurturing, in contrast with
Serbian culture, ‘father-centered’, destructive and bellicose. 95 In all appearances, the
divine was siding with Croats. As MacDonald explained, the apparition was not fortuitous,
“the Croats received this revelation because of their unique qualities as an ancient and
peace-loving people, and because they were the innocent victims of Serbian aggression. 96
God had sent them a sign indicating that they remained the chosen ones. It was a divine
election that Croat nationalists extended to their cause. In a conference in 1993, Tudjman
declared that “[t]he Madonna’s appearance” announced “the re-awakening of the Croatian
nation” and indicated that the Virgin was in favour of the Croatian nationalist cause. 97
In a manner strikingly similar to Serbian rhetoric, Croatian nationalists therefore
created fundamentally antithetical, irreconcilable identities. Croats were an enlightened,
civilised and peace-loving nation, an image created through mythology and modern
mythmaking, but also through a denial of the darker sides of Croat history. If Croats were
such a peaceful people, the Ustaša could not have been the murderous fiends they had
been made out to be. Dangerous intentions belonged only to the Serbs, who in their
barbarianism threatened Croats, threatened their faith, and threatened Western
civilisation. The confrontation between Serbs and Croats, in light of the Serbian threat, was
an absolute one, and as a result, demanded solidarity.
95
David Bruce MacDonald, Balkan Holocausts?, p. 119.
Ibid., p. 121.
97
Quoted in Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown,” p. 772.
96
356
In terms of politicising the Croat nationalist rhetoric, Croat ideologues and
politicians employed a strategy very similar to the one used by their Serbian counterparts.
Croat nationalists politicised issues by building into their nationalist platform the theme of
violated rights. They claimed the existence of ancient, elemental Croatian political rights,
which had been unjustly infringed upon. The situation demanded to be redressed.
To establish Croatia’s rights, Croatian ideologues pointed to the alleged continued
existence of Croatian political institutions, of a continued Croatian sovereignty for nearly a
thousand years. Croats grounded this belief in the state right myth. Croatia had developed
political institutions in the seventh century, the Banus and the Sabor, which had survived
conquests and domination for centuries. It had adopted and upheld Western legislative
and political tradition throughout the period. Croatian authors argued they could claim they
had had “a continuous state, even during the Austro-Hungarian period [...] Even then,
Croatian historians argued that the country had preserved the characteristics of its
constitutional statehood.” 98 These institutions and, more importantly, the fact that Croatia
belonged to the Western realm, had allowed it to develop civilised, evolved, Western
inspired traditions. Croats were a civilised people, a belief echoed in popular views. A
Croatian woman, for example, when interviewed explained that “Croatia is one of the
oldest and most advanced nations in Europe.” 99
The myth of Croatian state rights was eventually exploited for a political purpose. In
the eighteenth and nineteenth centuries, Croat intellectuals built on the myth to argue for
their right to self-determination. If Croatian institutions and traditions had endured for so
long, Croatia was certainly capable of ruling itself and deserved to be freed from AustroHungarian control. Sardamov explained, however, that though Croatian elites tied the myth
to grand ideals, the movement had begun for more instrumental reasons. The revival of
98
99
David Bruce MacDonald, Balkan Holocausts?, p. 115.
Franke Wilmer, The Social Construction of Man, the State, and War, p. 32.
357
the myth was “promoted by the Croatian nobility in their struggle to preserve old privileges
against encroachments from Budapest and Vienna.” 100 Whatever the reasons for the
myth’s instrumentalisation in the eighteenth and nineteenth centuries, it passed into
Croatian political rights’ repertoire.
Twentieth century Croatian ideologues also came to exploit the state right myth for
their own nationalist agenda. They revised it thoroughly, however. According to them,
though Croatia had enjoyed a continued autonomy and had been allowed to keep its
institutions for over a millennium, the creation of the first Yugoslavia put an end to this.
Worse than Austro-Hungarian domination, a Croat historian argued, “[t]he ‘Yugoslav’
period of Croat history is definitely the darkest and most humiliating [...] for the first time in
13 centuries the institutions of the BANUS and SABOR were completely abolished.” 101
But if the first Yugoslavia was blamed for having destroyed Croatia’s thousand
years old institutions, the Second Yugoslavia, Tito’s Yugoslavia, had also done its share to
oppress Croats. Part of Tudjman’s project of revising accounts of the Second World War
centered on reveal communist crimes against Croatian populations. As Bowman recalled,
Tudjman’s party, the HDZ, “paraded pictures of bone piles, asserting these were not the
skeletons of ‘Nazis’ or ‘quislings’ but of ‘Croatian victims’ of communist brutality.” 102 The
years of Communist rule under Tito had continued the oppression of Croats. According to
Croat ideologues, Yugoslavia had been created to rob Croats of their statehood, to stop
them from “enjoying their nationhood”. 103 To add insult to injury, integration into Yugoslavia
signified that Croats, bearers of ancient traditions of democracy and right, were to be
submitted to an illiberal government, infringing on basic human and collective rights. The
Yugoslav state reserved a particularly unfair attitude towards Croats. A number of Croat
ideologues pointed to the events of 1971-1972 as proof. The Maspok, the Croatian spring,
100
Ivelin Sardamov, “Identity’s Role in the Serbo-Croatian Conflict,” p. 464.
David Bruce MacDonald, Balkan Holocausts?, p. 115.
102
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 333.
103
Ibid.
101
358
during which demands for more autonomy on the part of Croat intellectuals had led to
repressive and massive purges on the part of Tito’s government indicated that equity,
human rights and democracy could never flourish in Yugoslavia. 104 As a result, not only
had Yugoslavia robbed Croats of their statehood and imposed a government at odds with
their political traditions, but it also seemed clearly biased against Croatians, intent on
“”punish[ing] the Croatian people for their previous attempt to realise themselves as a
nation.” 105
In Croatian history, Yugoslavia represented a loss of Croatia’s revered institutions,
the Banus and Sabor, and its continued sovereignty, a violation of principles and values at
the core of Croatian identity, democracy and rights. Yugoslavia represented a fall from a
golden age to a degraded present. But as Levinger and Lytle explained, in nationalist
rhetoric diagnosing losses and violations, diagnosing a degraded present, is often a
strategy that emphasises the need for action. 106 Such was the case with Croatian
nationalist rhetoric. The situation could not be tolerated. Its acceptance constituted a
further betrayal of what Croatia had stood for. Croatian elites, however, knew how to
recover Croatia’s glorious past and to reclaim Croatian rights. Croatians, on the one hand,
needed to distance themselves from Yugoslavia. They needed to recover their
sovereignty, even their independence. Increasingly, particularly starting late in 1990,
strategies for freeing Croatia from Yugoslavia were discussed. Read through the lenses of
the state right myth, “Tudjman and his colleagues [promised] to resurrect the Croatian
nation and bring about its independence.” 107
A central part of the ideologues’ platform also dealt with territorial claims. On the
one hand, Croatians needed to counter Serbian claims to the Krajina and Eastern
Slavonia. It mattered to press the continued Croatian control of these territories so as to
104
Sabrina P. Ramet, Thinking about Yugoslavia, p. 152.
Glenn Bowman, “Constitutive Violence and the Nationalist Imaginary,” p. 333.
106
Matthew Levinger and Paula Franklin Lytle, “Myth and Mobilisation,” p. 178.
107
David Bruce MacDonald, Balkan Holocausts?, p. 116.
105
359
argue “against Serbian claims that [Croatia’s present borders] represent mere Titoist
administrative demarcations.” 108 These territories were rightfully Croatian as attested by
Croatian historical accounts. Ironically, if the state right myth's revival in the late twentieth
century was intended, in part, to defuse Serbian territorial claims in Croatia, another
Croatian myth, the Antemurale Christianitatis, served to ground Croatian territorial claims
in Bosnia-Herzegovina. Croatians, having sacrificed themselves in the defence of the
Western realm, had lost many lives and territory. Croatian territory had been amputated
and left with rump borders. This had given it an unnatural, untenable shape, a “shape of
resistance.” 109 Croatia was, according to Tudjman, as “an apple with a bite taken out of
it.” 110 With Croatian sacrifices behind Croatian borders and the alleged unsustainability of
the territory, Croatia was entitled to new territory. The argument was even developed in a
textbook reputed to have been used in the Faculty of Political Science of the University of
Zagreb. The textbook asked:
Does Croatia have the right to assume such a geographical shape as will enable
her survival? Of course she does. For Croatia with cancer in her womb–or even
without a womb, for the Serbs and the Muslims have taken it–cannot exist longterm, never mind forever. Therefore she has a right to Bosnia-Herzegovina. 111
The Croatian nationalist platform therefore developed a plan to recover territories to which
Croatia was entitled historically because of the suffering endured, and even to prevent
further suffering, with the announced disintegration of Croatia because of its borders
unsustainability.
Croatian nationalists politicised their rhetoric by building on a language of injustice
and oppression, identifying Yugoslavia as the source of Croatia’s fall. The Yugoslav
episode had been a mistake for Croats, which had cost them their sovereignty and historic
state. To uphold Croats historic rights and entitlements, the situation needed to be
108
Ivelin Sardamov, “Identity’s Role in the Serbo-Croatian Conflict,” p. 464.
David Bruce MacDonald, Balkan Holocausts?, p. 118.
110
Ibid., p. 100.
111
Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown,” p. 773.
109
360
redressed. Standing strong behind their leaders, Croats could, as Tudjman claimed,
ensure “not to repeat that episode ever again!” 112
To ensure the impact of their rhetoric, Croatian nationalists deployed the same
simplification tactics Serb nationalists had used. They ensured the predominance of their
nationalist platform by limiting alternative, particularly moderate, rhetoric. They also
proposed simplified concepts, images, and more importantly identities, to facilitate
people’s identification with them.
In the 1990 elections in Croatia, Tudjman’s nationalist party, the HDZ, was elected
with 41.5 percent of votes. Thanks to the winner-take-all system instituted prior to the
elections by Croatian Communists, the HDZ won 58 percent of seats in the legislature. As
in other republics, the HDZ’s victory was not sweeping, but gave the party control of
Croatia’s political institutions. The HDZ quickly proceeded to consolidate its hold on power.
A central strategy to achieve this was the removal of opponents, particularly moderates,
who contested the HDZ’s nationalist stance. Croat nationalists silenced a number of their
detractors through threats. A Serbian party leader, in Croatia, who promoted open
dialogue between Serbs and Croats in the republic and the pacification of relations
between the two groups was, according to Oberschall, “vilified and threatened by both
Croat and Serb nationalists.” 113 The houses of moderate politicians were bombed. A
moderate police chief in Slavonia, who had been in charge of quelling tensions between
Serbs and Croats in the region, was even gunned down by an official of the HDZ. 114 These
various incidents created a sense of fear among moderates and intimidated their
supporters. People were tempted to dissociate themselves from the moderates for fear of
reprisals. When an independent moderate in the Zagreb legislature became the target of
112
David Bruce MacDonald, Balkan Holocausts?, p. 118.
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 995.
114
Ibid., p. 996.
113
361
the nationalists, her former students began distancing themselves because, as one
explained, “[i]t was dangerous to be seen talking to [her] in public.” 115
Taking control of the media also helped to subjugate alternative voices from the
public sphere. According to Renaud de La Brosse, the take over of state media had been
a well planned move on the part of political parties in Yugoslav republics. 116 With state
media controlled at the republican level, the winners of the elections 1990 would be given
free reign over their republic’s media. Though the HDZ had campaigned during the
elections on a platform that promised complete freedom of expression and free press in a
democratic Croatia, once in power it quickly broke its electoral promise. 117 Tudjman, as
Milošević had done in previous years in Serbia, quickly moved to take control of important
media outlets in Croatia. He ensured himself favourable media coverage, and with growing
tensions in Krajina in June 1991, he further tightened his control over the media. With the
explosion of hostilities between Serbs and Croats in the Krajina, Tudjman played the
patriotic card. He seized the opportunity, in the face of a Serb attack on the nation, to
eliminate journalists and media representatives critical of his party and of how they
handled the conflict in the Krajina by accusing them of being traitors to the nation. 118 Over
the following months, Tudjman and his supporters replaced directors and editors at the
state television, the RTZ, as well as intimidated and fired hundreds of journalists at the
RTZ and in other important media outlets. The nationalists finally succeeded in instituting a
partial censorship regime in Croatia in November 1991. Media output was to be surveyed
by a new committee composed of media experts and the Minister of Information. Instituted
through a presidential decree, “the committee could name the directors and editors of state
run media, censor information given to foreign journalists, and even ban media publishing
115
Ibid.
Renaud de La Brosse, “Les médias machines de guerre en ex-Yougoslavie,” p. 115.
117
Ibid., p. 117.
118
Ibid.
116
362
information that ‘harmed Croatian defence’ or risked revealing military secrets.” 119 The
elimination of moderates and the take over of the more prominent media in Croatia
ensured the HDZ and Croatian nationalists a near total control of the public realm. Croat
nationalists could disseminate their propaganda, with little information that risked
contradicting them. The nationalist platform dominated.
The control, and even suppression, of alternative voices in Croatia turned the
nationalist rhetoric into the domineering discourse. It is its formulation, however, that
worked towards its internalisation. Croatian nationalism centered on the promotion of
simple themes, simple versions of identities, even Manichean ones. These simple ideas
and images, many drawn from Croatian mythology and historical memories, made
engaging with them easier. There was little complexity to wrestle with; Croat ideologues
had worked out a platform in which clear role were attributed to all without exception and in
which issues were clarified. All Croats faced an unrelenting Serbian threat to their lives, to
their rights, to their traditions. These were in turn repeated throughout the days in political
rhetoric and in the media. Repetition, “the single most effective technique of persuasion,”
helped pass on these simple teleologies to Croat audiences, given them an overbearing
semblance of truth. 120
The wars in the former Yugoslavia began first and foremost as a war of words
between elites in the various republics.
Years before armed hostilities, each side
campaigned against those they identified as opponents, moderates elites, but principally
other nations and nationalities and their leaders. In this pursuit, they recruited the help of
strong allies, such as the media and their respective religious establishment. Together,
they formed powerful nationalists propaganda machines aimed at winning the support of
ethnic constituencies. For this, they weaved a narrative around a confrontation between
119
120
Personal translation. Ibid., p. 118.
Anthony Oberschall, “The Manipulation of Ethnicity,” p. 993.
363
antithetical, irreconcilable opponents. In their reading of the situation, it became about ‘us
versus them’, about the Serbs against the Albanians, Slovenes and Croats, about the
Croats against the Serbs, about Slovenes against Serbs, and the Bosniaks against the
Serbs and Croats. To strike a responsive, an emotive chord, ideologues tied political
discourses to the past and to each group’s previous losses at the hand of others. On all
sides, leaders put forth ideas of injustice perpetrated by others to encourage nationalist
sentiments. To facilitate the internationalisation of their rhetoric, elites ensured its
predominance, by eliminating alternative views, but also by creating simple notions and
images to which populations could relates. What had always been an extremely complex
identity environment witnessed a collapse of identities around caricatures of what they had
been. Equipped with this potent persuasive tool, elite stoked up tensions within their
societies, creating rifts between Yugoslav populations that would eventually explode into
open conflict.
364
CONCLUSION
Implications of the Research
The fundamental point this dissertation sought to make is that there is not just extant,
physical violence to conflict. Conflict is not only violence. Violence is an endpoint, the
explosion of hostilities. Before violence occurs, there is generally a slow–sometimes not so
slow–degenerative spiral towards hostilities. To understand conflict, it is necessary to
understand where it comes from, its background conditions, political, economic,
environmental, etc. But this degeneration also implies another form of intermediary step:
human agency translating background conditions into violence. The break out of hostilities
comes as a decision made by individuals, judging the situation acute enough, important
enough, to merit a violent reaction. This analysis focused on this decision, the choice to
turn to violence in instances of conflict directed from the top down. It focused on elite-led
mobilisation, or more specifically, it focused on elite-led felt mobilisation, the juncture at
which people are swayed, convinced by their leaders of the need for collective action,
including violent collective action. In other words, violence starts in the minds of people
before it takes a physical shape. Felt mobilisation towards violence is thus an intermediary
step before extant violence. It is just as part of understanding conflict as understanding
hostilities is. It is critical to understand psychological and social dynamics leading to
violence to understand conflict in its globality, its degeneration phase and extant violence.
Moving beyond the limited focus of violence has implications. It means looking at
broader sets of relations than those uniquely behind hostilities, and, implicitly then, at a
broader group of actors. To understand conflict, it is a mistake to focus only on the visible
actors of violence. Extant violence may be the feat of specific actors, such as combatants,
365
soldiers or thugs or marauding armed bands. 1 Who lies behind the broader spiral of
conflict? Who encouraged it? Who supported it? It is important to know who was directly
involved in the violence, but it should not come at the cost of overshadowing the role
played by other actors.
As this dissertation demonstrated with its focus on elite-led mobilisation, elites often
play a role in conflict. Elites, political authorities, are often in a position to decide of the
outcome of a crisis. John Mueller, for example, alluded to the common role played by
political authorities, whether recruiting thugs and armed bands to wreak havoc “under their
guidance” or letting them rein free. 2 The simple fact that political leaders are in a position
to hire groups to do their violent deeds or are the ones to sanction the free rein of these
marauders and armed bands proves they are important actors with vested interests in the
conflict. To understand conflict, just as it matters to know the perpetrators of the violence,
it matters to know who gave the orders. Who are these elites, political leaders and ethnocentric entrepreneurs? Why do they recruit for violence or allow the violence to explode?
In other words, what is their interest in violence? Understanding elites, leaders and
entrepreneurs who create the space for violence is as important as knowing who fills this
space.
Another actor often ignored is the population. Populations disappear in analyses of
conflict. When they are discussed, it is frequently in their condition of helpless victims or
fleeing refugees, exporting the negative consequences of war to neighbouring countries. If
populations are so peripheral, though, why do leaders deploy efforts to appeal to them,
particularly in times of crisis or conflict? Even powerful authoritarian regimes feel the need
to call on their people, and justify their actions and hold on the country through massive
1
For a piece arguing for the centrality of thugs and marauders in conflict, see John Mueller, “The Banality of
‘Ethnic War’,” International Security, vol. 25, no. 1, 2000, pp. 42-70.
2
John Mueller, “The Banality of ‘Ethnic War’,” p. 42.
366
propaganda machines. Political entrepreneurs still rely on propaganda to influence their
public.
It is a mistake to dismiss the role of populations in conflicts. They are a force to be
unleashed. Populations can be, just as combatants, thugs, or leaders are, the fuel of
conflict. If directly involved, they are implicit combatants who feed the violence through
combat or sporadic, insurgent or terrorist acts. Or they are the workers that produce the
food, fuel, and machines that make warfare possible. But populations can also be the
source of fundamental immaterial resources. Whether actively in support of their leaders or
passively condoning their actions, publics can be as dangerous as the few (or many) who
take up arms. Popular support is a source of influence and power. In both democratic and
non-democratic systems, social representation, popular backing, is power. It is the power
of the ‘Big Men’, with the many behind them. As Anthony Oberschall explained, “[a]n
armed, organized 10 per cent who control mass communications can have its way when
the majority supports it overtly or tacitly or is confused, and when the opposition is
unorganized, divided and scared.” 3
But populations are not simply the passive victims of elite ploys or fearful followers
either. That elites deploy recruitment strategies and propaganda is indicative of the fact
populations are constituted of individuals with the capacity to adjudicate. Belief in what
leaders say is not automatic. People have the capacity to weigh their options. They may
disbelieve what they are told by leaders. They can choose to internalise or to dismiss what
leaders are saying.
And this is the case whether it transpires in people’s outward
expressions or behaviours, or whether–in authoritarian conditions for example–it does not.
This is an important point to make, one on which this framework is predicated. Whether
talking about electoral campaigns in democratic countries or states of crisis requiring
3
Anthony Oberschall, “The Manipulation of Ethnicity: From Ethnic Cooperation to Violence and War in
Yugoslavia,” Ethnic and Racial Studies, vol. 23, no. 6, 2000, p. 986.
367
mobilisation, constituencies and communities are not infinitely gullible. Support must be
won; people must be swayed. For this reason, then, communication with them, appeals to
gain their support, must be thought through and strategised. Whether this is done aptly,
how appeals are constructed, can determine the success of the endeavour, whether elites
successfully convince populations and therefore win supporters.
Looking at a broader set of relations and actors, understanding conflict in a broader
sense, points to the complexity of conflict sequences. Collective violence is not
instantaneous. Violence is the result of intricate preceding steps. It is in these preceding
steps that occurs the breakdown of mechanisms in society keeping violence in check.
Space for violence is created at the confluence of varied background conditions, effects,
actors and their actions and interactions. Focusing solely on the phase of open hostilities
in conflict disregards the fact that violence is the endpoint of a sequence of degenerative
tension.
Because collective violence, fighting and hostilities, are born of this intricate
degenerative sequence, they are not as inevitable as may seem. To be sure, policing and
military forces can control armed combats. Addressing processes and factors underlying a
conflict are not an easy feat, however, as recent international interventions have
illustrated. More telling, is the fact that even instigators of a conflict can lose control of
what they started. Conflict sequences are such complex phenomena that often times elites
and groups who triggered tensions for their own interests create Frankenstein monsters.
They create tensions of which they lose control and which end up controlling them. In
Rwanda, Kayibanda and leaders of the first generation Parmehutu became bound to their
ethnocentric rhetoric, forced to pursue their Tutsi hatred campaign, recurrently employing
violence, at the cost of being perceived as weakened, or worse traitors to the cause. In the
1990s, a similar situation occurred, once chosen, the path of ethnocentrism constrained
Rwandese elites’ actions and discourses, imposing on them to adopt evermore radical
368
ways. In Yugoslavia, Miloš Vasić indicated, for example, that Milošević “himself was
surprised by the fact that the war of ethnic extermination gained such a momentum as to
make it a self-supporting suicidal machine.” 4 As a result, he found himself like “someone
who has jumped on the tiger of nationalism and is finding it difficult to get off again without
the tiger eating him.” 5
Collective violence and conflict are certainly not inevitable. But their complexity,
and their sources, makes them difficult to resolve, particularly to resolve them in depth at
their roots. Nebojša Popov offered another take on the debate of the inevitability of war.
Referring to the Yugoslav wars of the early 90s, but just as applicable to the Rwandese
genocide, Popov explained that “[t]he war was not unavoidable. A real choice existed: a
democratic process of change, or the violent destruction of society and the state.” 6
Although somewhat simplified–stakes are certainly not this clear at the start of crises–
Popov’s comment raises the point that violence can be the result of choices made by
actors and communities party to the conflict. Addressing the conflict, thus, entails undoing
many of these choices, undoing many of the factors and human decisions that constructed
the degenerative sequence towards conflict, a sequence that at every step risks being
reinforced, a sequence that often takes a lock-in, increasing returns, path dependent, form.
Extant violence implies that all preceding steps coalesced, interlocked to become this
violence. Resolving not just the violence but the conflict itself means that preceding steps,
preceding actions and decisions that lie behind the violence, be addressed to in order to
defuse the conflict in its globality. It means addressing background conditions, a point
often made, but also instrumental reasons for wishing conflict, invented or rekindled
4
Miloš Vasić, quoted in Sabrina P. Ramet, “A Theory about the Causes of the Yugoslav Meltdown: The
Serbian National Awakening as a ‘Revitalization Movement’,” Nationalities Paper, vol. 32, no. 4, 2004, p. 771.
5
David Owen, quoted in Dejan Jovic, “The Disintegration of Yugoslavia: A Critical Review of Explanatory
Approaches,” European Journal of Social Theory, vol. 4, no. 1, 2001, p. 106.
6
Nebojša Popov, “Traumatology of the Post-Party State”, in Nebojša Popov, ed. The Road to War in Serbia:
Trauma and Catharsis (Budapest: Central University Press, 2000).
369
antagonistic ethnocentrism, love for one’s group, hate for the other(s), etc. It means
addressing psychological and social dynamics of conflict.
The originality of this research is to raise the importance of factors and dynamics
often assumed or ignored in crises or conflicts, factors and dynamics somewhere between
background conditions and military/strategic variables. It delves into the psychological and
social dynamics of elite-led mobilisation to highlight that propitious background conditions
do not automatically lead to military/strategic decisions. When confronted with a crisis, all
strands of society are called upon to react to it. Populations and elites are called upon to
make choices, to decide what should be done. Different options exist when reacting to
background conditions. Conflict, collective violence, war, is a choice. It is one of the
options that can open up during a crisis. And it is an option that can be appealing under
certain circumstances.
It is fundamental that the reasons for this choice be understood, for why elites
choose war, and for why populations, in elite-led conflicts, choose to follow them. This is
what this research addressed. The originality of this research, then, is that through its
focus on social and psychological dynamics of elite-led mobilisation, it proposed an
interactionist approach, integrating both the psychological mechanisms of attachment to
the group and the interactions between leaders and potential followers, an approach in
which mobilisation comes at the point where elite strategies align with a people, a group’s,
sentiments, beliefs and choices.
Identifying Gaps
In its essence, then, this research addressed the interaction between elites and
populations. By doing so, however, it avoided developing each of these two concepts–
elites and populations–on its own. For the sake of simplicity and clarity, this analysis
treated publics and elites as distinct, detached categories of agents, one the potential
370
follower in need of convincing, the other the calculating head. But the population is a
diverse group, as are elites. Each concept would have been better served if further
developed individually. Each actor should have been defined better. The research equally
glossed over existing ties between elites and populations, or more specifically elites’
embeddedness in the population.
Populations are not one-dimensional. Groups, communities, do not suffer from the
‘unity of the crowd’ syndrome Le Bon claimed existed, though they often develop strong
shared norms for behaviour.
To start, there always is dissonance in how people
understand what they are presented with.
People negotiate meaning differently,
regardless of how simply and generically issues are cast. This necessarily means that
people also interpret differently. Presented with the same object or fact, for example,
people can develop a different sense of what they are or mean. On the social level, people
often need to negotiate with each other to come to an agreement on the meaning of issues
they are presented with. This negotiation process, whether at the individual or social level,
is not accounted for enough in this framework and in the analysis. People also engage
with issues and calls differently. Individual beliefs illustrate this point. Belief is not an
either/or process. People believe or disbelieve in varying degrees. In turn, these varying
degrees of belief can result in varying behaviour. Finally, people respond differently to
crises and conflicts. As discussed, individuals react to crises and conflicts through forms of
frustration, fear or anger, though the specific nature of their responses is left mostly
unexamined. Instead, the analysis concentrated on how these emotive stances were
hijacked by political entrepreneurs to drive collective action, invariant of the form popular
sentiments took. Further investigation of the dynamics of these very different emotive
states is warranted.
A similar criticism can be levelled at the concept of ‘elites’. Considering the
importance of elites in this research, the concept is underdeveloped. The theoretical
371
framework stressed the sources of elite behaviour. Political and power games were
described as imposing a type of competition that demands that political entrepreneurs
behave instrumentally. Yet describing their behaviour does not define elites. Nor does it
answer important questions: Where do elites come from? Who are they? How did they
become elites? Or, more importantly, how do they get to be perceived as elites?
These questions highlight a principal weakness of the analysis. Elites or political
entrepreneurs do not float above a population. Typically, they emerge from the population.
This raises a series of related difficulties: What are elites’ relationship to identity, cultural
and social backgrounds? As members of the population, they share its background and
are constrained by it. When building appeals, they refer to this background because it is
their ontology as well. This point was acknowledged, but not developed. If elites have
these backgrounds as the subtext of who they are and the world around them, how much
do they believe in their calls for mobilisation, and how much of it is merely a tool to
convince potential supporters? Where does conviction end and instrumentalism begin?
Similarly, with elites as members of the population, how detached from this population
must they be to manipulate them? What is involved in this qualitative change from member
of the population to elite? More critically, is there such a qualitative change in actuality?
These issues are not given their due in this research.
Elites, political and ethno-centric entrepreneurs, are too often treated as an
undefined, generic category. In reality, however, they are a varied category. Moreover,
different motives animate them. They can range from calculating demagogues to
benevolent visionaries, with numerous other categories in between. Arguably, these
varying motives may be the source of different behaviour, different strategies. It might also
make a difference in terms of what they demand from their supporters. More importantly,
the possibility that elites, amongst themselves as a group, are driven by different motives
is an issue in and of itself. How do they negotiate to reach an agreement on actions to
372
take? Interactions among elites are unduly addressed, despite contentious politics’ call for
dynamic frameworks addressing framing battles within and between groups of challengers.
These gaps call for greater nuance. Yet nuance often comes at the cost of
parsimony. Choices were made in terms of issues addressed to ensure the clarity of the
argument. And, while these are issues which should be looked at in future research, they
do not take away from the sequence of mobilisation developed over the course of this
research and from the three solidarity mechanisms identified. Indeed, as pertains to the
latter, while motives change, means to achieve a goal can be the same. Opposing,
politicising and simplifying are not hypothesised as absolute, necessary strategies. They
were developed as probable ones to attain group solidarity, which is a fundamental part of
mobilisation no matter what the motive is.
Directions for Future Research: Grounds for Generalisability
Rwanda and Yugoslavia are certainly more different than alike. The most obvious
difference between the cases is that they are on different continents and have experienced
very different historical trajectories. Recent Rwandese history centered on increased
efforts at state-building, the arrival of European colonisers, decolonisation and the
construction of a modern polity. The former Yugoslavia was also, to a certain extent,
submitted to external domination at the hands of the Habsburg and Ottoman Empires, but
then freed themselves to eventually become a socialist nation caught between the Eastern
and Western blocs during the Cold War. Their demographic situations were also different.
Rwandese groups lived intermixed throughout the country, though admittedly the
proportion of Tutsi is smaller in certain regions. Tutsi and Hutu share a common language,
a common religion, even common myths. In the former Yugoslavia, the situation was more
complicated. Groups populated certain areas in the country, which allowed them to form a
majority in certain republics, but to be a minority in others. The exception to this was
373
Bosnia-Herzegovina, where groups lived more intermixed that in the rest of the country.
Nations in the former Yugoslavia spoke different languages. 7 Yugoslav populations were
also divided by their faiths. Finally, all of them had developed their own set of myths and
different cultural references.
Another fundamental difference between Rwanda and the former Yugoslavia is the
impact authoritarian regimes on society. In Rwanda, the Kayibanda and Habyarimana
regimes were both in large part responsible for the continued dissemination of an ethnocentric ideology within Rwandese society. Both the Kayibanda and Habyarimana regimes
kept differences between Tutsi and Hutu alive by making them the central pillar of their
regime’s ideology, whether openly or tacitly through special discriminatory tactics. In the
former Yugoslavia, on the contrary, Marshall Tito’s regime fought to defuse nationalism, to
eliminate it as a force and source of contention between Yugoslav populations. He instead
championed a broader identity, Yugoslavism, which he believed would supersede
nationalism. Socialist Yugoslav ideology was therefore meant to replace nationalist
attachment. In both Rwanda and the former Yugoslavia, populations were submitted to
these regime ideologies for decades.
In both cases, however, despite their differences, similar strategies played out.
Both countries, in the 1980s, faced a similar ordeal. Their authoritarian regime’s hold on
the country began unravelling as a result of an unfolding crisis. More than a simple crisis,
problems Rwanda and the former Yugoslavia faced in the 1980s were multifaceted. An
economic crisis combined with their regime’s inability to handle problems adequately also
led to a crisis of legitimacy. Faced with growing internal discontent and under international
pressure, both countries were forced to liberalise their society and their political realm. In
both cases, with looming multi-party elections, competition between elites, old and new,
7
Some would argue that Serbo-Croatian is a language shared by Serbs and Croats, but written with different
alphabets. In the context of the crisis in the 1980s, both Serbs and Croats nationalists argued the contrary, that
is, that it is in fact two languages and strove to purify ‘their’ language of the ‘others’ influence by purging words
associated with the other group.
374
intensified. In Rwanda, as in the former Yugoslavia, a group of old and new elites chose
the path of ethnocentrism to win their constituents hearts and minds, the means to keep or
attain power.
In Rwanda, the three solidarity mechanisms developed by the research, opposition,
politicisation and simplification, were elements in the successful frames for support
elaborated by both the Kayibanda and the Habyarimana regimes. Faced with a legitimacy
crisis, these two regimes agitated anti-Tutsi sentiments and Hutu solidarity as a strategy to
(re)gain popular support. In each case, they presented Manichean depictions of Hutu and
Tutsi to foster antagonistic perceptions and relationship between these groups. They also
structured their discourses to politicise Hutu and Tutsi identities, focusing on issues of
rights and power. Finally, they simplified the identity environment by providing generic
understandings of communal groups and their relations.
In Yugoslavia, very similar dynamics could be observed. In the midst of a political
and economic crisis, old and new elites found themselves in an intense competition for
positions, standing and power. A large proportion of them chose to turn to ethnocentrism in
the hopes of winning popular support. In their appeals to their constituency, they too
reverted to opposition, politicisation and simplification strategies. Both Serbian and
Croatian leaderships, for example, adopting mirror approaches, portrayed group relations
as antagonistic. Ascribing threatening intentions to the others pressed the need for group
solidarity. Serbian and Croatian rhetoric also politicised the situation by focusing on
injustices committed against and entitlements of their group. Finally, they simplified
readings of the situation, and as a result simplified people’s choices, by ensuring their
viewpoint predominated and was simple to engage with.
Despite their differences, following a crisis, Rwanda and the former Yugoslavia
followed a similar trajectory. In both cases, political elites reacted to the crisis and parallel
liberalisation by appealing to their group. But, more importantly, in both cases, they
375
resorted to opposition, politicisation and simplification as means to rekindle group
solidarity. This thus establishes the plausibility of generalising identified mechanism.
Across two different cases, similar patterns were identified. There is thus reason to believe
that the mechanisms developed by this research could recur in other cases, that they
broad enough and generic enough to be applicable to other cases.
Follow-up research should concentrate on assessing the generalisability of the
framework. In order to assess the robustness of ideas and mechanisms described by this
research, other cases of elite-led mobilisation should be studied. Cases of successful
mobilisation should be analysed in depth in order to evaluate how elite strategies unfolded
and impacted on popular participation in the conflict. This would constitute an opportunity
to further trace processes involved in opposition, politicisation and simplification, as well as
their interactions.
Cases selected should not only be those where elites have been successful in
manipulating identity, however. An equally important step in order to better assess the
validity of this framework is to look at counterfactual cases. These include cases where the
solidarity fostering mechanisms described in this research were employed by elites, but
did not lead to a change in the state of mind of constituents, did not lead to ‘felt’ or extant
mobilisation. In a similar vein, the study of counterfactual cases should also involve cases
where none of the mechanisms described were employed, but where popular mobilisation
occurred nonetheless. Both types of counterfactual cases are an opportunity to delve
deeper into the complexity of communicating mobilisation. In the first instance, such cases
offers the possibility to study factors countering strategies to foster solidarity. The second
type of cases, on the other hand, are an occasion to get a better sense of alternatives
tactics behind elite-led mobilisation.
376
APPENDIX
Interview Questionnaires
ENGLISH
My name is Marie-Eve Desrosiers. I am a Ph.D. student at the University of Toronto, in
Canada, studying under the supervision of Professor Thomas Homer-Dixon. These are
my coordinates both in Rwanda while I am here, until the end of April, and in Canada,
should you wish to contact me. I have also included the contact information for my thesis
supervisor, Professor Homer-Dixon.
Let me begin by describing my research a bit. I am interested in studying the issue of how
people perceive themselves, their group and other groups and how that relates to
situations of conflict. I would like to know how feelings of belonging to a group and how
people see other groups affect how they interpret events and act during hard political
times. I want to study this in particular during the Rwandan civil war up to the events of
April 1994. During the war, many things were said about the different groups (Tutsi and
Hutu) in Rwanda. I am curious to know if that played a role in the conflict. I want to know
if what people heard about groups and group relations, from within their group or from
others, changed the way they saw people in their community from other groups and
changed how they interacted with them. I would really like to understand if what people
heard in Rwanda about groups (Hutu and Tutsi) had an impact on what they thought of the
situation, of the people involved, and what should be done. I am looking at this from the
perspective of groups and communities, and I do not need to know what group you belong
to, so I will not ask you if you are a Hutu or a Tutsi.
As a Rwandan, you have first hand experience of this, and that is why I would like to talk to
you: to know your perspective on these issues. I know these are complicated questions
and they address sensitive issues, and I do not underestimate the difficulty of answering
them. They might make you think of things you lived that were hard and hurtful. But I will
not ask about them. I would just like to know your perspective, but you can choose at any
time to stop this interview or not to answer some of my questions if you feel uncomfortable
or it is too difficult for you to talk about certain things. I value everyone’s perspective on
these issues, and for this reason, I hope you will understand that I need to ask everyone
the same questions. Also, nobody except for me will know what you have told me. I will be
the only one reading my notes or listening to the recordings. If you agree to be audiotaped, I will put a code on the audiotape and not your name, and I will be the only one to
have access to information to understand this code. My notes and tapes will be stored
securely and locked away and when I will finish my research, I will destroy everything. So,
now that you know a bit more about my research, I would like to invite you to participate in
it by answering some questions. This should take approximately an hour. Do you agree to
take part in my study? Do you mind if I audio-tape our conversation? It will help me
remember your answers. Are there any questions you would like to ask or comments you
would like to make before we begin?
377
BEFORE
a. 1 What was life like in Rwanda before the war? I know the two main groups in Rwanda
are the Hutus and Tutsis. I am not asking you what group you belong to, but do you have
a sense of what life was like in Rwanda before the war for Hutus? For Tutsis? What was
your sense of interactions between these groups (Tutsi and Hutu)? What were group
interactions like in public places i.e. church, work, etc. between groups? In personal
relationships, social gatherings, friendship, family across groups?
a.2 Did people of different groups get along well before the war? Were you aware of any
problems between groups before the war?
b.1 Do you recall any of the general perceptions found in Rwandan society about groups
and group relations before the war? If so, can you talk more about them?
b.2 People often have general perceptions about the groups and communities in their
society. They get these perceptions from different people, places and experiences they
have. How did you learn about the groups in Rwanda, their history and how the different
groups should interact? Did you learn about the groups in Rwanda and group interactions
from your family, friends or neighbours, society in general? If so what? In school? If so,
what?
b.3 What else did you learn in general about the history of Rwanda or of the different
groups in Rwanda in your community or school?
b.4 People have different identities and attach different feelings and ideas to them, as well
as a different importance. For example, I am Canadian, but I am also a Quebecoise,
someone from a French region of Canada. Canadians, and the ones that are Quebecois,
associate special meanings to these identities, for example in terms of history or how they
see people from their group or other groups and interact with them. I am not asking you
what group you belong to in particular. But do you think for people in Rwanda identities
like Rwandan or Tutsi and Hutu had particular meanings? Could you give me a sense of
what people generally thought they meant? What did you think of these identities and of
what you have learned about them?
CHANGE IN PERCEPTIONS OF COLLECTIVE IDENTITY
a.1 Did you feel things changed in terms of how ordinary people related to each other
between Hutu and Tutsi groups in Rwanda when the war started in 1990?
a.2 If so, when do you think relations between groups started to change?
b.1 How did things change in relations between people of the different groups? Were
interactions different in public places, i.e. church, work, etc.? In interpersonal relations, i.e.
neighbours, friends, family? How did things change in general?
b.2 Were you aware of changes in how people felt about their group (Hutu or Tutsi) or in
terms of what it meant to be part of one or the other. Did it mean a difference in how
378
people perceived the other group and their impressions of the other group? If so, in what
way? What did people start feeling about the other group?
c.1 If you believe things changed in how people perceived the other group, why do you
think it happened?
MAIN HYPOTHESIS: POLITICAL RHETORIC AND MANIPULATION OF PERCEPTIONS
a.1 Do you think certain people or groups played a particular role in bringing about this
change in terms of how people perceived each other across groups? Which ones? The
government? Elites? The media?
a.2 What was their role in bringing about theses changes in perceptions between groups?
What did they do to make these perceptions change?
a.3 Did these people promote certain ideas or perceptions about the groups and of
relations between groups? If so, which ones? What were they saying about the different
groups and group relations?
b.1 Do you think ordinary people believed what those different actors were saying at the
time?
b.2 Many reasons have been given to explain what happened in Rwanda. Do you think
this was one of the reasons behind what happened in Rwanda between 1990 and 1994?
Do you think that believing in what was said around that time, about the different groups
and their relations, is a reason that people acted the way they did and that the events of
1990-1994 took place?
b.3 Or were there other reasons for why people acted and events took place? If so, which
ones?
FOLLOW UP QUESTIONS ON PRESENT SITUATION
a.1 What is the situation like in Rwanda now in terms of interactions between groups now
that its has been ten years? How do groups perceive each other after these ten years?
a.2 How do groups perceive each other and interact with each other now that there is the
process of national reconciliation?
b.1 How is the process of national reconciliation going? Do you find it has been helpful in
reconstructing society? In reconstructing relations between the different groups? Has it
helped change the perceptions that existed between groups at the time of the events of
1990-1994?
FINAL COMMENTS
a.1 Is there any thing you would like to add?
379
a.2 Do you have any questions or comments about my research or the questions I have
asked you?
380
FRENCH
Mon nom est Marie-Eve Desrosiers. Je suis étudiante au doctorat à l’Université de
Toronto, au Canada, en science politique. J’étudie avec le Professeur Thomas HomerDixon. Voici mes coordonnées au Rwanda pendant que je suis ici, jusqu’à la fin du mois
d’avril, et mes coordonnées au Canada, si vous désirez me contacter. J’ai aussi inclus les
informations de contact de mon superviseur, le Professeur Homer-Dixon.
Laissez moi commencer par vous décrire un peu mes recherches. Je m’intéresse à la
façon dont les gens se perçoivent, voient leur groupe et voient les autres groupes, et à
comment cela a un rôle dans les situations de conflit. Je voudrais savoir comment les
sentiments de faire partie d’un groupe et comment on perçoit les autres groupes affectent
comment les gens interprètent les événements et comment les gens agissent durant des
temps politiques difficiles. Je veux étudier cette question en particulier durant la période
de la guerre civile jusqu’aux événements d’avril 1994. Durant la guerre, beaucoup de
choses ont été dites à propos des différents groupes (Tutsi et Hutu) au Rwanda. Je suis
curieuse de savoir si cela a joué un rôle dans le conflit. Je veux savoir si ce que les gens
ont entendu à propose des groupes et des relations entre les groupes a changé la façon
dont les gens percevaient les individus de leur communauté appartenant à d’autres
groupes et si cela a changé leur façon d’interagir avec eux. J’aimerais vraiment
comprendre si ce que les gens ont entendu au Rwanda à propos des groupes (Hutu et
Tutsi) a eu un impact sur ce qu’ils pensaient de la situation, des gens impliqués, et sur ce
qui devait être fait. Je regarde cela du point de vue des groupes et des communautés, et
je n’ai pas besoin de savoir à quel groupe vous appartenez, alors je ne vous demanderai
pas si vous êtes Hutu ou Tutsi.
En tant que Rwandais, vous avez l’expérience de cela, et c’est pour cela que j’aimerais
vous parler: pour connaître votre perspective, une perspective rwandaise sur la question.
Je sais qu’il s’agit de questions compliquées, et qu’elles touchent à des sujets sensibles.
Je ne sous-estime pas la difficulté d’y répondre. Elles pourraient vous faire penser à des
événements difficiles et blessants que vous avez vécus. Mais je ne vous poserai pas de
questions sur ces événements. Je veux tout simplement connaître votre point de vue
général, vous pouvez toutefois décider d’arrêter l’entrevue à n’importe quel moment ou
choisir de ne pas répondre à certaines questions si vous vous sentez mal à l’aise ou que
de répondre est trop difficile. Je respecte le point de vue de tous sur ces questions, et pour
cette raison, j’espère que vous comprendrez que je dois poser les mêmes questions à tout
le monde. Aussi, personne sauf moi ne saura ce que vous aurez dit. Je serai la seule à
lire mes notes ou à écouter les enregistrements. Si vous acceptez d’être enregistré, je
mettrai un code sur la cassette et non pas votre nom, et je serai la seule à avoir accès à
l’information pour comprendre ce code. Mes notes et cassettes seront entreposées de
façon sécuritaire et sous clé, et lorsque je terminerai mes recherches, je détruirai tout.
Maintenant que vous en connaissez plus sur ma recherche, j’aimerais vous inviter à y
participer en répondant à quelques questions. L’entrevue devrait prendre environ une
heure. Acceptez vous de participer à ma recherche? Acceptez vous que j’enregistre
l’entrevue? Cela m’aiderait à mieux me souvenir de vos réponses. Avez vous des
questions ou des commentaires avant que nous commencions?
381
AVANT
a. 1 Comment était la vie au Rwanda avant la guerre ? Je sais que les deux groupes
principaux au Rwanda sont les Hutu et les Tutsi. Je ne vous demande pas de quelle
groupe vous êtes, mais avez-vous une idée de comment la vie au Rwanda avant la guerre
était pour les Hutus ? Pour les Tutsis ? Quelle était votre impression des contacts entre
les deux groupes (Tutsi et Hutu)? Comment étaient ces contacts dans les lieux publics,
par exemple à l’église, au travail, etc. ? Dans les relations personnelles, les rencontres
sociales, les amitiés, les familles?
a.2 Est-ce que les gens des différents groupes s’entendaient bien avant la guerre? Y
avait-il des problèmes dans les relations entre les deux groupes avant la guerre?
b.1 Vous souvenez vous de certaines des perceptions générales par rapport aux différents
groupes et leurs relations au Rwanda avant la guerre ? Si oui, pouvez vous m’en parler
un peu ?
b.2 Les gens ont souvent des idées générales au sujet des différents groupes dans leur
société. Ils tiennent ces idées d’autres personnes, d’endroits, des expériences qu’ils ont
vécues. Comment avez-vous appris sur les différents groupes au Rwanda, leur histoire, et
comment les gens interagissent ? Avez-vous appris sur les groupes au Rwanda et les
relations entre les groupes par votre famille, vos amis, vos voisins, la société en
générale ? Si oui, pourriez vous me dire un peu ce que vous avez appris ? À l’école ? Si
oui, pourriez vous me dire un peu ce que vous avez appris ?
b.3 Avez vous appris d’autre chose en général sur l’histoire du Rwanda ou sur les
différents groupes au Rwanda dans votre communauté ou école?
b.4 Les gens ont différentes identités et leur attachent différents sentiments et idées, ainsi
qu’une importance différente. Par exemple, je suis Canadienne, mais je suis aussi
Québécoise, une personne d’une région particulière au Canada. Les Canadiens, et les
gens qui sont aussi Québécois, associent une signification particulière à ces identités, par
exemple en terme d’histoire ou comment ils perçoivent leur groupe ou les autres groupes
et interagissent avec eux. Je ne vous demande pas à quel groupe vous appartenez en
particulier. Mais pensez-vous que pour les gens du Rwanda, des identités comme
Rwandais, ou Tutsi et Hutu avaient une signification particulière ? Pourriez-vous me
donner une idée de ce que les gens pensaient généralement de ces identités ? Que
pensiez-vous de ces identités et de ce que vous aviez appris à leur sujet ?
CHANGEMENT DES PERCEPTIONS D’IDENTITÉ COLLECTIVE
a.1 Avez-vous senti que les choses changeaient en terme de comment les gens ordinaires
interagissaient entre Hutu et Tutsi au Rwanda quand la guerre a commencé en 1990 ?
a.2 Si oui, quand pensez-vous que les relations entre ethnies ont commencées à
changer ?
b.1 Comment les choses ont-elles changées dans les relations entre les gens de différents
groupes ? Est-ce que les interactions étaient différentes dans les endroits publics, par
382
exemple à l’église, au travail, etc. Dans les relations interpersonnelles, par exemple entre
voisins, amis et dans la famille ? Comment les choses ont-elles changées en général ?
b.2 Pensez-vous qu’il y a eu un changement par rapport à comment les gens se sentaient
par rapport à leur groupe (Hutu ou Tutsi) ou en terme de ce que ça signifiait d’être d’un
groupe ou de l’autre ? Est-ce que ça signifiait un changement par rapport à comment les
gens percevaient l’autre groupe et à leur impression de l’autre groupe ? Si oui, de quelle
façon ? Qu’est-ce que les gens ressentaient par rapport à l’autre groupe?
c.1 Si vous croyez que les choses ont changées par rapport aux perceptions que les gens
avaient de l’autre groupe, pourquoi pensez-vous que c’est arrivé?
HYPOTHÈSE PRINCIPALE : DISCOURS POLITIQUE ET PERCEPTIONS
a.1 Pensez-vous que certains individus ou groupes ont joué un rôle particulier pour
amener ce changement dans les perceptions dans gens par rapport à l’autre groupe ?
Lesquels ? Le gouvernement ? Les élites ? Les médias ?
a.2 Quel était leur rôle dans ce changement de perceptions entre les groupes? Qu’est-ce
qu’ils ont fait pour faire changer ces perceptions ?
a.3 Est-ce que ces gens ont fait la promotion de certaines idées ou perceptions par
rapport aux groupes et aux relations entre groupes ? Si oui, lesquelles? Que disaient-ils
par rapport aux différents groupes et leurs relations ?
b.1 Pensez-vous que les gens ordinaires croyaient ce que disaient ces gens à l’époque.
b.2 Plusieurs raisons ont été données pour expliquer ce qui est arrivé au Rwanda.
Pensez-vous que c’est une des raisons expliquant ce qui est arrivé au Rwanda entre 1990
et 1994 ? Pensez-vous que de croire en ce qui était dit à ce moment, sur les différents
groupes et leur relations, est une des raisons pour lesquelles les gens ont agi de la façon
dont ils ont agi durant les événements de 1990-1994.
b.3 Ou, y avait-il d’autres raisons derrière les actions des gens et derrière les
événements ? Si oui, lesquelles ?
QUESTIONS SUR LA SITUATION ACTUELLE
a.1 Comment est la situation aujourd’hui au Rwanda en terme d’interactions entre les
groupes maintenant que ça fait dix ans ? Comment les groupes se perçoivent-ils après
ces dix ans ?
a.2 Comment les groupes se perçoivent-ils et interagissent-ils maintenant qu’il y a un
processus de réconciliation nationale?
COMMENTAIRES FINAUX
a.1 Y a-t-il quelque chose que vous aimeriez rajouter ?
383
a.2 Avez-vous des questions ou des commentaires sur ma recherche ou sur les questions
que je vous ai posées ?
384
Responses to the Role of Beliefs in the Genocide
The following table presents a distribution of answers given to question b.1 “Do you think
ordinary people believed what those different actors were saying at the time?”. (A
discussion of beliefs was sometimes provided by respondents as an answer to question
b.2 “Many reasons have been given to explain what happened in Rwanda. Do you think
this was one of the reasons behind what happened in Rwanda between 1990 and 1994?
Do you think that believing in what was said around that time, about the different groups
and their relations, is a reason that people acted the way they did and that the events of
1990-1994 took place?).
Thirty-one individuals were interviewed, of which seven did not provide a clear answer to
this question. Clearly, twenty-four respondents do not constitute a representative sample.
Nonetheless, a majority of respondents indicated that the Rwandese population, or a
majority of the population, believed in ethno-centric rhetoric promoted between 1990 and
1994.
Beliefs, of course, are not an either/or phenomenon. Individual beliefs vary in degrees. To
get a sense of this, interview questions were left open-ended. Respondents allowed to
comtemplate and discuss questions, rather providing a yes or no answer. This permitted
them to qualify their statements and provide alternative explanations. Eight respondents
named one or more factors they believed, in parallel, to have played a role in participation
in the genocide (in response to b.2, or in response to b.3 ‘Or were there other reasons for
why people acted and events took place? If so, which ones?’).
TABLE APPENDIX.1: DISTRIBUTION OF ANSWERS ON BELIEFS
ANSWER
NUMBER OF
RESPONDENTS
YES
14
YES, BUT OTHER
8
FACTORS
PLAYED A ROLE
MAYBE
1
NO
1
DON’T KNOW /
NO ANSWER
7
TOTAL
31
Four respondents indicated that material gain (resources promised by leaders or
potentially gained from the fleeing or dead Tutsi) played a role in mobilisation of the Hutu
385
population. Three believed deference, or the respect of authorities, was a factor behind
popular participation in the 1994 genocide. Two raised factors tied to regionalism (infighting between regions and the wish to settle scores). One mentioned the rural-urban
divide. According to this respondent, people respondent differently whether they came
from cities or the countryside. Finally, one mentioned the fear or reprisal on the part of
authorities as a motive for participation.
386
BIBLIOGRAPHY
Adler, Emanuel. “Seizing the Middle Ground: Constructivism in World Politics,” European
Journal of International Relations, vol. 3, no. 3, 1997, pp. 319-363.
African Rights. Tribute to Courage (London: African Rights, 2002).
African Rights. Rwanda: Death, Despair and Defiance, Revised Edition (London: African
Rights, 1995).
Amnesty International. Gacaca: A Question of Justice, Rapport AFR 47/007/2002 (London:
Amnesty International, 2002).
Anderson, Benedict. Imagined Communities: Reflection on the Origin and the Spread of
Nationalism (New York: Verso Press, 1996).
André, Catherine and Jean-Philippe Plateau. “Land Tenure under Unbearable Stress:
Rwanda Caught in The Malthusian Trap,” Journal of Economic Behaviour and
Organisation, no. 38, 1998, pp. 1-47.
Arendt, Hannah. On Violence (Orlando: Fla., Harvest Books, 1970).
Arthur, W. Brian. “Competing Technologies, Increasing Returns, and Lock-In by Historical
Events,” Economic Journal, vol. 99, 1989, pp. 116-131.
Atran, Scott. Cognitive Foundations of Natural History: Towards an Anthropology of
Science (Cambridge, Mass.: Cambridge University Press, 1990).
Ayres, Jeffrey M. “Political Process and Popular Protests: The Mobilization against Free
Trade in Canada,” The American Journal of Economics and Sociology, vol. 55, no. 4,
1996, pp. 473-487.
Bailey, F. G. Stratagems and Spoils: A Social Anthropology of Politics (Toronto: Copp
Clark Publishing Company, 1969).
Baker, Bruce. “The Class of 1990: How Have the Autocratic Leaders of Sub-Saharan
Africa Fared under Democratisation?,” Third World Quarterly, vol. 19, no. 1, 1998, pp. 115127.
Barber, Bernard. “Theory of Collective Behaviour, by Neil J. Smelser,” History and Theory,
vol. 4, no. 2, 1965, pp. 264-270.
Barnett, Michael. “Culture, Strategy and Foreign Policy Change: Israel’s Road to Oslo,”
European Journal of International Relations, vol. 5, no. 1, 1999, pp. 5-36.
Bart, Annie. La presse au Rwanda. Production, diffusion et lecture depuis le début du
siècle, doctoral thesis, Université de Bordeaux III, Bordeaux, France, 1982.
Becker, Howard S. “The Chicago School, So-Called,” Qualitative Sociology, vol. 22, no. 1,
1999, pp. 3-12.
387
Bélanger, André-J. and Vincent Lemieux. Introduction à l’analyse politique (Montréal:
Presses de l’Université de Montréal, 1996).
Benford, Robert D. “‘You Could be the Hundredth Monkey’: Collective Action Frames and
Vocabularies of Motive within the Nuclear Disarmament Movement,” Sociological
Quarterly, vol. 34, 1993, pp. 195-216.
Benford, Robert D. and David A. Snow. “Framing Processes and Social Movements: An
Overview and Assessment,” Annual Review of Sociology, vol. 26, 2000, pp. 611-639.
Bennett, Andrew. “Causal Inference in Case Studies: From Mill’s Methods to Causal
Mechanisms,” paper presented at the 1999 Annual Meeting of the American Political
Science Association, Atlanta, United States, 1999.
Bennett, Andrew and Alexander L. George “Process Tracing in Case Study Research,”
paper presented at MacArthur Foundation Workshop on Case Study Methods, Belfer
Center for Science and International Affairs (BCSIA), Harvard University, Cambridge,
Mass., United States, October 17-19, 1997.
Berdal, Mats R. and David M. Malone. Greed and Grievance: Economic Agendas in Civil
War (Boulder, Co.: Lynne Rienner Publisher, 2000).
Berkeley, Bill. “Sounds of Violence: Rwanda’s Killer Radio,” New Republic, vol. 211, no, 89, 1994, pp. 18-19.
Berman, Bruce, Dickson Eyoh and Will Kymlicka. “Ethnicity and Democracy in Historical
and Comparative Perspective,” in Bruce Berman, Dickson Eyoh and Will Kymlicka, eds.,
Ethnicity and Democracy in Africa (Oxford: James Currey, 2004).
Bially Mattern, Janice. “The Power Politics of Identity,” European Journal of International
Relations, vol. 7, no. 3, 2001, pp. 349-397.
Borch, Christian. “Crowds and Economic Life: Bringing an Old Figure Back In,” paper
presented at the Centre for the Study of Invention and Social Process, Goldsmiths
College, London, United Kingdom, 21 November 2005, 28 pp.
Bougarel, Xavier. “Yugoslav Wars: The ‘Revenge of the Countryside’, Between
Sociological Reality and Nationalist Myth,” East European Quarterly, vol. 33, no. 2, 1999,
pp. 157-175.
Bowman, Glenn. “Constitutive Violence and the Nationalist Imaginary. Antagonism and
Defensive Solidarity in ‘Palestine’ and ‘Former Yugoslavia’,” Social Anthropology, vol. 11,
no. 3, 2003, pp. 319-340.
Bozić-Roberson, Agneza. “Words Before the War: Milošević’s Use of Mass Media and
Rhetoric to Provoke Ethnopolitical Conflict in Former Yugoslavia,” East European
Quarterly, vol. 38, no. 4, 2005, pp. 395-408.
Brass, Paul R. Ethnicity and Nationalism: Theory and Comparison (London: Sage
Publications, 1991).
388
Braude, Lee. “‘Park and Burgess’: An Appreciation,” American Journal of Sociology, vol.
76, no. 1, 1970, pp. 1-10.
Breslau, Daniel. “The Scientific Appropriation of Social Research: Robert Park’s Human
Ecology and American Sociology,” Theory and Society, vol. 19, 1990, pp. 417-446.
Brewer, Marilynn B. “The Many Faces of Social Identity: Implications for Political
Psychology,” Political Psychology, vol. 22, no. 1, 2001, pp. 115-125.
Brewer, Marilynn B. “The Psychology of Prejudice: Ingroup Love or Outgroup Hate,”
Journal of Social Issues, vol. 55, no. 3, 1999, pp. 429-444.
Brown, Bernard E. “Revolution and Anomie,” in Bernard E. Brown, ed., Comparative
Politics: Notes and Readings, Ninth Edition (Orlando, Fla.: Harcourt College Publishers,
2000).
Brown, Michael E. ed. International Dimensions of Internal Conflict (Cambridge: MIT
Press, 1996).
Brubaker, Roger and David D. Laitin. “Ethnic and Nationalist Violence,” Annual Review of
Sociology, vol. 24, no. 1, 1998, pp. 423-452.
Byrne, David. Complexity Theory and the Social Sciences: An Introduction (New York:
Routledge, 1998).
Calvert, Clay. “Hate Speech and Its Harms: A Communication Theory Perspective,”
Journal of Communication, vol. 47, no. 1, 1997, pp. 4-19.
Calvert, Randall. “Rationality, Identity and Expression,” in Ira Katznelson and Helen V.
Milner, eds., Political Science: The State of Discipline, 3rd Edition (New York: W.W. Norton
& Co., 2002).
Calvert, Randall. “Identity, Expression, and Rational Choice Theory,” revised version of a
paper presented at the Conference on the State of the Discipline, Washington, D.C.,
December 8-9, 2000, 70p
Campbell, David. National Deconstruction: Violence, Identity, and Justice in Bosnia
(Minneapolis: University of Minnesota Press, 1998).
Campbell, David. Writing Security: United States Foreign Policy and the Politics of Identity
(Minneapolis: University of Minnesota Press, 1998).
Canel, Eduardo. “New Social Movement Theory and Resource Mobilization Theory: The
Need for Integration,” in Michael Kaufman and Haroldo Dilla Alfonso, eds., Community
Power and Grassroots Democracy: The Transformation of Social Life (Ottawa: IDRC/Zed,
1997).
Canetti, Elias, translated by Carol Steward, Crowds and Power (London: Penguin Books,
1973).
389
Carnegie Commission on Preventing Deadly Conflict. Preventing Deadly Conflict
(Washington, D.C.: Carnegie Corporation of New York, 1997).
Carrothers, Robert M. and Denzel E. Benson. “Symbolic Interactionism in Introductory
Text-books: Coverage and Pedagogical Implication,” Teaching Sociology, vol. 31, 2003,
pp. 162-181.
Castells, Manuel. The Power of Identity. The Information Age: Economy, Society and
Culture. Volume II, Second Edition (Malden, Ma.: Blackwell Publishing, 2004).
Chapoulie, Jean-Michel. “Everett Hughes and the Chicago Tradition,” Sociological Theory,
vol. 14, no. 1, 1996, pp. 3-29.
Chazan, Naomi. “The New Politics of Participation in Tropical Africa,” Comparative Politics,
January, 1982, pp. 169-189.
Checkel, Jeffrey T. “Why Comply? Social Learning and European Identity Change,”
International Organization, vol. 55, no. 3, 2001, pp. 553-588.
Checkel, Jeffrey T. “The Constructivist Turn in International Relations Theory,” World
Politics, vol. 50, 1998, pp. 324-348.
Chrétien, Jean-Pierre. “RTLM Propaganda: The Democratic Alibi,” in Allan Thompson, ed.,
The Media and the Rwanda Genocide (London: Pluto Press/IDRC, 2007).
Chrétien, Jean-Pierre. in The Media and the Rwanda Genocide Symposium Report,
School of Journalism and Communication, Carleton University, Ottawa, 13 March 2004.
Chrétien, Jean-Pierre. “Mythes et Stratégies autour des Origines du Rwanda (XIXe-XXe
siècles), in Jean-Pierre Chrétien and Jean-Louis Triaud, Histoire d’Afrique: Les Enjeux de
Mémoire (Paris: Éditions Karthala, 1999).
Collier, David and James Mahoney. “Insights and Pitfall: Selection Bias in Qualitative
Research,” World Politics, vol. 49, no. 1, 1996, pp. 56-91.
Coppedge, Michael. “Thinkening Thin Concepts and Theories: Combining Large N and
Small in Comparative Politics,” Comparative Politics, vol. 31, no. 4, 1999, pp. 465-476.
Cortese, Anthony J. “The Rise, Hegemony, and Decline of the Chicago School of
Sociology, 1892-1945,” The Social Science Journal, vol. 32, no. 3, 1995, pp. 235-254.
Cohen, Jean L. “Strategy or Identity: New Theoretical Paradigms and Contemporary
Social Movements,” Social Research, vol. 52, no. 4, 1985, pp. 663-716.
Connor, Walker. “A Nation is a Nation, is a State, is an Ethnic Group is a …,” Ethnic and
Racial Studies, vol. 1, no. 4, 1978, pp. 377-400.
Daley, Patricia. “Ethnicity and Political Violence in Africa: The Challenge of the Burundi
State,” Political Geography, vol. 25, 2006, pp. 657-679.
390
Dallaire, Roméo. Shake Hands with the Devil: The Failure of Humanity in Rwanda
(Toronto: Random House Canada, 2004).
David, Paul A. “Clio and the Economics of QWERTY,” American Economic Review, vol.
75, no. 2, 1985, pp. 332-337.
Defleur, Melvin L. and Shearon A. Lowery. Milestones in Mass Communication Research,
3rd Edition (White Plains, N.Y.: Longman, 1994).
de Lame, Danielle. Une colline entre milles ou le calme avant la tempête. Transformations
et blocages au Rwanda rural (Tervuren: Musée royal de l’Afrique centrale, 1996).
Delanty, Gerard. Social Science: Beyond Constructivism and Realism (Minneapolis:
University of Minnesota Press, 1997).
Denitch, Bogdan. Ethnic Nationalism: The Tragic Death of Yugoslavia (Minneapolis:
University of Minnesota Press, 1996).
Derriennic, Jean-Pierre. Les guerres civiles (Paris: Presses de la Fondation nationale des
sciences politiques, 2001).
Des Forges, Alison. “Call to Genocide: Radio in Rwanda, 1994,” in Allan Thompson, ed.,
The Media and the Rwanda Genocide (London: Pluto Press/IDRC, 2007).
Des Forges, Alison. in The Media and the Rwanda Genocide Symposium Report, School
of Journalism and Communication, Carleton University, Ottawa, 13 March 2004.
Des Forges, Alison. “The Ideology of Genocide,” Issue: A Journal of Opinion, vol. 23, no.
2, 1995, pp. 44-47.
Des Forges, Alison. “Kings without Crowns: The White Fathers in Ruanda,” in Daniel F.
McCall, Norman R. Bennett and Jeffrey Butler, eds., Eastern African History (New York:
Praeger, 1969).
Deutsch, Karl. Nationalism and its Alternatives (New York: Knopf, 1969).
Devic, Ana. “Ethnonationalism, Politics, and the Intellectuals: The Case of Yugoslavia,”
International Journal of Politics, Culture and Society, vol. 11, no. 3, 1998, pp. 375-409.
Diamond, Larry. “Ethnicity and Ethnic Conflict,” The Journal of Modern African Studies,
vol. 25, no. 1, 1987, pp. 117-128.
Diani, Mario. “Linking Mobilization and Political Opportunities: Insights from Regional
Populism in Italy,” American Sociological Review, vol. 61, no. 6, pp. 1053-1069.
Dion, Douglas. “Evidence and Inference in the Comparative Case Study,” Comparative
Politics, vol. 30, no. 2, 1998, pp. 127-145.
Doder, Dusko and Louise Branson. Milosevic (New York: Free Press, 1999).
391
Doty, Roxanne Lynn. “Immigration and National Identity: Constructing the Nation,” Review
of International Studies, vol. 22, 1996, pp. 235-255.
Doty, Roxanne Lynn. “Sovereignty and the Nation: Constructing the Boundaries of
National Identity,” in Thomas J. Biersteker and Cynthia Weber, eds., State Sovereignty as
a Social Construct (New York: Cambridge University Press, 1996).
Duijvelaar, Christy. Beyond Borders: East-East Cooperation Among Environmental NGOs
in Central and Eastern Europe (Szentendre, Hungary: Regional Environmental Center for
Central and Eastern Europe, 1997).
Dunn, Kevin C. “Imagining Mobutu’s Zaïre: The Production and Consumption of Identity in
International Relations,” Millennium: Journal of International Studies, vol. 30, no. 2, 2001,
pp. 235-258.
Du Preez, Wilhelmus P. Genocide: The Psychology of Mass Murder (London:
Boyars/Bowerdean, 1994).
Eller, Jack David and Reed M. Coughlan. “The Poverty of Primordialism: The
Demystification of Ethnic Attachments,” Ethnic and Racial Studies, vol. 16, no. 2, 1993, pp.
183-201.
Elster, Jon. “A Plea for Mechanisms,” in Peter Hedstrøm and Richard Swedberg, eds.,
Social Mechanisms: An Analytical Approach to Social Theory (Cambridge: Cambridge
University Press, 1998.
Eltringham, Nigel. Accounting for Horror: Post-Genocide Debates in Rwanda (London:
Pluto Press, 2004).
Famiglietti, Antonio. “For a Realist Definition of Social Movements,” paper presented at the
Debating Realism(s) conference, Roskilde University, Denmark, 17-19 August 2001, 22p.
Farrell, Theo. “Constructivist Security Studies: Portrait of a Research Program,”
International Studies Review, vol. 4, no. 1, 2002, pp. 49-72.
Farro, Antimo L. Les mouvements sociaux: diversité, action collective et globalisation
(Montréal: Presses de l’Université de Montréal, 2000).
Fearon, James D. and David D. Laitin. “Violence and the Social Construction of Ethnic
Identity,” International Organization, vol. 54, no. 4, 2000, pp. 845-877.
Featherstone, Richard and Mathieu Deflem. “Anomie and Strain: Context and
Consequences of Merton’s Two Theories,” Sociological Inquiry, vol. 73, no. 4, 2003, pp.
471-489.
Fein, Helen. Genocide: A Sociological Explanation (London: Sage Publications, 1993).
Fierke, Karen M. “Links across the Abyss: Language and Logic in International Relations,
International Studies Quarterly, vol. 46, 2002, pp. 331-354.
392
Fine, Gary Alan. “Forgotten Classic: The Robbers Cave Experiment,” Sociological Forum,
vol. 19, no. 4, 2004, pp. 663-666.
Finnemore, Martha and Kathryn Sikkink. “International Norm Dynamics and Political
Change,” International Organization, vol. 52, no. 4, 1998, pp. 887-917.
Fisher, Kimberly. “Locating Frames in the Discursive Universe,” Sociological Research
Online, vol. 2, no. 3, 1997, pp. 1-32.
Foran, John, “Theories of Revolution Revisited: Toward a Fourth Generation,” Sociological
Theory, vol. 11, no. 1, 1993, pp. 1-20.
Franche, Dominique. Rwanda: généalogie d’un génocide (Paris: Éditions Les milles et une
nuits, 1997).
Freud, Sigmund. Group Psychology and the Analysis of the Ego (New York: Norton, 1990
[1921]).
Gagnon Jr., V.P. “Ethnic Nationalism and International Conflict: The Case of Serbia,”
International Security, vol. 19, no. 3, 1994/1995, pp. 130-166.
George, Alexander L. “Case Studies and Theory and Theory Development: The Method of
Structured, Focused Comparison,” in Paul G. Lauren, ed., Diplomacy: New Approaches in
History, Theory and Policy (New York: Free Press, 1979).
George, Alexander L., and Andrew Bennett. Case Studies and Theory Development in the
Social Sciences (Cambridge, Mass.: MIT University Press, 2004).
Giddens, Anthony. The Constitution of Society (Cambridge: Polity Press, 1984).
Gobineau, Joseph Arthur. Essai sur l’inégalité des races humaines (Paris, Éditions Pierre
Belfond, 1967 [1853-1855]).
Godina, Vesna V. “The Outbreak of Nationalism on Former Yugoslav Territory: A Historical
Perspective on the Problem of Supranational Identity,” Nations and Nationalism, vol. 4, no.
3, 1998, pp. 409-422.
Goffman, Erving. Frame Analysis: An Essay on the Organization of Experience
(Cambridge, Mass.: Harvard University Press, 1974).
Goldstone, Jack A. “Toward a Fourth Generation of Revolutionary Theory,” Annual Review
of Political Science, vol. 4, 2001, pp. 139-187.
Goldstone, Jack. “Initial Conditions, General Laws, Path Dependence, and Explanation in
Historical Sociology, American Journal of Sociology, vol. 104, no. 3, 1998, pp. 829-845.
Goodheart, Eugene. “The Power of Elias Canetti,” Partisan Review, vol. 67, no. 4, 2000,
pp. 613-621.
Goodwin, Jeff and James M. Jasper. “Caught in a Winding, Snarling Vine: The Structural
Bias of Political Process Theory,” Sociological Forum, vol. 14, no. 1, 1999, pp. 27-54.
393
Gourevitch, Philip. We Wish to Inform You that Tomorrow We Will Be Killed with Our
Families: Stories from Rwanda (New York: Picador, 1998).
Gramsci, Antonio, translated and edited by Q. Hoare and G. N. Smith. Prison Notebooks:
Selections (New York: International Publishers, 1971).
Griffiths, Paul E. “What is Innateness,” The Monist, vol. 85, no. 1, 2002, pp. 70-85.
Gruber, Thomas R. “A Translation Approach to Portable Ontology Specifications,”
Knowledge Acquisition, vol. 5, no. 2, 1993, pp. 199-220.
Guigni, Marco. “Le retour des institutions dans les sciences sociales,” Politiques et
Sociétés, vol. 21, no. 3, 2002, pp. 71-90.
Gurr, Ted Robert. Why Men Rebel (Princeton, N.J.: Princeton University Press, 1970).
Guzzini, Stefano. “A Reconstruction of Constructivism in International Relations,”
European Journal of International Relations, vol. 6, no. 2, 2000, pp. 147-182.
Hacking, Ian. The Social Construction of What? (Cambridge, Mass.: Harvard University
Press, 1999).
Haklai, Oded. “Linking Ideas and Opportunities in Contentious Politics: The Israeli
Nonparliamentary Opposition to the Peace Process,” Canadian Journal of Political
Science, vol. 36, no. 4, 2003, pp. 791-812.
Hall, Gregory O. “The Politics of Autocracy: Serbia under Slobodan Milosevic,” East
European Quarterly, vol. 33, no. 2, 1999, pp. 233-249.
Haslam, S. Alexander, John C. Turner, Penelope J. Oakes, Katherine Reynolds, Rachael
A. Eggins, Mark Nolan and Janet Tweedie. “When Do Stereotypes Become Really
Consensual? Investigating the Group-Based Dynamics of the Consensualization Process,”
European Journal of Social Psychology, vol. 28, 1998, pp. 755-776.
Hatzfeld, Jean. Une saison de machettes (Paris: Seuil, 2003).
Hayles, N. Katherine, ed. Chaos and Order: Complex Dynamics in Literature and Science
(New Practices of Inquiry) (Chicago: The University of Chicago Press, 1991).
Helmes-Hayes, Richard C. “Everett Hughes: Theorist of the Second Chicago School,”
International Journal of Politics, Culture and Society, vol. 11, no. 4, 1998, pp. 621-673.
Higiro, Jean-Marie Vianney. “Rwandan Private Print Media on the Eve of the Genocide,” in
Allan Thompson, ed., The Media and the Rwanda Genocide (London: Pluto Press/IDRC,
2007).
Hjelmar, Ulf. “Constructivist Analysis and Movement
Clarifications,” Acta Sociologica, vol. 39, 1996, pp. 169-186.
394
Organizations:
Conceptual
Hollis, Martin and Steve Smith. Explaining and Understanding International Relations
(Oxford: Clarendon Paperbacks, 1990).
Homer-Dixon, Thomas. The Upside of Down: Catastrophe, Creativity, and the Renewal of
Civilization (Toronto: Knopf Canada, 2006).
Homer-Dixon, Thomas F. Environment, Scarcity and Violence (Princeton: Princeton
University Press, 1999).
Homer-Dixon, Thomas. “Strategies for Studying Causation in Complex Ecological-Political
Systems,” The Journal of Environment and Development, vol. 5, no. 1996, pp. 132-148.
Homer-Dixon, Thomas. They and We: An Empirical and Philosophical Study of a Theory of
Social Conflict, doctoral thesis, Massachusetts Institute of Technology, Cambridge, Mass.,
United States, 1989.
Hopf, Ted. “The Promise of Constructivism in International Relations Theory,” International
Security, vol. 23, no. 1, 1998, pp. 171-200.
Horowitz, Donald L. Ethnic Groups in Conflict (Berkeley, Ca.: University of California
Press, 1985).
Human Rights Watch. Leave None to Tell the Story: Genocide in Rwanda (New York:
Human Rights Watch, 1999).
Huntington, Samuel. “The Clash of Civilizations?” Foreign Affairs, vol. 72, no. 3, 1993, pp.
22-49.
Inglehart, Ronald. The Silent Revolution: Changing Values and Political Styles among
Western Publics (Princeton, Princeton University Press, 1977).
Isaković, Zlatko. Identity and Security in Former Yugoslavia (Burlington, Vt.: Ashgate,
2000).
Jervis, Robert. “Timing and Interaction in Politics: A Comment on Pierson,” Studies in
American Political Development, vol. 14, Spring, 2000, pp. 93-100.
Jones, Bruce D. “Civil War, the Peace Process, and Genocide in Rwanda,” in Taisier M.
Ali and Robert O. Matthews, eds., Civil Wars in Africa: Roots and Resolution (Montreal:
McGill-Queen’s University Press, 1999).
Jovic, Dejan. “The Disintegration of Yugoslavia: A Critical Review of Explanatory
Approaches,” European Journal of Social Theory, vol. 4, no. 1, 2001, pp. 101-120.
Judah, Tim. “The Serbs: The Sweet and Rotten Smell of History,” Daedalus, vol. 126, no.
3, 1997, pp. 23-45.
Jung, Courtney. “Breaking the Cycle: Producing Trust Out of Thin Air and Resentment,”
Social Movement Studies, vol. 2, no. 2, 2003, pp. 147-175.
395
Kabanda, Marcel. “Kangura: the Triumph of Propaganda Refined,” in Allan Thompson, ed.,
The Media and the Rwanda Genocide (London: Pluto Press/IDRC, 2007).
Kagwi-Ndungu, Charity. in The Media and the Rwanda Genocide Symposium Report,
School of Journalism and Communication, Carleton University, Ottawa, 13 March 2004.
Kapferer, Bruce. “Remythologizations of Power and Identity: Nationalism and Violence in
Sri Lanka,” in Kumar Rupesinghe and Marcial Rubio C., eds., The Culture of Violence
(Tokyo: United Nations University Press, 1994).
Katzenstein, Peter J. ed. The Culture of National Security: Norms and Identity in World
Politics (New York: Columbia University Press, 1996).
Kaufman, Stuart J. “Symbolic Politics or Rational Choice: Testing Theories of Extreme
Ethnic Violence,” International Security, vol. 30, no. 4, 2006, pp. 45-86.
Kaufman, Stuart J. Modern Hatreds: The Symbolic Politics of Ethnic War (New York:
Cornell University Press, 2001).
Kellow, Christine L. and H. Leslie Steeves. “The Role of Radio in the Rwandan Genocide,”
Journal of Communication, vol. 48, no. 3, 1998, pp. 107-128.
Kimonyo, Jean-Paul. La participation populaire au Rwanda: de la révolution au génocide
(1959-1994), doctoral thesis, Département de science politique, Université du Québec à
Montréal, Montréal, Canada, September 2002.
King, Elisabeth. “Educating for Conflict or Peace: Challenges and Dilemmas in PostConflict Rwanda,” International Journal, vol. 60, no. 4, 2005, pp. 904-918.
Kiss, Endre. “Does Mass Psychology Renaturalize Political Theory? On the
Methodological Originality of ‘Crowds and Power’,” The European Legacy, vol. 9, no. 6, pp.
725-738.
Klandermans, Bert. “How Group Identification Helps to Overcome the Dilemma of
Collective Action,” American Behavioral Scientist, vol. 45, no. 5, 2002, pp. 887-900.
Klandermans, Bert, The Social Psychology of Protest (Oxford: Blackwell Publishers,
1997).
Lake, David A. and Donald Rothschild. “Containing Fear: The Origins and Management of
Ethnic Conflict,” International Security, vol. 21, no. 2, 1996, pp. 41-75.
Le Bon, Gustave. Aphorismes du temps présent (Paris: Les amis de G. Le Bon, 1978
[1913]).
Le Bon, Gustave. Psychologie des foules (Paris: Félix Alcan,1905), electronic version.
Lemaine, Gérard. “Social Differentiation and Social Originality,” European Journal of Social
Psychology, vol. 4, no. 1, 1974, pp. 17-52.
396
Lemarchand, René. “Les génocides se suivent mais ne se ressemblent pas : l’Holocauste
et le Rwanda,” in L’Afrique des Grands Lacs. Annuaire 2001-2002 (Antwerp: L’Harmattan,
2002).
Lemarchand, René. “Rwanda: The Rationality of Genocide,” Issue: A Journal of Opinion,
vol. 23, no. 2, 1995, pp. 8-11.
Lemarchand, René. Rwanda and Burundi (New York: Praeger, 1970).
Lepgold, Joseph and Alan C. Lamborn. “Locating Bridges: Connecting Research Agendas
on Cognition and Strategic Choice,” International Studies Review, vol. 3, no. 3, 2001, pp.
3-29.
Levi, Margaret. “A Model, a Method, and a Map: Rational Choice in Comparative and
Historical Analysis,” in Mark I. Lichbach and Alan S. Zuckerman, eds., Comparative
Politics: Rationality, Culture, and Structure (Cambridge: Cambridge University Press,
1997).
Levinger, Matthew and Paula Franklin Lytle. “Myth and Mobilisation: The Triadic Structure
of Nationalist Rhetoric,” Nations and Nationalism, vol. 7, no. 2, 2001, pp. 175-194.
Lieberman, Robert C. “Ideas, Institutions, and Political Order: Explaining Political Change,”
American Political Science Review, vol. 96, no. 4, 2002, pp. 697-712.
Liebowitz, Stan J. and Stephen E. Margolis. “Network Externality”, in Peter Newman, ed.,
The New Palgrave Dictionary of Economics and the Law (London: Palgrave MacMillan,
1998).
Lindblom, Charles E. “Political Science in the 1940s and 1950s,” Daedalus, vol. 126, no. 1,
1997, pp. 225-252.
Lizée, Pierre. Presentation during the ‘War to Peace Transitions’ panel, Eight
Peacebuilding Consultations, Ottawa, Canada, 19-20 January 2005.
Lipschutz, Ronnie D. “Negotiating the Boundaries of Difference and Security at
Millennium’s End,” in Ronnie D. Lipschutz, ed., On Security (New York: Columbia
University Press, 1995).
Longman, Timothy P. “Empowering the Weak and Protecting the Powerful: The
Contradictory Nature of Christian Churches in Rwanda, Burundi, and the Democratic
Republic of Congo,” African Studies Review, vol. 41, no. 1, 1998, pp. 49-72.
Longman, Timothy. “Christian Churches and Genocide in Rwanda,” paper presented at the
Conference on Genocide, Religion, and Modernity, United States Holocaust Memorial
Museum, Washington D.C., United States, 11-13 May 1997.
Lydall, Harold. Yugoslavia in Crisis (Oxford: Clarendon Press, 1989).
Lyman, Stanford M. “Symbolic Interactionism and Macrosociology,” Sociological Forum,
vol. 3, no. 2, 1988, pp. 295-301.
397
MacDonald, David Bruce. Balkan Holocausts? Serbian and Croatian Victim-Centred
Propaganda and the War in Yugoslavia (Manchester: Manchester University Press, 2002).
Mack, Michael. “How Literature Questions the Dogmas of Modernity,” Thesis Eleven, no.
82, 2005, pp. 109-117.
Mackie, J. L. “Causes and Conditions,” American Philosophical Quarterly, vol. 2, no. 4,
1965, pp. 245-263.
Macleod, Alex. “Les études de sécurité: du constructivisme dominant au constructivisme
critique,” Cultures et conflits, no. 54, 2004, pp. 13-51.
Mahoney, James. “Path Dependence in Historical Sociology,” Theory and Society, vol. 29,
2000, pp. 507-548.
Majstorovic, Steven. “Ancient Hatreds or Elite Manipulation? Memory and Politics in the
Former Yugoslavia,” World Affairs, vol. 159, no. 4, 1997, pp. 170-182.
Mamdani, Mahmood. When Victims Become Killers: Colonialism, Nativism and the
Genocide in Rwanda (Princeton: Princeton University Press, 2002).
Marsden, Paul. “Forefathers of Memetics: Gabriel Tarde and the Laws of Imitation,”
Journal of Memetics, vol. 4, 2000.
Martinussen, John. Society, State and Market: A Guide to Competing Theories of
Development (London: Zed Books, 1997).
Maslow, Abraham H. “A Theory of Human Motivation,” Psychological Review, vol. 50,
1943, pp. 370-396.
Masters, Daniel. “Support and Nonsupport for Nationalist Rebellion: A Prospect Theory
Approach,” Political Psychology, vol. 25, no. 4, 2004, pp. 703-726.
McAdam, Doug, John D. McCarthy and Mayer N. Zald. eds. Comparative Perspectives on
Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framing
(New York: Cambridge University Press, 1996).
McAdam, Doug, Sidney Tarrow and Charles Tilly. Dynamics of Contention (New York:
Cambridge University Press, 2001).
McClelland, John. “The Place of Elias Canetti’s Crowds and Power in the History of
Western Social and Political Thought,” Thesis Eleven, no. 45, 1996, pp. 16-27.
McGarty, Craig. “Social Identity Theory Does Not Maintain that Identification Produces
Bias, and Self-categorization Theory Does Not Maintain that Salience is Identification: Two
Comments on Mummendey, Klink and Brown,” British Journal of Social Psychology, vol.
40, 2001, pp. 173-176.
McSweeney, Bill. Security, Identity and Interest: A Sociology of International Relations
(New York: Cambridge University Press, 1999).
398
Medin, Douglas L. and Scott Atran. eds. Folkbiology (Cambridge, Mass.: MIT Press,
1999).
Meinzen-Dick, Ruth, Monica Di Gregorio and Nancy McCarthy. “Methods for Studying
Collective Action in Rural Development,” CAPRi Working Paper no. 33, International Food
Policy Research Institute, July 2004.
Mercer, Jonathan. “Anarchy and Identity,” International Organization, vol. 49, no. 2, 1995,
pp. 229-252.
Merleau-Ponty, Maurice, translated by Colin Smith. Phenomenology of Perception (New
York: Routledge, 2003).
Merton, Robert K. Social Theory and Social Structure (New York: Free Press, 1968
[1949]).
Merton, Robert K. “Social Structure and Anomie,” American Sociological Review, vol. 3,
no. 5, 1938, pp. 672-682.
Meyer, David S. “Protest and Political Opportunities,” Annual Review of Sociology, vol. 30,
2004, pp. 125-145.
Migdal, Joel S. State in Society: Studying How States and Societies Transform and
Constitute One Another (Cambridge: Cambridge University Press, 2001).
Mihata, Kevin. “Emergence and Complexity in Interactionism: Comments on David A.
Snow’s ‘Extending and Broadening Blumer’s Conceptualization of Symbolic
Interactionism,” Symbolic Interaction, vol. 25, no. 4, 2002, pp. 571-575.
Mische, Ann. “Interventions: Dynamics of Contention, Conversation with Charles Tilly
about his Recently Published Book, Dynamics of Contention, Co-Authored with Doug
McAdam and Sidney Tarrow,” Social Movement Studies, vol. 2, no. 1, 2003, pp. 85-96.
Moscovici, Serge. “The Phenomenon of Social Representations,” in Robert M. Farr and
Serge Moscovici, eds., Social Representations (London: Cambridge University Press,
1984).
Mpungwe, Ami R. “Crises and Response in Rwanda: Reflections on the Arusha Peace
Process,” in Monograph No. 36: Whither Peacekeeping in Africa? (Pretoria: Institute for
Security Studies, 1999).
Mucchielli, Roger. Psychologie de la publicité et de la propagande: Connaissance du
problème, applications pratiques (Paris: ESF, 1972).
Mueller, John. “The Banality of ‘Ethnic War’,” International Security, vol. 25, no. 1, 2000,
pp. 42-70.
Mulinda, Bernard. La liberté de la presse au Rwanda (1974-1994), licence thesis, Faculté
des lettres, Université nationale du Rwanda, Butare, Rwanda, December 1999.
399
Mummendey, Amélie and Hans-Joachim Schreiber. “Better or Just Different: Positive
Social Identity by Discrimination Against, or by Differentiation from Outgroups,” European
Journal of Social Psychology, vol. 13, no. 4, 1983, pp. 389-397.
Muukkonen, Martti. From Deviant Phenomenon to Collective Identity: Paradigm Shifts in
Social Movement Studies, master’s thesis, University of Joensuu, Joensuu, Finland, 1999.
National Unity and Reconciliation Commission. Report on the Evaluation of National Unity
and Reconciliation, Republic of Rwanda, Kigali, Rwanda, June 2002.
Neufeld, Mark. The Restructuring of International Relations Theory (Cambridge, Mass.:
Cambridge University Press, 1995).
Neumann, Iver B. Uses of the Other: ‘The East’ in European Identity Formation
(Minneapolis: University of Minnesota Press, 1999).
Newbury, Catharine. “Ethnicity and the Politics of History in Rwanda,” Africa Today, vol.
45, no. 1, 1998, pp. 7-24.
Newbury, Catharine. “Background to Genocide: Rwanda,” Issue: A Journal of Opinion, vol.
23, no. 2, 1995, pp. 11-17.
Newbury, Catharine. The Cohesion of Oppression, Clientship and Ethnicity in Rwanda
1860-1960 (New York: Columbia University Press, 1988).
Newbury, Catharine and David Newbury. “A Catholic Mass in Kigali: Contested Views of
the Genocide and Ethnicity in Rwanda,” Canadian Journal of African Studies, vol. 33, no.
2, 1999, pp. 292-328.
Newbury, David. “Precolonial Burundi and Rwanda: Local Loyalties, Regional Royalties,”
The International Journal of African Historical Studies, vol. 34, no. 2, 2001, pp. 255-314.
Newbury, David. “Ecology and the Politics of Genocide: Rwanda 1994,” Cultural Survival
Quarterly, vol. 22, no. 4, 1999, pp. 32-35.
Newbury, David. “Understanding Genocide,” African Studies Review, vol. 41, no. 1, 1998,
pp. 73-97.
Newman, David V. “Emergence and Strange Attractors,” Philosophy of Science, vol. 63,
1996, pp. 245-261.
Nkunzumwami, Emmanuel. La tragédie rwandaise: historique et perspectives (Paris:
Éditions L’Harmattan, 1996).
Nizurugero, J. “Question des réfugiés: argumentaire des régimes successifs et leur
connotation identitaire,” in Faustin Rutembesa, Josias Semujanga and Anastase Shyaka,
eds., Rwanda: Identité et citoyenneté (Butare: Éditions de l’Université Nationale du
Rwanda, 2003).
400
Noiseux, Yanick. Comment penser les mouvements sociaux dans les Amériques? Guide
de lecture (Montréal, Canada: Groupe de recherche sur l’intégration continentale, UQAM,
2004).
Oakes, Penelope. “Psychological Groups and Political Psychology: A Response to
Huddy’s ‘Critical Examination of Social Identity Theory’,” Political Psychology, vol. 23, no.
4, 2002, pp. 809-824.
Oakes, Penelope J. and John C. Turner. “Social Categorisation and Intergroup Behaviour:
Does Minimal Intergroup Discrimination Make Social Identity More Positive,” European
Journal of Social Psychology, vol. 10, 1980, pp. 295-301.
Oberschall, Anthony. “The Manipulation of Ethnicity: From Ethnic Cooperation to Violence
and War in Yugoslavia,” Ethnic and Racial Studies, vol. 23, no. 6, 2000, pp. 982-1001.
Office of the President of the Republic. The Unity of Rwandans: Before the Colonial Period
and Under the Colonial; Under the First Republic, Republic of Rwanda, Kigali, Rwanda,
August 1999.
Olson, Mancur. The Logic of Collective Action (Cambridge, Mass.: Harvard University
Press, 1971).
Organisation of African Unity. Rwanda: The Preventable Genocide, Report of the
International Panel of Eminent Personalities, July 2000.
Oxford Concise Dictionary of Current English, Seventh Edition (Oxford: The Clarendon
Press, 1983).
Park, Bernadette. “Some Thoughts on the Goals of Social Psychological Research on
Stereotyping,” Psychological Inquiry, vol. 3, no. 2, 1992, pp. 181-183.
Parson, Craig. “Showing Ideas as Causes: The Origins of the European Union,”
International Organization, vol. 56, no. 1, 2002, pp. 47-84.
Parsons, Talcott. “An Outline of the Social System,” in Craig Calhoun et. al., eds.,
Classical Sociological Theory (London: Blackwell Publishers, 2002).
Parsons, Talcott. The Social System (Glencoe, Ill.: Free Press, 1951).
Pavković, Aleksandar. The Fragmentation of Yugoslavia: Nationalism and War in the
Balkans (New York: St. Martin’s Press, 2000).
Payne, Rodger A. “Persuasion, Frames and Norm Construction,” European Journal of
International Relations, vol. 7, no. 1, 2001, pp. 37-61.
Percival, Valerie and Thomas Homer-Dixon. Environmental Scarcity and Violent Conflict:
The Case of Rwanda, Occasional Paper, Project on Environment, Population and
Security, American Association for the Advancement of Science and the University of
Toronto, June 1995.
401
Pierson, Paul. “Increasing Returns, Path Dependence, and the Study of Politics,”
American Political Science Review, vol. 94, no. 2, 2000, pp. 251-267.
Pierson, Paul. Politics in Time: History, Institutions, and Social Analysis (Princeton,
Princeton University Press, 2004).
Poggi, Christine. “Folla/Follia: Futurism and the Crowd,” Critical Inquiry, vol. 28, 2002, pp.
709-748.
Polletta, Francesca. “Contending Stories: Narratives in Social Movements,” Qualitative
Sociology, vol. 21, no. 4, 1998, pp. 419-446.
Polletta, Francesca. “’It Was Like a Fever...’ Narrative and Identity in Social Protest,”
Social Problems, vol. 45, no. 2, 1998, pp. 137-159.
Polletta, Francesca and James M. Jasper. “Collective Identity and Social Movements,”
Annual Review of Sociology, vol. 27, 2001, pp. 283-305.
Popov, Nebojša, ed. The Road to War in Serbia: Trauma and Catharsis (Budapest:
Central University Press, 2000).
Posen, Barry R. “The Security Dilemma and Ethnic Conflict,” Survival, vol. 35, no. 1, 1993,
pp. 27-47.
Potier, Elwis. “Imaginaire du contrôle des foules dans l’armée de terre française,” Cultures
& Conflits, no. 56, 2004, pp. 35-49.
Power, Samantha. “A Problem from Hell”: America and the Age of Genocide (New York:
Harper Collins, 2002).
Price, Richard and Christian Reus-Smit. “Dangerous Liaisons? Critical International
Theory and Constructivism,” European Journal of International Relations, vol. 4, no. 3,
1998, pp. 259-294.
Prunier, Gérard. The Rwanda Crisis: History of a Genocide (London: Hurst and Company,
1995).
Rabbie, Jacob M., Jan C. Schot and Lieuwe Visser. “Social Identity Theory: A Conceptual
and Empirical Critique from the Perspective of a Behavioural Interaction Model,” European
Journal of Social Psychology, vol. 19, 1989, pp. 171-202.
Ramet, Sabrina P. Balkan Babel: The Disintegration of Yugoslavia from the Death of Tito
to the Fall of Milosevic (Cambridge, Mass.: Westview Press, 2002).
Ramet, Sabrina P. “A Theory about the Causes of the Yugoslav Meltdown: The Serbian
National Awakening as a ‘Revitalization Movement’,” Nationalities Papers, vol. 32, no. 4,
2004, pp. 765-779.
Ramet, Sabrina P. Thinking about Yugoslavia: Scholarly Debates about the Yugoslav
Breakup and the Wars in Bosnia and Kosovo (Cambridge: Cambridge University Press,
2005).
402
Regan, Patrick M. and Daniel Norton. “Greed, Grievance, and Mobilization in Civil Wars,”
Journal of Conflict Resolution, vol. 49, no. 3, 2005, pp. 319-336.
Reporters sans frontières. Les médias de la haine (Paris: La Découverte, 1995).
Reyntjens, Filip. L’Afrique des Grands Lacs en crise: Rwanda, Burundi 1988-1994 (Paris:
Éditions Karthala, 1994.
Ricoeur, Paul. “Dialogue with R. Kearney,” in Richard Kearney, Dialogues with
Contemporary Continental Thinkers (Manchester: Manchester University Press, 1984), pp.
29-30.
Risse, Thomas. “‘Let’s Argue!’: Communicative Action in World Politics,” International
Organization, vol. 54, no. 1, 2000, pp. 1-39.
Ruggie, John Gerard. “What Makes the World Hang Together? Neo-Utilitarianism and the
Social Constructivist Challenge,” International Organization, vol. 52, vol. 4, 1998, p. 855885.
Ruiz Zafón, Carlos, translated by François Maspero. L’ombre du vent (Paris: Bernard
Grasset, 2001).
Rutayisire, Paul. “Les mécanismes de l’exclusion des Tutsi,” Revue africaine des livres,
vol. 1, no. 2, 2005, pp. 20-22.
Rutembesa, Faustin. “Contraintes économiques coloniales et développement des conflits,”
in Faustin Rutembesa, Josias Semujanga and Anastase Shyaka, eds., Rwanda: Identité et
citoyenneté (Butare: Éditions de l’Université Nationale du Rwanda, 2003).
Sardamov, Ivelin. “Identity’s Role in the Serbo-Croatian Conflict,” Peace Review, vol. 9,
no. 4, 1997, pp. 461-467.
Sachdev, Itesh and Richard Y. Bourhis. “Social Categorization and Power Differentials in
Group Relations,” European Journal of Social Psychology, vol. 15, 1985, pp. 415-434.
Schmitt, Carl, translated by George Schwab. The Concept of the Political (Chicago: The
University of Chicago Press, 1996).
Schorske, Carl E. “The New Rigorism in the Human Sciences, 1940-1960,” Daedalus, vol.
126, no. 1, 1997, pp. 289-309.
Schwartz, Stephen. “Beyond ‘Ancient Hatreds’: What Really Happened to Yugoslavia,”
Policy Review, vol. 97, 1999, pp. 39-51.
Segre, Sandro. “A Durkheimian Network Theory,” Journal of Classical Sociology, vol. 4,
no. 2, 2004, pp. 215-235.
Semujanga, Josias. “Formes et usages des préjugés dans le discours social du Rwanda,”
in Faustin Rutembesa, Josias Semujanga and Anastase Shyaka, eds., Rwanda: identité et
citoyenneté (Butare: Éditions de l’Université nationale du Rwanda, 2003).
403
Sells, Michael. “Crosses of Blood: Sacred Space, Religion, and Violence in BosniaHercegovina,” Sociology of Religion, vol. 64, no. 3, 2003, pp. 309-331.
Sewell, William H. Jr. “Three Temporalities: Towards an Eventful Sociology,” in Terence J.
MacDonald, ed., The Historic Turn in the Human Sciences (Ann Arbor: University of
Michigan Press, 1996.
Sewell, William H., Jr. “Ideologies and Social Revolutions: Reflections on the French
Case,” The Journal of Modern History, vol. 57, no. 1, 1985, pp. 57-85.
Shapiro, Michael J. “Warring Bodies and Bodies Politic: Tribal Warriors Versus State
Soldiers,” in Michael J. Shapiro and Hayward R. Alker, eds., Challenging Boundaries:
Global Flows, Territorial Identities (Minneapolis: University of Minnesota Press, 1996).
Sidanius, Jim. “The Psychology of Group Conflict and the Dynamics of Oppression: A
Social Dominance Perspective,” in S. Iyengar and W. McGuire, eds., Explorations in
Political Psychology (Durham, N.C.: Duke University Press, 1993).
Simon, Bernd and Bert Klandermans. “Politicized Collective Identity: A Social
Psychological Analysis,” American Psychologist, vol. 56, no. 4, 2001, pp. 319-331.
Slotin, Jenna. Unpublished manuscript, University of Toronto, Toronto, Canada, 2003.
Smelser, Neil. Theory of Collective Behaviour (New York: Free Press of Glencoe, 1963).
Smith, Anthony D. “The Ethnic Sources of Nationalism,” Survival, vol. 35, no. 1, 1993, pp.
48-62.
Snow, David A. and Robert D. Benford. “Master Frames and Cycles of Protest,” in Aldon
D. Morris and Carol McClurg Mueller, eds., Frontiers in Social Movement Theory (New
Haven, Conn.: Yale University Press, 1992.
Snow, David A., E. Burke Rochford, Jr., Steven K. Worden, Robert D. Benford. “Frame
Alignment Processes, Micromobilization, and Movement Participation,” American
Sociological Review, vol. 51, 1986, pp. 464-481.
Snyder, Jack. “Anarchy and Culture: Insights from the Anthropology of War,” International
Organization, vol. 56, no. 1, 2002, pp. 7-45.
Snyder, Richard C., H.W. Bruck and Burton Sapin. Decision-Making as an Approach to the
Study of International Politics (Princeton: Princeton University Press, 1954).
Sosa, Ernest. “Condition,” in Cambridge Dictionary of Philosophy, Second Edition
(Cambridge, Cambridge University Press, 1999, p. 171.
Steinberg, Marc W. “Tilting the Frame: Considerations on Collective Action Framing from a
Discursive Turn,” Theory and Society, vol. 27, 1998, pp. 845-872.
Sterling-Folker, Jennifer. “Realism and the Constructivist Challenge: Rejecting,
Reconstruction, or Rereading,” International Studies Review, vol. 4, no. 1, 2002, pp. 73-97.
404
Stewart-Steinberg, Suzanne R. “The Secret Power of Suggestion: Scipio Sighele and the
Postliberal Subject,” Diacritics, vol. 33, no. 1, 2003, pp. 60-79.
Stouffer, Samuel A., Edward A. Suchman, Leland C. DeVinney, Shirley A. Star and Robin
M. Williams, Jr. The American Soldier. Adjustment During Army Life (vol. 1) (Princeton,
N.J.: Princeton University Press, 1949).
Strauss, Anselm. “Blumer on Industrialization and Social Change,” Contemporary
Sociology, vol. 20, no. 2, 1991, pp. 171-172.
Swartz, Norman. Department of Philosophy, Simon Fraser University, The Concepts of
Necessary
Conditions
and
Sufficient
Conditions,
http://www.sfu.ca/philosophy/swartz/conditions1.htm, (last accessed 29 May 2007).
Tajfel, Henri, M.G. Billig, R. P. Bundy, and Claude Flament. “Social Categorization and
Intergroup Behaviour,” European Journal of Social Psychology, vol. 1, 1971, pp. 149-178.
Tajfel, Henri and A. L. Wilkes. “Classification and Quantitative Judgment,” British Journal
of Psychology, vol. 54, 1963, pp. 101-114.
Tarde, Gabriel. Les lois de l’imitation (Paris: Kimé Éditeur, 1993 [1890]).
Tarrow, Sidney. “Paradigm Warriors: Regress and Progress in the Study of Contentious
Politics,” Sociological Forum, vol. 14, no. 1, 1999, pp. 71-77.
Tarrow, Sidney. Power in Movement: Social Movement, Collective Action and Politics
(Cambridge, Mass.: Cambridge University Press, 1998).
Taylor, Michael. “Structure, Culture and Action in the Explanation of Social Change,”
Politics and Society, vol. 17, no. 2, 1989, pp. 115-162.
Thelen, Kathleen. “Timing and Temporality in the Analysis of Institutional Evolution and
Change,” Studies in American Political Development, vol. 14, Spring, 2000, pp. 101-108.
Thibault, Jean-François. “Représenter et connaître les Relations Internationales:
Alexander Wendt et le paradigme constructiviste,” Note de recherche no. 7, Centre
d’études des politiques étrangères et de sécurité, Université du Québec à Montréal, no. 7,
janvier, 1997.
Thomson, Susan M. “The Politics of Trauma in Post-Genocide Rwanda: Treatment for
Some at the Expense of the Many,” talk given at the Peacebuilding and Trauma Recovery:
Integrated Strategies in Post-War Reconstruction Conference, University of Denver,
Denver, United States, 25 February 2007.
Tilly, Charles. “Rhetoric, Social History, and Contentious Politics: Reply to Critics,”
International Review of Social History, vol. 49, no. 1, 2004, pp. 132-141.
Tilly, Charles. “Social Boundary Mechanisms,“ Philosophy of the Social Sciences, vol. 34,
no. 2, 2004, pp. 211-236.
405
Tilly, Charles. From Mobilization to Revolution (Reading, Mass.: Addison-Wesley, 1978).
Tilly, Charles and Sidney Tarrow. Contentious Politics (Boulder, Colo.: Paradigm
Publishers, 2007).
Tindall, David B. “From Structure to Dynamics: A Paradigm Shift in Social Movement
Research?,” Canadian Review of Sociology and Anthropology, vol. 40, no. 4, 2003, pp.
481-487.
Treviño, A. Javier. “Parsons’ Action-System Requisite Model and Weber’s Elective Affinity:
A Convergence of Convenience,” Journal of Classical Sociology, vol. 5, no. 3, 2005, pp.
319-348.
Turner, John C. “Social Comparison and the Social Identity: Some Prospects for
Intergroup Behaviour,” European Journal of Social Psychology, vol. 5, no. 1, 1975, pp. 534.
Turner, John C., with Michael A. Hogg, Penelope J. Oakes, Stephen D. Reicher, and
Margaret S. Wetherell. Rediscovering the Social Group: A Self-Categorization Theory
(Oxford: Basil Blackwell, 1987).
Turner, Ralph H. “Some Contributions of Muzafer Sherif to Sociology,” Social Psychology
Quarterly, vol. 53, no. 4, 1990, pp. 283-291.
Turner, Ralph H. and Lewis M. Killian. Collective Behavior, Fourth Edition (Englewood
Cliffs, N.J.: Prentice-Hall, 1993).
UNICEF, Rwanda–Statistics, http://www.unicef.org/infobycountry/rwanda_statistics.html,
(last accessed 29 May 2007).
United Nations, Final Report of the United Nations Commission of Experts Established
Pursuant to Security Council Resolution 780 (1992), The United Nations Commission of
Experts, December 1994.
United States Census Bureau. Population Pyramid Summary for Rwanda,
http://www.census.gov/cgi-bin/ipc/idbpyrs.pl?cty=RW&out=s&ymax=250, (last accessed
31 May 2007).
Uvin, Peter. “Reading the Rwandan Genocide,” International Studies Review, vol. 3, no. 3,
2001, pp. 75-99.
Uvin, Peter. Aiding Violence: The Development Enterprise in Rwanda (West Hartford,
Conn.: Kumarian Press, 1998).
Uvin, Peter. “Prejudice, Crisis and Genocide in Rwanda,” African Studies Review, vol. 40,
no. 2, 1997, pp. 91-115.
Uvin, Peter. “Tragedy in Rwanda: The Political Ecology of Conflict,” Environment, vol. 38,
no. 3, 1996, pp. 7-15.
406
Van Evera, Stephen. “Hypotheses on Nationalism and War,” in Sean M. Lynn-Jones and
Steven E. Miller eds., Global Dangers: Changing Dimensions of International Security
(Cambridge: MIT Press, 1995), pp. 252-285.
Vansina, Jan. Antecedents to Modern Rwanda: The Nyiginya Kingdom (Madison, Wisc.:
University of Wisconsin Press, 2004).
Verwimp, Peter. “Machetes and Firearms: The Organization of Massacre in Rwanda,”
Journal of Peace Research, vol. 43, no. 1, 2006, pp. 5-22.
Vider, Stephen. “Rethinking Crowd Violence: Self-Categorization Theory and the
Woodstock 1999 Riot,” Journal for the Theory of Social Behaviour, vol. 34, no. 2, 2004, pp.
141-166.
Voci, Alberto. “Relevance of Social Categories, Depersonalization and Group Processes:
Two Field Tests of Self-categorization Theory,” European Journal of Social Psychology,
vol. 36, 2006, pp. 73-90.
Voegelin, Eric. The New Science of Politics: An Introduction (Chicago: The University of
Chicago Press, 1987 [1952]).
Walker, Iain and Heather J. Smith. eds. Relative Deprivation Theory: Specification,
Development, and Integration (New York: Cambridge University Press, 2002).
Weber, Max. The Protestant Ethic and the Spirit of Capitalism (New York: Scribner's,
1958 [1904-1905]).
Weber, Max, translated by Talcott Parsons. The Theory of Social and Economic
Organization (New York: Free Press, 1947).
Webster’s Ninth New Collegiate Dictionary (Springfield, Mass.: Merriam-Webster
Publishers 1988).
Weeber, Stan C. and Daniel G. Rodeheaver. “Militias at the Millennium: A Test of
Smelser’s Theory of Collective Behavior,” The Sociological Quarterly, vol. 44, no. 2, 2003,
pp. 181-204.
Weldes, Jutta, Mark Laffey, Hugh Gusterson, and Raymond Duvall. eds. Cultures of
Insecurity: States, Communities, and the Production of Danger (Minneapolis: University of
Minnesota Press, 1999).
Welsh, David. “Domestic Politics and Ethnic Conflict,” Survival, vol. 35, no. 1, 1993, pp.
63-80.
Wendt, Alexander. Social Theory of International Politics (New York: Cambridge University
Press, 1999).
Wendt, Alexander. “Anarchy Is What States Make of It: The Social Construction of Power
Politics,” International Organization, vol. 46, no. 2, 1992, pp. 391-425.
407
Werbner, Richard, and Terence Ranger, eds. Postcolonial Identities in Africa (London: Zed
Books, 1996).
Williams, Robin M., Jr. “The Sociology of Ethnic Conflict: Comparative International
Perspectives,” Annual Review of Sociology, vol. 20, 1994, pp. 49-79.
Williams, Rhys H. “Constructing the Public Good: Social Movements and the Cultural
Resources,” Social Problems, vol. 42, no. 1, 1995, pp. 124-144.
Wilmer, Franke. The Social Construction of Man, the State, and War: Identity, Conflict, and
the Violence in the Former Yugoslavia (London: Routledge, 2002).
Wilmer, Franke. “Identity, Culture, and Historicity: The Social Construction of Ethnicity in
the Balkans,” World Affairs, vol. 160, no. 1, 1997, pp. 3-16.
Wind, Marlene. “Nicholas G. Onuf: The Rules of Anarchy,” in Iver B. Neumann and Ole
Waever, eds., The Future of International Relations: Masters in the Making? (London:
Routledge, 1997).
Woodward, Susan L. Balkan Tragedy: Chaos and Dissolution After the Cold War
(Washington, D.C.: The Brookings Institution, 1995).
World
Bank.
Rwanda:
Quick
Facts,
http://web.worldbank.org/WBSITE/EXTERNAL/COUNTRIES/AFRICAEXT/RWANDAEXTN
/0,,menuPK:368741~pagePK:141132~piPK:141109~theSitePK:368651,00.html,
(last
accessed 31 May 2007).
Young, Crawford. “Patterns of Social Conflict: State, Class and Ethnicity,” Daedalus, vol.
111, no. 2, 1982, pp. 71-98.
Young, Crawford. The Politics of Cultural Pluralism (Madison, Wisc.: The University of
Wisconsin Press, 1976).
Zehfuss, Maja. “Constructivism and Identity: A Dangerous Liaison,” European Journal of
International Relations, vol. 7, no. 3, 2001, pp. 315-348.
408
