Understanding the Age of Transitional Justice: Crimes, Courts, Commissions, and Chronicling 9780813597805

Since the 1980s, an array of legal and non-legal practices—labeled Transitional Justice—has been developed to support po

208 9 5MB

English Pages 250 [252] Year 2018

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Understanding the Age of Transitional Justice: Crimes, Courts, Commissions, and Chronicling
 9780813597805

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

UNDERSTANDING THE AGE OF TR ANSITIONAL JUSTICE

CENTER FOR THE STUDY OF

GENOCIDE

& HUMAN RIGHTS

GENOCIDE, PO­L ITI­C AL VIO­L ENCE, ­H UM AN RIGHTS SERIES Edited by Alexander Laban Hinton, Stephen Eric Bronner, and Nela Navarro Nanci Adler, ed., Understanding the Age of Transitional Justice: Crimes, Courts, Commissions, and Chronicling Alan W. Clarke, Rendition to Torture Lawrence Davidson, Cultural Genocide Daniel Feierstein, Genocide as Social Practice: Reor­ga­niz­ing Society ­under the Nazis and Argentina’s Military Juntas Alexander Laban Hinton, ed., Transitional Justice: Global Mechanisms and Local Realities ­after Genocide and Mass Vio­lence Alexander Laban Hinton, Thomas La Pointe, and Douglas Irvin-­Erickson, eds., Hidden Genocides: Power, Knowledge, Memory Douglas A. Kammen, Three Centuries of Conflict in East Timor Walter Richmond, The Circassian Genocide Victoria Sanford, Katerina Stefatos, and Cecilia M. Salvi, eds., Gender Vio­lence in Peace and War: States of Complicity Irina Silber, Everyday Revolutionaries: Gender, Vio­lence, and Disillusionment in Postwar El Salvador Samuel Totten and Rafiki Ubaldo, eds., We Cannot Forget: Interviews with Survivors of the 1994 Genocide in Rwanda Anton Weiss-­Wendt, A Rhetorical Crime: Genocide in the Geopo­liti­cal Discourse of the Cold War Ronnie Yimsut, Facing the Khmer Rouge: A Cambodian Journey

UNDERSTANDING THE AGE OF TR ANSITIONAL JUSTICE Crimes, Courts, Commissions, and Chronicling

edited by

n a nc i a dler

rutger s uni v er sit y p r ess

New Brunswick, Camden, and Newark, New Jersey, and London

Library of Congress Cataloging-­in-­Publication Data Names: Adler, Nanci, editor. Title: Understanding the age of transitional justice : crimes, courts, commissions, and chronicling / edited by Nanci Adler. Description: New Brunswick, New Jersey : Rutgers University Press, 2018. | Series: Genocide, po­liti­cal vio­lence, ­human rights | Includes bibliographical references and index. Identifiers: LCCN 2017055139 | ISBN 9780813597775 (hardback : alk. paper) | ISBN 9780813597768 (paperback : alk. paper) Subjects: LCSH: Po­liti­cal crimes and offenses. | Transitional justice. | Truth commissions. | BISAC: POLITICAL SCIENCE / Po­liti­cal Freedom & Security / ­Human Rights. | LAW / International. | SOCIAL SCIENCE / Criminology. | LAW / Judicial Power. Classification: LCC K5250 .U53 2018 | DDC 340/.115—­dc23 LC rec­ord available at https://­lccn​.­loc​.­gov​/­2017055139 A British Cataloging-­in-­Publication rec­ord for this book is available from the British Library. This collection copyright © 2018 by Rutgers, The State University of New Jersey Individual chapters copyright © 2018 in the names of their authors All rights reserved No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, or by any information storage and retrieval system, without written permission from the publisher. Please contact Rutgers University Press, 106 Somerset Street, New Brunswick, NJ 08901. The only exception to this prohibition is “fair use” as defined by U.S. copyright law. The paper used in this publication meets the requirements of the American National Standard for Information Sciences—­Permanence of Paper for Printed Library Materials, ANSI Z39​.­48​-­1992. www​.­rutgersuniversitypress​.­org Manufactured in the United States of Amer­i­ca

CONTENTS



Introduction: On History, Historians, and Transitional Justice nanci adler

1

part i: the complex relationship between truth and justice 1

Swinging the Pendulum: Fin-­de-­Siècle Historians in the Courts vladimir petrović

21

2

Time, Justice, and ­Human Rights: Statutory Limitation on the Right to Truth? william a. schabas

37

3

How Truth Recovery Can Benefit from a Conditional Amnesty jeremy sarkin

56

4

New Epistemologies for Confronting International Crimes: Developing the Information, Dialogue, and Pro­cess (IDP) Approach to Transitional Justice stephan parmentier , mina rauschenbach, and maarten van craen

75

part ii: the narrative of the trial rec­o rd 5

The Spark for Genocide? Propaganda and Historical Narratives at International Criminal Tribunals richard ashby wilson

6

The International Criminal Trial Rec­ord as Historical Source thijs b. bouwknegt

97

118

v

vi Contents

part iii: the afterlife of transitional justice pro­c esses 7

Narrating (In)Justice in the Form of a Reparation Claim: Bottom-­Up Reflections on a Postcolonial Setting—­ The Rawagede Case nicole l. immler

149

8

Collective and Competitive Victimhood as Identity in the Former Yugo­slavia christian axboe nielsen

175

9

Perpetrator-­Victims: How Universal Victimhood in Cambodia Impacts Transitional Justice Mea­sures timothy williams

194

10

Collective Crimes, Collective Memory, and Transitional Justice in Bangladesh kjell anderson

213

Acknowl­edgments 237 Notes on Contributors 239 Index 243

UNDERSTANDING THE AGE OF TR ANSITIONAL JUSTICE



INTRODUCTION On History, Historians, and Transitional Justice NANCI ADLER

Since Nuremberg, and more pronouncedly since the move t­oward demo­cratic rule in Africa and Latin Amer­i­ca in the 1980s, t­ here has been a rise in understanding that “bad pasts” need to be institutionally addressed in order to create “good ­futures.”1 It has now become the norm that postrepressive, postconflict, postauthoritarian successor states confront—or support the pro­cess of confronting—­the crimes of their pre­de­ces­sor repressive regimes, both for the victims and for society as a ­whole. We may mark such an advance in sociocultural evolution as the Age of Transitional Justice. The expectation that an array of ­legal and nonlegal mea­sures ­will be undertaken in the aftermath of mass po­liti­cal crimes is intended to provide a vital course correction to recurrent cycles of vio­ lence and impunity. Transitional justice has generated a variety of strategic and tactical approaches for redressing often irreparable harms. Th ­ ese have included: international criminal tribunals, national or local l­ egal proceedings, truth commissions, restitution, the accurate revision of history, public apologies, the establishment of monuments and museums, and official commemorations. Each of t­hese, in dif­fer­ent ways, attempts to incorporate the lived and narrated experiences of the victims, the witnesses, and the perpetrators. Such narratives bring the crimes into a shared public domain where they can be co-­processed and critically evaluated. The outcome of the convergence of multilevel narratives produced by many ­people and disparate groups can result only in an approximation of the full-­scale of the targeted events. Nevertheless, even as we recognize ­these limitations, real ­things happen to real p­ eople who frame them as au­then­tic experiences. While their narrative reconstruction of events is subjected to their interpretive framework, 1

2

n a nc i a dler

so, too, are the assessments of judges, truth commissioners, the public, the media, and opinion-­makers alike. Twenty years ­after the operation of one of the most prominent truth commissions—­the South African Truth and Reconciliation Commission—­and, now, as the International Criminal Tribunal for the former Yugo­slavia (ICTY) and the International Criminal Tribunal for Rwanda (ICTR) (both U.N. ad hoc tribunals) have wound down and closed, we are entering a new phase, one in which we are well-­positioned to scrutinize the pro­cesses as well as the products of the age of transitional justice, even if only as a work in pro­gress. Transitional justice is young in age and beset by growing pains, but we have gained some deeper understanding of the differential efficacy of transitional justice mechanisms, as well as the rec­ords they have produced. Pablo de Greiff, U.N. Special Rapporteur on truth, justice, reparation, and guarantees of non-­recurrence, has found cause for cele­bration in the fact that in the short span of twenty to thirty years, truth-­seeking and justice-­seeking have consolidated their efforts and have come to include more victim and gender participation. Th ­ ere is, however, less to be hailed when it comes to the field’s engagement with impact issues, that is, what changes, improvements, and/or reconciliation ­these efforts actually brought to the affected, divided socie­ties.2 Indeed, the consequences of transitional justice mechanisms are still unfolding. We are still in the early stages of exploring post–­transitional justice trajectories and identifying the impediments to pro­gress, but two obstacles that stand out are the challenges to amnesties and to some of the U.N. and Hague courts’ rulings. We can infer that t­ hese are reliable indicators of a larger disconnect and that more obstacles await. Early transitional justice models ­were designed to help reconstruct postrepressive socie­ties—­those recently liberated from repressive regimes and assumed to have thrown off their repressive culture. We w ­ ere confronted, however, with the fact that removing the repressor from the society did not remove the repression from the culture. Th ­ ese fragile socie­ties that are postconflict, but not postrepressive, require a strategic shift to a model more suited to t­ hose challenges.3 We are still uncertain about how to help transitional justice take root and enhance its overall credibility in the socie­ties where the crimes took place, or ­those from which the perpetrators emerged. For example, countries like Liberia, previously ruled by Charles Taylor, and Serbia, previously ruled by Slobodan Milosevic, are still governed by a culture of repression so entrenched in the defendants’ homeland that The Hague courts’ rulings are flouted by many of its citizenry.4 This volume provides a broad perspective on current institutionalized practices for confronting genocide, mass po­liti­cal vio­lence, and historical injustices. It goes beyond looking at the successes and limitations of con­temporary models. It also conveys the ways t­ hese injustices ­were experienced by the victims and analyzes the kinds of narratives that have been accorded credibility by gaining an

Introduction 3

audience in international criminal tribunals, ­trials, and truth commissions, and the public space. A better understanding of t­ hese narratives could contribute to identifying the c­ auses and consequences of repression and may inform transitional justice strategies.

addressing cultures of repression One of the systemic flaws in the early approach to transitional justice mechanisms was the expectation that achieving justice in postrepressive states would be correlated with achieving reconciliation, as it might have been if all the parties had agreed on the facts and the criteria for a just outcome, and if justice meant the same ­thing to dif­fer­ent ­peoples. In a number of cases, ­these models served as a functional starting point,5 but failed to reconcile or sufficiently address the competing narratives of history that persist in postrepressive socie­ties.6 This disconnect was not remedied by the laudable increase of victim participation in transitional justice proceedings.7 Indeed, repressive socie­ties often blur the boundary between perpetrators and victims, so the increased opportunity to tell stories also gave rise to competing narratives of who was a victim.8 This kind of dispute ­matters in dealing with the legacy of repression b­ ecause continued disagreement on facts and ­causes can marginalize victim groups, intensify rivalry, and facilitate the recycling of old repressions into new ones. For their part, the academic and practitioner debates that have addressed the difficulties of the reconciliation pro­cess have not arrived at a solution for—or an adequate understanding of—­the obstacles to reconciliation chronically regenerated by the enduring culture of repression in socie­ties aspiring to be postrepressive.9 This deficit stymies efforts to deal with the “hard” or non-­cases in which transitional justice is not on the state agenda (i.e., Rus­sia, see figure I.1), t­ hose in which it is restricted or resisted (i.e., Serbia), or t­ hose in which it is ambivalently implemented but its achievements are undermined by divisive narratives that justify the repression of rights, which can readily lead to renewed vio­lence (i.e., Rwanda).10

competing narratives One of the impor­tant questions that emerges when we study the “products” of this age of transitional justice is what kind of history is being written in the narratives recorded by judges and in the trial transcripts themselves. For example, while some Yugo­slavia Tribunal judgments have contributed to the pursuit of historical accuracy regarding the ­causes of the conflict(s),11 the “­legal narrative” of the courtroom, in contrast to the personal narrative, is purposefully driven ­toward a hard-­edged verdict that may tell a narrow story.12 It is likely to be true, but it is not the ­whole truth. In consequence, contending parties often enter and leave the courtroom with their own “truths” still intact, unmodified by exposure

4

n a nc i a dler

figure i.1. Remembering Stalin then and now. At the end of the Soviet era, in the absence

of official registration, the archives on the terror of the anti-­Stalinist organ­ization Memorial ­were stored in private apartments, like this one belonging to Nikita Okhotin, Moscow, 1990. (Photo [on left] by Rob Knijff.) Transitional justice was not on the post-­Soviet state’s agenda. 2017 calendar for sale, Moscow, 2017. (Photo [on right] by Nanci Adler.)

to other narratives of how events tran­spired. This in no way diminishes the value of the rec­ord produced. However, transitional justice mechanisms may benefit by recognizing the obstacles to change embedded in the implicit and explicit “truths” conveyed by the rival narratives that roil just ­under the surface. Some of ­these truths are related to ideological convictions,13 ­others to age-­old, state-­ sponsored xenophobia (aided and abetted by the media and by history and civics curricula). ­These coexisting “truths,”14 based on the dif­fer­ent perceptions of dif­fer­ent groups, provide an interpretive frame that structures and sustains their version of real­ity and impedes transitional justice from taking root. However, what historians might contribute to the goals of transitional justice is, in a word, “AND”—­the hitherto insufficiently considered likelihood that a richer, more functional truth emerges when competing narratives are reframed in a conceptual shift as contributing narratives. Elazar Barkan and o­ thers have pioneered the approach of historical dialogue between groups on dif­fer­ent sides of the po­liti­cal divide as a potential starting point to repair the schism of competing narratives.15 An example of how confounding this prob­lem can be is illustrated by the dueling narratives of the Serbs and Bosnians who inhabit former Yugo­slavia. The central theme of Serb narratives characterizes the catastrophic war as an internal conflict, triggered from the outside, and resulting in a civil war in which Serbs

Introduction 5

­ ere also victims.16 By contrast, the central Bosniak narrative frames ­these same w events as unbridled Serbian aggression against them.17 Now, years ­after the physical ­battle in former Yugo­slavia has ended, the divisiveness remains, perpetuated by competing narratives raging in and out of the courtroom and battling about whose version of truth is right or wrong rather than recognizing the accurate and inaccurate information in each version and potentially using this richer combination of information to head off f­ uture disputes.18 The ICTY rulings have had ­little influence on the perceptions and be­hav­ior of t­ hose in the affected regions. In 2012, Tomislav Nikolic, then newly-­elected Serbian president, denied that the Srebrenica massacre constituted genocide, despite the rulings of both the ICTY and the International Court of Justice.19 This denial reflected his constituents’ sentiments, proclaimed an ethos of repression impervious to international criticism, and reinforced Serbia’s re­sis­tance to transitional justice.20 Substantial criticisms regarding results have also been leveled against the International Criminal Tribunal for Rwanda (ICTR). Rwanda experienced colonial rule that followed and led the division of its population into antagonistic ethnic categories. ­Here, the suppression of individual ­human rights had many starting points and culminated in a one-­hundred-­day genocide that claimed the lives of approximately 800,000 Rwandan citizens, the overwhelming majority of whom ­were Tutsi. In the aftermath, several transitional justice mea­sures ­were implemented. ­These included the ICTR, the gacaca “lawn court” trial (traditional dispute resolution system within communities), and national courts. As ­these mechanisms wound down, Rwanda could claim to have achieved widespread justice, as evidenced by a total of one million Rwandans prosecuted.21 At pres­ent, however, ­there is just one official narrative on the “genocide against the Tutsis,” labeled the “healing truth.” The ICTR has been accused of facilitating, perhaps even aiding and abetting, this outcome.22 Competing narratives have been criminalized, and any referencing of ethnicity is forbidden. The regime justifies the restrictions it has imposed on po­liti­cal parties, civil society, and the media as a safeguard against the resurgence of ethnic vio­lence. Many maintain that ethnicity did not exist in precolonial Rwanda.23 Even as the twentieth anniversary of the genocide was being commemorated in 2014, much of it was being “dis-­memorated,” as the public discussion of this part of Rwandan history was limited to the state-­sponsored narrative.24 The suppression of broader, multiple points of view may undermine, rather than promote the pro­cesses of reconciliation.

lessons learned? ­ ese are but two of many similar examples demonstrating that the trial model is Th of limited success in reconciling competing narratives of the same events. Granted this limitation, tribunals excel at facilitating access to testimonies that

figure i.2. The archives of the more than 12,000 gacaca courts in Rwanda, Kigali, 2012. It

was estimated that the rec­ords, which ­were stored in some 19,000 boxes, comprised more than an estimated 60 million pages. (Photo by Thijs Bouwknegt.)

Introduction 7

might other­wise have been inaccessible. This alone makes the rec­ord they create an indispensable source that, like all ­others, must be subjected to critical scrutiny. Truth commissions, for their part, have been successful in documenting a  broad range—­and even competing narratives—of repression, but have not always been successful in addressing their trajectory. The privileging of truth over justice may have achieved consensus during the commission proceedings, but in their aftermath some victims and survivors have challenged the legitimacy of outcomes that allowed perpetrators to escape with impunity.25 Moreover, as Molly Andrews has argued, ­these testimonies ­were not immune to being ­shaped by the statement-­takers and givers, who influenced and even tainted the “truths” they produced before the commission.26 This failure of transitional justice to meet expectations reflects several insufficiently examined assumptions regarding the cultural milieu necessary for success. Among ­these are: the belief that a postrepressive government is synonymous with a postrepressive culture; a conflation of truth, justice, and fairness (even by some judges, ­lawyers, and commissioners) that offends the commonsense judgment of victims and survivors; the belief that the “truths” produced by court testimony can be at all immune to the social context of the courtroom; and the belief that the repressors’ acknowl­edgment of a crime is more impor­tant to the aim of reconciliation than our acknowl­edgment of the preceding xenophobia and dehumanization of the “other” embedded in the crime. Carol Gluck has observed that while addressing the past is not the province of the law, international tribunals have significantly facilitated the disclosure and transmission of narratives of past repression. The themes and plots of t­ hese stories, the emotions they express, and their value as au­then­tic sources, along with the pruning of ­these testimonies by the judicial pro­cess, influence the prospects for reconciliation in the affected socie­ties.27 Stories are accounts of what happened. They overlap with explanations but are not the same, so trying to account for p­ eople’s be­hav­ior by listening to their testimonies provides only one part of the tale. We have no single narrative of the past that can well explain why good ­people can so readily commit evil acts collectively, or why previously good neighbors turn against each other. We need to identify and analyze which history is being remembered by whom, and consider ­whether ­there can be, as U.N. Special Rapporteur on cultural rights Farida Shaheed, has termed it, a “multi-­voiced narrative of the past”28 that might contribute to a shared narrative of the past.29 Such a history might facilitate an understanding of what conditions triggered the vio­lence, how propaganda perpetuated them, as Richard Wilson has extensively researched, and what allowed some ­people to resist the repression. The opening of governmental rec­ords and state archives would be an essential component of this pro­cess, ­because, in the words of Kate Doyle of the National Security Archive, “Never Again” can start only by answering the question, “Never again what?”30 And that “what” should include

8

n a nc i a dler

the preliminary conditions that led to the malevolent events. Furthermore, since a repressive past influences the pres­ent sense of justice and injustice, it is worth considering, as does William Schabas in this volume, ­whether the pursuit of truth should have any temporal limitations.

overview The contributors to this volume scrutinize transitional justice pro­cesses from a variety of theoretical perspectives and consider some of the endemic issues that thwart postrepressive socie­ties, victims, transitional justice prac­ti­tion­ers, judges, and historians. They examine how institutionalized, global transitional justice pro­cesses and practices interact with bottom-up, local perceptions of justice, and they reflect on how some victims feel empowered and ­others, alienated by the  variety of outcomes ranging from sentencing through restitution, and acquittals. This volume is divided into three parts, each examining an overarching set of issues. Part I addresses “The Complex Relationship between Truth and Justice” and opens with the entry of historians into the courtroom in Vladimir Petrović’s “Swinging the Pendulum: Fin-­de-­Siècle Historians in the Courts.” In his analy­sis of the continuously negotiated partnership between history and law, he traces the role of historians in the courts, along with the role of history itself in ­legal proceedings. Petrović looks at how the use of historians as expert witnesses emerged and evolved, and poses the question of w ­ hether historians are properly suited and vetted to justify their increasing role as forensic contributors to international criminal tribunals. He recognizes that even long a­ fter a­ ctual ­battles end, many historians remain entrenched in historiographical ­battles regarding their ­causes and consequences. Using several case studies, Petrović examines the significant contributions of historians to the “undoing of the miscarriage of justice.” Fi­nally, he scrutinizes the complex relationship between history and law—­ two major contributors to the outcome of transitional justice. In his essay, “Time, Justice, and H ­ uman Rights: Statutory Limitation on the Right to Truth?” William Schabas explores the right to truth, an issue that is gaining increasing momentum in both international ­human rights law and approaches to transitional justice. The twin imperatives of investigating atrocities and holding perpetrators accountable are part and parcel of the right to justice. Accordingly, Schabas calls our attention to the prevailing judicial reluctance to look too far into the past. Examining the right to truth and the passage of time, Schabas reminds us of chronologically past, but culturally pres­ent events such as Katyn (the long-­denied and ­later marginalized1940 massacre of thousands of Polish officers by Stalin’s NKVD in a forest near Smolensk, Rus­sia), the Spanish Civil War, and the Armenian genocide. He points out that the Eu­ro­pean Court of ­Human Rights seemed to be employing a statutory limitation in its reluctance to

Introduction 9

deal with the latter case ­because a c­ entury had passed. In fact, such limitation is prohibited for atrocity crimes. In examining this and other cases, Schabas reflects on ­whether the right to truth should have temporal limitations and embeds this impor­tant question in related discussions on the rights to justice and to reparations for victims. ­Here, Schabas seems to find common ground with Carol Gluck’s observation that addressing the past is not the province of the law, by asking that we consider ­whether the search for the historical truth might not be better suited for nonjudicial institutions. In his chapter, “How Truth Recovery Can Benefit from a Conditional Amnesty,” Jeremy Sarkin explores another aspect of the right to truth. He examines the South African conditional amnesty pro­cess for lessons that can be learned for other pro­cesses in which amnesty could be given in exchange for the truth. He points out that the truth revealed in transitional justice pro­cesses is often limited, ­because truth commissions rely largely on victim testimonies. Sarkin argues that adding the perpetrators’ carefully scrutinized testimony regarding motive and per­for­mance would provide a more complete understanding. For example, such testimony could inform us of the claimed ­causes of vio­lence, the command structures, who was empowered to give the proximal ­orders, whence came the higher o­ rders, and what the connections w ­ ere between security officials and government personnel. This testimony, in turn, could facilitate our understanding of when and how states and individuals decide to perpetrate h­ uman rights violations. Sarkin promotes the benefits of conditional amnesty to enhance truth recovery, but cautions that its efficacy is dependent on the perpetrators’ perceived need for amnesty—­that is, their belief that without it, they would run the real risk of prosecution. The final contribution to this section emphasizes methodology. Stephan Parmentier, Mina Rauschenbach, and Maarten van Craen describe “New Epistemologies for Confronting International Crimes: Developing the Information, Dialogue, and Pro­cess (IDP) Approach to Transitional Justice.” The authors found that ­there was general agreement in both the public and private spheres that the past must be acknowledged by establishing the truth. However, the question of how to establish the truth has been contentious. A major reason for this contention is the coexistence of many truths, along with what their adherents and opponents would accept as credible methods of verification. To this end, the authors draw on several types of empirical research and quantitative surveys that they have designed to collect data on the attitudes, opinions, and expectations of populations affected by mass atrocities. Their survey aimed at understanding the extent to which establishing the truth was impor­tant to the respondents, and then by which means and pro­cesses they felt this should happen. The authors attempt to circumvent the biases created when transitional justice is managed by national or international elites by focussing

10

n a nc i a dler

their attention on local populations—an approach they label research “from below.” This has yielded insights into vari­ous strategies and mechanisms the respondents employed for dealing with the crimes of the past. The authors assert that quantitative surveys can provide an additional source of information on a range of transitional justice questions, including the victimization of persons during and ­after violent conflicts. They include specific issues related to postconflict justice, such as truth-­seeking, offender accountability, victim reparations, and reconciliation. They further claim that the nonadversarial act of asking a person for information constitutes a specific form of acknowl­edgment of his or her victimhood, even if that person may also belong to the group of offenders. Their approach supports the proposition that the act of facilitating a dialogue with a victim to gather information is validating and thus makes a contribution to transitional justice. Part II of this volume, “The Narrative of the Trial Rec­ord” addresses the kind of information that is revealed by the l­ egal pro­cesses transitional justice employs. In his examination of court transcripts in “The Spark for Genocide? Propaganda and Historical Narratives at International Criminal Tribunals,” Richard Wilson looks at how international courts from Nuremberg to the International Criminal Court have s­ haped our views of the origins and c­ auses of mass atrocities. The common thread ­running through the historical narratives the courts produce is the instigating role po­liti­cal propaganda plays in fomenting popu­lar participation in mass crimes. This is the explanation favored by judges to account for why neighbours turn against their neighbours, and Wilson argues that it has come to assume an “outsized role” in ­legal histories. Thus, even as Wilson describes how propaganda became one of the overarching narratives employed by international courts and expert witnesses to explain atrocities perpetrated during armed conflicts, he encourages scepticism by noting the paucity of empirical evidence to support this claim. Wilson’s examination of courtroom history as a site of contestation between ­legal actors pursuing ­legal outcomes draws our attention to the fundamental question regarding the kind of histories international courts produce. Furthering this discussion in “The International Criminal Trial Rec­ord as Historical Source,” Thijs Bouwknegt looks at the legacy of the Special Court for Sierra Leone and the International Criminal Tribunal for Rwanda. He evaluates the courts’ claims that they facilitate justice through their judicious use of extra-­ legal powers such as serving victims with closure, ending protracted wars, reconciling fractured socie­ties, and writing historical rec­ords. He focuses on the latter claim and addresses the prob­lems associated with the court rec­ord when Western models of justice are superimposed upon non-­Western cultures. Bouwknegt, like Wilson, grapples with the question of how to position court judgments in the larger historiography of mass vio­lence. He argues that t­ rials are more focussed on affirming theories than ascertaining truth and points to where

Introduction 11

some of the difficulties lie. Among the impediments that hobble most tribunals is the shortage of documentary, forensic, and other forms of tangible evidence. Another layer of unreliability is added for t­ rials connected with the oral socie­ties of sub-­Saharan Africa, where judges have to rely almost exlusively on eye-­and ear-­witness accounts of events that occurred five, ten, and even twenty years ago. Judges have recognized the questionable veracity of ­these testimonies that, even when honest, are merely the witnesses’ version of the truth. And while the trial transcript is a valuable source for historical research, it is, according to Bouwknegt, only a truncated rec­ord, as it cannot convey the full realities of the trial. It cannot, for example, convey the tonal effects of the witnesses or accusers, their beseeching looks or intimidating stares, or the effect of a closing statement delivered in song. In such circumstances, historians are tasked with reaching conclusions without the benefit of the courtroom context, as well as the potentially vital evidence not used in—or deliberately excluded from—­ the trial rec­ord. Part III, “The Afterlife of Transitional Justice Pro­cesses,” the final section of this volume, investigates some of the lingering tensions that transitional justice pro­cesses tried—­and often failed—to remedy. In “Narrating (In)Justice in the Form of a Reparation Claim: Bottom-­Up Reflections on a Post­colonial Setting,” Nicole Immler looks at the difficult, and sometimes fraught translation of transitional justice mechanisms from global to local scales. In 2011, the Dutch government was obliged to apologize and pay compensation for its repressive colonial policies during the Indonesia war of in­de­pen­dence (1945–1949). Immler’s case study, with its focus on the court-­ordered compensation granted to some of the ­widows of the Dutch military’s 1947 massacre in the village of Rawagede in Indonesia, brings the decolonization debate squarely into the transitional justice arena. The plaintiffs’ victory in this civil court case in The Hague set a pre­ce­dent that led to the filing of additional claims against the Dutch government, including ­those from the victims’ ­children. Much research regarding the Dutch military’s ­human rights violations in Indonesia in t­ hose years is ongoing,31 and much more is on the horizon.32 Initially, it seemed that transitional justice could mark the Rawagede ruling as a success ­because it had obtained acknowl­edgment from the repressor and reparations for the identified victims. However, as Immler explains, ­because local tradition required that the reparations earmarked for the ­widows be distributed to the ­whole village, the designated victims received only a small fraction of the compensation. This case illustrates some of the unanticipated complexities that arise when Western models are superimposed upon non-­Western cultures. In so ­doing, Immler’s investigation calls attention to the need to incorporate a deeper understanding of the facts on the ground into transitional justice approaches. Christian Axboe Nielsen’s chapter, “Collective and Competitive Victimhood as Identity in the Former Yugo­slavia,” describes how exclusivist narratives of

12

n a nc i a dler

victimhood impede postconflict, post–­transitional justice reconciliation. When a nation portrays itself in its politics, media, and academia as a victim, it seems self-­evident that the burden of postconflict reconciliation should be borne by the perceived perpetrators and their descendants. This ethos of repression recycles old, competing narratives of victimhood that confound the many transitional justice initiatives that have been implemented in the region to foster agency for confronting the past. Nielsen points to the “very high concentration” of repressive regimes, collective vio­lence, and mass atrocities that characterized the former Yugo­slavia between the years 1918 and 1991, only to be followed by a destructive, protracted state collapse. Each of the multiple communities that w ­ ere cobbled together for most of the twentieth ­century has its own memory of the violent events, and each asserts its primary victimhood, ongoing suffering, and martyrdom. Nielsen reminds us that ­these entrenched views have reinforced the “we”-­“they” divide in the region and caused re­sis­tance to the findings and rulings of the ICTY and other transitional justice mechanisms. On the other side of the globe, in the aftermath of a genocide, postconflict Cambodian society has narrowed the “we”-­“they” divide so much that nearly every­one is considered a victim. Timothy Williams’s chapter, “Perpetrator-­ Victims: How Universal Victimhood in Cambodia Impacts Transitional Justice Mea­sures,” explores the ways in which even former cadres of the Khmer Rouge, who had implemented genocidal policies, have become categorized as victims. They are well integrated into their communities, and the chief transitional justice mechanism in Cambodia—­the Extraordinary Chambers in the Courts of Cambodia (ECCC)—­has largely left them alone. As the hybrid tribunal aims at trying only the most responsible, highest leaders, the notions of mass victimhood and narrow perpetrator responsibility have taken even stronger hold in societal perceptions. The narrative of universal victimhood is further reinforced in memorial museum exhibitions. Williams believes that this compartmentalization of responsibility has facilitated stable and harmonious community relations in a society where former victims and former perpetrators often live side by side. He cautions, however, that by universalizing victimhood and downplaying agency, potential perpetrators of vio­lence in the ­future ­will not feel the need to take responsibility for their actions, which may in turn lead to them to the path of least resistance—­namely, participation. Fi­nally, the last chapter of this volume raises the question of how collective memory is s­ haped in the aftermath of collective crimes. It addresses the two core issues of who composed a given story and for what purpose. In “Collective Crimes, Collective Memory, and Transitional Justice in Bangladesh,” Kjell Anderson examines the multiple, contested memories of collective crimes committed during the “Bangladesh Liberation War” in 1971 and evaluates how transitional justice mechanisms such as the International Crimes Tribunal of Bangladesh

Introduction 13

framed and contextualized individual acts of vio­lence. Anderson calls attention to the moral priorities embedded in the officially favored memories that circulated in the aftermath of the collective crimes that occurred in Bangladesh. He argues that the tribunal may not have contributed to social solidarity, but rather fostered polarization by reinforcing the narratives perpetrators assert regarding their victimization. Anderson faults the transitional justice pro­cess in Bangladesh for promoting a par­tic­ul­ar narrative of the 1971 war, which facilitated e­ ither the denial of past atrocities or the assignment of collective blame to entire groups rather than individuals. In this way, he concludes, the ­trials may have contributed to deepening societal divisions rather than rebuilding a “moral community.” As a result, transitional justice failed to break the cycle of vio­lence. Rather it may have increased the risk of retaliation from t­ hose who felt alienated by the reinforcement of the state-­sponsored narrative coming out of the International Crimes Tribunal of Bangladesh.

reflection: how to frame the past This volume’s approach to history, historians, and transitional justice recognizes but also deepens William Faulkner’s admonition that “the past is not dead, it ­isn’t even past.” Considering the cases that ­were scrutinized in this collection, it seems transitional justice raises a number of issues and questions about the past that Faulkner overlooked. First, what actually happened in the past—it is not self-­ evident. This point is well-­captured by a post-­Soviet joke that observes, “It’s easy to talk about the pres­ent and the f­ uture, but the past keeps changing ­every day;” Second, how has this past been framed and interpreted by victims, repressors, and the amalgam of victim-­repressors that a totalitarian state produces to sustain itself? Third, how should we understand the role of volition when we recognize the power of repressive regimes to influence ordinary ­people to become repressors, and how do we assess the moral and ­legal issues surrounding contested culpability? Fourth, can we promote inclusiveness and reconciliation by re-­viewing competing narratives of repression as contributing narratives that could improve our understanding of the ­causes and consequences of repression? Fifth, what steps could be taken by transitional justice ­toward remediating irremediable harms? And lastly, how can we transform this narrative of events into a lessons-­learnable accurate history? The informed deliberations in this collection lead us to suggest that historians may be an underutilized tool in the repertoire of transitional justice mechanisms that could provide missing, perhaps invisible linkages. Systematic historical analy­sis could facilitate the move beyond competing narratives to the identification of what t­ hese themed stories tell us about how individuals and regimes perceived and recounted the repression. The discovery of impediments to

14

n a nc i a dler

transitional justice is not limited to the explicit stories that may be recounted in the courts and to the commissions, but can also be inferred from the implicit “truths” embedded between the lines of ­these narratives, as well as from viewpoints that have been silenced. For example, one of my research proj­ects involved interviews with Gulag survivors who had been illegally imprisoned, but remained loyal Communists. Counterintuitively, their unjust treatment by the Communist system was not experienced as disconfirming evidence, nor did it undermine their belief in the success of Communism, b­ ecause their belief was affirmed by the unrivaled meaning that a belief in Communism provided in their lives.33 Our work supports the social, scholarly, and po­liti­cal necessity of opening classified archives; collecting, contextualizing, properly placing, and managing the rec­ords produced by and used in international tribunals; and collating disparate official and personal narratives. Together, such materials could facilitate the “shared custody”34 of an inclusive “common past,” based on credible evidence and validated by informed public audiences.35 The instrumental role historians should play in transitional justice pro­cesses has been eloquently recognized by one prominent Rus­sian h­ uman rights defender: “Society and the state w ­ ill need to work together, and historians bear a special responsibility in this pro­cess.”36 The “never-­again” responses that follow mass atrocities rarely prevent recurrence b­ ecause they have not grasped what it is that should not recur. It is our hope that the reflections in this volume ­will energize the ongoing dialogue between transitional justice and seemingly inevitable injustices in the twenty-­first ­century. Transitional justice ­will necessarily be too ­little and too late, but a “lessons-­ learned” history and understanding of repression that it could generate has the potential to influence social practices that prevent—­rather than attempt to repair—­irreparable harms.

notes 1. ​See, among ­others, Ruti G. Teitel, “Transitional Justice Genealogy,” Harvard ­Human Rights

Journal 16 (Spring 2003): 69–94; Samuel P. Huntington, The Third Wave: Democ­ratization in the Late Twentieth C ­ entury (Norman: University of Oklahoma Press, 1991); Alexandra Barahona De Brito, Carmen Gonzalez Enriquez, and Paloma Aguilar, The Politics of Memory: Transitional Justice in Demo­cratizing Socie­ties (Oxford: Oxford University Press, 2001); Alexander Laban Hinton, ed., Transitional Justice: Global Mechanisms and Local Realities ­after Genocide and Mass Vio­lence (New Brunswick, N.J.: Rutgers University Press, 2010). 2. ​Pablo de Greiff, “Opening Remarks” (via Skype), International Conference on “Understanding the Age of Transitional Justice,” Film House, The Hague, December 9, 2015. 3. ​Pablo de Greiff, “Report of the Special Rapporteur on the Promotion of Truth, Justice, Reparation and Guarantees of Non-­Recurrence,” ­Human Rights Council, Thirtieth Session, United Nations, New York, September 4, 2015, 8. 4. ​Pablo de Greiff, “High-­Level Policy Dialogue on Guarantees of Non-­Recurrence—­From Aspiration to Policy: Challenges and Lessons in Preventing Mass Violations,” Ministry of Foreign Affairs, Stockholm, October 15, 2015.

Introduction 15 5. ​Rosemary Nagy, “Reconciliation in Post-­Commission South Africa: Thick and Thin

Accounts of Solidarity,” Canadian Journal of Po­liti­cal Science 35, no. 2 ( June 2002): 345. 6. ​Laurel E. Fletcher, Harvey M. Weinstein, and Jamie Rowen, “Context, Timing and the Dynamics of Transitional Justice: A Historical Perspective,” ­Human Rights Quarterly 31, no. 1 (February 2009): 166–169. 7. ​See UC Berkeley School of Law H ­ uman Rights Center, The Victims’Court? A Study of 622 Victim Participants at the International Criminal Court (Berkeley, Calif.: ­Human Rights Center, 2015), https://­www​.­law​.­berkeley​.­edu​/­w p​-­content​/­uploads​/­2015​/­04​/­VP​_­report​_­2015​_­final​ _­full2​.­pdf; Steven J. Stern and Scott Straus, eds., The H ­ uman Rights Paradox: Universality and Its Discontents (Madison: University of Wisconsin Press, 2014), 107–124. 8. ​De Greiff, “Opening Remarks.” 9. ​Susan Dwyer, “Reconciliation for Realists,” Ethics and International Affairs 13, no.  1 (March 1999): 81–98; Erin Daly, “Truth Skepticism: An Inquiry into the Value of Truth in Times of Transition,” International Journal of Transitional Justice 2, no. 1 (March 2008): 23–41; Nenad Dimitrijevic, “Justice beyond Blame: Moral Justification of (the Idea of) a Truth Commission,” Journal of Conflict Resolution 50, no. 3 ( June 2006): 268–382; Fiona C. Ross, “On Having a Voice and Being Heard: Some After-­Effects of Testifying before the South African Truth and Reconciliation Commission,” Anthropological Theory 3, no.  3 (September  2003): 325–341; Eugenia Zorbas, “What Does Reconciliation a­ fter Genocide Mean? Public Transcripts and Hidden Transcripts in Post-­Genocide Rwanda,” Journal of Genocide Research 11, no.  1 (2009): 127–147; Richard Ashby Wilson, “Judging History: The Historical Rec­ord of the International Criminal Tribunal for the Former Yugo­slavia,” ­Human Rights Quarterly 27, no. 3 (August 2005): 908–943. 10. ​Nanci Adler, “The ‘Bright Past,’ or Whose (Hi)story? Challenges in Rus­sia and Serbia ­Today,” Philosophy and Society 23, no. 4 (2012): 119–138. 11. ​See the discussion in Wilson, “Judging History,” 912, 922, 924. 12. ​See Nigel Eltringham, “ ‘ We Are Not a Truth Commission’: Fragmented Narratives and the Historical Rec­ord at the International Criminal Tribunal for Rwanda,” Journal of Genocide Research 11, no. 1 (2009): 59–60; see also Wilson, “Judging History.” 13. ​See, for example, Nanci Adler, Keeping Faith with the Party: Communist Believers Return from the Gulag” (Bloomington: Indiana University Press, 2012). 14. ​See Daly, “Truth Skepticism.” 15. ​See Elazar Barkan, “Truth and Reconciliation in History,” American Historical Review 114, no. 4 (October 2009): 899–913; see also Charles S. Maier’s thought-­provoking discussion, “­Doing History, ­Doing Justice: The Narrative of the Historian and of the Truth Commission,” in Truth v. Justice: The Morality of Truth Commissions, ed. Robert I. Rotberg and Denis Thompson (Prince­ton, N.J.: Prince­ton University Press, 2000), 261–278. 16. ​This, despite the efforts of the International Criminal Tribunal for the Former Yugo­slavia to set the rec­ord straight, overcome ambiguity, and “police a violent past,” as noted in Roland Kostic, Ambivalent Peace: External Peacebuilding, Threatened Identity and Reconciliation in Bosnia and Herzegovina, Report No. 78 (Uppsala: Department of Peace and Conflict Research, 2007), 33. See also Sonja Biserko, contribution to panel discussion “Letting Go of the Past, Losing a Hold on the F ­ uture,” Dudok, The Hague, February 11, 2008; Jelena Obradovic-­Wochnik, Ethnic Conflict and War Crimes in the Balkans: The Narratives of Denial in Post-­Conflict Serbia (London: I. B. Taurus, 2013). 17. ​Kostic, Ambivalent Peace, 256, 259. See also Selma Leydesdorff, Surviving the Bosnian Genocide: The ­Women of Srebrenica Speak (Bloomington: Indiana University Press, 2011). 18. ​See, among ­others, Impunity Watch, Action to Combat Impunity in Serbia: Options and Obstacles (Utrecht: Impunity Watch, 2008), 25, 61, 71. On cogent but irreconcilable narratives,

16

n a nc i a dler

see Wilson, “Judging History,” 908–942; Timothy Garton Ash, “A Nation in Denial,” Guardian, March 7, 2002; Jelena Obradovic-­Wochnik, “Knowledge, Acknowledgement and Denial in Serbia’s Responses to the Srebrenica Massacre,” Journal of Con­temporary Eu­ro­pean Studies 17, no. 1 (2009): 63–64. Regarding participation and apathy among the younger generation, see Jessica Greenberg, “ ‘­There’s Nothing Anyone Can Do about It’: Participation, Apathy, and ‘Successful’ Demo­cratic Transition in Postsocialist Serbia,” Slavic Review 69, no. 1 (Spring 2010): 41–64, particularly 54 and 63; Aryeh Neier, “Fifteen Years of Pro­gress,” International Herald Tribune, May 13, 2008. 19. ​See Jelena Grujic, “Reactions to Statements from Newly Elected President of Serbia, Tomislav Nikolic,” RECOM Initiative !Voice 7 (2012): 12–14. 20. ​See also Paul B. Miller, “Contested Memories: The Bosnian Genocide in Serb and Muslim Minds,” Journal of Genocide Research 8, no.  3 (September  2006): 317, 320–321; Obradovic-­ Wochnik, “Knowledge, Acknowledgement,” 64. In this climate of impunity, Mladić remained at large for fifteen years, and enraged protestors rallied in Belgrade’s streets in opposition to the 2008 extradition of Karadzic; see Dan Bilefsky, “Karadzic Sent to Hague for Trial Despite Violent Protest by Loyalists,” International Herald Tribune, July 30, 2008. The salience of a “culture of impunity,” also evidenced by recurrent incidents of po­liti­cal murders in Rus­sia, remains a fundamental impediment to transitional justice efforts. 21. ​Nigel Eltringham, “The Past Is Elsewhere: The Paradoxes of Proscribing Ethnicity in Post-­ Genocide Rwanda,” in Remaking Rwanda: State Building and H ­ uman Rights a­ fter Mass Vio­lence, ed. Scott Straus and Lars Waldorf (Madison: University of Wisconsin Press, 2011), 269–282. See also Timothy Longman, “Placing Genocide in Context: Research Priorities for the Rwandan Genocide,” Journal of Genocide Research 6, no. 1 (March 2004): 29–45. 22. ​Makau Mutua, “What Is the ­Future of Transitional Justice?,” International Journal of Transitional Justice 9, no. 1 (March 2015): 5. 23. ​That is, ethnicity was not the tradition in t­ hese class-­and clan-­based socie­ties; it was imposed by the Eu­ro­pe­ans in the scramble for and divide of Africa. See Eltringham, “The Past Is Elsewhere,” 269–282. See also Longman, “Placing Genocide in Context.” 24. ​The opening ceremony of the twentieth-­anniversary commemoration of the genocide in Kigali was carefully choreographed to reinforce this narrative. 25. ​Cody Corliss, “Truth Commissions and the Limits of Restorative Justice: Lessons Learned in South Africa’s Cradock Four Case,” Michigan State International Law Review 23, no. 2 (2013): 273–299; Khulumani Support Group, “Cradock 4 W ­ idows to Test NPA Ruling,” Johannesburg, July  20, 2007, http://­www​.­khulumani​.­net​/­khulumani​/­in​-­the​-­news​/­item​/­47​-­cradock​ -­four​-­widows​-­to​-­test​-­npa​-­ruling​.­html, accessed December 14, 2016. 26. ​For example, South Africa’s Truth and Reconciliation Commission laid the basis for an inclusive view of the traumatic past. However, it was criticized for having a po­liti­cal agenda that affected the statements made by victims and perpetrators alike. In its aftermath, the per­sis­ tence of competing narratives undermined some of its successes. See Molly Andrews, Shaping History: Narratives of Po­liti­cal Change (Cambridge: Cambridge University Press, 2007), 161. 27. ​Carol Gluck, Concluding Remarks, International Conference on “Understanding the Age of Transitional Justice: Narratives in Historical Perspective,” Peace Palace, The Hague, December 10, 2015. 28. ​Farida Shaheed, “High-­Level Policy Dialogue on Guarantees of Non-­Recurrence—­From Aspiration to Policy: Challenges and Lessons in Preventing Mass Violations,” Ministry of Foreign Affairs, Stockholm, October 15, 2015. 29. ​Eric Ketelaar, “Truth, Memories, and Histories in the Archives of the ICTY,” paper presented at the “Sixty Years Genocide Convention” conference, The Hague, December 8, 2008.

Introduction 17 30. ​Kate Doyle, “High-­Level Policy Dialogue on Guarantees of Non-­Recurrence—­From Aspi-

ration to Policy: Challenges and Lessons in Preventing Mass Violations,” Ministry of Foreign Affairs, Stockholm, October 15, 2015. 31. ​See Remy Limpach, De brandende kampongs van Generaal Spoor (Amsterdam: Boom Uitgevers, 2016); Gert Oostindie, Soldaat in Indonesië 1945–1950 (Amsterdam: Prometheus/Bert Bakker, 2015); Peter Romijn, Der lange Krieg der Niederlande: Besatzung, Gewalt und Neuorientorientiering in den vierziger Jahren (Göttingen: Wallstein Verlag, 2017). 32. ​In 2017, the NIOD Institute for War, Holocaust and Genocide Studies, embarked on a four-­ year, in­de­pen­dent, state-­funded collaborative research proj­ect with two other Dutch institutions to investigate the structural vio­lence committed by Dutch soldiers during the war on decolonization in the former Dutch East Indies; see Leiden University, “4.1 Million for Study on Dutch East Indies War of Decolonisation,” March 2, 2017, https://­www​.­universiteitleiden​ .­nl​/­en​/­news​/­2017​/­02​/­4​.­1​-­million​-­for​-­study​-­on​-­decolonisationJoint. 33. ​See Adler, Keeping Faith; Adler, “Enduring Repression: Narratives of Loyalty to the Party before, during, and ­after the Gulag,” Europe-­Asia Studies 62, no. 2 (March 2010): 211–234. 34. ​Ketelaar, “Truth, Memories, and Histories,” 13; Eric Ketelaar, “Archives as Spaces of Memory,” Journal of the Society of Archivists 29, no. 1 (2008): 9–27; Roland Kostic, “Nationbuilding as an Instrument of Peace? Exploring Local Attitudes ­toward International Nationbuilding and Reconciliation in Bosnia and Herzegovina,” Civil Wars 10, no. 4 (2008): 394–395. 35. ​For a discussion of unified narratives and national belonging, see Kay Schaffer and Sidonie Smith, “Conjunctions: Life Narratives in the Field of H ­ uman Rights,” Biography 27, no.  1 (2004): 6, 13, 19, 21. On truth production by ruling elites in post-­repressive states, see Brian Grodsky, “Producing Truth: The Politics of Investigating Past ­Human Rights Violations in Post-­Communist States,” World Affairs 169, no. 3 (Winter 2007): 125–133. 36. ​Arsenii Roginskii, “Pamiat’o Stalinizme,” in Istoriia stalinizma: itogi i prob­lem izucheniia, ed. E. Iu. Kandrashina (Moscow: ROSSPEN, 2011), 21–27.

1 • SWINGING THE PENDULUM Fin-­de-­Siècle Historians in the Courts V L A D I M I R PETROV I Ć

The extraordinary surge of ­human rights transgressions and the simultaneous rise in awareness about their disastrous effects remain one of the central paradoxes of our times. Torn between Auschwitz and Nuremberg, described both as the age of extremes and an age of transitional justice, the de­cades since World War II have been marked by an unpre­ce­dented demand to address historical wrongs. ­W hether through academic debates and truth commissions or lustration procedures and criminal ­trials, our encounters with the past invariably appear in a complex high-­profile context in which law, politics, history, and memory are intertwined. This tectonic shift has affected a number of disciplines that traditionally deal with the past, challenging their prac­ti­tion­ers to rethink the purpose of their work. The historical profession especially has been engaged in considering the implications of this change, which calls on historians to take an ethical stand regarding the atrocious past. A once marginal topic has achieved undisputed importance: “We are all historians of ­human rights,” proclaimed Linda Kerber, the president of the American Historical Association in 2006.1 However, such an understanding hardly translates into a consensus on the professional responsibilities of historians. Some have readily taken part in proceedings aimed at undoing the consequences of historical injustices, whereas ­others have expressed concerns. “Being an expert witness in a judicial case allows historians to use their historical training and historical judgment to benefit society,” wrote Charles Bolton, who took the stand in the 1981–1982 trial on creationism in Arkansas public schools. For Morgan Kousser, who at that time was starting his long involvement as a historical expert in the voting rights t­ rials, testifying was an opportunity to “do justice and scholarship at the same time.”2 ­Others, such as Henry Rousso, refused to perform such a role on the basis of a dif­fer­ent set of ethical considerations: “In my soul and conscience, I believe that historians cannot be ‘witnesses’ and that the role of ‘expert witness’ rather poorly suits the 21

22

v l a di mir petrov ić

rules and objective of a court trial. It is one t­ hing to try to understand history in the context of a research proj­ect or course lesson, with the intellectual freedom that such activities presuppose: it is quite another to try to do so ­under oath when an individual’s fate hangs in the balance.”3 The rich lit­er­a­ture devoted to the troubled relation between historical truth and ­legal justice confirms t­hese impressions of the polarized nature of the debate—­the pendulum does, indeed, swing very sharply.4 However, it remains largely unclear when the swing started. Historical expert witnessing, as David J. Rothman reminds us, grew prominent with the end of the Cold War, but “in fact it has been a common occurrence for some fifty years.”5 Hal Rothman concurs: “Since the 1950s, historians have been active in l­egal proceedings as experts and adversaries.”6 Indeed, it is difficult to overlook the importance of the postwar wave of war crimes ­trials or the desegregation efforts in the United States and the Native American claims cases appearing around the same time. However, obscured beneath ­these layers is impor­tant prehistory, as historians have navigated ­these murky ­waters at least since the end of the nineteenth ­century. The purpose of this chapter, then, is to unearth and bring into the discussion early instances of historical interventions into ­legal proceedings in order to highlight a structural and enduring (if ambiguous) connection between ­these disciplines in the pro­cess of confronting the past abuses of power.

history and the law How far does one go back into the troubled relations between history and law? It is worth noting that their complex interdependence can be traced back to their emergence in antiquity. The ancient Greek word for historian, ίστωρ (hístōr), denotes “the one who knows,” hence the expert, and in some contexts, even investigator and judge.7 This bond was also rediscovered during the Re­nais­sance: “Historical studies must be placed upon a solid foundation of law, and jurisprudence must be joined to history,” wrote François Baudouin in 1561.8 Ever since, ­there has hardly been a shortage of references to the proximity between t­ hese areas of ­human interest. In 1814, Joseph von Hormayr, for one, was praising “history, that pure two-­sided mirror of judgments past and judgments still to come, that inexhaustible source of universally beneficial experience, and incorruptible judge of ­those who have no other judge.”9 A similarly power­ful understanding was immortalized in Schiller’s and Hegel’s vision of Weltgeschichte als Weltgericht. This vision persisted in frequent, if implicit meta­phors of the tribunal of history or the judgment of history. However, ­these parallels ­were worlds apart from the aspirations of the found­ ers of historiography. “To history has been assigned the office of judging the past, of instructing the pres­ent for the benefit of ­future ages. To such high offices this work does not aspire: It wants only to show what actually happened [wie es



Swinging the Pendulum 23

eigentlich gewesen],” wrote Leopold von Ranke in his much-­quoted dictum, which has guided generations of prac­ti­tion­ers.10 His professional ideal, as described by Peter Novick, envisaged “the objective historian’s role” to that of “a neutral, or disinterested, judge; it must never degenerate into that of advocate or, even worse, propagandist. The historian’s conclusions are expected to display the standard judicial qualities of balance and evenhandedness. As with judiciary, t­ hese qualities are guarded by the insulation of the historical profession from social pressure or po­liti­cal influence, and by the individual historian avoiding partisanship or bias.”11 ­Lawyers seemed to concur. The ­legal value of historical arguments was challenged by as persuasive an authority as Frederic William Maitland, whose inaugural lecture at Cambridge University in 1888 offered a warning against the temptation to “mix up two dif­fer­ent logics, the logic of authority, and the logic of evidence. What the l­ awyer wants is authority and the newer the better; what the historian wants is evidence and the older the better. . . . ​A mixture of l­ egal dogma and ­legal history is in general an unsatisfactory compound. I do not say that ­there are not judgments and textbooks which have achieved the difficult task of combining the results of deep historical research with a luminous and accurate exposition of existing law—­neither confounding the dogma nor perverting the history; but the task is difficult.”12 In the view of the found­ers, for historiography to become a science, it had to be based on the meticulous, chronologically distanced investigation of as many available sources as pos­si­ble with the purpose of creating an objective account of past events. Lord Acton proclaimed in his famous 1895 inaugural speech “On the Study of History” that “history, to be above evasion or dispute, must stand on documents, not on opinions.” This was a path for historians to move from the biased role of hanging judge to becoming “an accepted tribunal, and the same for all.”13 The availability of sources and their impartial analy­sis was to be secured by a detachment created through the temporal remoteness between the researcher and the topic—­the famed historical distance. This detachment was supposed to result in a shielding gap between historians and the l­egal and po­liti­cal concerns of con­temporary society. The influential methodological textbook of Charles Langlois and Charles Seignobos echoed this idea: “It is an obsolete illusion to suppose that history supplies information of practical utility in the conduct of life.”14 However, the ink was still drying on their textbook as their ­recipe was to be tested in the most dramatic ­legal event that was shaking the French public and shaping its po­liti­cal and intellectual landscape—­the Dreyfus affair.

24

v l a di mir petrov ić

the dreyfus affair: historians and the protection of ­human rights Universally acknowledged as a major watershed in con­temporary French history, the Dreyfus affair was launched in 1894 with the allegation that confidential information from the French military general staff was leaked to the German embassy. The affair took off as the fin­ger of accusation was pointed ­toward Captain Alfred Dreyfus, a French artillery officer of Jewish origin. By the end of the year, he was arrested, tried for treason by a military tribunal, and sentenced to life imprisonment.15 However, such a prompt rendering of justice, surrounded by chauvinistic and anti-­Semitic campaigning, raised suspicions of foul play. It became apparent that Dreyfus had not been convicted on solid evidence, that the trial had not been transparent, and that he had been scapegoated and even framed by the top echelons of the French military. Hence the reopening of his case became a rallying cry for a group of intellectuals assembled around the newly formed Ligue des droits de l’homme (­Human Rights League), who pursued the case with growing zeal. In the ­legal quagmire surrounding the arrest of Dreyfus, written evidence of dubious authenticity had been hastily examined by inadequate specialists.16 Many grew convinced that the crucial documents, among ­others the infamous bordereau (a memo) found at the German embassy, ­were ­either forged or falsely attributed to Dreyfus. When Emile Zola started writing his articles in late 1897, culminating with the famous “J’accuse!,” he warned the French president that “France has this stain on its cheek. History ­will write that it is ­under your presidency that such social crime could be committed.”17 Zola hoped that not only history, but also historians would support him. Promptly accused of defaming the military and directed to appear in court, he wrote to a number of historians that he “would cite as witnesses before the jury of the Seine, scholars, paleographers, historians accustomed to ­these scientific methods.”18 His chances of finding allies in t­ hese quarters w ­ ere good. But Zola was not the first intellectual to speak publicly for Dreyfus. He was preceded by three weeks by Gabriel Monod (1844–1912), one of the most distinguished French historians, famed medievalist, and founder of a chair in history at the Ecole des hautes études. Monod was also creator of Revue historique. In his 1876 opening article of this venerable periodical, he wrote that “history, without assumptions to any profit apart from the one that we draw from the truth works in a discrete and certain manner ­toward the greatness of the nation, and at the same time t­ owards the pro­gress of the h­ uman race.”19 Two de­cades l­ater, he was challenged to translate this ideal into practice. In late 1897 Mathieu Dreyfus, the ­brother of the accused, showed letters with his ­brother’s handwriting to Monod, who compared them to the facsimile of the bordereau published in Le matin a year before. ­After initial graphological analy­sis Monod was convinced that Dreyfus could not be its author; he made this public in an open letter to Le temps,



Swinging the Pendulum 25

figure 1.1. Professor Gabriel Monod and Captain Alfred Dreyfus, 1910. (Photo © Ser­vice

des Archives de France, Mission aux commémorations nationale.)

thereby becoming the first among intellectuals to advocate for reconsideration of the case: “I have read carefully the facsimile of the bordereau attributed to Captain Dreyfus. . . . ​I believe I can determine that it was not written by his hand . . . ​and that he was a victim of most tragic judicial error. . . . ​I’d also add that the content of the document even increased my suspicion. I was ­silent u­ ntil

26

v l a di mir petrov ić

now . . . ​convinced that my personal conviction was not enough to secure the revision of the trial.”20 Together with Albert Sorel, his colleague from the French Institute, Monod joined the ranks of Dreyfusards, along with classicist Louis Havet, archeologist Salamon Reinach, chemist Emile Duclaux, poet Anatole France, educator Ferdinand Buisson, journalist Octave Mirbeau, librarian Lucien Herr, archivist Paul Viollet, essayist Edmond Fleg, and politician Léon Blum.21 As individuals with so many dif­fer­ent personalities, po­liti­cal convictions, and professions came together in an attempt to rectify what they perceived to be an injustice symptomatic of militaristic, chauvinistic, and clerical segments of French society, a vivid public debate gave rise not only to the Ligue des droits de l’homme, the first nongovernmental institution concerned with ­human rights, but also to the very concept of the public intellectual. Among them, historians ­were perhaps less publicly vis­i­ble, but legally they had the most relevant role. With graphology still in its cradle, it made sense to turn to historians, whose training included detecting forgeries in historical documents.22 As the case was about to be appealed, Monod started preparing his expert opinion for the Court of Cassation.23 Galvanized by Monod, many volunteered to serve as experts in several ­trials related to the affair. They included medievalists from Ecole des chartes, who enthusiastically responded to Zola’s invitation to offer their expertise at his defamation trial in early 1898. In addition, famed historians Arthur Giry and Paul Meyer and paleographer Auguste Molinier and his b­ rother, historian and Louvre curator Emile, applied the historical method to critique the sources of the disputed documents, arguing primarily that the author of the incriminating bordereau was not Dreyfus, but Major Ferdinand Esterhazy. Only Abel Lefranc, who worried he would not be seen as competent enough, did not accept Zola’s invitation.24 As the date of the trial approached, ­these figures also went public with their views, sometimes even forgetting necessary caution. As historian Ruth Harris writes, “Meyer, for example, told L’Aurore before Zola’s trial that even an ignoramus would be able to see that the writing on the bordereau belonged to Esterhazy, a statement that invited the question of why, in that case, expert skills ­were needed in court.”25 This provoked Robert de Lasteyrie, also an alumnus of Ecole des chartes and a member of Institute de France, to attack historians for testifying about a document “that they know only from a crude facsimile. . . . ​I can scarcely understand how they could have forgotten to such an extent all the critical traditions honored at the Ecole des chartes.” He also or­ga­nized a Manifesto of the Chartists published in L’Eclair on March  22, 1898, in order to ­counter the historians’ efforts. Still, ­those who intended to testify in the Zola ­trials ­were not easily deterred. Molinier claimed the “applicability of universal critical method: knowledge of thousands of documents from dif­fer­ent eras did indeed qualify someone to judge the bordereau.”26



Swinging the Pendulum 27

Historians exhibited the same determination on the stand. Arthur Giry, professor at Ecole des hautes études who testified in the Zola case, passionately asserted their professional authority over the documents: “The study and the composition of the text are naturally one impor­tant part in our branch of erudition; we teach our pupils to determine the age, the origin, and the provenance, to discern between au­then­tic, false, interpolated, and sincere documents. The method is always the same. It does not vary, b­ ecause we have such a par­tic­u­lar education that habituates us to observing in a text the most detailed particularities we can, as well as applying this method to the con­temporary text as much as to the Ancient text.”27 However, as persuasive as he was, he could not sway the judges, who w ­ ere not necessarily as impressed as his students. Paul Meyer, the director of the Ecole des chartes, received similar treatment from the presiding judge who reprimanded him for using notes. Meyer replied: “I am not reading. Mr. President, I am just a professor, and as I have a poor memory, I am always obliged to write down, in a sentence or two, the ideas I am about to convey.”28 At the very beginning of their expert role, historians w ­ ere to learn an impor­tant lesson: the courtroom was no classroom. Zola was convicted and fled to E ­ ngland before the verdict was rendered on his appeal. However, this development aroused more public sympathy for the revisionists and opened up the ave­nue for their original goal—­the review of the Dreyfus case. Historians ­were even blunter during this review, which was held from August to September 1899 in Rennes. Particularly, during the sixteenth session, held on August 30, Paul Meyer, August Molinier, and Arthur Giry had an opportunity to substantiate their argument that Esterhazy was the author of the bordereau.29 Emile Molinier also restated the arguments on the issue of the authenticity of the disputed documents, claiming on the basis of his professional authority that they could not possibly have been written by Dreyfus: “I w ­ ill allow myself to remind you on this topic that the rules of historical critique which I apply to the expertise of Commander Esterhazy’s paper are the same rules that historians have been applying to all forms of historical memoires during the previous centuries.” Challenging other authorities, he added: “You may perhaps tell me that I am not an expert. This is the third time that I have come to testify and the first in which I have offered my expertise in a case of writing. But if I allow myself to put this opinion in front of you with such a force, that is b­ ecause it is a result of very long personal study.”30 No less than ten graphological experts ­were again deposed at length, but the historians made a much better impression among the spectators.31 Jean Jaures wrote: “They ­were not divided as experts in the Dreyfus case, and they do not operate b­ ehind closed doors as did the experts in the Esterhazy case.”32 Young Julien Benda was even more impressed, highlighting the value of “patient historical research” and expressing “disgust with t­hose ‘men of letters’ who distorted reason to serve their po­liti­cal agenda. History and its rigor,” he asserted, “had been the discipline that turned him into a ‘natu­ral’ Dreyfusard.”33

28

v l a di mir petrov ić

In the short run, public reactions w ­ ere of ­little help, as the judges w ­ ere not so impressed. “The court listened to them with scrupulous but mistrustful attention, for it regards them primarily as ‘intellectuals,’ presumptuous pedants and as members of a cast who believe that they constitute an aristocracy of the intellect and are all more or less lost to patriotism,” observed Maurice Paleologue, a perceptive observer of the affair.34 Dreyfus was reconvicted and his sentence lowered to ten years. However, the tide was turning, as more evidence of Dreyfus’s innocence came to light, and ­after fierce public pressure, he was pardoned by the French president in 1899, fully rehabilitated by the court in 1906 and awarded the Legion d’Honneur and reintegrated into the French army in the same year. In the Dreyfus affair, historians contributed significantly to the undoing of the miscarriage of justice. As historian Isabel Noronha-­DiVanna writes, “Detectives at work, they thoroughly enjoyed showing the usefulness of their technique. . . . ​ It was a glorious time for t­hose men whose participation in po­liti­cal life so far had been so limited.”35 The rationale for their engagement was spelled out by Zola’s friend, one of the found­ers of the H ­ uman Rights League and one of the Sorbonne’s foremost professors of history, Gabriel Séailles: “Our teaching would lose all authority if we ­were not ready to confirm it by our deeds.”36 Through this confrontation with the most traditionalist parts of the French society, ­these intellectuals ­were also asserting their cosmopolitan attitudes and credentials. It is by no means a coincidence that Gabriel Monod was in the forefront of the initiative to or­ga­nize the International Historical Congress, the first structural cross-­border cooperation of historians. The inaugural meeting was held in The Hague in 1898 and, as Monod reported, with “no friction resulting from the presence of scholars of dif­fer­ent nationalities, languages and conceptions.”37 The first congress was held in Paris in 1900, with strong emphasis on comparative transnational historical research, which seemed to inaugurate broad international cooperation and underscore the emancipatory potential of studying of the past.

the friedjung affair: historians and the promotion of national interest In the Dreyfus affair, historians contributed significantly to the undoing of the miscarriage of justice by confronting disciplinary conservatism, nationalistic bias, and racial prejudice. However, such commitment was more an exception than the rule in this period. ­There ­were power­ful countertendencies. The jingoistic fervor blowing through Eu­rope was threatening to compromise the high standards of historical scholarship, as historians ­were tempted to put their skills and credentials in the ser­vice of the national cause. As Harold Berman reminds us, “The emergence in the nineteenth c­ entury of the so-­called scientific history, that is of systematic and painstaking research . . . ​coincided with the emergence



Swinging the Pendulum 29

of the most intense nationalism that Eu­rope had yet experienced. It was simply assumed that history meant national history.”38 As the fin-­de-­siècle period was coming to a close, the pendulum was decisively swinging in that direction. In late 1909, shortly a­ fter Dreyfus was rehabilitated, a libel suit was brought against well-­known Austrian historian Heinrich Friedjung (1851–1920).39 One of the first historians of the con­temporary period, Friedjung was the author of widely respected monographs on the recent history of Central Eu­rope. He entered po­liti­cal life as a pan-­German, but was excluded from the Greater German Party due to his Jewish origins. Eventually, he found a new cause in defending the Habsburg monarchy from perceived Slavic danger.40 To that end, he joined the campaign fought by a number of Vienna journals, which ­were closely aligned with the hawkish ele­ments in the Austro-­Hungarian elite. The aim was to discredit South Slavic politicians of the Empire, suspected of maintaining treasonous connections with Serbia to the detriment of Austria-­ Hungary and its monarch.41 Such articles appeared by the beginning of 1909 in the daily Reichpost, and ­were backed by Friedjung’s article “Österreich-­Ungarn und Serbien,” published in the Neue Freie Presse in March 1909. His essays presented an overview of con­temporary foreign policy, sprinkled with sensational claims and exclusive documents. In analyzing the impact of the change of Serbian dynasties and the implications for the region, Friedjung stressed that this change marked the beginning of instabilities caused by constant conspiracies launched against the Habsburg monarchy; ­these conspiracies w ­ ere helped by insiders in the monarchy, chiefly the representatives of the Croato-­Serbian co­ali­tion, presumed to have been receiving money from the Serbian government through the organ­ization “Slovene South,” whose ultimate purpose was the dismembering of the dual monarchy. His article was not only descriptive, but also prescriptive, stating that “it would be a cultural deed of the highest value if Austrian weapons would be ordered to annihilate the conspiratorial center in Belgrade and help the healthy ele­ments of Serbian society to prevail.”42 The warmongering article appeared at the peak of the crisis that erupted when Bosnia was annexed by Austria-­Hungary, bringing the Empire to the brink of war with Serbia, as well as with Rus­sia. As the crisis mounted, Miroslav Spalajković, official of the Serbian Ministry for Foreign Affairs whom Friedjung depicted as the key conspirator, reacted immediately. Denying the entire content of the article, Spalajković demanded “arbitration by experts on the documents.” Friedjung accepted the challenge and even raised the stakes, showing his readiness for no less than a trial: “While Mr. Spalajković has the right to deny every­thing even u­ nder the arbitrage, I am in a contrary position. . . . ​That is the difference between him, who was the soul of the war party against Austria-­Hungary in his Ministry, and me, as a historian who aspires to the truth.” Asked to provide evidence

30

v l a di mir petrov ić

for his accusations of the parliamentarians of the Croato-­Serbian co­ali­tion, Friedjung stated: “If I ­were to be called in front of the court, I would bring out my documentary material in its entirety in order to establish ­legal and historical facts.” He maintained that his article was based on a scholarly scrutiny, “just as any of the documents in my historical books. Having in mind the course of my studies, I believe that I am recognized for my ability to distinguish true historical sources and documents from false ones. I gather one ­will also acknowledge that I am not led by po­liti­cal passion and personal hatred in establishing the facts. Therefore I dare demand that my article is given full historical credibility.”43 Friedjung got what he called for, as fifty-­two parliamentarians sued him and the editors of Reichpost and Neue Freie Presse for libel. He had put his professional reputation at stake in guaranteeing the credibility of his interpretations and the reliability and authenticity of his sources. ­These he assembled in a volume, which he presented to a Landesgericht (Regional Court) in Vienna, where a jury trial was held between December 9 and 22, 1909.44 At the outset of the trial, which opened with his defense, Friedjung explained that “it is no business of the historian to reduce men’s words and deeds to the provisions of a penal code; his task is to examine documents, to establish facts and illustrate characters.” He finished his plea with utmost confidence: “Experts and historians are t­ here to give judgments if I did my job well. I think I was building on a sound ground, and let it be the base for jurors to reach their verdict.”45 However, it was exactly on this frontline that Friedjung started losing ground. In the course of the trial, it emerged that his essay was based on material that seemed to be confidential correspondence and documentation of high Serbian officials.46 Photographed and supplied to the Austrian legation in Belgrade through shady intelligence channels, they ­were taken at face value in Vienna due to their explosive character and ultimately their translations ­were leaked to Friedjung directly from the Austrian Foreign Office, on the instructions of its head, Count Ährenthal. Unable to read Serbian, Friedjung did not question the translations. As he said at the trial, “I was chosen to publish ­these protocols and reveal Serbian conspiracies as the intention was to use my name to elevate the importance of the publication. . . . ​I published the article as at the time the danger of war with Serbia was culminating. If it ­were not for that danger, I would not publish.”47 Asked to provide the originals of facsimiles of the documents at stake, he desperately attempted to obtain them, writing personally to Count Ährenthal and pleading that his professional ­career was at stake: “As in the article I explic­itly said that I have the documents at my disposal, I need them to cover myself as a historian. This is an impor­tant point which decides if the article is written by a historian or by a publicist. It has been only a few months since the Austrian Acad­emy of Sciences elected me as associate member, and t­here ­will be g­ reat surprise among their circles if I appear before jurors without convincing material.”48



Swinging the Pendulum 31

Despite the plea, he could pres­ent only three facsimiles and one transcription, whereas other documents ­were withheld by the Ministry, whose officials started doubting their authenticity and attempted to control po­liti­cal damage by cutting Friedjung lose. In real­ity, Ährenthal, who knew Friedjung from 1898 and assessed him to be a vain and naïve academic suitable for manipulation, used him accordingly.49 During the rest of the trial, witness a­ fter witness appeared, challenging the authenticity of the documents and calling attention to anachronisms in the text and inconsistencies of the content. Friedjung could only fume as he questioned them: “I hold it unheard of that witnesses from Serbia simply dismiss as mystifications and crude forgery all the documents used by the leading persons of this Monarchy in shaping policy ­toward Serbia.” Spalajković retorted: “­There are forgers all over the world, who deceive even politicians and ­people in the highest positions. This is what happened now as well, and I pity the victims of this deceit.”50 This view was supported by a Czechoslovak politician and public figure Tomáš Masaryk, for whom this case was yet another cause célèbre demonstrating pressure on the Slavic population by the Habsburg monarchy. He testified about his own inquiries in Belgrade into the nature of the documents, which he considered to be crude forgeries. In the final stage of the trial, this opinion was also supported through the expertise of Milan Reshetar (1860–1942), professor of Slavistics at Vienna University, and the court’s translator, Dr. Kraus. Their joint opinion related to a supposed copy of Spalajković’s report was read on December 20. Analyzing its grammar, diction, expressions, orthography, and style, they reached the conclusion that the author of the text had no thorough knowledge of the Serbian language and that the expressions used in the documents ­were highly aty­pi­cal for the vocabulary of Serbian diplomacy. They also claimed that the other documents submitted by Friedjung ­were likely to be forgeries. Another expert called by the court, Dr. Hans Übersberger (1877–1962), a reputable young historian and lecturer specializing in the history of Eastern Eu­rope at the University of Vienna, attempted to offer a counter-­version, insisting on “the methodological differences between Professor Reshetar and me. He is a philologist, I am a historian.”51 He was much more ready to consider the documents to be au­then­tic transcriptions—­but apparently without much success. The day ­after, the president of the court assisted in brokering a settlement between the parties, according to which the plaintiffs withdrew their charges in return for Professor Friedjung’s public apology and acknowl­edgment of the questionable nature of his sources. Even Friedjung’s admirers, like A.J.P. Taylor, who translated his seminal book into En­glish, ­were forced to conclude that “the contrast between Friedjung the historian and Friedjung the politician is at first sight overwhelming.”52 Indeed, Friedjung’s professional reputation was so badly bruised that even such a pro-­ Slav observer as historian R. W. Seton-­Watson felt compelled to console him ­after

32

v l a di mir petrov ić

the trial: “As a fellow soldier in an army of historians, I wanted to convey to my General that in my view the pro­cess did not damage your stature as a man and historian at the least. This cannot be said for t­hose who had put a historian of such a position in so a disgraceful manner.”53 However, in other writings Seton-­ Watson attributed this blunder to a par­tic­u­lar patriotic fever: “Not being in the position to defend his fatherland, sword in hand, he conceived it to be his plain duty as a historian and publicist, to place his pen at the ser­vice of Austria, and in so d­ oing he was only continuing his life’s work of strengthening the consciousness of his fellow citizens with an interpretation of their past history.”54 Such an attitude, which was not aty­pi­cal for historians of the time, and was becoming more extensive as the crisis leading to the First World War was mounting. According to Ernst Breisach, u­ ntil then most historians “agreed that history was an endeavor with the purely theoretical interest of reconstructing the past and without any practical interests, be it lessons, devotion, entertainment, or propaganda. Yet since 1914 adherents of both answers have confronted prob­ lems.”55 Older historians, like Heinrich Friedjung or Hans Übersberger, joined the ranks in writing jingoistic proposals for the postwar order, a­ fter the presumed victory of their country. Their younger colleagues went to the trenches.

new wine in old ­bottles: an enduring ethical challenge For historians across the front lines, the G ­ reat War hardly ended in 1918. The war­ time explosion of nationalist hysteria was followed by postwar resentment and bitterness. “The war posed a fundamental and sweeping challenge to the profession’s posture of disinterested objectivity,” observes Peter Novick.56 Indeed, world war was a sobering experience, and historians w ­ ere not spared its embittering effects. Many would stay entrenched in their prejudices long ­after the end of the First World War, fighting historiographical b­ attles about its c­ auses and consequences, which last to this very day.57 The interwar tensions, characterized by fervent nationalism, racism, anti-­Semitism, discrimination, and persecution, strengthened by economic difficulties, closures of national borders, and the building of authoritarian and totalitarian systems, contributed to the growing historiographical autarchy, which turned good intentions into false promises. It took another war of unparalleled devastation and de­cades of postwar reconstruction to build a consensus on the necessity to confront the legacy of a bad past. In that re­spect, torn between the protection of h­ uman rights and the promotion of national interest, the dynamics at play among fin-­de-­siècle historians ­were bound to reappear over and over during the age of extremes. This enduring ethical challenge gradually brought postwar historians to testify in the Holocaust-­ related ­trials in Germany, as well as in antidiscrimination cases in the United States during the 1950s and 1960s.58 Similar concerns are echoed in more con-



Swinging the Pendulum 33

temporaneous courtroom sagas, such as the Sears case, anti-­tobacco litigation, the Irving versus Lipstadt libel suit, and vari­ous other ­trials taking place in front of international criminal tribunals of our times.59 Yet, it is impor­tant to note that this explosion of historical forensic engagement is neither unconditional nor irreversible. The fragility of this ethical consensus regarding the obligations of historians is only too obvious in our own turn of the c­ entury, which commenced with the fall of the bipolar world order and the global advent of transitional justice mea­sures and ended in the smoke of terrorist attacks and protracted warfare on the ruins of failed states. ­These momentous events set a pendulum in motion again, bringing universalist commitments ­under the pressure of national allegiances. As old dilemmas are posed in new circumstances, historians are bound to partake and take a fresh look into their disciplinary pledges. “The historian would like to do justice; the judge must establish some version of history,” Charles Maier reminds us, adding a telling warning: “If good judges and historians shun t­ hese tasks, they w ­ ill be taken on by prejudiced or triumphalist ones.”60 Overcoming disciplinary conservatism and nationalistic bias is an ongoing challenge for a profession conceived as a cornerstone of the humanities, but not necessarily practiced according to that ideal. The only advantage con­temporary historians have is an opportunity to learn from the ­bitter experience of their pre­de­ces­sors.

notes 1. ​Linda K. Kerber, “We Are All Historians of H ­ uman Rights,” Perspectives: Newsmagazine of

the American Historical Association 44, no. 7 (October 2006): 3–4.

2. ​Charles Bolton, “The Historians as Expert Witnesses: Creationism in Arkansas,” Public His-

torian 4, no. 3 (Summer 1982): 59; Morgan J. Kousser, “Are Expert Witnesses Whores? Reflections on Objectivity in Scholarship and Expert Witnessing,” Public Historian 6, no. 1 (Winter 1984): 18. 3. ​Henry Rousso, The Haunting Past: History, Memory, and Justice in Con­temporary France (Philadelphia: University of Pennsylvania Press, 2002), 86. 4. ​Daniel A. Farber, “Adjudication of ­Things Past: Reflections on History as Evidence,” Hastings Law Journal 49, no. 4 (April 1998): 1009–1026; Reuel E. Schiler, “The Straw­horse­men of the Apocalypse: Relativism and the Historian as Expert Witness,” Hastings Law Journal 49, no. 4 (April 1998): 1169–1181; Carlo Ginzburg, The Judge and the Historian: Marginal Notes on a Late-­Twentieth-­Century Miscarriage of Justice (London: Verso, 1999); Norbert Frei, Dirk van Laak, and Michael Stolleis, eds., Geschichte vor Gericht: Historiker, Richter und suche nach Gerechtigkeit (Munich: Verlag C. H. Beck, 2000); Alain Wijffels, ed., History in Court: Historical Expertise and Methods in a Forensic Context (Leiden: Ius Deco Publications, 2001); Richard J. Evans, “History, Memory, and the Law: The Historian as Expert Witness,” History and Theory 41, no. 3 (October 2002): 326–345; Jonathan D. Martin, “Historians at the Gate: Accommodating Expert Testimony in Federal Courts,” New York University Law Review 78, no.  4 (October  2003): 1518–1543; Harriet Jones, Kjell Östberg, and Nico Randeraad, eds., Con­ temporary History on Trial: Eu­rope since 1989 and the Role of the Expert Historian (Manchester: Manchester University Press, 2007). Recently Public Historian 37, no. 1 (2015): 8–104, devoted

34

v l a di mir petrov ić

its forum to this topic. The latest addition to the growing lit­er­a­ture on historical expert witnessing is written by Ramses Delafontaine, Historians as Expert Judicial Witnesses in Tobacco Litigation: A Controversial ­Legal Practice (New York: Springer, 2015). 5. ​David J. Rothman, “Serving Clio and the Client: The Historian as Expert Witness,” Bulletin of the History of Medicine 77, no. 1 (Spring 2003): 25. 6. ​Hal K. Rothman, “Historian v. Historian: Interpreting the Past in the Courtroom,” Public Historian 15, no. 2 (Spring 1993): 39. 7. ​Henry George Liddell and Robert Scott, eds., A Greek-­English Lexicon (Oxford: Clarendon Press 1996), s.v. ίστωρ. See also Gerald Press, “History and the Development of the Idea of History in Antiquity,” History and Theory 16, no. 3 (October 1977): 280–296. See also François Hartog, “Le temoin et l’historien,” Gradhiva 27 (2000): 1–14. 8. ​Donald R. Kelley, Foundations of Modern Historical Scholarship: Language, Law, and History (New York: Columbia University Press, 1970), 116. 9. ​Joseph von Hormayr, “Austria and Germany,” in National Romanticism: The Formation of National Movements, ed. Balázs Trencsényi and Michal Kopeček (Budapest: Central Eu­ro­ pean University Press, 2006), 30. 10. ​Leopold von Ranke, Sämtliche Werke (Leipzig: Duncker und Humblot, 1885), 33:7. On the role of seminars of founding persons, see Michael Bentley, Modern Historiography: An Introduction (New York: Routledge, 1999), 71–79. The professionalization of historiography is scrutinized in a comparative perspective by Georg G. Iggers and Q. Edward Wanq, A Global History of Modern Historiography (London: Pearson, 2008), 128–133, and its propagation is treated by Rolf Torstendahl, The Rise and Propagation of Historical Professionalism (New York: Routledge, 2014), 3–67. 11. ​Peter Novick, That Noble Dream: The “Objectivity Question” and the American Historical Profession (Cambridge: Cambridge University Press, 1993), 2. 12. ​Frederic W. Maitland, “Why the History of En­glish Law Is Not Written,” in The Collected Papers of Frederic William Maitland, ed. H.A.L. Fisher (Cambridge: Cambridge University Press, 1991), 1:491–492. 13. ​John Edward Emerich Acton, Inaugural Lecture on the Study of History (London: Macmillan 1906), 16. 14. ​Charles V. Langlois and Charles Seignobos, Introduction to the Study of History (New York: Holt, 1898), 319. 15. ​The most complete set of transcripts of ­trials, other proceedings, and material related to the Dreyfus affair was reproduced in about 150 volumes published by Pierre-­Victor Stock. See Le pro­cess Dreyfus: Devant le conseile de guerre de Rennes, vols. 1–3 (Paris: P. V. Stock, 1900); La revision du procès Dreyfus: Enquête de la Cour de cassation (Paris: P. V. Stock, 1899). A comprehensive se­lection of the documents is available at Dreyfus Rehabilitated: France 1906, http://­ www​.­dreyfus​.­culture​.­fr​/­en​/.­ A comprehensive overview of the events is provided by George R. Whyte, The Dreyfus Affair: A Chronological History (London: Palgrave, 2008). 16. ​Louis Begley, Why the Dreyfus Affair ­Matters (New Haven, Conn.: Yale University Press, 2009), 98–99. 17. ​Emile Zola, “J’accuse,” L’Aurore, January 13, 1898, 1. More on Zola’s intervention can be found in Lida Maxwell, Public ­Trials: Burke, Zola, Arendt and the Politics of Lost C ­ auses (Oxford: Oxford University Press, 2014), 81–95. 18. ​Laurent Ferri, “Emile Zola et ‘ces messieurs de l’école des chartes’ dans l’affaire Dreyfus: Documents inédits,” Bibliothèque de l’école des chartes 164 (2006): 598. 19. ​G. Monod, “Du progrès des études historiques en France depuis le XVIe siècle,” Revue historique 1 (1876): 5–38.



Swinging the Pendulum 35

20. ​“Le 2 novembre 1897  l’historien Gabriel Monod analyse l’écriture de Dreyfus et défend

l’honorabilité du capitaine Dreyfus,” Le Temps, November 6, 1897, repr. in Dreyfus Rehabilitated, http://­www​.­dreyfus​.­culture​.­fr​/­fr​/­mediatheque​/­media​-­type1​-­html​-­Engagement​_­dreyfusard​ _­de​_­G​.­​_­Monod​.­htm. 21. ​On the role of intellectuals in the affair see Vincent Duclert, “De l’engagement des savants à l’intellectuel critique: Une histoire intellectuelle de l’affaire Dreyfus,” Historical Reflections / Réflexions historiques 24, no. 1 (Spring 1998): 25–62. 22. ​More on the state of French historiography around the time of the affair can be found in Pim den Boer, History as a Profession: The Study of History in France 1818–1914 (Prince­ton, N.J.: Prince­ton University Press, 1998), 224–359. On the role of historians in the Dreyfus case, see Isabel Noronha-­DiVanna, Writing History in the Third Republic (Cambridge: Cambridge Scholars, 2010), 42–47. 23. ​His account appears ­under the pseudonym Pierre Mole, Exposé impartial de l’affaire Dreyfus (Paris: P. V. Stock 1899). 24. ​Letters of ac­cep­tance and refusal can be found in Ferri, “Emile Zola,” 601–603. 25. ​Ruth Harris, Dreyfus: Politics, Emotion, and the Scandal of the C ­ entury (New York: Holt 2010), 143. 26. ​Bertrand Joly, “L’école des chartes et l’affaire Dreyfus,” Bibliothèque de l’école des chartes 147 (1989): 629–637; Harris, Dreyfus, 141–146. 27. ​Extensive excerpts from courtroom per­for­mances of historians in the Zola trial are available in Olivier Dumoulin, Le rôle social de l’historien (Paris: Albin Michel, 2003), 165–166. See also Jean-­Noël Jeanneney, La passé dans le prétoire (Paris: Seuil, 1998), 13–33. The full transcript of the Zola trial can be found in Le procès Zola, devant la Cour d’assises de la Seine et la Cour de Cassation, vols. 1–2 (Paris: P. V. Stock, 1898). 28. ​Quoted in Jeanneney, La passé dans le prétoire, 16. 29. ​The full rec­ord can be found in Le pro­cess Dreyfus, 3:3–55. 30. ​O. Dumoulin, Le rôle social de l’historien, 167. 31. ​ Le pro­cess Dreyfus, 2:297–496. 32. ​O. Dumoulin, Le rôle social de l’historien, 168. 33. ​Julien Benda, La jeunesse d’un clerc (Paris: Gallimard, 1968), 115. 34. ​Maurice Paleologue, An Intimate Journal of the Dreyfus Case (New York: Criterion, 1957), 271. 35. ​Noronha-­DiVanna, Writing History in the Third Republic, 46. 36. ​Whyte, Dreyfus Affair, 24. 37. ​Karl Dietrich Erdmann, ­Toward a Global Community of Historians: The International Historical Congresses and the International Committee of Historical Sciences, 1898–2000 (New York: Berghahn Books 2005), 10. 38. ​Harold J. Berman, Law and Revolution: The Formation of the Western ­Legal Tradition (Cambridge, Mass.: Harvard University Press, 1983), 17. 39. ​Biographical details are in Harry Ritter, “Progressive Historians and the Historical Imagination in Austria: Heinrich Friedjung and Richard Charmatz,” Austrian History Yearbook 19, no. 1 (1983): 45–90. 40. ​Friedjung’s speeches, articles, and correspondence are assembled in Franz Adlgasser and Margret Friedrich, eds., Heinrich Friedjung, Geschichte in Gesprächen Aufzeichnungen 1898– 1919, vols. 1–2 (Vienna: Böhlau, 1998). The evolution of Friedjung’s attitudes ­toward southern Slavs was the subject of a doctoral dissertation by Franz Graf, “Heinrich Friedjung und die sudslawische Frage” (Vienna: University of Vienna, 1950). 41. ​See Ranka Gašić, Novi kurs Srba u Hrvatskoj (Zagreb: Prosvjeta, 2001), 52–54; more on the context in which this pro­cess was happening can be found in Nicholas J. Miller, Between

36

v l a di mir petrov ić

Nation and State: Serbian Politics in Croatia before the First World War (Pittsburgh: University of Pittsburgh Press 1997), 125–134. For more on the po­liti­cal context, see Livia Kardum, “Aneksiona kriza i Friedjungov pro­cess,” Politička misao 1 (1993): 133–147. 42. ​Heinrich Friedjung, “Österreich-­Ungarn und Serbien,” Neue Freie Presse, March 25, 1909, 2–4. 43. ​Heinrich Friedjung, “Österreich-­Ungarn und Serbien,” Neue Freie Presse, April 10, 1909, 4–5. 44. ​Extensive analy­sis of the trial was written by Hodimir Sirotković, “Pravni i politički aspekti procesa Rundpost-­Fridjung,” Starine JAZU 52 (1962): 49–184. The trial generated enormous interest in the press and was extensively covered. Neue Freie Presse published almost the entire stenographic rec­ord: “Der Prozeß Friedjung vor den Wiener Geschwornen,” Neue Freie Presse, December 19, 21, 22, 1909; Obzor and Politika also reported extensively. 45. ​Sirotković, “Pravni i politički aspekti,” 121. 46. ​Heinrich Friedjung, Aktenstücke zur grosserbischen Propaganda in Österreich-­Ungarn, Vorgelegt dem Wiener Geschworenengericht von Heinrich Friedjung (Vienna: Selbstverlag 1909). 47. ​Sirotković, “Pravni i politički aspekti,” 90. 48. ​Ibid., 99–100. 49. ​Adlgasser and Friedrich, Heinrich Friedjung, 1:12. 50. ​The testimonies and particularly Spalajković’s role in the affair are analyzed by Zoran Bajin, “Miroslav Spalajković na Fridjungovom procesu,” Zbornik Matice srpske za istoriju 85 (2012): 101–102. 51. ​“Der Prozeß Friedjung vor den Wiener Geschwornen,” Neue Freie Presse, December 21, 1909, 11–13; “Fridjungov pretres,” Politika 8 (December 21, 1909): 2. 52. ​A.J.P. Taylor, Introduction, in Heinrich Fridejung, The Strug­gle for Supremacy in Germany 1859–1866 (London: Macmillan, 1935), xxviii. 53. ​R. W. Seton Watson to Heinrich Friedjung, December 25, 1909, in R. W. Seton-­Watson and the Yugo­slavs: Correspondence 1906–1941, ed. Hugh Seton-­Watson (London: British Acad­emy, 1976), 1:65. 54. ​R. W. Seton-­Watson, The Southern Slav Question and the Habsburg Monarchy (New York: Howard Fertis, 1969), 211. 55. ​Ernst Breisach, Historiography: Ancient, Medieval and Modern (Chicago: University of Chicago Press, 1994), 323. 56. ​Novick, That Noble Dream. 57. ​Keith Wilson, ed., Forging the Collective Memory: Government and International Historians through Two World Wars (Oxford: Berghahn Books, 1996). 58. ​Vladimir Petrović, The Emergence of Historical Forensic Expertise: Clio Takes the Stand (New York: Routledge, 2017), 85–161. 59. ​Thomas L. Haskell and Sanford Levinson, “Academic Freedom and Expert Witnessing: Historians and the Sears Case,” Texas Law Review 66 ( June 1988): 1629–1659; Richard Evans, “History, Memory, and the Law: The Historian as Expert Witness,” History and Theory 41, no. 3 (October  2002): 326–345; Ramses Delafontaine, Historians as Expert Judicial Witnesses in Tobacco Litigation (New York: Springer International, 2015); Richard A. Wilson, Writing History in International Criminal ­Trials (Cambridge: Cambridge University Press, 2011). 60. ​Austin Sarat and Thomas R. Kearns, eds., History, Memory and the Law (Ann Arbor: University of Michigan Press, 2002), 1.

2 • TIME, JUSTICE, AND ­H UM AN RIGHTS Statutory Limitation on the Right to Truth? WILLIA M A. SCHABA S

Although not explic­itly set out in the Universal Declaration of H ­ uman Rights and other foundational h­ uman rights documents, t­ here is a growing recognition of a fundamental “right to truth.” This is often expressed as a component of the requirement that justice be delivered for serious violations of h­ uman rights, including the core princi­ples of the right to life and the prohibition of torture. Transitional justice requires vari­ous mea­sures of accountability of which criminal prosecution of perpetrators is only one ele­ment. However, recent decisions of international courts and tribunals, including the Eu­ro­pean Court of ­Human Rights and the International Court of Justice, manifest a reluctance to extend the right to truth too far into the past. It is as if they have imposed a form of time bar or statutory limitation on the right to know the truth, even when the inquiry concerns the very atrocities—­namely, genocide, crimes against humanity, and war crimes—­whose prosecution cannot be restricted. That ­legal rights and entitlements are subject to temporal limitation is not a controversial proposition. In many ­legal systems, even a murder cannot be prosecuted a­ fter a certain lapse of time. During the 1960s, just two de­cades ­after the Nuremberg trial, many convicted and released war criminals ­were being reintegrated into society and returning to their professional networks, while ­others who bore responsibility for war­time atrocities w ­ ere still at large. Statutory limitation or prescription of prosecution in some countries threatened to block accountability. In order to address this, the United Nations undertook work on an international convention to deal with the issue. In November 1968, the General Assembly a­ dopted the Convention on Non-­ Applicability of Statutory Limitations to War Crimes and Crimes against Humanity.1 The treaty entered into force two years ­later a­ fter obtaining its tenth 37

38

w illi a m a . sc h a b a s

ratification.2 To date, fewer than sixty States have ratified the treaty. In 1974, the Council of Eu­rope ­adopted its own regional treaty on the same subject. Although the Eu­ro­pean Convention on the Non-­Applicability of Statutory Limitation to Crimes against Humanity and War Crimes is open to all forty-­seven members of the Council of Eu­rope, only three have ratified it.3 The lack of enthusiasm for ­these treaties might suggest equivocation on the subject by many States, but this does not appear to be the case. Article 29 of the Rome Statute of the International Criminal Court, a­ dopted in 1998, declares that genocide, crimes against humanity, war crimes, and the crime of aggression are “not to be subject to any statute of limitations.” ­There are 124 States Parties to the Rome Statute and several other States that have signed it. International judges have spoken of “a broad and recent consensus” that “the criminal punishability of crimes against humanity without any time-­limit can be considered as a princi­ple of customary international law, binding on all States.”4 The 1968 Convention may be viewed as the first foray of the United Nations into accountability for past atrocities. At its first session, in 1946, the General Assembly had a­ dopted a resolution affirming the “princi­ples of international law recognized by the Charter of the Nurnberg Tribunal and the judgment of the International Military Tribunal.”5 The same day, another resolution was ­adopted recognizing genocide as an international crime, confirming in its preamble that “[m]any instances of such crimes of genocide have occurred.”6 However, the United Nations was not then involved in the investigation of h­ uman rights violations. Its members would have shuddered at the idea that the organ­ization would concern itself not only with the pres­ent and the ­future but also with the past. By the 1990s, the U.N. ­human rights organs began to speak about a “right to truth.” This had already emerged at the regional level. In 1986, the Inter-­American Commission on ­Human Rights observed that “­every society has the inalienable right to know the truth about past events, as well as the motives and circumstances in which aberrant crimes came to be committed, in order to prevent repetition of such acts in the ­future.”7 In his 1997 report to the United Nations to the Sub-­Commission on the Protection and Promotion of H ­ uman Rights, Louis Joinet, a long-­time expert within the U.N. ­human rights system, wrote: This is not simply the right of any individual victim or his nearest and dearest to know what happened, a right to the truth. The right to know is also a collective right, drawing upon history to prevent violations from recurring in the ­future. Its corollary is a “duty to remember” on the part of the State: to be forearmed against the perversions of history that go ­under the names of revisionism or negationism, for the history of its oppression is part of a ­people’s national heritage and as such must be preserved. ­These, then, are the main objectives of the right to know as a collective right.8



Time, Justice, and ­Human Rights 39

Joinet listed the “right to truth” as Princi­ple 1 in the Set of Princi­ples for the Protection and Promotion of ­Human Rights through Action to Combat Impunity, where it was presented as a right of a “­people” rather than of individuals: “­Every ­people has the inalienable right to know the truth about past events and about the circumstances and reasons which led, through the consistent pattern of gross violations of h­ uman rights, to the perpetration of aberrant crimes. Full and effective exercise of the right to the truth is essential to avoid any recurrence of such acts in the ­future.”9 In the French version, he wrote: “Chaque peuple a le droit inaliénable.” Princi­ple 2, entitled “The Duty to Remember,” stated: “A p­ eople’s knowledge of the history of their oppression is part of their heritage and, as such, ­shall be preserved by appropriate mea­sures in fulfilment of the State’s duty to remember. Such mea­sures ­shall be aimed at preserving the collective memory from extinction and, in par­tic­u­lar, at guarding against the development of revisionist and negationist arguments.”10 Princi­ple 5 required that states “ensure the preservation of, and access to, archives concerning violations of h­ uman rights and humanitarian law.”11 A de­cade ­later, the U.N. Commission on ­Human Rights requested in­de­pen­ dent expert Diane Orentlicher to prepare an updated version of the Princi­ples. Professor Orentlicher made some minor changes to the text of Joinet’s first princi­ple: “­Every p­ eople has the inalienable right to know the truth about past events concerning the perpetration of heinous crimes and about the circumstances and reasons that led, through massive or systematic violations, to the perpetration of ­those crimes. Full and effective exercise of the right to the truth provides a vital safeguard against the recurrence of violations.”12 She also reformulated the text about “[a] ­people’s knowledge of the history of its oppression” and the importance of archives, saying that this was “part of its heritage and, as such, must be ensured by appropriate mea­sures in fulfilment of the State’s duty to preserve archives and other evidence concerning violations of h­ uman rights and humanitarian law and to facilitate knowledge of ­those violations.”13 Inevitably, the past—­indeed, the rather distant past—­surfaced as an issue at the Durban Conference on Racism, Racial Discrimination, Xenophobia, and Related Intolerance in September 2001. The Declaration a­ dopted by the Conference emphasized that “remembering the crimes or wrongs of the past” and “telling the truth about history” w ­ ere “essential ele­ments for international reconciliation and the creation of socie­ties based on justice, equality and solidarity.”14 The Declaration also stressed “teaching about the facts and truth of the history of humankind from antiquity to the recent past,” all of this “with a view to achieving a comprehensive and objective cognizance of the tragedies of the past.”15 It referred specifically to slavery, the slave trade, the transatlantic slave trade, apartheid, colonialism, and genocide and indicated the relationship between t­hese historic abuses and t­ hose of the pres­ent day.16 It also referred to “dark chapters in

40

w illi a m a . sc h a b a s

history.”17 The equivocal conclusion spoke of a “moral obligation on the part of all concerned States” to take mea­sures “to halt and reverse the lasting consequences of t­ hose practices.”18 Another deliberate ambiguity in the Declaration concerned the application of international criminal law to the past: “[S]lavery and the slave trade are a crime against humanity and should always have been so, especially the transatlantic slave trade.”19 The first U.N. ­Human Rights Council resolution on the right to truth declared that “the public and individuals are entitled to have access, to the fullest extent practicable, to information regarding the actions and decision-­making pro­cesses of their Government, within the framework of each State’s domestic ­legal system.”20 Furthermore, “States should preserve archives and other evidence concerning gross violations of ­human rights and serious violations of international humanitarian law.”21 The Resolution requested the Office of the High Commissioner for H ­ uman Rights to prepare a comprehensive study of the question, “including, in par­tic­u­lar, practices relating to archives and rec­ords concerning gross violations of h­ uman rights with a view to create guidelines on protecting archives and rec­ords concerning gross h­ uman rights violations.”22 Following the pre­sen­ta­tion of the Report of the Office of the High Commissioner, the H ­ uman Rights Council ­adopted a resolution that highlighted “the importance of preserving historic memory related to gross h­ uman rights violations and serious violations of international humanitarian law through the conservation of archives and other documents related to t­ hose violations.”23 Subsequently, the ­Human Rights Council noted a report on the subject by the Office of the High Commissioner that pointed to the importance of ensuring “that all archives pertaining to h­ uman rights are preserved and protected, and to enact legislation that declares that the nation’s documentary heritage is to be retained and preserved, and creates the framework for managing State rec­ords from their creation to destruction or preservation.”24 The General Assembly has justified the protection of archives as a mea­sure necessary to “facilitate knowledge” of gross violations of h­ uman rights and serious violations of international humanitarian law.25

the “procedural right to truth” at the eu­ro­pean court of h ­ uman rights Louis Joinet’s early discussion of the right to truth did not point to specific provisions of the Universal Declaration of ­Human Rights or the treaties. Nevertheless, his remarks ­were situated in a discussion of impunity in which two fundamental rights—­the right to life and the prohibition of cruel, inhuman, or degrading treatment—­are of par­tic­ul­ar importance. In this context, the Eu­ro­ pean Court of ­Human Rights has developed the notion of a “procedural dimension” to ­these rights.26 Accordingly, ­there is an obligation upon the State to



Time, Justice, and ­Human Rights 41

investigate violations of ­these fundamental rights even if it is not itself responsible for perpetrating them. The procedural obligation of the right to life, set out in Article 2 of the Eu­ro­pean Court of ­Human Rights, has been held to be “separate and autonomous” from the substantive obligation.27 According to the Eu­ro­pean Court of ­Human Rights, the procedural obligation entitles victims, their families, and heirs to know the truth about circumstances associated with a violation of the right to life, especially when this is linked to a large-­scale or massive violation of fundamental rights. For example, it has stressed the importance of this right “in the event of widespread use of lethal force against the civilian population during anti-­Government demonstrations preceding the transition from a totalitarian regime to a more demo­cratic system.”28 Although the issue arises in the context of individual rights, t­ here is some authority in the court’s jurisprudence for the recognition of a broader collective or social dimension to this right to know the truth. Several cases ­will illustrate this. In one decision regarding Romania, the court framed the importance of an investigation into events not only in light of the rights of the individual applicants but “in view also of the importance to Romanian society of knowing the truth about the events of December 1989.”29 In another instance, El Masri v. the “former Yugo­slav Republic of Macedonia,” in which government complicity in “extraordinary renditions” by the Central Intelligence Agency was involved, the ­Grand Chamber of the Eu­ro­pean Court of H ­ uman Rights underscored “the ­great importance of the pres­ent case not only for the applicant and his ­family, but also for other victims of similar crimes and the general public, who had the right to know what had happened.”30 In a concurring opinion in that case, several judges noted that the right to truth was implicit in the Eu­ro­pean Convention, in par­tic­ul­ ar within the procedural obligation of Articles 2 (right to life) and 3 (prohibition of torture and inhuman or degrading treatment or punishment), although they argued that it was best approached from the perspective of Article 13 (right to a remedy). They wrote: In practice, the search for the truth is the objective purpose of the obligation to carry out an investigation and the raison d’être of the related quality requirements (transparency, diligence, in­de­pen­dence, access, disclosure of results and scrutiny). For society in general, the desire to ascertain the truth plays a part in strengthening confidence in public institutions and hence the rule of law. For t­ hose concerned—­ the victims’ families and close friends—­establishing the true facts and securing an acknowl­edgment of serious breaches of ­human rights and humanitarian law constitute forms of redress that are just as impor­tant as compensation, and sometimes even more so. Ultimately, the wall of silence and the cloak of secrecy prevent t­ hese ­people from making any sense of what they have experienced and are the greatest obstacles to their recovery.31

42

w illi a m a . sc h a b a s

In Abu Zubaydah v. Poland, which concerned complicity of Polish authorities in the torture of a prisoner who was being transferred to Guantanamo by the United States Central Intelligence Agency, a chamber of the court wrote: “[W]­here allegations of serious ­human rights violations are involved in the investigation, the right to the truth regarding the relevant circumstances of the case does not belong solely to the victim of the crime and his or her f­ amily but also to other victims of similar violations and the general public, who have the right to know what has happened.”32 Another decision of the Eu­ro­pean Court, referring to killings by soldiers during the conflict in Northern Ireland, insisted upon the importance of a “reasoned decision available to reassure a concerned public that the rule of law had been respected.”33 Granting an application based upon Article 3 of the Eu­ro­pean Convention, a chamber of the Eu­ro­pean Court said it appreciated the fact that descendants of the Polish officers murdered at Katyn had “suffered a double trauma: not only had their relatives perished in the war but they ­were not allowed, for po­liti­cal reasons, to learn the truth about what had happened and forced to accept the distortion of historical fact by the Soviet and Polish Communist authorities for more than fifty years.”34 In another of the rendition cases, Al Nashiri, the court cited at length from the testimony of the U.N. Special Rapporteur on the promotion and protection of ­human rights and fundamental freedoms while counteracting terrorism and about the importance of the right to truth, although it did not use the expression in its ­actual holding. The Rapporteur told the court that “within the United Nations it was seen as a right which had two dimensions—­a private dimension and a public dimension. It was the consistent position of the UN mechanisms that where gross or systematic ­human rights violations w ­ ere alleged to have occurred, the right to know the truth was not one that belonged solely to the immediate victim but also to society.”35 In the Katyn case, the ­Grand Chamber also cited submissions, this time from the Open Society Justice Initiative, about the right to truth.36

katyn at the eu­ro­pean court If ­there is such a broad, collective, historical dimension to the right to truth, when does it actually begin? Or rather, to return to the discussion on statutory limitation with which this chapter began, is it time-­barred? Do the princi­ples prohibiting statutory limitation of crimes against humanity and war crimes provide l­egal muscle to help pry open the door to the more distant past in such a way that ordinary procedural and jurisdictional limitations do not apply in the same inflexible manner as they may when so-­called ordinary crimes are concerned? This is an issue that does not arise, at least not in the same way, as long as the right to truth is confined to the victims themselves or to their next of kin. From



Time, Justice, and ­Human Rights 43

the perspective of individual victims, the right presumably exists only as long as they or their offspring are themselves alive.37 Perhaps the right can even be transmitted to one or more succeeding generations. But when the right is associated with the “­people’s knowledge of the history of its oppression,” and when it is “aimed at preserving the collective memory from extinction” ­because it is “a collective right, drawing upon history to prevent violations from recurring in the ­future,” can t­ here be any logical basis for a temporal limitation on its scope? This difficult question was complicated even further by a 2013 ruling by a sharply divided ­Grand Chamber of the Eu­ro­pean Court of H ­ uman Rights, which imposed a form of statutory limitation on the right to truth. It barred the door to claims based upon the procedural obligation if the a­ ctual loss of life occurred prior to the adoption of the Eu­ro­pean Convention on ­Human Rights in November 1950. The case concerned the refusal of Rus­sia to release documents on the Katyn massacre, the mass murder of about 20,000 Polish officers that took place in 1940, ­after eastern Poland had been occupied by the Soviet Union following the German invasion of the western part of the country. At the insistence of the Soviet prosecutors, the Nazi defendants ­were accused of responsibility for the killings by the International Military Tribunal at Nuremberg. A ­ fter hearing two days of inconclusive evidence in the final days of the trial, the judges simply ignored the issue in their final ruling.38 The Soviet government persisted in its denial of responsibility for several de­cades. Only in 1990 did the Rus­sian leaders fi­nally acknowledge the truth. Yet years a­ fter the admission of guilt, the Rus­sian government continued to refuse to release files and documents related to the killings on grounds of national security. Relatives of the victims complained of Soviet intransigence before the Eu­ro­pean Court of ­Human Rights. It is well established in case law that from the moment that the Eu­ro­pean Convention enters into force for any given State, t­ here is an obligation upon the State Party to “secure to every­one” the rights and freedoms set out in Section I of the Convention. Beginning with this “critical date,” all of the State’s acts and omissions “not only must conform to the Convention but are also undoubtedly subject to review by the Convention institutions.”39 In other words, the obligations of a State Party to the Eu­ro­pean Convention are in princi­ple only prospective in nature. The corollary to this is rooted in the presumption that international treaties do not operate retroactively ­unless a dif­fer­ent intention appears from the treaty or is other­wise established, something confirmed in Article 28 of the Vienna Convention on the Law of Treaties. Consequently, the treaty provisions “do not bind a party in relation to any act or fact which took place or any situation which ceased to exist before the date of the entry into force of the treaty with re­spect to that party.”40 According to the Eu­ro­pean Court, “the Convention imposes no specific obligation on the Contracting States to provide redress for wrongs or damage caused prior to that date.”41 The G ­ rand Chamber has stated that this princi­ple is “beyond dispute.”42

figure 2.1. While at the Nuremberg Trial, a se­nior British diplomat, Sir Reader Bullard,

set out the case for Soviet responsibility. Despite their doubts, the British government was content to let the Germans take the blame. ­There is a minute on the Foreign Office reading: “It is difficult to know what action the Ambassador hopes might be taken on this tele­gram. His doubts may be well formulated—­and shared by many ­others—­but ­there could be no question of our ‘blowing’ the Rus­sian case ­either in public or in private, and in many ways, it might be as well that Katyn should be disposed of once and for all—­onto the Germans.” (Photo by William A. Schabas.)



Time, Justice, and ­Human Rights 45

The Eu­ro­pean Convention entered into force on September 3, 1953, following the deposit of the tenth ratification by a Member State of the Council of Eu­rope, as provided for by Article 59(3) of the Convention. For States that have ratified or acceded to the Convention subsequent to that date, the Convention enters into force on the date of deposit of its instrument of ratification or accession with the Secretary-­General of the Council of Eu­rope, pursuant to Article 59(4). However, the court may make exceptions to take into account facts that took place prior to the “critical date” in considering the procedural obligation associated with certain fundamental rights, notably ­those in Articles 2 and 3 of the Convention. Where a substantive violation of the right to life has taken place before the “critical date,” the Convention may nevertheless apply to the duty of the State Party to conduct an investigation and to ensure that mea­sures of accountability exist. In disappearance cases, even if the body has been found, “[t]his only casts light on one aspect of the fate of the missing person and the obligation to account for the disappearance and death, as well as to identify and prosecute any perpetrator of unlawful acts in that connection, w ­ ill generally remain.”43 Following divergences in the case law of the chambers of the Eu­ro­pean Court,44 in 2009 the ­Grand Chamber confronted the temporal jurisdiction issue with re­spect to the procedural obligation in a case dealing with medical malpractice. It recognized a procedural obligation pursuant to Article 2 of the Convention even if the a­ ctual killing took place prior to the entry into force of the Convention for the respondent State. Article 2 establishes that “[e]veryone’s right to life ­shall be protected by law” and that “[n]o one ­shall be deprived of his life intentionally.” By fifteen votes to two, the G ­ rand Chamber held that “the procedural obligation to carry out an effective investigation ­under Article 2 has evolved into a separate and autonomous duty” that constitutes “a detachable obligation arising out of Article 2 capable of binding the State even when the death took place before the critical date.”45 Nevertheless, the court said that “having regard to the princi­ple of ­legal certainty” this extension of the temporal jurisdiction of the court was “not open-­ended.”46 The ­Grand Chamber explained the limitations it was placing on the procedural obligation: 162. First, it is clear that, where the death occurred before the critical date, only procedural acts and/or omissions occurring ­after that date can fall within the Court’s temporal jurisdiction. 163. Second, ­there must exist a genuine connection between the death and the entry into force of the Convention in re­spect of the respondent State for the procedural obligations imposed by Article 2 to come into effect. Thus a significant proportion of the procedural steps required by this provision—­which include not only an effective investigation into the death of the person concerned but also the institution of appropriate proceedings for the purpose of determining

46

w illi a m a . sc h a b a s

the cause of the death and holding ­those responsible to account—­will have been or ­ought to have been carried out a­ fter the critical date. However, the Court would not exclude that in certain circumstances the connection could also be based on the need to ensure that the guarantees and the under­lying values of the Convention are protected in a real and effective manner.47

In this way, the ­Grand Chamber established an exception within an exception. The court could exercise jurisdiction with re­spect to a violation of the procedural obligation contained in Article 2 only if a “significant proportion of the procedural steps” had been undertaken ­after the “critical date.” However, this requirement could be waived if “certain circumstances” required the court “to ensure that the guarantees and the under­lying values of the Convention are protected in a real and effective manner.” This “under­lying values” exception has sometimes been called the “humanitarian clause.”48 ­These words amounted to an invitation to victims to formulate applications relating to investigation of historic violations of the right to life where the truth was still obscure, including the Katyn massacre. Within a few years such a case, Janowiec and ­Others v. Rus­sia, filed by relatives of ­those who had been murdered at Katyn, was presented to the Eu­ro­pean Court. In Janowiec, the G ­ rand Chamber of the Eu­ro­pean Court of H ­ uman Rights held that it could not consider the “under­lying values” exception if the loss of life had taken place prior to November 4, 1950, the date when the Eu­ro­pean Convention on ­Human Rights was signed at the Barberini Palace in Rome. It said that “a Contracting Party cannot be held responsible u­ nder the Convention for not investigating even the most serious crimes ­under international law if they predated the Convention.”49 The ­Grand Chamber explained that “the events that might have triggered the obligation to investigate ­under Article 2 took place in early 1940, that is, more than ten years before the Convention came into existence.” It said that “­there ­were no ele­ments capable of providing a bridge from the distant past into the recent post–­entry into force period.”50 At first instance, before the chamber, the Rus­sian judge had written that the Eu­ro­pean Convention on ­Human Rights had “arisen out of a bloody chapter of Eu­ro­pean history in the twentieth ­century” but that it was drafted as part of a pro­cess of reconstructing postwar Eu­rope and “not with the intention of delving into that black chapter.”51 The very substantial dissenting opinion of Judges Ziemele, De Gaetano, Laffranque, and Keller harshly criticized the majority for turning a “long history of justice delayed into a permanent case of justice denied.”52 With re­spect to the “under­lying values” exception in Šilih, the dissenters noted that the majority simply did not apply it. The majority’s decision “closes the Court’s door to victims of any gross ­human rights violation that occurred prior to the existence of the Convention,” they wrote.53 By failing to implement the “humanitarian clause,” the dissenters said that the court has failed “to fulfil the role for which it was intended: to provide a Court that would act as a ‘conscience’ for Eu­rope.”54



Time, Justice, and ­Human Rights 47

It is indeed hard to be a “conscience” if one refuses to even consider the past. Where ­else does “conscience” come from if not the past? In Janowiec and ­Others, it was argued that one ­factor justifying extension of the temporal jurisdiction of the court, on an exceptional basis, was the importance of truth-­seeking in order to enable nations to learn from their history and take mea­sures to prevent f­ uture atrocities.55 Reliance was placed upon the Updated Set of Princi­ples drafted by Professor Orentlicher, as well as on the International Committee of the Red Cross rules of customary international law56 and the case law of the Inter-­American Court of H ­ uman Rights.57 However, the ­Grand Chamber distinguished the procedural obligation of individual victims detailed in Article 2 from “other types of inquiries that may be carried out for other purposes, such as establishing a historical truth.”58 The four dissenting judges said that in addition to the rights of the individual victims, “it is equally clear that the obligation to investigate and prosecute ­those responsible for grave ­human rights and serious humanitarian law violations serves fundamental public interests by allowing a nation to learn from its history and by combating impunity.”59

the “passage of time” In recent rulings on the incompatibility of genocide denial legislation with freedom of expression, the Eu­ro­pean Court of H ­ uman Rights has referred to “the time ­factor.” A seven-­judge chamber said that when debates on historical issues are concerned, “[T]he passing of time makes it inappropriate to deal with certain remarks about historical events, many years on, with the same severity as just a few years before. That forms part of the efforts that ­every country must make to debate its own history openly and dispassionately.”60 The ­Grand Chamber in the same case, a­ fter citing an earlier pre­ce­dent in which the impugned speech concerned events forty years earlier, said that with re­spect to the Armenian genocide of 1915 “the lapse of time between the applicant’s statements and the tragic events to which he was referring was considerably longer, about ninety years, and at the time when he made the statements t­ here ­were surely very few, if any, survivors of ­these events.” The G ­ rand Chamber noted that some of the interveners in the case had emphasized “that this was still a live issue for many Armenians, especially ­those in the diaspora.” It said that “the time ele­ment cannot be disregarded. Whereas events of relatively recent vintage may be so traumatic as to warrant, for a period of time, an enhanced degree of regulation of statements relating to them, the need for such regulation is bound to recede with the passage of time.”61 The Eu­ro­pean Court has made similar pronouncements in cases dealing with Nazi sympathizers in war­time Switzerland,62 supporters of communism in modern-­ day Hungary,63 torturers in Algeria,64 and defense of French collaborators during the Second World War,65 as well as in cases dealing with individual reputation and image.66

48

w illi a m a . sc h a b a s

The Eu­ro­pean Court’s “passage of time” approach seems rather close to the concept of statutory limitation or prescription. In the Armenian genocide case, however, the court seemed to be saying to the descendants of victims of the “crime of crimes” that as a c­ entury had passed since the atrocities w ­ ere committed, perhaps they ­were insisting too much. Seven members of the G ­ rand Chamber dissented from the majority, specifically challenging its views on the passage of time issue. “Are we to infer that in twenty or thirty years’ time, Holocaust denial itself might be acceptable in terms of freedom of expression?,” wrote the dissenters. “How can this ­factor be squared with the princi­ple that statutory limitations are not applicable to war crimes and crimes against humanity?”67

the spanish civil war and the special rapporteur on truth The Special Rapporteur on the promotion of truth, justice, reparation and guarantees of non-­recurrence, a mandate established by the ­Human Rights Council in 2011, confronted some of t­ hese issues, but only indirectly, in his 2014 report on a mission to Spain. Pablo de Greiff welcomed efforts at the removal of symbols or monuments exalting the military uprising, the Civil War, and Franco’s dictatorship.68 Noting the importance of the teaching of history, he urged that it be “approached as a system of investigation rather than a mechanism for simply preserving data, [that] can train citizens in habits of analy­sis and critical reasoning.”69 He discussed a number of mea­sures u­ nder consideration for elucidating the truth about Spain’s history, including oral history proj­ects, a pos­si­ble truth commission, and access to archival materials. De Greiff was, of course, focusing his attention on the Spanish Civil War of 1936–1939, a conflict that took place prior to Spain’s membership in the United Nations, indeed, prior to the existence of the United Nations and, moreover, prior to the modern recognition in international law of fundamental h­ uman rights, including the right to life and the prohibition of ill treatment. ­W hether U.N. ­human rights mechanisms have the authority to examine issues relating to the Spanish Civil War is an issue that has also arisen before other bodies. In its report to the Committee on Enforced Disappearances, Spain explained that following its ratification of the International Convention for the Protection of All Persons from Enforced Disappearance, “certain organ­izations—­ associations and non-­governmental organ­izations (NGOs) which may or may not have consultative status—­have raised the issue of its applicability to enforced disappearances alleged to have taken place during the civil war and u­ nder the Franco regime and the need to abrogate or declare inapplicable the Amnesty Act (No. 46/1977) of 15 October.” Spain invoked Article 35(1) of the Convention that declares the committee to be competent “solely in re­spect of enforced disappearances which commenced ­after the entry into force of this Convention.”70 The



Time, Justice, and ­Human Rights 49

International Convention for the Protection of All Persons from Enforced Disappearance is often cited b­ ecause it is the only international h­ uman rights treaty with an explicit recognition of the right to truth. The final recital of the Convention’s preamble “[a]ffirm[s] the right of any victim to know the truth about the circumstances of an enforced disappearance and the fate of the dis­appeared person, and the right to freedom to seek, receive and impart information to this end.” Article 25(2) states: “Each victim has the right to know the truth regarding the circumstances of the enforced disappearance, the pro­gress and results of the investigation and the fate of the dis­appeared person. Each State Party s­ hall take appropriate mea­sures in this regard.” Recently, the Working Group on Enforced or Involuntary Disappearance has also considered Spanish cases relating to the civil war. Spain did not object to the Working Group’s authority to inquire into ­these cases.71 Previously, however, it seems that the Working Group declined to consider disappearances that occurred prior to the creation of the United Nations in 1945.72 In any event, enforced disappearance has special features b­ ecause of its continuous nature,73 something that cannot be said about a violation of the right to life or the prohibition of ill treatment. The Special Rapporteur on the promotion of truth did not speak to ­whether he had “jurisdiction” over the Spanish Civil War and Fascist Spain. Like the Working Group on Enforced or Involuntary Disappearances, his mandate says nothing specific about its temporal scope. Moreover, if an unresolved disappearance is a continuing violation, surely the same can be said about a partial or incomplete truth. The closest authority for the Special Rapporteur appears to be operative paragraph 1(b) of the relevant resolution: To gather relevant information on national situations, including on normative frameworks, national practices and experiences, such as truth and reconciliation commissions and other mechanisms, relating to the promotion of truth, justice, reparation and guarantees of non-­recurrence in addressing gross violations of ­human rights and serious violations of international humanitarian law, and to study trends, developments and challenges and to make recommendations thereon.”74

If this passage is read literally, he is on reasonably solid ground. But if his mandate applies to the distant past, then can this also be said about the other special procedures? Would t­ here not be strenuous opposition within the H ­ uman Rights Council and the General Assembly if the Special Rapporteur on extrajudicial, summary, or arbitrary executions ­were to investigate lynching in the United States prior to the Second World War or if the Special Rapporteur on torture ­were to inquire into the colonial practices of Britain, France, Belgium, and the Netherlands in the 1930s? Moreover, if the Special Rapporteur on the promotion

50

w illi a m a . sc h a b a s

of truth can inquire into the Spanish Civil War and its aftermath, can ­there be any logical reason why he could not also, within the context of his mandate, address issues of historical truth surrounding the Armenian genocide of 1915, the Irish famine of the 1840s, the transatlantic slave trade of the sixteenth, seventeenth, and eigh­teenth centuries, the St. Bartholomew’s day massacre of 1572, and the rape of the Sabine ­women of 750 b.c.e?

temporal limits on the genocide convention ­ ere is another recent example of judicial reluctance to look too far into the Th past, this time in a judgment of the International Court of Justice. The debate concerned the retroactive application of the Convention on the Prevention and Punishment of the Crime of Genocide. The Convention was a­ dopted by the U.N. General Assembly on December 9, 1948, a date that was henceforth to be known as the International Day of Commemoration and Dignity of the Victims of Genocide.75 The Convention’s preamble states that “at all periods of history genocide has inflicted ­great losses on humanity” and that its purpose is “to liberate mankind from such an odious scourge.”76 The preamble also makes reference to General Assembly Resolution 96(I), a­ dopted two years earlier, that affirmed that “[m]any instances of such crimes of genocide have occurred when racial, religious, po­liti­cal and other groups have been destroyed, entirely or in part.” In a 2008 ruling, the International Court of Justice clearly left the door open on the question of temporal scope of the Convention, noting “that t­here is no express provision in the Genocide Convention limiting its jurisdiction ratione temporis.”77 But in February 2015, in its final judgment in the case of Croatia v. Serbia, the court stated definitively that “the substantive provisions of the Convention do not impose upon a State obligations in relation to acts said to have occurred before that State became bound by the Convention.”78 Like the Eu­ro­ pean Court of ­Human Rights, the International Court of Justice pointed to the presumption against retroactivity in Article 28 of the Vienna Convention on the Law of Treaties. It acknowledged that the presumption was rebuttable, providing examples of international criminal law treaties with retroactive or retrospective effect, such as the Convention on the Non-­Applicability of Statutory Limitations to War Crimes and Crimes against Humanity. On the other hand, the International Court of Justice cited references in the travaux préparatoires of the Genocide Convention in which a few States suggested it was “intended to apply to acts taking place in the ­future and not to be applicable to ­those which had occurred during the Second World War or at other times in the past.”79



Time, Justice, and ­Human Rights 51

concluding remarks Neither the Eu­ro­pean Court of ­Human Rights nor the International Court of Justice can be faulted for reaching a conclusion that is not patently unreasonable. Their findings are in line with much of the academic commentary as well as with pre­ce­dent. Perhaps the only implausible ele­ment is the fixation of the Eu­ro­pean Court on November 4, 1950 as the starting date for the right to truth. To the extent it was deemed necessary to set a “critical date,” it might have been more logical to pick that of the entry into force of the Eu­ro­pean Convention, in line with the approach of the International Court of Justice, or the date of adoption of the Universal Declaration of H ­ uman Rights, on December 10, 1948, given that the Eu­ro­pean Convention, according to the final recital of its preamble, was intended “to take the first steps for the collective enforcement of certain of the Rights stated in the Universal Declaration.” But this is a detail. At a time when ­there is a growing movement within h­ uman rights law to affirm a right to truth and to insist upon the importance of historical memory, t­ hese two courts seem to have drawn a curtain on the past in or around the year 1950. Some ­will say in defense of t­hese judgments that the courtroom is not the place to adjudicate such ­matters. This message also emerges from a December 2014 decision of the Eu­ro­pean Court of H ­ uman Rights in an application filed by Yevgeniy Yakovlevich Dzhugashvili, a grand­son of Joseph Stalin. The applicant had unsuccessfully sued a Rus­sian journalist for defamation concerning an article about the Katyn massacre that described his grand­father as “a bloodthirsty cannibal.”80 Declaring the application inadmissible, the chamber said “it is an integral part of freedom of expression, guaranteed ­under Article 10 of the Convention, to seek historical truth. It is not the Court’s role to arbitrate the under­ lying historical issues, which are part of a continuing debate between historians.” It added that “[a] contrary finding would open the way to a judicial intervention in historical debate and inevitably shift the respective historical discussions from public forums to courtrooms.”81 It may well be that other institutions, such as truth commissions, are better suited to the complex task of detailed examination of the past. But the suspicion lingers that far from rising to the challenge of dealing with historical truth, the Eu­ro­pean Court of ­Human Rights and the International Court of Justice are turning their backs on the “bloody, black chapters” to which Judge Kovler referred in his dissent in Janowiec. Truth is often linked with reconciliation, as the names of many recent commissions bear out. But ­human experience suggests that sometimes reconciliation is better achieved without clarifying the truth or acknowledging it. Perhaps this is a leitmotif in the two judgments. That ­there is a right to truth applicable to recent de­cades seems ­today beyond dispute. The Janowiec ruling of the ­Grand Chamber of the Eu­ro­pean Court shuts out pre-1950 atrocities, but at the same time it seems to clear the way for inquiries

52

w illi a m a . sc h a b a s

into events of the second half of the twentieth ­century, even with re­spect to States that did not ratify the Eu­ro­pean Convention u­ ntil the 1990s. Nor does ­there appear to be much dispute about the distant past, when t­ here is l­ittle suggestion that States intentionally conceal documents and lock their archives, even if legitimate debates about historical events persist. With re­spect to t­hese very old atrocities of distant centuries and millennia, States are content to mutter apologies, with varying degrees of sincerity. But t­ here is a problematic grey zone, starting at about 1950 and g­ oing back several de­cades, perhaps a c­ entury, when the right to truth still seems to be impor­tant, yet its recognition encounters re­sis­ tance, as it did recently in Strasbourg and The Hague. The right to truth is closely related to the rights to justice and to reparations. In some cases, the hope of some compensation, of restored property, of financial gain, may lie at the heart of campaigns for historical truth. But the right to truth is also associated with a sense that knowledge of historical truth is intrinsically impor­tant. It contributes in a general way to justice, the rule of law, demo­cratic governance, and social wellbeing. For reasons that mere law cannot explain, many of us still feel emotionally connected to the not-­so-­distant past, to the grey zone. Some formulate this as an individual right, based on lineage to increasingly distant generations. The treatment of their immediate ancestors is impor­tant, although the concern generally grows cold as the time frame becomes more remote. More generally, the “­people,” to use the expression of the Joinet and Orentlicher Princi­ples, insist upon knowing the truth about this grey zone, much of it well before our birth yet somehow close enough to engage us directly. In his report on the Spanish Civil War, Special Rapporteur De Greiff has made an intriguing foray into this fog of history. The consequences of his initiative are difficult to assess. He w ­ ill doubtless receive many entreaties to look at other situations, cases, massacres, and atrocities. Only time ­w ill tell just how far back ­human rights law w ­ ill succeed in establishing a right to truth and w ­ hether, like the two prestigious courts, it ­will also decide to bolt the door to the past.

notes 1. ​UN Doc. A/RES/2392(XXIII). 2. ​(1970) 754 UNTS 73. 3. ​ETS 82, 1, https://­rm​.­coe​.­int​/­168007617f. 4. ​ Mocanu and o­ thers v. Romania [GC], nos. 10865/09, 45886/07, and 32431/08, Concurring

Opinion of Judge Pinto de Albuquerque, Joined by Judge Vučinić, §5, September 17, 2014.

5. ​UN Doc. A/RES/95(I). 6. ​UN Doc. A/RES/96(II) (emphasis added). 7. ​Inter-­American Commission on ­Human Rights, Annual Report, 1985–1986, OEA/Ser.L/V/

II.68, Doc. 8 rev. 1, p. 193. 8. ​Final report prepared by Mr. Joinet pursuant to Sub-­Commission decision 1996/119, UN Doc. E/CN.4/Sub.2/1997/20/Rev.1, para. 17.



Time, Justice, and ­Human Rights 53

9. ​Ibid., Annex II, Princi­ple 1. 10. ​Ibid., Princi­ple 2. 11. ​Ibid., Princi­ple 5. 12. ​Updated Set of princi­ples for the protection and promotion of h ­ uman rights through action

to combat impunity, E/CN.4/2005/102/Add.1, Princi­ple 2. 13. ​Ibid., Princi­ple 3 14. ​UN Doc. A/CONF.189/12 and Corr.1, chap. I, para. 106. 15. ​Ibid., para. 98. 16. ​Ibid., para. 99. 17. ​Ibid., para. 101. 18. ​Ibid., para. 102. 19. ​Ibid., para. 13. 20. ​Right to the truth, UN Doc. A/HRC/RES/9/11, PP 14. 21. ​Ibid., PP 15. 22. ​Ibid., UN Doc. A/HRC/RES/ 9/11, 0P 7. 23. ​Right to the truth, UN Doc. A/HRC/RES/12/12, PP 20. Also: Right to the truth, UN Doc. A/HC/RES/21/7, PP 20. 24. ​Right to the truth, UN Doc. A/HRC/RES/21/7, PP 20, OP 10. 25. ​Right to the truth, UN Doc. A/RES/68/165, para. 13. 26. ​ McCann and ­Others v. the United Kingdom, September 27, 1995, §§157–164, Series A no. 324. 27. ​ Šilih v. Slovenia [GC], no. 71463/01, §159, April 9, 2009. 28. ​ Association “21 December 1989” and ­Others v. Romania, nos. 33810/07 and 18817/08, §106, May 24, 2011. 29. ​Ibid., §130. Also: Şandru and O ­ thers v. Romania, no. 22465/03, §79, December 8, 2009; Vasiliaulskas v. Lithuania [GC], no. 35343/05, Dissenting Opinion of Judge Ziemele, §27, October 20, 2015. 30. ​ El Masri v. “the former Yugo­slav Republic of Macedonia” [GC], no. 39630/09, §191, ECHR 2012. 31. ​Ibid., Joint Concurring Opinion of Judges Tulkens, Speilmann, Sicilianos and Keller, §6. 32. ​ Abu Zubaydah v. Poland, no. 7511/13, §489, July 24, 2014. 33. ​ Kelly and ­Others v. the United Kingdom, no. 30054/96, §118, May 4, 2001. 34. ​ Janowiec and O ­ thers v. Rus­sia, nos. 55508/07 and 29520/09, §156, April 16, 2012. 35. ​ Al Nashiri v. Poland, no. 28761/11, §§481–483, July 24, 2014. 36. ​ Janowiec and O ­ thers v. Rus­sia, nos. 55508/07 and 29520/09, §124, April 16, 2012. 37. ​­Under some circumstances the Eu­ro­pean Court of ­Human Rights w ­ ill give standing to the next of kin in right to life cases in which the victim is no longer alive; see Fairfield and O ­ thers v. the United Kingdom (dec.), no. 24790/04, March 8, 2005. 38. ​William Schabas, “The Katyń Forest Massacre and the Nuremberg Trial,” in Historical Origins of International Criminal Law, ed. Morten Bergsmo, Cheah Wui Ling, Song Tianying, and Yi Ping (Brussels: Torkel Opsahl Academic Publishers, 2015), 3:249–297. 39. ​ Yağcı and Sargın v. Turkey, June 8, 1995, §41, Series A no. 319-­A ; Broniowski v. Poland [GC] (dec.), no. 31443/96, §§74 et seq., ECHR 2002-­X . 40. ​Vienna Convention on the Law of Treaties (1980), 1155 UNTS 331, art. 28. Also: Draft articles on Responsibility of States for Internationally Wrongful Acts, UN Doc. A/56/10, art. 13. 41. ​ Kopecký v. Slovakia [GC], no. 44912/98, §38, ECHR 2004-­IX. 42. ​ Varnava and ­Others v. Turkey [GC], nos. 16064/90, 16065/90, 16066/90, 16068/90, 16069/90, 16070/90, 16071/90, 16072/90, and 16073/90, §130, ECHR 2009; Blečić v. Croatia [GC], no. 59532/00, §70, ECHR 2006-­III. 43. ​ Palić v. Bosnia and Herzegovina, no. 4704/04, §46, February 15, 2011.

54

w illi a m a . sc h a b a s

44. ​ Moldovan and ­Others and Rostaş and O ­ thers v. Romania (dec.), nos. 41138/98 and 64320/01,

March 13, 2001. Followed by: Moldovan and O ­ thers v. Romania (no. 2), nos. 41138/98 and 64320/01, §102, ECHR 2005-­VII (extracts); Kholodov and Kholodova v. Rus­sia (dec.), no. 30651/05, September 14, 2006; Bălăşoiu v. Romania (dec.), no. 37424/97, September 2, 2003. 45. ​ Šilih v. Slovenia [GC], no. 71463/01, §159, April 9, 2009. See: E. Bjorge, “Right for the Wrong Reasons: Šilih v. Slovenia and Jurisdiction Ratione Temporis in the Eu­ro­pean Court of H ­ uman Rights,” British Yearbook of International Law 83, no. 1 (2013): 115. 46. ​ Šilih v. Slovenia [GC], no. 71463/01, §161. 47. ​Ibid., §§162–163 (interior reference omitted). 48. ​For example, Janowiec and ­Others v. Rus­sia [GC], nos. 55508/07 and 29520/09, Concurring Opinion of Judge Gyulumyan, October 21, 2013; Janowiec and O ­ thers v. Rus­sia [GC], nos. 55508/07 and 29520/09, Joint Partly Dissenting Opinion of Judges Ziemele, De Gaetano, Laffranque and Keller, §§7, 31, October 21, 2013. 49. ​ Janowiec and O ­ thers v. Rus­sia [GC], nos. 55508/07 and 29520/09, §151, October 21, 2013. 50. ​Ibid., §160. 51. ​Ibid., Joint Concurring Opinion of Judges Kovler and Yudkivska, April 16, 2012. 52. ​Ibid., Joint Partly Dissenting Opinion of Judges Ziemele, De Gaetano, Laffranque, and Keller, §36. 53. ​Ibid., §33. 54. ​Ibid. 55. ​Ibid., §§124, 126, October 21, 2013. 56. ​Jean-­Marie Henckaerts and Louise Doswald-­Beck, eds., Customary International Humanitarian Law, vol. 1 (Cambridge: Cambridge University Press, 2005), Rule 117. 57. ​I/A Court H.R., Case of Heliodoro-­Portugal v. Panama. Preliminary Objections, Merits, Reparations, and Costs. Judgment of August 12, 2008. Series C No. 186; I/A Court H.R., Case of Gomes Lund et al. (“Guerrilha do Araguaia”) v. Brazil. Preliminary Objections, Merits, Reparations, and Costs. Judgment of November 24, 2010. Series C No. 219. 58. ​ Janowiec and ­Others v. Rus­sia [GC], nos. 55508/07 and 29520/09, §143, October 21, 2013. However, see: ibid., Joint Partly Dissenting Opinion of Judges Ziemele, De Gaetano, Laffranque, and Keller, §§8, 9, October 21, 2013. 59. ​Ibid., §24. 60. ​ Perinçek v. Switzerland, no. 27510/08, §103, December 17, 2013. 61. ​Ibid., §250, October 15, 2015. 62. ​ Monnat v. Switzerland, no. 73604/01, §64, ECHR 2006-­X . 63. ​ Vajnai v. Hungary, no. 33629/06, §49, ECHR 2008. 64. ​ Orban and ­Others v. France, no. 20985/05, §52, January 15, 2009. 65. ​ Lehideux and Isorni v. France, September 23, 1998, §55, Eu­ro­pean Court of ­Human Rights, Reports of Judgments and Decisions 1998-­VII. 66. ​ Editions Plon v. France, no. 58148/00, §53, ECHR 2004-­IV; Smolorz v. Poland, no. 17446/07, §38, October 16, 2012; Hachette Filipacchi Associés v. France, no. 71111/01, §47, June 14, 2007. 67. ​ Perinçek v. Switzerland [GC], no. 27510/08, Joint Dissenting Opinion of Judges Spielmann, Casadevall, Berro, de Gaetano, Sicilianos, Silvis, and Kūris, §8, October 15, 2015. 68. ​Report of the Special Rapporteur on the promotion of truth, justice, reparation and guarantees of non-­recurrence, Pablo de Greiff, UN Doc. A/HRC/27/56/Add.1, paras. 27–33. 69. ​Ibid., para. 34. 70. ​Reports of States parties pursuant to article 29, paragraph 1, of the Convention due in 2012, Spain, UN Doc. CED/C/ESP/1, paras. 3–4. 71. ​Report of the Working Group on Enforced or Involuntary Disappearances, UN Doc. A/ HRC/13/31, paras. 479–502.



Time, Justice, and ­Human Rights 55

72. ​Report of the Working Group on Enforced or Involuntary Disappearances, UN Doc. E/

CN.4/2003/70, para. 247. 73. ​See the General Comment on Enforced Disappearance as a Continuous Crime of the Working Group on Enforced or Involuntary Disappearances, UN Doc. A/HRC/16/48, para. 39. 74. ​Special Rapporteur on the promotion of truth, justice, reparation and guarantees of non-­ recurrence, UN Doc. A/HRC/RES/18/7, OP 1(b). 75. ​Prevention of genocide, UN Doc. A/HRC/28/34, PP 22. 76. ​Convention on the Prevention and Punishment of the Crime of Genocide (1951) 78 UNTS 277, preamble. 77. ​ Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Croatia v. Serbia), Preliminary Objections, Judgment, I.C.J. Reports 2008, p. 412, para. 123. Also: Application of the Convention on the Prevention and Punishment of the Crime of Genocide, Preliminary Objections, Judgment, I.C.J. Reports 1996 (II), p. 617, para. 34. 78. ​ Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Croatia v. Serbia), Judgment, February 3, 2015, para. 100. 79. ​Ibid., para. 97. 80. ​ Dzhugashvili v. Rus­sia, no. 41123/10, December 9, 2014, para 5. 81. ​Ibid., para 33. See also Perincek v. Switzerland, no. 27510/08, §99, December 17, 2013; Irodalom Kft v. Hungary, no.  64520/10, §63, December  3, 2013; Putistin v. Ukraine, no.  16882/03, Concurring Opinion of Judge Lemmens, November 21, 2013.

3 • HOW TRUTH RECOVERY C AN BENEFIT FROM A CONDITIONAL A MNEST Y J EREMY SA RKI N

This chapter reviews the potential of conditional amnesties as a tool for truth recovery in postconflict socie­ties. The general position is that a variety of amnesties are widely used for dif­fer­ent types of crimes, often ­after mass crimes in transitional socie­ties. Conditional amnesties have also been employed for a variety of purposes. Generally, however, they are not often used in truth recovery pro­cesses in postconflict socie­ties. At pres­ent, truth recovery pro­cesses are mainly dependent on victims for truth-­telling, as perpetrators fear the consequences of participating. The only real use of a conditional pro­cess for truth recovery, in South Africa, was not very successful for a number of reasons. Among them: the incentives to induce p­ eople into applying for amnesties ­were insufficient, and the pro­cess was very bureaucratic, time-­consuming, and costly. This chapter explores the South African use of conditional amnesty to determine ­whether this mechanism could be useful for other truth pro­cesses. It argues that such an amnesty could be beneficial if an amnesty is to be granted anyway. However, for such a conditional amnesty to work, perpetrators have to need amnesty. They need to believe that without it, they run the real risk of prosecution. Thus, a workable conditional amnesty is dependent on an interconnected criminal justice system as well as ­viable investigations. This chapter examines how truth recovery pro­cesses elsewhere can be enhanced by learning the lessons from the South African pro­cess. It shows, for example, how, despite a w ­ hole range of prob­lems, hit squad activity was revealed through vari­ous tools that ensured that perpetrators disclosed the truth. It concludes with other relevant suggestions to further the pro­cess of truth recovery. 56



Truth Recovery and Conditional Amnesty 57

south africa’s amnesty pro­c ess as a useful lens In 2015, a South African newspaper article noted that “eigh­teen years ­after the TRC’s Special Hearing on Caprivi, the ghosts of Operation Marion’s actions, cover-­ups and crimes hover around our current prob­lems with vio­lence in postapartheid South Africa. It’s not clear what, if anything, can lay them to rest.”1 While the article raised questions about the nature of South African society twenty years ­after the Truth and Reconciliation Commission (TRC) and the legacy of the TRC pro­cess, the newspaper commentary focused on the ­human rights atrocities carried by the Inkatha Freedom Party (IFP) and the South African Defence Force (SADF) before and during South Africa’s transition to democracy. “Operation Marion” involved the training and ongoing support of a paramilitary unit to protect Inkatha and to act offensively against the United Demo­cratic Front (UDF)/African National Congress (ANC). The members of the paramilitary unit, some 206 men recruited by Inkatha, ­were secretly trained in the Caprivi Strip in Namibia during 1986. The offensive part of Operation Marion was cleared at the “highest po­liti­cal level and then Minister of Defence, Magnus Malan, was tasked with supplying the requested para-­military support.”2 ­Those recruits ­were then deployed for the IFP to attack their opponents. Thousands of ­people ­were killed and a tremendous amount of vio­lence was fomented as a result of their activities. ­After South Africa’s transition in 1994, vari­ous transitional justice pro­cesses including an amnesty and reparations commission—­and most notably the South African Truth and Reconciliation Commission—­were established to deal with the country’s past. While the 2015 newspaper report is accurate in that t­ here w ­ ere extensive cover-­ups and vari­ous steps taken by perpetrators to hide the truth, it mentions only the Special Hearing3 held by the TRC as the entire means to deal with Operation Marion and hit squad activity that occurred in the country. However, much occurred in South Africa to deal with the violations that resulted from Operation Marion. A special hearing was impor­tant in delving into what Operation Marion did, along with its consequences. It was impor­tant as, legally speaking, the only attempt to hold the perpetrators accountable occurred in the criminal court case of S v. Peter Msane and Nineteen O ­ thers (commonly called the Malan trial),4 which ended in failure. This was the case that ensued a­ fter the massacre of thirteen ­people at the home of UDF activist Victor Ntuli at KwaMakhutha, south of Durban, in January 1987. The massacre took place in the context of increasing rivalry and conflict in the KwaZulu-­Natal region5 between the IFP led by Chief Mangosuthu Buthelezi, then chief minister of the KwaZulu homeland,6 and the UDF, which was aligned with the ANC.7 The massacre was portrayed in the media at the time as a brutal terrorist attack carried out by the UDF/ANC,8 although in real­ity, it was perpetrated by IFP members trained by

58

jere my s a rkin

the South African Defence Force (SADF), which had been deployed in support of Inkatha in KwaZulu-­Natal. The Malan trial, however, ended up not proving t­ hese facts, but instead acquitting former Minister of Defence Magnus Malan and his many co-­accused on all charges in 1996. The Malan acquittals prob­ably had the effect of ensuring that the risk of prosecutions against t­hose who had committed apartheid-­era po­liti­cal crimes lost its believability.9 It meant that “without that stick, the carrot of amnesty is useless.”10 The outcome of the case had g­ reat significance for the work of the TRC, as the cut-­off point for amnesty applications was only six weeks a­ fter the acquittals ­were handed down.11 The acquittals meant that perpetrators waiting for its outcome before deciding ­whether or not to apply for amnesty before the cut-­off date had l­ittle incentive to do so.12 It is also impor­tant to note that Malan advised former members of the security ser­vices against applying for amnesty, stating that being put on trial was preferable to having to testify before the TRC.13 This compromised the emergence of details of security force involvement in gross ­human rights violations, as well the involvement of politicians in t­ hese types of activities. What is ignored in the 2015 newspaper report is that the amnesty pro­cess, as part of the TRC (i.e., in addition to the Special Hearing pro­cess), also dealt extensively with Operation Marion. In fact many ­people involved in Operation Marion applied for amnesty u­ nder the TRC’s conditional amnesty arrangement. ­Those who testified corroborated a number of facts that the court in the Malan trial had dismissed. Thus, the conditional amnesty pro­cess brought a number of the participants in Operation Marion into the amnesty pro­cess. They testified that what the state alleged in the Malan trail was in fact true, despite the acquittal of all the accused. They testified at their vari­ous amnesty hearings that hit squads ­were established by the apartheid state and that they carried out a range of atrocities, including assassinations. ­These perpetrators came forward, applied for amnesty, and told what they knew about what had occurred, and their corroboration of each other’s accounts meant that the truth that emerged was on much firmer ground. Thus, the benefits of South Africa’s conditional amnesty are well illustrated by the very pro­cess it took to get to the bottom of what Operation Marion was, what it intended, and whom it involved. The conditional amnesty pro­cess in South Africa is a useful lens through which to examine conditional amnesty pro­cesses in general, how they can be beneficial to truth recovery, w ­ hether they are valid in international law, and w ­ hether ­there are advantages to pursuing such an option. The South African model, it is suggested ­will not be repeated in the form that was used in that case, for a variety of reasons. Although the model did not work entirely well, much can be learned and many pitfalls avoided by examining the ways in which its tools have been used. In ­doing so, this chapter also offers suggestions for how conditional amnesties can be used for truth recovery and how to avoid the prob­lems that have tainted the model.



Truth Recovery and Conditional Amnesty 59

are conditional amnesties legally valid in international law? Amnesties are used for a variety of purposes, including for po­liti­cally motivated crimes.14 They are frequently used in transitional socie­ties to excuse acts committed to bring about or to stop such a transition.15 Although ­pardons and amnesty as concepts differ in nature—­with p­ ardons usually being individually granted postconviction while amnesties are usually reserved for groups of p­ eople prior to conviction—­there is a similarity in that each excuses illegal conduct. The type of amnesty ­adopted by a country is generally consistent with the type of transformation that the state goes through.16 Amnesties can occur b­ ecause a previous government wants to give immunity to its own operatives, ­because a new government wants to absolve its p­ eople, or ­because the amnesty emerges from a negotiated peace pro­cess. Blanket amnesties are less likely in the latter case than in situations in which ­there has been an overthrow of the previous regime or ­those in which the state reforms itself. While an amnesty law is domestic in nature and applicable only in the country in question, it is supposed to comply with international law, which, for example, determines that it is not granted for serious international crimes like genocide and crimes against humanity.17 This m ­ atter of compliance has been the subject of much attention over the last twenty years,18 partly ­because of the extensive use of amnesties.19 Thus, whereas between 1945 and 2011, t­ here w ­ ere 537 amnesty pro­cesses in 127 countries, 398 of ­these took place in the period ­after 1979.20 Research conducted on the then 193 member states of the United Nations reveals that the constitutions of 186 of them refer to ­either amnesty or ­pardon and that only seven of them forbid amnesties or ­pardons for certain serious crimes.21 Louise Mallinder therefore suggests that in view of how many countries use amnesties, it is not pos­si­ble to make the argument that customary international law does not outlaw amnesties in general.22 Charles Trumbull argues similarly that the involvement of mediator states in discussing and agreeing to amnesties implies that they do not believe that amnesties violate customary international ­legal dictates.23 Within the international and regional ­human rights systems, t­ here is a general consensus that amnesty for certain offences, such as war crimes, crimes against humanity, breaches of the Geneva Conventions, genocide, and torture is legally and morally unacceptable.24 Nonetheless, some amnesties that do not exclude such types of crimes have been a­ dopted in some places. For example, Af­ghan­i­stan enacted a general amnesty law, the “Law on Public Amnesty and National Stability” in 2007, which came into ­legal force in 2009 and provided amnesty for all crimes, without exception, and thus including international crimes.25 This is just one of many examples. While many international instruments t­oday are read as not permitting amnesty especially for serious international crimes and ensuring that the duty to

60

jere my s a rkin

prosecute is effected by states, t­ here is in real­ity no certainty that amnesties in general are barred.26 Some statutes in international law such as the Rome Statute, the treaty that established the International Criminal Court (ICC), are ­silent about amnesties.27 Th ­ ere can, however, be no doubt that the ICC ­will not be bound by an unconditional amnesty granted by a state, and may decide to investigate and prosecute in such cases. Nevertheless, in some cases, amnesty is specifically available. Thus, Article 6(5) of Protocol II to the Geneva Conventions of 1949 unambiguously obliges states to make efforts to grant far-­reaching amnesties upon the conclusion of hostilities in internal armed conflicts.28 The ramifications of this provision are widely disputed. One key question concerns which types of amnesties are permissible and which are not.29 Conditional amnesties tend to fall in the former category.30 Thus, for example, the Extraordinary Chambers in the Courts of Cambodia have held that “certain conditional amnesties such as ­those providing for some form of accountability have also met widespread approval.”31 The South African granting of conditional amnesty has been the most widely referred to when the issue of the compatibility of conditional amnesties with international law is discussed.32 John Dugard has argued that the South African conditional amnesty merits international ac­cep­tance.33 Similarly, Grace Fiddler argues that while some international instruments have introduced specific limitations on amnesties, ­there is no ban against amnesties if they are conditional in nature and a pro­ cess is set in place to investigate the crimes that ­were committed.34 The Belfast Guidelines on Amnesty and Accountability state in this regard that “amnesties are more likely to be viewed as legitimate where they are primarily designed to create institutional and security conditions for the sustainable protection of h­ uman rights, and require individual offenders to engage with mea­sures to ensure truth, accountability and reparations.”35 In fact, while the guidelines indicate that the imposition of more conditions may lead to fewer perpetrators partaking in the amnesty pro­cess, they emphasize that it should expand the lawfulness and acceptability of an amnesty in terms of international law.36 It is certainly true that conditional amnesties for nonserious international crimes are far more legitimate, far more credible, and far more transparent than blanket amnesties.37 They also offer more to victims than blanket or limited amnesties—­namely, truth. Additionally, with regard to perpetrators and accountability, it has been noted that “a lawful amnesty which requires the per­ for­mance of certain obligations . . . ​may in fact be preferable to de facto impunity where the vast bulk of perpetrators are untouched by any ­legal pro­cess.”38 Thus, amnesties can have positive effects if they are conditional and if they address impunity. They can be established in conjunction with lustration and vetting pro­cesses.



Truth Recovery and Conditional Amnesty 61

using amnesties for truth recovery Considered narrowly, amnesty can enable perpetrators to escape from justice and accountability and thereby promote impunity.39 However, the issue of amnesty should also be viewed through a wider lens, “as impor­tant institutions in the governance of mercy, the reassertion of state sovereignty and . . . ​the return of law to a previously lawless domain.”40 Thus, the question of the validity and benefits of amnesty must not be evaluated in isolation. Within the discipline of transitional justice, amnesty is usually seen to undermine one of its pillars—­–­justice, or accountability. However, this is only one of its five pillars, the ­others being truth, reparations, reconciliation, and guarantees of nonrepetition. Moreover, as Martha Minow has stated, “Conditional amnesties do not foreclose truth-­seeking, but instead promote it.”41 Generally, truth recovery in proceedings such as truth commissions relies on victims whose knowledge is useful in constructing at least a partial picture of what occurred the past. However, this knowledge is usually limited to what occurred to them specifically, and even ­there, they may not have much information. Additionally, t­ here may be no survivors who can testify to what occurred. Although it is true that patterns of perpetrator conduct and motivations can be discerned from victim testimonies, much more could be learned from direct testimonies of perpetrators, including not only who planned or ordered a given violation, who was involved, who provided information, and so on, but also, what motivated the perpetrators, what caused the vio­lence, what the connection was between security officials and government personnel, and how the system worked in general. Such information can become available only if perpetrators ­were incentivized to tell the truth by being granted amnesty. Thus, conditional amnesty pro­cesses have the potential to provide answers and provide substantially more truth than presently occurs in investigations that rely almost exclusively on victim testimony. This is not to argue that amnesty should be more regularly provided, or that it should be provided for serious international crimes. Where amnesty is to be granted, it should be conditional. Conditional amnesties have been used in a variety of places (including the United States, Syria, Mozambique, South ­Korea, Palestine, Algeria, and many ­others) to ensure that a person who seeks amnesty has to meet certain criteria including providing details of what they did. Some conditional amnesties demand a payment of a fine and ­others an agreement not to commit a similar offence in the ­future. Sometimes weapons have to be handed over. Usually amnesty is provided in exchange for information about the acts for which amnesty is sought. Victims have a right to the truth, and amnesty is one means by which the state can enable them to learn it, as the South African conditional amnesty pro­cess reveals to some extent.

62

jere my s a rkin

south africa’s amnesty pro­c ess as a means to obtain truth Providing amnesty was an arrangement that emerged from South Africa’s negotiated settlement between primarily the out­going National Party and the incoming African National Congress although ­there w ­ ere many po­liti­cal parties involved in the pro­cess.42 That t­ here was not a blanket amnesty was a major benefit for the society.43 The South African Constitutional Court in its AZAPO decision44 found that “[t]he amnesty contemplated is not a blanket amnesty against criminal prosecution for all and sundry, granted automatically as a uniform past of compulsory statutory amnesia. It is specifically authorised for the purposes of effecting a constructive transition t­owards a demo­cratic order.45 Thus, it was useful that a conditional amnesty pro­cess was created.46 This ensured that applicants had to meet vari­ous conditions to qualify for amnesty.47 ­These conditions, however, ­were not as extensive as they could have been. They included requiring that a person apply, publicly testify if necessary, and truthfully state what he or she had done; in addition, the violation had to have been committed for a po­liti­cal purpose.48 It is pos­si­ble to add a range of other conditions, including participation in disarmament, demobilization, and reintegration pro­cesses and in reconciliation or other pro­cesses in which victims are involved; specifying the involvement of other ­people, mandatory public testimony, and testifying in a given pro­cess or trial; and relinquishing effects unlawfully obtained and paying or contributing to reparations.49 Linking amnesty to the revelation of information about what occurred was a useful strategy for obtaining truth.50 Thus, t­ hose who applied for amnesty, met the criteria, and told the ­whole truth would be granted immunity from prosecution; ­those who did not apply for amnesty or whose applications failed would not.51 Moreover, the pro­cess demanded full disclosure of the truth about an event before amnesty was granted for the crimes committed during that event. Thus, the pro­cess aimed to find out as much as pos­si­ble about the h­ uman rights abuses committed in the past and to satisfy the needs of individual victims for information about what had occurred to them or their families. The TRC itself noted that truth was a core issue and that it was “guided in its work by a desire to understand the events of the recent past. It is not seeking vengeance or retaliation. It is trying to bring about the national unity and reconciliation, on the basis of knowing the truth about the past.”52 The courts also saw the goal of truth discovery as a primary concern of the pro­cess. In the Constitutional Court AZAPO decision in 1995,53 in which victims sought to overturn the decision of Parliament to give perpetrators both criminal and civil amnesty, the constitutionality of the amnesty law was tested,54 the then



Truth Recovery and Conditional Amnesty 63

new Constitutional Court made it clear that knowledge about the past was a major goal and in fact was more impor­tant than t­ rials b­ ecause truth, which the victims of repression seek so desperately to know is, in the circumstances, much more likely to be forthcoming if ­those responsible for such monstrous misdeeds are encouraged to disclose the ­whole truth with the incentive that they ­will not receive the punishment which they undoubtedly deserve if they do. Without that incentive t­here is nothing to encourage such persons to make the disclosures and to reveal the truth which persons in the positions of the applicants so desperately desire. With that incentive, what might unfold are objectives fundamental to the ethos of a new constitutional order.55

The TRC has argued that without the promise of amnesty, perpetrators would prob­ably not have provided their narratives.56 The commission viewed full disclosure as a component in its restorative justice approach57 in that: “amnesty in return for public and full disclosure (as understood within the broader context of the Commission) suggests a restorative understanding of justice, focusing on the healing of victims and perpetrators and on communal restoration.”58 In this regard, the TRC noted in its final report that the “amnesty pro­cess was also a key to the achievement of another objective, namely eliciting as much truth as pos­si­ble about past atrocities. The primary sources of information w ­ ere the perpetrators themselves who, without the option of applying for amnesty, would prob­ably not have told their side of the story.”59 In other words, according to the TRC, most of the impor­tant information came from perpetrators. This needs to be a crucial consideration when evaluating the benefits of conditional amnesties. However, only 7,100 ­people in South Africa applied for amnesty, and only about 1,100 ­people ­were successful. Many perpetrators did not apply, while many of ­those who did ­were common criminals not entitled to amnesty (about 4,500) or participants in the liberation movements.60 Archbishop Tutu noted in this regard that “we ­were fortunate that many police came seeking amnesty, but sadly the military, represented in the old South African Defence Force (SADF), hardly cooperated with the Commission at all. This left a considerable gap in our truth-­ gathering pro­cess. . . . ​­There is much truth that the nation still needs to know if our healing and reconciliation are to be lasting and effective.”61 Or, as then Minister of Justice Dullah Omar put it, “The perpetrators of apartheid crimes have not been as forthcoming as they should have been.”62 Many of t­ hose who may have required amnesty believed that they did not need it ­because they would not be prosecuted or ­because the state would not be able to prosecute them successfully and thus did not apply. If the carrot and stick approach did not bring as many perpetrators into the pro­cess as expected, it is also true that amnesty ­will not necessarily ensure that perpetrators ­will come forward, for a variety of reasons.63

64

jere my s a rkin

Some do not think they need it while ­others rather wish to brave the possibility of a prosecution. Despite the low number of ­people who applied for amnesty, ­there ­were many benefits from the pro­cess. One was that much more was learned by the public which other­wise would have remained hidden. Holding public hearings was also beneficial insofar as accountability was concerned. However, ­there was ­little public attention to the amnesty pro­cess. To some extent this was fueled by the belief that the TRC had finished its work in 1998. This was indeed true for the rest of the TRC, and the formal handing over of the final TRC report occurred in October 1998 with much fanfare. However, the public part of the amnesty pro­cess and the amnesty hearings, only r­eally got ­going in 1998 when the rest of the TRC came to an end. While public hearings occurred for about 2,100 cases, about 5,000 cases remained completely opaque. While 2100 cases w ­ ere “public,” they ­were subject to much less interest partly b­ ecause the pro­cess was often outside of the general public’s view. While the public in theory could attend hearings, it was given ­little information about when and where they ­were to take place. The media covered only a few of the high-­profile cases, with many cases receiving very ­little coverage. In fact, most of ­those high-­profile cases occurred before 1998 so the years from 1998 to 2001 saw only sporadic bursts of information. The extent of what occurred during the apartheid years remained outside of public consciousness ­because of the scant media attention in the years ­after 1998. It also remained hidden ­because the majority of the 7,100 amnesty cases w ­ ere dealt with in private in the offices of the amnesty committee. They ­were deci­ded merely on the basis of the written application of the perpetrator. It is true that many ­were dismissed b­ ecause of an alleged absence of a po­liti­cal motivation. However, open and public hearings would have advanced public knowledge and interest in the pro­cess. Only about two thousand cases had public hearings, mostly a­ fter 1998, when rest of the TRC had closed down. Among the most prominent cases ­were the attempt to get collective amnesty for thirty-­seven top ANC leaders, the killings of activists known as the Cradock Four (popu­lar community leaders/anti-­apartheid activists from the town of Cradock), and the killings of the abducted anti-­apartheid activists known as the Pebco (Port Elizabeth Black Civic Organisation) Three in 1985. One of the most controversial grants of amnesty, which was done without a hearing, was to Archbishop Tutu’s son Trevor for stating that he had a bomb at an airport. Public hearings w ­ ere for cases in which ­there was a gross ­human rights violation. ­Those cases, in theory, allowed a wide range of South Africans—­particularly white South Africans, many of whom claimed not to have known—to hear what abuses had been committed in their name, if they knew when the hearings ­were taking place. Of critical importance is the fact that ­these hearings ­were public in nature even if not many attended. That public testimony was, in effect, a form of punishment of ­those who applied for amnesty, ­because it was, at some level an exercise



Truth Recovery and Conditional Amnesty 65

figure 3.1. The Truth and Reconciliation Commission of South Africa Report in five volumes.

(Photo by Rob Knijff.)

in public shaming. In this sense, t­ here was at least some mea­sure of accountability, and at least some victims received some acknowledgement about what they had endured.64 For a number of participants, the pro­cess was cathartic. The degree of openness of the TRC did not typify the Amnesty hearings, and in fact, some perpetrators took l­egal action to avoid their names being mentioned. Moreover, if they had not been given notice, the TRC was loath to allow victims in general from naming names. The pro­cess was not good at ensuring that victims ­were told when amnesty hearings w ­ ere being held, which almost entirely excluded them from being pres­ent, or being able to object to an amnesty being granted. The media did not follow the amnesty pro­cess in any meaningful way, and besides some high-­profile cases such as the St. James church attack, the Heidelberg Tavern assault, and the killing of American Fulbright student Amy Biehl, relatively few received any significant media attention. Looking back, many of the reasons for victims not being informed had to do with TRC disor­ga­ni­za­tion and at times with the limited capacity of the rooms where the amnesty proceedings w ­ ere held, rather than any specific intent to

66

jere my s a rkin

exclude victims. However, the fact that choices w ­ ere made not to prioritize giving information to victims, or finding them and keep them abreast of their cases, does speak volumes about the way that victims’ needs w ­ ere dealt with in general. If truth is to r­ eally benefit t­ hose who have suffered h­ uman rights violations, then it is impor­tant that they are actually able to participate in the pro­cess of obtaining it. Lip ser­vice cannot be given to victims’ needs in a pro­cess that is actually designed to be for victims. ­There has been extensive debate about how much truth the TRC uncovered.65 Some have questioned w ­ hether what the TRC reported was altogether correct and ­whether it contributed to uncovering the truth about the past. Th ­ ese claims ­were often made for po­liti­cal reasons to undermine the pro­cess. Crucially they misunderstand the role of the TRC pro­cess, how it emerged as a result of the negotiations, and why it was accepted that it would have been difficult to conduct ­trials when the po­liti­cal w ­ ill and evidence w ­ ere not pres­ent. What is true is that the amnesty pro­cess, had it been more successful, would have aided greater truth recovery. It was not that fruitful ­because so few ­people came forward and applied. Many who did apply did not fit into its remit, did not qualify, and ­were not granted amnesty. The pro­cess was cumbersome, bureaucratic, slow, and costly. It relied on investigations that for a variety of reasons w ­ ere inadequate, including lack of funding, and ­there was ­little opportunity to test the veracity of perpetrators’ testimonies. Th ­ ere is therefore a widely held perception, based to some extent on real­ity, that many perpetrators tailored their versions of what they did to what was already known, and ­were complicit with other applicants in determining what to reveal and what not to.

getting the truth about operation marion and avoiding the effect of the malan trial On the substantive issues about the establishment of Operation Marion and the creation of hit squads by the apartheid state and the Inkatha Freedom Party (IFP), the role of the amnesty pro­cess was useful. It played a crucial role in countering the not guilty findings of the court in the Malan trial and in learning what tran­spired as a result of Operation Marion. The mechanism used by the TRC was a special closed section 29 hearing, set up by legislation that provided that “[t]he Commission may for the purposes of or in connection with the conduct of an investigation or the holding of a hearing, . . . ​by notice in writing call upon any person to appear before the Commission and to give evidence or to answer questions relevant to the subject m ­ atter of the investigation or the hearing.” This meant that the TRC could call perpetrators or o­ thers to testify even if they did not apply for amnesty. ­Those who testified could not be prosecuted on the basis of the information they provided, but they could be prosecuted for perjury if the information they provided was not true. H ­ ere, private examination of t­ hese per-



Truth Recovery and Conditional Amnesty 67

sons helped make it more likely that they would give full and detailed information. The law provided that “no person other than a member of the staff of the Commission or any person required . . . ​to give evidence s­ hall be entitled or permitted to attend any hearing conducted in terms of this section.” This ensured that the information remained confidential. The use of the section 29 TRC pro­cess was a useful one as a fair amount of truth was obtained during ­these hearings. Interestingly, while the Malan trial ended in an acquittal for all the accused, the TRC found in its final report in 1998, mainly as a result of the section 29 hearing, that: “In 1986, the SADF force conspired with Inkatha to provide Inkatha with a covert, offensive paramilitary unit (or “hit squad”) to be deployed illegally against persons and organisations perceived to be opposed to or enemies of both the South African government and Inkatha. The SADF provided training, financial and logistical management and behind-­the-­scenes supervision.”66 This finding was made in 1998 before the amnesty pro­cess had dealt with many cases. In this regard, it must be remembered that the main part of the TRC completed its work in 1998 but the amnesty pro­cess continued for another three years.67 The amnesty pro­cess also contributed to getting to the truth about the hit squads. A number of persons involved in the actions for which the Malan trial accused ­were acquitted applied for amnesty. One of ­those was former Sergeant André Cloete,68 who applied for amnesty for the KwaMakutha events dealt with at the Malan trial, in which he had appeared as a state witness. A ­ fter the trial, he was not indemnified for his role in supplying weapons, training attackers, or other aspects of the attack. Hence his amnesty application, which detailed the role of the state in covert activities and confirmed the security forces’ support of the IFP. The Cloete application was supported by other applications, including that of Jan Anton Nieuwoudt, who was given amnesty for providing military training for “offensive action within KwaZulu-­Natal given by applicant to a group of Inkatha Freedom Party members during or about the ­middle of 1986 at or near Hippo Base, Caprivi, Namibia.” This is exactly the issue that had been denied at the Malan trial. On the basis of the amnesty application, however, the amnesty committee found that the offenses ­were criminal in nature and committed with a po­liti­cal objective and that the truth had been told. Thus, amnesty was granted. Another amnesty application case, which revealed the truth concerning the issues dealt with at the Malan trial, was that of Daluxolo Words­worth Luthuli and six ­others. The decision in this case did not discuss the merits of the vari­ous applications, the evidence, or why amnesty was granted. It did, however, contain a long list of crimes committed by each person, including many murders also dealt with at the Malan trial.69 Luthuli alone was given amnesty for more than twenty events, many involving murders and IFP attacks against their opponents. One of the events in which Luthuli participated was the murder and attempted murder of an unidentified number of ­people and the burning down of about thirty ­houses “during a retaliatory attack in 1989 or 1990.” In other words, this was an offensive

68

jere my s a rkin

attack for which the unit was established as alleged in the trial. Luthuli was also given amnesty for providing military training to IFP ­people during March 1986 and December 1993. Again, this was to establish the hit squads alleged in the Malan trial. Thus, while the Malan trial permitted denials and repudiations of the role of the apartheid state in the vio­lence and specifically the establishment of hit squads for the IFP to attack their opponents, the amnesty pro­cess and the section 29 pro­cesses ­were useful tools in getting the perpetrators to testify and uncovering the truth, which the courts had not been able to do.

protecting the truth and making it more difficult for lies to be told While an amnesty pro­cess can assist in truth recovery, t­ here is a danger that applicants w ­ ill not tell the truth. Thus, ways have to be found to prevent applicants from believing they can avoid the truth. In the South African pro­cess vari­ous applicants, particularly ­those not in prison and ­those with access to ­legal repre­sen­ta­tion, had discussions about ­whether to apply and, if so, how to complete their applications in ways that benefited them. As a result, perpetrators often applied when ­others involved in the same incident also applied. It is further clear that t­hose who applied for amnesty carefully focused on ensuring consistency between the applications and testimonies, not to ensure veracity, but to ensure that all applicants had the same or similar versions of what occurred. In this way, full disclosure did not necessarily prove to be an obstacle to meeting the criteria for amnesty. Complicity to avoid the truth occurred not only in the applications, but also in the public hearings. This complicity was aided by the practice during the hearings of allowing one perpetrator or applicant to testify to the events or incidents, while the other applicants simply confirmed what that applicant had said. Often t­ here was very l­ ittle testing of the veracity of the circumstances and the occurrences ­unless t­ here was vigorous protest from the evidence leader of the committee or by victims. Thus, a variety of mechanisms are needed in such conditional amnesty pro­ cesses to protect truth recovery. One of ­these is an interconnected criminal justice system, which works in conjunction with—­and is legally empowered to do so—­and alongside the conditional amnesty pro­cess. This is vital, as without it perpetrators can say what­ever they wish. Indeed, a designated parallel criminal pro­cess to deal with such cases and provide backup to the amnesty pro­cess was lacking in the South African TRC amnesty pro­cess. The two pro­cesses did not work in sync, and in fact often worked in diametrically opposing ways. Often the state officials in the formal justice system opposed what the TRC was d­ oing and undermined it. They also refused to coordinate, coming up with vari­ous reasons why the criminal justice system’s in­de­pen­dence and role would be undermined



Truth Recovery and Conditional Amnesty 69

if it cooperated with the TRC and would not even discuss working methods. This disconnect severely undermined the ability and effectiveness of the TRC to encourage ­people to apply for amnesty. As a result of the infighting and lack of collaboration between the two pro­cesses, the wrong message was given to t­ hose who w ­ ere assessing w ­ hether or not to apply for the amnesty win­dow that was available u­ ntil 1996. Had ­there been a more purposeful role of ­those in the prosecuting authorities, and had they been committed to ­doing all in their power to pursue ­these cases, more p­ eople would have applied for amnesty. ­There would also have been more applications if t­here was seen to be a real intent by the authorities to investigate more apartheid-­era crimes and if the prosecuting authorities ­were seen to be cooperating with the TRC. Had t­ here been a dedicated prosecuting unit focused on pursuing t­ hose who ­ought to have been applying for amnesty, t­ here would have been greater incentive for t­hose who w ­ ere still contemplating applying to do so. Although some assistance did come, the Investigation Task Unit (ITU) investigating hit squads in KwaZulu-Natal and the Office of the Attorney General of the Transvaal, it was quite limited scope ­because the ITU was coming to the end of its work at the beginning of the amnesty pro­cess, and the relationship between the Transvaal Attorney-­General’s office and the TRC was less than positive. In addition, the lack of capacity as far as resources and staff within the parallel criminal pro­cess compromised how much truth was discovered. The ­legal provisions with re­spect to nondisclosure w ­ ere weak. Thus, se­nior leaders and operatives did not come forward in significant numbers to reveal their stories and apply for amnesty. Th ­ ese capacity and l­egal provision deficit f­ actors must affect the extent to which the truth or sufficient truth is known and w ­ hether selected prosecutions might not assist, to some degree, in greater truth discovery. A useful tool to discourage applicants from not telling the truth is a system that permits revocation of an amnesty if it l­ater emerges that the applicant was dishonest. Th ­ ere would need to be a pro­cess in place for this, as without it, applicants would believe—as they did in the South Africa—­that they could get away with being less than truthful. A revocation of amnesty would mean that the information furnished by a culprit would be able to be used in a ­later prosecution. Such an amnesty annulment would encourage full disclosure but would circumvent the ­wholesale complete dependence on investigations, which are usually few and far between.

conclusion Truth recovery is a major goal of transitional justice pro­cesses in postconflict socie­ ties. To learn as much truth as pos­si­ble, as many p­ eople as pos­si­ble need to participate in a pro­cess that attempts to ascertain the fullest picture of what occurred in the past. Excluding a large sector, namely ­those who committed the violations,

70

jere my s a rkin

means that much truth w ­ ill remain unknown. Despite this, in very few truth and reconciliation commission pro­cesses, do perpetrators come forward and participate in any significant numbers. Besides Timor-­Leste, the only real exception to perpetrators participating in TRCs around the world in large numbers has been the South African pro­cess. As a result of its offer of amnesty for truth, thousands of perpetrators entered the pro­cess. However, as noted, this was prob­ably a relatively small percentage of t­ hose who o­ ught to have applied for amnesty. It was largely the result of an in­effec­tive pro­cess that a relatively small percentage of the total number of perpetrators came forward to apply for amnesty. Many culprits realized that the state did not have the capacity or the w ­ ill to pursue t­hose who did not make use of the amnesty pro­cess. The Malan trial, which saw the acquittal of the twenty-­three accused just before the cut-­off for amnesty applications, cemented that view. Despite this, and in fact b­ ecause of the prob­lems that beset the South African TRC, much can be learned from its amnesty pro­cess. The benefits are manifest, but much more could have been achieved. A critical lesson to success is that ­these pro­cesses require coordination between the vari­ous institutions involved, including a TRC, especially its investigation arm, as well as the criminal justice system. A further lesson is that the carrot and stick method can play a crucial role in any conditional amnesty pro­ cess. It should be remembered that ­people ­will only apply for amnesty if they believe that need it. If they believe that they have no need for amnesty, as it unlikely that the state is willing to hold them accountable, they w ­ ill not apply. Thus, the stick is necessary ­because without the state making the right noises and taking appropriate steps, such as investigating and prosecuting a number of perpetrators during the pro­cess, perpetrators ­will not see the necessity to apply for amnesty. Without regular public coordination and harmonization of a TRC amnesty pro­cess and state prosecution, perpetrators ­will not get the impression that ­there is any real intent to prosecute. Thus, in theory, a conditional amnesty in exchange for truth makes sense. In practice, ­there are many pitfalls to getting this to work. The model is tainted ­because the South African conditional amnesty for truth did not work all that well, and so it is unlikely to be replicated. The South African pro­cess was too bureaucratic, too costly, too time consuming, and too cumbersome. If truth recovery is to be part of a conditional amnesty pro­cess, then the pro­cess should be less administratively difficult and more likely to persuade ­people to apply. If providing the truth is a criterion for the granting of amnesty, persuading ­people not to lie is also crucial. Even though a conditional amnesty in exchange for truth can be beneficial, that is not to say that ­there should always, or even often, be an amnesty. In cases of a negotiated outcome, without which ­there would not be peace or a transition, amnesty ­ought to be used optimally. An alternative to a conditional amnesty pro­



Truth Recovery and Conditional Amnesty 71

cess, or even in conjunction with it, involves search and seizure powers that can subpoena perpetrators to testify ­under pain of criminal prosecution and severe punishment. This was done in the South African pro­cess, and it was a useful tool. It could be implemented elsewhere. Further hitherto insufficiently used tools to aid with truth are vetting and lustration programs where ­those who hold a variety of public offices, especially in the security forces, are examined to determine w ­ hether they o­ ught to hold such positions in the ­future. ­These pro­cesses have facilitated the collection of much information, but they have not been as successful in learning more truth for the society as a w ­ hole and the specific victim in par­tic­ul­ar.

notes 1. ​Robyn Leslie and Nala Ramohlokoane, “Caprivi: The Ghosts of Operation Marion Still

Haunt South Africa,” Daily Maverick, September  8, 2015, http://­www​.­dailymaverick​.­co​.­za​ /­section​/s­ outh​-­africa​/­. 2. ​South Africa State Security Council, December 20, 1985, and February 3, 1986, minutes (on file with the author). 3. ​­These hearings ­were provided for in terms of Section 29 of the Promotion of National Unity and Reconciliation Act. Individuals, often perpetrators, supposedly with information ­were subpoenaed u­ nder oath to reveal the information they had in a closed hearing. The information obtained could not ­later be used to prosecute the person who revealed the information. 4. ​ S v Msane Case number CC1/96, Durban and Coast Local Division (1996) (on file with the author). 5. ​See Jeremy Sarkin, “South Africa: An Investigation into the Natal Vio­lence,” Review 45 (1990): 7–11, and Jeremy Sarkin and Howard Varney, “Traditional Weapons, Cultural Expediency and the Po­liti­cal Conflict in South Africa: A Culture of Weapons and a Culture of Vio­ lence,” Criminal Justice Journal 6, no. 2 (May 1993): 2. 6. ​KwaZulu was one of the bantustans, or “self-­governing territories” created by the apartheid state. 7. ​At that time the ANC was a banned organ­ization in South Africa. 8. ​ Daily News, January 21, 1987. 9. ​Dorothy Shea, The South African Truth Commission: The Politics of Reconciliation (Washington, D.C.: United States Institute of Peace, 2000), 5. 10. ​Timothy Garton Ash, “True Confessions,” New York Review of Books 44, no. 12 ( July 1997): 38. 11. ​See Jeremy Sarkin, “The ­Trials and Tribulations of the South African Truth and Reconciliation Commission,” South African Journal on ­Human Rights 12, no. 4 (1996): 617–640. 12. ​Catherine Jenkins, “ ‘They Have Built a L ­ egal System without Punishment’: Reflections on the Use of Amnesty in the South African Transition,” Transformation: Critical Perspectives on Southern Africa 64, no. 1 (2007): 50. 13. ​SA TRC Press Release Truth and Reconciliation, Statement concerning General Magnus Malan, December 8, 1996 (on file with the author). 14. ​Juan Carlos Portilla, “Amnesty: Evolving 21st ­Century Constraints u ­ nder International Law,” Fletcher Forum of World Affairs 38, no. 1 (2014): 169. 15. ​Max Pensky, “Amnesty on Trial: Impunity, Accountability, and the Norms of International Law,” Ethics & Global Politics 1, nos. 1–2 (2008): 1–40.

72

jere my s a rkin

16. ​On a range of vari­ous amnesties see Margaret Popkin and Nehal Bhuta, “Latin American

Amnesties in Comparative Perspective: Can the Past Be Buried?” Ethics and International Affairs 13, no. 1 (1999): 99–122. 17. ​Faustin  Z. Ntoubandi, Amnesty for Crimes against Humanity u ­ nder International Law (Leiden: Brill, 2007). 18. ​Antje Du Bois-­Pedain, Transitional Amnesty in South Africa (Cambridge: Cambridge University Press, 2007), 300. 19. ​Renée Jeffery, Amnesties, Accountability, and H ­ uman Rights (Philadelphia: University of Pennsylvania Press, 2014). 20. ​Louise Mallinder, “Amnesties’ Challenge to the Global Accountability Norm? Interpreting Regional and International Trends in Amnesty Enactment,” in Amnesty in the Age of ­Human Rights Accountability, ed. Francesca Lessa and Leigh  A. Payne (Cambridge: Cambridge University Press, 2012), 79. 21. ​Josepha Close, “The International ­Legal Status of Amnesties Granted for Serious Crimes: Historical and Con­temporary Perspectives,” PhD dissertation (Middlesex University, 2016). 22. ​Mallinder, “Amnesties’ Challenge,” 69, 70–71. 23. ​Charles P. Trumbull IV, “Giving Amnesties a Second Chance,” Berkeley Journal of International Law 25, no. 2 (2017): 283, 291. 24. ​Andreas O’Shea, Amnesty for Crime in International Law and Practice (The Hague: Kluwer Law International, 2002). 25. ​Lisa J. Laplante, “Outlawing Amnesty: The Return of Criminal Justice in Transitional Justice Schemes,” ­Virginia Journal of International Law 49, no. 4 (2009): 918. 26. ​See also Ronald Slye, “The Legitimacy of Amnesties u ­ nder International Law and General Princi­ples of Anglo-­American Law: Is a Legitimate Amnesty Pos­si­ble?” ­Virginia Journal of International Law 43, no. 1 (2002): 173. 27. ​Dwight G. Newman, “The Rome Statute, Some Reservations concerning Amnesties, and a Distributive Prob­lem,” American University International Law Review 20, no.  2 (2005): 293–357, 304. 28. ​Vari­ous states and courts have used this to rationalize the use of amnesties. See, for example, Azanian ­Peoples Organ­ization (AZAPO) and O ­ thers v. President of the Republic of South Africa and ­Others, 8 BCLR 1015 (South African Constitutional Court’s Decision 1996). 29. ​John Dugard, “Dealing with Crimes of a Past Regime, Is Amnesty Still an Option?” Third Manfred Lachs Memorial Lecture, The Hague, 1999. 30. ​Priscilla B. Hayner, Unspeakable Truths (New York: Routledge, 2011), 115. 31. ​Decision on Ieng Sary’s Rule 89 Preliminary Objections (Ne bis in idem and amnesty and ­pardon), 00219-09-2007 (Extraordinary Chambers in the Courts of Cambodia Trial Chamber 2011). Available at: http://­www​.­worldcourts​.­com​/­eccc​/­eng​/­decisions​/­2011​.­11​.­03​_­Co​_­Pro​ secutors​_­v​_­Ieng​_­Sary​.­pdf#search​=­%22decision on sary%22. 32. ​Hendrik J. Lubbe, Successive and Additional Mea­sures to the TRC Amnesty Scheme in South Africa: Prosecutions and Presidential ­Pardons (Antwerp: School of H ­ uman Rights Research Series, Intersentia, 2012), 57:25. 33. ​John Dugard, “Dealing with Crimes of a Past Regime: Is Amnesty Still an Option?” Leiden Journal of International Law 12, no. 4 (2000): 1001–1014. 34. ​Grace Fiddler, “Using a Conditional Amnesty and Truth and Reconciliation Commission as a Transitional Justice Mechanism in Syria,” George Washington Institute of International Law Review 47, no. 4 (2015): 893, 901. 35. ​ The Belfast Guidelines on Amnesty and Accountability (Belfast: Transitional Justice Institute, 2013). 36. ​Ibid., 17



Truth Recovery and Conditional Amnesty 73

37. ​David Weissbrodt, “Review of Andreas O’Shea, Amnesty for Crime in International Law

and Practice,” American Journal of International Law 97, no. 1 (2003): 227. 38. ​Kieran McEvoy and Louise Mallinder, “Amnesties in Transition: Punishment, Restoration, and the Governance of Mercy,” Journal of Law and Society 39, no. 3 (2012): 410. 39. ​Aryeh Neier, War Crimes: Brutality, Genocide, Terror, and the Strug­gle for Justice (New York: Times Books, 1998). 40. ​McEvoy and Mallinder, “Amnesties in Transition,” 410, 440 (italics in original). 41. ​Martha Minow, Between Vengeance and Forgiveness (Boston: Beacon Press, 1998), 56. 42. ​Erin Daly and Jeremy Sarkin, Reconciliation in Divided Socie­ties: Finding Common Ground (Philadelphia: Pennsylvania Studies in H ­ uman Rights, University of Pennsylvania Press, 2007). 43. ​Erik Doxtader and Charles Villa-­Vicencio, eds., The Provocations of Amnesty: Memory, Justice, and Impunity (Claremont, South Africa: David Philip Publishers 2003). 44. ​ A zanian ­Peoples Organ­ization (AZAPO) and O ­ thers v. President of the Republic of South Africa and ­Others (CCT17/96) [1996] ZACC 16; 1996 (8) BCLR 1015; 1996 (4) SA 672 ( July 25, 1996). 45. ​Ibid., AZAPO, para. 32. 46. ​Jeremy Sarkin, Carrots and Sticks: The TRC and the South African Amnesty Pro­cess (Antwerp: Intersentia, 2004). 47. ​The Republic of South Africa, Promotion of National Unity and Reconciliation Act No. 34 of 1995, http://­www​.j­ ustice​.­gov​.­za​/­legislation​/­acts​/­1995​-­034​.­pdf. 48. ​Anurima Bhargava, “Defining Po­liti­cal Crimes: A Case Study of the South African Truth and Reconciliation Commission,” Columbia Law Review 102, no. 5 (2002): 1304. 49. ​ The Belfast Guidelines, 17–18. 50. ​James L. Gibson, “Truth, Justice, and Reconciliation: Judging the Fairness of Amnesty in South Africa,” American Journal of Po­liti­cal Science 46, no. 3 (2002): 540–556. 51. ​Mahmood Mamdani, “Amnesty or Impunity? A Preliminary Critique of the Report of the Truth and Reconciliation Commission of South Africa,” Diacritics 32, no. 3 (2002): 33–59. 52. ​South African Truth and Reconciliation Commission, Truth and Reconciliation Commission of South Africa Report (Part 1: Background and Explanation of Key Terms), http://­www​ .­justice​.­gov​.­za​/­trc​/­report​/­finalreport​/­Volume%201​.­pdf. 53. ​ A zanian ­Peoples Organ­ization (AZAPO) and O ­ thers v. President of the Republic of South Africa and ­Others, 8 BCLR 1015 (South African Constitutional Court’s Decision 1996). 54. ​Lubbe, Successive and Additional Mea­sures, 52 55. ​AZAPO, para. 17 56. ​ TRC Report, vol. 1, chap. 5, para. 62. 57. ​In the Sierra Leonean context, see Kirsten Ainley, Rebekka Friedman, and Chris Mahony, eds., Evaluating Transitional Justice: Accountability and Peacebuilding in Post-­Conflict Sierra Leone (Basingstoke: Palgrave Macmillan, 2015), 55–76. 58. ​ TRC Report, vol. 1, chap. 5, para. 55. 59. ​South African Truth and Reconciliation Commission, Truth and Reconciliation Commission of South Africa Final Report, 1:120. 60. ​Jeremy Sarkin, “An Evaluation of the South African Amnesty Pro­cess,” in Truth and Reconciliation: Did the TRC Deliver?, ed. A. Chapman and H. van der Merwe (Philadelphia: University of Pennsylvania Press, 2008), 93–108. 61. ​Desmond Tutu, No ­Future without Forgiveness (London: Ryder, 1999), 188. 62. ​Reported in Jim Lehrer, “Facing the Past,” PBS Newshour, April 8, 1997, http://­www​.­pbs​ .­org​/­newshour​/­bb​/­africa​/­april97​/­south​_­africa​_­4​-­8​.­html. 63. ​William Schabas and Shane Darcy, Truth Commissions and Courts (Dordrecht, The Netherlands: Kluwer Academic, 2004), 153.

74

jere my s a rkin

64. ​See Du Bois-­Pedain, Transitional Amnesty in South Africa. 65. ​See a major criticism of how much truth was uncovered by Audrey Chapman and Patrick

Ball, “The Truth of Truth Commissions: Comparative Lessons from Haiti, South Africa, and Guatemala,” ­Human Rights Quarterly 23, no. 1 (2001): 1. 66. ​ TRC Final Report, vol. 3, chap. 3, subs. 26, para. 182. 67. ​Martin Coetzee, “An Overview of the TRC Amnesty Pro­cess,” in The Provocations of Amnesty, ed. C. Villa-­Vicencio and Erik Doxtader (Cape Town: David Philip 2003), 115–120. 68. ​On Cloete’s case, see Truth Commission, Amnesty Decision AM 5726/97, http://­sabctrc​ .­saha​.­org​.­za​/­documents​/­decisions​/­59257​.­htm​?­t​=­%2Bcloete+%2Bmc&tab​=­hearings. 69. ​Howard Varney and Jeremy Sarkin, “Failing to Pierce the Hit Squad Veil: A Critique of the Malan Trial,” South African Journal of Criminal Justice 141, no. 2 (1997): 141–161.

4 • NEW EPISTEMOLOGIES FOR CONFRONTING INTERNATIONAL CRIMES Developing the Information, Dialogue, and Pro­cess (IDP) Approach to Transitional Justice S T E P H A N PA R M E N T I E R , M I N A R AUSCH EN B ACH, A N D M A A RT E N VA N C R A E N

Transitional justice has established itself in the last two de­cades as a booming field of study in several disciplines including law, po­liti­cal science, history, psy­chol­ogy, anthropology, sociology, lit­er­a­ture studies, and criminology, to name just a few. Genocide, crimes against humanity, and war crimes have attracted the attention of scholars who are interested in describing and explaining how po­liti­cal, economic, social, and intellectual elites deal with the legacy of large-­scale ­human rights abuses in the aftermath of po­liti­cal regime transitions or outside of them. The academic world has produced a broad stream of publications, or­ga­nized small seminars and large symposia, submitted tiny and huge grant applications to conduct new research, witnessed the defense of master’s ­theses and doctoral dissertations, and engaged in many other activities. Transitional justice is equally high on the agenda of prac­ti­tion­ers and policy-­ makers who are interested in seeing criminal courts prosecute and try offenders of international crimes, develop victim reparation programs, undertake several forms of truth-­telling exercises, and engage in efforts t­oward reconciliation between former enemies. National and international donors have incorporated transitional justice in their policies on ­human rights, peace-­building, and economic reconstruction. Notwithstanding—­and maybe ­because of—­the steep 75

76

pa r mentier , r ausc henb ac h, a nd va n c r a en

and rapid expansion of this field of study, the last de­cade has witnessed the emergence of many critiques of transitional justice. One of the most fundamental relates to the production of knowledge about transitional justice issues in a variety of countries and situations. In many cases, academic knowledge as well as local and central policies are developed by elites “from above,” without duly taking into account the attitudes, opinions, and expectations of populations on the ground. Not only does this approach result in a very partial understanding of real­ity in complex settings, but this “disconnect” also frequently leads to failures in the implementation and impact of such policies. The seminal book Transitional Justice from Below: Grassroots Activism and the Strug­gle for Change edited by Kieran McEvoy and Lorna McGregor1 has called attention to exploring and developing transitional justice “from below.” In the last de­cade, the outcomes of a dozen or so empirical studies have allowed for a deeper understanding of the attitudes, opinions, and expectations of populations in relation to ­these very complex issues. This chapter is composed in the same vein and aims at contributing to an epistemology “from below” in the field of transitional justice. For this purpose, it first summarizes the key features of population-­based research conducted in a quantitative manner and then reports on some salient findings of our own survey administered in Bosnia in 2006. We conclude by highlighting the strengths and weaknesses of t­ hese methods, and in ­doing so, we advocate for the development of an approach to transitional justice issues that is grounded in I(nformation), D(ialogue), and P(rocess).

population-­based research on transitional justice In the last de­cades, several population-­based studies on issues of transitional or postconflict justice have been produced. In essence, they can be said to take two forms. The vast majority of ­these studies are of a quantitative nature and are aimed at obtaining a broad overview of the opinions of the general population, or a specific subsection thereof. The latter is the case when no representativity can be attained (e.g., in the absence of a reliable population census), or when specific groups such as victims of sexual vio­lence, internally displaced persons, sentenced offenders, and so on are being specifically targeted. In addition, qualitative types of empirical research are burgeoning. In them, small numbers of persons are being interviewed with an objective of providing “thick” forms of understanding with regard to their experiences, attitudes, knowledge, affections, be­hav­ior, and expectations in relation to vio­lence and postviolence. While some of t­ hese studies use in-­depth interviewing and open questionnaires, other such studies take the form of oral history accounts that allow the interviewed to freely speak about their experiences and ideas.2 In this contribution, we w ­ ill limit ourselves to focusing only on quantitative research that tries to compose a representative



The IDP Approach to Transitional Justice 77

image of transitional justice issues and to singling out the issue of truth-­seeking as a prime field of consideration. Quantitative Surveys: Aims and Features In general terms, quantitative surveys in transitional justice aim to provide more information about ­people’s opinions relating to international crimes and serious ­human rights violations, dealing with the past, and (re-)constructing the ­future. In the last ten years or so, several such surveys have been conducted in a wide variety of countries, including (but not limited to) Af­ghan­i­stan, Bosnia, Cambodia, the Central African Republic, Colombia, the Demo­cratic Republic of Congo, ­Kenya, Liberia, Serbia, Sri Lanka, and Uganda. While the precise format of the surveys may differ substantially, their overall research design shares some common features that are worth highlighting:

• As to the content, surveys of this kind most often ask questions about two major aspects: (1) what the profile of the respondents is (i.e., age, gender, ethnicity, religion, educational level, economic status, experience of victimization, involvement in violent conflict, and so on), and (2) what the respondents think and feel and how they act in relation to issues of transitional justice. • In terms of instruments, surveys of this kind most often make use of semi-­ structured questionnaires with a combination of closed questions (sometimes using scales) and open questions. • The administration of the questionnaire can take several forms, but the main tendency is to approach respondents on a face-­to-­face basis and train interviewers accordingly; other models include delivering questionnaires to respondents and picking them up ­later, and to a much lesser extent, questionnaires sent by post. • The group of respondents can be very dif­fer­ent, although surveys of this kind invariably target specific subsections of the general population of a given country or region, and often aim at composing a representative sample. • Nearly all quantitative surveys in transitional justice have involved a combination of one or more international experts and research teams, as well as local staff members to co-­design the questionnaire, co-­conduct the interviews, and co-­analyze the results. A Key Aspect of Transitional Justice: Truth-­Seeking As transitional justice is often associated with the fight against impunity, it is worthwhile to highlight the reports of the two special experts entrusted by the former U.N. Commission on ­Human Rights (now ­Human Rights Council) to systematize and improve the princi­ples and practices for the b­ attle against impunity. Both Louis Joinet and Diane Orentlicher consider the “right to know” about the facts of the past, both for individuals and for society at large, to be one

78

pa r mentier , r ausc henb ac h, a nd va n c r a en

of the three major pillars of this current fight.3 The other two are the “right to justice,” in the sense that perpetrators are prosecuted and tried before a court of law, and the “right to reparation” for victims for the harm inflicted upon them. The “right to know” thus belongs to the core aspects of transitional justice around the world and takes the form of vari­ous initiatives and mechanisms for truth-­seeking and truth-­telling. One of the prime mechanisms that can effect the right to know are truth commissions.4 ­These differ from criminal prosecutions and ­trials in that their main objective is to produce as many facts about the past as pos­si­ble using nonjudicial procedures. In this sense, they contribute primarily to “historical justice” rather than “­legal justice.”5 Truth commissions come in dif­fer­ent shapes and colors, but in general they adhere to some basic princi­ples distinguished by Priscilla Hayner: (1) they are concerned with past events, not pres­ent ones; (2) they investigate patterns of events, rather than individual events, over a certain period of time; (3) they engage in a direct and broad manner with the p­ eople who are affected and collect information about their perceptions and needs; (4) they are temporary, not permanent, bodies that are supposed to produce a final report; and (5) they enjoy the support and/or the powers provided by the state being investigated.6 The latter feature is generally regarded as crucial b­ ecause truth commissions operate within a state context and rely on public resources for their functioning and public support for the implementation of their recommendations. Th ­ ose truth-­telling initiatives that lack state support belong to a wide range of other, unofficial forms of h­ uman rights investigations that include reporting and pos­si­ble actions by NGOs, trade u­ nions, churches, and the like.7 Truth commissions may use very dif­fer­ent terminologies that reflect their specific objectives and operational methods. The names can refer to a specific set of victims, like the National Commission on the Dis­appeared in Argentina, or a specific set of objectives, like the Truth and Reconciliation Commission in South Africa, the Commission for Historical Clarification in Guatemala, and the Equity and Reconciliation Commission in Morocco.8 In our own heuristic model of postconflict justice, truth and truth-­seeking constitute one of the key issues that new elites have to address ­after a regime change or in the broader frame of addressing large-­scale ­human rights abuses.9 Other key issues include the accountability of offenders, reparations for victims, and reconciliation between former enemies.

a population-­based survey in bosnia and herzegovina Following the general framework of truth-­seeking and the specific format of truth commissions, we would like to illuminate some findings of a quantitative survey conducted in Bosnia and Herzegovina. This survey was administered in



The IDP Approach to Transitional Justice 79

2006 as part of a larger research proj­ect in Bosnia-­Herzegovina (hereafter BiH) and in Serbia focusing on mass victimization and restorative justice and was supported by the Research Council of the University of Leuven. As highlighted elsewhere,10 we first provide a short sketch of the violent conflict in BiH and of how transitional justice was practiced or thought of when this survey data was collected. We then report on the main results of the survey in relation to the issue of truth-­seeking. War and Postwar in Bosnia and Herzegovina In early 1992, following the breakup of Yugo­slavia, Bosnia and Herzegovina (BiH) plunged into a devastating war that lasted for almost four years and took an enormous toll on its population, infrastructure, and cultural heritage. Recent estimates point to around 100,000 deaths and to 2.2 million displaced ­people. Atrocities such as mass murders, extrajudicial executions, torture, rape, illegal detention, forced displacement, looting, and destruction of religious and cultural sites are fairly well documented, though responsibilities for them continue to be contested. As reports of such wanton destruction and humanitarian crisis became widespread, the concern of the international community grew. However, it was not ­until December 1995 that the Dayton Peace Agreements ­were signed, officially ending the war in BiH. ­These agreements created a complex po­liti­cal and territorial structure in order to respond to the parties’ demands. They divided the country into two entities: the Federation of Bosnia and Herzegovina, inhabited predominantly by a Bosnian Muslim and Bosnian Croat population, and the Republika Srpska, inhabited predominantly by Bosnian Serbs. The agreements also provided for a far-­reaching intervention of the international community in both civilian and military affairs in postwar BiH. The involvement of the international community in transitional justice issues began even before the end of the war, with the creation in 1993 of the International Criminal Tribunal for the former Yugo­slavia (ICTY) by Resolution 827 of the U.N. Security Council. The establishment of this judicial mechanism while the war was still raging in BiH was from the onset justified by evoking its intended contribution to vari­ous goals, such as the restoration and maintenance of peace or putting an end to impunity through the promotion of deterrence. Its main targets of prosecution ­were ­those bearing the highest levels of po­liti­cal and military responsibility for the international crimes committed during the conflict.11 In the 1990s, justice-­ making at the international level was favored over national justice mechanisms, given the latter’s lack of resources and of functional institutional structures. Yet, in the face of the sheer mass of criminal cases to be pro­cessed, by 2003 the necessity for complementary trial-­based mechanisms to be devised at the level of national jurisdictions was rapidly acknowledged. Vari­ous national courts with a jurisdiction over international crimes have since been created, including the

80

pa r mentier , r ausc henb ac h, a nd va n c r a en

War Crimes Chamber of the State Court of Bosnia and Herzegovina (hereafter War Crimes Chamber [WCC]) and sixteen other entity courts across BiH. Thus, even ­after the ICTY closed and transferred some of its activities to the Mechanism for International Tribunals (MICT), war crimes prosecutions are likely to continue within the national courts for an extended period of time, given the number of existing cases.12 In addition to t­hese trial-­based mea­sures, nonjudicial transitional justice mea­sures (e.g., memorialization initiatives, fact-­finding efforts, mechanisms for documenting war crimes, and attempts to establish a national and regional truth commission) have also been developed or envisaged in BiH. While some initiatives are considered to have achieved their goal locally, many ­others have been criticized for having failed to attain their purported objectives in terms of reconciliation, for reflecting one-­sided and polarized interpretations of the past, or even for having increased interethnic divisions.13 The RECOM, a regional truth-­ finding commission involving a co­ali­tion of civil society actors from several Balkan countries including BiH, well illustrates such dynamics.14 Since its launch in 2008, this initiative has been mired in internal disagreements due to its initiators’ conflicting interests in justice, grounded in competing po­liti­cal orientations, exclusive definitions of victimhood, or a tendency to focus solely on specific constituents’ needs. Adding to t­ hese divides, this proj­ect also suffered from the outset from a lack of support from the po­liti­cal and public sphere. Despite ­these many challenges, t­ here are still hopes that the RECOM, which has been stagnating since the drafting of its statute, may in the near f­uture move forward and contribute to the integration of competing narratives of victimhood into a comprehensive account of the past. Some have argued that BiH should start with a national truth-­finding commission before engaging in a regional mechanism for ­doing so. Such contentions are based on BiH’s previous experiments with national commissions.15 The Srebrenica Commission was the first one created by the Republika Srpska government in 2003 to investigate the atrocities committed in Srebrenica between July 10 and 19, 1995. While it did establish significant facts concerning the mass atrocities that occurred in Srebrenica, such as the number of individuals killed during this ten-­day period, it fell short of recognizing that ­these crimes amounted to a genocide. Two other failed attempts can also be mentioned: the Sarajevo Commission set up in 2006 and the Bijeljina one established in 2008. Both commissions stopped functioning in the early stages of their implementation due to divisions about the nature of their mandates and their scope. Thus, at the time when our BiH population-­based survey was administered in 2006, transitional justice initiatives, w ­ hether judicial or nonjudicial, did not benefit from much support at the grassroots level and w ­ ere increasingly criticized within scholarship. Th ­ ere w ­ ere concerns that transitional justice was externally imposed by vari­ous actors, ­whether international or national, with insufficient consideration for context-­specific expectations of justice and for the needs of



The IDP Approach to Transitional Justice 81

their intended beneficiaries.16 This imposition from above was considered all the more problematic in that it exacerbated the power strug­gles for recognition between competing narratives of the past in this complex postwar matrix. Recent scholarship shows that such divisions and strug­gles for recognition of multiple and contrasting understanding of the past are also pres­ent in the current Bosnian justice landscape.17 We can thus safely assume that truth-­finding and the acknowl­ edgment of vari­ous understanding of the past ­were prime concerns back in 2006 and that they remain significant in ­today’s Bosnian memory politics. Methodology of the Survey The quantitative survey made use of a self-­administered written questionnaire distributed to nine hundred respondents all over Bosnia and Herzegovina. As described at length in other publications, the methodology consisted of four key steps:18

Creating an instrument of data-­gathering. The questionnaire was developed by the members of the Leuven research team in the period between February and May 2006, on the basis of a number of sources. The printed questionnaire included a total of thirty-­eight questions, some with mutually excluding answers, some with multiple pos­si­ble answers, and some with options for respondents to rank their level of agreement with alternative answers given to the formulated question. All of ­these questions ­were closed or semi-­closed with one exception: the very last question, “What does reconciliation mean to you?” allowed the respondents to give their personal opinion in a very open and indeterminate way. 2. Determining an adequate sampling method. The absence of a population census in BiH before the war and the impossibility of creating an accurate national representative sample prompted the choice of a “quota sampling method.”19 With this method, we divided the target population—­the citizens and residents of Bosnia—­into subgroups or strata according to four criteria of par­tic­ u­lar interest to us: geo­graph­ic­ al distribution, religious affiliation, age, and gender. Regarding the religious affiliation of the respondents, which is often considered a key ele­ment in the violent conflict, we deci­ded to give equal weight to the voices of each religious group and opted for a “disproportionate sample” with an underrepre­sen­ta­tion of Muslims (Bosnian Muslims) and Orthodox (Bosnian Serbs) and an overrepre­sen­ta­tion of Catholics (Bosnian Croats) in relation to the estimate composition of the country. 3. Collecting the data through the distribution of the questionnaire. The data was collected in the course of May and June 2006 by students of the University of Sarajevo, BiH, all of whom had some background in social science research methods. They distributed the questionnaires in their home towns or villages to persons fitting the se­lection criteria mentioned above and also following 1.

82

pa r mentier , r ausc henb ac h, a nd va n c r a en

the “snowball method,” wherein the data collectors could rely on referrals by some respondents in order to find additional respondents whenever necessary. All questionnaires w ­ ere handed to the respondents in person and retrieved from them ­after a few days, with sufficient guarantees for anonymity. Of the total number of 900 questionnaires distributed, 855 w ­ ere returned duly filled, thus providing a 94,4 ­percent response rate, a high number typical of “con­ve­nience sampling methods.”20 The open question on reconciliation was filled out by 650 respondents. 4. Analyzing the data gathered. With the help of Stastical Package for the Social Sciences (SPSS) software, univariate, bivariate, and multivariate analyses, ­factor analy­sis, cluster analy­sis, and tests of the reliability of scales w ­ ere performed. All figures presented in the following paragraphs are the result of this four-­step pro­cess of data gathering and data analy­sis. It should be clear from the outset that all questions w ­ ere asked to all the respondents in the survey and not only to ­those who might be legally or so­cio­log­ic­ ally qualified as “victims.” The reason is that in violent conflicts like the Yugo­slav one it is extremely difficult, if not impossible, to make such clearcut distinctions between ­those who are only perpetrators and t­ hose who are only victims, since individuals can and sometimes do assume dif­fer­ent roles in dif­fer­ent phases of a conflict. As a result, our research should be qualified as a “population-­based survey,” not as a mere victim survey. As indicated elsewhere, the questionnaire itself consisted of questions centered on two major parts:21 Experiences of victimization. As it is very likely that each person’s experiences during the war may have an impact on his or her current opinions of and attitudes ­toward issues of postconflict justice, a first set of questions related to the victimization suffered by the respondents. They probed for (a) dif­fer­ent types of victimization, including direct (e.g., physical injuries on the respondent) and indirect (e.g., having lost ­family members); (b) several categories of victimization, including physical, material, and psychological; and (c) the temporal framework, by relating the types and categories to two time periods, during and a­ fter the war. Overall, the responses to ­these questions allowed us to conclude that emotional suffering, both during and a­ fter the war, came out as the prime form of victimization, followed by material harm and physical harm. 2. Opinions about key issues of postconflict justice. The second part of the questionnaire had vari­ous questions and subquestions u­ nder four dif­fer­ent headings: seeking truth, establishing accountability for offenders, providing reparation for victims, and promoting reconciliation. Th ­ ese headings corresponded to the heuristic model that was developed by the Leuven team and served as its conceptual framework for the survey. 1.



The IDP Approach to Transitional Justice 83

The General Meaning of Truth-­Seeking in Bosnia and Herzegovina The violent conflict in BiH and other parts of ex-­Yugoslavia has to a large extent been fueled by the manipulation of facts and public opinion by all sides, and one of the main obstacles to dealing with the past remains the continued denial of the facts, coupled with multiple and contradictory versions of the truth. This was the case back in 2006 and remains equally so more than a de­cade ­later. For ­these reasons, truth-­seeking can be considered a crucial mechanism to establish the foundations of a shared ­future for the citizens of Bosnia and Herzegovina. ­Here, we limit ourselves to the most salient findings related to truth-­seeking, while much more information on a variety of topics can be found in the original survey report.22 The survey first tried to understand w ­ hether establishing the truth was impor­tant to the respondents. Of the total number of 850 respondents, a vast majority—89 ­percent—­considered the establishment of the truth as one of the main functions of prosecutions. An equally large majority of 85 ­percent said that it would make them feel better if the truth about all facts and events of the war ­were known. It then went on to ask questions about the means and pro­cesses for the truth to be established, with par­tic­ul­ ar attention to opinions about informal, small-­scale initiatives of truth-­telling. Out of 850 respondents, the largest group (44 ­percent) answered that both victims and perpetrators should be given a chance to tell their stories, while 22 ­percent said only victims should be given a chance to tell their stories, and a small group (2 ­percent) said only the perpetrators should be given a chance to tell their stories. Thus, two thirds (68  ­percent) of the respondents underscored the importance of p­ eople telling about their war experiences, as opposed to 15 ­percent who w ­ ere of the opinion that it was not impor­tant to do so and an almost equal (and considerable) number (18 ­percent) of respondents who did not know what to answer. Given the strong religious cleavages during the violent conflict in BiH, it was worth investigating if ­these opinions would differ according to religious affiliation. In fact, 80 ­percent of the Muslim respondents said that it was impor­tant for p­ eople to describe their war experiences; relatively speaking, this was a more sizeable proportion than the other religions. Only 7 ­percent of the Muslim respondents considered it not impor­tant, and 10 ­percent did not know what to answer. Also, more Muslim respondents than ­those belonging to other religions thought that only victims should be given a chance to tell their stories (32  ­percent, against 18  ­percent of Catholics and 17 ­percent of Orthodox). The Forums for Truth-­Telling The respondents w ­ ere also asked where they thought the stories about the war should be told and to rate their most preferred and least preferred places for truth-­ telling. We aggregated the values “agree” with “strongly agree” on the one hand,

84

pa r mentier , r ausc henb ac h, a nd va n c r a en

and the ones of “disagree” and “strongly disagree” on the other hand. As ­table 4.1 indicates, the most preferred places for truth-­telling w ­ ere the courts (83 ­percent), followed by a truth and reconciliation commission (TRC) (71  ­percent), then psychologists (62 ­percent), and then public events (such as roundtables or workshops). Slightly less than half of the respondents preferred small groups in the community in which members of other ethnic groups ­were pres­ent (48 ­percent), 34 ­percent thought ­these stories should be told only to f­ amily and friends, and a smaller number (27 ­percent) said they should be told in small groups in the community in which members of other ethnic groups w ­ ere not pres­ent. The missing cases per type of response range between 9 ­percent and 16 ­percent. In terms of religious affiliations, as shown in ­table  4.2, ­there was a relative majority of Muslims among ­those who thought that the stories should be told in the courts, although the difference between the three main religious groups was not very significant. Differences w ­ ere more notable in the case of “public events” and “small groups in the community in which members of other ethnic groups are pres­ent,” which ­were supported by a majority of Muslims. ­There also was a significant difference between the percentage of Muslims who supported truth-­ telling in a TRC (75 ­percent) and that of Orthodox respondents (62 ­percent). The latter ­were quite supportive of telling t­ hese stories only to ­family and friends (40 ­percent, in comparison with 28 ­percent Muslims and 35 ­percent Catholics). ­Because our research proj­ect was particularly focused on the potential of restorative justice a­ fter violent conflict and mass victimization, we devoted considerable attention to collecting opinions related to setting up a pos­si­ble truth and reconciliation commission in BiH. This was all the more relevant, given the establishment of the Srebrenica Commission in 2003 and the discussions about two other commissions, which ­were subsequently established in 2006 and 2008, as mentioned previously. First of all, the respondents ­were asked if they ­were aware of the proj­ect to set up a truth and reconciliation commission in BiH; its objective was to determine to what extent this proj­ect had been disseminated at that time. Out of a total number of 845 respondents, an absolute majority (58 ­percent) said they had not heard about it, while a minority (42 ­percent) said they had. Given the uncertain status of the Bosnian TRC proj­ect at the time, the other questions explored the knowledge of respondents about truth and reconciliation commissions in general. When asked w ­ hether they knew what a truth and reconciliation commission was, two-­thirds (66 ­percent) of the respondents said they did not know, while one-­third (34 ­percent) said they knew (total number of respondents equalling 841). Of ­those indicating knowledge of a TRC (388 respondents), a large majority of about three-­quarters (73  ­percent) responded that it could be useful, with a subgroup of 31 ­percent saying that it could be useful in general, and another of 42 ­percent indicating that a TRC would not be useful for them

­table 4.1

The IDP Approach to Transitional Justice 85

Preferred Places for Truth-­Telling—­Survey BiH 2006

Courts1 TRC Psychologists Public events2 Small community groups with members of other ethnic groups Only ­family and friends Small community groups without members of other ethnic groups

Strongly disagree

Disagree

Agree

Strongly agree

­Don’t know

5 9 14 24 28

5 5 10 10 10

8 12 13 13 12

75 59 48 40 37

7 16 15 14 14

40 45

11 12

10 10

24 18

15 16

source: Marta Valinas, Stephan Parmentier, and Elmar Weitekamp, “ ‘Restoring Justice’ in Bosnia and Herzegovina: Results of a Population-­Based Survey,” Working Paper No. 31 (Leuven: Leuven Centre for Global Governance Studies, 2009), https://­ghum​.­kuleuven​.­be​/­ggs​/­publications​/­working​_­papers​ /­new​_­series​/­wp31​-­40​/­wp31​.­pdf. 1. Missing cases: Courts 9%; TRC 10%; Psychologists 11%; Public events 13%; Multi-­ethnic community group 13%; ­Family and friends 16%; Mono-­ethnic community groups 16%. 2. In the questionnaire, “roundtables or workshops” ­were given as examples for public events.

­table 4.2

Cross-­Tabulation between Places for Truth-­Telling and Religious Affiliation—­Survey BiH 2006

Courts Public events Small community groups with members of other ethnic groups Small community groups without members of other ethnic groups TRC Psychologists Only ­family and friends

Catholic

Muslim

Orthodox

83 48 44

90 66 60

75 44 39

23

38

19

73 59 35

75 64 28

62 61 40

source: Valinas, Parmentier, and Weitekamp, “ ‘Restoring Justice’ in Bosnia and Herzegovina.” note: Values shown are percentages within religious affiliation and for responses “agree” and “strongly agree” taken together.

personally but for o­ thers. A small group (13 ­percent) thought the TRC would not be useful, and another small group (14  ­percent) said they did not know the answer to this question. Th ­ ese figures are spelled out in t­ able 4.3 below. In terms of religious affiliation, no major variation among ­these answers could be identified. The only exception was that Orthodox respondents, who found that the TRC could be useful, tended to find it more useful for ­others than for them

86

pa r mentier , r ausc henb ac h, a nd va n c r a en

­table 4.3

Usefulness of a TRC in the Eyes of ­Those Respondents Knowing a TRC—Survey BiH 2006

Yes, not personally but possibly for ­others Yes No ­Don’t know Total

n

%

161 122 50 55 388

42 31 13 14 100

source: Valinas, Parmentier, and Weitekamp, “ ‘Restoring Justice’ in Bosnia and Herzegovina.”

personally, whereas among Catholic and Muslim respondents, a very similar number found the TRC useful ­either for themselves personally or for ­others. Also, slightly more Orthodox respondents (18 ­percent) reported that they did not find a TRC useful, as compared to Muslims (12 ­percent) and Catholics (8 ­percent). Furthermore, the respondents ­were asked what they thought a truth and reconciliation commission should be responsible for ­doing (results in ­table 4.4). All but one of the alternative answers presented to the respondents ­were based—­ according to an article in the Bosnian press23—on the contents of the draft law ­under preparation at the time of this survey in 2006. Given the controversy surrounding this topic, it came as a surprise that a ­great majority of the respondents seemed to agree with the ele­ments of the proposed mandate of a Bosnian truth commission: 80 ­percent thought it should “analyze what led to the ethnic distrust and to the war”; 80 ­percent thought it should “establish the number and identity of the victims”; 78 ­percent agreed it should “establish the moral responsibility of organ­izations and institutions in the course of the war”; 77  ­percent thought it should “establish the role of actors outside Bosnia and Herzegovina in the war”; 75 ­percent ­were of the opinion that it should establish “the moral responsibility of individuals in the course of the war”; 74 ­percent thought it should “promote reconciliation”; 71 ­percent said it should “provide a forum for ­people to tell their experiences”; and fi­nally 63 ­percent said it should “establish the role of individuals who refused to take part in the vio­lence and protected other individuals.” The alternative answer to this question and not included in the mandate of the draft law was ­whether the commission should “provide a forum for p­ eople to tell their experiences.” The responses w ­ ere corroborated by ­those given to a previous question on where experiences of the war should be told, with 71 ­percent of the respondents saying that this should be done in a truth and reconciliation commission. When interpreting ­these figures, it should be kept in mind that not more than 40 ­percent of the total group of respondents actually responded to ­these questions, thus leaving a large margin for alternative interpretations.

­table 4.4

The IDP Approach to Transitional Justice 87

Mandate of a TRC for Bosnia—Survey B­ iH 2006

Establish the number and identity of victims Establish the role of actors outside Bosnia-­Herzegovina in the war Analyze what led to ethnic distrust and to the war Promote reconciliation in Bosnia-­Herzegovina Establish the moral responsibility of organ­izations and institutions in the course of the war Establish the moral responsibility of individuals in the course of the war Provide a forum for ­people to tell their experiences about the war Establish the role of individuals who refused to take part in the vio­lence and protected other individuals or groups

Strongly disagree

Disagree

Agree

Strongly agree

­Don’t know

9

3

8

71

8

9

4

6

70

10

7

4

11

69

10

10

6

8

66

10

7

4

14

64

10

10

5

14

61

10

10

7

11

60

12

17

6

12

51

14

source: Valinas, Parmentier, and Weitekamp, “ ‘Restoring Justice’ in Bosnia and Herzegovina.”

Based on the respondents’ opinions in relation to the most impor­tant tasks of a truth and reconciliation commission, we also arranged a list of their responses according to their priorities:24 1.

Establish the number and identity of victims. Establish the role of actors outside of Bosnia in the war. 3. Analyze what led to the ethnic distrust and the war. 4. Promote reconciliation. 5. Establish the moral responsibility of organ­izations and institutions in the course of the war. 6. Establish the moral responsibility of individuals in the course of the war. 7. Provide a forum for ­people to tell of their experiences. 8. Establish the role of individuals who refused to take part in the vio­lence and protected other individuals. 2.

88

pa r mentier , r ausc henb ac h, a nd va n c r a en

Summary Conclusions on Truth-­Seeking Stemming from the Population-­Based Survey in Bosnia and Herzegovina From the many quantitative data collected through our population-­based survey in BiH in 2006, a number of salient findings in relation to the mechanism of truth-­seeking can be reported. First of all, it was striking to find out that respondents reported a high degree of emotional suffering, both during the war and afterward, and even placed this type of harm over material and physical harm. While it may have been expected in violent conflicts of this type, it still could be called surprising that none of the national or international mechanisms to deal with the past had addressed this harm in any way, but ­were instead heavi­ly focused on trying and punishing perpetrators of the crimes committed. Second, it became clear that the debate on truth-­seeking was characterized by a significant degree of controversy. The vast majority of respondents attached high importance to the establishment of the truth, both in judicial proceedings and other forums. But the question of how to establish the truth continued to divide the opinions of the respondents, with a larger group thinking inclusively that both victims and perpetrators should be given a chance to tell their experiences, a smaller group considering only the victims, and another group downplaying the importance of telling stories at all. ­These responses seem influenced by the religious affiliation of the respondents. Third, on the issue of the forums for truth-­telling, the answers suggested that respondents tended to f­avor more institutionalized and formalized forums of truth-­seeking, like courts or a truth commission, which provide an official validation of the stories told. However, more than half of the respondents also believed that public events and community groups, especially ­those with a multiethnic composition, would be good places to share such experiences. Again, most striking was the high value attached to psychologists as listeners for their stories, which would be congruent with the high degrees of emotional harm and psychological trauma that persisted in the postwar period. Also, some significant patterns could be observed along religious lines, with Muslims preferring truth-­ telling forums that w ­ ere public or involved members of other ethnic groups, and Orthodox respondents favouring more private settings. Fourth, only one-­third of the respondents seemed to know about the proj­ect to set up a truth and reconciliation commission in Bosnia and Herzegovina, despite the fact that discussions had been ongoing for quite some time. And despite their limited knowledge and the many controversies surrounding the issue, a ­great majority of ­those respondents who responded to ­these questions (around 40 ­percent of the w ­ hole sample due to high missing data levels on t­ hese questions) agreed with the proposed mandate of a Bosnian Truth Commission.



The IDP Approach to Transitional Justice 89

t­ oward the “idp approach” in transitional justice The preceding sections have clearly demonstrated the aims and operational methods of quantitative surveys that collect data about international crimes and transitional justice with large populations or specific sections thereof. Using our own survey in Bosnia and Herzegovina in 2006, they have also highlighted some possibilities for exploiting very rich data, as well as vari­ous limitations of such interpretations. It is therefore worthwhile highlighting the strengths and weaknesses of such surveys and reflecting on the possibility of developing a more systematic approach to transitional justice. It is crystal clear that a quantitative approach to transitional justice has several advantages. First of all, it allows many and diverse “voices from the field”25 to be captured and acknowledged and maybe even a representative view of a population, alongside of the opinions of national and international elites. In consequence, quantitative surveys provide an additional source of information about all aspects questioned, including the victimization of persons during and ­after violent conflict, and specific issues of postconflict justice, like truth-­seeking, offender accountability, victim reparations, and reconciliation. The very act of asking information from ordinary ­people also constitutes a specific form of acknowl­edgment of their victimhood, even if they also belong to the group of offenders. Depending on the precise administration of the surveys, the information generated can be extremely useful in formulating new initiatives and general policies to be designed and implemented by local, national, and international bodies. On the other hand, surveys of this nature also suffer some weaknesses that should be taken into account. E ­ very survey is context-­specific, not only b­ ecause it focuses on specific issues, but also b­ ecause of its specific methodological choices. Overall, transitional justice issues remain very complex and require a “thick” understanding of many layers of experiences, attitudes, opinions, and expectations, which general and superficial surveys cannot deliver. Moreover, all of ­these findings may change over time and thus require a dynamic interpretation as ­people change their minds constantly. The general consequence is that survey findings can hardly be compared, which undermines the potential for engaging in cross-­case and cross-­time comparisons. ­These difficulties also generate prob­lems when trying to feed survey information into a pro­cess of policy-­making. With the glass half empty, some may be tempted to throw the remaining ­water out and get rid of quantitative surveys altogether. In our view, such an approach would come down to throwing out the baby with the bathwater and thus losing on all accounts. For this reason, we prefer to consider the glass as half full and in need of more fillings. Hence, our ambition is to develop an approach that ­will make maximum use of quantitative surveys without being hampered by their limitations. The proposed IDP approach serves this purpose well, as it is

90

pa r mentier , r ausc henb ac h, a nd va n c r a en

grounded on three pillars. The first pillar is information (I) and acknowledges that survey findings constitute very in­ter­est­ing sources of understanding that complement mainstream thoughts or elite views. The second pillar, dialogue (D), assumes that pure information is insufficiently sophisticated to be applied right away and requires some form of exchange through individual and collective means and media in order to increase the understanding of all. The third pillar, pro­cess (P), is related to the other two and is concerned with which survey data can be delivered and discussed and may fi­nally serve for practice and policy-­making. We advocate the development of an IDP approach to transitional justice for the same two basic reasons that public opinion research in general can be considered useful:26 it provides additional social legitimacy to policy-­making, and it likely increases the effectiveness of societal institutions. For t­hese high ambitions to be fulfilled, it would make sense to further develop the empirical cycle of combining both quantitative and qualitative approaches.27

concluding reflections In conclusion, we would like to highlight the central role played by understandings of the past when assessing justice stakeholders’ needs and expectations in the aftermath of a violent conflict. Historical memories are likely to be particularly significant ­drivers in highly divided socie­ties such as BiH, where justice claimants constitute a multifaceted cluster of stakeholders with heterogeneous needs and claims.28 Such heterogeneity in justice concerns is often grounded in dif­fer­ent repre­sen­ta­tions of the violent past and contingent on identity divides. ­These conflicting identities and repre­sen­ta­tions are often suggested to have generated and promoted intergroup conflict in the first place and to be instrumentalized in order to maintain divisions a­ fter the war. Given their significance, we argue that ­these pro­cesses should be given more consideration when assessing, through grassroots perceptions, the potential of transitional justice mea­sures to contribute to the restoration of social ties. Thus, we contend that capturing the dif­fer­ent meanings afforded to justice and their complexity requires devising quantitative surveys that account for the role of par­tic­ul­ ar understanding and experiences of the past in determining how ­people perceive justice needs in the pres­ent and their implications for the ­future. This allows for a better assessment of the role of understandings of the past in shaping the many nuances in meanings that can be attributed to justice. As argued elsewhere,29 transitional justice scholarship would also benefit from a deeper understanding of the dynamic and historical pro­cesses grounding perceptions of justice in postwar settings by using mixed method approaches involving both qualitative and quantitative methodologies. Fi­nally, when assessing the potential of transitional justice mea­sures to contribute to the restoration of social ties within highly divided socie­ties such as



The IDP Approach to Transitional Justice 91

BiH, one should also bear in mind that such initiatives are significant channels for the expression and dissemination of narratives of the past. Yet, within divided postwar settings, they are also likely to constitute politicized sites of contestation of the violent past. Transitional justice pro­cesses have been criticized for privileging dominant narratives of the past, which tend to marginalize the experiences of subportions of the collectives in the name of whom they w ­ ere developed.30 Such hegemonic pro­cesses run the risk of imposing authoritative readings of the past that can create or maintain hierarchies of victimhood,31 as well as deny the past suffering for other unrecognized justice claimants.32 When exploring justice perceptions in the aftermath of atrocity, accounting for the existence of vari­ous and often conflicting or marginalized needs and interests is thus, in our view, essential. ­Doing so is likely to increase the chances of justice-­making that is contextually appropriate and fits the diversity of experiences and claims of justice stakeholders.

notes 1. ​Kieran McEvoy and Lorna McGregor, eds. Transitional Justice from Below: Grassroots Activ-

ism and the Strug­gle for Change (Oxford: Hart Publishing, 2008). 2. ​Mina Rauschenbach, Stef Scagliola, Stephan Parmentier, and Franciska de Jong, “The Perfect Data-­Marriage: Transitional Justice Research and Oral History Life Stories,” Transitional Justice Review 1, no. 4 (2016): 7–58, http://­ir​.­lib​.u­ wo​.c­ a​/­tjreview​/­vol1​/­iss4​/­2. 3. ​Question of the Impunity of Perpetrators of H ­ uman Rights Violations (Civil and Po­liti­cal), Revised Final Report Prepared by Mr. Joinet Pursuant to Sub-­Commission Decision 1996/119, UN Doc. E/CN.4/Sub.2/1997/20/Rev.1; In­de­pen­dent Study on Best Practices, Including Recommendations, to Assist States in Strengthening Their Domestic Capacity to Combat All Aspects of Impunity, by Professor Diane Orentlicher, UN Doc. E/CN.4/2004/88. 4. ​Stephan Parmentier and Monica Aciru, “The Whole Truth and Nothing but the Truth? On the Role of Truth Commissions in Facing the Past,” in Facing the Past: Amending Historical Injustices through Instruments of Transitional Justice, ed. Peter Malcontent, Series on Transitional Justice (Cambridge: Intersentia Publishers, 2016), 21:225–246. 5. ​Ruti Teitel, Transitional Justice (New York: Oxford University Press, 2000). 6. ​Priscilla Hayner, Unspeakable Truths: Transitional Justice and the Challenge of Truth Commissions (London: Routledge, 2011). 7. ​Stephan Parmentier, “Transitional Justice,” in The Cambridge Companion to International Criminal Law, ed. William Schabas (Cambridge: Cambridge University Press, 2016). 8. ​Extensive information about numerous truth commissions can be found on the website of the United States Institute of Peace: https://­www​.­usip​.­org​/­publications​/­2011​/­03​/­truth​-­comm​ ission​-­digital​-­collection. 9. ​Stephan Parmentier, “Global Justice in the Aftermath of Mass Vio­lence: The Role of the International Criminal Court in Dealing with Po­liti­cal Crimes,” International Annals of Criminology 41, nos. 1–2 (2003): 203–224; Stephan Parmentier and Elmar Weitekamp, “Po­liti­cal Crimes and Serious Violations of H ­ uman Rights: T ­ owards a Criminology of International Crimes,” in Crime and ­Human Rights, Series in Sociology of Crime, Law and Deviance (Amsterdam: Elsevier, 2007), 9:109–144. 10. ​Marta Valinas, Stephan Parmentier, and Elmar Weitekamp, “ ‘Restoring Justice’ in Bosnia and Herzegovina: Results of a Population-­Based Survey,” Working Paper No.  31 (Leuven:

92

pa r mentier , r ausc henb ac h, a nd va n c r a en

Leuven Centre for Global Governance Studies, 2009), https://­ghum​.­kuleuven​.b­ e​/­ggs​/­publi​ cations​/­working​_­papers​/­new​_­series​/­w p31​-­40​/­w p31​.­pdf. 11. ​Such objectives are illustrated in the prosecution of Radovan Karadzic who was found guilty of genocide and war crimes committed during the Bosnian War of 1992–1995 and sentenced to forty years in jail. 12. ​Based on estimates as of 2014, roughly 1,200 cases remained to be pro­cessed; see http://­ www​.­balkaninsight​ .­c om​ /­e n​ /­a rticle​ /­b osnia​ -­f ailing​ -­to​ -­reduce​ -­n umber​ -­o f​ -­w ar​ -­c rimes​ -­investigations, and the projected deadline for this to be accomplished was 2023, according to the War Crimes pro­cessing proj­ect (see http://­www​.­osce​.o­ rg​/­bih​/­106868​?­download​=­true). 13. ​Martina Fischer and Ljubinka Petrovic-­Ziemer, Dealing with the Past in the Western Balkans: Initiatives for Peacebuilding and Transitional Justice in Bosnia-­Herzegovina, Serbia and Croatia, Berghof Report, No. 18. (Berlin: Berghof Foundation, 2013); Arnaud Kurze and Iva Vukusic, “Afraid to Cry Wolf: ­Human Rights Activists’ Strug­gle of Transnational Accountability Efforts in the Balkans,” in Transitional Justice and Civil Society in the Balkans, ed. Olivera Simic and Zala Volcic (New York: Springer, 2013), 201–216. 14. ​Kurze and Vukusic, “Afraid to Cry Wolf.” 15. ​Janine Natalya Clark, International T ­ rials and Reconciliation: Assessing the Impact of the International Criminal Tribunal for the Former Yugo­slavia (New York: Routledge, 2014). 16. ​Xavier Bougarel, Elissa Helms, and Ger Duijzings, The New Bosnian Mosaic: Identities, Memories and Moral Claims in a Post-­War Society (Aldershot: Ashgate, 2007); Laurel E. Fletcher and Harvey M. Weinstein, “Vio­lence and Social Repair: Rethinking the Contribution of Justice to Reconciliation,” ­Human Rights Quarterly 24, no. 3 (2002): 573–639. 17. ​Nicolas Moll, “Fragmented Memories in a Fragmented Country: Memory Competition and Po­liti­cal Identity-­Building in T ­ oday’s Bosnia and Herzegovina,” Nationalities Papers 41, no. 6 (2013): 910–935. 18. ​Marta Valinas, Stephan Parmentier, and Elmar Weitekamp, “ ‘Restoring Justice’ ”; Stephan Parmentier, Marta Valinas, and Elmar Weitekamp, “How to Repair the Harm ­after Violent Conflict in Bosnia? Results of a Population-­Based Survey,” Netherlands Quarterly of H ­ uman Rights 27, no. 1 (2009): 27–44. 19. ​Alan Bryman, Social Research Methods (Oxford: Oxford University Press, 2008). 20. ​Ibid., 100. 21. ​Valinas, Parmentier, and Weitekamp, “ ‘Restoring Justice.’ ” 22. ​Ibid. 23. ​Nerma Jelacic and Nidžara Ahmetašević, “Experts Query Draft Law on How the Pro­cess Would Work, while Victims Complain They Have Not Been Consulted,” Balkan Insight, March 31, 2006. 24. ​This list is drawn on the basis of the sum of the percentages of t­ hose who said they “agreed” or “strongly agreed” with the given option. This method was used since the distribution of answers for each option among the pos­si­ble dif­fer­ent ratings was not, in statistical terms, a “normal distribution” and therefore it was not advisable to use the statistical means. 25. ​Stephan Parmentier and Piero Sullo, “Voices from the Field: Empirical Data on Reconciliation in Post-­War Bosnia and their Relevance for Africa,” in Victimological Approaches to International Crimes: Africa, ed. Rianne Letschert, Roelof Haveman, Anne-­Marie De Brouwer, and Antony Pemberton (Cambridge: Intersentia Publishers, 2011), 335–352. 26. ​Stephan Parmentier, Geert Vervaeke, René Doutrelepont, and Georges Kellens, eds., Public Opinion and the Administration of Justice: Popu­lar Perceptions and Their Implications for Policy-­Making in Western Countries, Proceedings of the International Colloquium, Leuven, September 25–27, 2003 (Brussels: Politeia Press, 2004). 27. ​Rauschenbach et al., “Perfect Data-­Marriage.”



The IDP Approach to Transitional Justice 93

28. ​Mina Rauschenbach, “Symbolic Forms of Transitional Justice for Social Restoration in Bos-

nia-­Herzegovina?,” paper presented at the “Trauma, Memory, and Healing in the Balkans and Beyond,” conference in Sarajevo, July 12–14, 2016. 29. ​Rauschenbach et al., “Perfect Data-­Marriage.” 30. ​McEvoy and McGregor, Transitional Justice from Below. 31. ​Neil Ferguson, Mark Burgess, and Ian Hollywood, “Who Are the Victims? Victimhood Experiences in Postagreement Northern Ireland,” Po­liti­cal Psy­chol­ogy 31, no. 6 (2010): 857–886. 32. ​Rezarta Bilali and Michael Ross,“Remembering Intergroup Conflict,” in The Oxford Handbook of Intergroup Conflict, ed. Linda. R. Tropp (New York: Oxford University Press, 2012), 123–135.

5 • THE SPARK FOR GENOCIDE? Propaganda and Historical Narratives at International Criminal Tribunals R I C H A R D A S H BY W I L S O N

As a result of the diffusion of the anti-­Tutsi propaganda, the killings “started off like a ­little spark and then spread.” —­Kayishema and Ruzindana Trial Chamber, §282

While prosecuting crimes against humanity, international criminal tribunals from the International Military Tribunal (IMT) at Nuremberg to the International Criminal Court (ICC) have also furnished accounts of the origins and ­causes of mass atrocities. The historical narratives of international tribunals exhibit a common feature that has not been remarked upon, and that is the central role they assign to po­liti­cal propaganda in explaining mass crimes. In par­tic­ u­lar, judges have looked to inciting speech to answer one of the most vexing questions at international criminal tribunals: Why did neighbors turn against neighbors and commit extreme acts of persecution and collective vio­lence, even in contexts characterized by long periods of coexistence. This chapter states that the evidence introduced in t­ rials does not support the claim made in international tribunals’ judgments that propaganda incited a population to commit genocide and/or crimes against humanity.1 It observes that the evidence from eyewitnesses for the causal role of propaganda is often slender and unconvincing. Victims and other fact witnesses often have l­ittle insight into the mobilization of perpetrators by leaders. Insiders and material perpetrators appearing for the prosecution frequently repudiate their original testimony ­after a concerted program of intimidation and bribery. Expert witnesses called by the prosecution to educate the courtroom on the context and consequences of propaganda have had a variable impact on the outcome. Judges at the International Criminal Tribunal for Rwanda (ICTR) have been receptive to prosecution expert testimony on the meaning of certain expressions in local languages and 97

98

ric h a rd a shby w ilson

the overall role of media in fomenting genocide. At the International Criminal Tribunal for the former Yugo­slavia (ICTY), judges have often balked at expert evidence on propaganda and refused to recognize it as germane to a criminal trial. This has had implications for the ability of the prosecution to document the deleterious effects of inciting speech, as was amply illustrated in the acquittal of Serb nationalist politician Vojislav Šešelj in 2016. Given the relative paucity of evidence for the directly causal role of propaganda, why have international courts emphasized po­liti­cal propaganda and inciting speech when explaining mass atrocities? What other kinds of historical narratives do ­these references to propaganda displace? How does the model of direct causation customarily used in criminal law shape the histories that international courts write of conflicts? Such questions are particularly relevant as the ICTY and ICTR Appeals Chambers conclude their final cases, and the historical legacy of t­ hese two international criminal tribunals assumes greater import. In answering, we must consider the unique model of causation used in criminal law, the apo­liti­cal nature of propaganda as an historical explanation, and the moral expressivist function of criminal courts.

propaganda in international criminal law Before addressing the approaches to propaganda of international criminal tribunals, it is necessary to briefly summarize the ongoing debate regarding the character and quality of l­egal histories of collective vio­lence. ­There exists a longstanding skepticism about law’s ability to write history on the part of scholars from diverse disciplines and theoretical persuasions.2 In the 1961 trial of Nazi bureaucrat Adolph Eichmann, Hannah Arendt famously protested the prosecution’s forays into the past, calling them “bad history and cheap rhe­toric.”3 For Arendt, bringing history into the courtroom undermined the right of the accused to due pro­ cess and procedural fairness, and with it the credibility of the law. In the French Holocaust t­ rials of Paul Touvier, Maurice Papon, and Klaus Barbie, critics such as French historian Henry Rousso contended that the aims and methods of law and history are so radically dissimilar that whenever courts try to address historical explanations, they inevitably fail.4 The inevitable outcome of l­egal enquiry into historical m ­ atters, it is argued, is a distorted version of history. While they highlight impor­tant distinctions between criminal law and academic history, ­these arguments are overdrawn, and they construe “law” and “history” as disembodied abstractions floating untethered in the intellectual firmament. In examining how history has been handled by the parties at three modern international tribunals (ICTY, ICTR, and ICC), we find instead that courtroom history appears not as a clash of rarified princi­ples, but as a terrain of contestation between ­legal actors who pursue defined and concrete ­legal outcomes. Prosecutors and defense counsel invoke historical evidence, arguments and expertise



The Spark for Genocide? 99

not out of a higher calling or to accomplish nonlegal (or chapeau) objectives such as leaving a historical rec­ord for posterity, but to achieve tangible and pressing ­legal objectives related to the immediate trial they are in. To fully appreciate l­egal histories of armed conflicts and mass atrocities, observers might usefully direct their attention to what the adversarial parties actually do with history during a trial. For instance, prosecutors turn to historical narratives to help their (inevitably) incomplete cases to cohere (or at least appear to cohere), and defense counsel summon up past atrocities as part of a necessity defense or tu quoque defense.5 In some instances, the historical narrative that emerges from an international criminal trial is thorough, relatively comprehensive, and, most importantly, accurate enough to provide an evidentiary bulwark against historical revisionists who invariably deny the atrocities committed by their group. One thread not fully addressed in the lit­er­a­ture is the predilection of international judges, exhibited in many t­ rials in many dif­fer­ent international tribunals, for historical accounts of popu­lar mobilization. What is undeniable is that propaganda has emerged as one of the main explanations for intercommunal mass atrocities at a variety of international tribunals. The four international courts reviewed ­here—­the IMT at Nuremberg, the ICTY, the ICC, and the ICTR—­were established in quite distinct historical moments over nearly six de­cades, and yet their judgments attribute a con­spic­u­ous role to propaganda when accounting for the origins and ­causes of international crimes.6 It is now commonly accepted that the IMT at Nuremberg furnished an impoverished historical account of the Holocaust in that it did not centrally address the Nazi regime’s program to exterminate Eu­ro­pean Jews.7 So as not to violate the princi­ple of non-­retroactivity (nullum crimen sine lege), Article 6(c) of the 1945 London Charter defined crimes against humanity “in execution of or in connection with any crime within the jurisdiction of the Tribunal,” namely war crimes and crimes against peace. The wording of Article 6(c) obliged prosecutors to approach the Holocaust not as a separate crime, but as an extension of Germany’s war of aggression against its neighbors. The only judgments in which the IMT judges addressed the existence of widespread German popu­lar support for the Holocaust and the longstanding anti-­Semitism in German society ­were the verdicts on two propagandists, the head of the Radio Division of the Propaganda Ministry, Hans Fritzsche,8 and the Nazi publisher Julius Streicher.9 Founder and publisher of weekly tabloid Der Stürmer (The Attacker), a viciously anti-­Semitic German weekly, Julius Streicher was convicted and hanged in 1946 for “persecution on po­liti­cal and racial grounds in connection with war crimes,” punishable as a crime against humanity ­under the IMT Charter.10 The tribunal found that the defendant had knowledge of Hitler’s policy of extermination of the Jews, yet he continued to incite Germans “to murder and extermination at the time when Jews in the East ­were being killed ­under the most horrible conditions.”11

100

ric h a rd a shby w ilson

figure 5.1. Defendant Julius Streicher, the former editor of Der Stürmer on the stand at the International Military Tribunal at Nuremberg, April 26–29, 1946. (Photo from the United States Holocaust Memorial Museum, courtesy of the National Archives and Rec­ords Administration, College Park, Mary­land.)

The challenge for the prosecution was to create a theory of the case when no pre­ce­dent existed in international law for holding propagandists criminally accountable. The trial briefs and memoranda held in the archives of Thomas J. Dodd, executive trial counsel at Nuremberg, reveal that the prosecution construed the propaganda defendants as liable for crimes against humanity as accomplices



The Spark for Genocide? 101

and abettors. The prosecutors’ brief on Streicher stated that: “[d]efendant Streicher is an accessory to the persecution of the Jews within Germany and in occupied territories . . . ​[who] actively supported, recommended, and promoted the program of extermination.”12 The prosecution refrained from claiming that Streicher’s propaganda directly caused specific genocidal actions, and instead referred to how propaganda prepared the population psychologically for an orchestrated state-­sponsored program of extermination of the Jews. A May 1946 memo from Harriet Zetterberg Margolies to Thomas Dodd lays out the prosecution’s case theory: “Streicher helped to create, through his propaganda, the psychological basis necessary for carry­ing through a program of persecution which culminated in the murder of six million men, ­women, and c­ hildren.”13 In the fleeting Streicher judgment, the IMT judges broadly accepted the prosecution’s case theory, and in convicting Streicher of incitement, they converged on the psychological state of the German populace: “In his speeches and articles, week a­ fter week, month ­after month, [Streicher] infected the German mind with the virus of anti-­Semitism, and incited the German ­people to active persecution.”14 The IMT decision postulated a connection, not between the defendant’s speech acts and concrete genocidal acts, but between the defendant’s speech acts and the attitudes of the German public, through the meta­phor of viral contagion. The account in Streicher is the only one provided in all of the Nuremberg judgments for why so many ordinary Germans actively endorsed and participated in a program to exterminate the Jews. The decision regards propaganda as a background condition, jointly sufficient with other contextual conditions, rather than as directly causal ­factor that is necessary and sufficient in itself. Prosecutors and judges constructed Streicher’s criminal responsibility upon accomplice liability, an inherently vicarious form of liability that relies upon other more direct forms of material commission. Judges also emphasized intention and psychological states, rather than material causation and a demonstrable nexus between propagandistic speech and par­tic­ul­ ar material crimes. From the very first case at the ICTY, propaganda became one of the overarching explanations for why neighbors became killers during the 1991–1995 conflict in the former Yugo­slavia. Perhaps this emphasis on popu­lar participation was inevitable, given that the first defendant brought to trial at the ICTY was the low-­ranking Duško Tadić, a Bosnian Serb civilian who in­de­pen­dently volunteered to torture, murder, and forcibly transfer civilian non-­Serbs in the Prijedor area of Bosnia-­Hercegovina. The Tadić Trial Chamber judgment opens with one of the longest historical treatises thus far in an international criminal tribunal judgment. Tadić dedicates extensive discussion to the Serbian propaganda campaign in Bosnia, observing that by spring 1992, all the media in Bosnia was Serb controlled.15 Radio and tele­vi­sion pounded out the same unrelenting message that Serbs w ­ ere about to be overwhelmed by “Ustasha” (Fascist) Croats and fundamentalist Muslims, and ordinary Serbs had l­ ittle choice but to join the

102

ric h a rd a shby w ilson

army in an all-­out war to save themselves from genocide. The judgment acknowledges centuries of conflict in the region but places more emphasis on how a relentless and manipulative propaganda conjured up and aggravated historical grievances: “Many witnesses speak of good intercommunal relations, of friendships across ethnic and coincident religious divides, of intermarriages and of generally harmonious relations. It is only subsequent events that may suggest that beneath that apparent harmony always lay buried ­bitter discord, which skillful propaganda readily brought to the surface, with terrible results.”16 ­After Tadić, ICTY prosecutors charged vari­ous defendants such as Bosnian Croat politician Dario Kordić and Bosnian Serb politician Radoslav Brđanin with crimes against humanity in part based on their speeches. Prosecutors cited public speeches and propagandistic broadcasts as evidence of the defendants’ participation in a joint criminal enterprise.17 For instance, the Trial Chamber in Brđanin identified propaganda as a primary driver of the internecine Balkans conflict: Prior to the outbreak of the armed conflict, the SDS [Serbian Demo­cratic Party] started waging a propaganda war which had a disastrous impact on the ­people of all ethnicities, creating mutual fear and hatred and particularly inciting the Bosnian Serb population against the other ethnicities. Within a short period of time, citizens who had previously lived together peacefully became enemies and many of them, in the pres­ent case mainly Bosnian Serbs, became killers, influenced by a media, which by that time, was already ­under the control of the Bosnian Serb leadership.18

Claims about the role of propaganda in conflict are not the exclusive dominion of prosecutors, and ­here we should recall the earlier theoretical point that historical narratives at international tribunals emerge from the strategic l­egal imperatives of the adversarial parties in a trial. ICTY defense counsel have also referred to the misleading effects of propaganda in order to mitigate the responsibility of their clients such as Serbian politician Milan Babić. The Babić Trial Chamber accepted defense counsel’s arguments regarding the impact of Serb propaganda: “Babić stated that during the events, and in par­tic­ul­ ar at the beginning of his po­liti­cal ­career, he was strongly influenced and misled by Serbian propaganda, which repeatedly referred to an imminent threat of genocide by the Croatian regime against the Serbs in Croatia, thus creating an atmosphere of hatred and fear of the Croats. Ultimately this kind of propaganda led to the unleashing of vio­lence against the Croat population and other non-­Serbs.”19 ­After Streicher, a number of ICTR Trial Chamber judgments (Akayesu, Kayishema and Ruzindana, Nahimana, Ruggiu, and Bikindi) all asserted that the defendants’ inciting speech played a causal role in prompting o­ thers to commit genocide and that therefore they deserved to be convicted for more direct forms



The Spark for Genocide? 103

of commission than mere accomplice liability. The ­legal judgment that granted greatest prominence to propaganda in explaining the Rwandan genocide was Prosecutor v. Ferdinand Nahimana, Jean-­Bosco Barayagwisa, Hassan Ngeze, also known as the “Media Trial.”20 The three defendants in the Media Trial all owned major Rwandan media outlets before the genocide. Ferdinand Nahimana and Jean-­Bosco Barayagwisa ­were both government ministers and found­ers of the in­de­pen­dent radio station Radio Télévision Libre des Milles Collines (RTLM). Hassan Ngeze was the owner and editor of Kangura (Wake O ­ thers Up), a newspaper widely distributed in Rwanda before 1994. The three defendants in the Media Trial w ­ ere convicted of incitement to commit genocide, in a verdict that asserted a clear causal connection between speeches and radio broadcasts and subsequent public vio­lence.21 The ruling asserts no less than sixteen times that speech acts directly caused genocidal killings, using language such as: Many of the individuals specifically named in RTLM broadcasts ­after 6 April 1994 ­were subsequently killed. . . . ​W hile the extent of causation by RTLM broadcasts in t­ hese killings may have varied somewhat, depending on the circumstances of ­these killings, the Chamber finds that a causal connection has been established by the evidence.22 Without a firearm, machete, or any physical weapon, he [Nahimana] caused the deaths of thousands of innocent civilians.23

Causal claims ­were not confined to material fact-­finding but also permeated the section reviewing the law of genocide. The Nahimana Trial Chamber attributed a causal role generally to “the media”24 and noted that while the downing of the Rwandan president’s plane may have triggered the genocide, “RTLM, Kangura, and CDR ­were the bullets in the gun.”25 At the International Criminal Court (ICC), the pattern has continued of charging propagandists with playing an essential role in fomenting popu­lar participation in a program of ethno-­nationalist vio­lence. Charles Blé Goudé, youth leader and supporter of former President of the Ivory Coast Laurent Gbagbo, is at the time of this writing standing trial for co-­perpetrating, ordering, inducing or soliciting, aiding and abetting, and contributing in any other way to the commission of crimes against humanity, in part as the result of his “violent rhe­toric and hate speech.”26 In another ICC trial, ­Kenyan radio broadcaster Joshua Arap Sang, a supporter of co-­accused Vice President William Ruto, was originally indicted for having “contributed, within the meaning of article 25(3)(d)(i) of the Statute, to the commission of the crimes against humanity of murder, deportation or forcible transfer of population and persecution” on the basis of his broadcasts during the 2007

104

ric h a rd a shby w ilson

­ enyan presidential elections.27 Sang’s guilt or innocence hinged on ­whether K the prosecution could demonstrate that ­there was a causal nexus between his broadcasts on his daily radio show and the commission of the crimes. Since no accused has yet to be convicted u­ nder Article 25(3)(d)(i) of the ICC statute, ­there is some uncertainty as to what level of contribution to the commission of crimes is required (e.g. essential contribution, substantial contribution, and so on). In Lubanga, the judges indicated that the actus reus requirement for co-­perpetrating according to Article 25(3)(a) necessitates a “coordinated essential contribution made by each co-­perpetrator resulting in the realisation of the objective ele­ments of the crime.”28 Essential contribution implies “but-­for causation” where an act is a necessary and sufficient cause of another act. This is a very high bar of causation indeed, and Judge Adrian Fulford strenuously argued in Lubanga that the Trial Chamber’s test of essential contribution is “unrealistic and artificial”29 and “imposes an unnecessary and unfair burden on the prosecution.”30 Advocating plain language and a lesser test, Judge Fulford recommended simply requiring “a contribution to the crime, which may be direct or indirect, provided ­either way ­there is a causal link between the individual’s contribution and the crime.”31 Most inciting speech acts, and even quite effective ones, ­will fall short of the threshold of constituting an essential (but-­for) contribution to the commission of crimes. A ­ fter the Ruto and Sang trial had begun, the prosecution petitioned the court to give notice that the ­legal characterization of the charges may change. The prosecution identified the possibility of convictions based on Article 25(3) (b), which focuses on ordering, soliciting, or inducing the commission of the offence, and Article 25(3)(c), which refers to aiding and abetting. A lower threshold of causation than essential contribution is likely to apply for t­hese lesser forms of criminal responsibility.32 In each of the instances reviewed thus far—­the persecution of Jews in Germany, the murder of Serbs, Croats and Muslims in the former Yugo­slavia, the genocide of Tutsis in Rwanda, and interethnic election vio­lence in K ­ enya and the Ivory Coast—­international tribunals have turned to propaganda to explain how po­liti­cal leaders mobilized followers to persecute their erstwhile friends and neighbors. Propaganda is not the only historical cause identified by international courts to explain the outbreak of internal or international armed conflict. Tribunals have also contemplated a range of other ­factors, including economic collapse, constitutional crises, and the demise of dominant ideologies such as Soviet-­style socialism.33 Yet when it comes to comprehending widespread popu­ lar participation in vio­lence and the cultural resonances of a government’s program of religious or ethnic persecution, po­liti­cal propaganda has come to assume an outsized role in international ­legal histories.



The Spark for Genocide? 105

mind the gap: evidentiary weaknesses in propaganda ­trials Given the forceful declarations made in international criminal tribunal judgments regarding the deleterious effects of propaganda, one might expect the supporting evidence for ­these claims to be extensive and reliable. This is seldom the case, however, and it could be argued that the chasm between the claims regarding propaganda’s consequences and the evidence presented to support such claims is wider than in any other area of international criminal law. A noteworthy feature of international criminal t­ rials is the seeming inability of prosecution teams to locate and incorporate witnesses, ­either bystanders or material perpetrators, who w ­ ill testify about the instigating effect of propaganda on themselves or other insiders. This may be a conscious strategy, or it may result from f­ actors beyond prosecutors’ immediate control. In the former category, we could note that Nuremberg prosecutors consciously avoided introducing eyewitness or perpetrator testimony into their propaganda cases. In Streicher, they deci­ded not to pres­ent evidence in their possession from an earlier German trial of a Stormtrooper who murdered a Jew and who testified that the offence was incited by a (fabricated) story about ritual murder published in Der Stürmer.34 The prosecution team anticipated Streicher’s defense rejoinder that no direct evidence had been introduced to prove that his propaganda did, in fact, influence ­those directly participating in persecution or extermination by reminding the tribunal that causation was “tangential to the charge brought against Streicher.”35 Instead, prosecutors construed inciting speech as an inchoate crime (in which the conduct itself is the crime) and concentrated the courtroom’s attention on the explicit message contained in Julius Streicher’s words themselves. ­Here, the primary prosecution exhibit was the May 1939 issue of Der Stürmer, in which Streicher declared, “The Jews in Rus­sia must be killed. They must be exterminated root and branch.”36 The explic­itly genocidal intentionality contained in ­these words, not the consequences of the words, served as the basis for Streicher’s conviction for persecution. In contrast, the consequences of inciting propaganda w ­ ere the centerpiece of the ICTR’s Trial Chamber judgment in Nahimana. In the factual findings section on RTLM broadcasts, Nahimana identified three types of evidence connecting the radio station’s transmissions to the genocide. First, some of the Tutsi individuals specifically mentioned by name in broadcasts before and ­after April 6, 1994, ­were subsequently killed.37 The judgment cited the hearsay testimony of one eyewitness who reported being accosted in the street by an attacker who referred to the anti-­Tutsi content of an RTLM broadcast.38 Fi­nally, the nexus between broadcasts and killings was found in the threat perceived by individuals who had been named in RTLM broadcasts.39

106

ric h a rd a shby w ilson

Despite the conventional wisdom at the ICTR and in journalistic and h­ uman rights circles that the pro-­government radio and press w ­ ere full of daily calls for genocide in 1994, compelling evidence for genocidal mens rea was not presented by the prosecution in Nahimana.40 Former ICTR attorney Alexander Zahar remains “puzzled by the court’s inability to come up with a single example—­ broadcast on RTLM or printed in Kangura—of a blatant call on Hutu to hunt down and destroy the Tutsi ethnic group.”41 He notes that while the thirty-­seven fragments of RTLM cited in Nahimana are “brutal,” full of ethnic animus and deserving of a penal response, in his view “the fragments do not read like direct and public incitement to commit genocide.”42 For Zahar, it is remarkable that “the prosecutor was not able to arrange for a single witness to testify that he or she was incited by appeals or hints . . . ​to commit genocide.”43 It is worth noting ­here that a population, especially during war­time, may not be consciously aware of the insidious and unconscious effects of propaganda on their beliefs and actions. At the same time, Zahar is justified in his expectation that ICTR prosecutors could have located at least a handful of perpetrators who participated in the manual slaughter of over 800,000 Tutsis and moderate Hutus, ­were aware of the effects of RTLM radio broadcasts on their be­hav­ior, and could testify to the Trial Chamber about the effects of the propaganda.44 The Nahimana Appeals Chamber pinpointed t­hese lacunae in the evidence and determined that ­there was “no evidence that Appellant Nahimana played an active part in the broadcasts a­ fter 6 April 1994 which instigated genocide. Furthermore, the rec­ord contained no evidence that Nahimana had, before 6 April 1994, given instructions to RTLM journalists to instigate the killing of Tutsi.”45 Acquitting Nahimana on the charge of instigating genocide, the Appeals Chamber rejected the Trial Chamber’s evidence of a link between RTLM broadcasts and acts of genocide, declaring such evidence, “at the very least, tenuous.”46 Nevertheless, Nahimana’s conviction for direct and public incitement to commit genocide, an inchoate crime that does not require proof of causation, was upheld by the Appeals Chamber. Tenuous evidence for the causal effects of inciting speech has also plagued other propaganda t­ rials, in some cases resulting in the acquittal of the accused. In the trial of Serb nationalist politician Vojislav Šešelj at the ICTY, prosecutors sought testimony from insiders, former high-­ranking Šešeljevci (or Šešelj’s Men), to corroborate the theory that the defendant exerted unquestioned authority over his followers and that his public speeches and broadcasts instigated crimes against humanity. At the beginning of the trial, six of Šešelj’s Men gave statements to this effect, but over the course of the trial this cornerstone of the prosecution’s case disintegrated. All of the insiders, to a man, withdrew their testimony and refused to testify in The Hague, or ­else they appeared to testify and recanted on the stand. In its closing brief,47 the prosecution argued that the testimonial evidence of recanting witnesses should not be rejected by the Trial Chamber, and it fur-



The Spark for Genocide? 107

nished corroborating evidence to bolster many aspects of their statements. And while the insider statements ­were formally accepted into evidence by the Trial Chamber, their probative value was grievously harmed by the spate of recantations, and this is one explanation why the prosecution case ultimately failed. Other reasons include a seemingly conscious desire on the part of the majority, Judge Lattanzi dissenting, to construe the evidence in a way favorable to the accused and to ignore compelling evidence of a widespread and systematic attack on a civilian population. Prosecution witnesses who ­later recant are a feature of all types of international criminal tribunal cases. While our sample is too small to draw iron-­clad conclusions, this phenomenon does seem more endemic in cases where propaganda and speech form the centerpiece of the prosecution case, in part b­ ecause the accused are often high-­level po­liti­cal officials or popu­lar “opinion leaders” with ­either continued government backing or a wide base of over-­zealous supporters. In 2015, the Trial Chamber hearing the Ruto and Sang case at the International Criminal Court ordered nine prosecution witnesses to testify a­ fter they stopped communicating with the prosecution or recanted their statements.48 Compelling recalcitrant witnesses to testify in Ruto and Sang against their w ­ ill via a video link from Nairobi proved counterproductive. As one witness ­after another defected to the defense or continually claimed to not remember any pertinent facts he or she had given statements on, prosecutors scrambled to petition the court to declare a succession of witnesses as hostile witnesses.49 Prosecutors alleged that the witnesses recanted their earlier testimony only ­after the prosecution’s witness list was disclosed to the defense and they ­were approached and bribed by persons aligned with the defendants.50 Amidst swirling allegations of bribery, intimidation, and murder, ­trials of po­liti­cal leaders at international criminal tribunals share many of the attributes of major or­ga­nized crime cases in domestic settings. In 2016, the Trial Chamber, by majority, vacated the charges against Mr. Ruto and Mr. Sang, having regard, at least in part, to the “disturbing level of interference with witnesses, as well as inappropriate attempts at the po­liti­cal level to meddle with the trial and to affect its outcome.”51 Prosecutors at international tribunals have supplemented fact witness testimony with evidence from expert witnesses, generally academic social scientists, who have expertise in the study of language, media, and propaganda. ­There is ­great variation, possibly greater than in any other area of expertise, in w ­ hether expert testimony on propaganda is admitted as evidence and accepted at an international court. At Nuremberg, no such experts w ­ ere called. Generally speaking, at the ICTR, prosecution expert testimony on propaganda has been favorably received by the judges, while at the ICTY it has often been spurned; it is too early to tell which disposition ­will emerge at the ICC. At the ICTR, the judges’ knowledge of Rwandan history and culture was formed largely by two prosecution expert witnesses who between them appeared

108

ric h a rd a shby w ilson

for the prosecution over a dozen times before the tribunal: Rwandan linguist Dr. Mathias Ruzindana and ­Human Rights Watch adviser Dr. Alison Des Forges.52 ­These two experts’ viewpoints became an accepted part of the official court history of the Rwandan genocide. Ruzindana appeared in the first case of the tribunal, in which the defendant Jean-­Paul Akayesu was convicted of direct and public incitement to commit genocide, an international crime established nearly fifty years earlier in the 1948 U.N. Convention on the Prevention and Punishment of Genocide.53 Des Forges repeatedly emphasized how propaganda prepared the population for genocide: “RTLM had an enormous impact in encouraging the killing of the Tutsis and of o­ thers who might support and protect the Tutsis during this genocide.”54 Ruzindana reiterated this message, and the judges cited Ruzindana in Nahimana regarding the genocidal meaning of the machete on the cover of the newsletter Kangura published by Nahimana’s co-­defendant, Hassan Ngeze.55 Consolidated in successive Trial Chamber decisions, Ruzindana’s analy­sis of the genocidal communications in the Rwandan media became an accepted fact of common knowledge at the tribunal. The circumstances w ­ ere quite dissimilar at the ICTY, where over time judges became more resistant to expert witnesses who did not deal with ballistics, forensics, and other technical ­matters. It was not this way at the outset. In the first ­trials, the prosecution successfully called historians and social scientists to explain the historical origins and contours of the conflict to the judges who ­were uninformed about Balkans history.56 The defense brought their own experts to engage in a “­battle over the first paragraph,” with each side seeking to shape the framing of the crimes in the opening paragraph of the trial judgment. For at least the first ten years of t­ rials at the ICTY, t­ hese framing debates ­were a focal point of ­legal contestation, as judges ­adopted the position of one party or another, or, as occasionally happened, combined them into a composite picture of events. By 2008, ICTY judges w ­ ere adopting a decidedly more skeptical approach to expert testimony on historical ­matters.57 This was not an especially propitious moment to call a social science expert to educate the court on the propagandistic speech acts of the notorious ICTY defendant, former Serbian Vice President Vojislav Šešelj. The prosecution team commissioned an eminent scholar of Eu­ro­pean social movements, Dr. Anthony Oberschall, to analyze the content of 290 speeches. Oberschall submitted a well-­ crafted report, applying widely accepted theories of nationalism and so­cio­log­ic­ al discourse analy­sis to the material he had been given by prosecutors. His status as expert was rejected by the judges on the grounds that he did not possess the requisite expertise to analyze the propagandist speeches of the defendant.58 As a compromise, the report was admitted and the bench allowed him to testify as a fact witness, which makes ­little sense since a fact witness must, by definition, have been pres­ent at the scene of the crime, and Oberschall did not visit the for-



The Spark for Genocide? 109

mer Yugo­slavia during the war. This was but one of the irregular decisions made in the trial by Presiding Judge Antonetti. In so­cio­log­ic­ al analy­sis, government propaganda functions as a macrolevel condition rather than a direct cause essential to the outcome. Therefore, a social scientist’s view of cause and effect may not conform to criminal law’s conventional linear, billiard-­ball model of causation, as we see in the Prosecution’s Closing Brief in Šešelj in passages such as: “Šešelj’s public ‘hate speech’ on 6 May 1992 directly caused a number of Croats to leave Hrtkovci. His speech also triggered and facilitated a massive campaign of intimidation, harassment and vio­lence directed against the local Croat population, causing further displacements of Croats from Hrtkovci and surrounding areas.”59 While not suggesting for a moment that this statement is untrue, in chaotic war­time conditions, ­there are often multiple in­de­pen­dent actors, and any one violation is the product of a complex set of ­causes and conditions. When crimes emerge from a tangled web of causation, it is an arduous undertaking to unravel causation with the precision required to attribute individual criminal responsibility. Social science expert evidence is likely to have limitations in proving the direct causation and the actus reus that directly underpins the liability of an accused who is charged with instigating crimes. Nonetheless, as we saw in Nahimana, social and historical research can have probative value when the accused is charged with inchoate speech crimes (such as incitement to genocide and hate speech as a form of persecution) where the meaning and intentionality of the speaker are at issue, and it has potential utility when a defendant is charged with accomplice liability, which does not require such direct causation. International tribunals have frequently attributed a causal role to propaganda when the accused is charged primarily on the basis of speech acts, as well as in cases where speech is not part of the actus reus but propaganda is a way of speaking about the general context of the crimes. Yet the prominence of propaganda in international criminal trial judgments is often not matched by the evidential basis presented by the prosecution in specific t­rials. Testimony from material perpetrators or insiders is ­either absent or recanted on the stand, eyewitness evidence of propaganda’s influence is often based on hearsay, and experts, while crucial at some tribunals, have been absent or excluded at ­others. Th ­ ere is a lack of clarity and consistency on what expertise on propaganda is admissible, and the threshold of eyewitness and expert evidence would prove causation in propaganda ­trials. New and innovative prosecutorial strategies have emerged to illuminate the deleterious effects of the propagandist’s words, but t­hese approaches have not yet been accepted in international t­ rials. What is in­ter­est­ ing is that the wide variation at the level of trial rules of procedure and evidence (e.g., from the IMT to the ICTR) has not greatly altered the degree to which jurists at successive international tribunals reference propaganda as an ele­ment

110

ric h a rd a shby w ilson

of macrocausation in their rulings. Propaganda keeps resurfacing, in dif­fer­ent epochs and on dif­fer­ent continents, as the preferred ­legal explanation for why ordinary ­people become killers or complicit bystanders who are willing to tolerate crimes committed in their name.

international law’s histories: identifying ­ causes and expressing morals I asserted at the outset that the study of international tribunal histories of armed conflict and genocide must begin with the agency of the main social actors in ­legal pro­cesses. In par­tic­u­lar, we might pay attention to the strategies that prosecutors and defense attorneys adopt in their case theory and courtroom arguments in pursuit of specifically ­legal goals. Overstressing the divergence between ­legal and historical ways of knowing hinders our understanding of why historical narrative has been an enduring feature of international criminal t­ rials and how defense and prosecution teams integrate historical accounts into their cases. Opposing parties to international criminal t­rials instrumentalize history, the judges do the same in their judgments, and each group reaches for historical narrative to advance its own special aims. However, this approach needs some qualification ­because prosecutors, judges, and defense attorneys do not enjoy f­ree rein, and ­there are ingrained ways of knowing in a criminal trial that must conform to established rules of evidence and tests for causation. ­Legal epistemology implies a cognitive template that influences the kinds of histories that international tribunals write. The trick is to keep an eye on what l­ egal actors are seeking to achieve with historical evidence in any given trial, while still keeping firmly in view the structural rules of the game in which all actors operate (the doctrine, established procedures, epistemology, and so on). ­There are at least three explanations for the prominence of propaganda in international tribunals’ historical narratives on mass atrocities, in ascending order of abstraction. The first relates to the prevailing cognitive model of causation in Anglo-­American and civil law. Criminal law conventionally applies strict tests to determine relations of cause and effect beyond a reasonable doubt. ­Legal doctrine conventionally divides causation into two types: material causation and ­legal causation. The material cause or “cause-­in-­fact,” is the a­ ctual cause of an event. Establishing cause-­in-­fact conventionally requires convincing evidence of direct causation that passes the “but-­for” or sine qua non test, a counterfactual method of proof to identify the cause without which the event could not have occurred.60 ­Legal causation, or “proximate” cause, refers to the cause that has been selected as a ­matter of policy or statute as the most legally relevant. As a rule, criminal courts are a­ dept at identifying the proximate cause that is temporally and physically adjacent to the commission of the crime. They can also distinguish without g­ reat fanfare remote historical ­causes that, while ger-



The Spark for Genocide? 111

mane to explaining why the crime was committed, are less germane to attributing blame. Historical research has a very dif­fer­ent conception of cause and effect than criminal law, and what is a remote cause for attorneys might constitute a compelling cause for historians. At the ICC, ­Kenyan experiences of Eu­ro­pean colonialism, however relevant to the ethnic politics of ­Kenya ­today, occupy a secondary tier of remote ­causes. At the ICTR, the 1959–1962 Revolution against the Tutsi monarchy in Rwanda could arguably have been one material cause of the 1994 genocide of Tutsis, but it is too historically distant to qualify as a proximate cause to which criminal responsibility could attach. What criminal courts generally find harder to discern are c­ auses in the temporal ­middle ground, between distant and adjacent ­causes. It is also pos­si­ble that ­these intermediate ­causes might constitute a­ ctual material c­ auses that contributed to the harm or injury. Propaganda lies in an indefinite and ambiguous terrain of causation, as it often arises in the crescendo to an armed conflict and lays the basis for, in the words of the Streicher decision, “the requisite psychological and po­liti­cal conditions for aggressive war.” Note the use of the term “conditions” ­here, rather than any reference to direct ­causes. ­Legal phi­los­op­ hers generally perceive conditions to be distinct from ­causes, insofar as a cause must satisfy the sine qua non test, whereas conditions can be jointly sufficient with other concurrent circumstances to effectuate a result.61 At international criminal tribunals, the category of propaganda is a black box of indeterminate causation, a container into which judges consign all that is cultural, contextual, and historically conditioning. References to propaganda allow international judges to pronounce upon prior conditions as if they ­were proto-­ causes and in so ­doing to discover the historical origins of collective vio­lence and answer the per­sis­tent question, “Why did so many normal citizens engage in mass atrocities?” In using causal language, albeit meta­phorical causal language, judges can widen the scope of liability beyond what is conventionally allowed by the proximate cause test, which must be properly evidenced, but without reaching so far back as to summon up remote ­causes. Judges pronounce on propaganda to address the macrocausation issue, but as we have seen, ­there is usually not enough evidence in the trial rec­ord to warrant such a determination. To be clear, the argument ­here is not that government propaganda and inciting speech are irrelevant in a conflict, only that international law has not begun to define and conceptualize in a rigorous and well-­evidenced manner what their effects are. Yet international tribunal judgements constantly pronounce upon propaganda’s corollaries with g­ reat authority and certitude. When, as in rare instances such as Šešelj, they reject any role of inciting speech in the commission of crimes, it is left unclear what further evidence, beyond that already provided by the prosecution, the Trial Chamber would require to determine causation. This area of international criminal law is beset by uncertainty over the threshold of evidence conventionally required to fulfil the ele­ments of the crimes.

112

ric h a rd a shby w ilson

Second, references to propaganda allow international judges to attribute general responsibility for a conflict in a way that places blame squarely on the po­liti­ cal authorities at the time. Why was ­there such popu­lar support for a genocidal regime in Germany, Rwanda, and Yugo­slavia? ­Because shameless leaders harboring wicked intentions whipped up the unruly mob with their goading, inciting speech. Of course, this account applies, at a superficial level, to all the scenarios we have considered. Yet as a general and overarching explanation, this account mistakenly relies on a model of speech acts and ­human be­hav­ior that deprives all actors (except the accused) of any subjectivity and agency and treats the population as mindless automatons. Crucially, meta­phorical language closes the evidentiary gap and superficially resolves prosecutorial tribulations in documenting the chain of m ­ ental causation in international propaganda ­trials. Judicial meta­phors convert the desires, beliefs, and intentions of speakers and their audience into the kind of physical objects that make up the material-­object world, and thus also transform subjectivities and interpersonal social relations into material-­object world relations. In this way, judges erroneously solve the intransigent prob­lem of ­mental causation, which requires the kind of evidence that international criminal tribunals have thus far sorely lacked—­namely, reliable insider testimony on the stand. Constructing speech acts as bullets, poison, or viruses, and thereby reconfiguring ­mental causation as physical causation, allows judges and prosecutors to assert a mechanical relationship between the ­mental states of actors and between ­mental states and observed be­hav­ior. When courts perceive speech and intentionality as analogous to guns, poison, fire, and viruses that conform to physical laws of cause and effect, then they can align speech and intentionality with the criminal law’s billiard-­ball model of causation. Unlike h­ umans, physical objects like guns, fire, and poison do not contain inner subjectivity or agency. They are merely objects knocked this way and that by forces acting externally upon them. In contrast, social scientists conducting perpetrator interviews in Kigali, Rwanda, have found that the population did not always hear the message in the way that it was intended, nor did Rwandans blithely act in conformity with the intentions of po­liti­cal leaders.62 International law’s historical account of propaganda is exculpatory t­owards the po­liti­cal status quo, since it sidesteps intractable prob­lems such as the perennial competition over scarce economic resources, the racial legacy of colonialism, and the ethnic and national character of states in the postcolonial order. Blaming propaganda is more straightforward than addressing past social, economic, and po­liti­cal tensions that are still salient in the pres­ent, since d­ oing so makes no demands on the postconflict state and society. Worse still, international criminal tribunal narratives can coincide with the interests of postconflict regimes such as Kagame’s Rwanda, which lay the blame for popu­lar mobilization on genocidal propaganda and then pass laws such as the 2008 law prohibiting “genocide ideology” to stifle dissent and jail government critics.



The Spark for Genocide? 113

Third and fi­nally, judgments on the role of propaganda are, as we have seen, full of meta­phors, analogies, and other forms of figurative speech, and so any analy­sis must acknowledge the “expressivist” function of law.63 Expressivism in law, argue normative ­legal theorists, serves to legitimize new laws that shift social norms in a certain direction. This seems to fit our case h­ ere. The passages dealing with propaganda in international criminal judgments and prosecutor’s closing briefs are where prosecutors and judges are at their most colorful and loquacious, in a way that they are not in a routine decision on, say, war crimes. This contrast results from the precariousness of the new international law of propaganda. The suppression of inciting speech and propaganda for war is a novel and emergent area of international criminal law, in which t­here is only a sliver of ­legal pre­ce­dent, and the evidence presented during ­trials is often fragile and incomplete. Moreover, freedom of speech is a topic where ­there is perhaps the widest variation in global l­egal practices, and the bench in any given trial at an international criminal tribunal may include judges from countries that are near-­absolutist in their freedom of speech protections (the United States, Norway) and also from countries where the ­legal regulation of hate speech is a m ­ atter of settled law (Germany, New Zealand). Moral denunciations are at the heart of the enterprise of criminal law. Modern U.S. criminal law textbooks usually begin with a definition of crime of the type found in Dressler and Garvey’s Criminal Law: [A crime] is not simply anything which a legislature chooses to call a “crime.” It is not simply antisocial conduct which public officers are given responsibility to suppress. It is not simply any conduct to which a legislature chooses to attach a “criminal” penalty. It is conduct, which, if duly shown to have taken place, ­will incur a formal and solemn pronouncement of the moral condemnation of the community.64

International criminal law is no dif­fer­ent in this re­spect, and Drumbl applauds the expressivism of international criminal ­trials when they morally condemn and stigmatize: “­Trials can educate the public through the spectacle of theatre—­ there is, ­after all, pedagogical value to per­for­mance and communicative value to dramaturgy. This per­for­mance is made all the more weighty by the real­ity that, coincident with the closing act, comes the infliction of shame, sanction and stigma upon the antagonists.”65 Moral condemnations are more abundant in criminal law than any other area of law, or official state discourse for that m ­ atter. The sovereignty of the modern state has historically been conceived and established, perhaps in its most elemental function, through its ability to denounce and punish in the name of the po­liti­cal community. What is new and dif­fer­ent in the setting of international criminal law is the paucity of in­de­pen­dent international ­legal pre­ce­dent on some of the crimes being adjudicated, and the lack of

114

ric h a rd a shby w ilson

consensus on speech crimes on the part of jurists coming from distinctive national jurisdictions. Substantive law and community values are combined into a heady mixture in international criminal t­ rials, with the desired result being the legitimization of the law and the position of t­ hose who proscribe be­hav­ior in the name of the “international community,” itself a precarious imaginary.66 The sovereignty of the international justice order to condemn and penalize international crimes such as hate speech as persecution, and direct and public incitement to commit genocide is still nascent. Historical accounts of propaganda that are replete with judges’ rhetorical flourishes serve to consolidate and enlarge the scope of that conduct which falls ­under their contested jurisdiction. Judgments on propaganda turn into opportunities for jurists to engage in moral denunciation and express the international community’s opprobrium on irredentist conflicts and the slaughter of innocents, and in the pro­cess, make new international law.

notes I thank Nanci Adler, Thijs Bouwknegt, Predrag Dojčinović, and Vladimir Petrović for their helpful comments on this essay. Any errors are my own. This chapter is an abridged adaptation of Richard Ashby Wilson, “Propaganda and History in International Criminal ­Trials,” Journal of International Criminal Justice 14, no. 3 (2016): 519–541. 1. ​On propaganda for war, see Michael Kearney, The Prohibition of Propaganda for War in International Law (Oxford: Oxford University Press, 2007). The discussion of propaganda in this essay includes not only propaganda for war, but also the international crimes of direct and public incitement to commit genocide, hate speech as a form of persecution, disseminating propaganda as part of a joint criminal enterprise, instigating crimes against humanity on the basis of speech acts, and co-­perpetrating crimes against humanity on the basis of speech acts. 2. ​Richard Ashby Wilson, Writing History in International Criminal ­Trials (Cambridge: Cambridge University Press, 2011), 1–12; Vladimir Petrović, The Emergence of Historical Forensic Expertise: Clio Takes the Stand (New York: Routledge, 2016). 3. ​Hannah Arendt, Eichmann in Jerusalem: A Report on the Banality of Evil (New York: Viking Press, 1965), 19. 4. ​Henri Rousso, “Letter to the President of the Bordeaux Assizes Court,” October 6, 1997, in The Papon Affair: Memory and Justice on Trial, ed. Richard J. Golsan (New York: Routledge, 2000), 193–194. 5. ​For an empirical elucidation of t­ hese claims, see Wilson, Writing History in International Criminal ­Trials. 6. ​The IMT was created in 1945 by the United States, United Kingdom, Soviet Union, and France. The ICTY (1993) and ICTR (1994) ­were established by the U.N. Security Council as ad hoc tribunals. The ICC (2002) was established ­after Senegal became the sixtieth member state to sign the 1998 Rome Statute. 7. ​See Michael Marrus, The Holocaust in History (Dartmouth, N.H.: University Press of New ­England, 1988); Michael Marrus,The Nuremberg War Crimes Trial 1945–46 (Boston: Bedford Press, 1997); Donald Bloxham, Genocide on Trial: War Crimes ­Trials and the Formation of Holocaust History and Memory (Oxford: Oxford University Press, 2001).



The Spark for Genocide? 115

8. ​Fritzsche was acquitted on all counts by the tribunal. Judicial Decisions, “International Mili-

tary Tribunal (Nuremberg), Judgment and Sentences,” American Journal of International Law 41, no. 1 (1947): 172, 327–329. 9. ​Ibid., 294–296. 10. ​Ibid., 294–296, 331. 11. ​Ibid., 295–296, 333. 12. ​ The Individual Responsibility of the Defendant Julius Streicher, April 30, 1946, Archives & Special Collections at the Thomas  J. Dodd Research Center, http://­archives​.l­ib​.­uconn​.­edu​ /­islandora​/­object​/2­ 0002%3A1942#page​/­1​/­mode​/­2up. 13. ​ Office of U.S. Chief of Counsel Memorandum for Mr. Dodd, May 29, 1946, Archives & Special Collections at the Thomas  J. Dodd Research Center, http://­archives​.l­ib​.­uconn​.­edu​ /­islandora​/­object​/2­ 0002%3A1953#page​/­1​/­mode​/­2up. 14. ​“International Military Tribunal,” 294. On Streicher, see M. Eastwood, “Hitler’s Notorious Jew-­Baiter: The Prosecution of Julius Streicher,” in Propaganda, War Crimes and International Law: from Speakers’ Corner to War Crimes, ed. Predrag Dojčinović (Abingdon: Routledge, 2012), chapter  6; Telford Taylor, The Anatomy of the Nuremberg ­Trials: A Personal Memoir (Boston: ­Little, Brown, 1992). 15. ​ Tadić (IT-94-1-­T), Trial Chamber (TC), May 7, 1997, §88. 16. ​Ibid., §64. 17. ​ Brđanin (IT-99-36-­T), Trial Chamber, September 1, 2004, §§80, 323–332. See also Popović et  al. (IT-05-88-­T), Trial Chamber, June  10, 2010, §§1812–1821, describing the role of Milan Gvero in the widespread promotion and dissemination of propaganda. 18. ​ Brđanin TC, §80. 19. ​ Babić (IT-03-72-­S), Trial Chamber, June 29, 2004, §24(g). 20. ​ Nahimana et al. (ICTR-99-52-­T), Trial Chamber, December 3, 2003. For ­legal commentary on Nahimana / the Media Trial, see Susan Benesch, “Inciting Genocide, Inciting ­Free Speech,” World Policy Journal 21, no.  2 (2004): 62; Gregory  S. Gordon, “A War of Media, Words, Newspapers, and Radio Stations: The ICTR Media Trial Verdict and a New Chapter in the International Law of Hate Speech,” ­Virginia Journal of International Law 45, no. 1 (2004– 2005): 1; Catharine A. Mackinnon, “Prosecutor v. Nahimana, Barayagwiza, and Ngeze, Case No. ICTR 99-52-­A ,” American Journal of International Law 103, no. 1 (2009): 98; Alexander Zahar, “The ICTR’s ‘Media’ Judgment and the Reinvention of Direct and Public Incitement to Commit Genocide,” Criminal Law Forum 16, no. 1 (2005): 33. 21. ​In both the Anglo-­American and civil law traditions, incitement is an inchoate crime, and as such, the speech act itself is the crime. Inchoate crimes are crimes primarily of mens rea. 22. ​ Nahimana TC §482. 23. ​Ibid., §1099. 24. ​Ibid., §952. 25. ​Ibid., §953. 26. ​Decision on the Confirmation of Charges against Charles Blé Goudé, Blé Goudé (ICC02/11-02/11-186), Pre-­Trial Chamber, December  11, 2014, §94 (see also §§49–50). Mr.  Blé Goudé is charged u­ nder Arts 25(3)(a), 25(3)(b), 25(3)(c), and 25(3)(d) of the Statute, see §§158, 166, 171, 181. Hateful and inciting rhe­toric is but one of a number of ele­ments in the prosecution case against Mr. Blé Goudé. 27. ​Decision on the Confirmation of Charges Pursuant to Article 61(7)(a) and (b) of the Rome Statute Sang, Ruto, Kosgey and Sang (ICC-01/09-01/11), Pre-­Trial Chamber, January 29, 2012, §§366–367(a)–(d). His radio broadcasts ­were the main ele­ment of the prosecution case against Sang although they w ­ ere not the only evidence brought against him; see Prosecution’s

116

ric h a rd a shby w ilson

consolidated response to the “Corrigendum of Ruto Defence Request for Judgment of Acquittal” and “Sang Defence ‘No Case to Answer’ Motion,” Ruto and Sang (ICC-01/09-01/11-2000-­ Red2), Trial Chamber, November 20, 2015, §§307–326 (radio broadcasts), 327–359 (other evidence). 28. ​See Decision on Confirmation of Charges, Lubanga (ICC-01/04-01/06), Pre-­Trial Chamber, January 29, 2007, §346, which sets out the ele­ments of the crime of co-­perpetration. 29. ​ Lubanga (ICC-01/04-01/06-2842), Trial Chamber, March 14, 2012, Separate Opinion of Judge Adrian Fulford, §17. 30. ​Ibid., §3. 31. ​Ibid., §16.c. 32. ​Prosecution’s Request for notice ­under regulation 55(2) of possibility of variation with re­spect to individual criminal responsibility of Mr. Joshua Arap Sang, Ruto and Sang (ICC01/09-01/11-1951), Trial Chamber, September 8, 2015. 33. ​See Tadić TC, §§64–79. 34. ​ Defense of Streicher, Archives & Special Collections at the Thomas J. Dodd Research Center, http://­archives​.­lib​.­uconn​.­edu​/­islandora​/­object​/­20002%3A1953#page​/­1​/­mode​/­2up; Office of U.S. Chief of Counsel Memorandum for Mr. Dodd. 35. ​ Individual Responsibility of the Defendant Julius Streicher, 15. 36. ​“International Military Tribunal,” 295. 37. ​ Nahimana TC, §478. 38. ​Ibid., §479. 39. ​Ibid., §480. 40. ​On the deep-­seated prob­lems with eyewitness testimony at the ICTR, see Nancy Combs, Fact-­Finding without Facts: The Uncertain Evidentiary Foundations of International Criminal Convictions (Cambridge: Cambridge University Press, 2010). 41. ​Zahar, “The ICTR’s ‘Media’ Judgment,” 37–38. 42. ​Ibid., 38. 43. ​Ibid., 41. 44. ​In the 2016 Ruto and Sang decision vacating the charges, Judge Fremr also observed that “[c]are, in par­tic­u­lar, must be taken to ensure that proof of a criminal broadcast or publication does not depend mostly or entirely on the oral evidence of witnesses whose own biases and sense of offence about the subject m ­ atter of discussion may have clouded their perception.” In this context, it was noted that the prosecution had produced no direct evidence of the content of Mr. Sang’s broadcasts, but only the testimony of witnesses who claimed to have listened to the broadcasts. The m ­ atter was further complicated by the alleged use of “coded language.” See Ruto and Sang, Reasons of Judge Fremr, §141. 45. ​ Nahimana (ICTR-99-52-­A), Appeals Chamber (AC), November 28, 2007, §597. 46. ​Ibid., §513. 47. ​ Prosecution’s Closing Brief, Šešelj (IT-03-67-­T), Trial Chamber, February  5, 2012, §§651–670. 48. ​For instance, Decision on Prosecution Request for Issuance of a Summons for Witness 727 of February 17, 2015, Ruto and Sang (ICC-01/09-01/11-1817), February 17, 2015, §§37–38. 49. ​Trial Hearing, Ruto and Sang (ICC-01/09-01/11), September 25, 2014. 50. ​Ibid., September 9, 2014, 70–71. 51. ​See Ruto and Sang, Reasons of Judge Fremr, §§147–148 52. ​Des Forges appeared eleven times at the ICTR, and Ruzindana appeared twice as a prosecution expert in Akayesu and Nahimana. 53. ​ Akayesu (ICTR-96-4-­T), Trial Chamber, September 2, 1998. 54. ​ Nahimana AC, §733.



The Spark for Genocide? 117

55. ​Ibid., §161. 56. ​Wilson, Writing History in International Criminal ­Trials, 71–76. 57. ​Decision on the Defense Motion to Exclude the Expert Reports of Patrick Treanor, Perišić

(IT-04-81-­T), Trial Chamber, October 27, 2008. 58. ​Decision Regarding the Admission of Evidence Presented during the Testimony of Anthony Oberschall, Šešelj (IT-03-67-­T), Trial Chamber, January 24, 2008. 59. ​Prosecution’s Closing Brief, Šešelj, §486. This quotation alludes both to direct commission by speech acts and to a direct causal nexus between speech acts and material acts of persecution. 60. ​Herbert Lionel Adolphus Hart and Tony Honoré, Causation in the Law (Oxford: Clarendon Press, 1985), 110–114. 61. ​­Unless they are a “necessary and sufficient condition,” in which case the distinction between ­causes and conditions does not hold; see ibid., 17–20. 62. ​See Charles Mironko, “The Effect of RTLM’s Rhe­toric of Ethnic Hatred in Rural Rwanda,” in The Media and the Rwanda Genocide, ed. Allan Thompson (London: Pluto Press, 2007), 125– 135; Scott Straus, The Order of Genocide: Race, Power, and War in Rwanda (Ithaca, N.Y.: Cornell University Press, 2006); Scott Straus, “What Is the Relationship between Hate Radio and Vio­lence? Rethinking Rwanda’s ‘Radio Machete,’ ” Politics & Society 35, no. 4 (2007): 609. 63. ​For a defense of judicial expressivism, see Cass Sunstein, “On the Expressive Function of the Law,” Pennsylvania Law Review 144, no. 5 (1995): 2021; Mark A. Drumbl “The Expressive Value of Prosecuting and Punishing Terrorists: Hamdan, the Geneva Conventions, and International Criminal Law,” George Washington Law Review 75, no. 5/6 (2007): 1165. 64. ​Joshua Dressler and Stephen  P. Garvey, Cases and Materials in Criminal Law (St.  Paul, Minn.: West Academic Publishing, 2012), 2. 65. ​Mark A. Drumbl, Atrocity, Punishment and International Law (Cambridge: Cambridge University Press, 2007), 175. 66. ​Dressler and Garvey, Cases and Materials in Criminal Law, 599, quoting State v. Johnson (1979): “Any l­egal standard . . . ​must reflect the under­lying princi­ples of substantive law and community values.”

6 • THE INTERNATIONAL CRIMINAL TRIAL REC­O RD AS HISTORIC AL SOURCE T H I J S B. B O U W K N E G T

The ­people who asked o­ thers to die for ideas w ­ ere the last ones to do so themselves. —­Alain Mabanckou, Broken Glass (2009)

Historians do not only study the past, they also study the dealing with the past, increasingly including transitional justice rituals such as criminal ­trials.1 As some of the first modern-­day international(-­ized) ad hoc tribunals and hybrid courts have fulfilled their mandates, the time has come to have a first glance at what they have actually accomplished or not, appraise their legacy, and plunge into the vast archives they have inadvertently produced along the way. Informed by a rising interest in the legacy of international criminal tribunals by ­legal scholars and tribunal protagonists in a debate that is dubbed “Legacy Talk,”2 taking ­wholesale stock is an empirically impossible proj­ect since from the outset, all ­these courts—or at least their po­liti­cal architects, l­egal agents, and principled campaigners—­have ascribed to the cosmopolitan international justice proj­ect a myriad of ambitious functions.3 Besides the primary purpose of investigating, charging, prosecuting, judging, and sentencing individuals for their acts or omissions constituting genocide, crimes against humanity, and war crimes committed in the recent or remote past, they have also claimed to possess extralegal powers such as healing the world,4 eradicating impunity for gross h­ uman rights violations, deterring potential génocidaires, ending protracted wars, serving victims with closure, reconciling fractured socie­ties, ascertaining truth, and writing historical rec­ords. Adding to what is mostly a theoretical discussion among nonhistorians,5 this chapter confines its scope to the alleged truth-­finding and history-­writing functions of the interna118



The Trial Rec­ord as Historical Source 119

tional criminal tribunals, as well as the archival rec­ord they have left b­ ehind. Based on my thirteen years of first-­hand trial observations at and analy­sis of court rec­ ords from the International Criminal Tribunal for Rwanda (ICTR), the Special Court for Sierra Leone (SCSL), and the International Criminal Court (ICC),6 it focuses on three interrelated questions: how to understand the invocation of historical narratives in international criminal t­ rials; how to position court judgments in the larger historiography on mass vio­lence; and how to approach the courts’ ­trials and the trial rec­ords as historical sources.

history in the courtroom History with a Capital “H” Beginning in January 2016, the former president of Côte d’Ivoire, Laurent Gbagbo, alongside former youth minister Charles Blé Goudé, has stood trial before the International Criminal Court (ICC) on charges of war crimes and crimes against humanity allegedly perpetrated between December 2010 and April 2011. A criminal trial in essence, the proceedings are informed by such a complex reading and litigation of Côte d’Ivoire’s recent past that the trial has turned into a platform for broad discussions on history. Madam President, my entire life, and this is a known fact not only back in Côte d’Ivoire but throughout Africa, and throughout France, throughout po­liti­cal France notably, I have been fighting for democracy. I asked my counsel only last week, and I said that I wanted to bring you all the books that I’ve written, and they said that it was too late to introduce ­these books, but once we have finished, what­ever the result may be, what­ever you decide, I ­w ill send a batch of books written by Gbagbo to the Office of the Prosecution, and I ­will send you also a batch of my books, ­because, well, that is the man that I am.7

In February 2013, Gbagbo told the Pre-­Trial Chamber of the ICC that he would share with them his scholarly publications. His promise came a­ fter several days of hearings on the question of w ­ hether the four charges of crimes against humanity against him ­were to be confirmed and he should be sent to trial. As in most international criminal proceedings, the arguments and case narratives of the prosecution and defense w ­ ere confrontational. Regarding what occurred in the five months a­ fter Côte d’Ivoire’s hotly contested elections in November  2010, which left an estimated three thousand civilians dead, they ­were diametrically competing. Forever defiant and self-­aggrandizing, Gbagbo grasped the moment to lecture Chief Prosecutor Fatou Bensouda and the judges about the po­liti­cal history of West Africa in general and Côte d’Ivoire in par­tic­ul­ ar. In so d­ oing, and speaking mostly in the third-­person from b­ ehind his court-­booth-­turned-­lectern,

120

thi js b. bo u w knegt

Gbagbo reinforced not only his position as the all-­knowing leader and central agent in recent Ivoirian history but also his supreme expertise as a history professor.8 In his opinion, Bensouda, a Gambian jurist who had previously prosecuted crimes in Rwanda and the Demo­cratic Republic of the Congo (DRC),9 had distorted the facts and “constructed a mere caricature of the history of Côte d’Ivoire, which made it impossible for them to fully grasp the issues at stake or to understand the real­ity of the crisis in this country.”10 For international prosecutors, academic discussions of history are not the prime focus. Instead, the intention is to substantiate the individual criminal responsibility of the accused whom they presume to be “most responsible” for the crimes perpetrated by o­ thers on the ground. In that pro­cess, however, they cannot escape invoking history and employing ­legal language to frame the complex realities of mass vio­lence—­and vice versa. Since they are not trained historians, they ­ought to be diligent, realistic and modest in what they can truly accomplish. Enter the trial against Gbagbo—­who was tried alongside his alleged mouthpiece, former student leader Charles Blé Goudé.11 One day before the trial in January  2016, Bensouda told journalists “that the purpose of the trial . . . ​is to uncover the truth through purely a ­legal pro­cess . . . ​, for the sake of ­doing justice for the victims; and to prevent mass atrocities recurring in the f­ uture.”12 On first notice her avowal was innocuous, blending as it did into the ever-­repetitive hum of platitudes made by the agents of global justice. A ­ fter all, as a “civil servant of humanity,” she represents the victims, and as prosecutor she is tasked with the burden of proof, ­after investigating incriminating and exonerating circumstances.13 But a vigilant reading of the prosecution’s case theory warrants caution since she puts forward a feeble historical narrative: Upon assuming the Presidency of Côte d’Ivoire in October 2000, Gbagbo harboured the objective of retaining power by, inter alia, repressing or violently attacking ­those who challenged his authority.  In the following years, knowing that a freely-­contested presidential election was inevitable, Gbagbo and the Inner Circle jointly conceived and implemented a common plan to keep him in power by all means, including by committing the crimes charged (“Common Plan”). By 27 November 2010, the implementation of the Common Plan had developed to include a State or organisational policy aimed at a widespread and systematic attack against perceived Ouattara [his opposing presidential candidate] supporters.14

Based on an unsophisticated general hypothesis about the workings of the African state,15 a case theory commences two years before the start of the ICC’s temporal jurisdiction from July 1, 2002, the allegations culminate in the core charge that from November  2010 “Gbagbo and members of the Inner Circle jointly planned, organised, coordinated, ordered, induced, authorised and allowed vari­



The Trial Rec­ord as Historical Source 121

ous mea­sures to implement the Common Plan and the crimes charged. In pursuance of the Common Plan, pro-­Gbagbo forces attacked, killed, injured, raped and persecuted hundreds of civilians.”16 ­There is no doubt that the exact criminal incidents that ­were being retroactively prosecuted ­were not only committed in the past, but also not in historical isolation.17 Additionally, they took place in the immediate context of the first presidential elections in a de­cade, a preceding civil war, prior po­liti­cal and ethnic animosity, and anti-­Western sentiments.18 Facing ­these social, po­liti­cal, and historical dimensions, in Bensouda’s quest for the truth, history does actually appear to m ­ atter. What she wants to show the Trial Chamber is that Gbagbo—­and his wife Simone, also a trained historian19—­had always had an insatiable appetite for power. Once they ­were served the main dish (the presidency), the ­couple was not about to break bread with ­others. Moreover, Laurent Gbagbo’s intent to commit crimes was, according to Bensouda, partly demonstrated by “his historical repression of his po­liti­cal opposition.”20 In rebuttal, Gbagbo’s defense squad zealously picked up on the prosecution’s historical tone, accentuated its own pitch, and strategically put history at center stage: Madam President, your honours, ­behind all judicial proceedings is a story, a story of dates, places and events, the story of one man, President Gbagbo. . . . ​In international proceedings, it is also history with a capital “H” that is being written. In fact, the situations in which the Court must intervene require it to take on board historical, so­cio­log­ic­ al and po­liti­cal context without which the case cannot be fully understood. It is also a history of a country, a region and a ­people and its sufferings that is being written.21

“passé du futur” For Gbagbo the history professor, his criminal trial was not only an arena in which he had to defend himself from criminal charges, but even more a public lecture hall in which he set out to set the historical rec­ord straight.22 Outmaneuvered by Gbagbo and Blé Goudé’s oratorical finesse and sensitive to the reaction their defense sparked among the many “pro-­Gbagbo” Ivoirians on the public gallery, the prosecution, through lead Prosecutor Eric MacDonald, sought to temper the role of history: Iw ­ ill now highlight some of the historical background and context that lead to the post-­election vio­lence. This context is not to establish the history of Ivory Coast. It is not the purpose of this trial. This context is relevant to describe the creation of the “Common Plan” and, more importantly, it shows evidence of

122

thi js b. bo u w knegt

Mr Gbagbo and Mr Blé Goudé’s intent and knowledge of past vio­lence and how their methods in the past evolve over the years. It ­will serve also as pattern evidence, the “passé du futur,” the awareness of their actions.23

A minute ­later, however, MacDonald spiraled back to 1993, the year Côte d’Ivoire’s first president, Félix Houphouët-­Boigny, passed away and nationalistic, ethnic, and xenophobic po­liti­cal discourse (Ivoirité) engulfed the po­liti­cal and social arena.24 From the moment Gbagbo became president in 2000, alleged the prosecutor, unfolded a “pattern of repression of opposition with repeated allegations of crimes committed by pro-­Gbagbo forces; a pattern of denial of t­hese crimes by members of Mr Gbagbo’s inner circle; a pattern of failure to hold anyone accountable for ­these crimes; a pattern of divisive identity-­based politics and the use of speech by Blé Goudé and o­ thers to mobilize the youth and incite them to vio­lence. You w ­ ill see how t­ hese patterns are repeated in 2010.”25 In order to demonstrate t­hese patterns, including Gbagbo’s “intentionality” to commit crimes against humanity in the ­future as well as his awareness that ­these would be “committed in the ordinary course of events,”26 MacDonald told the bench they ­will “hear evidence about the crimes themselves from Ivoirian civilians, overview evidence on the historical and po­liti­cal origins of the crisis, expert evidence, and the evidence of many insider witnesses.”27 Thus, in what appeared to be an inevitable U-­turn, the prosecution invoked historical patterns and even vowed to pres­ent evidence to substantiate history. For Paolina Massidda, victims’ representative—­who, according to her, “are the very sad raison d’être of the proceedings”—­the historical latitude of the trial was even wider: “We must all keep in mind ­after all that the history that we are ­going to try to retrace h­ ere in this court room reflects not only the past, but also the pres­ent and the ­future.”28 Echoed but twisted by the defense, that reasoning figured in a pre­sen­ta­tion in which a positive view of Gbagbo in the past was to carry over to how he was to be perceived in the pres­ent and into the f­uture. In the defense’s version, Gbagbo, “as all ­people of Côte d’Ivoire and all Africans know,” is a true demo­crat who has always “promoted a multiparty system” and has even “established a remarkable system of ­free mandatory education.”29 All he did was fight to protect democracy from armed rebel forces, the country’s former colonizer France, and armed groups that supported po­liti­cal rival Alassane Ouattara, the current president. The defense said “that the Prosecution is so uncomfortable with their scenario, which has no foundation, that to keep this scenario alive, they remain s­ ilent about all the high-­level events of Côte d’Ivoire history of t­ hose years.”30 By omitting charges31 against the “other side” to the conflict “and re-­ writing history,” the prosecution showed that they “did . . . ​not interest themselves in understanding the history of Côte d’Ivoire” and that ­because “the more the Prosecution attempts to provide explanations, the more they find themselves bogged down in contradictions of history.”32 The defense strategy, then, in



The Trial Rec­ord as Historical Source 123

attacking the prosecutor’s limited vision of Côte d’Ivoire’s history, was to shift the Trial Chamber’s focus away from the m ­ atter at hand: Gbagbo’s alleged individual criminal responsibility for four events in the months between November  2010 and April  2011. Like the prosecutor, the defense abused history by opportunistic framing of past events, operationalizing historical rhe­toric, and selecting witnesses to corroborate the defense’s version of the truth.

“and who is jesus?” Before the three judges in this case—­nonhistorians from Italy, the Dominican Republic, and Trinidad and Tobago lay a massive challenge. They ­were to sift through this plethora of competing statements about the past to make findings—­ based exclusively on evidence presented at trial and beyond any reasonable doubt—on a distant history and the role therein of the accused.33 But truth finding, like law in itself, is not an exact science; in the words of one of Charles Dickens’s characters, “it is an ass—­a idiot.”34 A medley of moral agreements, rituals, and interpretations, the law and its application is essentially self-­referential in that it is predominantly about words and their enactment. What appears from the Gbagbo proceedings is that international ­trials, which argue the law, are layered word games.35 They are the battlegrounds of competing narratives about and perspectives on real conflicts from the past—­not so much about the past itself. Exhibiting the most plausible and persuasive account of past events and casting doubts over the version of the c­ ounter party is at the heart of the trial. As theaters of thought, ­these hearings compose i­ magined worlds, which are exposed and captured in audible and graphic documentary texts, constituting the trial rec­ord. It is then up to the “trier of fact” or “adjudicator”36 to apply the evidentiary standard of “beyond any reasonable doubt” to get as close to truth as pos­si­ble.37 Gbagbo’s judges ­were faced with an arduous chore, especially when witness testimony turned the trial into almost carnivalesque opera extending to world history. At first, the proceedings moved forward routinely, with four victims of the alleged crimes presenting their harrowing testimonies. Then, however, the prosecution suffered a surprising setback when it was the turn of their fifth witness, Mohammed Sam Jichi, better known as “Sam l’Africain.” As a former “insider,” he was to testify against Gbagbo and corroborate the prosecution’s case theory. On the stand, however, he turned “hostile” and instead of the incriminating story he had told ICC investigators, his testimony attempted to exonerate Gbagbo: “He is a professor. He knows the history of Africa. He knows the history of the world. And it was he who could defend the African sovereignty, not only in Côte d’Ivoire, he could defend Africans in general. . . . ​He was a ­great head of state. . . . ​ That’s my personal analy­sis. And in the investigations and in many documents you ­will read that this is the truth what I say to you.”38

124

thi js b. bo u w knegt

Nodding in agreement, Gbagbo the historian could certainly subscribe to this narrative. But moments ­later, the witness drifted, saying: “When I see the history of President Gbagbo it reminds me a l­ ittle of that of Jesus and Barnabas. . . . ​ It’s history repeating  itself. . . . ​This is my analy­sis. This is what’s happening to Gbagbo, Jesus and Barnabas.”39 Playing along, Gbagbo’s ­lawyer then staunchly asked, “And who is Jesus?” only for the presiding judge to interrupt: “I think ­we’re ­going a ­little bit too far with this questioning on the Holy Bible. We should come back a bit to the facts. Please.”40 For the bench, all ­these questions about history more broadly ­were window-­dressing, in­ter­est­ing, perhaps, but irrelevant to answer ­whether or not Gbagbo committed the crimes as charged. Nevertheless, as the perspective widened, so did expectations of the proceedings.

tribunalization of historical injustice When it suits them, judges, prosecutors, and activists happily embrace their role as peace-­and history-­makers, in the name of the victims. But as soon as such expectations look far-­fetched, or are necessarily betrayed, the same ­people quickly state that this is not what courts are for. Can you have it both ways?41 As Thierry Cruvellier observed, tribunals are chameleon-­like, they reformulate their core purposes and manage outsider’s expectations as they go, adjusting to the ebb and flow of the dynamics in the courtroom. Sometimes they claim to write history, only to l­ater claim that strictly applying the law is their work. Can history and justice be written with the same pen? Historians like Gbagbo are rare defendants at the international criminal tribunals,42 arenas that adjudicate agency within the wider spectrum of historical events.43 In essence, they are occupied with the past from the perspective of the pres­ent and directed at the ­future, and as such, they are inherently anachronistic institutions that render opinions about the past from the moral and ­legal perspective of the pres­ent. Continuing with the creation of historical narratives in the age of transitional justice, as well as their uses and abuses, the discussion focuses ­here on the trial rec­ords and archives of international criminal tribunals and their promises and pitfalls as historical sources—­a question that itself becomes pertinent as some tribunals come to a close. In December 2015, the ICTR delivered its last ever judgment, in the so-­called Butare trial, and its residual institution, the United Nations’s Mechanism for International Criminal Tribunals (MICT), began to take shape.44 The ICTR was the first tribunal to close its doors. Other tribunals, like the Special Court for Sierra Leone (SCSL), the Special Panels for Serious Crimes (SPSC) in East Timor, the ICTY and the Extraordinary African Chambers (EAC) have subsequently concluded their mandates. At the same time, a plethora of courts, like



The Trial Rec­ord as Historical Source 125

the ICC, the Extraordinary Chambers in the Courts of Cambodia (ECCC), and the Special Tribunal for Lebanon (STL) are in full operation. At this transitional juncture, ­lawyers and judges make room for archivists and historians. Generally, archivists become the custodians of the court rec­ord, and then historians stand in line to consult the source materials for research into the ­trials themselves as well as the conflicts they have adjudicated.45 But recently, some international courts have started to include not only archivists during the proceedings, but also historians, who, in studying ongoing ­trials, also observe the enactment and unraveling of history therein. For several reasons, this is increasingly the case in the broader field of what is fashionably, albeit ambiguously and controversially, termed “transitional justice,”46 an all-­encompassing field that inherently deals with the violent, repressive, or other­wise burdensome past and the politics of memory.47 Truth and reconciliation commissions that disentangle recent and remote historical injustices are a fine example thereof.48 But courts are, too. They are also occupied with crimes from the past. A cursory overview across the globe shows a rec­ord number of criminal and civil cases that address historical injustices.49 In the mid-2010s, t­ rials concerned crimes g­ oing back in time as far as the Holocaust in the 1940s and extending to places in all parts of the world.50 From the mid-1990s, the world has seen a tribunalization of historical injustice: crimes from the past have increasingly become a ­matter for prosecution. Arguably, all ­these atrocity ­trials deal with history or at least events of g­ reat historical significance. History—or events from the past—is both background and foreground in international criminal justice. Essentially, it is historical wrongs that spark ­these justice initiatives, which in their turn invoke history, historical sources, or historical testimony.51 This correlation occurs at least at four dif­fer­ent levels. First, as a starting point, historically relevant sources concerning large-­ scale ­human rights abuses can themselves lead to ­trials. In Argentina, Guatemala, and Chad, for example, prosecutors have built criminal cases on the rec­ords of truth commissions,52 as well as on formerly secret archives of dictators.53 On the other hand, history books are often among the first references for court staff at international tribunals, who must understand the culture and customs, the social and po­liti­cal landscapes, and historical backgrounds of the conflicts and situations with which they are dealing. At the ICTR, all new staff was virtually required to read Alison Des Forges’ standard work on Rwanda history, before entering any courtroom.54 Likewise, the internal libraries at the tribunals have all collected history books, historical dictionaries, or other relevant sources, such as h­ uman rights reports on the conflict situation. Second, at the pretrial phase, during investigations and preparations of indictment, historians or “experts” on the history of par­tic­ul­ar contexts are often consulted.55 Sometimes they are hired as con­sul­tants to prosecutors, advising them on pos­si­ble sources, contextual information, and case theories. At the ICTY, historians ­were formally represented in the Leadership Research Team, which was

126

thi js b. bo u w knegt

headed by historian Patrick  J. Traynor.56 On the other side of the spectrum, defense teams also regularly consult historians during pretrial investigations and preparations, to c­ ounter claims made by the prosecution’s historians. Yet, this practice is generally not as formalized as it is on the prosecution side. Third, historians are often called by parties during trial, mostly as experts on the context of the alleged crimes. They have featured at all the tribunals and some became well-­known, including Christian Axboe Nielsen, who has a chapter in this book.57 Many became regular witnesses, such as, inter alia, the late Alison Des Forges in ten ­trials at the ICTR,58 David Chandler in two ­trials at the ECCC, Stephen Ellis in one trial at the SCSL, and Gérard Prunier in one trial at the ICC. Robert Donia perhaps tops the list, with testimony in at least fifteen ­trials at the ICTY.59 Historical expert witnessing is not a new phenomenon introduced by the tribunals60 and has remained a continuing practice at the international courts. In the Hissène Habré case before the Extraordinary African Chambers (EAC), Chadian historian Arnaud Dingammadji extensively testified in 2015.61 The fourth, last, and perhaps most typical time for historians to step in is ­after the trial, when they enter the archives to analyze the trial legacies or write histories on the adjudicated conflicts using the trial rec­ord. In this capacity, they have the advantage of being able to take more of a bird’s-­eye point of view.

forced marriage, disfigured offspring But it o­ ught not be just the post-­trial rec­ord that should interest historians; the ­trials themselves as well as the pretrial proceedings are also impor­tant. In t­ hese specific areas, historians assess the retroactive pro­cesses, methodologies, and subsequent pos­si­ble biases involved in fact-­finding and ascertaining the truth in the past. Moreover, understanding the ­trials, such as that of Laurent Gbagbo in which history and historical narratives ­were themselves invoked and debated, is crucial to gaining a perspective on the very birth of the trial rec­ord and its credentials. In the pro­cess, two questions are pertinent: Why was the trial rec­ord created? And how and by whom was it created? Let us begin with the purposes of trial rec­ords in criminal tribunals, which are are tasked with rendering justice in cases in which individuals are suspected of international crimes: genocide, crimes against humanity, and war crimes—­ nothing ­else.62 They do so by eliciting, collecting, presenting, questioning, and reviewing testimony about the past as well documents from the past. Th ­ ere are only two pos­si­ble outcomes of the trial: a narrative of individual guilt or a narrative of individual innocence.63 Th ­ ere is one prob­lem, however. Courts—or at least their creators or protagonists—­“ have always shown a remarkable lack of modesty as to what l­ ittle they can actually accomplish.”64 They have claimed to possess extra-­legal powers such as eradicating impunity for gross ­human rights violations, deterring potential génocidaires, ending protracted wars, serving victims



The Trial Rec­ord as Historical Source 127

with closure, reconciling fractured socie­ties, ascertaining truth and writing historical rec­ords,65 or even healing the world. But courts are not made for that and if they attempt any of t­hese tasks, they risk becoming a “rudderless ship tossed about by the waves” and “end in complete failure.”66 Courts unravel history on the one hand and make history on the other, but they do not necessarily write history. By nature, the tribunals themselves are historical phenomena with retrospective mandates to submit a narrative repre­sen­ta­tion of individual transgression within a par­tic­ul­ar historical context and within a specific ­legal framework. For their part, “judges are acutely aware that their judgement w ­ ill inevitably be viewed as making history and their judgement ­will itself be subject to historiographical scrutiny.”67 Thus, enacting justice and writing history should not be confused. “In the last 2500 years, since the beginnings in ancient Greece of the literary genre we call “history”, the relationship between history and law has been very close,” writes Carlo Ginzburg.68 It is not just an etymological coincidence that “historian” derives from “one who knows” or “judges,”69 but more impor­tant is that the pursuit of truth is central to both professional inquirers. Their dealing with the past, however, diverges. Judges must judge and historians can judge. In addition, ­there is an impor­tant difference between l­egal truth and historical truth,70 and between the texts of court verdicts t and t­ hose of history. Judges submit a narrative repre­sen­ta­tion of individual transgression in a par­tic­ul­ar historical time period within a ­legal framework, thus producing a very specific and constrained kind of normative agency history. Historical writing, on the other hand, focuses on agency as well as structure.71 And whereas historical scholarship takes into account meso-­and macro-­perspectives, judges not only are confined to the micro-­ perspective of the single accused, but also further straitjacketed by temporal, geo­graph­i­cal, and ­legal bars. ­There is thus an inherent relationship between international criminal justice and history, but it is an uneasy one.72 In this forced marriage, the offspring can be disfigured. ­After questioning ­whether international courts delivered innovative and significant understandings of the origins and c­ auses of armed conflict and ­whether or not their historical inquiries undermined due pro­cess and v­ iolated the rights of the accused, anthropologist Richard Wilson, who has a chapter in this volume, concluded that “courts of law produce mediocre historical accounts of the origins and ­causes of mass crimes.”73 As Wilson emphasizes, proving past international crimes demands understanding the conflicts as such, the po­liti­cal and social contexts in which they took place, and how they fit the demanding l­ egal criteria of, for instance, the Genocide Convention. Yet the complex endeavors of judging international crimes and writing history of an armed conflict cannot be characterized by “­either harmonious accord or inherent contradiction.”74 Compared to professional historians, international judges are restrained in vari­ous ways. First, they cannot freely and widely pick their topics and commence their own investigations.

128

thi js b. bo u w knegt

Tribunals prosecute only a limited number of very specific crimes (genocide, crimes against humanity, and war crimes), and by exclusion of other crimes narrow the prism through which they may look at events from the past. Second, tribunals are also constricted in time. The best example is the ICC, which may not deal with events prior to July 2002. At the ICTR, the temporal jurisdiction spanned only the calendar year 1994, thus including half a year a­ fter the genocide but excluding the period of civil war that started in October 1990.75 A third limitation is geography. In Freetown, for instance, at the SCSL, prosecutors and judges ­were to adjudicate crimes only within Sierra Leone, not in Liberia, the country where its prime suspect, Charles Taylor, came from.76 Fourth, ­there are also bars concerning the types of perpetrators. Tribunals are supposed to deal with the most se­nior leaders suspected of being “most responsible” at the macro level, thus barring agency at the meso-­and micro-­levels.77 Generally, prosecutors make a se­lection out of this already exclusive group of agents, and so the majority of perpetrators, particularly at the mid-­and lower levels walk away, uninvestigated, unprosecuted, and unsentenced.78 Fifth, although some of the trial chambers and international criminal tribunals undertake judicial site visits, most are largely dependent on the investigative work of o­ thers, mainly third-­party investigators, prosecutors, and defense ­lawyers.79 Judges are triers of fact, but in most cases they are not in­de­pen­dent seekers of evidence.80 And last, judges have no room for nuance as they can only reach two pos­si­ble conclusions: guilty or not guilty. All ­these ­legal straitjackets distort the historical narrative, as judges have increasingly acknowledged. At the ICTY, for instance, Judge Alphonse Orie told a defence witness, who was a historian: “I want you to understand it, that ­we’re not writing history of war but that we are preparing a judgement, what­ever that judgement ­will be, on an indictment against one accused.”81 Although addressing interpretations of history and limiting the scope of myths and lies about the past, “­trials cannot s­ ettle conflicting interpretations of history” and should therefore “not be expected to write history.”82

overpromised, underdelivered A main issue in international ­trials, as shown by the Gbagbo proceedings, remains the way in which prosecutors, defense counsel, victims’ representatives, witnesses, and judges use nonlegal evidence and historical narratives, and for what reasons: as rhetorical window-­dressing or as material integral to case theories?83 Along the way, the more prosecutorial theories, defense alibis, victim statements, witness testimonies, and court reasoning one hears, the more sceptical one becomes about their meaning and truthfulness. Indeed, “the more t­ rials you follow, the more you start to disbelieve every­one: witnesses, the police, judges, prosecutors, defence ­lawyers, and victims.”84 In the live setting of the



The Trial Rec­ord as Historical Source 129

international criminal trial, elicited, constructed and contested narratives often outweigh the forensic truth in the “Rankean” sense of “wie es eigentlich gewesen,” or “how ­things actually ­were.”85 As the Gbagbo proceedings show, t­ rials are not necessarily driven by a “Rankean rage” to establish the full truth regarding what happened and developing an in­de­pen­dent historiography.86 Much more, they are about testing delicately constructed case theories. The danger of such practices for the historical rec­ord is that ­these theories can turn to deceit, and the public at large leaves that experience in confusion, disbelief, and disappointment.87 As initiators of proceedings, prosecutors often raise high expectations regarding what ­trials can uncover. In the Eichmann trial in 1961, the Israeli Prosecutor Gideon Hausner believed that they could “superimpose on a phantom a dimension of real­ity,” a “living rec­ord of a gigantic h­ uman and national disaster.”88 In Nuremberg and at modern tribunals, it has been no dif­fer­ent.89 In fact, prosecutors cannot get around entering the realms of history. Progressing and escalating over time, the crimes they prosecute are the outcomes of a preconceived plan, a policy, or a sequence of choices and decisions. At the international criminal tribunals, dwelling on the recent or more remote past to infer a criminal mind set or a pattern thus becomes a necessary evil. But unlike historians who have a professional duty to address the past with rigorous, academic, and dispassionate caution, prosecutors cherry-­pick, infer, and frame—or arguably even mutilate—­ the historical narrative in such a way that best fits the charges and tells a coherent story. In effect, they litigate truth more than they uncover truth. Considering the fact that the authoritativeness of their stories are overpromised by their protagonists, overrated by their readers, and distorted by their critics, the consequences of this courtroom storytelling can be devastating, particularly when the evidence turns out to be unavailable, fluid, unconvincing, unreliable, manipulated, or pointing in a dif­fer­ent direction. Judgments as a result, may very well be unsatisfactory, insofar as they are based on prosecutor and defense narratives that do not necessarily conform to generally accepted versions of history, as narrated by prosecutors. This happened, for example, in the flagship cases of the ICTR and the SCSL, where prosecutors sought to adjudicate entire conflicts through elaborate charges against individuals, stumbled, and failed to unveil the rationales ­behind the Rwandan genocide and the civil war in Sierra Leone.90 In Freetown, the prosecutor’s narrative—­that Charles Taylor, “The God­ father,” engaged in a criminal conspiracy with Revolutionary United Front (RUF) leader Foday Sankoh in Libya in the late 1980s to conquer West Africa, terrorize its p­ eople, and enrich themselves with rough diamonds from Sierra Leone—­was reduced by the judges to a short tale of a temporary assailant and planner between late 1998 and early 1999.91 The judges could factually establish that on a handful occasions Taylor provided advice and material to rebels within the timespan of four months. In Arusha, prosecutors, informed by the writings of their con­sul­tant historian Alison Des Forges, framed the 1994 genocide as a

130

thi js b. bo u w knegt

form of “tropical Nazism”92—as if it followed the same pro­cess as the Holocaust in Eu­rope in the 1940s. In an attempt to generate a historical rec­ord of the Rwandan genocide,93 they charged Col­o­nel Théoneste Bagosora with a conspiracy to commit genocide on the case theory that from late 1990 to July 1994, he had executed a “Machiavellian Plan” to exterminate all Tutsis and their Hutu “accomplices.”94 Despite eigh­teen days of testimony by Des Forges,95 Bagosora was found guilty of genocide, crimes against humanity, and war crimes committed only between April 6 and 9, 1994, and his role of “mastermind” was reduced to “proj­ect man­ag­er.”96 Thus, while the historical lead-up to events in 1994 w ­ ere crucial to the prosecution’s understanding of the genocide, it appears that on the basis of testimony from 242 witnesses, nearly 1,600 exhibits, and around 4,500 pages of submissions from the prosecution and defence, the ICTR judges w ­ ere not able (beyond any reasonable doubt) to corroborate historiography as charged by the prosecution. They ruled that several ele­ments commonly considered to be crucial in the planning of the 1994 massacres ­were “not supported by sufficiently reliable evidence” or did “not necessarily demonstrate criminal intent.”97 Despite their acknowl­edgment that “the pro­cess of a criminal trial cannot depict the entire picture of what happened in Rwanda,”98 the judgment was received in Rwanda as a revisionist piece of writing and has indeed been cited by a range of genocide deniers and malevolent historians.99 To treat a trial judgment—­which is a response to the historically framed case theories—as an objective and authoritative account of history can have a devastating social impact. Cognizant, however, of the disappointment their judgment might cause, the judges carefully wrote: “Other or newly discovered information, subsequent t­ rials or history may demonstrate a conspiracy involving the Accused prior to 6 April to commit genocide. This Chamber’s task, however, is narrowed by exacting standards of proof and procedure, the specific evidence on the rec­ord before it and its primary focus on the actions of the four accused in this trial. In reaching its finding on conspiracy, the Chamber has considered the totality of the evidence, but a firm foundation cannot be constructed from fractured bricks.”100

acts of faith Implicitly, Bagosora’s judges picked up on a crucial deficiency in the tribunals’ fact-­finding: the volatility of the available evidence. Whom and what to believe is a question one is left with ­after the courtroom experience, which, beyond the ­legal questions at stake, is also “a social encounter, where dif­fer­ent modes of being and dif­fer­ent worldviews might conceivably collide.”101 As the Gbagbo case showed, ­trials often turn out less about ascertaining the truth and more about narrative and proving a case theory: Who pres­ents the best, most credible and convincing version of events? Prosecutors employ their version of the truth as a vehicle to secure convictions, ­lawyers do so to best serve their clients, and



The Trial Rec­ord as Historical Source 131

judges apply the law and regulations to the evidence put in front of them and form a “reasoned” opinion. But the truths presented, heard, and deliberated at ­trials are extremely delicate. With the relative exceptions of Nuremberg, the ICTY, and the EAC, ­there is a critical shortage of documentary, forensic, or other tangible types of evidence at most tribunals, including the ICC, which up to 2018 dealt strictly with cases in oral socie­ties in sub-­Saharan Africa. In practical terms, this means that trial judges have to rely almost exclusively on witness accounts given five, ten, sometimes twenty or even more years ­after the crime has occurred.102 In other words, the data concerning atrocity crimes are based on ephemeral stories drawn from the fallible memory of witnesses who ­were close to the vio­lence103—in what Nancy Coombs has termed a procedure of “fact-­finding without facts” in a ground-­breaking study.104 This Pandora’s box105 in what ­really is an existential “truth-­finding crisis,” has been opened in par­tic­ul­ar in the Congo and ­Kenya cases at the ICC.106 The consequences can be devastating. In the latter case, which concerned the 2008 postelectoral crimes against humanity in K ­ enya, witnesses ceased to cooperate, dis­appeared, or recanted their prior statements en masse, and as a result, all ­trials ­were vacated.107 In case judges render an international criminal judgment, its foundation may be “uncertain” and “often be ­little more than an act of faith.”108 It is a statement that recalls a reflection by Norwegian Judge Erik Møse, who presided over the ICTR’s Bagosora trial, that “it is the responsibility of the judge to listen to the testimonies of the witnesses. Each pres­ents his or her version of the truth. Our task is to get as close to it as pos­si­ble.”109

­legal labyrinth of history ­ rials and their judgments may lead to historically impoverished narratives, yet T it is also the case that previously unknown facts are exposed in testimonial stories, guilty pleas, and evidence.110 For instance, the ICTR established beyond ­legal dispute that during 1994 t­here was a genocidal “campaign of mass killing intended to destroy, in ­whole or at least in very large part, Rwanda’s Tutsi population.”111 In arriving at ­these kinds of significant historical conclusions, courts collect a wealth of evidence through such means as forensic investigations, DNA tests, or ballistics analy­sis that have been carried out by professional experts on demand of tribunals.112 But they also produce historical sources: particularly witness testimony. Among the most in­ter­est­ing achievements of the international courts is that they have established a massive repository of oral history of mass vio­lence, although, again, through the very limited lens of criminal procedure.113 The trial rec­ords produced, a legacy of documents, can be invaluable sources for historians, but should not be the only sources consulted. Understanding how the trial rec­ord was established, who established it, and ­under  what conditions are also impor­tant considerations. We also should

132

thi js b. bo u w knegt

know what made it into that rec­ord, what did not, and how ­these decisions ­were made. What is publicly available is only a part of the trial rec­ord. Confidentiality agreements and sometimes censorship through redactions, witness protection, or agreements with third parties are problematic bars for historians. Getting to know as much as pos­si­ble about the trial and its context requires an interdisciplinary and creative approach, outside of the academic armchair comfort zones. The con­temporary historian must at times turn into an anthropologist or investigative journalist conducting field work, observing ­trials, even participating in them. Many experts from the ­legal field, but also from the social sciences, have taken part in ­trials. Having worked as a researcher at the ICTR and ICC and having attended, monitored, and observed over fifty trial proceedings for about a de­cade, including more recently the ICC’s trial against Laurent Gbagbo, I know that the trial itself and the immediate social, po­liti­cal, and historical context in which it takes place teach us a lot about the way case theories, evidence, and the past are discussed, often beyond what is given in the trial rec­ord. At the same time, though, it is virtually impossible to digest every­thing in a trial. ­After twenty years, the trial rec­ords of the tribunals have grown into an impenetrable labyrinth for newcomers.114 Aside from the millions of pages of evidence, testimony, motions, decisions, and judgments, the ­trials against Charles Taylor and Radovan Karadzic for example, also generated thousands of pages of transcripts requiring months to read through. Even the experienced judge, assisted by a dozen staff members, needs time. In the ten-­year trial against the Serbian nationalist politician Vojislav Šešelj, Judge Mandiaye Niang, who replaced a colleague, took one and a half years to “familiarize” himself with the case rec­ord, on the basis of which he had to pronounce a judgment.115 Likewise, Robert Donia, a historian who testified for the ICTY’s prosecution and wrote a biography of Karadzic, for instance, explained that over the years, he was able to sift through only a segment of the evidence.116

journalism as the first draft of history In tandem with the continuous creation of trial rec­ords, or even in lieu of them, observers’ accounts can be crucial to our understanding of ­trials. A pioneer in this re­spect is Hannah Arendt. In her journalistic observations, repre­sen­ta­tions, and opinions in The New Yorker about the Eichmann trial in Jerusalem, she not only showed that ­trials are dynamic pro­cesses and historical events in their own right, but also that they involved a certain banality, as it became laborious everyday business.117 Even Arendt attended only a part of the trial. Yet, in all re­spects, ­trials are real-­time per­for­mances, which require real-­time observation and study—­their meaning gets lost when read from the armchair at university offices or ­legal libraries. Unfortunately for the post-­trial researcher, most of the courtroom drama and theatrics cannot be captured in ­legal filings, written testimonies, and documents.



The Trial Rec­ord as Historical Source 133

From the transcribed trial rec­ord, we cannot see the tears, hear the voices of witnesses, or feel the tension in the courtroom and public galleries. The transcript does not rec­ord the applause of Gbagbo supporters in the ICC’s public gallery when “their defendant” made a court entrance, for instance.118 It does not reflect how the Rwandan singer Simon Bikindi sang his closing statements at the ICTR,119 and or how Thomas Lubanga, through his intense staring, confused a former child soldier witness, who accordingly recanted his testimony at the ICC.120 ­These trial events are often not even captured on the video or audio recordings, which are edited and distributed by the courts. Legally speaking, t­hese kinds of events may be irrelevant, but for the social scientists and historians who need to contextualize their sources, they are crucial. In this re­spect, one might recall the popu­lar saying that journalism is the first draft of history. At the ICTY, ­trials ­were continuously covered by journalists from the region as well as by the international press corps.121 At the ICC, which deals with an increased case load from many situation countries, the rec­ord on reporting by “locals” is mixed and by international journalists increasingly or­ga­nized and specialized.122 For the more obscure or distant trial settings coverage was much more sparse. It is due to the endeavors of con­temporary in­de­pen­dent investigative journalists like Thierry Cruvellier that we know at least some critical details about the many ­trials and errors at the ICTR in Arusha between the crucial years of 1997 and 2002.123 Like Arendt and Cruvellier, many other investigative reporters provide critical insight into the daily realities of the trial, including vital data that can hardly be distilled from the trial rec­ord or from the preselected official “outreach” of the tribunals.124 Observers also highlight the critical moments in ­trials, which in daily real­ity are ceaselessly boring, banal, and disillusioning. But ­there is something ­else. The post-­trial researcher can, of course, use official court transcripts, videos, and audio from the trial rec­ord to corroborate basic facts, such as dates and names. But regrettably, the trial rec­ord is not always complete.125 This is due to a variety of ­causes, including situations in which certain evidence was not used at trial, documents w ­ ere confidential, parts of the trial took place b­ ehind closed doors, documents ­were ­later redacted, or, in the most extreme cases ­trials ­were censored. For instance, only ­those pres­ent at the delivery of judgment in Charles Taylor’s trial, including myself, witnessed the strong disagreement from the alternate judge, Malick Sow, from Senegal, who had sat through the entire trial and said that he found the prosecution case was “not proven beyond any reasonable doubt,” that “­there had been no deliberations,” and that he feared the international justice system “was heading to disaster.”126  But during his statement, the tribunal was quick to cut off the microphone and lower the blinds in the public gallery. Afterward, his statements ­were deleted from the court’s video rec­ord and from the court transcript. His name was initially not even published on the written judgment.127

134

thi js b. bo u w knegt

lacunae ­ ese deliberate erasures from the trial rec­ord are disturbing. But ­there are other, Th more ordinary, ele­ments that we ­will not find in the transcripts or videos. ­These include judges and other court staff falling asleep during the proceedings or arguing among themselves b­ ehind the screens. But a more impor­tant consideration is the fact that we cannot scrutinize judges’ deliberations or look at their notes. ­These are private and confidential. Similarly, we cannot see the vari­ous draft judgments with dif­fer­ent scenarios that are usually produced by a batch of ­legal officers and interns. ­These are prob­ably electronically deleted or physically destroyed. Also, we cannot read about the nonverbal power plays in the courtroom—­for example, between the accused and witnesses—in the written transcript. Nor can post-­trial researchers deduce from the trial rec­ord that some ­lawyers and prosecutors ­were on Facebook, Twitter, or Instagram while their colleagues ­were cross-­examining their witnesses—­not a trivial ­matter when it comes to war crimes ­trials. What we do know about such circumstances comes mainly from in­de­pen­dent observers or journalists, who witnessed them from the courts’ public galleries. Thus, we should always read the trial rec­ord in tandem with in­de­pen­dent and impartial first-­hand reports and observations, preferably from outside of the ­legal realm, b­ ecause t­ hese can provide critical context and information other­wise not available from officially distributed court rec­ords. Now, shifting away from the trial itself, ­there is another lacuna in the trial rec­ ord. It concerns the rec­ords that ­were collected, selected, and produced in the lead up to the ­trials: the investigations and the gathering of evidence. All tribunals start with sending investigators into the field. They conduct the first interviews, locate documents in dusty archives, follow leads, stumble upon vari­ous crime scenes, conduct exhumations, talk to low-­level perpetrators, tap phones, collect materials from intelligence agencies, government agencies, and U.N. peacekeeping missions, conduct forensic analy­sis, and more. While all this collecting of facts, evidence, and testimony may be one of the most in­ter­est­ing and essential aspects of the work of a tribunal, it is also the least transparent part of the tribunal’s work and seldom finds its way to the public trial rec­ord. This is the case both for the prosecution and for the defense, and the result is a black hole in the trial rec­ord. Often we do not know anything about the investigators, their professional background, their methodologies, and even their own reliability and objectivity.128 We know l­ ittle about decisions as to how to investigate, when to investigate, where to investigate, and whom to investigate—­and just as l­ittle about decisions as to when not to investigate, where not to investigate, and whom not to investigate. In the court rec­ords and tribunal archives, we do not find minutes of meetings of investigative teams, decisions by chief investigators, or investigator’s notebooks. In addition, the trial rec­ord does not include testimonies that w ­ ere collected during the initial investigations but ­were not introduced as evidence in the ­trials.



The Trial Rec­ord as Historical Source 135

More impor­tant is that the rec­ords of the prosecution, the defense, and the judges’ chambers are not included in publicly available archives, so we miss a ­great deal of valuable information. It is only quite recently that we have become more aware of the methodologies applied in ­these vital areas of work.129 The ICC, in its three judgments, has opened the door for in­ter­est­ing insights into the prosecutor’s investigations in the Demo­cratic Republic of the Congo,130 particularly, in the Lubanga case.131 Likewise, we also continue to learn more in the K ­ enyan cases as ­lawyers have been calling for an in­de­pen­dent investigation of the Office of the Prosecutor’s modus operandi of alleged witness tampering.132 Thus, as the ­trials continue, the trial rec­ord expands and our understanding of it increases. However, impor­tant information from the ­trials and tribunals may never reach the public at large, or the research community.

conclusion In the age of transitional justice, historians increasingly study how the past is dealt with, including the tribunalization of remote and recent historical injustices. Essentially, it is historical wrongs that catalyze atrocity t­ rials, which in their turn summon historical sources, historical testimony and historical narratives. History and international criminal justice are also interlinked in cases such as the crimes against humanity in Côte d’Ivoire, the genocide in Rwanda, and the war crimes in Sierra Leone, in which judgments require understanding the roots of conflicts as such and the po­liti­cal and social contexts in which they took place. During and a­ fter ­these ­trials, the expanding rec­ords ultimately become part of the histories of mass vio­lence and historiographies of the individual conflicts. This chapter has shown, first of all, that international tribunals, which are themselves historical phenomena, select, collect, and create a par­tic­u­lar and complex set of sources. Through the prism of law, t­ rials produce diverse accounts of mass vio­lence. They are also theaters of thought about the past, consisting of mentally constituted and legally straitjacketed stories, which are captured in audible, vis­i­ ble, and documentary texts. Second, this chapter has also shown that prosecutors, ­lawyers, and judges purposely distort the historical rec­ord in the pro­cess of adjudicating agency in the larger context of mass vio­lence. The Gbagbo trial demonstrates that history is used in case theories and abused as rhetorical window-­ dressing for both l­egal and extra-­legal purposes. In writing judgments based solely on evidence elicited during the trial, judges are straitjacketed into applying a legalistic style of story-­telling to a ­legal form of agency history. Third, while international t­ rials deal with the past, look at (criminal) acts performed in the past, uncover sources from the past, hear testimony about the past, and generate opinions of the past, the ability of the international criminal justice system to judge history is limited. Thus, this chapter argued that court judgments should not be considered objective versions of history and that they should instead be treated

136

thi js b. bo u w knegt

with extreme caution since their authoritative worth is overpromised by their authors, overrated by their readers, and distorted by their critics and po­liti­cal opportunists in ways that can cause disillusionment, distrust on the part of its recipients, and possibly delegitimization of the international criminal justice system. In fact, despite being mutually dependent, international criminal justice and history are not written with the same pen and ink. They do dispel misunderstandings of the past, as well as lies about it, and they do make significant findings that are beyond dispute. However, tribunals, which operate in self-­referential ­legal frameworks, as hermetically sealed oratorical rituals of adversarial trial proceedings, and require a standard of proof beyond any reasonable doubt, should not to be treated as protohistorians. This role is much better suited to truth commissions, which are tasked with laying bare ­causes, courses, and consequences of violent conflict. Tribunals, on the other hand, render truths and judgments that are ­shaped by the circumstances in which they are written and are final. That is the biggest difference between judges and historians: the historian’s work never ends. The anthropologist Michael Jackson elucidates this point quite nicely. “In the pro­cess of historical explanation,” he writes, “accounts of the past continually evolve in response to the needs of the pres­ent, in dialogue with o­ thers and with our own imagination.”133 Thus, in history, nuance, new insight and revision are the rule rather than an exception. Through historical investigation, what are taken to be facts can be shown to be semi-­truths or lies, or vice versa. Discovery of new facts, debates about them, and reinterpretations of them continually shift understanding of historical events. Judges know this and have increasingly acknowledged it in trial judgments. The conclusions reached from the totality of the trial rec­ord in lieu of evidence that was not tendered in the t­ rials can thus be dif­fer­ent than the collection of the tribunal’s individual judgments, which, in turn, imposes a huge responsibility on historians.

notes 1. ​Antoon de Baets, “Truth Commissions as Protohistorians,” V.V.N.—­Berichten, Tijdschrift van

de Vereniging voor de Verenigde Naties 117, no. 4 (2002): 1 (in Dutch), 3–19. 2. ​For insightful discussion on the topic, see Sara Kendall and Sarah Nouwen, “Speaking of Legacy: ­Toward an Ethos of Modesty at the International Criminal Tribunal for Rwanda,” ­Legal Studies Research Paper Series, Paper No. 20/2016 (Cambridge University, April 2016). 3. ​Kamari Maxine Clarke, Fictions of Justice: The International Criminal Court and the Challenge of L ­ egal Pluralism in Sub-­Saharan Africa (Cambridge: Cambridge University Press, 2009), 63–65. 4. ​The ceremonial opening of the permanent premises of the International Criminal Court (ICC) on April 19, 2016, attended by Ban Ki Moon, King Willem Alexander, and other dignitaries, ended with a live per­for­mance of artists singing Michael Jackson’s song “Heal the World.” 5. ​See, for instance, Carsten Stahn, Re-­Constructing History through Courts: Legacy in International Criminal Justice, SSRN, June 9, 2015, https://­ssrn​.c­ om​/­abstract​=2­ 616491 (accessed on February 21, 2017).



The Trial Rec­ord as Historical Source 137

6. ​Between 2003 and 2016, the author attended hearings in the war crimes cases of vari­ous sus-

pects: over twenty at the ICTR; nine at the SCSL; and sixteen at the ICC. He worked as a researcher at the ICTR (2004) and the ICC (2006), was a trial monitor at the SCSL and ICC for Amnesty International, and worked as a full-­time war crimes t­rials correspondent for Radio Netherlands Worldwide and the International Justice Tribune. 7. ​International Criminal Court (ICC), Pre-­Trial Chamber (PTC) I, Situation: Republic of Côte d’Ivoire. In the Case of The Prosecutor v. Laurent Gbagbo: Confirmation of Charges Hearing Transcript, ICC-02/11-01/11, The Hague, February 28, 2013, 47. 8. ​Gbagbo holds a BA in philosophy and a PhD in history. He was a professor of history and geography at the University of Abidjan, where he ­later worked as a researcher and became director at the Institute of History, Art, and Archaeology of Africa (IHAAA). Cyril  K. Daddich, Historical Dictionary of Côte d’Ivoire (The Ivory Coast), 3rd ed. (London: Rowman & Littlefield, 2016), 261. 9. ​See Thijs Bouwknegt, “Bensouda: An African Heritage,” International Justice Tribune 142 (December 21, 2011): 4. 10. ​ICC, PTC I, Prosecutor v. Laurent Gbagbo: Confirmation of Charges Hearing Transcript, 44–45. 11. ​ICC, Trial Chamber (TC) I, Situation in the Republic of Côte d’Ivoire, The Prosecutor v. Charles Blé Goudé: Decision on Prosecution requests to join the cases of The Prosecutor v. Laurent Gbagbo and The Prosecutor v. Charles Blé Goudé and Related ­Matters, ICC-02/11-02/11, The Hague, March  11, 2015; ICC, PTC I, Prosecutor v. Laurent Gbagbo: Decision on the Confirmation of Charges against Laurent Gbagbo, ICC-02/11-01/11, The Hague, June 12, 2014. 12. ​ICC, Office of The Prosecutor (OTP), “ICC Prosecutor’s Statement at Press Conference, ahead of the Trial-­Start of the Prosecution’s Case against Messrs. Laurent Gbagbo and Charles Blé Goudé,” Press Release, January 27, 2016. 13. ​“The Prosecutor ­shall: In order to establish the truth, extend the investigation to cover all facts and evidence relevant to an assessment of ­whether ­there is criminal responsibility ­under this Statute, and, in ­doing so, investigate incriminating and exonerating circumstances equally.” Rome Statute of the International Criminal Court, A/CONF/183/9*, July 17, 1998, art. 54 (a). 14. ​ICC, TC I, Situation in the Republic of Côte d’Ivoire, The Prosecutor v. Laurent Gbagbo and Charles Blé Goudé: Corrected Version of Prosecution’s Pre-­Trial Brief, ICC-02/11-02/11, The Hague, July 28, 2015, para. 5. 15. ​Skype interview with Dr. Scott Straus, February 26, 2016. 16. ​ICC, TC I, The Prosecutor v. Laurent Gbagbo and Charles Blé Goudé: Corrected Version of Prosecution’s Pre-­Trial Brief, para. 7. 17. ​In making its case for crimes against humanity (a systematic and widespread attack), the prosecution relied on acts committed against civilians during the course of thirty-­eight incidents but charged only four thereof. See for details: ibid., paras. 288–358. 18. ​Scott Straus, Making and Unmaking Nations: War, Leadership, and Genocide in Modern Africa (Ithaca, N.Y.: Cornell University Press, 2015), 123–168; Mike McGovern, Making War in Côte d’Ivoire (London: Hurst, 2011). 19. ​Daddich, Historical Dictionary of Côte d’Ivoire, 264. Simone Gbagbo, as argued in the prosecution’s case, was a key contributor to the “common plan” and as such also f­ aces ICC charges. Côte d’Ivoire’s judiciary, however, chose to prosecute her at home and not hand her over. She went to trial for crimes against humanity in May  2016. ICC, PTC III, Prosecutor v. Simone Gbagbo: Warrant of Arrest for Simone Gbagbo, ICC-02/11-01/12, The Hague, February 29, 2012; Robey Corey-­Boulet, “Gbagbo’s Trial Is the Latest Sign of Victor’s Justice in Côte d’Ivoire,” World Politics Review, May  4, 2016, http://­www​.­worldpoliticsreview​.­com​/­articles​/­18669​ /­gbagbo​-­s​-­trial​-­is​-­the​-­latest​-­sign​-­of​-v­ ictor​-s­ -​ ­justice​-­in​-­cote​-­d​-­ivoire.

138

thi js b. bo u w knegt

20. ​ICC, TC I, The Prosecutor v. Charles Blé Goudé: Corrected Version of Prosecution’s Pre-­Trial

Brief, para. 439. 21. ​ICC, PTC I, Situation: Republic of Côte d’Ivoire. In the Case of The Prosecutor v. Laurent Gbagbo: Confirmation of Charges Hearing Transcript, Case no. ICC-02/11-01/11, The Hague, February 19, 2013, 7. 22. ​During his detention, Gbagbo co-­authored with French journalist François Mattei the biography For Truth and Justice: Revelations on a French Scandal, in which he sketches a po­liti­cal self-­portrait that Gbagbo would like his readers to see. See Sophie T. Rosenberg, “A Portrait from The Hague: All You Need to Know about What Laurent Gbagbo Wants You to Know,” Justice in Conflict, January  20, 2016, https://­justiceinconflict​.­org​/­2016​/­01​/­20​/­a-​ ­self​-­portrait​ -­from​-­the​-­hague​-­all​-y­ ou​-­need​-­to​-­know​-­about​-­what​-­laurent​-­gbagbo​-­wants​-­you​-­to​-­know. 23. ​ICC, TC I, Prosecutor v. Laurent Gbagbo and Charles Blé Goudé: Trial Transcript, ICC-02/1101/15, The Hague, January 28, 2016, 70. 24. ​Straus, Making and Unmaking Nations, 131–138. 25. ​ICC, TC I, Prosecutor v. Laurent Gbagbo and Charles Blé Goudé: Trial Transcript, 72–73. 26. ​ICC, TC I, Prosecutor v. Laurent Gbagbo and Charles Blé Goudé: Corrected Version of Prosecution’s Pre-­Trial Brief, paras. 439–440. 27. ​ICC, TC I, Prosecutor v. Laurent Gbagbo and Charles Blé Goudé: Trial Transcript, 64. 28. ​Ibid., January 29, 2016, 59. 29. ​Ibid., February 1, 2016, 11–12. 30. ​Ibid., 12, 14. 31. ​Bensouda swore, however, “This is our first case to reach trial in the Situation of Côte d’Ivoire. Th ­ ere ­will be ­others as our in­de­pen­dent and impartial pursuit to hold t­hose most responsible for the post-­election vio­lence in the country, irrespective of po­liti­cal affiliation or side, remains firm. We ­will not falter, Mr President, ­until this work is done” (ibid., January 28, 2016, 50). 32. ​Ibid., 12, 43, 66. 33. ​As of April 2016, five prosecution witnesses, out of a promised 138, had testified in the trial. Four ­were victims, one was a former ally of Gbagbo. 34. ​Charles Dickens, Oliver Twist (London: Richard Bentley, 1839), 3:279. 35. ​Folkert Jensma, “Het recht gebruikt taal als toverstokje, met eigen spreuken,” NRC Handelsblad, October 26, 2013. 36. ​All the tribunals and courts relevant to this study—­the ICTR, SCSL, and ICC—­work with panels of three judges. At some other jurisdictions, like the Extraordinary Chambers in the Courts of Cambodia (ECCC), the bench is constituted by five judges. 37. ​­After hearing the evidence, ICC judges still have a controversial opportunity, ­under regulation 55, to recharacterize the charges and accordingly adjust the historical narrative in line with the evidence. See Thijs B. Bouwknegt, “Lubanga Trial Ends with (Un)Reasonable Dissent? Some Observations,” Newsletter Criminology and International Crimes 9, no. 2 (2014): 12–13; Carsten Stahn, “Justice Delivered or Justice Denied? The Legacy of the Katanga Judgement,” Journal of International Criminal Justice 12, no. 4 (2014): 809–834. 38. ​ICC, TC I, Prosecutor v. Laurent Gbagbo and Charles Blé Goudé: Trial Transcript, March 15, 2016, 73–74. 39. ​Ibid., 74. 40. ​Ibid., 74–75. 41. ​Philip Gourevitch, “Mass Murder Relies on P ­ eople like Us: An Interview with Thierry Cruvellier,” New Yorker, May 15, 2014. 42. ​­There have been many highly educated defendants but only Ferdinand Nahimana, who was tried by the ICTR for public incitement to commit genocide, was a trained historian.



The Trial Rec­ord as Historical Source 139

United Nations International Criminal Tribunal for Rwanda (UNICTR), The Prosecutor v. Ferdinand Nahimana, Jean Bosco Barayagwiza & Hassan Ngeze: Judgement and Sentence, Case No. ICTR-99-52-­T, Arusha, December 3, 2003. 43. ​The ICTR dealt with crimes that took place in the context of the Rwandan civil war and genocide between 1990 and 1994; the SCSL dealt with crimes that took place in the context of the Sierra Leone civil war between 1991 and 2002. In all the ICC cases, the court deals with events ­after July 1, 2002, the date the court became operative. 44. ​UNICTR, Appeals Chamber (AC), The Prosecutor v. Pauline Nyiramasuhuko, Arsène Shalom Ntahobali, Sylvain Nsabimana, Alphonse Nteziryayo, Joseph Kaynabashi and Élie Ndayambaje: Judgement, ICTR-98-42-­A , Arusha, December 14, 2015; United Nations Security Council (UNSC), “Security Council Press Statement on Closure of International Criminal Tribunal for Rwanda,” Press Release, SC/12188-­AFR/3296-­L/3249, December 31, 2015. 45. ​Some significant studies based on trial rec­ords already tran­spired: André Guichaoua, Rwanda. De la guerre au genocide. Les politiques criminelles au Rwanda (1990–1994) (Paris: La Découverte, 2010); Robert  J. Donia, Radovan Karadžić: Architect of the Bosnian Genocide (New York: Cambridge University Press, 2014). 46. ​For an insightful critique, see Makau Mutua, “What Is the F ­ uture of Transitional Justice?,” International Journal of Transitional Justice, Special Issue 9, no. 1 (2015): 1–9. 47. ​For a fresh insight in this debate, see David Rieff, In Praise of Forgetting Historical Memory and Its Ironies (New Haven, Conn.: Yale University Press, 2016). 48. ​Priscilla B. Hayner, Unspeakable Truths: Transitional Justice and the Challenge of Truth Commissions, 2nd  ed. (New York: Routledge, 2011); Trudy Huskamp Peterson, Final Acts: A Guide to Preserving the Rec­ords of Truth Commissions (Baltimore: Johns Hopkins University Press, 2005). 49. ​Antoon de Baets differentiates between recent historical injustice (past genocides, crimes against humanity, war crimes, and historical crimes) and remote historical injustices (crimes of the past similar to genocides, crimes against humanity, and war crimes). Recent historical injustices include “injustice of which at least some perpetrators or (direct and indirect) victims are still alive,” whereas remote historical injustices include “injustice of which all perpetrators and (direct and indirect) victims are dead.” The latter may also be defined as historical crimes, which fall ­under the scope of ethics and history, while the former may fall within the scope criminal and civil law and h­ uman rights and humanitarian law. See Antoon de Baets, “Historical Imprescriptibility,” Storia della Storigrafica, nos. 59–60 (September 2011): 128–149, 132. 50. ​For example: Dutch war crimes in Indonesia (late 1940s), colonial abuses in ­Kenya (1950s), mass killings in Bangladesh (1970s), the “killings fields” of Cambodia (1970s), death flights and enforced disappearances in Argentina (1970s–1980s), the genocide of the indigenous Ixil Maya’s in Guatemala (1980s), ethnic persecutions and widespread torture in Chad (1980s), the Balkan wars (1990s), the genocide in Rwanda (1994), civil war in Sierra Leone (1991– 2002), the Indonesian occupation and in­de­pen­dence of East Timor (1999). More recent cases investigated include mass vio­lence in the Demo­cratic Republic of the Congo (DRC), Uganda, Central African Republic, ­Kenya, Côte d’Ivoire, Libya, Sudan (Darfur), Mali, Georgia, Burundi, Lebanon, and Kosovo. At the time of writing, initiatives ­were undertaken in relation to Colombia, South Sudan, Sri Lanka, Syria, and many ­others. 51. ​Vasuki Nesiah, “Delimiting Accountability: Writing History Out of Justice,” Rapaport Center Workshop on ­Human Rights and Justice, November 5, 2007. 52. ​Comisión Nacional sobre la Desaparición de Personas (CONADEP), Nunca Más: Informe de la Comisión Nacional sobre la Desaparición de Personas (Buenos Aires: Eudeba, 1984); Comisión para el Esclarecimiento Histórico, Guatemala, memoria del silencio (Guatemala City: Comisión para el Esclarecimiento Histórico, June 1999); and Ministère Tchadien de la

140

thi js b. bo u w knegt

Justice, Les crimes et détournements de l‘ex président Habré et de ses complices: rapport de la Commission d’Enquete Nationale (Paris: L’Harmattan, 1993). Th ­ ese reports ­were used in Argentina’s junta t­ rials, Guatemala’s trial against Efrain Rioss Montt, and the prosecution of Chad’s former President Hissène Habré, respectively. 53. ​Antoon de Baets, “The Dictator’s Secret Archives: Rationales for Their Creation, Destruction, and Disclosure,” in Scholarly Environments: Centres of Learning and Institutional Contexts 1600–1960, ed. Alasdair MacDonald and Arend H. Huussen Jr. (Louvain: Peeters, 2004), 181–196. 54. ​Alison Des Forges, Leave None to Tell the Story: Genocide in Rwanda (New York: H ­ uman Rights Watch, 1999). 55. ​Doris Buss, “Expert Witnesses and International War Crimes ­Trials,” in Narratives of Justice In and Out of the Courtroom: Former Yugo­slavia and Beyond, ed. Dubravka Zarkov and Marlies Glasius (Heidelberg: Springer, 2014), 23–45. 56. ​But we also see it at the national level, in domestic cases of universal jurisdiction. For instance, the special Team on International Crimes within the Dutch Police Force has two historians on staff, on a permanent basis. Interview, Jeroen Toor, Team Leader, International Crimes Unit, April 30, 2014. 57. ​United Nations International Criminal Tribunal for the former Yugo­slavia (UNICTY), TC III, Prosecutor v. Radovan Karadžić: Trial Transcript, Case no. IT-95-5/18-­T, The Hague, July 6, 2011. 58. ​According to the United Nations, Dr. Des Forges testified in ten cases before the ICTR, spending a total of over one hundred days on the stand; see Outreach Programme on the Rwanda Genocide and the United Nations, Judicial Archive of the International Criminal Tribunal for Rwanda (ICTR), http://­www​.u­ n​.­org​/­en​/­preventgenocide​/­rwanda​/­resources​/­podcasts​ .­shtml. According to o­ thers, she testified in eleven t­ rials. The author verified the following ten ­trials, involving twenty-­four accused individuals: Akayesu (February 1997), Media (May 2002), Butare ( June 2004), Military 1 (September-­November 2002), Military 2 (September 2006), Gatsumbitsi (August 2003), Renzaho (March 2007), Ndindabahizi (September 2003), Zigiranyirazo (March 2006), and Rwamakuba ( July 2005). 59. ​Donia, Radovan Karadžić. 60. ​Vladimir Petrović, The Emergence of Historical Forensic Expertise: Clio Takes the Stand (New York: Routledge, 2016). 61. ​Chambres Africaines Extraordinaires (CAE), La Chambre d’Instruction, Ordonnance de designation d’expert, Case No 01/13, Dakar, August 2, 2013; CAE, Rapport d’expertise sur la contexte historique du Tchad sous le regime de Hissein Habre de Arnaud Dingammadji, Case Document D1235, N’Djamena, December 18, 2013; Trust Africa, “Summary of the Fifth Hearing Held on 9 September 2015,” The Public Prosecution versus Hissène Habré, http://­www​.­trustafrica​.o­ rg​ /­i mages​ /­I CJ​ _­reports​ /­E AC​ -­% 20Trial%20Hearing%20Report%20​ -­% 209th%20Septem ber%202015%20​-­%20English​.­pdf. 62. ​This ­legal positivist stance was famously popularised by Hannah Arendt, who, a­ fter observing part of the Eichmann trial in Jerusalem in 1961 wrote: “The purpose of a trial is to render justice and nothing ­else; even the noblest of ulterior purposes—­such as ‘the making of a rec­ ord of the Hitler regime which would withstand the test of history,’ . . . ​—­can only detract from the law’s main business: to weigh the charges brought against the accused, to render judgment, and to mete out due punishment” (Arendt, “A Reporter at Large: Eichmann in Jerusalem—­V,” New Yorker, March 16, 1963, 58–134, 101). She had earlier written: Justice demands that the accused be prosecuted, defended, and judged, and that all the other questions, though they may seem to be of greater import—of “How could it happen?” and “Why did it happen?,” of “Why the Jews?” and “Why the Germans?,” of “What was the role of other nations?” and “What was the extent to which the Allies



The Trial Rec­ord as Historical Source 141

shared the responsibility?,” of “How could the Jews, through their own leaders, cooperate in their own destruction?” and “Why did they go to their death like lambs to the slaughter?”—be left in abeyance. Justice insists on the importance of Adolf Eichmann, the man in the glass booth built for his protection: medium-­sized, slender, middle-­ aged, with receding hair, ill-­fitting teeth, and near-­sighted eyes, who throughout the trial keeps craning his scraggy neck ­toward the bench (not once does he turn to face the audience), and who desperately tries to maintain his self-­control—­and mostly succeeds, despite a ner­vous tic, to which his mouth must have become subject long before this trial started. On trial are his deeds, not the sufferings of the Jews, not the German ­people or mankind, not even anti-­Semitism and racism. (Arendt, “A Reporter at Large: Eichmann in Jerusalem—­I,” New Yorker, February 16, 1963, 40). 63. ​Although it should be noted that sometimes judges are becoming creative in this re­spect. ICC Judge Bruno Cotte, for instance, remarked during the reading of the acquittal judgement in the case of Matthieu Ngudjolo Chui from Congo: “Dès lors, déclarer qu’un accusé n’est pas coupable ne veut pas nécessairement dire que la Chambre constate son innocence” [the fact of deciding that an accused is not guilty does not necessarily mean that the Chamber finds him innocent]. See ICC, Résumé du jugement rendu en application de l’article 74 du Statut dans l’affaire Le Procureur c. Mathieu Ngudjolo le 18 décembre 2012 par la Chambre de première instance II, The Hague, December 18, 2012, para. 6. 64. ​Gourevitch, “Mass Murder Relies on P ­ eople like Us.” 65. ​See, also, Thijs Bouwknegt, “Nothing but Verdicts: The Purpose of Tribunals,” Justice in Conflict, September  25, 2013, https://­justiceinconflict​.o­ rg​/­2013​/­09​/­25​/­nothing​-b­ ut​-­verdicts​ -­the​-­purpose​-­of​-­tribunals. 66. ​District Court of Jerusalem, Attorney General v. Adolf Eichmann: Judgement, Case No. 40/61, Jerusalem, December 11, 1961, para. 2. 67. ​Mark Osiel, “Ever Again: ­Legal Remembrance of Administrative Massacre,” University of Pennsylvania Law Review 144, no. 2 (December 1995): 522. 68. ​He continues: “True, the Greek word historia is derived from medial language, but the argumentative ability it implied was related to the judicial sphere (Carlo Ginzburg, “Checking the Evidence: The Judge and the Historian,” Critical Inquiry 18, no. 1 [Autumn 1991]: 79). 69. ​The word “history” derives from the ancient Greek ἱστορία (historía, “learning through research, narration of what is learned”), from ἱστορέω (historéō, “to learn through research, to inquire”), from ἵστωρ (hístōr, “the one who knows, the expert, the judge”). 70. ​Antoon de Baets, “­After the Genocide: Truth Strategies of Judges and Historians,” Tijdschrift voor geschiedenis 116, no. 2 (May 2003): 212–230 (in Dutch). 71. ​Ibid., 213. 72. ​Dubravka Zarkov and Marlies Glasius, eds., Narratives of Justice In and Out of the Courtroom (Heidelberg: Springer, 2014); William Schabas, “History, International Justice, and the Right to Truth,” in William Schabas, Unimaginable Atrocities: Justice, Politics, and Rights at the War Crimes Tribunals (Oxford: Oxford University Press, 2002), 153–162; Fergal Gaynor, “Uneasy Partners—­Evidence, Truth and History in International ­Trials,” Journal of International Criminal Justice 10, no. 5 (October 2012): 1–19. 73. ​Richard Ashby Wilson, Writing History in International Criminal T ­ rials (Cambridge: Cambridge University Press, 2011), 1. 74. ​Ibid., 13. 75. ​United Nations Security Council (UNSC), Resolution 955. Annex: Statute of the International Tribunal for Rwanda, S/RES/955 (1994), November 8, 1994. 76. ​ Agreement between the United Nations and the Government of Sierra Leone on the Establishment of a Special Court for Sierra Leone & Statute of the Special Court for Sierra Leone, Freetown,

142

thi js b. bo u w knegt

January 16, 2002, annexed to UNSC, Report of the Planning Mission on the Establishment of the Special Court for Sierra Leone, S/2002/246, March 8, 2002. 77. ​On the completion strategy of the ICTY and ICTR, see United Nations Security Council (UNSC), Resolution 1534 (2004), S/RES/1534 (2004), March 26, 2004. 78. ​ICC, OTP, Paper on Some Policy Issues before the Office of the Prosecutor, September 2003. 79. ​With the exception of the ECCC and EAC, t­ here are no investigative judges at the tribunals. 80. ​At the ICC, some trial chambers choose to apply a more inquisitorial approach, calling their own witnesses and visiting crimes scenes. In the Katanga trial the chamber said: “Aside from the opportunity thus afforded to the Chamber to gain a better understanding of the context of the events before it for determination, the main purpose of the site visit was to enable the Chamber to conduct the requisite verifications in situ of specific points and to evaluate the environment and geography of locations mentioned by witnesses and the Accused persons” (ICC, TC II, Situation in the Demo­cratic Republic of the Congo in the Case of The Prosecutor v. Germain Katanga: Judgment Pursuant to Article 74 of the Statute, ICC-01/04-01/07, The Hague, March 7, 2014, paras. 106–108). 81. ​UNICTY, TC I, Prosecutor v. Ratko Mladić: Trial Transcript, IT-09-92-­T, The Hague, December 1, 2015, 42068. 82. ​Richard Ashby Wilson, Writing History in International Criminal T ­ rials (Cambridge: Cambridge University Press, 2011). 83. ​Ibid., 15. 84. ​Thierry Cruvellier, The Master of Confessions: The Making of a Khmer Rouge Torturer (New York: HarperCollins, 2014), 80. 85. ​Leopold von Ranke, “Histories of the Latin and Germanic Nations from 1494–1514,” in The Va­ri­e­ties of History: From Voltaire to the Pres­ent, ed. Fritz Stern (Cleveland, Ohio: World, Meridian, 1956), 57. 86. ​For a discussion on the “Rankean rage” in periods of deep-­rooted po­liti­cal transitions from repression, see Antoon de Baets, Censorship of Historical Thought: A World Guide, 1945–2000 (Westport, Conn.: Greenwood Press, 2002), 22. 87. ​Gourevitch, “Mass Murder Relies on P ­ eople like Us.” 88. ​Gideon Hausner, Justice in Jerusalem (New York: Harper & Row, 1966), 291–292. 89. ​For Robert G. Storey, U.S. trial counsel at the International Military Tribunal (IMT), “the purpose of the Nuremberg trial was not merely, or even principally, to convict the leaders of Nazi Germany. . . . ​Of far greater importance, it seemed to me from the outset, was the making of a rec­ord of the Hitler regime which would withstand the test of history” (quoted in W. R. Harris, “Foreword,” Tyranny on Trial—­The Evidence at Nuremberg [New York: Transaction, 1997], vii). 90. ​See Thijs Bouwknegt, “Unravelling Atrocity: Between Transitional Justice and History in Rwanda and Sierra Leone,” in Genocide: New Perspectives on Its Cause, Course and Consequences, ed. Uğur Ümit Üngor (Amsterdam: Amsterdam University Press, 2016), 217–250. 91. ​Special Court for Sierra Leone (SCSL), The Prosecutor of the Special Court v. Charles Ghankay Taylor: Prosecution Final Brief, Case No. SCSL-2003-01-­T, The Hague, April 8, 2011, and SCSL, TC II, The Prosecutor of the Special Court v. Charles Ghankay Taylor: Judgement Transcript, Case No. SCSL-2003-01-­T, Leidschendam, April 26, 2012, 49676–49677. For a comprehensive study of prosecution and defense discourses, see Marlies Glasius and Tim Meijers, “Constructions of Legitimacy,” International Journal of Transitional Justice 6, no. 2 (2012): 229–252. 92. ​The term was coined in Jean-­Pierre Chrétien, “Enjeux: Un ‘Nazisme tropical’ au Rwanda? Image ou logique d’un génocide,” Vingtième siècle. Revue d’histoire, no.  48 (October–­ December 1995): 131–142.



The Trial Rec­ord as Historical Source 143

93. ​Don Webster, “The Uneasy Relationship between the ICTR and Gacaca,” in Remaking

Rwanda: State Building and H ­ uman Rights a­ fter Mass Vio­lence, ed. Scott Straus and Lars Waldorf (Madison: University of Wisconsin Press, 2011), 185–186. 94. ​UNICTR, Prosecutor versus Théoneste Bagosora et al.: Amended Indictment, ICTR-96-7-­I, Arusha, July 31, 1998, pp. 2–11 and para. 5.2. Some minor details w ­ ere altered in the final operative indictment, but the main accusation was set in stone. UNICTR, The Prosecutor versus Théoneste Bagosora: Amended Indictment, ICTR-96-7-­I, Kigali, August 12, 1999. 95. ​Prosecutor Chile Eboe-­Osuji introduced Des Forges as an expert witness “[T]o give the story of the history of Rwanda . . . ​so that when she is done . . . ​we w ­ ill call other factual witnesses and the evidence of such factual witnesses w ­ ill be put now in the context of what you would have heard from Ms. Des Forges, both in terms of the history of the country and in terms of the . . . ​context in which each of ­these witnesses ­will be coming to testify.” UNICTR, TC III, The Prosecutor of the Tribunal against Théoneste Bagosora, Anatole Nsengiyumva, Gratien Kabaligi & Aloys Ntabakuze: Trial Transcript, Case No. ICTR-98-41-­T, Arusha, September 3, 2002, n.p; ibid., September 2, 2002, n.p. 96. ​Tribunal expert witness, André Guichaoua, quoted in Thierry Cruvellier, “ICTR: Rwandan Genocide—­No Masterplan,” Radio Netherlands Worldwide International Justice, December 19, 2011. 97. ​UNICTR, TC I, The Prosecutor v. Théoneste Bagosora, Gratien Kabiligi, Aloys Ntabakuze & Anatole Nsengiyumva: Judgement and Sentence, Case No. ICTR-98-41-­T, Arusha, December 18, 2008, §12. 98. ​Ibid., §5. 99. ​For a full analy­sis of the Taylor and Bagosora cases, see Thijs Bouwknegt, Cross-­Examining the Past: Transitional Justice, Mass Atrocity ­Trials and History in Africa, PhD dissertation (University of Amsterdam, October 2017), 217–250. 100. ​UNICTR, Prosecutor versus Théoneste Bagosora et al.: Judgement & Sentence, December 18, 2008, para. 1221 101. ​Tim Kelsall, Culture ­under Cross-­Examination: International Justice and the Special Court for Sierra Leone (Cambridge: Cambridge University Press, 2009), 17. 102. ​Martin Witteveen, “Dealing with Old Evidence in Core International Crimes Cases: The Dutch Experience as a Case Study,” in Old Evidence and Core International Crimes, ed. Morten Bergsmo and Cheah Wui Ling (Beijing: Torkel Opsahl Academic Epublisher, 2012), 65–108, 71–72. 103. ​For an analy­sis, see Robert Cryer, “A Long Way from Home: Witnesses before International Criminal Tribunals,” International Commentary on Evidence 4, no. 1 (2006): art. 8. 104. ​Nancy Combs, “Grave Crimes and Weak Evidence: A Fact-­Finding Evolution in International Criminal Law,” Harvard International Law Journal 58, no.  1 (Winter 2017): 47–125; Nancy Amoury Combs, Fact-­Finding without Facts: The Uncertain Evidentiary Foundations of International Criminal Convictions (Cambridge: Cambridge University Press, 2010). 105. ​See also Thijs B. Bouwknegt, “How Did the DRC Become the ICC’s Pandora’s Box?,” African Arguments, March  5, 2014, http://­africanarguments​.­org​/­2014​/­03​/­05​/­how​-­did​-­the​-­drc​ -­become​-­the​-­iccs​-­pandoras​-­box​-­by​-­thijs​-b­ ​-­bouwknegt. 106. ​Caroline Buisman, “Delegating Investigations: Lessons to Be Learned from the Lubanga Judgement,” Northwestern Journal of International ­Human Rights 11, no. 3 (2013): 30–82. 107. ​ICC, OTP, Situation on the Republic of K ­ enya. In the Case of The Prosecutor v. Uhuru Muigai Kenyatta: Notice of Withdrawal of the Charges against Uhuru Muigai Kenyatta, Case No. ICC-01/09-02/11, The Hague, December 5, 2014; ICC, TC V(a), Situation in the Republic of ­Kenya. The Prosecutor v. William Samoei Ruto and Joshua Arap Sang: Public Redacted Version of

144

thi js b. bo u w knegt

Decision on Defence Applications for Judgments of Acquittal, Case no. ICC-01/09-01/11, The Hague, April 5, 2016. 108. ​Cruvellier, Master of Confessions, 80. 109. ​Beate Arnestad, dir., Telling Truths in Arusha, documentary film (Norway: SF Norge A/S, 2010). 110. ​UNICTY, “Address by Chief Prosecutor Carla del Ponte at the Conference on ‘Establishing the Truth about War Crimes and Conflicts,’ Zagreb, Croatia, February  8–9, 2007,” Press Release, February 15, 2007. 111. ​UNSC, Report on the Completion of the Mandate of the International Criminal Tribunal for Rwanda as at 15 November 2015, S/2015/884, November 17, 2015, para. 55; UNICTR, Appeals Chamber, The Prosecutor versus Eduard Karemera, Mathieu Ngirumpatse, Joseph Nzirorera: Decision on Prosecutor’s Interlocutory Appeal Decision on Judicial Notice, Case No. ICTR-98-44-­ AR73(C), The Hague, June 16, 2006. 112. ​Erin Jessee, “Forensic Investigations,” in Encyclopaedia of Transitional Justice, ed. Lavinia Stan and Nadya Nedelsky (Cambridge: Cambridge University Press, 2013), 1:27–32; Caroline Buisman, “Ascertainment of the Truth in International Criminal Justice,” PhD dissertation (Brunel University, 2012), 199–200. 113. ​For instance, a total of approximately 11,100 witnesses, including multiple and video-­linked testimonies and support persons, have testified and assisted the (7,700 witnesses) and ICTR (witnesses (3,400). United Nations Mechanism for International Criminal Tribunals (UNMICT), “About the MICT,” http://­www​.­unmict​.­org​/­en​/­about​/­witnesses (accessed May 9, 2016). 114. ​See the Judicial Rec­ords and Archives Database ( JRAD), which provides access to all MICT public judicial rec­ords, as well as to the public judicial archive rec­ords of the ICTR. The JRAD includes filings from parties and non-­parties to t­ rials; exhibits tendered in court; transcripts and audiovisual recordings of court hearings. Available at: http://­jrad​.­unmict​.­org. 115. ​UNICTY, TC II, “Annexe 2—­R appel de la procédure,” Le Procureur c/ Vojislav Sešelj: Jugement, IT-03-67-­T, The Hague, March 31, 2016, 5–6. 116. ​Donia, Radovan Karadžić; “Interview: Robert Donia,” New Books in Genocide Studies 6 (February 2015), podcast at http://­newbooksnetwork​.­com​/­robert​-­j​-­donia​-r­ adovan​-­karadzic​ -­architect​-­of​-­the​-­bosnian​-­genocide​-­cambridge​-­up​-­2014​-­3. 117. ​Hannah Arendt, “A Reporter at Large: Eichmann in Jerusalem,” published in The New Yorker five parts: February 16, 1963, 40–113, February 23, 1963, 40–111, March 2, 1963, 40–91, March 9, 1963, 48–131, and March 16, 1963, 58–134. 118. ​ICC, TC I, Prosecutor v. Laurent Gbagbo and Charles Blé Goudé: Trial Transcript, January 28, 2016. 119. ​As observed by the author on September 30, 2009. 120. ​ICC, TC I, Situation in the Demo­cratic Republic of the Congo: The Prosecutor vs. Thomas Lubanga Dyilo: Trial Transcript, Case No. ICC-01/04-01/06, The Hague, January 28, 2009. 121. ​Since 1998 SENSE has been continuously producing daily news reports and a weekly tele­v i­sion program entitled The Tribunal for TV networks in Bosnia and Herzegovina, Croatia, Serbia, and Montenegro. See http://­www​.­sense​-­agency​.­com​/­home​/­home​.­4​.­html​?­verz​ =­2. Thomas Verfuss, “Trying Poor Countries’ Crimes in a Rich City: Prob­lems of the Press from the Former Yugo­slavia,” Journal of International Criminal Justice 2, no. 2 (2004): 509–515. 122. ​For example, ­there is an Association of Journalists at the ICC (AJICC). Amongst several other media, ­there is a specialized website, which monitors at the proceedings at the ICC, in the Guatemalan genocide trial against Efrain Rioss Montt as well as the Khmer Rouge ­trials at the ECCC: Open Society Justice Initiative, International Justice Monitor, http://­www​ .­ijmonitor​.­org.



The Trial Rec­ord as Historical Source 145

123. ​Cruvellier has published broadly on international criminal ­trials in the magazines Diplo-

matie Judiciaire and the International Justice Tribune (IJT), whose archives are available at https://­www​.­justicetribune​.­com. Cruvellier’s reporting in Arusha formed the basis for his book on the ICTR: Thierry Cruvellier, Court of Remorse: Inside the International Criminal Tribunal for Rwanda (Madison: University of Wisconsin Press, 2010). 124. ​For published books by journalists, see Stéphanie Maupas, Le Joker des puissants: Le g­ rand roman de Cour pénale internationale (Paris: Don Quichote Editions, 2016); Tjitske Lingsma, All Rise: The High Ambitions of the International Criminal Court and the Harsh Real­ity (Amsterdam: Ipso Facto, 2017). 125. ​Richard A. Wilson, Robert J. Donia, and Saskia Baas, “Safeguarding The Hague Tribunal’s Unique War Archives,” Balkan Transitional Justice, March 30, 2016, http://­www​.­balkaninsight​ .­com​/­en​/­article​/­safeguarding​-­the​-­yugoslavia​-­tribunal​-­s​-­unique​-­war​-­archives​-­03​-­29​-­2016. 126. ​El Hadj Malick Sow, Oral Statement (Leidschendam, April 26, 2012) on file with author. Also see an interview with Sow: “Justice Sow: Charles Taylor Should Have Walked F ­ ree,” New African, no. 523 (December 2012): 46–53. 127. ​William Schabas, “What Happened to Judge Sow?,” PhD Studies in ­Human Rights, May 25, 2012, http://­humanrightsdoctorate​.­blogspot​.­nl​/­2012​/­05​/­what​-­happened​-­to​-­judge​-­sow​.­html. 128. ​At the ICC, for instance, this is confidential, including the Operations Manual. Author’s email exchange with Michel de Smedt, the ICC’s Head of Investigations, May 15, 2015. 129. ​For instance, details regarding investigations tran­spired through the testimony provided by former investigators: ICC, TC I, Situation: Demo­cratic Republic of the Congo. In the Case of The Prosecutor v. Thomas Lubanga Dyilo. Rule 68 Deposition: Trial Transcript, Case No. ICC-01/0401/06, November 16, 2010; ICC, TC I, Situation in the Demo­cratic Republic of the Congo: The Prosecutor vs. Thomas Lubanga Dyilo: Transcript Rule 86 Deposition, Case No. ICC-01/0401/06, The Hague, November  17, 2010; ICC, TC I, Situation in the Demo­cratic Republic of Congo: Situation in the Demo­cratic Republic of the Congo. In the Case of The Prosecutor v. Germain Katanga and Mathieu Ngudjolo Chui: Trial Transcript, Case No. ICC-01/04-01/07, The Hague, November 25, 2009. See also Buisman, “Ascertainment.” 130. ​Bouwknegt, “How Did the DRC Become the ICC’s Pandora’s Box?” 131. ​ICC, TC I, Situation: Demo­cratic Republic of the Congo. In the Case of The Prosecutor v. Thomas Lubanga Dyilo: Judgment Pursuant to Article 74 of the Statute, The Hague, March 16, 2012; Caroline Buisman, “Delegating Investigations: Lessons to Be Learned from the Lubanga Judgement,” Northwestern Journal of International ­Human Rights 11, no. 3 (2013): 30–82. 132. ​ICC, TC V (A), Situation in the Republic of ­Kenya in the Case of The Prosecutor v. William Samoei Ruto and Joshua Arap Sang: Public Redacted Version of “Ruto defence request to appoint an amicus prosecutor,” ICC-01/09-01/11, The Hague, May 2, 2016. 133. ​Michael Jackson, The Politics of Storytelling: Vio­lence, Transgression, and Intersubjectivity (Copenhagen: Museum Tusculanum Press, 2002), 15.

7 • NARR ATING (IN)JUSTICE IN THE FORM OF A REPAR ATION CL AIM Bottom-­Up Reflections on a Postcolonial Setting—­The Rawagede Case NICOLE L. I M M LER

Ever since a landmark decision in September 2011 in the Civil Court in The Hague, we know their ­faces and have heard their voices: Pak Saih and Ibu Sariman, Ibu Layem, and Ibu Cawi; a survivor and three el­derly ­widows, whose husbands ­were executed by the Dutch military in a small village in Java called Rawagede (­today Balongsari) in the year 1947.1 At that time the Dutch ­were fighting a decolonization war, trying to keep “their” colony, while Indonesia—­ after the Japa­nese occupation during World War II had ended Dutch rule—­had declared its in­de­pen­dence and an end to 350 years of colonial power on August 17, 1945. Only international pressure and the realization that the war could not be won fi­nally pushed the Dutch to formally acknowledge Indonesian sovereignty in December 1949. The ­widows and one survivor (and their respective representatives) sued the Dutch government to acknowledge colonial faults, and won. In the Dutch media, we see individuals’ ­faces, but b­ ehind them are many other ­people, who (feel they) have the same history. They represent ­those who never spoke with a journalist or testified in The Hague. Insofar as we do hear from a few, who, as new claimants and their subsequent court cases attest, tell the story of a larger collective, the question arises: which collective? What do we know about t­ hese ­widows and their ideas of justice? What do they expect from the Netherlands seventy years ­after the decolonization war came to an end? And what about what happened ­after the historical injustice was acknowledged (by a court in several decisions), an 149

150

nicole l . i m mler

apology was made (by an ambassador and two ministers), and development aid (for the village) and compensation (to some individuals) was given? As claimed by the h­ uman rights movement, do such mea­sures provide victims with recognition, and more—­namely, emancipation of individuals and transformation of socie­ties, in the sense of overcoming old identity-­positions inherited from the past? Transitional justice has been called the hegemonic, or dominant discourse of our time for redressing historical injustices, but as critics say, this statement is based more on faith than facts.2 While the lit­er­a­ture and practice are characterized by strong beliefs in transitional justice’s toolbox of reparative and restorative mea­sures, ­there is much less empirical knowledge about the impact of ­these mea­sures on the individual and social levels.3 Most of the attention is devoted to bringing ­those mea­sures into place, not to monitoring their effects.4 This chapter aims to do the latter. Based upon field work, it critically assesses the effects of such reparative mea­sures in the “Rawagede case” in order to draw lessons for similar cases of historical injustice and gain a better understanding of how justice is experienced from bottom-up. In par­tic­u­lar, it explores the challenges of trying court cases on behalf of a group of individuals in a postcolonial setting and asks ­whether the individually-­based l­ egal approach is adequate for colonial crimes. While in the beginning transitional justice was dominated by a ­legal approach focused on the perpetrators (as in the Nuremberg t­ rials), t­ oday, especially since the Truth and Reconciliation Commission in South Africa, one listens more to the victims, not only ­because of a belief in the healing effect of truth-­telling, but even more, ­because anthropologists who have criticized the top-­down normativity of the field have demanded more attention to local contexts. Indicative of this trend of moving from retributive to restorative justice and reaching beyond “the legalist paradigm”5 is the research on transitional justice from a historical perspective presented in this volume. Carol Gluck has recognized this shift in noting that “our vocabulary has changed,”6 while Stephan Parmentier has argued that transitional justice should start in the villages rather than in the commissions.7 This brings to mind the famous phrase in Eleanor Roo­se­velt’s speech at the United Nations in 1958: “Where, ­after all, do universal rights begin? In small places, close to home. . . . ​­Unless t­ hese rights have meaning t­ here, they have l­ittle meaning anywhere.”8 While h­ uman rights are understood to be a ­matter of international law and national obligation, they should be more a m ­ atter of local implementation. This case study illustrate this point.

compensation as a dialogue—­but between whom? In the transitional justice debate, monetary compensation is seen as the most effective transitional mea­sure ­because of its “potential direct impact on victims,” including restoring their dignity—at least when connected to other mea­sures



Narrating (In)Justice—­The Rawagede Case 151

such as apologies.9 While in the l­egal and public discourse compensation is often associated with closure and the end of a successful recognition pro­cess, historians have highlighted a dif­fer­ent interpretation: namely, the importance of compensation in the sense of entering into a relationship—­“the law can be a beginning for reconnection”10—­and also as a part of a po­liti­cal negotiation pro­cess.11 Th ­ ese scholars tend to emphasize that crucial to such procedures is negotiating history between formerly opposed parties: “The discourse of restitution encourages governments to admit that their policies w ­ ere unjust and discriminatory and to negotiate with their victims over morally right and po­liti­cally feasible solutions.”12 For Elazar Barkan, historical identity is a negotiated identity, resting on the notion of reciprocity, which involves the recognition and consequent transformation of perpetrators and victims through the pro­cess of negotiating and struggling with historical injustice. He aims for a historical dialogue,13 inviting all parties to express their views on the violent past, aiming not at reconciliation but at rapprochement. According to Barkan, reparation is not just an instrument allowing individuals and communities to transform their (historical) positions by being recognized, but also an instrument dealing with pluralism, acknowledging diversity and par­tic­ul­ar identities. This means “domination and power disparities are not eliminated but are supplemented by negotiation and agreement among unequal parties.”14 What does this idea of dialogue mean in a postcolonial setting, and in par­tic­ u­lar for the Dutch-­Indonesian case? In this chapter, it ­will be argued that studying compensation claims not as an instrument of dialogue between a former colonizer and the colonized (the dominant perspective), but as a dialogue within local communities ­will contribute to a better understanding of whose justice and what kind of justice we are talking about. Elsewhere I have shown the way in which the Rawagede case is a Dutch debate about postcolonial Netherlands.15 This chapter shows the experiences from the Indonesian side: the consequences of ­those pro­cesses for local ­people ­there, and the meaning the victims and their ­family members attribute to them. Exploring some of the main narratives circulating among ­family, community, village, and neighbors, and on the (inter)national level, as well, w ­ ill show to what extent compensation is a rewarding experience when applied in a non-­Western, postcolonial setting. This bottom-up data w ­ ill force us to confront some of “our” (theoretical) assumptions of justice, scrutinize the nature of h­ uman rights discourse, and assess the promises of transitional justice activism from a critical historical and postcolonial stance.

prehistory: the court case In September of 2011 the Civil Court obliged the Dutch government to apologize and to pay compensation (20,000 euros) to nine w ­ idows and one survivor of the 1947 massacre in the village of Rawagede (­today Balongsari) in West Java.

152

nicole l . i m mler

At that time soldiers from the Royal Netherlands East Indies Army (KNIL) who ­were searching for re­sis­tance fighters executed a majority of the male population aged fifteen to sixty, many of whom ­were civilians. The numbers of victims in Rawagede vary: while Dutch official documents mention 120 executions, Indonesian sources cite 431. But Rawagede is only one of many of places where such vio­lence occurred—­albeit up u­ ntil now it has been the best documented one. Altogether it is estimated that about 100,000 Indonesians (civilians and soldiers) and about 6,000 KNIL soldiers died in the four years of diplomatic dispute and armed conflict (1945–1949); ­these numbers include the victims of Indonesian groups/parties caught in between.16 The violent excesses by the Dutch military in the former Dutch East Indies ­were called “deliberate and ruthless” in a U.N. report from 1948. Th ­ ese impressions ­were officially confirmed by the Dutch government only in 1969, ­after a veteran’s testimony was aired on national tele­vi­sion. The so-­called Excessennota listed seventy-­six “excesses,” while recently revealed documents show many more, as well as the structural nature of the vio­lence.17 In 1995, when the first TV documentary about “mass executions on Java” was shown,18 a ­legal case was still considered “unfeasible.”19 In 2007, however, when the documentary The Massacre of Rawagede was aired,20 the Committee for Dutch Debts of Honor (KUKB) started to prepare a l­egal claim on behalf of the Indonesian civilian victims with the help of h­ uman rights l­awyer Liesbeth Zegveld.21 Although the Dutch government kept insisting that consideration of the case was barred, in 2009 it allocated 850,000 euros to the village to improve the infrastructure. It was not explic­itly called reparation money, and a commemoration plaque at the planned school building—­a symbolic gesture demanded by the Dutch ­widows’ representatives—­ was rejected. This neglect, said ­lawyer Liesbeth Zegveld, motivated them to precede with their l­egal claims, which aimed for individual recognition: “We d­ on’t agree with this kind of satisfaction.”22 When survivor Saih, one of the w ­ idows, and another ­widow’s son gave their testimonies in front of the state’s l­ awyer in the courtroom in The Hague in 2009, the moment felt historical.23 When Dutch national TV broadcasted that they had won their claim, it seemed as though justice was done, the more so when Dutch ambassador Tjeerd de Zwaan officially apologized for the first time and shook hands at the local commemoration day (December 9, 2011) at the victims’ monument in Rawagede. The payment of 20,000 Euros to each of the plaintiffs followed. However, soon the front page of the Volkskrant was reporting: “Rawagede ­w idows plundered.”24 The village had taken part of the w ­ idows’ money, and they and their families ­were heavi­ly threatened. In the Netherlands, the ­widows’ representative, the Dutch-­Indonesian activist Jeffry Pondaag from the KUKB and ­lawyer Liesbeth Zegveld, who had both fought this case for several years, saw this action as a violation of the ­widows’ personal rights and a form



Narrating (In)Justice—­The Rawagede Case 153

of corruption. They filed a complaint at the Indonesian ­Human Rights Commission, which, however, remained unanswered. When Pondaag visited the village to insist that the money was meant for the ­widows, he needed police protection. Although scared a­ fter having had shirt torn by some of the village youth, he downplayed the incident and expressed understanding in view of the poor living conditions, the lack of education, and the emotionally laden subject. “They are descendants, mostly c­ hildren, their families ­were also executed t­here. So basically their suffering is neglected.”25 ­These unintended consequences made clear that we know l­ittle about the ­widows’ experiences of the recognition procedure and the impact it had on their lives, their ­family’s lives, and the village community. Consequently, in October of 2015, I went to Rawagede with the documentary photographer Suzanne Liem to examine what had happened.26 It was a pressing ­matter, as the Rawagede case had begun to serve as a quasi-­model for other Indonesian w ­ idows, whose husbands ­were victims of similar summary military executions, such as t­hose in Sulawesi in 1946–1947.27 Moreover, ­children of executed Indonesians who had claimed equal treatment had won their claim in The Hague in March 2015.28 “It is about the same incident, same place, same pro­cess,” said Shafiah Paturusi (born 1933) and Andi Monji (born 1938) in their testimony in regard to executions in the villages of Suppa and Bulukumba in South Sulawesi, where about 460 ­people died and ­houses w ­ ere burned down.29 As many more claims are in the making,30 the question of what we can learn when listening to the experiences of some families in Rawagede remains urgent.

fieldwork in rawagede: was the compensation a blessing? Close-­Up: A ­Family Story First we went to Ibu Tijeng (born 1927). Five years before, she had dreamed of rebuilding her own ­house if she ­were to receive some money: “I harbor no feelings of revenge against the Dutch who killed my husband. But I hope the Dutch could help me to rebuild my ­house. As I had no money to support it, we demolished it. The stones are now piled up in the garden of my grand­daughter.”31 When we arrived, we could see a half-­finished ­house in the formerly empty backyard ­behind the ­house. Tijeng, who was in her eighties, sick, and nearly blind, lay on her bed in her grand­daughter’s living room, with her d­ aughter Karmas (born 1949) sitting beside her. She said, “When my f­ather was shot, I was still in a crib. I ­don’t remember my ­father’s face.”32 Karmas’s son Luus (b. 1971) heard about what had happened when he was about seven years old: “Nimong, my grand­father, was only a farmer. It’s normal if he had been shot as a soldier. . . . ​ ­There ­were also soldiers, you know. While t­ hose farmers w ­ ere civilians. He was

154

nicole l . i m mler

about to go to the rice field when he was shot.”33 Grand­mother Tijeng still remembers that night: “It had rained heavi­ly the night before. It was flooded where my husband was shot; a lot of bodies lay t­ here. . . . ​In this village, t­ here was nobody left. They ran away. It was so quiet. ­Women left, scared that it would be their turn a­ fter all men had been killed. . . . ​I ran away to Panyingkiran for one month.”34 Before she left, she buried her husband in her garden: “I wrapped him in cloths, dug a hole in my backyard and, crying incessantly, buried him between the mango and coconut trees.”35 Much l­ater he was transferred to the village’s monument. The d­ aughter remembers this: “I was prob­ably forty years old when we removed the bones. We w ­ ere told to collect the bones and move them to the monument.” Her m ­ other was pleased with the reburial: her husband was then “with his friends ­there together, not ­here in the garden by himself,” and that meant that he “died as a martyr.” Still, she never learned why he was shot: “They came and t­ here was chaos. I ­didn’t know why they shot him. The Dutch took his life and that’s it. . . . ​It was a terrifying experience. I ask for peace. That’s all.”36 According to her grand­son Luus, she had a certain way of talking about the past. “She always says that if my grand­father had not been shot by the Dutch, my life would be dif­fer­ent. . . . ​‘You would have a lot of w ­ ater buffaloes, a vast field,’ she said.” He distances himself from this view. “That is the story from a long time ago, it’s not good to keep thinking about it. Let’s open a new page.” Although Luus has known the story of his grand­father since he was a child, it has been only since the establishment of the Rawagede Foundation in 1995 that he began to situate it in the larger picture of the “freedom-­fighting history” of which Rawagede was part. Luus described how the Foundation’s head, Sukarman, gathered all the victims’ w ­ idows (around twenty w ­ ere still alive) and went to Jakarta with them, to the Dutch embassy and elsewhere, “four or five times in total” to tell their story. “They took Tijeng to meetings. She was still in good condition at that time. They introduced her to p­ eople. Sometimes t­ here ­were guests from wherever, who wanted to see her.” Karmas also described how her ­mother “was taken h­ ere and t­here,” but could not remember whom they met, while Tijeng herself mentioned the embassy only for the “good food” and to say how “tiring” the trip was. Nevertheless, telling the story had a clear purpose: to create support for herself and the other ­widows. Her grand­daughter felt that Tijeng worked hard for her recognition, and that is why she objected to the money having been divided up and ­going to the village: “I was disappointed. It should be Uuk’s [Tijeng’s] right only. In the end, the compensation money we fi­nally received was too ­little ­after so many years of fighting for the money. They dragged Uuk to go to Jakarta, or drag her to attend the meetings; sometimes Uuk was not in good condition. Fi­nally, when the money arrived, it had to be shared. It w ­ asn’t worth it.”37 Every­one we interviewed referred to having had to share the money as a disturbing experience. They spoke less of what they had achieved with the money



Narrating (In)Justice—­The Rawagede Case 155

they received than that they immediately had lost half of it. “The money d­ idn’t even stay overnight at my ­house, I ­didn’t even have time to put it in my cupboard as p­ eople ­were ­there waiting for the money already.” Luus, who works in road construction, felt powerless when confronted by village officials, who “phoned up continuously even before the day we went to the bank,” saying, “We like to take half of it as we want the money to be shared.” As he was scared, he did not accompany his grand­mother to the bank, but the husband of his s­ ister did. Rawagede is a place with “no ATM,” where many ­people have no bank accounts, which is why an individual account was opened for each ­widow for the purpose of the money transfer. Karmas went to the bank in the nearby town of Karawang with her m ­ other and son-­in-­law. “We took the money, 100 million rupiah [roughly 6,000 Euros], the other 100 million we took l­ater, bit by bit, not the w ­ hole sum. When we came back, p­ eople from the village already waited for us ­here. ­There was a civil servant from the village. They asked Tijeng to give her thumbprint, but she refused. She said: ‘No, I ­don’t want to sign it. This money belongs to me.’ She was told to sign it.” The ­daughter further recalled: “The 100 million was given to the ­people who waited ­here. It was said that they would divide the money in the Foundation. We w ­ ere told that if we w ­ ouldn’t share the money, they would come h­ ere. We ­were scared. We d­ idn’t want to give the money actually. The money belongs to us, but many p­ eople wanted it, too. Of course, we ­were scared. We gave the money cash on that bamboo tray to three ­people, who came ­here.”38 ­These three w ­ ere the village head and two civil servants, accompanied by “the village police.” One grand­child added: “First my ­father was angry at them and wanted to keep the money. But at the end he surrendered to the mass.” While Luus described it as an ad-­hoc action the f­ amily was not prepared for, grand­mother Tijeng remembered a meeting about this issue at the Foundation when she still had her eyesight. “For the meetings, someone took and dropped me home with a motorbike. . . . ​­There w ­ ere a lot of ­people, the room was full. I ­didn’t understand what they said.” And her d­ aughter added cynically: “She was like a puppet doll, sitting quietly. Usually it was Pak Dewan [Sukarman, representing the Foundation] who talked.” Then she asked with a laugh: “How could she speak publicly? She ­can’t speak Bahasa,” the local dialect. Although in this meeting the sharing was discussed, the w ­ idow’s ­will was not respected, the grand­ son explains: “In the beginning Ibu Tijeng proposed 50 million. They refused, as what they wanted was 100 million. At the end it was coercion, it w ­ asn’t a discussion to reach a consensus. . . . ​­There’s no physical fight, but they intimidated us by saying that ­there ­will be a mass mobilization.” The ­family discussion made clear that it was not the sharing per se that was the prob­lem. As Luus said, “I know that my grand­father was not the only one killed over ­there, but o­ thers as well. I understand the feeling . . . ​so that’s why I would have given something for them as well. But I d­ idn’t like the way they forced us to do so without any discussion.”39

156

nicole l . i m mler

The ­family was convinced that the new ­house would have been completed if t­ here had not been the pressure to give half of the money away. But ­there ­were also more urgent needs, which Luus listed: “We took it for many t­ hings, mostly to pay off our debts. Thirty million was used for the h­ ouse. ­There ­were ­people who borrowed money, but ­didn’t pay it back. And much money was used only to buy food. Since we d­ on’t have our own rice field, we buy it.” Tijeng, who worked as coolie in the fields when she was young, added: “The money was used for the kids, for food. I loved to eat anything. I like papaya. If I wanted to eat chicken, roasted chicken I would buy it.” The grand­daughter, who like the other grandchildren had received 500,000 rupiah (about 35 euro), confirmed giggling: “I used it to buy snacks; meatballs for my c­ hildren.” This, however, was a luxury of the past; once the money was gone, they borrowed again. Still, for Ibu Tijeng the judges’ decision remained an impor­tant gesture: “I am happy with the decision of the judges, even though I had to share the money with ­others. I could not do anything about it, but I was happy I received the money. . . . ​I’ve forgiven the Dutch for what has happened, I ­don’t have grudges. It’s safe and t­ here’s peace now. What happened, it’s in the past now. We’ve forgiven each other.”40 While she claimed to be “happy,” her grand­daughter quoted her as saying, “I am the one who is tired, but o­ thers benefited from it.” Her grand­ daughter described her immediate ­family as “traumatized by what happened,” while relatives who received a share from the village (5.2 million Rupiah, about 350 Euros) w ­ ere satisfied: “My u­ ncle’s son, he was very happy. He d­ idn’t demand more.”41 In her view, her grand­mother should have been the main beneficiary. Luus, on the other hand, was convinced that the grand­mother did this in the first place for the ­family, to compensate the loss of the grand­father with whom life would have been much easier for the ­children and grandchildren. The Village: Circulating Narratives of (In)Justice ­After Tijeng’s f­ amily, we visited the other w ­ idows, as well as the f­ amily members of the deceased w ­ idows, and heard similar stories about feelings of recognition, but also about personal threats, from both the village heads and ­family members, all demanding their share. Nonetheless, we saw that some of ­those ­widows had acquired homes. Ibu Wanti (b. 1925) owned the ­house she formerly rented, and Ibu Cawi (1928–2014), who had always wanted her “own door,” got a small ­house next to her son’s place, which she enjoyed for two years before she passed away. Saih (1923–2011) had hoped that “the Dutch government wants to help the villa­ gers to develop Balongsari and ­will compensate any descendant,”42 but he died before the conclusion of the court case. His f­ amily invested in a shop, where the grand­daughter cleaned and refilled ­water ­bottles—­a signifier of modern Indonesia. ­These ­were substantial changes. Invested money does not dissolve, but in most families, basic needs still occupied their minds and our talks.



Narrating (In)Justice—­The Rawagede Case 157

figure 7.1. Ibu Wanti, map of the deaths in Rawagede, 2015. (Photo by Nicole L. Immler.)

Beyond ­these concrete results and what the ­widows or their families did or did not gain materially, what meaning did they attribute to the money and to the other mea­sures that ­were part of this recognition pro­cess? To explore this, the following section summarizes the main narratives of a perceived (in)justice circulating in the village.

158

nicole l . i m mler

The Monument: “My Grand­mother Is My Hero” While having lunch at the food stalls in front of the monument, one hears many stories about “the money”: who profited, for whom it arrived too late, and among whom it caused f­amily trou­bles. The monument, overlooking the cemetery, displays the history of the decolonization war and its victims. While life-­size sculptures reenact the murder of villa­gers by KNIL soldiers in 1947, the walls also memorialize ­those who ­were killed by “the mob,” partisan groups and local gangs that extorted the village during the same period of revolution ­after the Japa­nese surrender in August 1945 and the emergence of a power vacuum. ­Here one reads that the history was complex and involved vio­lence from several parties. The monument is the only place in the village where one can sit down and eat. Th ­ ose who have small food stalls in front of it make a living from the sociality of this place, while unemployed youth hang around to take care of a few parked cars. Three of ­these young men have grand­fathers buried in the cemetery. ­Family members look a­ fter the graves. The area, one explained, had been a swamp, but ­after it was cleaned up, it became a place to meet and earn some money. Sukarman regularly comes along ­eager to share historical details. It is impossible to avoid him when visiting the village. As local historian and head of the Rawagede Foundation, he speaks for the past and acts as mediator between the ­widows and their visitors: press, politicians, and researchers alike. For him, his own ­family history was the beginning. Sukarman grew up in the village, but heard about what had happened in 1947 only as an adult. It was then that his m ­ other, Ibu Cawi (b. 1928), told him that “her first husband, Bitol, was shot by the Dutch and buried at the monument. Just six months ­after they got married, he was killed. Sukardi [her second husband] was a survivor from the slaughter in the river. . . . ​ He was still single at that time and married to my m ­ other in 1948. I was born in 1949.”That he did not know this history while passing by the cemetery ­every day has haunted him. Breaking the silence has been his passion ever since. Sukarman went on: “The idea for the Rawagede Foundation already crystalized when I was village head in 1977–1978. I wrote a book about Rawagede in 1990, when I was a senator in the local parliament and got into a fight with the district head ­because he wanted to move the grave to the heroes’ cemetery in Pancawati, but I wanted it h­ ere.”43 The attention the book gained led the military to support the monument and the establishment of the Rawagede Foundation in 1995. As its president, Sukarman organizes the annual commemoration of the massacre on December  9. Since then much has happened. “Recently we have also had visitors from outside the district,” Sukarman says, highlighting this evidence of local fame. When a bus arrives, he enthusiastically welcomes pupils who come to visit the monument. He organizes many educational programs. Listening to one, in the presence of military, one realizes that not all memory-­teaching is a lesson in history. When civil



Narrating (In)Justice—­The Rawagede Case 159

victims are turned into heroes, we see regionalism and nationalism rewriting history, nurturing the g­ rand national narrative of Merdeka (or freedom), meaning: “We defeated the Dutch as one nation.” This slogan—­used during the strug­gle for in­de­pen­dence to unite the dif­fer­ent groups in Indonesia in one spirit to gain freedom from the Dutch colonial government—is still alive in public life and leaves an imprint on the youth. One grand­child is proud of her heritage: “My grand­mother is famous. You can find her on YouTube.” Th ­ ere you can find also the pop song “My Grand­ mother, My Hero,” written by the grand­child of a Rawagede ­widow, who had remarried in Jakarta.44 Only ­after seeing the court case of Rawagede on TV did she return to visit the grave of her first husband. Both examples illustrate the national mind-­set: celebrating the in­de­pen­dence and its heroes, rather than commemorating its civil victims. “We Have Forgiven the Dutch” A phrase often used by the ­widows is “We have forgiven the Dutch,” when they are addressed by journalists, the Dutch ambassador, Minister of Foreign Affairs Bert Koenders, or by us. For Tijeng the past is the past (“I forgave the Dutch for what has happened”), and the same is true for her grand­son Luus and ­others. But the Dutch are not forgotten—­not when in­de­pen­dence is nationally celebrated each August and commemorated at the village level in December. However, we encountered no negative feelings about the Dutch; rather, what we found was a kind of nostalgia. Survivor Saih, for one, dreamed about meeting Queen Beatrix and shaking hands,45 while child victim Warjo’s wish to be on a boat in the Amsterdam canals did come true.46 This nostalgia is part of an ambivalent postcolonial relationship. “Money Always Helps” Some attitudes are however clear: “Money always helps.” “Yeah, it’s money, so yes, we can take it, we can profit from it, we can benefit from it.” Tijeng’s grand­ son Luus was aware that it enabled him to start anew in life: “Thank God that I’ve got the money, ­because I could pay my debt of forty million [rupiah]. I sold gas and kerosene. I was bankrupt.”47 Although he profited im­mensely, his feelings have remained ambivalent—­thankful, on the one hand, while but at the same resentful on the other, particularly of what he perceived to be village authorities’ unjust treatment of his ­family. As in other families, money was always welcomed and needed to improve poor living conditions. However, one gets the impression that it may not ­matter where it came from or ­whether the amounts ­were small or large. When another ­widow, Ibu Rasem, was given her compensation, she was able to keep it for herself. This indicates that the grandchildren community was already satisfied with the share they had received; thus it was not the amount as such that mattered.

160

nicole l . i m mler

One ­thing applied to all families: the moment money became available, it was immediately spent, mostly on a h­ ouse, to pay off debts, or to buy food and pres­ ents for ­family members, but not all “pres­ents” ­were given out of ­free ­will, but rather for lack of choice. Thereby the material capital was turned into symbolic capital, stabilizing relationships that one might need in the ­future. As Ibu Wanti put it: “If I have no money for the bus t­ oday, I ask my neighbor; maybe I can pay him back tomorrow, a­ fter I have sold some fruits.”48 “We ­Couldn’t Say No” Ibu Wanti (b. 1925) lost her husband when she was in her twenties. She remembered: “I saw bodies along the tramway. I found my husband’s body in Sumur Bor [the ­water well]. He had been lined up. Th ­ ere w ­ ere three lines, each line was twenty ­people. So, ­there ­were sixty ­people [shot]. . . . ​He was shot through his head and back. He used to work on his own rice field that was inherited by his parents. I did not receive anything, so I went to live with my parents again.”49 That same day her ­house and rice barn ­were burned, w ­ hether by the Dutch or fighters or by the in­de­pen­dence movement (“Lukas’ boys”) remains unclear. The effects for her ­were the same: she had to start all over from scratch. ­Today Ibu Wanti lives close to the river, half a day’s travel from Rawagede, in the fishing village Sedari. On the walls, one can still see the remnants of the last flood. She is happy about having visitors and proud to explain that the compensation enabled her to buy the ­house the ­family had rented up till then. But the fact that she lived in a remote area did not help her to keep the money to herself. “When Ibu Wanti, ­after fetching the money, was in her ­family’s ­house in Rawagede, we ­were surrounded by village staff. We ­couldn’t say no,” her ­daughter Sophia explained. “We gave 100 million to the village, then we gave as much as 40 million to my ­mother’s late husband’s ­family, who lives in Rawagede.” Ibu Wanti nodded: “I had to, other­wise they w ­ ouldn’t recognize me as a native of Rawagede, while I am.”50 The “Hereafter”: Financing Funerals While none of the w ­ idows managed to realize her dream of g­ oing to Mekka, some did use their money for religious purposes, giving some to the mosque or to orphans or saving it for the last meal. “The money is impor­tant ­because first, we need food, when ­we’re still alive, and when we die, we need money too, for the funeral,”51 explained victim Warjo (b. 1936). The traditional duty to invite the ­whole ­family—­who often come from afar and stay for a week—to the funeral burdens not just the ­widows but also the young. Thus, although in many cases the money came too late to have made a qualitative difference in the w ­ idows’ lives, it did help the next generation with their funeral duties and thereby benefited them.



Narrating (In)Justice—­The Rawagede Case 161

“The Money Should Be Personal”: An “Incomplete, Imperfect Pro­cess” As a boy, Warjo, whose ­father had been shot, had seen the executed villa­gers piled up in the marketplace: “Then came another group of villa­gers. The soldier told them to stand and lined up at that spot, then shot them . . . ​drrrrttt rrtttt. . . . ​ They shot ­until ­there was a big pile of bodies in the market area. . . . ​I think ­there ­were about three layers of bodies in that pile! That was not including the bodies on the road and also in the field.” ­After he received the money for his ­mother, Ibu Layem, who had died during the court procedure, he fled the village to escape the obligation to share. But then, fearing the consequences, he gave up his re­sis­ tance two weeks l­ater. “What would have happened if I keep insisting? Maybe I win, but my ­family might become their target and be a victim. ­People ­were fighting like in a war. ­There w ­ ere many of them. Even ­those, who ­were not part of the heirs joined, maybe around fifty ­people.”52 For him the compensation was an imperfect justice; he was convinced that “the money should be personal.” ­Because his ­father had been killed, he had not been able to attend a school. “Deep in my heart, I still keep grudges that 100 million w ­ ere taken from me. That made the pro­cess kind of incomplete, imperfect.” While he described himself as a victim of collective pressures, villa­gers described him as the “bad guy,” ­running off with the money, cheating his ­family. “­Free Money” While Warjo believed that the survivors deserved the money individually, villa­ gers referred to the compensation as “­free money” and “a gift,” suggesting that no one had worked for it and that is should therefore be available to every­body. Besides, “participating” is considered legitimate, as a local farmer explained to us: “All ­people who feel involved in a pro­cess, even if they are only watching, want to be paid for having supported this pro­cess.”53 ­Here cultural issues w ­ ere at stake, as Sukarman’s ­daughter Sri told us: “That’s the nature of Indonesian p­ eople: when money comes, they claim a share, they claim they have rights.”54 What from one side is felt as “robbing” (Luus), from another side is felt as “participating” only. “Do It Differently Next Time” Due to t­ hese trou­bles, several ­people begged us—­identifying us with the Dutch government—to “please do it differently next time”: “­There has to be a discussion prior to it.”55 It is the lack of transparency in the pro­cess that caused the dismay. Equity was an issue, too, as in the case of the ­family of Ibu Taswi, who died during the court procedure, but whose side of the ­family received much more than the executed f­ ather’s side, even though the latter had looked a­ fter his ­widow. We often heard the plea for a dif­fer­ent procedure from villa­gers, but hardly any concrete suggestions ­were offered. Therefore, we went to the village representatives to ask their views on the ­matter.

162

nicole l . i m mler

The 181 In September 2011, when rumors first circulated that money would come, a village meeting was or­ga­nized and the approved applicants w ­ ere invited. Only half of them went, however; the o­ thers feared they would have to agree to share the money. “­Those nine w ­ idows represent to us the families of all victims, and the money they received must be shared,” said Mamat, the village head.56 Thus, part of the money would be collected and divided among other victims’ families, who believed that they had as much right to it as the living ­widows. The fact that the Rawagede Foundation represents all 181 civil victims’ families reinforced this position, as Sukarman stressed: “Collectivity is an unwritten custom, the money has to be divided”—­and divided specifically among 181 victims’ families. Sukarman was proud that the idea was his, but also admitted that it had been compelled by the grandchildren of the victims, who had told him, “If you w ­ ill not share the money, we w ­ ill burn your h­ ouse.” For Sukarman, the m ­ atter was both personal—­his ­mother, Ibu Cawi, was one of the ­widows—­and public, since as the Foundation’s head, he was expected to act in the name of all victims. Sukarman’s preferred model for sharing, however, would have been dif­fer­ent from what actually tran­spired. “It would have been better if the Dutch gave the money to Foundation Rawagede and let us h­ andle the m ­ atter, split it evenly for every­body. ­Because for ­every donation, we always give the same sum to every­one, ­either the ­widows, c­ hildren, or grandchildren . . . ​181 ­people.” In that case, he himself would have received a smaller portion, but as he said, “The most impor­tant ­thing is peace between us, keeping the solidarity between us villa­gers.”57 The 181 are more than names on Sukarman’s list and graves in the cemetery. The part of the village that surrounds the monument and is where most victim families live is known as the “monument block.” ­There the past is not the past, but a lived presence. Would taking a collective approach right from the beginning, along the lines of a the “village-­model,” have been better for all?

a neighbor’s perspective “The Monument Block Mentality” One has to leave the quarter to get a sharper view of the monument block. Across the river, a young man, about thirty years old, works for the Cooperative, an initiative set up by the development organ­ization HIVOS, led by local businessman Suparta (chosen by the village elders), and financed by the Dutch development aid allocated to the village. The young administrator describes how a­ fter long debate the money was invested in four facilities: a permanent market (a hall with shops to rent), a vocational school (SMK), the improvement of the local health clinic, and the Cooperative that provides microcredits to w ­ omen and uses some



Narrating (In)Justice—­The Rawagede Case 163

figure 7.2. ­Children of the victims, Sulawese, 2015. (Photo by Nicole L. Immler.)

of the profit for donations to the mosque, old p­ eople, orphans, and so on, as well as to “the heirs of the tragedy.”58 At the beginning, half of the w ­ idows’ families joined the Cooperative, which counts about 1,600 members, but they ­were excluded ­after not paying back their loans. “­Because they think to join the Coop is part of the grant money scheme, they d­ idn’t want to pay their loans,” said the administrator. “They know that the money came from the Dutch, so no need to pay it.” For him it is not a question of having no money, but of attitude: Why pay back money that you feel belongs to you anyway? “The Rawagede ­people are known for such a mentality . . . ​the monument block. . . . ​Some of them w ­ eren’t heirs, but they followed the attitude of the heirs.” He attributes part of this mentality to the mono­poly of the two foundations (the ­widows’ foundation and the microcredit foundation/Coop, both key players in the village in sharing), led by the “so-­called heirs’ leaders.”59 Our interviewee, himself one of the heirs, had no illusion about the compensation. For him it contributed more to a victim attitude than to ­women’s agency. While Sukarman believed his 181-­model could bring justice, and Suparta’s Coop provided a similar benefit program for the same (victim) community, for our young in­for­mant, the local practice is more of a prob­lem than a solution. Instead of believing in benefit programs, he believed in the international microcredit model, treating all participants equal.

164

nicole l . i m mler

From “Graveyard” to a “Place in Indonesian History” Both the Rawagede Foundation and the Coop can boast achievements—­the former for in its role in creating the monument and memorializing the history, the latter for successfully managing a portion of the Dutch development aid. Villa­gers are proud. The market that was once an hour’s motorbike ­ride away is now close, making a difference in every­body’s life; the same goes for the school that ­will open. Before t­ hese changes, say the villa­gers, “this was a sleepy place, so quiet and dead calm. It was as if . . . ​it was like a common graveyard.” At that time, “Nobody wanted to move h­ ere,” but “when a w ­ oman from h­ ere married a Javanese man, now they stay ­here.”60 The ­widows remember how dark and frightening the monument area used to be. Now the young also value its beauty and reputation: “I’m proud of Rawagede! Beforehand, nobody knew it. The monument was dirty with garbage. Now, it is clean and well-­maintained. . . . ​The Rawagede monument has a place in Indonesian history.”61 This transformation is apparent even in the name of the village: Rawa Gede (Big Swamp) became Balong Sari (Beautiful Pond)62—­a success story. However, t­here are also stories ­people do not (dare to) tell, stories that are hidden, or silenced. They emerge only from incidental remarks, between the lines in the interviews, or in dif­fer­ent places, such as in a neighboring village.

another neighbor’s perspective “The Rawagede Story Is Written from Only One Source” One is not supposed to talk about the Indonesian rebels, who plundered rice barns or the ­uncle, who was killed by villa­gers as an alleged “spy” b­ ecause the Dutch liked to sit in his “clean garden,” explains a h­ uman rights activist from a neighboring village: The prob­lem of the Rawagede story is that it is written from only one source. If you go ­there to get to know what is ­going on, all ­people say, “go to Pa Sukarman,” he is the former village head and was in local parliament, and during the Suharto era he had a close relation with the military group, the Suharto party (Golkar). His story is one-­sided. If you want to ask the ­people, they are afraid, afraid to tell a dif­fer­ent story. They always think they have no importance, they feel they are stupid, they know l­ ittle, [feel] no re­spect for their story. . . . ​I am living h­ ere, I want to know: my ­uncle was killed, but the ­whole ­family is not saying a word. . . . ​The history is ­really dark, we are in the dark.63

­ ose experiences, of villa­gers killing villa­gers, or of plundering and murdering Th rebel groups (such as the Darul Islam ­people,64 or the Indonesian nationalists



Narrating (In)Justice—­The Rawagede Case 165

who killed collaborators) may be a vivid memory in many families, but they are wiped out of the “story of Rawagede.” ­Here doubt is raised about w ­ hether t­ hose stories we have heard are the most meaningful for the w ­ omen. Are they their stories, or just the ones Rawagede is now well known for? Or are ­these the same? By extension, is the “village model” suggested by Sukarman indeed a better solution in terms of balancing individual and collective needs, or does it rather continue a certain style of managing not just the village, but also its history?65

global-­local: translating international transitional justice to the village level While the legalistic discourse surrounding this case has its value in the Netherlands, in that it encourages a debate about its colonial past,66 in Indonesia the legalistic paradigm has caused prob­lems on the village level, as the Rawagede case illustrates. It also highlights the inherent tensions between local customs and international law, norms, or practices when the ­legal acknowledgement of injustice goes hand in hand with a form of individualization that can harm existing social structures. This difficulty is acknowledged in the field of transitional justice, and not only vis-­à-­vis Indonesia. For too long, the international reconciliation discourse overlooked the orientation ­toward collectives or communities. Thus, scholars such as Birgit Bräuchler have argued that both the ­human rights and the reconciliation concepts need to be translated to the local level and to take into account the collective and cultural dimension of a community in order for them to be meaningful to its members.67 Whereas in the West the community is assumed to be healthy when the individual is, it is the other way around in many non-­Western socie­ties.68 Sukarman’s action—­initiating a sharing pro­cess according to his village model—­can be described as a translation of a decision of international law to the local level, and in a way that anticipates social realities. Such a conversion of “universalistic ­human rights into local understandings of social justice” has often resulted in a “limitation on the transformative power of h­ uman rights.”69 This “limitation,” namely bound by local constraints, is indeed felt in Rawagede. In his efforts to mediate between Indonesia and the Netherlands, Sukarman has felt misunderstood by both his own p­ eople and the Dutch actors. H ­ ere it is impor­ tant to note that he was not just translating an international ­legal decision into local practice; he was applying a specific h­ uman rights discourse that differs from that of Eu­rope. That is, whereas in Eu­rope the strug­gle for h­ uman rights means that protecting individuals from state vio­lence and acknowledging individual victimhood are foremost, in former colonies and in the vari­ous decolonization strug­gles in which they have engaged, h­ uman rights means the right of

166

nicole l . i m mler

self-­determination,70 such that the civil and po­liti­cal rights of individuals are subordinated to collective aims.71 Historian Katharine McGregor considers the ­human rights discourse that acknowledges individual suffering to be a welcome tool for reaching beyond national history-­writing in Indonesia. This is characterized by seeing “lives lost as part of a national contribution to the in­de­pen­dence strug­gle and therefore as national sacrifice.”72 She shows that in terms of the global h­ uman rights discourse and the strength of memory activism, the Rawagede case has resonated in the Indonesian po­liti­cal system by triggering a “domestic debate” as to how “to deal with its own cases of historical injustice.” It has also shifted memory discourse in Sulawesi from a nationalist into an individual framework, which acknowledges civilians as victims and not just in­de­pen­dence heroes.73 What does the Rawagede case tells us in this re­spect? Do the compensations provide individual recognition, as well as emancipation of victims and transformation of local discourses?

­ ere the apology and the compensations w meaningful and supportive? At the time of our visit (October 2015) village life in Rawagede seems to be back to normal. “Money is gone, conflicts are gone,” summarized Sukarman’s ­daughter Sri about the troubling episode in her f­ amily’s life. We often heard “peace is back,” a phrase emphasizing the turmoil created by the w ­ hole pro­cess, but also calling attention to how strongly village identity and personal identity are intertwined: one’s own peace seems also to be the peace of the village. Is then the success of the village likewise one’s own success? The village Rawagede has become transformed: it is “famous” and consequently more prosperous. Has this meant that the life for the ­widows and their f­ amily changed for the better? Recognition The w ­ idows said they felt recognized. They have told their stories; the law declared their husbands innocent and acknowledged their suffering in front of their community and the world; the money symbolizes the irrevocable admission that a crime had been committed; with the money they fulfilled some personal needs and desires, although the joy of being able to share with one’s immediate relatives was overshadowed by the imposed responsibility to share with distant ­family or the victim community. Po­liti­cal phi­los­o­pher Nancy Fraser defines recognition as a “reciprocal relation between subjects. . . . ​One becomes an individual subject only by virtue of recognizing, and being recognized by, another subject.”74 Taking this insight, as well as understanding recovery and repair as reconceptualizing the self-­image, acquiring a positive self-­identity, and experiencing a sense of self-­esteem and re­spect,75



Narrating (In)Justice—­The Rawagede Case 167

then the Rawagede case seems like a success story for both the collective and the individual. ­After all it is the relationship between individuals and their po­liti­cal community that is seen as key in reparation theory, which also rests on the public recognition that “the person is not only the subject of his or her own actions, but the object of the actions of o­ thers.”76 The narrative of pro­gress on a collective level means that the ­widows and descendants can situate their personal experience in it and thereby turn their experience into a step in a recovery pro­cess. However, at the same time the w ­ idows’ stories remain subordinated to the heroic nationalistic history directed by the Foundation.77 This impression was confirmed at the commemoration ceremony on December 9, 2011, when the Dutch ambassador offered an “excuse” for the misdeeds committed. The w ­ idows sat at the side, as the subjects of discussion, while the main protagonists ­were men, the local elite, who did the talking—­and in Bahasa, not in Sundanese, the local dialect of the old p­ eople in the village. A visitor called it “a puppet show.”78 While for Sukarman the ceremony, when “this history was acknowledged” by the Dutch, was a historical moment, neither the w ­ idows nor their families saw it that way. Only when asked did one w ­ idow mention that “we w ­ ere invited” and ­there was “­free food.” The younger generation had higher expectations: “An apology is only words, it needs a physical substance.” Words make no difference in village life. For Warjo “cheap talk” is not enough.79 Emancipation-­Transformation Thus, how did the symbolic and the material recognition relate to one another? The compensation, although we have seen some success stories, often made no real difference in ­people’s lives. “Every­body knew that we received lots of money, but ­we’re still poor; d­ on’t have money,” summarized Tijeng’s d­ aughter.80 Her son did not have the resources for his own most talented son to continue school.81 ­Here insights from the field of development aid seem similarly true for the field of transitional justice: financial aid is of l­ ittle use when “dropped at a place” and not linked to training about how to make use of it in an emancipatory way, so that it provides a sense of agency. A local farmer favored structural support: “The poverty is too deep, then it is difficult to see that the money can help to get out. They have no land, they have no health insurance, and the c­ hildren do not study continuously.”82 In a place where ­people are poor and have debts, it is difficult issue for a grant of money to provide a feeling of recognition and have a transformative power. Would a collective pro­cess have been better? Pablo de Greiff has pinpointed the impor­tant difference between collective and individual mea­sures: although development aid appears “to allow due recognition to be given to entire communities” and gives “the impression of making it pos­si­ble to reach goals of justice as well as development,” the disadvantage is

168

nicole l . i m mler

that such programs have “a very low reparative capacity, for they do not target victims specifically, and what they normally try to achieve is to satisfy basic and urgent needs, which makes their beneficiaries perceive such programs, correctly, as ones that distribute goods to which they have rights as citizens, and not necessarily as victims.”83 Moreover, De Greiff asserts, t­ hese mea­sures “do nothing to promote re­spect for p­ eople as individuals rather than as members of marginal groups.”84 This can indeed be observed in Rawagede: the victim families did not feel addressed by the microcredit initiative. Expressions used by f­amily members describing the ­widows as “cash machines for the village” are revealing. The way the ­widows’ families described how village officials collected the money, the disrespect for the w ­ idows (treated “like a puppet doll”), and the pressure from the village police (“not known as nice ­people”) is informative about the power hierarchies not only in the village, but also within families, where ­women and younger ­family members do not have equal say. The narratives above show that t­ here has been recognition but not much transformative justice in the w ­ idows’ lives. Poverty, local power relations and, gender inequalities seem not to have changed. Historian Selma Leydesdorff has similarly observed with regard to the Srebrenica w ­ idows that the compensation they received is part of a specific, l­egal discourse that is “centred on the indictment,” but often “neglecting the social and psychological dimensions of the victims’ emotional needs.” She emphasizes: “Monetary compensation never restores lost social and cultural capital.”85

conclusion Rawagede has achieved its place in history over the last de­cades, even if it is a dif­ fer­ent one on both sides of the ocean: in the Netherlands it is the epitome of colonial disaster; in Indonesia it is the symbol of heroic re­sis­tance.86 On the basis of field research, this chapter has assessed to what extent the reparations provided by the Dutch government to the w ­ idows of Rawagede had been effective and to what extent they correspond to local ideas of recognition and justice. Shifting the focus from ­here (the court in The Hague) to ­there (a village in West Java) has revealed a more nuanced view than the Dutch media image of “the village robbing the ­w idows,” which implies corruption and thus undermines our understanding of the compensation pro­cess. Have we seen corruption, or the translation of a Dutch court decision to local circumstances? Both are true. The “village model” has some validity in that is based upon shared experiences and feelings; at the same time, it assumes a romantic idea of village community and solidarity that has l­ittle to do with the social realities of strict gendered hierarchies and few choices. According to theory, reparation processes—­mirroring



Narrating (In)Justice—­The Rawagede Case 169

the relationship between individuals and their po­liti­cal community—­should provide recognition, and promote civic trust and social solidarity, which means a “forward looking perspective.”87 Enforced “sharing” by village authorities in the name of “social solidarity,” however, undermines the “civic trust” in them as in the community. If repair is defined as a pro­cess in which the person can develop hope and a feeling of control, that control is exactly what the ­widows lacked: they ­were not involved in the planning (asked about their ideas of justice), the decision-­making, or in the sharing pro­cess. This caused feelings of being victimized again. It is also evident that in a village such as Rawagede (which has about 3,000 residents ­today), where many p­ eople are somehow related to or dependent on each other, where poverty and scarcity are an everyday real­ity, and where cultural, social and economic positions are intensely intertwined, all is a ­matter of give and take. In such a context one has to ask ­whether the l­ egal approach, based upon individual cases and focused on individual suffering, is the right approach to colonial crimes. Vio­lence in the decolonization war was not an individual experience; ­whole families, villages, and regions have been affected by it.88 Most of the claimants in Sulawesi lost several ­family members in what has been called the “40,000 victims deaths by Westerling,” a magic number that shape real­ity for many ­people. Andi Monji, who as boy witnessed his f­ather being shot, represents many child-­ victims and fights against the exclusive recognition of the ­widows: “I told the court it is about victims, not a victim. . . . ​I ­will not share, but fight for the o­ thers.”89 ­Legal claims demand, however, a kind of evidence that challenges local systems of archived knowledge. Although the Committee for Dutch Debts of Honor (KUKB) put enormous effort into reaching out to potential candidates and collecting the required data with the help of regional coordinators, the real­ity of illiteracy, the limits of personal and public archives, and the lack of support on national level, have the pro­cess selective and ad hoc. “The individual claims should have been an icebreaker so that something collectively could happen,” explained ­lawyer Liesbeth Zegveld. But she is also aware that the pro­cess needs to be “much broader, more collective and po­liti­cal.”90 But how might this be accomplished? Scholars who argue that ad hoc compensation might not serve the victims, have suggested that po­liti­cal pro­cesses and state legislation may be more effective than fighting l­egal ­battles.91 But then, we might ask, what would a ­widow in one of t­ hose villages gain from such an approach? So far one can learn at least one lesson from the Rawagede case: if money is given to individuals, more attention must be given to what happens before and ­after compensation. Such attention means an anthropological assessment of the needs and expectations of recipients, and then a monitoring of the ­whole pro­ cess, together with local civil society partners, so that former victims are not

170

nicole l . i m mler

revictimized by the pro­cess, and so that the decision-­making and the sharing are self-­directed and create agency. Thus, ­there must be a clear communication of and transparency about the procedures beforehand and afterwards, which should not reside in the hands of some activist individuals. In order to strengthen the awareness of compensation as a social pro­cess, it is essential to modify our idea of compensation. Whereas in ­legal and po­liti­cal discourse, compensations are largely associated with closure, t­ hese mea­sures can cause complex dynamics in communities. Our research suggests that we reconsider the nature of the parties in a reparation pro­cess and take the idea of compensation as a dialogue between two opposing parties along the simplistic victim/perpetrator or colonized/colonizer binary, much further b­ ecause more players are involved. In villages such as Rawagede, compensations trigger a multiple dialogue within the community: conflicts within individuals (between individual desires and collective and religious duties), within families (between generations and close and distant ­family members), and within the victim community (the monument block versus the rest of the village). Moreover, this dialogue extends to neighboring villages, regional and national government, and even global society (in the relations between the Indonesian diaspora and the former colonizer). Exploring reparation as a social pro­cess on a village level revealed the relational nature of experiencing justice. I hope that this research contributes to a better understanding of the need to further localize transitional justice pro­cesses. It is also a reminder that the central aim of t­ hose reparative mea­sures is an improvement of the life of the w ­ idows and their families, within their own social environment.

notes This research was made pos­si­ble by a Marie Curie Intra-­European Fellowship, within the Seventh Eu­ro­pean Community Framework Program, for the proj­ect Narrated (In)Justice (No: 626577), and conducted within the program Understanding the Age of Transitional Justice: Narratives in Historical Perspective at the NIOD Institute for War, Holocaust, and Genocide Studies in Amsterdam. With many thanks to my colleagues, who pushed my thoughts with their expertise. 1. ​De Rechtspraak [Ruling, the Hague Court], https://­www​.­recht​.­nl​/­rechtspraak​/­​?­ecli​ =­ECLI:NL:RBSGR:2011:BS8793 (September 14, 2011). 2. ​Oskar N. T. Thoms, James Ron, and Roland Paris, “State-­Level Effects of Transitional Justice: What Do We Know?,” International Journal of Transitional Justice 4, no. 3(2010): 329–354. 3. ​Tricia D. Olsen, Leigh A. Payne, and Andrew G. Reiter, Transitional Justice in Balance: Comparing Pro­cesses, Weighing Efficacy (Washington D.C.: United States Institute of Peace Press, 2010). 4. ​Klaus Neumann and Janna Thompson, “Introduction: Beyond the Legalist Paradigm,” in Historical Justice and Memory, ed. Klaus Neumann and Janna Thompson (Madison: University of Wisconsin Press, 2015), 3–24, 20. 5. ​Ibid., 10 and 17ff. See also the claim upon ­lawyers to develop a “thicker” understanding of transitional justice by reflecting more about the consequences of legalistic discourses in such



Narrating (In)Justice—­The Rawagede Case 171

contexts and trying to be more imaginative in overcoming them Kieran McEvoy, “Letting Go of Legalism: Developing a ‘Thicker’ Version of Transitional Justice,” in Transitional Justice from Below: Grassroots Activism and the Strug­gle for Change, ed. Kieran McEvoy and Lorna McGregor (Oxford: Hart Publishing, 2008), 15–45. 6. ​Carol Gluck, Concluding Remarks, Conference on “Understanding the Age of Transitional Justice: Narratives in a Historical Perspective,” The Hague, December 10, 2015. 7. ​Stephan Parmentier, “Understanding the Age of Transitional Justice: Narratives in a Historical Perspective,” Conference, The Hague, December 10, 2015. 8. ​Eleanor Roo­se­velt, “In Your Hands,” Speech, 10th Anniversary of the Universal Declaration of ­Human Rights, New York, March 27, 1958. 9. ​Pablo se Greiff, “Justice and Reparations,” in The Handbook of Reparations, ed. Pablo de ­Greiff (Oxford: Oxford University Press, 2006), 451–477. 10. ​Selma Leydesdorff, “Why Compensation Is a Mixed Blessing,” in The Genocide Convention: The Legacy of 60  Years, ed. Harmen van der Wilt, Jeroen Vervliet, Göran Sluiter, and Johannes Houwink ten Cate (Leiden: Martinus Nijhoff, 2012), 105–114, 114. 11. ​Peter Malcontent, “Financial Compensation as a Po­liti­cal Pro­cess,” in Facing the Past: Amending Historical Injustices through Instruments of Transitional Justice, ed. Peter Malcontent (Cambridge: Intersentia, 2016), 265–284. 12. ​Elazar Barkan, The Guilt of Nations: Restitution and Negotiating Historical Injustices (New York: W. W. Norton, 2000), 318. 13. ​Barkan positions the historian as part of this dialogue pro­cess; Elazar Barkan, “Historical Dialogue: Beyond Transitional Justice and Conflict Resolution,” in Historical Justice and Memory, ed. Klaus Neumann and Janna Thompson (Madison: University of Wisconsin Press, 2015), 185–201, 185 and 200. 14. ​Barkan, The Guilt of Nations, 321. 15. ​I called it the “diaspora ele­ment of reparation claims” that members of the Indonesian diaspora in the Netherlands translate their perceived neglect of themselves and their history into a claim on behalf of the formerly colonized. An alternative ­human rights perspective argues that the claims go beyond acknowledging historical injustice to address the presence of a colonial mindset in Eu­rope; see Nicole L. Immler, “­Human Rights as a Secular Imaginary in the Field of Transitional Justice: The Dutch-­Indonesian ‘Rawagede Case,’ ” in Social Imaginaries in a Globalizing World, ed. Hans Alma and Guy Vanheeswijk (Berlin: De Gruyter, forthcoming). 16. ​Gert Oostindie, Soldaat in Indonesië, 1945–­1950: Getuigenissen van een oorlog aan de verkeerde kant van de geschiedenis (Amsterdam: Prometheus, 2015), 26. 17. ​Ibid.; Rémy Limpach, De brandende kampongs van generaal Spoor (Amsterdam: Boom, 2016); in En­glish, Rémy Limpach, “Business as Usual: Dutch Mass Vio­lence in the Indonesian War of In­de­pen­dence, 1945–49,” in Colonial Counterinsurgency and Mass Vio­lence: The Dutch Empire in Indonesia, ed. Bart Luttikhuis and A. Dirk Moses (London: Routledge, 2014), 64–90. 18. ​ Excessen van Rawagede, documentary on RTL4, originally aired on August 17, 1995. 19. ​Cees Fasseur, quoted in “Rechtszaak Indonesische slachtoffers tegen Nederland bij voorbaat verloren,” Trouw, Augustus 17, 1995. 20. ​“Het spoor terug: Het bloedbad van Rawagede,” episode of the Dutch documentary TV program Onvoltooid verleden tijd, originally aired on December 9, 2007, https://­www​.­v pro​.­nl​ /­speel~POMS​_­VPRO​_2­ 06515~ovt​-­09​-d­ ecember​-­2007​-­uur​-­2​-­31​-m ­ in~​.­html. 21. ​Foundation Committee of Dutch Debts of Honor, http://­www​.­kukb​.n ­ l​/­new​/­index​.­php. 22. ​Liesbeth Zegveld in a discussion on “Reparations for Historical Wrongs,” The Hague Institute of Global Justice, October 14, 2016.

172

nicole l . i m mler

23. ​Pres­ent ­were Saih with his son, Ibu Sariman, Sukarman with his ­daughter, and Warjo, a

child survivor representing his ­mother, Ibu Layem. 24. ​Michel Maas, “Rawagede-­weduwen geplukt,” Volkskrant, December 24, 2011. 25. ​Jeffry Pondaag, interviewed by the author in Hemskerk, the Netherlands, November 13, 2014. 26. ​This chapter is based upon more than forty semi-­structured interviews conducted by the author with the help of a translator in October 2015 with victims and ­family members, including eight families in Rawagede and twelve in Sulawesi, five local experts/activists. 27. ​Based upon an informal agreement with the Dutch government, announcement published in September 2013 at https://­zoek​.­officielebekendmakingen​.­nl​/­stcrt​-­2013​-­25383​.­html. 28. ​De Rechtspraak [Ruling, The Hague Court], https://­recht​.­nl​/­rechtspraak​/­​?­ecli​ =­ECLI:NL:RBDHA:2015:2442 (March 11, 2015). 29. ​Both interviewed by the author at the Vrije Universiteit, Amsterdam, August 19, 2014. 30. ​In Rawagede, eleven w ­ idows and one survivor received compensation. Around fifty child-­ victims are currently preparing a claim. Twenty-­three w ­ idows from South Sulawesi received compensation, and sixteen are in the current court case. 31. ​Suzanne Liem went to Rawagede already in 2010 to portray the w ­ idows and their stories. See Suzanne Liem, “De weduwen van Rawagede en de heer Saih bin Sakam, interviews en fotografie,” unpublished manuscript, 2010, partly published in Trouw, August 23, 2010. 32. ​Karmas, interview with author, Balongsari, October 20, 2015. 33. ​Luus, interview with author, Balongsari, October 20, 2015. 34. ​Ibu Tijeng, interview with author, Balongsari, October 20, 2015. In the following all quotations are from the same ­family interview. 35. ​Liem, “De weduwen van Rawagede,” 4. 36. ​Ibu Tijeng, interview with author. 37. ​Grand­child, interview with author, October 20, 2015. 38. ​Karmas, interview with author. 39. ​Luus, interview with author. 40. ​Ibu Tijeng, interview with author. 41. ​Grand­daughter, interview with author. 42. ​Liem, “De weduwen van Rawagede,” 16. 43. ​Sukarman, interview with author, October 20, 2015. 44. ​Group Wali, Nenekku Pahlawanku, Nagaswara, ­music video, https://­www​.­youtube​.­com​ /­watch​?­v​=­CU6rPv4xXhQ. 45. ​Esther Bootsma, “Koningin ontvangt overlevende bloedbad niet,” NOS, November 9, 2010. 46. ​Warjo, interview with author, October 22, 2015. 47. ​Luus, interview with author. 48. ​Ibu Wanti, interview with author, Sedari, October 24, 2015. 49. ​Ibid. 50. ​Sophia, interview with author in Sedari, October 24, 2015. 51. ​Warjo, interview with author. 52. ​Ibid. 53. ​Paul ter Weel, local farmer in Pasirawi, West Java, interview with author, October 19, 2015. 54. ​Sri, interview with author, October 23, 2015. 55. ​Luus, interview with author. 56. ​Mamat, interview with author, October 23, 2015. 57. ​Sukarman, interview with author, October 20 and 23, 2015. 58. ​Didi and Ujar interview with author, October 22, 2015.



Narrating (In)Justice—­The Rawagede Case 173

59. ​Ibid. 60. ​Luus, interview with author. 61. ​Mustari, son of Warjo, interview with author, October 22, 2015. 62. ​Ibu Cawi: “Ik herinner mij dat Rawagede bekend stond als het dorp van Randa (weduwe)

Tekdung (beng beng), dat is Sundanees voor: ‘dorp van de weduwen van de neergeschotenen’ ” [“I remember that Rawagede known as the village of Randa (­widow) Tekdung (beng beng), that is Sundanese for: ‘village of the ­widows of the executed’ ”]; Liem, De weduwen van Rawagede, 2010, 4. 63. ​Tati Krisnawaty, interview with author, Pasirawi, October 19, 2015. 64. ​Darul Islam was an organ­ization fighting for Indonesia as an Islamic state, first in West Java during the revolution and l­ ater also in South Sulawesi. 65. ​Tati Krisnawaty, interview 66. ​Larissa van den Herik, “Addressing ‘Colonial Crimes’ through Reparations? Adjudicating Dutch Atrocities Committed in Indonesia,” Journal of International Criminal Justice 10, no. 3 (2012): 693–705. See also Immler, “­Human Rights as a Secular Imaginary.” 67. ​Birgit Bräuchler, “Introduction: Reconciling Indonesia,” in Reconciling Indonesia: Grassroots Agency for Peace, ed. Birgit Bräuchler (London: Routledge, 2009), 3. 68. ​Erin Daly and Jeremy Sarkin, Reconciliation in Divided Socie­ties: Finding Common Ground (Philadelphia: University of Pennsylvania Press, 2007). 69. ​Peggy Levitt and Sally Engle Merry, “The Vernacularization of ­Women’s H ­ uman Rights,” in ­Human Rights F ­ utures, ed. Leslie Vinjamuri, Jack Snyder, and Stephen Hopgood (Cambridge: Cambridge University Press, forthcoming). 70. ​Joseph R. Slaughter, ­Human Rights, Inc.: The World Novel, Narrative Form, and International Law (New York: Fordham University Press, 2007). 71. ​For insights into the complexity of the h ­ uman rights discourse, see Bradley R. Simpson, “Self-­Determination, ­Human Rights, and the End of Empire in the 1970s,” Humanity: An International Journal of H ­ uman Rights, Humanitarianism, and Development 4, no.  2 (2013): 239–260. 72. ​Katherine McGregor, “From National Sacrifice to Compensation Claims: Changing Indonesian Repre­sen­ta­tions of the Westerling Massacres in South Sulawesi, 1946–47,” in Colonial Counterinsurgency and Mass Vio­lence: The Dutch Empire in Indonesia, ed. Bart Luttikhuis and A. Dirk Moses (London: Routledge, 2014), 283. 73. ​Ibid., 295–302, 300. 74. ​Nancy Fraser, “Rethinking Recognition,” New Left Review 3 (May–­June 2000): 107–120, 109. 75. ​Nora Jacobson and Dianne Greenley, “What Is Recovery? A Conceptual Model and Explication,” Psychiatric Ser­vices 52, no. 4 (2001): 482–485, 483. 76. ​De Greiff, “Justice and Reparations,” 460. 77. ​Tati Krisnawaty, e-­mail message to the author, October 6, 2016. 78. ​Paul ter Weel, conversation with the author in Delft, October 4, 2016. 79. ​De Greiff, “Justice and Reparations,” 461. 80. ​Karmas, interview with author. 81. ​Luus, interview with author. 82. ​ter Weel, interview with author, October 19, 2015. 83. ​De Greiff, “Justice and Reparations,” 470. 84. ​Ibid. 85. ​Leydesdorff, “Why Compensation Is a Mixed Blessing,” 105. 86. ​More research necessary to fill out complexities in ­these black-­and-­white stories.

174

nicole l . i m mler

87. ​De Greiff, “Justice and Reparations,” 466. 88. ​New historical research confirms the structural character of this vio­lence. See Oostindie,

Soldaat in Indonesië, and Limpach, De brandende kampongs. 89. ​Monji, interview with author, Suppa-­Pinrang, South Sulawesi, October 13 and 14, 2015. 90. ​Conversation ­after the last court case on January 27, 2016, in The Hague. 91. ​See Van den Herik, “Addressing ‘Colonial Crimes’ through Reparations?”

8 • COLLECTIVE AND COMPETITIVE VICTIMHOOD AS IDENTIT Y IN THE FORMER YUGO­S L AVIA CH RISTI A N A X BO E N I ELSEN

In May 2015, events took place throughout Eu­rope to commemorate the seventieth anniversary of the end of the Second World War. Although the Rus­ sian occupation of the Crimean peninsula and the ongoing crisis in Ukraine injected a discordant note and led the United States, Canada, and EU members to decline participation in a ­grand military parade in Moscow on May 9, most commemorations ­were held in the spirit of comity, reconciliation, and Eu­ro­pean partnership. They contrasted starkly with the acrimonious concurrent events in Croatia and Serbia, where controversies from the Second World War once again took center stage. In Croatia, commentators traded barbs regarding the commemoration of the Bleiburg massacre of Croat Fascists and civilians by the Communist Partisans in May 1945. Seen through the eyes of the po­liti­cal Right, Bleiburg and the subsequent repression committed by the Socialist Yugo­slav dictatorship invalidated claims by the po­liti­cal Left that May 1945 marked a liberation. In Serbia, the High Court in Belgrade officially rehabilitated the leader of the royalist Chetnik forces during the Second World War, Dragoljub “Draža” Mihailović.1 His trial and execution in 1946 w ­ ere determined to be invalid, and therefore the Court proclaimed that he could be rehabilitated and exonerated. Meanwhile, efforts to rehabilitate the quisling leader of German-­controlled Serbia, Milan Nedić, also proceeded apace. ­These rehabilitations deeply divided the Serbian public, and even government officials issued contradictory statements. ­Those in Serbia who viewed Socialist Yugo­slavia as a less than wholly negative experience regarded the decision as a shameful and historically illiterate exoneration of collaborators with Fascist occupiers.2 175

176

c hristi a n a x boe nielsen

figure 8.1. Croat participant at Bleiburg commemoration with sign invoking the

“Holocaust” allegedly committed by Yugo­slavia against the Croat nation, 2016. (Photo by Christian Axboe Nielsen.)

In sum, casual observers arriving in Croatia or Serbia in early May 2015 could be excused for thinking that the Second World War had not ended, but was rather still being fought with other means. It is hardly surprising, therefore, that the Croat po­liti­cal commentator Jelena Lovrić remarked on a strangely beneficial, though apparently insufficient byproduct of the con­temporary refugee and mi­grant crisis in Eu­rope: “Instead of Ustashas and Partisans, black and red zombies, ­imagined Yugo­slavs and similar bogeymen constructs . . . ​the drama of the refugees from the East who, completely expectedly, are also flowing into Croatia could be the key to the result of the coming elections.”3 Subsequent events dur-



Collective and Competitive Victimhood 177

ing and a­ fter the Croatian elections showed Lovrić’s hopes to be naïve. In January 2016, a new Croatian government was formed. Although it lost a vote of confidence less than six months ­later, this government was widely perceived both at home and abroad as being obsessed with the politics of the past. This chapter explores the construction and history of exclusivist narratives of victimhood in the former Yugo­slavia. In the years 1918 to 1991, from Slovenia in the northwest to Kosovo and Macedonia in the southeast, the territory encompassed first by the Kingdom of Yugo­slavia and l­ater Socialist Yugo­slavia witnessed a very high concentration of repressive regimes punctuated by bouts of collective vio­lence and mass atrocities. This la­men­ta­ble track rec­ord culminated with a particularly destructive and protracted state collapse, the consequences of which are still being felt ­today. Perhaps the most extraordinary aspect of the memory of ­these events is the manner in which multiple communities in the former Yugo­slavia assert that their own community was the primary or even sole victim of the Yugo­slav experience and its aftermath. That is, by selectively appropriating and instrumentally applying po­liti­cal and cultural modalities that have become widespread since the end of the Second World War—­and particularly since the beginning of the boom in memory studies and museums commemorating mass vio­lence—­the successor states of Yugo­slavia have tended to cultivate an exclusive notion of national victimhood. This has led to a double discursive hegemony, as it ­were, in which what dominates is not just the nation, but more specifically the (ongoing) suffering or martyrdom of that nation. In politics, the media, and academia, the status of the nation as primary or exclusive victim at times dominates the identity and hence also the agency of that nation. Thus conceived, the resulting identity militates against owner­ship of socioeconomic and other challenges, which are perceived almost exclusively as caused by erstwhile enemies of the nation and their lingering remnants. This not only impedes postconflict reconciliation and makes socie­ties in the former Yugo­slavia much less cognitively receptive to the findings of the International Criminal Tribunal for the Former Yugo­slavia (ICTY), but also breeds a passivity that shirks the confrontation with the real challenges facing ­these states and socie­ties. For the purposes of this chapter, I define competitive victimhood as a pro­cess in which the in-­group is not only focused exclusively on its own real and alleged suffering, but also aspires to demonstrate and assert that this suffering is greater than that of other groups. The in-­group w ­ ill hence tend to deny, question, or belittle the suffering of other groups. While competitive victimhood typically takes place among ethnic groups that perceive themselves to be historical rivals, competitive victimhood can also be transmitted internationally as a way to gain visibility for the group and stake claims of po­liti­cal and territorial legitimacy. Although competitive victimhood often emanates from past wrongs committed against the group, it is also frequently invoked with re­spect to pres­ent grievances or even as a harbinger of f­ uture misfortune, should the cohesion or leadership of

178

c hristi a n a x boe nielsen

the in-­group be reduced. In Bosnia and Herzegovina in 1990, the nationalist parties founded to contest the first multiparty elections all played on past collective vio­lence and deployed the threat or fear of genocide in order to mobilize voters in a time of systemic crisis.4 And while the suffering that forms the foundation of competitive victimhood necessarily encompasses the deaths or injuries of individual ­human beings, the salience and utility of ­these victims to prac­ti­tion­ers of  collective victimhood derives primarily from their collective identity. Thus, a Serb, Croat, or Bosniak victim of an atrocity holds interest for nationalist Serb, Croat, or Bosniak politicians primarily ­because the victim symbolizes the collective suffering or martyrdom of the nation. As Elissa Helms recounts in her compelling study of victimhood in Bosnia, “Some of my educated, consciously anti-­ nationalist (and arguably jaded) acquaintances exasperatedly complained that it was as if some of their leaders w ­ ere happy that so many Bosniacs had died in the war.”5 Any detailed and critical examination of the narratives of collective victimhood reveals an obsession with the group rivaled only by a disinterest in the individual humanity of the victims, who are symbolically thrust forward as a justification for the group’s past suffering and pres­ent and f­ uture claims. The perpetrator-­victim dichotomy radiates with sublime clarity. “We” can never be perpetrators, but only victims; “they” can only be perpetrators, but never victims. N ­ eedless to say, in this narrative, victims can never have been perpetrators themselves in the past, and each victim epitomizes the victim group. As a corollary, in criminal ­trials, the victim as witness is primarily of interest as a performer in a ritual designed to demonstrate the suffering of the group, not as a ­bearer of evidence to be examined critically in a l­egal environment.6 It is informative to contrast the attention devoted by mass media in the former Yugo­slavia to some of the most notorious atrocities of the wars of Yugo­slav succession (Vukovar, Srebrenica, and so on) with the relative lack of interest that ­these same media outlets have shown for the testimony of a­ ctual victims and survivors of t­ hese crimes. Before looking at the specific case(s) of the former Yugo­slavia, it might be helpful to cast a brief glance at international developments in transitional justice. In the wake of the mass atrocities of the Second World War, researchers devoted a tremendous amount of attention to analyzing the identity, organ­ization, and be­hav­ior of perpetrators. Indeed, in the de­cades immediately following the war, not much attention was paid to the victims of t­ hese crimes. The same was true of the Nuremberg T ­ rials, where perpetrators and their self-­incriminating documentation took center stage, while victims largely figured as an afterthought, if at all. At the same time, the Nuremberg ­Trials also applied the princi­ple of individual rather than collective criminal responsibility. Only in the early 1960s, with the trial of Adolf Eichmann in Jerusalem, did the focus begin to move away from perpetrators. As Lawrence Douglas has shown, leading Israeli officials regarded the Eichmann trial as an opportunity to advance



Collective and Competitive Victimhood 179

a ­grand narrative justifying the existence of the state of Israel.7 Meanwhile, in Holocaust scholarship, the 1960s witnessed the beginning of a shift not only to research about victims, but also to the creation of collective understandings of the Holocaust in which victims and their stories outranked perpetrators. In the United States, this has been described in depth by Peter Novick.8 Throughout the 1970s and 1980s, particularly in the light of atrocities committed in Latin Amer­ i­ca and apartheid in South Africa, broader notions of transitional justice, including nonjudicial remedies and expanded victim participation, continued to percolate. In the 1990s, the genocide in Rwanda and the wars of Yugo­slav succession compelled the establishment of two ad hoc international tribunals. In carry­ing out their work, ­these tribunals ­were occasionally criticized for not ­doing enough to integrate victims into their work or to disseminate and explain their work to victims’ communities and organ­izations.9 In 2002, the permanent International Criminal Court (ICC) began to operate, and its Rome Statute incorporated a formal degree of victim participation not previously seen anywhere in international criminal justice. In brief, both in scholarship and in international criminal justice, it could be argued that whereas the focus in the first de­cades a­ fter the Second World War was primarily on perpetrators, victims’ interests have been on the rise during the past two de­cades. The question I wish to raise h­ ere is w ­ hether this has brought with it unintended and potentially harmful consequences. In par­tic­u­lar, is ­there a link between the focus on victims and the propagation of narratives of collective suffering, which can mutate into exclusivist national narratives? And if this is the case, should we as critical scholars not also turn our attention to addressing and preventing ­these consequences?

victims, perpetrators, and perceptions of justice In ­trials of genocide, war crimes, and crimes against humanity, a thin and often invisible line has separated the narratives of individual victims from the collective identity ­because of which they ­were often persecuted or killed.10 In part this is a function both of the crimes and the ­legal requirements for prosecuting them. Crimes against humanity require demonstrating that the under­lying actions that contributed to the commission of crimes ­were widespread and systematic.11 Genocide goes even further. Not only does the 1948 Genocide Convention restrict the definition of genocide to certain protected groups, but it also requires the existence of a special and specific intent to destroy this group in ­whole or in part, as such. ­These l­ egal requirements tend to channel the focus away from individual ­human beings as victims and t­oward victim groups, particularly ethnic groups and nations. By ­doing so, they reify and reproduce group identities, often in a way that inadvertently confirms the heavi­ly reductionist and negative images or Feindbilder propagated by the perpetrator group. Moreover, by including certain

180

c hristi a n a x boe nielsen

groups, the convention excludes o­ thers, notably po­liti­cal groups, thereby creating a hierarchy of group identities. In criminal justice, this topic leads us back to fundamental questions about the purpose of criminal ­trials. Are they supposed to lead primarily to the prosecution, conviction, and punishment of perpetrators, or are they supposed to deliver some form of vengeance and/or “closure” for the victims—or both? ­Here it may be useful to contemplate a comparison with domestic criminal justice. Modern domestic criminal justice systems seek to address criminal wrongdoing against individuals and groups. This can be attempted or achieved through the conviction and punishment of the perpetrators and by granting a voice to the victims through their own testimony. In some systems, civil ­legal procedures also exist, which permit the victims to sue perpetrators for financial damages or compensation of material loss as well as physical and m ­ ental harm. Yet domestic criminal systems as well as society as a ­whole also generally tend to recognize that ­there are limits to the role and even the rights of the victim. (The most basic example is that criminal justice systems do not permit, and are supposed to obviate the need for, vigilante justice on the part of the victim and his or her relatives or community.) Roughly put, victims are at some point expected to be able to achieve “closure” and then henceforth to transcend this pro­cess and resume their previous roles in society.12 Certainly t­here are cases in which victims (or their relatives) of notorious crimes—­sexual vio­lence, for example—­assume advocacy roles that can last long a­ fter crimes have been committed. In cases where justice is delayed or denied, this role can endure much longer than in cases where justice is seen to be properly and expeditiously administered. Generally, however, society does not tolerate “eternal victims,” particularly if ­these are seen to be exploiting their status as victims for personal, po­liti­cal, or economic gain. Nor does society generally tolerate victims’ claims to exclusivity at the expense of other victims. Simply stated, the suffering of victims of attempted murder or aggravated assault does not trump that of victims of rapes or even burglaries, though most ­people would recognize that victims of violent crimes have suffered more than ­those of nonviolent crimes. In this sense, it is useful to contrast the acceptable but limited roles of victimhood that most socie­ties grant individual victims with the open-­ended and rather unlimited space afforded to collective victimhood in the former Yugo­slavia. Moreover, in considering this point, we might also recall that, since the Nuremberg ­Trials, collective guilt has been discarded in ­favor of individual criminal responsibility. To be sure, since the writings of Karl Jaspers, it has been recognized that broader categories of ethical and moral responsibility apply.13 Germany is the paragon in this re­spect, and its leading politicians regularly remind their citizens that ­there is no expiration date for Vergangenheitsbewältigung. The ad hoc U.N. tribunals and the ICC have also upheld the princi­ple of  individual criminal responsibility, though prosecutors, judges, and spokes­



Collective and Competitive Victimhood 181

persons for ­these institutions have often had to compete with public perceptions in both perpetrator and victim communities that collective guilt is at play. Conversely, a growing international understanding has emerged since the end of the Cold War that socie­ties that w ­ ere occupied by Nazi Germany, such as Poland, the Netherlands, or Denmark, should not enjoy an exemption from confronting past episodes in which the ethnic majority in t­ hose countries engaged in criminal be­hav­ior or inflicted suffering on minorities.14 France, the United Kingdom, Italy, and other countries have also had to confront colonial oppression. The beginnings of this discussion are evident in the Netherlands, where the ­human rights violations committed by the Dutch army in Indonesia’s war of in­de­pen­dence are leading issues in the decolonization debate.15 Outside Eu­rope, socie­ties like the United States, Canada, and Australia are starting to confront the oppression of native populations and, in the U.S. case, slavery. In the G ­ reat Lakes region of Africa “a sense of collective victimhood exists among most ethnic groups in the region and has contributed to what is perceived as legitimate defensive or preemptive vio­lence.”16 And in Japan, an oft-­reluctant po­liti­cal leadership recognizes that atonement must be made for the war crimes and crimes against humanity committed by the Japa­nese Empire in the 1930s and 1940s. Many of ­these reckonings have been greatly delayed and, at least initially, involved considerable controversy. Yet it can be stated that modern demo­cratic socie­ties recognize that it is the obligation of ­every society to adopt a critical interpretation of the past in the interests of refining democracy and h­ uman rights in the pres­ent and ­future.17 This is the basis on which the field of transitional justice emerged.

the case of (the former) yugo­slavia Much of Yugo­slav historiography from late Socialist Yugo­slavia and its successor states can seem obsessed with the documentation of its own suffering. This doubtless has something to do with the very high levels of po­liti­cal oppression and mass vio­lence witnessed in the Balkans since the beginning of the nineteenth ­century and, in par­tic­ul­ ar, during the Second World War. Indeed, history as a discipline in the Balkans in large part emerged as a national endeavor, charting the heroic emergence of Christian nation-­states from the ashes of the accursed “Ottoman yoke.”18 Hence, even prior to the establishment of Yugo­slavia, a pre­ce­dent had been set for the denial or marginalization of the suffering of o­ thers. Massacres of Balkan Christians by Ottoman forces w ­ ere carefully documented and incorporated into the national narrative, while massacres against Muslims or deportations of Muslims w ­ ere depicted as a necessity, to the degree that they w ­ ere addressed at all. As Mark Mazower has written, “In the popu­lar imagination the vio­lence ran only one way,” and the be­hav­ior and concerns of Christian Eu­ro­pean powers reinforced this bias among the elites of the emerging Balkan nation-­states.19 Subsequently, they often ascribed setbacks and challenges in the socioeconomic

182

c hristi a n a x boe nielsen

and po­liti­cal development of Southeastern Eu­rope to structural deficiencies inherited from centuries of Ottoman rule. Some of this view can even be experienced anecdotally ­today, often in combination with the presumptive, fatalistic, and self-­orientalizing pronouncement: “Ovo je Balkan” (“This is the Balkans.”) While working in the Archive of Vojvodina in 2013, I complained about irregularities in the per­for­mance of public transport, upon which an archivist immediately reacted by saying, “What do you expect? We have been ­under the Turks for so long.” When browsing the historiographies of the former Yugo­slavia, one is struck by the number of works by Croat historians on Croat victims, by Serb historians on Serb victims, and so on. Works by, for example, Kosovo Albanian authors on Serb or Bosniak victims are rare, and Slovene historians show ­little interest in Croat victims. Why is it so difficult to identify historians and politicians in the former Yugo­slavia who are willing to confront and discard notions of collective victimhood? Why is past suffering instead regarded as a zero-­sum game, in which ­there is room only for “our” own suffering, and where any admission of the suffering of the Other is seen as a negation of “ours”? Why do so few historians choose to document and analyze the past suffering of the Other, or even deny their suffering entirely or dismiss it as insignificant in the light of “our own” suffering? The deeply flawed establishment a­ fter the First World War of the Kingdom of Serbs, Croats, and Slovenes (officially renamed the Kingdom of Yugo­slavia in 1929) as well as its subsequent history of fragile parliamentary politics and royal dictatorship, was hardly conducive to harmonious interethnic coexistence. This protracted and unhappy experiment in crafting a constitutional monarchy took place on the heels of the First World War, which had famously been ignited in Sarajevo and then pitted the South Slavs against each other in the strug­gle between the Austro-­Hungarian Empire and Serbia. Of all combatant nations, Serbia suffered the highest proportionate loss of ­human life during the First World War. This “Calvary or Golgotha of the Serb nation” and subsequent “resurrection,” as it was termed both contemporaneously and by nationalist historians, led Serb elites to demand a leading role in the po­liti­cal life of interwar Yugo­slavia. For Croats and other non-­Serbs, the result became “Serb hegemony,” invalidating the country’s legitimacy and fatally eroding any previously existing possibilities for the emergence of a common Yugo­slav idea.20 Much worse was the Second World War, which destroyed Yugo­slavia, exposing its population to foreign Fascist occupation, catastrophic and atrocious misrule by Croat Fascists, and a protracted civil war within the larger international conflagration.21 While Communist forces succeeded in establishing a second, Socialist Yugo­slavia, the Communist takeover also witnessed a maelstrom of mass atrocities such as Bleiburg. Together with the first de­cade of intensive Stalinist rule and repression in Yugo­slavia, ­these massacres came to constitute a kind of original sin. Opponents of the regime both during and a­ fter Yugo­slavia seemed incapable of



Collective and Competitive Victimhood 183

acknowledging that, while Yugo­slavia remained a Communist Party state ­until its collapse, the level of repression in the 1970s and 1980s was dramatically lower than in the first de­cade ­after the Second World War. On the other hand, the League of Communists of Yugo­slavia ensured that any investigations of Communist crimes remained taboo. One should also not underestimate the role of nationalist diasporas, where openly Fascist counternarratives of collective victimhood had flourished.22 Only beginning in the late 1970s did the historiography, like the country, began to disintegrate into mutually hostile nationalist camps. As the final days of Tito’s life approached at the end of the 1970s, restive currents could be detected among historians. At first hesitantly and then increasingly boldly, historians challenged the founding myths of Socialist Yugo­slavia, ultimately even broaching the previously most taboo subject of all, the role of Tito. Several years ­later, in 1986, a group of leading scholars in Serbia produced a draft memorandum that was leaked to the press before its completion.23 Collating and synthesizing grievances of the Serbs, the authors of the memorandum concluded that the Serb nation was contemporaneously suffering nothing less than a genocide, particularly in Kosovo. Elsewhere in Yugo­slavia, other nations continued to develop their own national identities in ways that often involved dwelling upon past and pres­ent oppression.24 Perhaps particularly a­ fter the promulgation of the decentralizing 1974 constitution, the 1970s saw the emergence of public accusations of each group’s own suffering as part of the larger game of asserting national identity and rights in the competitive arena of Yugo­slavia. Culturally and educationally, the extensive decentralization of the country meant that school textbooks focused predominantly on the “own” culture in the respective regions of the federation.25 The contested federal re­distribution of funds to lesser developed areas of Yugo­slavia provided ample fuel for competitive grievances. Hence, even arguably positive developments such as hard currency proceeds from tourism or the ability of Yugo­ slav workers to seek employment outside the country as Gastarbeiter [mi­grant workers] could also be depicted as symptoms of Croat suffering. Seen from this perspective, large-­scale economic emigration and the re­distribution of hard currency income perpetuated the martyrdom of the Croat nation. On the level of Yugo­slavia as a ­whole, events that ­were at first glance marks of prestige and pride, such as the 1984 Winter Olympics, exacerbated fierce “inter-­republican rivalries” and “opened up a dangerous space in Yugo­slav society, a space that exposed some of the economic and po­liti­cal issues that would become tragically unworkable by the end of the de­cade.”26

184

c hristi a n a x boe nielsen

from “brotherhood and unity” to a competition of suffering All of this can be said to have culminated in an extraordinary situation in which ­every single constituent nation in Yugo­slavia, as well as the Albanians, the largest non-­Slavic ethnic minority in the country, asserted themselves to be the primary victims of Yugo­slavia. Interethnic communication in Kosovo, while it had never lived up to the “brotherhood and unity” ideal of Communist Yugo­slav propaganda, disintegrated in the 1980s into a dialogue of the hostile and deaf.27 This tapestry of collective suffering provided the backdrop for the collapse of the country. Moreover, the positioning of the collective self as an exclusively suffering group that practiced only self-­defense and never perpetrated criminal atrocities, militated against consensus over necessary reforms at the end of Yugo­slavia. ­Later, exclusivist notions of suffering contributed to one-­sided understandings of the vio­lence that accompanied the country’s dissolution. Ironically, particularly in Serbia, the obsession with victimhood also set the stage for interpretations of state collapse in which only the Serbs, despite being allegedly heavi­ly disadvantaged in Yugo­slavia, acted to save the common state. This distorted approach to history led the satirical Serbian online newspaper Njuz​.­net to quip in 2012 that Serbia had proclaimed a national day called “But what did they do to us.”28 If the writing of history in the nineteenth ­century had as its primary objective the assertion of the primacy of one’s own nation, as Dubravka Stojanović and ­others have argued, then the writing of history in the late twentieth ­century seemed to emphasize the suffering of one’s own nation.29 And since history is a broader enterprise that extends to museums, documentary and feature films, magazines, blogs, and all manner of “history culture,” this suffering is also highlighted and reproduced ­there. A good con­temporary example can be located in the Macedonian national history museum, part of the megalomaniacal prestige proj­ect that is called “Skopje 2014.” The lengthy, triple-­barreled official name of the museum says it all: “The Museum of the Macedonian Strug­gle for Statehood and In­de­pen­dence / The Museum of the VMRO [Internal Macedonian Revolutionary Organ­ization] / the Museum of Victims of the Communist Regime.” ­Here we see that Macedonian victimhood is, if not eternal, at least of very long duration, transcending multiple fiendish foreign regimes. And the experience does not stop with the names of the museum. Visitors can see the museum only by participating in lengthy guided tours (in Macedonian or English—no photo­ graphy is allowed!), where they are regaled with tales of dastardly feats perpetrated by Turks, Serbs, Greeks, Albanians, and Bulgarians against innocent Macedonians. When I took the tour in 2015, none of the approximately twenty participants asked any questions, and the official guide at no point stopped to solicit any questions from his captive audience. The truth of “our” collective suffering is apparently expected to be absorbed and internalized uncritically.



Collective and Competitive Victimhood 185

Particularly in Croatia and Serbia, the regimes of the 1990s pedaled the notion of national reconciliation between Fascist and Partisan traditions as a necessity. In Croatia, President Franjo Tuđman, himself both a former Partisan general and a dissident nationalist historian, argued that, with the very existence of Croatia and Croats at stake, the nation had to bury the po­liti­cal hatchet and unite. Similarly, in Serbia, the 1990s witnessed bizarre co­ali­tions of barely reformed Communists such as Slobodan Milosevic with extreme nationalist Serbs such as ­those from the ranks of Vojislav Šešelj’s Serb Radical Party. This led to an artificial, uneasy, and temporary ceasefire in Croatia and Serbia, which lasted as long as ontological security was not attained. Yet since at least the death of Tuđman in 1999 and the removal of Milosevic in 2000, this co­ali­tion has collapsed and given way to protracted debates about the crimes of the Second World War and Socialist Yugo­slavia.

settling the books Since the late 1990s, several well-­meaning international and nongovernmental organ­izations have attempted to tackle collective victimhood by sponsoring the writing of joint history textbooks. Such attempts typically seek ­either to write history as a product of the “lowest common denominator” (analogous to what ­lawyers call “adjudicated facts”) or to pres­ent former Yugo­slav students with the novelty of examining and commenting upon conflicting narratives regarding the same historical events.30 In practice, such other­wise admirable efforts have had ­little traction on the ground, not least b­ ecause national (or in Bosnia and Herzegovina entity-­level or cantonal) educational ministries exert an iron grip on the choice of textbooks and curricula.31 And problematic textbooks are merely symptomatic of a much bigger issue: educational systems that to this day neither encourage critical thinking nor prepare students for it. It remains nearly unimaginable that ethnic Albanian students would be asked to consider the history of Kosovo from a Serbian point of view, or that ethnic Macedonian students might be encouraged by their teacher to pres­ent the ethnic Albanian or Bulgarian view of Macedonia’s recent history as part of a group exercise in class. Moreover, it is not just that the textbooks and educational system presume a mono-­ethnic point of view, it is more precisely that they pedal monoethnic martyrdom and victimhood. This even seeps into the teaching of broader world historical issues. As Jelena Subotić has observed, Serb pupils confronting the Holocaust in history textbooks are taught that Serbs ­were the primary victims of the Holocaust, and that if Jews or Roma happened to be killed, it was b­ ecause “they w ­ ere supporting their Serbian friends.”32 In this way, the Holocaust becomes a symbol of international recognition of suffering that Balkan nationalists—­including ­those who openly espouse anti-­Semitic views—­aspire to achieve. It was thus not surprising to see a participant at the Bleiburg commemoration in 2016 carry­ing a loquacious

186

c hristi a n a x boe nielsen

oversized sign that proclaimed Bleiburg to be the “Holocaust of the Croat nation.”33 Only a few meters away, a brisk sale of extreme nationalist and fascist memorabilia was being conducted, while in the background prominent ministers from the Croatian government and the Roman Catholic Church held speeches. Competitive victimhood has extremely direct and negative consequences for transitional justice. As the term “competitive” suggests, t­ rials at the ICTY tend to be portrayed by both regional media and politicians as a competition in which the winners are ­those who see fewest perpetrators from their own ranks convicted of serious international crimes and, conversely, who receive the ICTY’s confirmation of their own status as victims. The most facile manifestation of this occurs when the number of persons indicted or convicted by the tribunal is sorted by ethnicity. Yet the more consistent prob­lem that transcends the tribunal is the abject lack of interest in perpetrators from one’s own ranks and in the victims from among the ranks of the ­others. A related prob­lem is the willful misinterpretation of ICTY judgments to confirm or reject dominant national narratives. In the Croatian context, this has mainly revolved around the legitimization of the war from 1991 to 1995, which in Croatian is referred to officially as the Homeland War (Domovinski rat). Both the convictions of the Croat generals Ante Gotovina and Mladen Markač and their subsequent acquittals on appeal ­were flagrantly misinterpreted within a discourse of collective and competitive victimhood by media, politicians, and government ministers in Croatia. Th ­ ere is a widespread perception in Croatia that recognizing Serb victims or Croat perpetrators ­will somehow invalidate or weaken Croatia’s claim to in­de­pen­dence.34 In the first instance, the convictions ­were interpreted as a criminalization of the Homeland War, and as an attempt by the ICTY to (re)write Croatian history. Conversely, the acquittals ­were heralded as international recognition that the Homeland War had been defensive in nature and was a just and legitimate war. Yet crimes w ­ ere undoubtedly committed by Croat perpetrators before, during, and a­ fter Operation Storm, the August  1995 military operation that defeated the Croatian Serbs’ self-­proclaimed republic and led to the departure or expulsion of the vast majority of Croatian Serbs. Besides the obvious observation that the Croatian government was perfectly happy to accept the ICTY’s version of history if it confirmed Croat nationalist views, both reactions demonstrated a complete misunderstanding of the purpose of the trial. ­After all, all accused at the ICTY are on trial for their individual criminal responsibility, and the judgments in their cases speak to ­whether they and the forces ­under their command committed crimes, and not to the legitimacy of the war as such. The common expectation among the general public and, in par­tic­u­lar, the ruling elites is that the tribunal must validate “the truth,” which can only be the chaste victimhood of the in-­group relative to its enemies and aggressors.35 As Ana Mijić has observed, “the Tribunal can also be regarded as an arena of b­ attle for the preservation of the ‘own’ identity and the ‘own’ perception of real­ity.”36 In



Collective and Competitive Victimhood 187

this vein, the current conservative Croatian government has even gone so far as to announce that it ­will file an amicus brief in the case of the Bosnian Croat leadership (Prlić et al.), b­ ecause the judgment in this case demonstrated the extensive involvement of the leadership of the Croatian state in the war in Bosnia and Herzegovina.37 The most pronounced form of competitive victimhood in transitional justice manifests itself with re­spect to genocide, generating what I have elsewhere termed a “myopic focus” on the genocide charge.38 This has had problematic consequences for investigations and prosecutions at the ICTY, as well as subsequently in the reception of the verdicts of the ICTY in the former Yugo­slavia.39 Moreover, the obsession with the genocide charge as the perceived gold medal in the “victim Olympics” also propelled Bosnia and Herzegovina, Croatia and Serbia to pursue rulings of genocide at the International Court of Justice. While the case filed by Bosnia and Herzegovina had considerable merit, the mutually contesting lawsuits filed by Croatia and Serbia had much more to do with politics than with the crimes that had been committed during the war in Croatia between 1991 and 1995.40 Paradoxically, both the groups purporting to represent the interests of victim communities and t­hose that support the accused perpetrators often end up agreeing that convictions for “only” crimes against humanity or war crimes are tantamount to acquittals. For example, in March 2016, the former Bosnian Serb president Radovan Karadzic was convicted of crimes against humanity and war crimes for crimes committed in Bosnia and Herzegovina in 1992 and 1995, but was found guilty of genocide only for Srebrenica in 1995, and not for municipalities where other crimes ­were committed in 1992. In an official statement, the Congress of the Islamic Community in Bosnia and Herzegovina declared that it was “incomprehensible” that the tribunal had not reached the “only reasonable conclusion” that genocide had been committed in 1992. Hence, the tribunal “has not fulfilled its mission of establishing the truth and justice for the crimes of genocide and other victims.”41 The journalist Nidžara Ahmetašević contrasted the Congress’s outrage with the lack of a genocide conviction for 1992 with the Congress’s apparent lack of condemnation for certain crimes committed by foreign Muslim combatants during the war and for the fate of ­those Muslims who ­were victims of the war and currently continue to live in penury. Meanwhile, in Serbia, while some impor­tant steps have been taken to accept responsibility for Serbia’s extensive involvement in the war in Bosnia and Herzegovina, the Serbian state has insisted upon formulations that tend to emphasize that all sides committed atrocities during the war, that the conflict was primarily a civil war, and that genocide was not committed. All of t­hese postulates are belied by repeated judgments of the ICTY. Such considerations ­were in the foreground when the Serbian parliament debated and fi­nally in 2010 succeeded in passing a declaration on Srebrenica.42 And Prime Minister Aleksandar Vučić has

188

c hristi a n a x boe nielsen

subsequently suggested that ­there should be a common day of commemoration for all victims of the wars of Yugo­slav succession. This notion has been brusquely dismissed by non-­Serb politicians in the region. On the other side of Bosnian politics, the president of Republika Srpska in Bosnia and Herzegovina, Milorad Dodik, continues to specialize in competitive victimhood. In Dodik’s worldview, any criticism of his rule or Republika Srpska is immediately equated with an attack on the Serb nation in Bosnia and Herzegovina. In March 2016, on the eve of the verdict in the Karadzic case, Dodik attended the renaming of a student dormitory a­ fter Karadzic.43 When Zdravko Tolimir, who had been convicted of genocide, war crimes, and crimes against humanity at the ICTY, passed away in the ICTY Detention Unit, Dodik also did not miss the opportunity to mourn Tolimir as a “patriot.”44 Dodik’s attitude ­toward the ICTY should perhaps not seem surprising, but it is worth noting that in the mid-1990s Dodik was among the rare Bosnian Serb politicians who called for Radovan Karadzic and Ratko Mladić to turn themselves in and stand trial at the ICTY. And for many years the Ministry of Internal Affairs of Republika Srpska cooperated closely with the ICTY in tracking and apprehending fugitive indictees. ­There can be ­little doubt that Dodik dislikes the ICTY, but as with many politicians in the former Yugo­slavia, his antipathy ­toward the tribunal is not fundamentally about its successes or failures as an institution of international criminal justice. Rather, competitive victimhood is a way of equating the po­liti­cal leadership of a po­liti­cal entity with the nation’s suffering, thereby inoculating the po­liti­cal leadership against criticism. By fostering a constant siege mentality, in which the slightest concession or lapse w ­ ill lead to imminent destruction of the nation-­state (or, in Dodik’s case, a pseudo-­state), critics from the ingroup can be accused of treachery, while the arguments of external critics can be blithely ignored or parried in order to increase the cohesion of the ingroup. Maintaining a competition among rival victimhoods is hence instrumental to fending off attempts to criticize or dismantle pervasive clientelism and state capture in the former Yugo­slavia. And t­ here is an additional advantage to mobilizing the (dead) victims of the past in order to legitimate the rule of a par­tic­u­lar politician or regime: the dead do not speak and therefore cannot protest their instrumentalization.45 Competitive victimhood in the former Yugo­slavia routinely careens into a chain reaction, in which the vituperative claims of one politician unleash a series of counterclaims from the opposing side. In the spring of 2016, Franjo Komarica, who is the Roman Catholic bishop of Banja Luka and also the president of the Bishops’ Conference of Bosnia and Herzegovina, attended the Bleiburg commemoration in Austria. In his speech, Komarica claimed that Bosnian Croats in Banja Luka in the 1990s had become the victims of a “new Bleiburg.” Dodik immediately reacted by claiming that Komarica had insulted the victims of Ustasha crimes, naming several notorious massacres in the Banja Luka area during the Second World War.46



Collective and Competitive Victimhood 189

Fascinatingly and problematically, this phenomenon of competitive victimhood extends even to intra-­national rivalries. Among Bosniaks, the verdicts of the ICTY have contributed to the creation of a kind of hierarchy between the victims of Srebrenica and ­those from other municipalities. As the massacres that ensued ­after the fall of Srebrenica in July 1995 are the only crimes to have been pronounced to constitute genocide in the Bosnian war, an at times provocative (self-)perception exists that Srebrenica victims somehow both domestically and internationally enjoy a higher status than other victims.

­toward a critical approach to victimhood It is impor­tant to note again that the phenomena described ­here are by no means restricted to the Balkans. Psychologists conducting research on competitive victimhood and collective victimization in the United States have observed that “some authors have argued that the United States has in recent years become a nation of victims, with members of all groups seeking the title of victimhood.”47 Recent public controversies concerning the display of the Confederate b­ attle flag and the removal of Confederate monuments have also shown how the United States, a country generally considered to be living in the pres­ent and the f­ uture, can at least periodically become consumed by historical wrongs.48 And any suggestions that President Barack Obama might have apologized for past crimes committed by the United States of Amer­ic­ a brought with them accusations of the abandonment of American exceptionalism. Irish history, M ­ iddle Eastern history, and East Asian history are other arenas where competitive victimhood also plays a power­ful con­temporary role.49 As Daniel ­Sullivan and his co-­authors argue, “Taken together, t­ hese [cases can] suggest that competitive victimhood is contingent on the threat implied by in-­group responsibility for outgroup victimization and does not follow from information suggesting that the outgroup itself or a third group is responsible for the outgroup’s victimization.”50 Yet at the same time, few social psychological studies have been undertaken to explore how the circle of competitive victimhood can be broken.51 It may seem to some that if past wrongs and the pro­cess of international criminal justice provide so much fuel for opportunistic politicians in the former Yugo­slavia, then it would be better to impose a moratorium on discussions of the past and to avoid further war crimes ­trials. Yet taking responsibility for the past is part and parcel of asserting accountability in the pres­ent. Socie­ties that shirk confrontation with unpleasant, traumatic, and violent episodes in the past are much less likely to enjoy leadership that ­w ill recognize and tackle serious economic, social, and po­liti­cal challenges in the pres­ent. This is why I disagree with David Rieff, who has argued that “historical memory is rarely as hospitable to peace and reconciliation as it is to grudge-­keeping, dueling martyrologies and enduring enmity.”52 The prob­lem is not so much that p­ eople seem predisposed

190

c hristi a n a x boe nielsen

to remembering past grievances and nurturing or exploiting ­these, but that they approach narratives of victimhood uncritically. This seems particularly true in times of crisis. If victim identity is the primary identity marker, particularly collectively, then prob­lems are bound to emerge and multiply. In conclusion, it seems that, regardless of the imperfections and justified criticisms of the vari­ous verdicts of the ICTY, the tribunal’s work cannot be cognitively assimilated in the former Yugo­slavia in the absence of a paradigmatic shift away from collective victimhood. Entire educational systems in the region are hardwired to produce the uncritical reiteration of discourses of collective and competitive victimhood. ­Until this changes, l­ittle e­ lse ­will. Unfortunately, ­there is meager incentive for this change to occur, ­because politicians of all stripes thrive on an uncritical public whose dissatisfaction with chronic social and economic prob­lems can be easily parried by mobilizing voters based on real and perceived grievances against the Other.

notes 1. ​V. Crnjanski Spasojević and M. Veljković, “Đeneral Draža nije zločinac,” Večernje novosti,

May  14, 2015, http://­www​.­novosti​.­rs​/­vesti​/­naslovna​/­drustvo​/­aktuelno​.­290​.­html:547976​ -­Rehabilitovan​-­Draza​-­Mihailovic. 2. ​For a summary of historical, l­egal, and ethical arguments against rehabilitation of figures from the Second World War, see Srđan Milošević, Istorija pred sudom: Interpretacija prošlosti i pravni aspekti u rehabilitaciji Kneza Pavla Karađorđevića (Belgrade: Edicija Reč, 2013). 3. ​Jelena Lovrić, “Merkel ili Orban: Tko je uzor Predsjednici?,” Jutarnji list, September 20, 2015, http://­www​.­jutarnji​.­hr​/­svicarci​-­i​-­izbjeglice​-u­ mjesto​-­partizana​-­i​-u­ stasa​/­1420917​/­. 4. ​Christian Axboe Nielsen, “Surmounting the Myopic Focus on Genocide: The Case of the War in Bosnia and Herzegovina,” Journal of Genocide Research 15, no. 1 (2013): 23–24. For a penetrating case study of the phenomenon of past, pres­ent, and ­future vio­lence as agent, see Max Bergholz, Vio­lence as a Generative Force: Identity, Nationalism and Memory in a Balkan Community (Ithaca, N.Y.: Cornell University Press, 2016). 5. ​Elissa Helms, Innocence and Victimhood: Gender, Nation, and ­Women’s Activism in Postwar Bosnia-­Herzegovina (Madison: University of Wisconsin Press, 2013), 5. 6. ​Fuyuki Kurasawa, “A Message in a ­Bottle: Bearing Witness as a Mode of Transnational Practice,” Theory, Culture & Society 26, no. 1 (2009): 92–111; Michael Humphrey, “From Victim to Victimhood: Truth Commissions and T ­ rials as Rituals of Po­liti­cal Transition and Individual Healing,” Australian Journal of Anthropology 14, no.  2 (2003): 171–187; Catherine  M. Cole, “Per­for­mance, Transitional Justice and the Law: South Africa’s Truth and Reconciliation Commission,” Theatre Journal 59, no. 2 (2007): 167–187. 7. ​Lawrence Douglas, The Memory of Judgment: Making Law and History in the T ­ rials of the Holocaust (New Haven, Conn.: Yale University Press, 2001). 8. ​Peter Novick, The Holocaust in American Life (New York: Houghton Mifflin, 2000). 9. ​Refik Hodžić, “Living the Legacy of Mass Atrocities: Victims’ Perspectives on War Crimes ­Trials,” Journal of International Criminal Justice 8, no. 1 (2010): 113–136. 10. ​For a comparison of the effects of the Eichmann trial and ­those of the ICTY in Bosnia, see Isabelle Delpla, “In the Midst of Injustice: The ICTY form the Perspective of Some Victim Associations,” in The New Bosnian Mosaic: Identities, Memories and Moral Claims in a Post-­War



Collective and Competitive Victimhood 191

Society, ed. Xavier Bougarel, Elissa Helms and Gerlachus Duijzings (Aldershot: Ashgate, 2007), 231–232. 11. ​For the state of the art on the definition and structure of crimes against humanity, war crimes, and genocide, see International Criminal Court, Ele­ments of Crimes, 2011, https://­ www​.­icc​-­cpi​.­int​/­NR​/­rdonlyres​/­336923D8​-­A6AD​-­40EC​-­AD7B​-­45BF9DE73D56​/­0​/­Elements​ OfCrimesEng​.­pdf. 12. ​For a thorough discussion of paradigms about victims and the ethics politics of their stories, see Diana Tietjens Meyers, Victims’ Stories and the Advancement of H ­ uman Rights (Oxford: Oxford University Press, 2016). 13. ​Karl Jaspers, Die Schuldfrage: Ein Beitrag zur deutschen Frage (Zu­rich: Artemis Verlag, 1946). 14. ​Particularly in Poland, where Poles suffered enormously ­under both German and Soviet occupation, conservative politicians have angrily rejected any suggestions of Polish culpability for any crimes committed in the twentieth c­ entury. See, for example, Timothy Snyder, “Poland vs. History,” New York Review of Books, May 3, 2016, http://­www​.­nybooks​.­com​/­daily​ /­2016​/0­ 5​/0­ 3​/p­ oland​-­vs​-­history​-­museum​-­gdansk​/­. 15. ​Leiden University, “4.1 Million for Study on Dutch East Indies War of Decolonisation,” March  2, 2017, https://­www​.­universiteitleiden​.­nl​/­en​/­news​/­2017​/­02​/­4​.­1​-­million​-­for​-­study​ -­on​-d­ ecolonisationJoint. 16. ​Johanna Ray Vollhardt and Rezarta Bilali, “The Role of Inclusive and Exclusive Victim Consciousness in Predicting Intergroup Attitudes,” Po­liti­cal Psy­chol­ogy 36, no. 5 (2015): 489–506. 17. ​See Pablo de Greiff, “Report of the Special Rapporteur on the Promotion of Truth, Justice, Reparation and Guarantees of Non-­Recurrence,” H ­ uman Rights Council, Thirtieth Session, September 4, 2015. 18. ​While ­there is no reason to depict the Ottoman Empire as a multicultural Disneyland, it is equally problematic to reduce the Ottoman experience to despotism, forced conversions, and repression. On the particularly contested nature of Bosnia and Herzegovina, see Edin Hajdarpasic, Whose Bosnia? Nationalism and Po­liti­cal Imagination in the Balkans, 1840–1914 (Ithaca, N.Y.: Cornell University Press, 2015). 19. ​Mark Mazower, The Balkans: A Short History (New York: Random House, 2000), xxxvii. 20. ​Christian Axboe Nielsen, Making Yugo­slavs: Identity in King Aleksandar’s Yugo­slavia (Toronto: University of Toronto Press, 2014). 21. ​For a succinct overview of the debates on the numbers and identity of victims in Yugo­slavia during the Second World War, see Jozo Tomasevich, War and Revolution in Yugo­slavia, 1941– 1945: Occupation and Collaboration (Palo Alto, Calif.: Stanford University Press, 2001), 718–750. 22. ​Ivo Banac, “Historiography of the Countries of Eastern Eu­rope: Yugo­slavia,” American Historical Review 97, no. 4 (1992): 1084–1104; Mate Nikola Tokić, “Landscapes of Conflict: Unity and Disunity in Post–­Second World War Croatian Émigré Separatism,” Eu­ro­pean Review of History 16, no. 5 (October 2009): 739–753. 23. ​Jasna Dragović-­Soso, “Saviours of the Nation”: Serbia’s Intellectual Opposition and the Revival of Nationalism (Montreal: McGill-­Queen’s University Press, 2002); Jasna Dragović-­Soso, “Rethinking Yugo­slavia: Serbian Intellectuals and the ‘National Question’ in Historical Perspective,” Con­temporary Eu­ro­pean History 13, no. 2 (2004): 170–184. 24. ​For example, on the evolution of Bosnian Muslim identity during this period, see Iva Lučić, “In the Ser­vice of the Nation: Intellectuals’ Articulation of the Muslim National Identity,” Nationalities Papers 40, no. 1 (2012): 23–44. 25. ​Andrew Wachtel, Making a Nation, Breaking a Nation: Lit­er­a­ture and Cultural Politics in Yugo­slavia (Stanford, Calif.: Stanford University Press, 1998), 173–189. 26. ​Kate Meehan Pedrotty, “Yugo­slav Unity and Olympic Ideology at the 1984 Sarajevo Winter Games,” in Yugo­slavia’s Sunny Side: A History of Tourism in Socialism (1950s–1980s), ed.

192

c hristi a n a x boe nielsen

Hannes Grandits and Karin Taylor (Budapest: Central Eu­ro­pean University Press, 2010), 335–367. 27. ​Julie A. Mertus, Kosovo: How Myths and Truths Started a War (Berkeley: University of California Press, 1999). 28. ​http://­www​.n ­ juz​.­net​/­u​-­srbiji​-­obelezen​-­dan​-­fraze​-­a​-­sta​-­su​-­oni​-­nama​-­radili​/­. 29. ​Dubravka Stojanović, “Vinovnici i žrtve u srpskim udžbenicima istorije,” Buka, September 20, 2013. 30. ​Within the former Yugo­slavia, the most well-­known initiative was the series of meetings between Croat and Serb historians or­ga­nized by the Friedrich Ebert Stiftung. ­These meetings led to a series of publications with the title Dijalog povjesničara/istoričara. See also the “Historija, istorija povijest” proj­ect of the Eu­ro­pean Association of History Educators (EURO CLIO) at http://­euroclio​.­eu​/­2016​/­06​/­member​-­project​-­historija​-­istorija​-­povijest​-­teacher​ -­training​-­zagreb​/­. Internationally, a mammoth but problematic effort resulted in the publication of Charles W. Ingrao and Thomas A. Emmert, eds., Confronting the Yugo­slav Controversies: A Scholars’ Initiative (West Lafayette, Ind.: Purdue University Press, 2009). 31. ​Ana Mijić, Verletzte Identitäten: Der Kampf um den Opferstatus im Bosnisch-­Herzegowinischen Nachkrieg (Frankfurt: Campus, 2014), 22. On the disproportionate focus on Serb victims in Serbian textbooks, see Sandra Obradović, “­Don’t Forget to Remember: Collective Memory of the Yugo­slav Wars in Present-­Day Serbia,” Peace and Conflict: Journal of Peace Psy­chol­ogy 22, no. 1 (2016): 14–15. 32. ​Jelena Subotić, lecture, University of North Carolina Chapel Hill, Center for Eu­ro­pean Studies, November  22, 2013, https://­www​.­youtube​.­com​/­watch​?­v​=5­ fCdqVDfAu0&index​ =­1&list​=­FLWzIDix4wIzG8D0QXaJ4p2w. See also Jelena Subotić, “Stories States Tell: Identity, Narrative, and ­Human Rights in the Balkans,” Slavic Review 72, no. 2 (Summer 2013): 306–326. 33. ​Together with a group of researchers from vari­ous countries, I attended the official commemoration of Bleiburg in Austria on May 14, 2016. Footage of the commemoration is available at https://­www​.y­ outube​.­com​/­playlist​?­list​=­PL5r7KviIkuKkXaOYqaqmyir3T9jes​_­a2X. 34. ​Vjeran Pavlaković, “Croatia, the International Criminal Tribunal for the Former Yugo­ slavia, and General Gotovina as a Po­liti­cal Symbol,” Europe-­Asia Studies 62, no.  10 (2010): 1707–1740. 35. ​On the gendering of victimhood in narratives about the war in Bosnia and Herzegovina, see Helms, Innocence and Victimhood. 36. ​Mijić, Verletzte Identitäten, 31. 37. ​Interestingly, the former Croatian minister of justice, Ante Šprlje, seemed oblivious to the possibility that filing such a claim would only implicitly prove the depth of Croatia’s involvement in the war in Bosnia and Herzegovina. Goran Jungvirth, “Slučaj Prlić i drugi: Hrvatska želi braniti Tuđmana u Hagu,” Radio Slobodna Evropa, March 22, 2016, http://­www​.­slobodnaevropa​ .­org​/­content​/­slucaj​-­prlic​-­i​-­drugi​-­hrvatska​-­zeli​-b­ raniti​-­tudjmana​-­u​-­hagu​/­27628440​.­html. 38. ​Nielsen, “Surmounting the Myopic Focus on Genocide,” 21–39. 39. ​See Timothy ­Waters, The Milošević Trial: An Autopsy (Oxford: Oxford University Press, 2014). 40. ​It is worth noting that the ­legal team representing Serbia also appealed to the symbolic standard of the Holocaust by comparing the Wannsee Conference, which historians consider to be the definitive meeting initiating the “Final Solution of the Jewish Question” in Eu­rope, with a meeting of the Croatian leadership prior to Operation Storm in August 1995. While t­ here can be no doubt that war crimes and crimes against humanity ­were committed by Croatian forces against Croatian Serbs during and a­ fter Operation Storm, this par­tic­ul­ar comparison is very disproportional. International Court of Justice, Verbatim Rec­ord of Public Sitting, the Hague March 14, 2014, 54.



Collective and Competitive Victimhood 193

41. ​Congress of the Islamic Community, quoted in Nidžara Ahmetašević, “Sabor Islamske

zajednice protiv haškog tribunala,” Buka, May 9, 2016, http://­www​.­6yka​.­com​/­novost​/­104901​ /­sabor​-­islamske​-­zajednice​-­protiv​-­haskog​-­tribunala. 42. ​Jasna Dragović-­Soso, “Apologising for Srebrenica: The Declaration of the Serbian Parliament, the Eu­ro­pean Union and the Politics of Compromise,” East Eu­ro­pean Politics 28, no. 2 (2012): 163–179. 43. ​M.L.D., “Dodik na Palama otvorio studentski dom nazvan po Radovanu Karadžiću,” Nezavisne novine, March 20, 2016, http://­www​.­nezavisne​.c­ om​/­novosti​/­drustvo​/­Dodik​-n­ a​-­Palama​ -­otvorio​-s­ tudentski​-d­ om​-­nazvan​-­po​-­R adovanu​-K ­ aradzicu​/­360529. 44. ​“Dodik povodom smrti generala: Zdravko Tolimir je bio patriota, ovo je veliki gubitak za srpski narod,” Kurir, February  9, 2016, http://­www​.­kurir​.­rs​/­crna​-­hronika​/­dodik​-­povodom​ -­smrti​-­generala​-­zdravko​-­tolimir​-­je​-­bio​-­patriota​-­ovo​-­je​-­veliki​-­gubitak​-­za​-­srpski​-­narod​-­clanak​ -­2128251. 45. ​Katherine Verdery, The Po­liti­cal Lives of Dead Bodies: Reburial and Postsocialist Change (New York: Columbia University Press, 1999). 46. ​“Reagovanja iz RS na izjavu biskupa Komarice da je Banjaluka Blajburg,” Buka, May 18, 2016, http://­www​.6­ yka​.­com​/­novost​/­105407​/­reagovanja​-­iz​-r­ s​-­na​-­izjavu​-­biskupa​-­komarice​-d­ a​ -­je​-­banjaluka​-­blajburg. 47. ​Daniel S ­ ullivan, Mark  J. Landau, Nyla  R. Branscombe and Zachary  K. Rothschild, “Competitive Victimhood as a Response to Accusations of Ingroup Harm ­Doing,” Journal of Personality and Social Psy­chol­ogy 102, no. 4 (2012): 778. 48. ​Dell Upton, What Can and C ­ an’t Be Said: Race, Uplift, and Monument Building in the Con­ temporary South (New Haven, Conn.: Yale University Press, 2015). 49. ​See, for example, Carol Gluck, “Operations of Memory: ‘Comfort ­Women’ and the World,” in Ruptured Histories: War, Memory, and the Post-­Cold War in Asia, ed. Sheila Miyoshi Jager and Rana Mitter (Cambridge, Mass.: Harvard University Press, 2007), 47–77. 50. ​­Sullivan et al., “Competitive Victimhood,” 792. 51. ​Luca Andrighetto, Silvia Mari, Chiara Volpato, and Burim Behluli, “Reducing Competitive Victimhood in Kosovo: The Role of Extended Contact and Common Ingroup Identity,” Po­liti­cal Psy­chol­ogy 33, no. 4 (2012): 513–529. 52. ​David Rieff, Against Remembrance (Melbourne: Melbourne University Press, 2011), 55–56.

9 • PERPETR ATOR-­V ICTIMS How Universal Victimhood in Cambodia Impacts Transitional Justice Mea­sures T I M OT H Y W I L L I A M S

In Cambodia, forty years ­after the seizure of power by the Khmer Rouge, across wide parts of society, t­here is a consensus that almost every­one was a victim of the regime. This narrative sees only a few individuals from the highest echelons of power as “perpetrators” bearing responsibility for the entire range of atrocities, while every­one ­else is conveyed the status of “victim.” Surprisingly, this categorization of victimhood is also assigned to most former cadres of the Khmer Rouge, who implemented the genocidal policies of the regime and upheld the system of oppression. It is within this societal context that the hybrid tribunal of the Extraordinary Chambers in the Courts of Cambodia (ECCC) is trying only the highest leaders—­those deemed “most responsible” for the crimes of the Khmer Rouge—­lending credence to this societal perception of mass victimhood and very narrow perpetrator responsibility. This chapter discusses how the societal narrative of universal victimhood gained traction and how this was assisted by the fact that cadres w ­ ere often stationed away from home, so that their victims did not know them, and the dynamics of a transition a­ fter the regime that allowed a fresh start for most ­people. Further, it points to how the horrific experiences of the cadres themselves allowed them to perceive themselves as victims. Fi­nally, this chapter explores the interaction between ­these societal ascriptions of victimhood and perpetration on the one hand, and the ECCC and other forms of dealing with the past on the other, arguing that they mutually reinforce each other and are central to the stable and harmonious community relations that exist ­today.

194



Perpetrator-­Victims 195

introduction On November 23, 2016, the Supreme Court Chamber of the ECCC announced its verdict, which upheld large parts of the judgment in cases against Nuon Chea and Khieu Samphân, the Khmer Rouge’s former ideologue-­in-­chief and head of state, respectively.1 With this verdict they became the second and third ­people ­after Kaing Guek Eav, better known as “Duch” (the head of the infamous S-21 prison), to have been brought to justice by this hybrid tribunal. The fact that only three p­ eople have been convicted by this international tribunal for crimes committed during Demo­cratic Kampuchea, as the regime of the Khmer Rouge between 1975 and 1979 was called, is surprising, given that between 1.7 and 2.2 million ­people of a population of 7.9 million died, half of them from vio­lence, half from starvation, overwork, and related diseases and infections.2 Compared with other cases such as the 1994 genocide in Rwanda or the Holocaust, three judicial sentences are not very many. However, within the societal discourse on perpetration and victimhood that has prevailed since the end of Demo­cratic Kampuchea, t­ hese numbers are more fathomable, as t­ here is a broadly accepted understanding that most p­ eople ­under the rule of the Khmer Rouge w ­ ere victims, including many of the low-­and middle-­ranking cadres of the Khmer Rouge themselves. The notion that most Cambodians ­were victims has manifest implications for the country’s memory of the past and for how vari­ous actors choose to deal with this memory. Consequently, this chapter does not focus on perpetrator-­ victim dichotomies, which are salient in many postconflict communities, or on ideas of a hierarchy of victimhood; rather, it considers the presence of nearly universal victimhood in post-­genocide Cambodia. Who is classified as a perpetrator, who can legitimately claim to be a victim, and how ­these groups interact with each other have tangible consequences for social relations in postconflict socie­ties. Furthermore, po­liti­cal framing can foster, counteract, or modulate t­ hese ascriptions, particularly in the context of transitional justice programs, which can construct an authoritative narrative of the past. This narrative and how it overlaps with former cadres’ own perceptions, on the one hand, and with transitional justice programs in Cambodia on the other, are the focus of this chapter. It is based on seven months of field work conducted in 2014, 2015, and 2017 in eleven provinces across Cambodia. In the course of this field research, I conducted in-­depth interviews with fifty-­eight former cadres of the Khmer Rouge. In addition, this study draws on conversations held with vari­ ous actors in the field of transitional justice, including local and international staff at the ECCC and in relevant NGOs, and is complemented by in-­depth analy­sis of vari­ous sites of memory. The chapter begins by briefly presenting vari­ous ways that Cambodia has approached its past. Then, ­after illustrating how low-­level former cadres of the Khmer Rouge understand themselves as victims of the Khmer Rouge regime, it

196

ti m othy w illi a ms

examines the reasons for the widespread ac­cep­tance of this ascription, as well as the fissures in the dominant narrative that can emerge at the local level. Fi­nally, this chapter discusses how this societal narrative is impacted by and itself impacts transitional justice pro­cesses in Cambodia.

remembering and dealing with the past in cambodia In the years since 1979, when Demo­cratic Kampuchea ended, vari­ous pro­cesses of dealing with the past have developed. H ­ ere I w ­ ill ­here very briefly sketch the main memory interventions, including memorials, judicial pro­cesses, reparations, civil society engagement, and education, as well as other more informal forms of memory. First, memorials play an impor­tant part in the government narrative of dealing with the past. Almost immediately ­after liberating the country, the new government sought to legitimate its presence by opening up S-21, a former security center, as the Tuol Sleng Genocide Museum. It is what Patrizia Violi refers to as a “trauma site museum,” in which the visitor is in the very space where the atrocities occurred and which can produce a “deeply disturbing experience.”3 The core exhibition has remained relatively fixed in both aims and ­actual content, focusing mostly on a shocking pre­sen­ta­tion of the space as it was found by the Viet­nam­ese in 1979.4 While t­ here have been some changes to the memorial, the key messages have remained the same over time,5 including a story of national suffering, rather than individual victimhood.6 Similarly, Choeung Ek, the fields in which most of the prisoners of Tuol Sleng ­were killed, has been experienced by visitors as a “profoundly disturbing place that evoked difficult and complex emotions.”7 Particularly harrowing in both spaces are the pre­sen­ta­tions of skulls and bones, which are controversial in that Cambodian Buddhist tradition the dead are supposed to be cremated to set the spirit ­free to be reincarnated.8 This is a good example of the challenge of designing ­these memorials for both a local and an international audience, with dif­fer­ ent aims connected to both.9 ­These emotional encounters provide visitors with a sense of being “dutiful tourists,”10 who feel that their visit makes them “better global citizens.”11 With the exception of a show trial against Pol Pot and Ieng Sary in absentia, which was conducted directly a­ fter the fall of the regime, for many years t­ here was no judicial attempt to confront the atrocities of Demo­cratic Kampuchea. In the pro­cess of Western involvement in the country, the idea of a tribunal matured.12 The ECCC is the result of extensive negotiations between the United Nations and the Cambodian government, which have been criticized as having involved a politicization of the justice pro­cess.13 Broadly, Alexander Laban Hinton emphasizes that in the global “Transitional Justice imaginary” the ECCC is central to the idea that the transitional justice pro­cess can constitute a form of societal heal-



Perpetrator-­Victims 197

ing.14 While the tribunal can be seen as writing a definitive version of history and rendering justice for Cambodian society, reparations are a particularly salient feature of how its work reaches beyond the walls of the court itself. The court did not make much use of the possibility of awarding reparations in case 001 against Duch,15 but several meaningful reparations ­were included in the verdict of case 002/1 and 002/2, both against Nuon Chea and Khieu Samphân. ­These reparations proj­ects have and w ­ ill include mobile exhibitions on the evacuation of Phnom Penh and on forced marriages, a play on civil courage called “The Courageous Turtle,” a traditional apsara dance per­for­mance on forced marriage called “Phka Sla,” as well as vari­ous forms of reconciliation proj­ects, testimonial therapy, and a memorial at S-21 (which ­will be discussed in more detail below). In the wake of the ECCC’s establishment, many NGOs have been focusing on the Khmer Rouge past and developing a myriad of programs, including documenting and monitoring the court, engaging in outreach work that the tribunal has failed to do itself, facilitating victim participation at the ECCC, psychosocial support, and developing and implementing vari­ous reparations proj­ects.16 Furthermore, using the societal momentum generated by the court, school education on the period of Demo­cratic Kampuchea has been introduced, and new efforts are underway to make this knowledge digitally accessible and appealing to young ­people.17 Even as official memorialization is predominantly tied to Tuol Sleng and Choeung Ek, and most transitional justice proj­ects focus on ­these and the ECCC, a number of the sites of forced l­abor, security centers, and mass graves that many p­ eople witnessed or survived during the regime of Demo­cratic Kampuchea are left unmarked; in this way the official narrative privileges t­ hose victims who ­were killed over ­those who endured many other horrific experiences.18 Nonetheless, vari­ous culturally-­specific forms of remembering and honoring the dead are practiced throughout the country, albeit not always tied into the more officially sanctioned forms discussed above.19 While Cambodia appears to have a bustling and active memoryscape, it is impor­tant to emphasize that (besides Tuol Sleng and Choeung Ek), most of ­these proj­ects have emerged since the founding of the court and are predominantly victim-­centric; as such, they are largely accepted by the government as long as they do not run ­counter to their narrative of the past. Yet, they also coexist with a marked desire by the government for collective amnesia exemplified by Prime Minister Hun Sen’s plea at a press conference in 1998 to “dig a hole and bury the past.”20

198

ti m othy w illi a ms

former khmer rouge cadres’ own narratives of victimhood Given ­these vari­ous ways in which Cambodians have been dealing with the past, how do vari­ous former cadres of the Khmer Rouge remember and interpret their own actions during Demo­cratic Kampuchea? To address this question, I have focused on low-­level cadres who served in local militias, as guards and interrogators at security centers, and as communal chiefs and other low-­level leaders who deci­ded the fate of t­ hose u­ nder their control. Th ­ ese foot-­soldiers of the system, who implemented the policies of the Khmer Rouge leadership, all had a role in the genocide. Nevertheless, in interviews ­these former cadres seldom spoke of themselves as perpetrators, but instead portrayed themselves—­either explic­itly or implicitly—as victims of the Khmer Rouge. ­These individuals’ claims to victimhood w ­ ere based on what they perceived as a common experience of victimhood during Demo­cratic Kampuchea. They appealed to the fact that during the regime they, like most other p­ eople in the population, suffered greatly. Almost all of them spoke of receiving terrible food and always being hungry.21 As one former Khmer Rouge soldier remarked, “We could not eat enough food. . . . ​[All] we ate [was rice] porridge and we worked so hard. The other [hardships endured u­ nder the regime] did not r­ eally m ­ atter as much. The food was the most difficult ­thing. Th ­ ere was not enough food.”22 In addition, cadres mentioned being constantly tired and overworked.23 “It was difficult,” a former S-21 prison guard said, “­because what­ever they assigned us to do, we had to finish it. They did not care about how hard it was and how tired we ­were.”24 As a regime with a Communist ideology, the Khmer Rouge emphasized collectivization, but as the Khmer Rouge cadres see it, the pro­cess of collectivization caused them ­great suffering as well. For example, a former guard at S-21 prison lamented that “it was not good ­under the Khmer Rouge ­because we ­were living ­under the collective: we worked collectively, we lived collectively, and even our property and clothes w ­ ere collective.”25 Furthermore, almost every­one in Cambodia, including Khmer Rouge cadres, lost f­amily members during this period—­both ­those who ­were arrested and killed and ­those who died from the terrible conditions.26 Fi­nally, former cadres claimed that they ­were continuously “fearful for their security,”27 knowing that anyone who was deemed to be associated with the former government of General Lon Nol, or who was labeled an internal e­ nemy, CIA spy, or Viet­nam­ese agent, was targeted for elimination. In the words of one former cadre who participated in the evacuation of Phnom Penh, “I was just worried about my life. When they said ‘­There is no gain for keeping [this person], no loss for weeding [them] out,’ I thought about myself and [wondered] when it would be my turn. I was worried about my security.” Any act of re­sis­tance or rebellion, no m ­ atter how small, sufficed for an individual



Perpetrator-­Victims 199

to be seen as an internal e­ nemy and thus against the Khmer Rouge revolution. As such, many cadres ­were actually killed. In other cases of genocide, this notion that someone from the perpetrator group could reasonably be considered a victim is unusual, although not unheard of. For example, Hutus who refused to participate in the killing of Tutsis in Rwanda could be killed for this refusal,28 while low-­level Ottoman soldiers could also be shot for refusing to participate in the deportations and massacres of the Armenians.29 However, in the Cambodian context, ­there was no clear demarcation between the in-­group and the out-­group, and Viet­nam­ese, Cham, Thai, and other ethnic groups w ­ ere targeted, the majority of victims w ­ ere from the majority ethnic group of the Khmer. Any ethnic Khmer individual could become a victim during Demo­cratic Kampuchea if he or she was suspected of being an internal ­enemy of the revolution, a category to which one was assigned with no evidentiary basis.30 Thus, many cadres of the Khmer Rouge, who suffered in ways that ­were similar to other members of society at that time, have subsequently felt that their claims to victimhood are legitimate. It is impor­tant to reiterate that this feeling of victimhood stems both from the horrific living conditions u­ nder which they suffered im­mensely at the time and the loss of many loved ones at the hands of the Khmer Rouge regime. ­These experiences are similar to t­ hose of the broader population. So, too, was the constant fear of being identified as an internal ­enemy and thus having to act in any way the regime commanded. The difference, however, is that whereas the broader population was forced to engage in hard l­abour, the Khmer Rouge cadres ­were forced to perpetrate many acts of brutality on behalf of the regime, and it is the compulsory nature of t­ hose acts, as well as their cruelty, that former cadres of the Khmer Rouge also point to when explaining why they see themselves as victims of the regime.

the dominant societal narrative of universal victimhood and narrow responsibility The government-­sponsored narrative about how to read the past has developed over time and manifests itself in the vari­ous memorialization proj­ects introduced above.31 This narrative stands in stark contrast to ­those of other postgenocidal socie­ties: it pins blame for the Khmer Rouge atrocities firmly on t­ hose in the highest echelons of power, while portraying all other actors, including low-­ and mid-­level cadres of the Khmer Rouge, as victims of the regime.32 Within this narrative, which pervades much of society, ­there is an understanding that in Cambodia ­under the Khmer Rouge almost every­one was a victim of the regime at the time, and almost every­one in society was acting u­ nder duress. This is not to say that individual tensions do not exist at all in local communities, but that

200

ti m othy w illi a ms

for the most part, former Khmer Rouge are accepted as victims. As such, only a few high-­ranking individuals, such as Pol Pot, Ieng Sary, Nuon Chea, and Khieu Samphân, as well as Ta Mok and Son Sen (reductively often referred to simply as “the Pol Pot–­Ieng Sary clique”) are seen as the “perpetrators,” and t­ hese few individuals bear the responsibility for the myriad of atrocities that occurred.33 This narrow ascription of responsibility is also vis­i­ble in the pres­ent colloquial usage of “the Pol Pots” to refer to Khmer Rouge cadres, which assigns responsibility for their actions directly to the highest leader. While in other cases, primary responsibility has also been pinned on the high-­level actors, such as the Nazi leadership ­after the Holocaust, or the Ittihad ­after the Armenian genocide, what makes Cambodia dif­fer­ent is that not only can the low-­level cadres deny their responsibility, but they can also define themselves as victims of the regime on equal footing to the rest of the population. I s­ hall now turn to the question of how Cambodian society has come to accept this narrative of universal victimhood and very narrow responsibility for the atrocities of Demo­cratic Kampuchea. An in­ter­est­ing approach to thinking about perpetration and victimhood is provided by Erica Bouris’s concept of complex po­liti­cal victims34 and also as adapted by Erin Baines’s concept complex po­liti­cal perpetrator.35 The idea of complex po­liti­cal actors, both perpetrators and victims, recognizes the often “choiceless decisions”36 that low-­level perpetrators are faced with during genocides, especially when the individuals have come to understand vio­lence as an expression of their po­liti­cal agency.37 In most cases of genocide, perpetrators are demonized and portrayed as monsters, but this neglects the complexity of their actions, including their own feelings of victimization as well as the acts of rescue and support for the victims in which some perpetrators engaged. In other words, a black and white picture of perpetrators and victims ignores the grey zones of agency and the social realties of mass vio­lence. What is distinct about Cambodia, however, is that the grey zones are reduced in complexity to the advantage of the low-­level cadres. Thus, rather than ­these perpetrators being demonized, their acts of perpetration are minimized; their agency is recognized as constrained to the degree that they forfeit responsibility, while their experiences of victimhood are foregrounded. The result of this simplification is that their actions are exonerated and their victimhood is affirmed. This societal attitude is exemplified through the ease with which former perpetrators live in their communities t­ oday. Of the fifty-­eight former cadres of the Khmer Rouge regime whom I interviewed, only one man spoke of hostility ­toward him in his community,38 while all ­others ­were integrated. My empirical research led me to conclude that ­there are five main reasons that ­these men and ­women are accepted not as perpetrators but as victims that should be welcomed by the rest of society. First, as mentioned above, the horrific experiences of the cadres themselves ­were comparable to t­hose of non–­Khmer Rouge individuals at the time. Insofar as the former cadres suffered from hunger and



Perpetrator-­Victims 201

exhaustion, the loss of ­family members, and the constant fear of being arrested and killed, they, too, ­were victims. Second, many Cambodians have bought into the idea of perpetrators having been forced into a “choiceless decision,”39 given that in the extremely coercive context of Demo­cratic Kampuchea very ­little re­sis­tance to ­orders was pos­si­ble.40 This was experienced by cadres of the Khmer Rouge, as well as by every­one e­ lse who had to work in hard ­labor groups and who could also have been killed for any form of re­sis­tance. According to this logic, all Cambodians ­were assigned roles that they had to fulfil, and so being forced to participate as a cadre can be seen as equivalent to being forced to participate in agricultural or factory working groups. All, thus, ­were victims of coercion. Third, the Khmer Rouge operated with a very strong division of ­labor among low-­level cadres, each of whom was assigned a very specific task, such as bringing detainees food, bringing them ­water, being in charge of the key for their locks, taking them from the holding cells to the interrogation rooms, and so on.41 ­Because their tasks ­were divided up so minutely, their responsibility for their actions was diffused. Thus, t­ hose arresting, detaining, transporting, interrogating, or torturing suspected internal enemies did not feel responsible for the ultimate fate of their victims—­they did not feel responsible for the killing that inevitably occurred at the end of a long pro­cess in which they played a necessary step. Hence, this diffusion of responsibility also makes it easier for society ­today to see ­these individuals not as perpetrators, but as fellow Cambodians forced into ­these roles. Fourth, due to the nature of hierarchy within Ângkar, the organ­ization of the Khmer Rouge, low-­level cadres could displace responsibility up the chain of command. While such displacement occurs in many cases, the nature of Khmer Rouge leadership was somewhat dif­fer­ent from that of the National Socialists in Nazi Germany, the Ittihad in the Ottoman Empire, or other po­liti­cal structures with orga­nizational membership. Instead the concept of hierarchy within Ângkar meant that when an individual spoke about Ângkar it only ever meant ­those who ­were higher than him-­or herself within the Khmer Rouge hierarchy. As a former S-21 guard explained: “Ângkar was t­ hose ­people in higher positions who gave the ­orders. Ângkar was not a small group. Anyone who was higher was called Ângkar.”42 Thus, members of the ordinary population would regard any Khmer Rouge cadre as Ângkar, but the cadre would not see him-­or herself or anyone below as Ângkar. This status was reserved for ­those higher up than oneself in the hierarchy.43 Thus, individuals themselves did not actually feel they w ­ ere part of Ângkar: they w ­ ere subject to Ângkar’s w ­ ill, but did not feel they embodied it. In this way, the authority of Ângkar could be absolute and demand unconditional obedience. Moreover, this understanding of authority also meant that members of the Khmer Rouge could relinquish responsibility to Ângkar and thus be more willing participants, b­ ecause they did not feel individual culpabil-

202

ti m othy w illi a ms

ity for their acts. In this way, the hierarchical structure that was established during Demo­cratic Kampuchea to facilitate authority and coercion has worked up to the pres­ent, with many ­people viewing the former cadres not as individually responsible, but at the mercy of ­those higher up. Fifth, and last, a­ fter recruitment, cadres w ­ ere often reassigned to locations far away from their homes and w ­ ere often split up from any friends or acquaintances with whom they had been recruited. In this way, the members of the communities in which they w ­ ere acting and the ­people they w ­ ere victimizing through their actions did not previously know them. And, even more impor­tant is that in the dynamics of transition, as the regime collapsed with the invasion of the Viet­ nam­ese, most cadres ­were able to have a fresh start when they returned to their home village. As one former commune militiaman and subsequently collective committee member explained, “When we ran, they [the Khmer Rouge] told us that the Viet­nam­ese came and would kill us all, so every­one ran. But some ­people escaped back to their home villages and t­ here w ­ ere no prob­lems, no killing. Then they went to call their relatives living in the jungle back.”44 The other members of their home communities may have known that they had been Khmer Rouge cadres, but they did not know anything about their specific crimes. Moreover, ­these returning individuals had also incurred losses ­under the Khmer Rouge, especially through the deaths of f­ amily members whom other members of the community had also known, so the attention could be focussed on the returning cadres’ victimhood. While ­these f­actors, taken together, allow for the propagation of universal victimhood and the exculpation of perpetration, it should be noted that this narrative is not without occasional fissures. The most obvious manifestation came in the form of the revenge killings that occurred in the immediate aftermath of the regime. Although l­ittle is known about t­ hese killings, my interviewees suggested that they killings occurred primarily in the localities where the cadres had been working, within a few hours of the downfall of the regime, and that they did not continue ­after the cadres left.45 A second fracture in the narrative of universal victimhood emerges when individuals still live in the communities in which they had been cadres ­under Demo­cratic Kampuchea. The only interviewee who spoke of any prob­lems in his community was not actually a full-­fledged Khmer Rouge cadre, but had aspired to be one and had taken part in arresting some victims and other such activities in his home community. As he had never been recruited formally, he was also not reassigned to another location, and ­today is confronted with the families of t­ hose he helped arrest.46 It is precisely the anomalies of this case—­namely, that he stayed in same location, that his agency was heightened ­because he volunteered to assist—­that underscore the main argument that u­ nder normal circumstances when new recruits w ­ ere assigned to other communities, it was easier for them to reintegrate themselves in their home communities and assume the status of victim ­after the regime. Insofar as t­ here



Perpetrator-­Victims 203

are tensions between former Khmer Rouge cadres and ­others in their home communities, they tend to remain below the surface, masked by the precepts of Cambodian culture against open conflict and resulting in what Laura McGrew terms a more “shallow coexistence” rather than “deep reconciliation.”47

transitional justice, dealing with the past, and universal victimhood As elaborated above, the first judicial proceedings to deal with the Khmer Rouge period ­were undertaken immediately a­ fter the regime was toppled, when the Viet­nam­ese liberators sentenced Pol Pot and Ieng Sary (both in absentia) ­under the ­People’s Revolutionary Tribunal in 1979. ­After that, however, the new Vietnamese-­installed government found it useful to proclaim nigh universal victimhood and forge social unity by reintegrating former cadres into government forces, as well as welcoming defectors from the Khmer Rouge and other civil war parties. Furthermore, ­in doing so, this was a po­liti­cally impor­tant means for legitimating the newly installed government, many members of which had been mid-­ level cadres of the Khmer Rouge in the early stages of the regime.48 Since ­today’s government includes many members with an ambiguous past, it is impor­tant to the po­liti­cal leadership that no one below the highest echelons of power during the Khmer Rouge be seen as having any responsibility for the atrocities that w ­ ere perpetrated. Indeed, present-­day Cambodia’s transitional justice mechanisms are upholding and reinforcing t­ hese narratives and are promoting the narrow societal ascriptions of perpetrator and the broad understanding of victims. This section ­will discuss two examples introduced above in more detail in order to show how this narrative manifests itself in two varied forms of dealing with the past: the ECCC’s inclusion of civil parties and the exhibition and recent memorial built at Tuol Sleng Genocide Museum, which are by far the most prominent examples of judicial and memorial practices in Cambodia. ECCC ­After a long negotiation pro­cess between the government of Cambodia and the United Nations, the ECCC was founded with the mandate of prosecuting only the most se­nior leaders of the Khmer Rouge and ­those deemed “most responsible” for the crimes committed during Demo­cratic Kampuchea. So far, only five ­people have been investigated and charged for their role in the atrocities; of ­these, only three have been convicted (Kaing Guek Eav alias Duch in case 001; Khieu Samphân and Nuon Chea in case 002/1 and ongoing in 002/2 as of February 2017), while the charges against two w ­ ere dropped due to ill health (Ieng Sary and Ieng Thirith in case 002). Although other individuals, including some slightly less high-­level cadres, have been investigated for cases 003 and 004, their cases are unlikely to move forward, b­ ecause the government has signaled that it

204

ti m othy w illi a ms

­ ill not accept any more t­ rials and that it believes t­ hese go beyond the mandate w of the ECCC. This strategy of trying only ­those considered “most responsible” for the Khmer Rouge atrocities, and the absence of any other prosecution, ­either by international or hybrid judiciaries or by local or traditional courts, such as gacaca, implicitly vindicates t­hose not tried.49 This, in turn, reinforces the narrative of general victimhood of most cadres who w ­ ere themselves subject to the regime and the very narrow perpetrator responsibility of t­ hose at the top. Furthermore, in line with an emerging orientation ­toward victims in global transitional justice pro­cesses,50 the ECCC allows victims to register as civil parties and affords them broad rights within the proceedings.51 This participation has allowed some victims to tell their stories publicly and contributed to many victims seeing the ECCC in a positive light—­even as empowering.52 But while taking part in the proceedings has been found to reduce negative emotions for victims,53 this victim inclusion has also brought with it vari­ous prob­lems related to the disjuncture between telling emotional stories and a judicial setting that necessarily focuses on evidence, causing psychological stress, mismanaged expectations, and so on.54 In the context of this chapter, it is particularly noteworthy that that some former cadres have registered as civil parties to the cases, thereby casting themselves legally as victims during the ­trials. While figures vary substantially, a reasonable estimate is that the number of civil parties with a Khmer Rouge background is about 10  ­percent of the total of 3,800.55 Moreover, Cambodian society has largely accepted their status as victims, even when former cadres ­were called to testify in court about the atrocities they had committed, precisely b­ ecause they ­were seen as victims of the regime who ­were forced ­under duress to commit ­these acts by the leaders who w ­ ere on trial. A prominent example is that of Bou Meng, one of the seven survivors of S-21. Bou Meng was a cadre of the Khmer Rouge before he was imprisoned at S-21; nonetheless, he is one of the most vocal civil parties and is deputy president of the Victims Association of Demo­cratic Kampuchea (Ksaem Ksan). Tuol Sleng Genocide Museum A second, prominent example of transitional justice mea­sures can be found in Tuol Sleng Genocide Museum, the educational and memorial site opened where S-21, Demo­cratic Kampuchea’s most infamous detention fa­cil­i­ty, had been located. In terms of “dark tourism,”56 it is the premiere attraction in Phnom Penh along with the Killing Fields at Choeung Ek and has become a major reference point for the cruelty of the Khmer Rouge regime. Although the exhibition itself does not explic­itly state this at the site, around 78 ­percent of the victims of S-21 ­were former cadres of the party who ­were deemed to be “microbes burrowing in the Party” and trying to undermine the revolution and ­were purged a­ fter 1976.57



Perpetrator-­Victims 205

As victims of the purges, they are both former Khmer Rouge and victims of the system. This fact is wholly in line with the narrative of universal victimhood. For the many Western visitors, this deconstruction of the dichotomy of perpetrator and victim could be perceived as problematic, and so the number of former Khmer Rouge cadres who w ­ ere tortured and killed at S-21 is not included in the exhibition or in the recently introduced audio guide. Instead, the audio guide simply tells listeners that more than 12,000, and possibly even as many as 20,000 ­people, ­were brought ­there, that most w ­ ere Lon Nol supporters and “new ­people” such as students, teachers, and foreigners, and that over 150 staff members from the site w ­ ere also arrested (as even they could be seen as traitors), as ­were at least 89 c­ hildren.58 While t­ hese details are not wrong, the museum consciously forwards the omission of not stating that the overwhelming majority of the prisoners ­were purged cadres of the Khmer Rouge. In turn, the curation of the museum invites anyone who is displayed on its iconic black-­and-­white photo­graphs to be seen as a victim. And b­ ecause ­there is a veritable sea of anonymous photo­graphs (names are not labeled even where they are known), any semblance of individuality is removed, and collective suffering is emphasized instead.59 At the same time, except for its focus on Duch, who ran S-21 and was mainly responsible for the atrocities committed ­there, the permanent exhibition and the audio guide remain almost ­silent on the other perpetrators; ­there are no pictures of them apart from stylized paintings by the former prisoner and artist Vann Nath, and t­ here is no text about them. Insofar as several temporary exhibitions do actually include discussions of such perpetrators of S-21, they uniformly pres­ent them as victims of the regime who lived difficult lives. For example, the exhibition “Stilled Lives: Photo­graphs from the Cambodian Genocide,” which pres­ents the stories of eleven former cadres of the Khmer Rouge, is part of a broader proj­ect including fifty-­one individuals in which victimization of cadres is writ large.60 Observing that “prison was the fate for most of this exhibition’s subjects,” and noting that a majority of ­those individuals ­were executed during the regime, the accompanying text states: “Death seemed preordained for nearly all of ­those who joined the revolution.”61 Another temporary exhibition entitled “Victims and Perpetrators? Testimony of Young Khmer Rouge Comrades” pres­ ents the stories of thirteen former cadres who had been guards at S-21 itself in similar terms, beginning the first panel with the statement: “­These photos examine a group of ­children who became victims of the Khmer Rouge regime.”62 A related development at the museum was the creation of a new memorial that was built on its grounds in March 2015, as part of a reparations proj­ect suggested by the ECCC in its verdicts. In the pro­cess of planning the memorial, a debate arose as to w ­ hether the marble panels should include the names of all ­those killed at the site or just the names of t­hose who ­were not Khmer Rouge cadres. If the latter position ­were taken, the panels would reflect only a fraction

206

ti m othy w illi a ms

figure 9.1. Personal file of a prisoner at Security Center 21 (S21), also known as Tuol Sleng, Phnom Penh, Cambodia, 2014. (Photo by Thijs Bouwknegt.)

of t­hose commonly accepted as victims of S-21, since a large majority of t­hose killed had been former Khmer Rouge cadres. This debate emerged during civil party consultations, a pro­cess by which civil parties to the case have the right to be consulted about reparations proj­ects. While most stakeholders agreed that all names should be inscribed in the panels, Youk Chhang, director of the Docu-



Perpetrator-­Victims 207

mentation Centre of Cambodia (DC-­Cam) argued against this and brought Chum Mey, president of Ksaem Ksan, on board. Ultimately, Chum Mey returned to his original position of supporting all names being inscribed, and the narrative remained as it was. The outcome is in line with the official narrative of universal victimhood, quite literally ­etching in stone the societal consensus that sees the former cadres predominantly as victims. While the debate itself demonstrates how the official narrative can—at least temporarily—be undercut by individual initiatives, such as that of Youk Chhang, ultimately t­hese interventions do not actually manifest in tangible results that ­counter the official narrative.

conclusion In this chapter I have discussed how perpetration and victimhood is ascribed in Cambodia from vari­ous perspectives, including that of the former perpetrators themselves, as well as that of society as a ­whole. Most former cadres of the Khmer Rouge view themselves as victims of the regime, and this narrative is accepted more broadly in society for a number of reasons. Like ­others throughout the country, the cadres themselves also experienced hunger, overwork, loss of ­family, and constant fear. The idea that every­one was assigned a role they did not want resonates with non-­cadres, too, so that all buy into the narrative that participation was a “choiceless decision.” Moreover, the very strong division of l­abor among low-­level cadres, as well as the overwhelming authority of Ângkar all allow responsibility to be diffused among the perpetrator group and displaced up the hierarchy, thereby distancing the cadres from culpability. Fi­nally, ­because cadres predominantly worked in other areas of the country, they ­were seldom the ­actual victimizers of their neighbors ­today, which, in turn, helps the latter to see former cadres as victims, particularly given their ongoing ­family suffering. This narrative of universal victimhood, in which only a few high-­level perpetrators ­were responsible for every­thing, is both manifest in and strengthened by the way the country remembers its past, as exemplified in the judicial work of the ECCC and the memorial work at Tuol Sleng in Cambodia. Although small fissures can emerge in this narrative of universal victimhood and although ­these should not be overemphasized, they do provide in­ter­ est­ing ave­nues for ­future research, particularly of how such momentary breaks are quickly varnished over in some fashion and the narrative of universal victimhood is reinstated, allowing for what Laura McGrew describes as vari­ous forms of coexistence, if not full reconciliation.63 For a country so deeply ravaged by de­cades of violent conflict, this coexistence in relative harmony provides a strong incentive to accept and support this narrative and to repair any fissures that appear. Yet even as the narrative of universal victimhood may have a positive societal impact ­today, the legitimating of universal victimhood and the downplaying of

208

ti m othy w illi a ms

agency can hold dangers for the f­ uture, for if vio­lence w ­ ere to return to Cambodia (for example in the form of pogroms against the Viet­nam­ese minority), individuals ­will see no reason to resist ­orders or speak out against the collective action. Rather, they ­will be able to participate in the expectation that they ­will not need to take responsibility for their actions, just as they and their compatriots did not need to do so in the Khmer Rouge past. With a discourse that pins all responsibility on the highest leaders and absolves t­ hose below them, individuals are more likely to take the path of least re­sis­tance and participate. Fi­nally, perpetration and victimhood and their close interrelationship in the Cambodian context is enlightening for broader discussions on the complexity of determining suitable transitional justice mea­sures. With such large grey zones in categorizing actors, the Cambodian case is certainly instructive as an example of one end of the continuum on how inclusive victimhood ascriptions can be framed.

notes 1. ​Extraordinary Chambers in the Courts of Cambodia, Press Office, “Supreme Court Cham-

ber Quashes Part of the Convictions, Affirms Life Imprisonment for Nuon Chea and Khieu Samphan in Case 002/01,” November 23, 2016, https://­www​.­eccc​.­gov​.­kh​/­en​/­articles​/­supreme​ -­court​-­chamber​-q­ uashes​-­part​-­convictions​-­affirms​-­life​-­imprisonment​-­nuon​-­chea​-­and​-­khie. 2. ​Ewa Tabeau and They Kheam, “Demographic Expert Report: The Khmer Rouge Victims in Cambodia, April 1975–­January 1979: A Critical Assessment of Major Estimates,” Extraordinary Chambers in the Courts of Cambodia, 2009, 19. 3. ​Patrizia Violi, “Trauma Site Museums and Politics of Memory: Tuol Sleng, Villa Grimaldi and the Bologna Ustica Museum,” Theory, Culture & Society 29, no. 1 (2012): 36–75. 4. ​Paul Williams, “Witnessing Genocide: Vigilance and Remembrance at Tuol Sleng and Choeung Ek,” Holocaust and Genocide Studies 18, no. 2 (2004): 234–254. 5. ​Stephanie Benzaquen, “Looking at the Tuol Sleng Museum of Genocidal Crimes, Cambodia, on Flickr and YouTube,” Media, Culture & Society 36, no. 6 (2014): 790–809. Caitlin Brown and Chris Millington, “The Memory of the Cambodian Genocide: The Tuol Sleng Genocide Museum,” History Compass 13, no. 2 (2015): 31–39. 6. ​Peter Manning, “Governing Memory: Justice, Reconciliation and Outreach at the Extra­ ordinary Chambers in the Courts of Cambodia,” Memory Studies 5, no. 2 (2011): 170. 7. ​Louis Bickford, Transforming a Legacy of Genocide: Pedagogy and Tourism at the Killing Fields of Choeung Ek (New York: International Center for Transitional Justice, 2009), 8. 8. ​On this controversy, see Wynne Cougill, “Buddhist Cremation Traditions for the Dead and the Need to Preserve Forensic Evidence in Cambodia,” Documentation Centre of Cambodia, http://­www​.­d.​ ­dccam​.o­ rg​/­Projects​/­Maps​/­Buddhist​_­Cremation​_­Traditions​.­htm (accessed January 31, 2017); Rachel Hughes, “Fielding Genocide: Post-1979 Cambodia and the Geopolitics of Memory,” PhD dissertation (University of Melbourne, 2006), 105–107. 9. ​Shannon Davis and Jacky Bowring, “Connecting with Tragedy through Landscapes of Memory: Memorial Design, Tourism, and the Post-­Genocide Memoryscapes of Cambodia, Rwanda, and Germany,” Memory Connection 1, no. 1 (2011): 377–391. 10. ​Rachel Hughes, “Dutiful Tourism: Encountering the Cambodian Genocide,” Asia Pacific Viewpoint 49, no. 3 (2008): 318–330.



Perpetrator-­Victims 209 11. ​Bickford, Transforming a Legacy, 7. 12. ​On the pro­cesses leading to the establishment of the ECCC, see George Chigas, “The Poli-

tics of Defining Justice a­ fter the Cambodian Genocide,” Journal of Genocide Research 2, no. 2 (2000): 245–265. 13. ​Kheang Un, “The Khmer Rouge Tribunal: A Po­liti­cally Compromised Search for Justice,” Journal of Asian Studies 72, no. 4 (2013): 783–792. 14. ​Alexander Laban Hinton, “Transitional Justice Time: U ­ ncle San, Aunty Yan, and Outreach at the Khmer Rouge Tribunal,” in Genocide and Mass Atrocities in Asia: Legacies and Prevention, ed. Deborah Mayersen and Annie Pohlman (London: Routledge, 2013), 86–99. 15. ​Renée Jeffery, “Beyond Repair?: Collective and Moral Reparations at the Khmer Rouge Tribunal,” Journal of ­Human Rights 13, no. 1 (2014): 103–119. 16. ​Christoph Sperfeldt, “Cambodian Civil Society and the Khmer Rouge Tribunal,” International Journal of Transitional Justice 6, no. 1 (2012): 149–160; Manning, “Governing Memory,” 167. 17. ​Interviews conducted in January 2017 with representatives of the Documentation Center of Cambodia and Bophana Audiovisual Resource Center about their website and app, respectively, which are currently both in development. 18. ​James A. Tyner, Gabriela Brindis Alvarez, and Alex R. Colucci, “Memory and the Everyday Landscape of Vio­lence in Post-­Genocide Cambodia,” Social & Cultural Geography 13, no. 8 (2012): 853–871. 19. ​Inger Agger, “Calming the Mind: Healing ­after Mass Atrocity in Cambodia,” Transcultural Psychiatry 52, no. 4 (2015): 543–560; Anne Yvonne Guillou, “An Alternative Memory of the Khmer Rouge Genocide: The Dead of the Mass Graves and the Land Guardian Spirits [neak ta],” South East Asia Research 20, no. 2 (2012): 207–226; Alexander Laban Hinton, “Justice and Time at the Khmer Rouge Tribunal: In Memory of Vann Nath, Painter and S-21 Survivor,” Genocide Studies and Prevention: An International Journal 8, no. 2 (2014): 7–17; Judy Ledgerwood, “Buddhist Ritual and the Reordering of Social Relations in Cambodia,” South East Asia Research 20, no. 2 (2012): 191–205. 20. ​Quoted in David Chandler, “Cambodia Deals with Its Past: Collective Memory, Demonisation and Induced Amnesia,” Totalitarian Movements and Po­liti­cal Religions 9, nos. 2–3 (2008): 355–369. 21. ​Among many other interviews, for example, interview with a former Khmer Rouge soldier, conducted in October  2014  in Prey Veng province; interview with a former Khmer Rouge commune militiaman, conducted in September 2014 in Banteay Meanchay province; interview with a former Khmer Rouge military messenger, conducted in September 2014 in Pailin province. 22. ​Interview with a former soldier in September 2014 in Battambang province along the Thai border. 23. ​Among many other interviews, for example, interview with a former Khmer Rouge commune militiaman and subsequently collective committee member, conducted in August 2014 in Kampong Chhnang province; interview with a former Khmer Rouge messenger, conducted in December 2014 in Takeo province; interview with a former S-21 guard and interrogator, conducted in November 2014 in Kampong Chhnang province. 24. ​Interview with a former S-21 guard and interrogator in November 2014 in Kampong Chhnang province. 25. ​Interview with a former S-21 guard and interrogator in October 2014 in Kampong Chhnang province. 26. ​Among many other interviews, for example, interview with a former S-21 guard, conducted in October 2014 in Kampong Chhnang province; interview with a former Khmer Rouge militiaman, conducted in January 2015 in Takeo province.

210

ti m othy w illi a ms

27. ​Among many other interviews, for example, interview with the former Khmer Rouge chief

of a hard l­ abor site, conducted in August 2014 in Battambang province; interview with several former S-21 guards, conducted in September 2014 in Kampong Chhnang province; interview with a former Khmer Rouge village chief and soldier, conducted in October 2014 in Kampong Thom province. 28. ​Ravi Bhavnani, “Ethnic Norms and Interethnic Vio­lence. Accounting for Mass Participation in the Rwandan Genocide,” Journal of Peace Research 43, no.  6 (2006): 657. Luke Fletcher, “Turning Interahamwe: Individual and Community Choices in the Rwandan Genocide,” Journal of Genocide Research 9, no. 1 (2007): 32: Scott Straus, The Order of Genocide: Race, Power, and War in Rwanda (Ithaca, N.Y.: Cornell University Press, 2006), 140. 29. ​Michael Mann, The Dark Side of Democracy: Explaining Ethnic Cleansing (Cambridge: Cambridge University Press, 2005), 164. 30. ​David Chandler, A History of Cambodia, 4th ed. (Boulder, Colo.: Westview, 2008), 267. David Chandler, ­Brother Number One: A Po­liti­cal Biography of Pol Pot (Boulder, Colo.: Westview, 1999), 123–130. 31. ​Chandler, “Cambodia Deals with Its Past,” 355–369. 32. ​Peter Manning, “Reconciliation and Perpetrator Memories in Cambodia,” International Journal of Transitional Justice 9, no. 3 (2015): 386–406. 33. ​Manning shows that some former Khmer Rouge cadres living in former strongholds of the Khmer Rouge try to portray their former leaders as not “bad men,” pointing to the fact that they ­were all trying to be patriots fighting for their country. Manning, “Reconciliation and Perpetrator Memories,” 398. I would argue that this difference in perspective can—­among other ­factors—be explained by the difference in temporal perspective with ­these former Khmer Rouge men and ­women mostly also speaking about the civil war against Lon Nol and the war against the Viet­nam­ese occupation a­ fter 1979, with less focus on the period of Demo­cratic Kampuchea. 34. ​Erica Bouris, Complex Po­liti­cal Victims (Sterling, Va.: Kumarian Press, 2007). 35. ​Erin Baines, “Complex Po­liti­cal Perpetrators: Reflections on Dominic Ongwen,” Journal of Modern African Studies 47, no. 2 (2009): 163–191. 36. ​Bouris, Complex Po­liti­cal Victims: Chris Coulter, “Female Fighters in the Sierra Leone War: Challenging the Assumptions?,” Feminist Review 88, no. 1 (2008): 54–73. 37. ​Baines, “Complex Po­liti­cal Perpetrators,” 163. See also Erin Jessee, Negotiating Genocide in Rwanda: The Politics of History (New York: Palgrave Macmillan, 2017). 38. ​Multiple interviews with a former assistant to a commune chief, conducted in August, September, and November 2014 in Battambang province. 39. ​Baines, “Complex Po­liti­cal Perpetrators,” 165. 40. ​This is a recurring theme across almost all interviews, for example, in most detail in an interview with a former S-21 guard, conducted in August 2014 in Kampong Chhnang province; interview with a former Khmer Rouge soldier, conducted in September 2014 in Pailin province; interview with a former Khmer Rouge commune militiaman and subsequently collective committee member, conducted in August 2014 in Kampong Chhnang province. 41. ​Among many other interviews, for example, interview with a former Khmer Rouge militiaman, ­later also militia group leader, conducted in September 2014 in Battambang province; interview with a former Khmer Rouge soldier and militiaman, conducted in August  2014  in Kampong Chhnang province; interview with a former Khmer Rouge commune militiaman, teacher, security guard, and interrogator, conducted in October 2014 in Svay Rieng province; interview with a former S-21 guard, conducted in August 2014 in Kampong Chhnang province. 42. ​Interview with a former S-21 guard in November 2014 in Takeo province.



Perpetrator-­Victims 211 43. ​This is a recurring theme across almost all interviews, for example, in most detail in an inter-

view with a former Khmer Rouge militiaman, l­ater also militia group leader, conducted in September 2014 in Battambang province; interview with a former S-21 guard and interrogator, conducted in November 2014 in Kampong Chhnang province; interview with a former Khmer Rouge commune militiaman, teacher, security guard, and interrogator, conducted in October  2014  in Svay Rieng province; interview with a former S-21 guard, conducted in November 2014 in Takeo province. 44. ​Interview with a former Khmer Rouge commune militiaman and subsequently collective committee member in August 2014 in Kampong Chhnang province. 45. ​For example, interview with a former chief of a female mobile unit in August 2014 in Battambang province. 46. ​Multiple interviews with a former assistant to a commune chief, conducted in Battambang province in August, September, and November 2014. 47. ​Laura McGrew, “Pathways to Reconciliation in Cambodia,” Peace Review 23, no. 4 (2011): 514–521. 48. ​Julie Bernath, “ ‘Complex Po­liti­cal Victims’ in the Aftermath of Mass Atrocity: Reflections on the Khmer Rouge Tribunal in Cambodia,” International Journal of Transitional Justice 10, no. 1 (2016): 50. 49. ​Manning, “Reconciliation and Perpetrator Memories,” 387; Manning, “Governing Memory,” 169. 50. ​Thorsten Bonacker, “Global Victimhood: On the Charisma of the Victim in Transitional Justice Pro­cesses,” World Po­liti­cal Science Review 9, no. 1 (2013): 97–129. 51. ​Thorsten Bonacker, Wolfgang Form, and Dominik Pfeiffer, “Transitional Justice and Victim Participation in Cambodia: A World Polity Perspective,” Global Society 25, no. 1 (2011): 113–134. 52. ​Patrick Hein, “The Multiple Pathways to Trauma Recovery, Vindication, and National Reconciliation in Cambodia,” Asian Politics & Policy 7, no. 2 (2015): 207; Heather Ryan and Laura McGrew, Per­for­mance and Perception: The Impact of the Extraordinary Chambers in the Courts of Cambodia (New York: Open Society Justice Initiative, 2016); Eric Stover, Mychelle Balthazard, and K. Alexa Koenig, “Confronting Duch: Civil Party Participation in Case 001 at the Extraordinary Chambers in the Courts of Cambodia,” International Review of the Red Cross 93, no. 882 (2011): 503–546; Nadine Stammel, Sebastian Burchert, Sopheap Taing, Estelle Bockers, Christine Knaevelsrud, The Survivors’ Voices: Attitudes on the ECCC, the Former Khmer Rouge and Civil Party Participation (Berlin: Berlin Center for the Treatment of Torture Victims, 2010); Phuong  N. Pham, Patrick Vinck, Mychelle Balthazard, Judith Strasser, and Chariya Om, “Victim Participation and the Trial of Duch at the Extraordinary Chambers in the Courts of Cambodia,” Journal of ­Human Rights Practice 3, no. 3 (2011): 264–287; Judith Strasser, Thida Kim, Silke Studzinsky, and Sopheap Taing, A Study about Victims’ Participation at the Extraordinary Chambers in the Courts of Cambodia and Gender-­Based Vio­lence u­ nder the Khmer Rouge Regime (Phnom Penh: Transcultural Psychosocial Organ­ization Cambodia [TPO], 2015). 53. ​Renée Jeffery, “The Forgiveness Dilemma: Emotions and Justice at the Khmer Rouge Tribunal,” Australian Journal of International Affairs 69, no. 1 (2015): 35–52. 54. ​Rudina Jasini, Victim Participation and Transitional Justice in Cambodia: The Case of the Extraordinary Chambers in the Courts of Cambodia (ECCC) (Utrecht: Impunity Watch, 2016); Mahdev Mohan, “The Paradox of Victim-­Centrism: Victim Participation at the Khmer Rouge Tribunal,” International Criminal Law Review 9, no. 5 (2009): 1–43. 55. ​Bernath, “Complex Po­liti­cal Victims.”

212

ti m othy w illi a ms

56. ​John Lennon and Malcolm Foley, Dark Tourism: The Attraction of Death and Disaster (Lon-

don: Continuum, 2000).

57. ​Quoting co-­prosecutor Robert Petit in Bernath, “Complex Po­liti­cal Victims,” 56. 58. ​Tuol Sleng Genocide Museum audio guide, track 5. 59. ​Rachel Hughes, “Nationalism and Memory at the Tuol Sleng Museum of Genocide Crimes,

Cambodia,” in Contested Pasts: The Politics of Memory, ed. Katharine Hodgkin and Susannah Radstone (London: Routledge, 2003), 183. 60. ​“Stilled Lives: Photo­graphs from the Cambodian Genocide,” exhibition by Wynne Cougill with Pivoine Pang, Chhayran Ra, and Sopheak Sim (visited on January 9, 2017). 61. ​Introduction panel of “Stilled Lives.” 62. ​“Victims and Perpetrators? Testimony of young Khmer Rouge Comrades,” exhibition by Meng-­Try Ea and Sorya Sim (visited on January 9, 2017). 63. ​McGrew, “Pathways to Reconciliation.”

10 • COLLECTIVE CRIMES, COLLECTIVE MEMORY, AND TR ANSITIONAL JUSTICE IN BANGL ADESH K JELL ANDERSON

The recent war crimes ­trials at the International Crimes Tribunal (ICT) in Bangladesh have been much criticized for their alleged po­liti­cal biases and procedural shortcomings. The ICT is a special court set up by Bangladesh in 2009 to prosecute individuals for the war crimes, crimes against humanity, and genocide, which occurred in Bangladesh during the 1971 Liberation War. This chapter argues that the ICT should be understood as an ele­ment in the broader pro­cess of remaking the collective memory of mass vio­lence in Bangladesh. Mass vio­lence creates a profound social rupture in the socie­ties where it occurs. Such a rupture is experienced both individually and collectively. In cases where vio­lence is directed at civilian populations, such as in genocide, entire categories of persons are excluded from the universe of moral reciprocity and the benefits of belonging to the po­liti­cal community. The individual memories of victims and perpetrators alike are s­haped by the collective context and the instrumentality of memory as a means for reframing past crimes. For perpetrators, the primary consideration is often the reduction of moral culpability for wrongful acts. For victims, on the other hand, memory involves the restoration of moral frameworks and reintegration into the moral community. Transitional justice may also challenge perpetrator narratives and validate ­those of the victims. During the 1971 conflict in Bangladesh hundreds of thousands of ­people ­were killed as the Pakistani Army and its allies fought a Bengali nationalist rebellion seeking in­de­pen­dence for East Pakistan.1 The way in which this killing is viewed is a product of positionality. For Pakistani forces, it is an excess committed amidst excesses by the “other” side; for Bengali nationalists, 1971 was a genocide, 213

214

k jell a nder son

the culmination of longer term pro­cesses of marginalization and persecution of Bengalis by the state of Pakistan. This chapter analyzes narratives of the 1971 atrocities, while also examining the role of the ICT in remaking collective memory. It hypothesizes that collective crimes engender collective memory and that memories themselves constitute moral frameworks. Thus, where memory is contested, morality is also contested, and the existence of multiple, contested collective narratives means that moral frameworks (in the sense of how past wrongs are viewed) are also contested. In assigning individual criminal responsibility, the ICT is also seeking to build a counternarrative of past vio­lence. The collective memories of courts are produced through judicial procedures, which strive to ensure fairness and objectivity; yet, as the ICT shows, t­ hese judicialized collective memories, like other forms of collective memory, are also subject to po­liti­cal instrumentality.

Collective Crimes and Collective Memory Collective crimes are pursuant to collective objectives. Th ­ ese collectives are constituted around a politicized identity field—­whether it be ethnic, racial, religious, nationalist, or ideological in some other form. The crimes also target individuals by virtue of their belonging to another identity group. Collective crimes include genocide, crimes against humanity, and war crimes. War crimes may not be focused on ethnic groups, but they retain the po­liti­cal dimension common to other collective crimes: the victims of vio­lence (­enemy soldiers) are killed as representatives of an opposing po­liti­cal group. The collective vio­lence in Bangladesh in 1971 arguably involved war crimes, genocide, crimes against humanity, and other ­legal and illegal forms of vio­lence. Some key features of collective crimes are: They are mass crimes, with a massive number of perpetrators, victims, and bystanders. Perpetration also generally takes place in group settings. 2. They are po­liti­cal crimes committed pursuant to a po­liti­cal ideology or policy objective. 3. They are conformity crimes, wherein individual perpetrators are encouraged or facilitated by the state or state-­like authorities. 4. They are intergenerational crimes. The per­sis­tence of identity among ethnic and national groups ensures that past events continue to carry significance and po­liti­cal saliency in the pres­ent. Moreover, genocide is often justified by narratives of past vio­lence; ­these narratives produce continuities of vio­lence, which serve to reframe actions of the out-­group as being hostile or other­wise dangerous, when viewed through the lens of past atrocities.2 ­These narratives also claim that the destruction of ­future generations of the out-­group may be the only means to ensure the survival of the in-­group in the pres­ent. 1.



Collective Crimes, Collective Memory 215

­ ese unique characteristics of collective crimes also mean that the memory Th (and denial) of t­ hese crimes becomes politicized and collectivized. In accounting for their own crimes, perpetrators must also account for the crimes of their peers. Moreover, even uninvolved members of the “perpetrator group” (the group identified with the perpetrators) are often faced with collective responsibility for the crimes committed. Collective memory is a contested concept. Memory is reflective, unique to each individual. So how can memory be produced, or even exist collectively? The production of collective memory is an interactive pro­cess: collective memory is produced by individuals but collective memory also produces individual memories. The effect of the collective on the individual occurs through the incorporation of external narratives (which are produced and disseminated collectively) into internal narratives of the past (or memories). In any pro­cess of collective memory-­making, ­there are always ele­ments that are emphasized and ele­ments that are forgotten, memories that are deemed more legitimate and memories that are deemed less. This is true on both the individual level and the collective level. Th ­ ese collective memories of the past also function as a way of constituting moral fields. They serve to exonerate, reframe, or lionize the actions of specific individuals, as well as the group as a w ­ hole. The past is particularly contentious in states, such as Bangladesh, where collective memory is po­liti­cally instrumental in the pres­ent—­legitimizing and delegitimizing po­liti­ cal actors. The group, the nation, can give individuals transcendence—­the opportunity to be part of something larger than themselves, something through which they ­will live on even ­after their deaths. However, the benefits of belonging to the group hold only if the group is perceived to have positive qualities. Involvement of the group in past atrocities may be a “stain” on the pres­ent, a collective burden to be borne, or better yet, forgotten. This collective burden may, of course, be personalized where the atrocities have occurred in living memory, as in Bangladesh. In such cases the perpetrator is not a theoretical “other” but rather somebody known to you personally, perhaps your own f­ ather, or ­sister. Moreover, living perpetrators may still be subject to judicial procedures. The po­liti­cal culture of Bangladesh continues to be defined by this “living memory” of mass atrocities, as well as the instrumentalization of memory on the collective and individual level. The in­de­pen­dence of Bangladesh in 1971, through the Liberation War, continues to carry g­ reat significance for Bangladesh, Pakistan, and India alike. In the following section, I give an overview of the events of 1971 to provide some context for the discussion to follow.

216

k jell a nder son

bangladesh, 1971 Bangladesh was born through violent strug­gle. The shape of this vio­lence—­the perpetrators and the victims—­remains contested not only between Bangladesh, India, and Pakistan but within Bangladesh itself. In Bengali iterations of this story, it is Pakistan that is the opposing force, the oppressor in Bangladesh’s “Liberation War” from Pakistan. The vio­lence committed by the “Liberation forces” (Mukti Bahini) is not part of this narrative; or, more accurately, it is that illegitimate (unheroic) forms of vio­lence, such as atrocities, are excluded. And what of ­those Bengalis who chose to side with the state of Pakistan against their ethnic kin? They are seen by the in­de­pen­dence movement as both exceptional and traitors. Their acts undermine the narrative of heroic re­sis­tance against an oppressive and abusive power. With the creation of Pakistan in 1947, the country was to have two component parts—­West Pakistan and East Pakistan (now respectively the states of Pakistan and Bangladesh) separated by more than 2,000 kilo­meters of India. East Pakistan was culturally distinct from the West—it operated as a sort of Bengali nation-­ state, albeit with a substantial minority of half a million non-­Bengali Muslims (by the 1960s). Th ­ ere w ­ ere also about ten million Hindus in East Pakistan out of a total population of seventy-­five million ­people.3 Bengali identity arguably existed alongside Pakistani identity, with only 29 ­percent of East Pakistani university students self-­identifying as Bengali in a 1964 poll.4 Nonetheless, identity politics dominated relations between West and East Pakistan in the 1950s. A ­ fter the first partition, roads in Dhaka w ­ ere renamed to remove Hindu names, many Hindus lost civil ser­vice jobs (and ­were replaced by West Pakistani Muslims), and use of the Bengali language was discouraged.5 The choice of Urdu as Pakistan’s official language was seen by many in East Pakistan as a suppression of Bengali cultural and po­liti­cal identity. This gave rise to the Bengali language movement in 1952 and the official recognition of Bengali as a national language in 1954. By the 1960s West Pakistan had come to dominate the po­liti­cal and economic life of Pakistan, despite East Pakistan being more populous and accounting for most of the foreign exchange earnings (through jute exports). A new Pakistani official narrative also emerged, emphasizing the “­great privations they [Muslims] had under­gone at the hand of the Hindus over a long period of their history.”6 The loyalty of the Bengalis was also challenged, as they w ­ ere presented as being too “Hinduised” and too closely tied to India. The Bengali language itself was seen as being “Hindu-­like” with “many Sanskrit-­derived words.”7 This xenophobia was accompanied by a paranoia that escalated in tandem with the strength of the Bengali nationalist movement; for example, a contemporaneous Pakistan government document entitled “Pakistan Nationalism” spoke of a country “ridden with enemies” and argued Bengalis ­were “susceptible to



Collective Crimes, Collective Memory 217

tremendous influence from West Bengal” (meaning India).8 The Pakistani military (dominated by West Pakistanis such as Punjabis and Baluchis) intervened repeatedly in Pakistan, dismissing an elected government in 1954, preventing elections in 1958, and blocking the newly elected Pakistani parliament from convening in 1971.9 In 1970 the pro-­East Pakistan in­de­pen­dence Awami League won a national majority and General Yahya Khan (then president of Pakistan) intervened.10 First, the Awami League was banned and then President Khan cryptically gave General Tikka Khan (the commander of the Eastern Command) an order to “sort them out.”11 The British High Commissioner in Islamabad similarly noted that “Punjabi and Pathan contempt for the Bengali has risen to the surface and ­there is much talk of ‘teaching them a lesson.’ ”12 The increasing tensions in East Pakistan led to a protest movement among pro-­independence Bengalis. Th ­ ese initial limited acts of rebellion w ­ ere, however, soon met by a brutal reprisal. On March 25, 1971, the Pakistani Army initiated Operation Searchlight; initially it focused on the killing of targeted intellectuals, students, Hindus, and Awami League members.13 It was “an attempt to eradicate the ­future Bengali leadership.”14 ­After this first phase of attacks on Bengali elites, the army shifted to attacking Hindus.15 On March 28, Archer Blood (the U.S. Consul General in Dhaka) sent a cable to Washington entitled “Selective Genocide,” which contended that ­there w ­ ere “systematic” attacks against Awami League supporters, Hindus, and Bengalis.16 On May 14 Blood sent another cable entitled “The Slaughter of Hindus” in which he wrote of “a common pattern of Army operations whereby troops entered a village, enquired where the Hindus lived, and then killed the male Hindus.”17 Similarly, U.S. Senator William Saxbe read a letter written by Jon E. Rohde (an American doctor in East Pakistan) before the U.S. Senate testifying that the army of West Pakistan was undertaking the “systematic elimination of the intelligent­sia” and “the annihilation of the Hindu population.”18 Biharis (the non-­Bengali minority of Muslims from Bihar and other Indian states) largely supported the army.19 Some of them joined paramilitary groups (Razakar) fighting against the separatists such as the Al Shams and Al Badr.20 They did this b­ ecause, as Al Badr Chief Maulana Matiur Rahman Nizami (­later to become a Jamaat member of parliament before being executed for war crimes in 2016, following a trial at the ICT) said: “In this hour of national crisis it is the duty of ­every Razakar to carry out his national duty to eliminate ­those who are engaged in war against Pakistan and Islam.”21 The Razakars carried out extensive acts of vio­lence. According to contemporaneous British intelligence reports, “The Biharis are on the rampage. [They] appear to have been incited to riot by the Army, who are turning a blind eye to their activities.”22 The Pakistani Army also acted with ­great brutality. When Lieutenant General Amir Abdullah Khan Niazi took over command as governor and martial law

218

k jell a nder son

administrator of East Pakistan in April of 1971, he reportedly said, “What have I been hearing about the shortage of rations? Are ­there not any cows and goats in this country? This is e­ nemy territory. Get what you want.”23 The operations of the Pakistani forces in Bangladesh ­were essentially a brutal counterinsurgency against the pro-­independence Mukti Bahini guerillas. ­There seemed to be a general failure to distinguish between armed insurgents and the general civilian population: effectively all Bengalis and all Hindus ­were held to be “miscreants” and fair game for the use of excessive vio­lence. Journalist Anthony Mascarenhas wrote at the time: “In the [Pakistani] officers mess at night I have listened incredulously as other­wise brave and honourable men proudly chewed over the day’s kill. ‘How many did you get?’ ”24 ­There may have been a genocidal intent pres­ent among the Pakistani forces for certain aspects of the mass killing that took place. On certain occasions, the Pakistani Army checked w ­ hether the men in their custody w ­ ere circumcised, and all noncircumcised men (Hindus) w ­ ere killed.25 The army also a­ dopted genocidal language by referring to the operation as a “cleansing pro­cess.”26 The targeted nature of the vio­lence and rhe­toric contributed to the International Commission of Jurists concluding at the time that t­ here was a prima facie case for “acts of genocide.”27 However, the overall characterisation of the vio­lence must also account for the presence of other types of mass killing and atrocities (i.e., politicide, crimes against humanity, war crimes, and general po­liti­cal repression).28 ­There ­were allegations of Mukti Bahini abuses against West Pakistanis, Biharis, and pro-­ Pakistan ele­ments in Bangladesh.29 Qutubuddin Aziz, a Pakistan Movement30 stalwart and Pakistani diplomat, went so far as to write of “the Awami League’s rebellion and genocide and the Mukti Bahinis’ reign of terror” as well as “the massacre of more than half a million non-­Bengalis and pro-­Pakistan Bengalis by the Awami League.”31 Th ­ ese claims are grossly inflated, although t­ here do seem to have been massacres and crimes committed by the Mukti Bahini and other pro-­independence forces. Which aspects of the 1971 vio­lence one emphasizes is still largely driven by positionality, as demonstrated through the collective memory narratives of 1971.

collective memory of 1971 Collective memory meets the needs of the po­liti­cal pres­ent. In Bangladesh, culpability for past crimes has become intimately related to con­temporary electoral politics and prospects. Moreover, individuals involved in the perpetration of mass atrocities have occupied impor­tant po­liti­cal positions (largely in parties opposed to the Awami League) since 1971. In addition, ­trials have become pos­si­ ble only when accountability has come to be aligned with the interests of ruling



Collective Crimes, Collective Memory 219

elites; the association of elites with claims of past heroism serves as a source of po­liti­cal legitimacy. Each of the three primary narratives of the 1971 war is po­liti­cally instrumental in its own way.32 The mainstream Bengali narrative of oppression and genocide at the hand of Pakistani perpetrators establishes the Awami League as being the memory holder (and power holder) of the legacy of the 1971 war. It also associates Pakistanis and Biharis with the perpetration of atrocities, undermining their po­liti­cal legitimacy. In fact, the term “Bihari” has almost become synonymous with the Razakar paramilitaries, and many Biharis remain po­liti­cal exiles, stateless, and lacking citizenship rights in e­ ither Bangladesh or India. The Pakistani narrative of war excesses, discussed in greater detail below, is directed at both domestic and international audiences to justify past atrocities as an inevitable consequence of warfare, or to completely deny their occurrence. The “loss” of Bangladesh is thus rendered as a po­liti­cal tragedy at the hands of Indian imperialists. Fi­nally, the Indian narrative is one of humanitarian intervention in the face of Pakistani atrocities. Such a narrative establishes India as an altruistic actor fighting a just war against Pakistani brutality. We ­will focus ­here particularly on Bangladeshi and Pakistani narratives of 1971. The Bangladeshi narrative has not remained constant. Rather it has evolved in tandem with po­liti­cal developments in Bangladesh. In the years a­ fter the war (1971–1975) Sheikh Mujibur Rahman constructed a state centered on the Awami League as a singular po­liti­cal force. Personal po­liti­cal legitimacy was constructed (and continues to be constructed) based primarily on one’s role in the Liberation War. This “revolutionary legitimacy” occurs when “a coup or a protracted strug­gle has brought a new leadership or a new po­liti­cal entity to in­de­pen­dence, and ­those who come to rule it base their right to do so on having led or participated in that strug­gle.”33 This also involves the repudiation of the ancien régime. The new regime is presented as a redeemer, dispelling the darkness of the past in ser­vice of a new national vision. The same narrative occurs, for example, in Paul Kagame’s postgenocide, RPF (Rwanda Patriotic Front) Rwanda. Revolutionary legitimacy also entails counterrevolutionary illegitimacy, the labeling of po­liti­cal opponents as regressive and counterrevolutionary, as traitors to the cause. The Liberation War Museum in Dhaka pres­ents something close to the Bengali–­ Awami League narrative, in writing that “the strug­gle for a separate homeland manifested itself right from 1948 through a continuous, united and popu­lar strug­ gle for democracy, autonomy and for the upholding of its secular cultural identity.”34 The museum also states: “In an attempt to crush the nationalistic movement in East Pakistan, the Pakistani Military Junta unleashed a systematic genocide against Bengalee ­people on the fateful night of March 25, 1971. The Junta received support from a handful of religion-­based local parties and religious fundamentalists.”35

220

k jell a nder son

Moreover, it contends that Indian intervention occurred only “­after Pakistan attacked and bombed airfields in the western part of India,” thereby excluding Indian involvement in the training and arming of the Mukti Bahini. This version of history portrays the Pakistanis as relentless oppressors of the Bengalis, and in so d­ oing, it removes the agency of the Bengalis, as well as ignoring Bengali acts of vio­lence. It also casts pro-­Pakistani Bengalis and Biharis as being exceptional and traitorous, a recurrent theme that is also reflected in the narratives produced by the ICT through its judgments. The Awami League’s iteration of the historical rec­ord is distorted through both its exclusive focus on pro-­Pakistan vio­lence and its inflation of the numbers of Bengali dead. It is often posited that as many as three million ­people died when the real­ity is prob­ably closer to hundreds of thousands.36 This does not mean, of course, that such vio­lence did not constitute genocide or crimes against humanity. But in the emphasis on certain events and the obscuring of ­others, we see the shaping of a collective memory that is responsive to po­liti­cal needs—­namely, the legitimization of the Awami League and delegitimization of its opponents. The assassination of Sheikh Mujibur Rahman (the Awami League’s founding ­father of Bangladesh) in 1975 challenged the party’s hegemony. In the ensuing period of military rule u­ nder Ziaur Rahman and then o­ thers, the war continued to be a source of legitimacy; however, culpability for crimes during the war was increasingly suppressed, as the regime came to rely on Islamist factions to reinforce its own power base. Moreover, memory of the war came to focus increasingly on the experience of the Bengali victims without identifying the perpetrators. This entailed a narrative shift from a “Pakistani occupation force” to just the “occupation force.” The current ICT ­trials may be a refutation of this perpetrator anonymity, although only certain types of perpetrators (namely Razakars) may be brought before the court. In 1971 Ziaur issued an indemnity ordinance giving amnesty to ­those involved in the vio­lence, as well as in the assassination of Rahman. He also shifted the focus of Bangladeshi national identity from linguistic nationalism to religious nationalism constructed around Islamic identity. Rahman was involved in the foundation of the Bangladesh National Party (BNP) in 1978, and he also allowed leaders of the banned Jamaat-­e-­Islami to run for office. Th ­ ere ­were po­liti­cal incentives for rehabilitating local collaborators and aligning oneself with religious factions in Bangladesh.37 It must be noted, however, that this appeal to religious princi­ple was not the sole purview of the BNP; a­ fter all, Bangladesh is a predominately Muslim country, and the Awami League has failed to repeal the amended provisions of the constitution since assuming power in 2008. With the Awami League’s electoral victory in 2008, their narrative of the war once again supplanted other narratives. ­After the election, a War Crimes Fact-­ Finding Committee identified 1,600 suspects for the crimes of 1971, and the



Collective Crimes, Collective Memory 221

International Crimes Tribunal (ICT) was created months ­later. In August of 2013 the Jamaat Party’s registration was also cancelled, only months before a general election. This deregistration, as well as the creation of the ICT, contributed to the opposition boycott of the January 2014 elections, which w ­ ere handily won by the Awami League. Pakistan tends to frame the 1971 events as a po­liti­cal dispute or an armed conflict (rather than genocide), focusing on the culpability of the Awami League, Mukti Bahini, and India. For example, the official website of the Pakistan Army focuses entirely on the war with India, all but ignoring the mass deaths of civilians or the brutal repression of pro-­independence forces. The Indian forces are referred to as “aggressors”; for the Pakistani forces, on the other hand, “this conflict was their finest hour”; they ­were “fighting against overwhelming odds.”38 This narrative emphasizes Pakistani heroism, portraying India as power­ful but immoral, while minimizing the atrocities committed by the Pakistani forces. A similar narrative arises from the Hamoodur Rahman Commission, an official Pakistani government commission of inquiry formed in 1974 to investigate the events of 1971. While the commission’s report remains classified, the Supplementary Report, issued a­ fter the release of Pakistani prisoners of war from India in 1974, was ­later declassified ­after portions ­were leaked. The report downplays army atrocities and diverts attention to professional misconduct, especially corruption and incompetence. For example, the report cites “the moral aspect of the ­causes of our defeat in the 1971 war” and denounces “lust for wine and ­women and greed for land and ­houses” among Pakistani officers.39 Through ­these competing narratives we are left with very dif­fer­ent stories as to what happened in 1971 and who is responsible.

identity and the formation of collective memory The competing narratives of the 1971 war are po­liti­cally instrumental while also contributing to the construction of national identity. As sociologist Kiyoteru Tsutsui writes, “Collective memory is an essential building block of social solidarity, which is based primarily on the nation in modern socie­ties.”40 The creation of collective identity requires collective memory. This memory provides a per­sis­tent sense of who we are as social beings. As Maurice Halbwachs argues, “­There is no point in seeking where . . . ​[memories] are preserved in my brain or in some nook of my mind to which I alone have access: for they are recalled by me externally, and the groups of which I am a part at any time give me the means to reconstruct them.”41 Groups such as the Bengalis are also communities of memory, ­imagined communities, to use Benedict Anderson’s phraseology. In the exercise of identity-­making, individual acts, particularly when they are admirable, are projected and reproduced through the group. Yet, memory is also constituted

222

k jell a nder son

through silence. One could argue that the transitional justice pro­cess in Bangladesh is a challenge to the silence, which has prevailed since 1971; consequently, it is an act of reclaiming and redefining history. History, it may be argued, is “the remembered past to which we no longer have an ‘organic’ relation.”42 Yet such a definition of history functions as historicism by framing history as something disconnected from individual memories. In relatively recent historical events, such as Bangladesh in 1971, many individuals do remember the vio­lence as a first-­hand experience, although their accounts are also undoubtedly ­shaped by collective memory. Individual memory may, in fact, be inseparable from cultural memory, as “­there is no such ­thing as pre-­cultural individual memory; but neither is ­there a Collective or Cultural Memory (with capital letters) which is detached from individuals and embodied only in media and institutions.”43 The individual draws from cultural repertoires in reconstructing his or her own memories, in what Frederick Bartlett called “schemes,” while also authoring collective memory.44 The construction of collective memory is never a completely successful hegemonic pro­cess: even where pro­cesses seek to consolidate memory ­there w ­ ill always be pluralistic memories, particularly where “official memory” conflicts with individual memory.45 As ethnic groups are politicized and new nation-­states are constituted, ­there is prominence given to the formative moment, the creation moment constituted through the po­liti­cal group. It is the moment when po­liti­cal identity is realized, although nationalist movements may already have a sense of identity long before they achieve in­de­pen­dence. This formative moment is often rooted more in myth than in real­ity. It can sometimes center on survival in the face of ­great evil. The fact of survival gives reason and legitimacy to the group’s existence, and the group’s state-­making. Such identity then becomes constructed around an image of victimization; this identity has, as a corollary, an oppositional “other” that is determined to cruelly destroy the very existence of the group. For example, the government of Bangladesh described the creation of Bangladesh in its 2013 Universal Periodic Review submission as “a nation born of a heroic War of Liberation with the sacrifice of 3 million martyrs and honour of 200,000 ­women.”46 By focusing on female victims, who are considered the most vulnerable in a patriarchal context, the victimization of Bengalis is brought into stark relief. The rape of ­these ­women remains privately shameful, even while ­there is a “fetishization of this public secret.”47 Through trying perpetrators of atrocities the fact of survival is reaffirmed. Asymmetrical victimization plays a similarly strong role in the creation of collective identity for Jews (through the Shoah), and Armenians (through the Aghet). Identity and the self-­esteem of the individual and group become intimately linked to a notion of unjust and asymmetrical suffering. Memory is constituted through acts of remembrance and forgetting in order to emphasize this character



Collective Crimes, Collective Memory 223

of the formative moment; individuals and groups alike ­will resist counternarratives. For example, seventy-­one Bangladeshi civil society organ­izations issued a joint declaration in 2007  in which they argued: “Remarking that no genocide was committed, and neither did any incident or war crime take place in ’71 and that our liberation war was actually a ‘civil war’ is tantamount to defying the Constitution of Bangladesh, deriding the freedom fighters, disparaging the supreme sacrifice of nearly three million martyrs, and last but not the least, denying the very existence of a sovereign Bangladesh.”48 Accordingly, the recharacterization of the Liberation War is presented as nothing less than a denial of the existence of Bangladesh. Collective memory of mass atrocities often distinguishes between: perpetrators, “core victims” (­those who belong to the in-­group), and “peripheral victims” (­those victims outside of the in-­group). The victimization of the peripheral victims is de-­emphasized or overlooked b­ ecause it seems to challenge the binary image of victim/perpetrator. The peripheral victims are often completely forgotten. In Bangladesh, the mainstream perspective (Bengali, Awami-­League) identifies the perpetrators as West Pakistanis and Razakars (collaborators) and the victims as Bengalis. The peripheral victims, such as Biharis, are excluded from this narrative as are t­ hose killed by the liberation forces. Through this narrative Bengali victimization is sanctified, and the Islamists and non-­Bengalis are condemned. The peripheral victims are ignored in ser­vice of the po­liti­cal needs of the core victims. In turn, in Pakistan we see a very dif­fer­ent narrative, which frames the atrocities in 1971 as being war excesses committed in ser­vice of holding a country together ­under siege from Bengali militants. The Rahman Commission reports that “during and ­after the military action excesses ­were indeed committed on the ­people of East Pakistan.”49 This passage illustrates the ways in which narrative mitigates culpability, particularly through use of the passive voice (“excesses w ­ ere indeed committed”) and the focus on “military action” rather than genocide or atrocities. Furthermore, the commission argues that the “surrender in East Pakistan has indeed been a tragic blow to the nation. By the act of surrender Pakistan stood dismembered.”50 The consolidation and proliferation of memory are countervailing tendencies involving a tension between hegemonic collective memory and heterogeneous individual memories, which may or may not reflect the official narrative. The transmutation of individual memory to collective memory is often coercive: acts of individual memory that diverge from the collective memory are threatening, perhaps even disloyal to the group or the state itself. We are left with a contradiction between the occurrence of multiple forms of vio­lence and unidimensional narratives. For example, in her work, historian Yasmin Saikia quotes a Mukti Bahini who attempted to rape a young Bihari ­woman; she concludes, “His story has no place in a Bangladesh that revels in the glory of victory in 1971. Perpetrators

224

k jell a nder son

­ ere the Pakistani ‘­others.’ ”51 Such divergent memories function as what Fouw cault called “countermemories,” which challenge dominant discourses, even while they may be grounded in personal experience.52

reframing narratives of collective crimes During and ­after genocide, memory is also politicized and collectivized. Individual perpetrators incorporate external narratives (which are produced and disseminated collectively) into internal narratives of the past (memory). The past self must be reconciled with the pres­ent.53 In accounting for their own crimes, perpetrators must also account for official narratives and the crimes of their peers. Individual and collective memory constitute an interactive relationship in which individual memory is ­shaped by collective narratives but collective memory is also constructed by individuals, including “memory entrepreneurs” who attempt to shift collective narratives.54 Memory may be forcibly imposed on countries through external po­liti­cal pressures and judicial procedures or by applying global ­human rights norms, such as the Nuremberg Tribunal or the International Criminal Tribunal for the Former Yugo­slavia (ICTY). In the case of the 1971 atrocities, however, ­there was l­ ittle external pressure to prosecute or other­wise recognize the crimes. The collective nature of perpetration may also facilitate perpetrator neutralization, as perpetrators reconceptualize their crimes as relatively insignificant and themselves as lacking agency when placed within the collective context of mass perpetration. Individual memory, therefore, is also subject to po­liti­cal instrumentality through formal narrative control by the authorities.55 The individual is encouraged, or even required, to produce a personal narrative that aligns with the official collective narrative. The collective memory of mass atrocities is also ­shaped by the outcome of ­those atrocities; where perpetrators remain in power, they continue to propagate narratives which justify their vio­lence. Alternatively, state policies may construct systems of denial (“synchronised distortions,” as Stanley Cohen terms them).56 Where transitional justice mechanisms exist in ­these circumstances, they often recognize victimization without attributing perpetrator culpability, or they focus on the prosecution of a few token perpetrators. Even where the perpetrating regime is removed from power, ruling elites are often associated with the prior regime, ­whether it be by virtue of their holding formal positions in the state apparatus or their association with corruption generated by the “deep state.” The outcomes of postatrocity memory are not set in stone in ­either case; civil society plays a crucial role. As Tsutsui writes, “The stronger the po­liti­cal mobilization by critics of perpetration, the more likely the national discourse w ­ ill feature ac­cep­tance of guilt; conversely, the stronger the mobilization by defenders of perpetration, the more likely re­sis­tance to guilt w ­ ill prevail.”57 The



Collective Crimes, Collective Memory 225

significant civil society pressures in Bangladesh in f­ avor of the creation of a tribunal have dictated that national discourse in Bangladesh is now focused on accountability for the 1971 vio­lence. However, this accountability, as we argue elsewhere, is limited only to certain types of perpetrators and perpetration. Narratives that deny or reframe crime at the collective level also absolve individuals of moral and ­legal culpability. I have identified ­here five collective memory narratives that reframe collective crimes:58 1.

Claim of ambiguity; Denial of the crime; 3. Claim of equivalency; 4. Use of ledger meta­phor; 5. Claim of higher purpose. 2.

Through the claim of ambiguity, it is argued that past vio­lence is unclear. Past crimes may also be denied entirely, or their character may be denied. It is common, for instance, to replace ac­cep­tance of genocide with the ac­cep­tance of lesser forms of crime. On a practical level, both the claim of ambiguity and denial of the crime may involve the destruction or censorship of incriminating documents, the targeting of witnesses, or the reburial of victims in secondary mass graves. The relative paucity of official fact-­finding on the atrocities of 1971 in both Bangladesh and Pakistan is indicative of significant po­liti­cal interests opposed to their unveiling. The claim of equivalency is often a “master narrative” that reframes vio­lence on the collective level. This narrative argues that all sides committed acts of vio­ lence, and thus that one side is no more culpable than another. Genocide is, by its very nature, asymmetrical; vio­lence is directed at the complete destruction of a group and this requires a large power differential. Equivalency moves vio­lence from the genocide frame (where the perpetrator and victim group are relatively clear) to a conflict frame (where extensive violations ­were committed by all sides). The claims of ambiguity, denial, and equivalency are all prevalent in Pakistani narratives of 1971. The claim of equivalency also denies the victim by arguing that the victims brought vio­lence on themselves through their own wrongful acts, or that they ­were committing or planning to commit atrocities. Equivalency may also take the form of evasion, arguing that “every­one suffered,” or even that it was a crime where the world was the victim rather than specific groups or individuals.59 This kind of misplaced universalism, what Karl Jaspers called “metaphysical guilt,” was used ­after the Holocaust, to recast the Holocaust as a ­human evil rather than a German evil.60 In Bangladesh, as in other cases, focusing on the multiplicity of victimization may also obscure asymmetrical patterns and scales of vio­lence, as in the systematic targeting of Hindus.

226

k jell a nder son

Similarly, both individuals and groups may use the ledger meta­phor to claim that, ­because one has done more good than harm over time, the vio­lence in question is neither definitional nor characteristic. Such a narrative recontextualizes atrocities as being an anomaly and minor harm when weighed against the essential virtue of the individual or collective. It also assumes that war crimes are committed only by ­those who are fundamentally malevolent. Fi­nally, perpetrator collective memories may justify vio­lence as serving a necessary or a higher purpose. Pakistani narratives often frame the vio­lence of 1971 as being for a good cause (keeping the country together) even while sometimes decrying “war excesses.” This type of narrative is typical where the perpetrators of vio­lence have prevailed. This is partly the case in Bangladesh ­because, although Pakistan ultimately lost the war for East Pakistan, the state of Pakistan endures. The narratives utilized often relate to conditions at the time of perpetration: if crimes w ­ ere also committed by members of the victim group, this facilitates the denial of the victim; if vio­lence was justified at the time of perpetration, this may facilitate the claim of higher purpose. In Pakistan, we see all t­ hese frames, while the Awami League Bengali narrative is entirely focused on Bengali victimization. For example, in the chapter of its report entitled “Alleged Atrocities by the Pakistan Army” the Rahman Commission first turns to the “misdeeds of Awami League Militants” and notes: “Let it not be forgotten that the initiative in resorting to vio­lence and cruelty was taken by the militants of the Awami League.”61 It further argues that “a large number of West Pakistani officers ­were butchered by the erstwhile Bengali colleagues,” and decries the “inhuman crimes committed on the helpless Biharis, West Pakistanis and patriotic Bengalis living in East Pakistan during that period.”62 The intention of recounting ­these atrocities is not “in justification of other atrocities,” but rather to “enable the allegations to be judged in their correct perspective.” It goes on to state: “The crimes committed by the Awami League miscreants ­were bound to arouse anger and bitterness in the minds of the troops, especially when they . . . ​­were also subjected to the severest of humiliations. . . . ​Tales of ­wholesale slaughter of families of West Pakistani officers and personnel of several units had also reached the soldiers who ­were ­after all only h­ uman, and reacted violently in the pro­cess of restoring the authority of the central government.”63 In fact, the commission is justifying the acts of the Pakistan army and its allies as “only h­ uman” while demonizing pro-­ independence forces as “miscreants” and reminding us that the higher purpose of the operation was to restore the authority of the central government. The report also argues that “the atrocities committed by them [the Awami League] on their own ­brothers and ­sisters must, therefore, be always kept in view.”64 The narratives of individual perpetrators are derived from t­hese collective narratives. However, where perpetrators are removed from power structures, their narratives shift over time from justifying the overall rightfulness of the vio­



Collective Crimes, Collective Memory 227

lence to minimizing their own role. This shift also reflects individual adaptation to evolved moral frameworks. It is no longer deemed acceptable to celebrate vio­lence, so the individual must reframe or minimize his or her own role. The framing of past vio­lence by individuals and collectivities is often dependent on po­liti­cal and judicial pro­cesses.

the international crimes tribunal, collective memory, and transitional justice The b­ attle over the collective memory of 1971 has now entered the judicial sphere through the ICT. The roots of this tribunal are in 1972, when the Bangladesh government issued the Bangladesh Collaborators (Special Tribunals) Ordinance to try “collaboration” with Pakistan in the commission of atrocities and other criminal acts (including both petty offences and serious crimes). By March of 1973 seventy-­three tribunals had been set up across the country. Six months of t­ rials took place u­ nder the Collaborators Ordinance. The t­ rials ­were then halted, and a few months ­later, a general amnesty was issued for collaborators so that “the entire country could “join in the task of nation building.”65 ­Those released included members of Islamist po­liti­cal parties such as the Jamaat e Islami, the Muslim League, and the Nizam e Islam.66 In total 30,471 individuals ­were charged u­ nder the ordinance, 2,848 w ­ ere brought to trial, and 752 ­were sentenced (with sentences ranging from a few years imprisonment up to the death penalty). A more comprehensive attempt at ­trials was made through the International Crimes (Tribunals) Act of 1973, which set up t­ rials in accordance with international criminal law, rather than Bangladeshi domestic law and procedure. Less than two hundred ­people ­were set to be tried ­under this new procedure for international crimes such as genocide. However, the ICT did not actually function ­until de­cades ­later. Impunity was further entrenched when in 1976 President Ziaur revoked the Collaborators Ordinance, gave cabinet posts to several leading collaborators, and restored citizenship to a number of alleged war criminals. When the ICT was revived in 2009, in the wake of the Awami League’s electoral victory, the law was amended to give the prosecution the right to appeal sentences and impose a time limit of sixty days for the disposal of appeals by the Appellate Division of the Supreme Court. The original act was primarily targeted at the Pakistani Army, but the ICT would have no jurisdiction over army officials who resided in Pakistan or abroad, so the new prosecutions w ­ ere to focus exclusively on Bangladeshi collaborators. Amendments to the act also expand jurisdiction beyond members of “armed, defense or auxiliary forces” to include “individuals or groups of individuals” (i.e., civilians and organ­izations). The requirement for a military judge to sit on the Tribunal was also removed, and Bengali was added as a procedural language.

228

k jell a nder son

Ultimately the tribunal has focused almost exclusively on alleged perpetrators who are opposition politicians. The tribunal has indicted nine leaders of the Jamaat-­e-­Islami and two leaders of the Bangladesh National Party. This apparent po­liti­cal focus has been much criticized. For example, a November  2015 U.S. State Department diplomatic cable obtained by Wikileaks argued that “­there is ­little doubt that hard-­line ele­ments within the ruling party believe that the time is right to crush Jamaat and other Islamic parties.”67 A report from a ­People’s Inquiry Commission published three years before the Awami League’s electoral victory notes: “In recent months, cadres of the Jamaat-­e-­Islami . . . ​ have been waging a violent, brutal campaign against ­those agitating for ­trials of war criminals. . . . ​Against such a backdrop, the Commission felt it imperative that leading collaborators, killers and war criminals of 1971 w ­ ere brought to trial.”68 Thus, ­there was a po­liti­cal impetus b­ ehind the creation of the tribunal, which also affected the way in which it functioned. The death penalty was not initially included in the tribunal’s sentencing, but was added partly b­ ecause of the Shahbag mass protests in 2013 (where thousands of Bangladeshis took to the streets in the Shahbag neighborhood of Dhaka in support of the death penalty following the sentencing of ICT defendant, Abdul Quader Mollah, to life imprisonment). Transitional justice pro­cesses may arguably serve several impor­tant social purposes such as deterrence and the reconstitution of moral frameworks. As Gerry Simpson argues, “The very idea of lawlessness is put on trial.”69 ­Trials may also pres­ent the possibility of reintegrating victims and perpetrators into the same moral framework (the law). The suffering of the victims is recognized, and the perpetrators are identified and condemned in an official record-­making pro­ cess. However, the presence of multiple competing memory narratives also leads to the presence of multiple competing moralities. Who is responsible, and what are they responsible for? Transitional justice trial pro­cesses are always selective, partly due to procedural and jurisdictional constraints (including powers of arrest) and partly driven by political-­instrumental considerations. The reconstitution of the moral community w ­ ill not occur where justice is seen as being unfair; in such scenarios, courts may contribute to intergroup conflict. Notions of collective blame alienate t­ hose within the “perpetrator group” who might be inclined to contribute to the peace-­building proj­ect. Furthermore, as Lawrence Sherman argues, where systems are perceived as unfair and illegitimate, ­trials may actually contribute to criminal be­hav­ior as a mode of re­sis­tance.70 Collective retaliation may occur where institutions are deemed to be unfair on the collective level, such as where t­ here is systemic discrimination or oppression.71 This retaliation can occur against another collectivity, ­whether or not that collectivity is the source of the unfairness in question. In other words, the perceived unfairness of the ICT ­trials may contribute to further social conflict. The ­trials in Bangladesh have been much criticized by h­ uman rights NGOs and international observers. The critique of the ICT centers on the politicization



Collective Crimes, Collective Memory 229

of the t­ rials, as well as inadequate safeguards for the rights of the accused. The dangers posed by poor pro­cesses are magnified when one ­factors in the frequent application of the death penalty by the ICT. The tribunal also appears to lack in­de­pen­dence and to be ­under the close po­liti­cal control of the government of Bangladesh.72 Where collective memory is highly polarized, responsibility for past crimes may be attributed to entire po­liti­cal, religious, or ethnic groups. This is arguably the case in Bangladesh, where the ICT has come to focus only on crimes committed by groups sympathetic to Pakistan. The selectivity of courts, both in terms of prosecutorial strategy and the demarcation of jurisdiction, may function to focus on certain types of victimization and certain categories of victims. This may occur, for instance, through temporal jurisdiction (as atrocities in certain periods may have been committed primarily by one group).73 The prob­lem of the collectivization of responsibility also arises through proportionality in charging: that is, should individuals from all “sides” be charged? Should this proportionally reflect the responsibility of each side for the crimes? In the case of the ICT only Razakars (collaborators) have been charged for the atrocities of 1971. Even if this is just, the tribunal must also “articulate the collaborators’ alleged responsibility within the overall context of the use of force by Pakistani armed forces in Bangladesh.”74 As this chapter has argued, individual and collective narratives of vio­lence correspond closely with po­liti­cal interests. In situations where perpetrators have “won” a­ fter the commission of mass atrocities (i.e., Indonesia), perpetrator narratives that continue to justify the rightfulness of the acts of vio­lence ­will persist. This is, to some degree, the way that the crimes of 1971 are framed in Pakistan, as well as among some members of the Jamaat e Islami in Bangladesh. In Bangladesh, ­these denialist narratives have officially been supplanted by another narrative, driven partly by the instrumental use of historical memory as a tool of legitimacy-­building by the Awami League. The narratives produced by the ICT closely parallel t­hose of the Awami League. For example, the judgment of Abul Kalam Azad cites Bangladesh Jihad Watch, which describes itself as “a website which exposes the Talibanization of the state of Bangladesh.”75 Moreover, in this judgment and many o­ thers, members of pro-­Pakistan paramilitary groups are always described through the emotionally loaded term of “collaborator.” The war is also referred to in the Azad judgemnt as “our war of liberation,” thereby already setting out the positionality of the court as being sympathetic to one party to the conflict.76 Furthermore, the Kamaruzzaman judgment uses highly emotive language to argue that “it is evinced that Jamat e Islami played a key role. . . . ​[T]he objective was to crush the Bengali nationalists, by maintaining unholy alliance with the Pakistani army.”77 In other words, they committed the “original sin” of being counterrevolutionaries opposed to the in­de­pen­dence of Bangladesh. Such a

230

k jell a nder son

po­liti­cal opinion would not normally be addressed by a court prosecuting international crimes. The martyrdom and victimization of the Bengali ­people is also central to the war’s narrative, as constituted through the court. In the Khan and Uddin trial judgment the court concludes: “In the present-­day world history, conceivably no nation paid as extremely as the Bangalee nation did for its self-­determination,”78 while the Syed Mohammed Qaiser judgment cites “homage to the blood of millions of patriotic martyrs and innocent defenceless ­people.”79 The unity of Bangladesh in opposition to West Pakistani oppression is also emphasized: “The Pakistani army had no friends in Bangladesh—­except a few traitors who took stance against the war of liberation and they belonged to the ideology of pro-­ Pakistan po­liti­cal parties, e.g. Muslim League, the Convention Muslim League, the Jamaat-­e-­Islami ( JEI) and the Nizami-­i-­Islami.”80 Thus, opposition po­liti­cal parties are traitors to Bangladesh. The presumption of innocence seems at serious risk in such an approach. Perhaps we should not be surprised that the ICT, a largely secular institution created by the state of Bangladesh, should take a position sympathetic to official narratives. However, in Bangladesh the fundamental identity of the state is itself contested between secularists and Islamists, and the court’s approach seems to be influenced by this po­liti­cal conflict. We must also note that it is not unusual to consider entire organ­izations as criminal. The same approach has been taken with regard to the Nazi party in the Nuremberg Tribunal, as well as in lustration pro­cesses in numerous states. However, in ­these cases the party in question had been removed from power rather than being active in electoral politics. The approach of the ICT runs the risk of making the court nothing more than a tool for the construction of partisan po­liti­cal narratives, in part through the elimination of opposition leaders through the death penalty. For de­cades Bangladesh’s approach to the vio­lence of 1971 centered on impunity. ­Those responsible for crimes ­were given amnesty without any requirement that they disclose their crimes or face the victims. However, in the post-­ICT period ­there has been a shift to collective shaming without reintegration, as entire categories of persons and po­liti­cal ideologies are condemned and cast out of the moral community without any restorative pro­cess. Unfortunately, the ICT t­ rials seem as concerned with the advancement of po­liti­cal narratives as they are with fact-­finding or the attribution of criminal responsibility for the atrocities of 1971.

conclusion: memory and transitional justice in bangladesh In Bangladesh, as in many other states, memory has become a battleground for the advancement of con­temporary po­liti­cal agendas. It constitutes morality in the sense that it contextualizes and neutralizes individual acts of vio­lence, while



Collective Crimes, Collective Memory 231

also legitimizing or delegitimizing individuals and groups associated with past vio­lence. This has contributed to a distorted transitional justice pro­cess, which seems to be driven by a par­tic­u­lar narrative of the 1971 war. ­Under such conditions ­trials may further social divisions, rather than rebuild a shared moral community, which is an essential ele­ment of postatrocity peace-­building. It is neither surprising, nor unusual, that collective memory of 1971 is contested, yet such contestation may become damaging where narratives deny past atrocities or assign collective blame to entire groups. It is undeniable that grave atrocities ­were committed in Bangladesh in 1971 and that ­these atrocities ­were committed mostly by Pakistan and its allies. Moreover, the de­cades of impunity for the atrocities of 1971 in Bangladesh had negative impacts on the Bangladeshi polity: through impunity, extreme vio­lence and fanat­ic­ ism ­were abided as a legitimate mode for the exercise of po­liti­cal power. This tendency may have contributed to the extremist turn in Bangladeshi politics in recent years, as well as the ongoing polarization and po­liti­cal instability within Bangladesh. Yet, the t­ rials of the ICT have not done much to build social solidarity. Rather, they have fed polarization and reinforced distorted perpetrator narratives asserting their own victimization. Even if Tsutsui is correct that collective memory is a “building block of social solidarity” and t­ rials themselves contribute to collective memory, Mark Osiel is also correct when he notes that ­there is a risk of “sacrificing the defendants’ rights on the altar of social solidarity.”81 Beyond the procedural dangers of such approaches to justice, they may paradoxically undermine the very goals of seeking justice by generating collective alienation and retaliation. ­There is a broader question h­ ere about the appropriate role of courts as history-­makers. Scholars such as Osiel have argued against t­rials being a producer of history, contending that such goals are inimical to the proper functioning of courts, which necessarily focus on individuals.82 Perhaps the dangers are greatest when courts seek to be an instrument of rebalancing narratives, rather than just assigning individual guilt. In seeking to produce a counternarrative and focusing only on certain incidents of vio­lence, courts sometimes flatten the sharp edges of history. In seeking to tell the untold story, courts are removed from their positivist aspirations, which may be illusory, but nonetheless limit the politicization of proceedings. The use of courts to build counternarrative risks replacing one distorted narrative with another. Such a result could potentially increase social conflict in Bangladesh, a society riven by deep divisions over the past.

232

k jell a nder son

notes With thanks to Christian Gerlach for his helpful comments on my draft, as well as to Dirk Moses for his lit­er­a­ture suggestions. 1. ​Please note that the spelling “Bengalee” is used interchangeably with “Bengali.” While I have chosen the spelling “Bengali,” the alternative spelling occurs in some quotations. 2. ​See, for example, Aaron T. Beck, Prisoners of Hate: The Cognitive Basis of Anger, Hostility, and Vio­lence (New York: Harper, 2000); and Branimir Anzulovic, Heavenly Serbia: From Myth to Genocide (London: Hurst, 1999). 3. ​Peter R. Kann, “A Nation Divided,” Wall Street Journal, July 23, 1971, n.p. 4. ​Christian Gerlach, Extremely Violent Socie­ties (New York: Cambridge University Press, 2010), 124. 5. ​Sydney H. Schanberg, “West Pakistan Pursues Subjugation of Bengalis,” New York Times, January 14, 1971. 6. ​Declassified file from the Pakistan Ministry of Education, cited in Yasmin Saikia, “Insaniyat for Peace: Survivor’s Narrative of the 1971 War in Bangladesh,” Journal of Genocide Research 13, no. 4 (2011): 499. 7. ​Ibid., 484. 8. ​Ibid. 9. ​Ben Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven, Conn.: Yale University Press, 2007), 573. 10. ​National ­People’s Inquiry Commission, Report on the Findings of the ­People’s Commission of Inquiry into the Activities of the War Criminals and the Collaborators (Dhaka: Ekattorer Ghatak Dalal Nirmul Committee, 2005), 4. The Awami League won 75 ­percent of the vote and 98 ­percent of parliamentary seats in East Pakistan. 11. ​A.A.K. Niazi, The Betrayal of East Pakistan (Karachi: Oxford University Press, 1998), 46. 12. ​Angela Debnath, “British Perceptions of the East Pakistan Crisis 1971: ‘Hideous Atrocities on Both Sides?,’ ” Journal of Genocide Research 13, no. 4 (2011): 428. 13. ​Sonia Nishat Amin, “A Case Study of the Trauma of 1971: Martyred Intellectual Selina Parveen,” International Conference on Genocide, Truth and Justice, Dhaka, Bangladesh, March 1–2, 2008, 54. 14. ​“Death in East Pakistan,” editorial, Eve­ning Star, April 17, 1971. 15. ​Schanberg, “West Pakistan Pursues Subjugation of Bengalis.” 16. ​Kiernan, Blood and Soil, 575. 17. ​Ibid. 18. ​Jon E. Rohde, “Recent Events in East Pakistan,” extract from the Rec­ord of the United States Senate, April 29, 1971. Reprinted from the Rec­ord of the U.S. Senate as “Recent events in East Pakistan” in Sheelendra Kumar Singh et  al. ed., Bangladesh Documents (Madras: B. N. K. Press 1971), 1:349–351. 19. ​Kann, “A Nation Divided.” 20. ​­There ­were also some Bengali Razakars, though it is difficult to arrive at accurate numbers for the composition of the militias. 21. ​ Daily Sangram, September 15, 1971, as cited in Bina D’Costa, Nationbuilding, Gender and War Crimes in South Asia (London: Routledge, 2011), 157. 22. ​Debnath, “British Perceptions,” 428. 23. ​Hamoodur Rahman Commission, Hamoodur Rahman Commission Supplementary Report (Lahore, 1974), 14. 24. ​Anthony Mascarenhas, “Genocide,” Sunday Times, June 13, 1971, cited in Debnath, “British Perceptions,” 435.



Collective Crimes, Collective Memory 233

25. ​Ibid. 26. ​Ibid. 27. ​International Commission of Jurists, “The Events in East Pakistan, 1971” (Geneva, 1972), 97. 28. ​The term “politicide” was coined by Barbara Harff to mean po­liti­cal mass killing. See Bar-

bara Harff, “No Lessons Learned from the Holocaust? Assessing Risks of Genocide and Po­liti­cal Mass Murder since 1955,” American Po­liti­cal Science Review 97, no. 1 (2003): 57–73. 29. ​Niazi, Betrayal of East Pakistan, 42. 30. ​The Pakistan Movement, or Tehrik-­e-­Pakistan, was a religious nationalist movement in the 1940s seeking the creation of the state of Pakistan (i.e., a dif­fer­ent in­de­pen­dence movement than the one seeking the creation of Bangladesh). 31. ​Qutubuddin Aziz, Blood and Tears (Karachi: United Press of Pakistan, 1974), ii–­iii. 32. ​Ahmed Dawood, “Rethinking the Big Lies from 1971,” Express Tribune, December 16, 2011. 33. ​Laurie Brand, Official Stories: Politics and National Narratives in Egypt and Algeria (Stanford, Calif.: Stanford University Press, 2014), 14–15. 34. ​Liberation War Museum, “The Emergence of Bangladesh,” http://­www​.l­iberation​warmu​ seumbd​.o­ rg​/­emergence​-o­ f​-b­ angladesh​/­, accessed June 7, 2016. 35. ​Ibid. 36. ​A demographic study by the Ford Foundation in 1975 came up with an estimate of 500,000 “excess deaths” (deaths that would not have happened other­wise). This encompasses non-­ Bengali deaths, and other war-­related c­ auses of death, such as famine. See George T. Curlin, Lincoln C. Chen, and Syed Babur Hussain, “Demographic Crisis: The Impact of the Bangladesh Civil War (1971) on Births and Deaths in a Rural Area of Bangladesh,” Population Studies 30, no. 1 (March 1976): 103. 37. ​Donald Bleacher, “The Politics of Genocide Scholarship: The Case of Bangladesh,” Patterns of Prejudice 41, no. 5 (2007): 491. 38. ​Pakistan Army, “1971 War,” https://­www​.­pakistanarmy​.­gov​.­pk​/­AWPReview​/­TextContent​ .­aspx​?­pId​=­197, accessed June 7, 2016. 39. ​Rahman Commission Report, 9. 40. ​Kiyoteru Tsutsui, “The Trajectory of Perpetrators’ Trauma: Mnemonic Politics around the Asia-­Pacific War in Japan,” Social Forces 87, no. 3 (2009): 1391. See also Ernest Gellner, Nations and Nationalism (Oxford: Blackwell, 1983). 41. ​Maurice Halbwachs, On Collective Memory (Chicago: University of Chicago Press, 1992), 38. 42. ​Jeffrey K. Olick and Joyce Robbins, “Social Memory Studies: From ‘Collective Memory’ to the Historical Sociology of Mnemonic Practices,” Annual Review of Sociology 24, no. 1 (1998): 111. 43. ​Astrid Erll, “Cultural Memory Studies: An Introduction,” in Cultural Memory Studies: An International and Interdisciplinary Handbook, ed. Astrid Erll and Ansgar Nünning (New York: De Gruyter, 2008), 5. 44. ​Frederic C. Bartlett, Remembering: A Study in Experimental and Social Psy­chol­ogy (Cambridge: Cambridge University Press, 1932). 45. ​James Young refers to “collected memory” rather than “collective memory” to account for the fragmentary and individualistic nature of memory. See James Young, The Texture of Memory (New Haven, Conn.: Yale University Press, 1994), xi. 46. ​Universal Period Review Report, Government of Bangladesh (2013), 2, https://­documents​ -­dds​-­ny​.­un​.­org​/­doc​/­UNDOC​/­GEN​/­G13​/­107​/­07​/­PDF​/­G1310707​.­pdf ​?­O penElement, accessed June 7, 2016. 47. ​Nayanika Mookherje, “Remembering to Forget: Public Secrecy and Memory of Sexual Vio­lence in the Bangladesh War of 1971,” Journal of the Royal Anthropological Institute 12, no. 2 (2006): 446.

234

k jell a nder son

48. ​Shahriar Kabir, ed., “Declaration of 71 Organ­izations: Try the War Criminals of ’71 at a

Special Tribunal” (Dhaka: Forum for Secular Bangladesh and Trial of War Criminals of 1971, 2008), 15. 49. ​Rahman Commission, 36. 50. ​Ibid., 77. 51. ​Yasmin Saikia, “History on the Line: Beyond the Archive of Silence: Narratives of Vio­ lence of the 1971 Liberation War of Bangladesh,” History Workshop Journal 58, no. 1 (2004): 285. 52. ​See Michel Foucault, Language, Counter-­Memory, Practice: Selected Essays and Interviews (Ithaca, N.Y.: Cornell University Press, 1977). 53. ​Erll, “Cultural Memory Studies,” 6. 54. ​Tsutsui, “The Trajectory of Perpetrators’ Trauma,” 1390. See also Robin Wagner-­Pacifici and Barry Schwartz, “The Vietnam Veterans Memorial: Commemorating a Difficult Past,” American Journal of Sociology 97, no. 2 (September 1971): 376–420. 55. ​James A. Holstein and Jaber F. Gubrium, The Self We Live By: Narrative Identity in a Postmodern World (New York: Oxford University Press, 2000), 157. 56. ​Stanley Cohen, States of Denial: Knowing about Atrocities and Suffering (Cambridge: Polity Press, 2008), 45. 57. ​Tsutsui, “The Trajectory of Perpetrators’ Trauma,” 1396. 58. ​­These schema build on Tsutsui’s “Seven Responses to the Collective Trauma of Perpetration,” as well as my own iteration of moral neutralization theory for genocide. See Tsutsui, “The Trajectory of Perpetrators’ Trauma,” 1392–1393, as well as Kjell Anderson, The Criminology of Genocide Perpetrators: Killing without Consequence (London: Routledge, 2017). 59. ​Tsutsui, “The Trajectory of Perpetrators’ Trauma,” 1393. 60. ​Ibid. 61. ​Rahman Commission, 20. 62. ​Ibid., 21. 63. ​Ibid., 22. 64. ​Ibid., 33. 65. ​Sara Hossain, “A Long and Winding Road; Justice and Accountability for War Crimes in 1971,” paper, International Conference on Genocide, Truth and Justice, Liberation War Museum, Dhaka, March 1, 2008, 65. 66. ​Ibid. 67. ​Joseph Allchin, “The Midlife Crisis of Bangladesh,” Foreign Policy, December 21, 2012. Available at: http://­foreignpolicy​.­com​/­2012​/­12​/­21​/­the​-­midlife​-­crisis​-­of​-­bangladesh​/­. 68. ​Committee for Resisting Killers and Collaborators of ’71, “Report on the Findings of the ­People’s Inquiry Commission on the Activities of the War Criminals and the Collaborators” (Dhaka: Ekattorer Ghatak Dalal Nirmul Committee, 2005), 9. 69. ​Gerry Simpson, Law, War and Crime (Cambridge: Polity Press, 2007), 13. 70. ​Lawrence W. Sherman, “Defiance, Deterrence, and Irrelevance: A Theory of the Criminal Sanction,” Journal of Research in Crime and Delinquency 30, no. 4 (1993): 448. See also Kjell Anderson, “Punishment as Prevention? The International Criminal Court and the Prevention of International Crimes,” in The International Criminal Court and Africa: Ten Years On, ed. Evelyn Ankumah (The Hague: Intersentia, 2016), 466–488. 71. ​Sherman, “Defiance, Deterrence, and Irrelevance,” 467. 72. ​“Discrepancy in Dhaka,” Economist, December 8, 2012. 73. ​Burundi is a good example of this: ­there, episodes of mass killing in 1972 and 1988 ­were Tutsi-­led, while atrocities arising from the 1993–2005 civil war involved both Tutsi and Hutu perpetrators.



Collective Crimes, Collective Memory 235

74. ​Morten Bergsmo and Elisa Novic, “Justice ­after De­cades in Bangladesh: National ­Trials

for International Crimes,” Journal of Genocide Research 13, no. 4 (2011): 505. 75. ​ Prosecutor v. Moulana Abdul Kalam Azad (ICT-­BD 05), January 21, 2013, para. 5. 76. ​Ibid., para. 316. 77. ​ Prosecutor v. Muhammad Kamaruzzaman (ICT-­BD 03), May 9, 2013, para. 578. 78. ​ Prosecutor v. Ashrafuzzaman Khan, Naeb Ali Khan, and Chowdhury Mueen Uddin (ICT-­BD 01), November 3, 2013, para. 11. 79. ​ Prosecutor v. Syed Mohammed Qaiser (ICT-­BD 04), December 23, 2014, para. 17. 80. ​Ibid., para 19. 81. ​Mark Osiel, Mass Atrocity, Collective Memory, and the Law (New Brunswick, N.J.: Transaction, 2012), 61. 82. ​Ibid.

ACKNOWL­E DGMENTS

This volume emerged from a research program whose aim was to integrate transitional justice into the more mainstream research areas of War, Holocaust, and Genocide Studies. The Royal Netherlands Acad­emy of Arts and Sciences (KNAW) offered five years of generous seed-­funding to my institution, the NIOD Institute for War, Holocaust, and Genocide Studies, to support this timely initiative. In the first place, I would like to thank Peter Romijn, Marjan Schwegman, Wichert ten Have, and Theo Mulder for entrusting me with this impor­tant proj­ect, which has broadened the scope of my thinking about the legacy of mass repression. I would also like to gratefully acknowledge the contributing authors for their diligence, willingness to engage with my critical comments, patience, and conscientiousness, but mostly for the excellent work they have done. I believe the questions they raise are just the beginning of the discussion. I look forward seeing it unfold. My sincerest appreciation goes to Mark Freeman, Carol Gluck, Stephan Parmentier, Wouter Veraart, and Richard Wilson, who ­were always on call to offer expert advice, read proposals, and brainstorm about pres­ent and f­uture directions of the field of transitional justice and my own research. Early on, a number of expert meetings helped us in determining the gaps in the field. Some of the participants are mentioned above, but I would also like to acknowledge o­ thers who assisted in this endeavor: Antoon de Baets, Berber Bevernage, Predrag Dojcinovic, Herman von Hebel, and Luc Huyse. The team at Rutgers University Press, especially Lisa Banning, has been exceptional—­engaged and highly supportive of this proj­ect from early on, making for an all-­around wonderful experience. Lastly, I would like to express my gratitude to the external referees for their careful scrutiny, enthusiasm, and very helpful comments and suggestions.

237

NOTES ON CONTRIBUTORS

nanci adler is Professor of Memory, History, and Transitional Justice at the

University of Amsterdam and Program Director of Genocide Studies at the NIOD Institute for War, Holocaust, and Genocide Studies (Royal Netherlands Acad­ emy of Arts and Sciences). She has authored Keeping Faith with the Party: Communist Believers Return from the Gulag (2012), The Gulag Survivor: Beyond the Soviet System (2002), and Victims of Soviet Terror: The Story of the Memorial Movement (1993), and published numerous scholarly articles and edited volumes on po­liti­cal rehabilitations, the memory of mass repression, and the consequences of Stalinism. Her current research focuses on transitional justice and the legacy of Communism. k jell anderson is an interdisciplinary scholar specializing in the study of

mass vio­lence and mass atrocities. He has held positions at the National University of Ireland, National University of Rwanda, University of Amsterdam, and Leiden University, and was a postdoctoral fellow in the Transitional Justice Program of the NIOD Institute for War, Holocaust, and Genocide Studies. He is the author of Perpetrating Genocide: A Criminological Account (2017) and numerous scholarly articles on genocide, mass atrocities, conflict, and transitional justice. His current research focuses on perpetrators of mass atrocities and the trial of Dominic Ongwen at the International Criminal Court. thijs b. bouwknegt is a historian and former journalist specialized in African

Affairs, International Criminal Justice, Transitional Justice, and Genocide Studies. He is Researcher at the NIOD Institute for War, Holocaust, and Genocide Studies and Lecturer at the University of Amsterdam and University of Utrecht. Since 2003, Bouwknegt has attended, monitored, and reported on the t­ rials at the International Criminal Tribunal for the former Yugo­slavia (ICTY), the International Criminal Tribunal for Rwanda (ICTR), the Special Court for Sierra Leone (SCSL), the Extraordinary Chambers in the Courts of Cambodia (ECCC), the Special Tribunal for Lebanon (STL), the International Criminal Court (ICC), and a range of national jurisdictions in Eu­rope and Africa. nicole l. immler is Associate Professor in the program on Globalisation and

Dialogue Studies at the University of Humanistic Studies in Utrecht. She is co-­editor of the special issue “Reconciliation & Memory: Critical Perspectives” (Memory Studies 5, no. 3 [2012]), the author of Das Familiengedächtnis der Wittgensteins (2011), and numerous scholarly articles on “the afterlife” of Holocaust restitution pro­cesses, including how reparations are experienced, the generational aspect 239

240 Contributors

in compensation politics, the interplay of individual and collective memory, and narratives of (in)justice. Her current research focuses on transitional justice and the legacy of colonialism. christian a xboe nielsen is Associate Professor of Southeast Eu­ro­pean Stud-

ies at Aarhus University in Denmark. He has worked as an analyst at the International Criminal Tribunal for the Former Yugo­slavia and the International Criminal Court, appeared as an expert witness in several international and domestic criminal and civil cases, and served as con­sul­tant to the Special Tribunal for Lebanon, the Ministry of Justice in Canada, and the Federal Prosecutor in Germany. Nielsen authored, among ­others, Making Yugo­slavs: Identity in King Aleksandar’s Yugo­slavia (2014), a Handbook on Assisting International Criminal Investigations (2011), and numerous articles in scholarly journals. His current research focuses on the police in Socialist Yugo­slavia and competitive victimhood in the historiographies of the former Yugo­slavia. stephan parmentier studied law, po­liti­cal science, and sociology at the Uni-

versities of Ghent and Leuven and sociology and conflict resolution at the Humphrey Institute for Public Affairs, University of Minnesota–­Twin Cities. He currently teaches sociology of crime, law, and ­human rights in the Faculty of Law of the University of Leuven and previously served as the Academic Secretary of the Faculty of Law (2002–2005) and Head of the Department of Criminal Law and Criminology (2005–2009). In 2010, Parmentier was elected Secretary-­ General of the International Society for Criminology (reelected in August 2014). He is on the Advisory Board of the Oxford Centre of Criminology and the International Center for Transitional Justice, is the founder and co-­general editor of the international book series on transitional justice (Intersentia Publishers), and editor of the Restorative Justice International Journal. vl adimir petrović is Se­nior Academic Researcher at the Center for the Study

of Eu­rope at Boston University and at the Institute of Con­temporary History in Belgrade. He held a postdoctoral position in the Transitional Justice Program of the NIOD Institute for War, Holocaust, and Genocide Studies in Amsterdam. Petrović authored The Emergence of Historical Forensic Expertise: Clio takes the Stand (2017), two monographs in Serbian, and around fifty articles, book chapters, and essays in Serbian and in En­glish. His current research interests are in the field of transitional justice, particularly in examining the role of historical narratives in war crimes t­ rials. mina r auschenbach is Lecturer at the Institute of Social Sciences of the Uni-

versity of Lausanne and a Research Associate at the Leuven Institute of Criminology (LINC) at the Catholic University of Leuven. Her current research interests focus on nonjudicial forms of transitional justice, on the role of identity and collective understandings of the past  in relation to  transitional  justice pro­cesses,

Contributors 241

and to diaspora transnational activism in postconflict settings. Rauschenbach’s publications include “Accused for Involvement in Collective Vio­lence: The Discursive Reconstruction of Agency and Identity by Perpetrators of International Crimes” (Po­liti­cal Psy­chol­ogy, 2015) and “The Perfect Data-­Marriage: Transitional Justice Research and Oral History Life Stories” (Transitional Justice Review, 2016). jeremy sarkin is Professor of Law at the University of South Africa and Distin-

guished Visiting Professor of Law and member of the Center for Research in Law and Society (CEDIS), Nova University Faculty of Law in, Lisbon. He was a member (2008–2014) and Chairperson-­R apporteur (2009–2012) of the U.N. Working Group on Enforced or Involuntary Disappearances. He served as an acting judge in 2002 and 2003 in South Africa, and as National Chairperson of the ­Human Rights Committee of South Africa from 1994 to 1998. Some of his books are Germany’s Genocide of the Herero (2011), Reparations for Colonial Genocides (2009), ­Human Rights in African Prisons (2008), and Reconciliation in Divided Socie­ties: Finding Common Ground (with Erin Daly, 2007). willia m a . schabas is Professor of International Law at Middlesex Univer-

sity, London, Professor of International Criminal Law and ­Human Rights at Leiden University, and Emeritus Professor of ­Human Rights Law at the National University of Ireland, Galway. He is the author of many books and articles on the abolition of capital punishment, genocide, ­human rights, and the international criminal tribunals. Schabas was a member of the Sierra Leone Truth and Reconciliation Commission and Chairman of the U.N. Commission of Inquiry on the 2014 Gaza Conflict. He is an Officer of the Order of Canada, a member of the Royal Irish Acad­emy, and holds several honorary doctorates. ma arten van cr aen is a postdoctoral researcher at the Leuven Institute of

Criminology of the University of Leuven. His research is supported by the Research Foundation-­Flanders (FWO). Van Craen’s interests include internal and external procedural justice, institutions’ legitimacy, citizens’ attitudes ­toward government institutions, policing, and ethnic minorities. His work has appeared in journals such as Justice Quarterly, Journal of Experimental Criminology, Criminal Justice and Be­hav­ior, and Eu­ro­pean Journal of Criminology. timothy willia ms is a research fellow at the Center for Conflict Studies at

Marburg University, Germany. His research deals with vio­lence, focusing on its dynamics particularly at the micro-­level, as well as its consequences for postconflict socie­ties. He has conducted extensive field research in Cambodia and has been awarded the Emerging Scholar Prize of the International Association of Genocide Scholars in 2017, as well as being named the Raphael Lemkin Fellow of the Armenian Genocide Memorial and Institute in 2015. Prior to his doctoral research, he studied at Mannheim University and at the London School of Economics. Williams has published in Terrorism and Po­liti­cal Vio­lence, Interna-

242 Contributors

tional Peacekeeping, Genocide Studies and Prevention, and Transitional Justice Review, among ­others. richard ashby wilson is the Gladstein Distinguished Chair of ­Human

Rights and Professor of Law and Anthropology at the University of Connecticut Law School. He is the author or editor of eleven books, including Writing History in International Criminal T ­ rials (2011), which was selected by Choice as an “Outstanding Academic Title.” He is the recipient of fellowships from the National Endowment for the Humanities, the Institute for Advanced Studies, Prince­ton, and the Russell Sage Foundation. Wilson’s latest book, Incitement on Trial: Prosecuting International Speech Crimes (2017), advocates a preventative approach to inciting speech and proposes a new risk-­assessment model for evaluating which types of speech are most likely to cause vio­lence.

INDEX

Note: Italicized page numbers indicate illustrations. Page numbers followed by t indicate tables. Abu Zubaydah v. Poland, 42 Acton, Lord, 23 Afghanistan, amnesty law in, 59 Africa. See South Africa, conditional amnesty and truth recovery African National Congress (ANC), 57, 62, 64 Ahmetašević, Nidžara, 187 Ährenthal, Count Alois Lexa von, 30–31 Akayesu, Jean-Paul, 102, 108 Albania, 182, 184–185 ambiguity, collective crimes and claim of, 225 Amnesty Act, in Spain, 48 amnesty law. See conditional amnesty, truth recovery and Anderson, Benedict, 221 anti-Stalinist organization, 4 Antonetti (judge), 109 apartheid, 57, 58, 63–64, 66, 68, 179. See also Malan trial; South Africa Arendt, Hannah, 98, 132 Argentina, 78, 125 Armenian genocide, 8, 50, 199, 200, 222; “passage of time” and right to truth, 47, 48 Awami League, 217, 218–221, 226, 227, 228, 229 Azad, Abul Kalam, 229 AZAPO decision, in South Africa, 62 Babić, Milan, 102 Bagosora, Théoneste, 130–131 Baines, Erin, 200 Bangladesh. See collective crimes and collective memories, Bangladesh and Bangladesh National Party (BNP), 220, 228 Barayagwisa, Jean-Bosco, 103 Barbie, Klaus, 98 Barkan, Elazar, 4, 151 Bartlett, Frederick, 222 Baudouin, François, 22 Belfast Guidelines on Amnesty and Accountability, 60 Benda, Julien, 27

Bensouda, Fatou, 119–121 Berman, Harold, 28–29 Biehl, Amy, 65 Biharis, 217, 218, 219, 220, 223, 226 Bijeljina Commission, in BosniaHerzegovina, 80 Bikindi, Simon, 102, 133 blanket amnesties, 59, 60, 62 Blé Goudé, Charles, 103, 119, 120, 121–122 Bleiburg massacre and commemoration of, 175, 176, 182–186, 188 Blood, Archer, 217 Blum, Léon, 26 Bolton, Charles, 21 Bosnia-Herzegovina (BiH): competing narratives and, 4–5; competitive and collective victimhood and, 178; meaning of truth-seeking in, 83; methodology of survey, 81–82; qualitative population-based research and, 78–88; survey forums for truth-telling, 83–87, 85t, 86t, 87t; war and postwar situation in, 79–81 Bou Meng, 204 Bouris, Erica, 200 Bräuchler, Birgit, 165 Brđanin, Radoslav, 102 Breisach, Ernst, 32 Buisson, Ferdinand, 26 Butare trial, 124 Buthelezi, Mangosuthu, 57 Cambodia. See universal victimhood, in Cambodia Cawi, Ibu, 149, 156, 158, 162 Central Intelligence Agency, 41, 42 Chad, 125, 126 Chandler, David, 126 Choeung Ek fields, 196, 197, 204 “choiceless decisions,” in Cambodia, 200, 201, 207 Chum Mey, 207

243

244 Index Cloete, André, 67 Cohen, Stanley, 224 collective crimes and collective memories, Bangladesh and, 213–235; collective memories of 1971 war, 218–221; collective memory as contested concept, 215; identity and formation of collective memory, 221–224; International Crimes Tribunal and, 213, 214, 217, 227–230, 231; key features of collective crimes, 214–215; Liberation War (1971) from Pakistan, 213, 214, 216–218; outcomes of wars and reframing of narratives of collective crimes, 224–227 colonialism: 11, 49; Kenya and, 111; Rwanda and, 2, 112. See also reparation claims, Rawagede and Commission on Human Rights, of UN, 39, 40, 77 Committee for Dutch Debts of Honor (KUKB), 152, 169 compensation, for victims. See reparation claims, Rawagede and competing narratives, transitional justice, 3–5, 13 competitive and collective victimhood, 175–193; collective guilt concept discarded for individual responsibility, 180; critical approach to, 189–190; education and, 185, 190; former Yugoslavia and, 175–179, 181–189; negative consequences for transitional justice, 186–190; perceptions of justice and, 179–181; perpetrator-victim dichotomy and, 178; post–World War II rehabilitations and, 175–176; purpose of criminal trials, 180; transitional justice developments, 178 conditional amnesty, truth recovery and, 56–74; benefits of, 64; places used in, 61; protecting of truth and discouraging of lies, 68–69; South Africa’s process and, 57–58, 62–68; South Africa’s process, lessons of, 56, 69–71; validity in international law, 59–60 conformity crimes, as collective crimes, 214 Congress of the Islamic Community in Bosnia and Herzegovina, 187 Convention on Non-Applicability of Statutory Limitations to War Crimes and Crimes against Humanity, of UN, 37–38, 50

Convention on the Prevention and Punishment of Genocide, of UN, 50, 108 Coombs, Nancy, 131 Côte d’Ivoire, International Criminal Court and, 103–104, 119–125, 132, 133, 135 Council of Europe, 36, 38, 45 Cradock Four, in South Africa, 64 crime, legal definition of, 113 Criminal Law (Dressler and Garvey), 113 criminal tribunals. See propaganda and historical narratives, international criminal tribunals and Croatia, 102, 175–177, 182, 185, 186–187 Croatia v. Serbia, 50 Cruvellier, Thierry, 124, 133 Dayton Peace Agreements (1995), 79 decolonization: human rights and, 166 167; Rawagede and, 11, 149, 158, 159, 181 De Gaetano (judge), 46, 47 de Greiff, Pablo, 2, 48, 52, 167–168 Democratic Kampuchea. See universal victimhood, in Cambodia Democratic Republic of the Congo (DRC), 120, 131, 135 Denmark, 181 Des Forges, Alison, 108, 124, 125, 129–130 Dickens, Charles, 123 Dingammadji, Arnaud, 126 Dodd, Thomas J., 100–101 Dodik, Milorad, 188 Donia, Robert, 126, 132 Douglas, Lawrence, 178–179 Doyle, Kate, 7 Dressler, Joshua, 113 Dreyfus, Alfred, 24–25, 25, 28, 29 Dreyfus, Mathieu, 24 Dreyfus affair, historians and protection of human rights, 24–28, 32 Drumbl, Mark A., 113 Dugard, John, 60 Durban Conference on Racism, Racial Discrimination, Xenophobia, and Related Intolerance, 39–40 Dzhugashvili, Yevgeniy Yakovlevich, 51 East Timor, 124 Ecole des chartes, Dreyfus affair and, 26–27 Eichmann, Adolph, 98, 129, 132, 178–179

Index 245 El Masri v. the “former Yugoslav Republic of Macedonia,” 41 Ellis, Stephen, 126 equivalency, collective crimes and claim of, 225 Esterhazy, Ferdinand, 26–27 European Convention on the NonApplicability of Statutory Limitation to Crimes against Humanity and War Crimes, 38, 41–43, 45–46, 51–52 European Court of Human Rights, procedural right to truth and, 37, 40–47, 51–52 expert witnesses. See historians, testimony in courts “expressivist” function of law, 113 Extraordinary African Chambers (EAC), 124, 126, 131 Extraordinary Chambers in the Courts of Cambodia (ECCC), 12, 60, 125, 194, 195, 196–197, 203–204, 207 Faulkner, William, 13 Federation of Bosnia and Herzegovina. See Bosnia-Herzegovina (BiH) Feindbilder (negative images), competitive and collective victimhood, 179 Fleg, Edmond, 26 Foucault, Michel, 224 France, Anatole, 26 Franco, Francisco, 48 Fraser, Nancy, 166 Friedjung, Heinrich, 29–31 Friedjung affair, historians and national interest, 28–32 Fritzsche, Hans, 99 Fulford, Adrian, 104 gacaca courts, 5, 6, 204 Garvey, Stephen P., 113 Gbagbo, Laurent, 103, 119–124, 126, 128–129, 130, 132–133, 135 Gbagbo, Simone, 121 Geneva Conventions, amnesty and, 59–60 Genocide Convention, 50, 179 Ginzburg, Carlo, 127 Giry, Arthur, 26, 27 Gluck, Carol, 7, 150 Gotovina, Ante, 186

Grand Chamber, of ECHR, 41–43, 45–48, 51–52 Guatemala, 78, 125 Habré, Hissène, 126 Habsburg monarchy, 29, 31 Halbwachs, Maurice, 221 Hamoodur Rahman Commission, 221, 223, 226 Harris, Ruth, 26 Hausner, Gideon, 129 Havet, Louis, 26 Hayner, Priscilla, 78 Hegel, G. W. F., 22 Helms, Elissa, 178 Herr, Lucien, 26 higher purpose claim, collective crimes and, 225, 226 Hinton, Alexander Laban, 196 historians, testimony in courts, 21–36; human rights and Dreyfus affair, 24–28, 32; law and history, 21–23; national interest and Friedjung affair, 28–32; obligations of historians, 32–33; tribunals and, 124–126 historical narratives. See propaganda and historical narratives, international criminal tribunals and Holocaust. See International Military Tribunal (IMT) at Nuremberg Hormayr, Joseph von, 22 Houphouët-Boigny, Félix, 122 human rights, historians and Dreyfus affair, 24–28, 32 Human Rights Council, of UN, 40 Hun Sen, 197 Hutus, 106, 130, 199. See also International Criminal Tribunal for Rwanda (ICTR) Ieng Sary, 196, 200, 203 Ieng Thirith, 203 India, Bangladesh and, 215, 216–217, 219–220, 221 Indonesia. See reparation claims, Rawagede and I(nformation), D(ialogue), and P(rocess) approach, to transitional justice, 76, 89–90 Inkatha Freedom Party (IFP), in South Africa, 57–58, 66, 67–68

246 Index Inter-American Commission on Human Rights, of UN, 38 Inter-American Court of Human Rights, 47 intergenerational crimes, as collective crimes, 214 International Committee of the Red Cross, 47 International Convention for the Protection of All Persons from Enforced Disappearance, 48–49 International Court of Justice, 5, 37, 50–51, 187 International Crimes Tribunal (ICT) of Bangladesh, 213, 214, 217, 227–230 International Crimes (Tribunals) Act (1973), 227 International Criminal Court (ICC), 126, 128, 131, 179, 180; Congo and, 131, 135; CÔte d’Ivoire and, 103–104, 119–125, 132–133, 135; Kenya and, 103–104, 111, 131, 135; propaganda and historical narratives and, 97–98, 103–104, 107; Rome Statute, 38, 60, 179 International Criminal Tribunal for Rwanda (ICTR), 2, 5, 128; nonlegal evidence and historical narratives, 129–131; propaganda and historical narratives and, 97–98, 102–103, 105–108, 112; trial records and, 119, 124–126, 128–133, 135 International Criminal Tribunal for the former Yugoslavia (ICTY), 2, 79, 80, 128; competing narratives and, 3–5; competitive and collective victimhood and, 175–179, 181–189; propaganda and historical narratives and, 98, 101–102, 106–107, 108–109; trial records, as historical source, 125–126 International Day of Commemoration and Dignity of the Victims of Genocide, 50 International Historical Congress, 28 International Military Tribunal (IMT) at Nuremberg, 37, 38, 43, 44; collective versus individual responsibility, 178, 180; propaganda and historical narratives and, 97, 99–101, 105, 107 Investigation Task Unit (ITU), in South Africa, 69 Ivory Coast. See Côte d’Ivoire Jackson, Michael, 136 Jamaat-e-Islami, 217, 220–221, 227, 228, 229, 230

Janowiec and Others v. Russia, 46–47, 51 Japan, 149, 158, 181 Jaspers, Karl, 180, 225 Jaures, Jean, 27 Jichi, Mohammed Sam, 123 Jihad Watch, in Bangladesh, 229 Joinet, Louis, 38–39, 40, 52, 77–78 Kagame, Paul, 112, 219 Kaing Guek Eav (Duch), 195, 197, 203, 205 Kamaruzzaman judgment, 229–230 Kangura (Wake Others Up) (newspaper), 103, 106, 108 Karadzic, Radovan, 132, 187, 188 Katyn massacre, right to truth and, 42–47, 44, 51 Keller (judge), 46, 47 Kenya, 103–104, 111, 131, 135 Kerber, Linda, 21 Khan, Tikka, 217 Khan, Yahya, 217, 230 Khieu Samphân, 195, 197, 200, 203 Khmer Rouge, universal victimhood and, 194, 195–196, 197, 198–208 Kingdom of Serbs, Croats, and Slovenes, 182 Koenders, Bert, 159 Komarica, Franjo, 188 Kordić, Dario, 102 Kosovo, 177, 182–185 Kousser, Morgan, 21 Kovler (judge), 51 Kraus, Dr. (translator), 31 KwaMakutha events, in South Africa, 67 KwaZulu-Natal region, 57–58, 67, 69 Laffranque (judge), 46, 47 Langlois, Charles, 23 Lasteyrie, Robert de, 26 Lattanzi (judge), 107 Layem, Ibu, 149, 161 League of Communists of Yugoslavia, 183 ledger metaphor, collective crimes and, 225, 226 Lefranc, Abel, 26 “Legacy Talk,” 118. See also trial records, as historical source legal causation (“proximate” cause), 110 Leuven, University of, 79, 81, 82 Leydesdorff, Selma, 168

Index 247 Liberia, 2, 128 Libya, 129 Liem, Suzanne, 153 Ligue des droits de l’homme (Human Rights League), 24 Lon Nol, 198, 205 Lovrić, Jelena, 176–177 Lubanga, Thomas, 104, 133, 135 Luthuli, Daluxolo Wordsworth, 67–68 MacDonald, Eric, 121–122 Macedonia, 41, 177, 184–185 Maier, Charles, 33 Maitland, Frederic William, 23 Malan, Magnus, 57, 58 Malan trial, 57, 66–68, 70 Mallinder, Louise, 59 Markač, Mladen, 186 Masaryk, Tomáš, 31 Mascarenhas, Anthony, 218 Massacre of Rawagede, The (documentary), 152 mass crimes, as collective crimes, 214 Massidda, Paolina, 122 material causation (“cause-in-fact”), 110 Mazower, Mark, 181 McEvoy, Kieran, 76 McGregor, Katharine, 166 McGregor, Lorna, 76 McGrew, Laura, 203, 207 Mechanism for International Criminal Tribunals (MICT), of UN, 80, 124 Memorial (anti-Stalinist organization), 4 Meyer, Paul, 26, 27 microcredit initiative, in Rawagede, 162–163, 168 Mihailović, Dragoljub “Draža,” 175 Mijić, Ana, 186 Milosevic, Slobodan, 2 Minow, Martha, 61 Mirbeau, Octave, 26 Mladić, Ratko, 188 Molinier, Auguste, 26–27 Molinier, Emile, 26, 27 Mollah, Abdul Quader, 228 Monji, Andi, 153, 169 Monod, Gabriel, 24–26, 25, 28 Møse, Erik, 131 Mukti Bahini, 216, 218, 220, 221, 223 Muslim League, 227, 230

Nahimana, Ferdinand, 102, 103, 105–106, 108, 109 national interest, historians and Friedjung affair, 28–32 Nedić, Milan, 175 Netherlands. See reparation claims, Rawagede and Neue Freie Presse, 29, 30 Ngeze, Hassan, 103, 108 Niang, Mandiaye, 132 Niazi, Amir Abdullah Khan, 217–218 Nielsen, Christian Axboe, 126 Nieuwoudt, Jan Anton, 67 Nikolic, Tomislav, 5 Nizam e Islam, 227, 230 Nizami, Maulana Matiur Rahman, 217 Njuz​.n­ et (newspaper), 184 Noronha-DiVanna, Isabel, 28 Northern Ireland, 42 Novick, Peter, 23, 32, 179 Ntuli, Victor, 57 Nuon Chea, 195, 197, 200, 203 Obama, Barack, 189 Oberschall, Dr. Anthony, 108–109 Omar, Dullah, 63 “On the Study of History” (Acton), 23 Open Society Justice Initiative, 42 Operation Marion, in South Africa, 57–58, 66 Orentlicher, Diane, 39, 47, 52, 77–78 Orie, Alphonse, 128 Osiel, Mark, 231 “Österreich-Ungarn und Serbien” (Friedjung), 29 Ottoman empire, 181–182, 199, 201 Ouattara, Alassane, 122 Pakistan, and Bangladesh war of liberation, 213–214, 215, 216–221, 223–224, 225–227, 229–231 Paleologue, Maurice, 28 Papon, Maurice, 98 pardons, amnesties differ from, 59 Parmentier, Stephan, 150 Paturusi, Shafiah, 153 Pebco (Port Elizabeth Black Civic Organisation) Three, 64 People’s Revolutionary Tribunal, in Cambodia, 203

248 Index perpetrator-victim dichotomy, in former Yugoslavia, 178 Poland, 42, 43, 181 political crimes, as collective crimes, 214 Pol Pot, 196, 200, 203 Pondaag, Jeffry, 152–153 population-based research, on transitional justice: advantages and limitations of quantitative approach, 89; in BosniaHerzegovina, 78–88; common aims and features of, 76–77; I(nformation), D(ialogue), and P(rocess) approach to, 76, 89–90; qualitative and quantitative forms of, 76–78; truth-seeking as key aspect of, 77–78 postcolonialism, 112, 150–151; nostalgia and, 159. See also reparation claims, Rawagede and propaganda and historical narratives, evidentiary weaknesses and, 105–110; expressions of legal causation and moral judgments, 110–114; International Criminal Court (ICC), 97, 98, 103–104, 107; International Criminal Tribunal for Rwanda (ICTR), 97–98, 102–103, 105–106, 107–108; International Criminal Tribunal for the former Yugoslavia (ICTY), 97–98, 101–102, 106–109; international criminal tribunals and, 97–117; International Military Tribunal (IMT) at Nuremberg, 97, 99–101, 105, 107; legal definition of crime and, 113; propaganda in international criminal law, generally, 98–99 Prosecutor v. Ferdinand Nahimana, Jean-Bosco Barayagwisa, Hassan Ngeze (“Media trial”), 103 Protection and Promotion of Human Rights report to UN, 39, 52 Prunier, Gérard, 126 Qaiser, Syed Mohammed, 230 Qutubuddin Aziz movement, 218 Radio Télévision Libre des Milles Collines (RTLM), 103, 105–106, 108 Rahman, Sheikh Mujibur, 219, 220 Rahman, Ziaur, 220, 227 Ranke, Leopold von, 22–23 “Rankean rage,” trials and truth and, 129 Rasem, Ibu, 159–160

Rawagede (Balongsari). See reparation claims, Rawagede and Rawagede Foundation, 154–155, 158, 162, 164, 167 Razakars, 217, 220, 223, 229 RECOM truth-finding commission, 80 Reichpost, 29, 30 Reinach, Salamon, 26 reparation claims, Rawagede and, 149–174; compensation as social process, 170; court case history, 151–153; emancipation-transformation and, 167–168; impact on individuals and village, 153–162, 168–169; monetary compensation measures, 150–151; neighbors’ perspectives on monetary compensation, 162–165; Rawagede name change, 164; recognition and, 166–167; transitional justice and villages, 165–166 Republika Srpska, 79, 80, 188 Reshetar, Milan, 31 Revue historique, 24 Rieff, David, 189 right to justice, 78 right to know, 77–78 right to reparation, 78 right to truth, statutory time limitations and, 37–55; history of declarations of, 37–40; Katyn massacre and, 42–47, 51; “passage of time” approach to, 47–48; “procedural,” at European Court of Human Rights, 40–42; Spanish Civil War and, 48–50; temporal limits on Genocide Convention, 50 Rohde, Jon E., 217 Romania, 41 Rome Statute, of International Criminal Court, 38, 60, 179 Roosevelt, Eleanor, 150 Rothman, David J., 22 Rothman, Hal, 22 Rousso, Henry, 21–22, 98 Royal Netherlands East Indies Army (KNIL), 152 Russia. See Katyn massacre Ruto, William, 103, 104, 107 Ruzindana, Dr. Mathias, 108 Rwanda. See International Criminal Tribunal for Rwanda (ICTR) Rwanda Patriotic Front (RPF), 219

Index 249 S v. Peter Msane and Nineteen Others. See Malan trial S-21 prison, in Cambodia, 195, 198, 201, 204–206, 206. See also Tuol Sleng Genocide Museum Saih, Pak, 149, 152, 156, 159 Saikia, Yasmin, 223–224 Sang, Joshua Arap, 103–104, 107 Sankoh, Foday, 129 Sarajevo Commission, 80 Sariman, Ibu, 149 Saxbe, William, 217 Schabas, William, 8 Schiller, Friedrich, 22 Séailles, Gabriel, 28 Seignobos, Charles, 23 Serbia, 2, 3, 29–31, 50; competing narratives and, 4–5; victimhood and, 175–176, 182, 183, 184, 185, 187. See also Bosnia-Herzegovina (BiH) Šešelj, Vojislav, 98, 106–107, 108–109, 111, 132 Seton-Watson, R. W., 31–32 Shaheed, Farida, 7 Sierra Leone. See Special Court for Sierra Leone (SCSL) Šilih v. Slovenia, 46 Simpson, Gerry, 228 “Skopje 2014” museum exhibition, 184 slavery and slave trade, 39–40, 50, 181 Son Sen, 200 South Africa, conditional amnesty and truth recovery, 57–58; lessons of, 56, 69–71; process of, 60, 62–68 South African Constitutional Court, AZAPO decision, 62 South African Defence Force (SADF), 57–58, 63, 67 Sow, Malick, 133 Spalajković, Miroslav, 29–30 Spanish Civil War, right to truth and, 48–50, 52 Special Court for Sierra Leone (SCSL), 119, 124, 126, 128, 129, 135 Special Panels for Serious Crimes (SPSC), in East Timor, 124 Special Rapporteur on cultural rights. See Shaheed, Farida Special Rapporteur on truth, justice, reparation, and guarantees of non-recurrence. See de Greiff, Pablo

Special Tribunal for Lebanon (STL), 125 Srebrenica massacre and Srebrenica Commission, 5, 80, 84, 168, 187, 189 Stalin, Joseph, 4, 51 “Stilled Lives: Photographs from the Cambodian Genocide” exhibition, 205 Stojanović, Dubravka, 184 Streicher, Julius, 99–101, 100, 105, 111 Stürmer, Der (The Attacker), 99, 105 Subotić, Jelena, 185 Sukarman, Pak Dewan, 154–155, 158–159, 162, 163–167 Sukarman, Sri, 161, 166 Sullivan, Daniel, 189 Suparta (Rawagede businessman), 162–164 Tadić, Duško, 101–102 Ta Mok, 200 Taswi family, 161 Taylor, A.J.P., 31 Taylor, Charles, 2, 128, 129, 132, 133 Tijeng family, 153–156, 159, 167 time limits (statutory). See right to truth, statutory time limitations and Tito, Josip Broz, 183 Tolimir, Zdravko, 188 Touvier, Paul, 98 transitional justice, 75–93; competing narratives and, 3–5, 13; conflicting understandings of past and, 90–91; culture of repression and, 2–3; current “age” of, 1–3, 21, 124, 135; expansion of field of study, 75–76; failure to meet expectations, 5, 7–8; framing of past and, 13–14; I(nformation), D(ialogue), and P(rocess) approach to, 76, 89–90; new phase of, 2; pillars and, 61; variety of approaches to redressing harm, 1. See also population-based research, on transitional justice Transitional Justice from Below: Grassroots Activism and the Struggle for Change (McEvoy and McGregor eds.), 76 Traynor, Patrick, 126 trial records, as historical source, 118–145; Côte d’Ivoire and ICC, 119–124, 132–133, 135; historical records in courtrooms, 119–124; limitations of and omissions from trial records, 131–136; Rwanda and ICTR, 119, 124–126, 128–133; tribunals and phases of,

250 Index trial records, as historical source (cont.) 124–126; use of nonlegal evidence and historical narratives, 128–131; writing history distinguished from enacting justice, 127–128 Trumbull, Charles, 59 truth and reconciliation commission, Bosnia-Herzegovina and, 84, 86–88 Truth and Reconciliation Commission (TRC), in South Africa, 2, 57–58, 62–70, 65, 78, 150 truth commissions, population-based research and, 78 truth recovery. See conditional amnesty, truth recovery and truth-seeking, as key aspect of populationbased research, 77–78 Tsutsui, Kiyoteru, 221–222, 224, 231 Tuol Sleng Genocide Museum, 196, 197, 203, 204–207 Tutsis, 5, 104, 105–106, 108, 111, 130, 131, 199. See also International Criminal Tribunal for Rwanda (ICTR) Tutu, Bishop Desmond, 63 Tutu, Trevor, 64 Übersberger, Dr. Hans, 31, 32 Uddin judgment, 230 United Democratic Front (UDF), in South Africa, 57 United Nations: Commission on Human Rights (Human Rights Council), 39, 40, 77; Convention on Non-Applicability of Statutory Limitations to War Crimes and Crimes against Humanity, 37–38, 50; Convention on the Prevention and Punishment of Genocide, 50, 108; General Assembly, 37–38, 40, 49–50; Inter-American Commission on Human Rights, 38; Mechanism for International Criminal Tribunals (MICT), 80, 124; Protection and Promotion of Human Rights report to, 39, 52; Special Rapporteurs of, 2, 7, 48, 52, 167–168 United States, competitive victimhood and, 189 Universal Declaration of Human Rights, 37, 40, 51

universal victimhood, in Cambodia, 194–212; Cambodian society’s acceptance of, 199–203; ECCC and, 12, 60, 125, 194, 195, 196–197, 203–204, 207; fractures in narratives of, 202–203; Khmer Rouge cadres and, 194, 195–196, 197, 198–208; processes of dealing with past, 196–197; transitional justice mechanisms and, 203–207; Tuol Sleng Genocide Museum and, 196, 197, 203, 204–207 Vann Nath, 205 Vergangenheitsbewältigung, 180–181 victimhood. See competitive and collective victimhood; universal victimhood, in Cambodia “Victims and Perpetrators? Testimony of Young Khmer Rouge Comrades” exhibition, 205 Victims Association of Democratic Kampuchea (Ksaem Ksan), 207 Vienna Convention on the Law of Treaties, 43, 50 Violi, Patrizia, 196 Viollet, Paul, 26 Vučić, Aleksandar, 187–188 Wanti, Ibu, 156, 157, 160 War Crimes Chamber of the State Court of Bosnia and Herzegovina (WCC), 80 Warjo (Rawagede survivor), 159, 160, 161, 167 Weltgeschichte als Weltgericht (Schiller and Hegel), 22 Wilson, Richard, 7, 127 Winter Olympics (1984), 183 Working Group on Enforced or Involuntary Disappearance, Spain and, 49 Youk Chhang, 206–207 Yugoslavia: naming history of, 182. See also International Criminal Tribunal for the former Yugoslavia (ICTY) Zahar, Alexander, 106 Zegveld, Liesbeth, 152, 169 Ziemele (judge), 46, 47 Zola, Emile, 24–27 Zwaan, Tjeerd de, 152