Nội dung

FA HSRC APARTHEID (Converted)-1 5/28/04 1:22 PM Page 1 Free download from www.hsrcpress.ac.za C Composite M Y CM MY CY CMY K OVERCOMING Can truth reconcile a divided nation? APARTHEID FA HSRC APARTHEID (Converted)-1 5/28/04 1:22 PM Page 2 Free download from www.hsrcpress.ac.za C M CM MY CY CMY K OVERCOMING Can truth reconcile a divided nation? APARTHEID James L Gibson Composite Y FA HSRC APARTHEID (Converted)-1 5/28/04 1:22 PM Page 3 C M Y CM MY CY CMY K Free download from www.hsrcpress.ac.za Published in South Africa by HSRC Press Private Bag X9182, Cape Town, 8000, South Africa www.hsrcpress.ac.za First published in the USA by the Russell Sage Foundation 112 East 64th Street, New York, NY 10021 © 2004 Russell Sage Foundation All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the Russell Sage Foundation. Cover design by Jenny Young Production by comPress Distributed in South Africa by Blue Weaver Marketing and Distribution, PO Box 30370, Tokai, Cape Town, 7966, South Africa. Tel: +27 +21-701-4477 Fax: +27 +21-701-7302 email: booksales@hsrc.ac.za Distributed in the United States by CUP Services 750 Cascadilla Street, PO Box 6525, Ithaca, NY 14851 Tel: (800) 666-2211 or (607) 277-2211 Fax: (800) 688-2877 or (607) 277-6292 www.russellsage.org Distributed worldwide, except South Africa and the United States, by NBN Plymbridge Estover Road, Plymouth, PL6 7PZ, United Kingdom Tel: +44 +1752-20-2301 Fax: +44 +1752-20-2333 In South Africa ISBN 0 7969 2071 0 PB In the rest of the world ISBN 0 87154-312-5 HB Composite Free download from www.hsrcpress.ac.za This book is dedicated to the pillars of South Africa’s truth and reconciliation process, Charles Villa-Vicencio, Alec Boraine, and Desmond Tutu, who accomplished far more for South Africa than even they dreamed possible. Free download from www.hsrcpress.ac.za Free download from www.hsrcpress.ac.za CONTENTS About the Author ix Preface and Acknowledgments xi CHAPTER 1 Does Truth Lead to Reconciliation? 1 CHAPTER 2 Apartheid’s Legacy in Contemporary South Africa: Experiences, Attributes, and Attitudes of the Sample 28 CHAPTER 3 South African Collective Memories 68 CHAPTER 4 Interracial Reconciliation 117 CHAPTER 5 Truth, Reconciliation, and the Creation of a Human Rights Culture 176 Tolerance: The Minimalist View of Reconciliation 213 CHAPTER 7 Judging the Fairness of Amnesty 258 CHAPTER 8 The Legitimacy of the Political Institutions of the New South Africa 289 Lessons for South Africa’s Future and for the World 328 Appendix A: The Design of the Survey 347 Appendix B: The Questionnaire 353 Notes 409 References 441 Index 457 CHAPTER 6 CHAPTER 9 Free download from www.hsrcpress.ac.za Free download from www.hsrcpress.ac.za ABOUT THE AUTHOR JAMES L. GIBSON is Sidney W. Souers Professor of Government at Washington University in St. Louis, fellow at the Centre for International and Comparative Politics and Professor Extraordinary in Political Science at Stellenbosch University, South Africa, and Distinguished Visiting Research Scholar at the Institute for Justice and Reconciliation, South Africa. Free download from www.hsrcpress.ac.za Free download from www.hsrcpress.ac.za PREFACE AND ACKNOWLEDGMENTS While this book was being written, a low-scale civil war was taking place in American political science. Dubbed the “perestroika” movement, some academic political scientists were voicing a number of complaints against “scientific” political science. Many of these protestations were foolish and ill informed; some, however, were not. Supporters of the perestroika movement argue that contemporary political science in the United States is too little concerned with politics. By this the critics mean that political scientists are too focused on methods and theory and have devoted too little attention to studying and analyzing the important political issues and controversies of our time. My own view is that it is impossible to expend too much effort on method and theory, given the range of unresolved methodological and theoretical problems characterizing our field. But at the same time, the complaint that politics is too often ignored has some validity. Consequently, this study is motivated in part by my desire to take important theories of political psychology and intergroup relations and advanced methodological techniques and marry them to what I judge to be one of the most important questions facing transitional polities: does truth lead to reconciliation? This book therefore purports to contribute to both policy analysis and theory building and scientific hypothesis testing. The policy question is obvious: I seek to assess whether South Africa’s truth and reconciliation process in fact achieved the goals it set for itself. In particular, I examine the degree to which South Africans are “reconciled,” and then ask whether evidence exists to suggest that the truth and reconciliation process contributed to this reconciliation. Mine is not a strong research design from the point of view of policy analysis, since I am forced to rely on cross-sectional empirical evidence collected near the end of the truth and reconciliation process. But certainly one objective of this book is to draw Free download from www.hsrcpress.ac.za xii Preface and Acknowledgments some conclusions about whether the South African experiment ought to be attempted in other deeply divided societies. Generally, with some important caveats, I conclude that the process did indeed contribute to reconciliation and therefore that others may wish to borrow from the South African experience in trying to come to terms with their own repressive pasts. In addressing the hypothesis that truth leads to reconciliation, I mobilize a body of social scientific theories in an attempt to give a full accounting of why some South Africans are reconciled and others are not. For instance, an important contribution of this book lies in its test of the so-called contact hypothesis. Social psychologists have long put forth the hypothesis that positive interracial attitudes emerge from interpersonal interactions between people of different races. This hypothesis has rarely been tested outside Western developed democracies; adding evidence from a multiracial political system like South Africa can contribute significantly to expanding the generalizability of the theory. This is just one example of the multiple ways in which theory is mobilized and tested in my attempt to understand individual-level reconciliation in South Africa. My dual purposes—scientific hypothesis testing and policy evaluation—are not incompatible. The perestroika complaint that science ignores politics is a statement only about contemporary trends and says nothing about the inevitability of such a disjuncture. A central assumption of my research is that science must inform policy and that only through rigorous scientific inquiry can we begin to draw conclusions about a question as broad (and as profound) as that of whether truth is associated with reconciliation. Thus, my goal is to add something to our understanding of the policy issues surrounding reconciliation and transitional justice, while also making a contribution to several specific bodies of social scientific theory. No research project of this scale is solely the product of the efforts of an individual scholar, and this one is no exception. Without the support of the Law and Social Sciences Program of the National Science Foundation (grant SES 9906576), this project could not possibly have been conducted.1 I am especially indebted to program director Marie Provine for supporting this research. As Free download from www.hsrcpress.ac.za Preface and Acknowledgments xiii I have noted in other contexts, I admire no institution or agency more than I admire NSF. Its commitment to truth is as pure as any of which I am aware. This project is a collaborative effort with Amanda Gouws of the Department of Political Science at Stellenbosch University. Since the day I first set foot in South Africa, Amanda has been an invaluable colleague and a dear friend. We do not see eye to eye on all issues of South African politics—what two strong-minded scholars would ever fulfill that condition?—but I have learned immeasurably from my association with Amanda, and for that I am deeply thankful. I also acknowledge my appreciation to the Department of Political Science at Stellenbosch University for naming me a “Professor Extraordinary in Political Science” while this research was being undertaken. I also acknowledge my great debt to Chris Willemse. In addition to being a dear friend, Chris has given me a second set of eyes and ears when it comes to South African political history and contemporary affairs. Countless conversations with Chris have begun with the query “Chris, why is it that . . . ?” I have rarely encountered nonacademics with such keen insights. This project has profited immeasurably from his wisdom. This project soon found its natural home at the Institute for Justice and Reconciliation (IJR) in Cape Town (where I am Distinguished Visiting Research Scholar). My association with the IJR is somewhat curious in that the institute has a policy-oriented action mission, while I consider myself nothing more or less than a scientist. Perhaps because we are so different, our collaboration has been so fruitful. I deeply admire the director of the institute, Charles Villa-Vicencio, whose strength of commitment and intensity of effort in building a multiracial and democratic South Africa has never failed to amaze me. I have learned profusely from my association with the institute. As a white person from outside South Africa, it is not always possible to understand the tragedy of apartheid from books, from statistics, or even from historical cinema and the arts. From my long and intense discussions with Charles and the institute staff, including Helen Macdonald, Paul Haupt, Nyameka Goniwe, Fanie du Toit, Erik Doxtader, and Zola Sonkosi, I have filled in at Free download from www.hsrcpress.ac.za xiv Preface and Acknowledgments least some of the many gaps in my knowledge. For that, I am deeply appreciative.2 Most of the writing of this book took place while I was a visiting scholar at the Russell Sage Foundation in New York. I am first indebted to my university, and especially to executive vice chancellor Ed Macias for allowing me the opportunity to spend a year at the Foundation. Just as Manhattan is a special and unique place, so too is the Russell Sage Foundation. I have rarely encountered such a strong commitment to scholarship and such noble and well-meaning people. Throughout my year at Sage I profited immensely from my discussions (not always or even usually on South Africa) with Jim Sidanius, John Hagan, Bob Hauser, Kay Deaux, Peter Katzenstein, and my many lunch companions over the year. I am especially indebted to Eric Wanner, president of the Russell Sage Foundation, for allowing me the chance to complete this book while in residence in one of the most wonderful cities in the world. Portions of chapters 3 and 4 were presented in several different colloquia, including as a talk at Columbia University, Rutgers University, Fordham University, Indiana University, Rice University, City University of New York, the New York Law School, the University of Texas, the University of Colorado, and State University of New York at Stony Brook. Versions of this talk were also given at the New York Political Psychology Association (2002) and at the Sixth National Political Science Congress of the Chilean Political Science Association, held May 8–9, 2002, in Santiago. My paper “Does Truth Lead to Reconciliation? Testing the Causal Assumptions of the South African Truth and Reconciliation Process” is the recipient of the Sage Paper Award for the best paper in the field of comparative politics, presented at the annual meeting of the American Political Science Association in 2001 by the Comparative Politics Organized Section. A version of chapter 5 was presented to the sixtieth annual meeting of the Midwest Political Science Association, held April 25–28, 2002, in Chicago, as “Truth, Reconciliation, and the Creation of a Human Rights Culture in South Africa: An Investigation of the Effectiveness of the Truth and Reconciliation Process,” and it is published as “Truth, Reconciliation, and the Cre- Free download from www.hsrcpress.ac.za Preface and Acknowledgments xv ation of a Human Rights Culture in South Africa” in the 2003 issue of the Law and Society Review (vol. 38, no. 1). Chapter 7 of this book is drawn from Gibson (2002b). That paper is the 2003 recipient of the McGraw-Hill Award, awarded by the Law and Politics Organized Section of the American Political Science Association in recognition of the best journal article on law and courts written by a political scientist and published during the previous calendar year. In the original article, I acknowledged the contributions of a number of people, including those members of the IJR staff mentioned earlier and Eric Lomazoff, Marc Hendershot, and Christine Lamberson, who provided valuable research assistance on this project. I am also thankful to James Alt, Ronald Slye, and John T. Scott for comments on an earlier version of that article. Finally, I very much appreciate the advice and assistance provided by Kathleen McGraw on many aspects of that piece. I also acknowledge and appreciate the comments of Steven Ellmann, New York Law School, and Anthony Marx, Columbia University, on the material presented in chapter 3. Four people read the entire manuscript and offered copious suggestions for changes: Walter Murphy, Princeton University; Gunnar Theissen, Free University of Berlin; Alfred L. Brophy, University of Alabama; and Monika Nalepa, Columbia University. The care with which these friends read the manuscript improved it immeasurably. I have not always accepted the advice of my critics, but I nonetheless appreciate deeply every comment I received. I am under no illusion that this book, even with its overall conclusion that truth does contribute to reconciliation, will change the politics of transitional societies. Much of the truth and reconciliation process in South Africa is perhaps unique to the conditions created by apartheid and to the specific individuals who directed the transition process. But if this book makes any contribution at all toward creating a world in which those who were formerly bitter political enemies become more reconciled with each other, I will be deeply gratified. Thus, the ultimate argument of this book is that truth is powerful and democracy will profit if we only (to paraphrase John Lennon) “give truth a chance.” James L. Gibson Washington University in St. Louis Free download from www.hsrcpress.ac.za Chapter 1 Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? Perhaps no country in history has so directly and thoroughly confronted its past in an effort to shape its future as has South Africa. Working from the explicit assumption that understanding the past will contribute to a more peaceful and democratic future, South Africa has attempted to come to grips with its apartheid history through its truth and reconciliation process. This bold undertaking to mold the country’s fate consumed much of the energy and many of the resources of South Africa during the initial days of its attempted transition to democracy. The gargantuan task of addressing the past has been under the supervision of South Africa’s Truth and Reconciliation Commission (TRC). Established in 1995, the TRC spent roughly five years examining and documenting atrocities committed during the struggle over apartheid.1 At one level, the TRC was extraordinarily successful: it held countless hearings, interviewed thousands of victims of apartheid, granted amnesty to roughly 850 human rights violators, and produced a massive, five-volume Final Report.2 And in terms of uncovering detailed evidence of what happened under specific circumstances—as in determining exactly what happened to the “Cradock Four”3 —the TRC seems to have been effective as well (but see Jeffery 1999, who complains about numerous inaccuracies and bias in the TRC’s history of several specific incidents). In many respects, and according to most observers, South Africa’s truth and reconciliation process appears to have been phenomenally successful. Indeed, the world has acknowledged the success of South Africa’s TRC through the numerous attempts that have been made to replicate its truth and reconciliation process in other troubled areas of the globe. Truth commissions modeled on the South African Free download from www.hsrcpress.ac.za 2 Overcoming Apartheid experience have proliferated, and one of the leaders of South Africa’s experiment has created a major institute in New York to assist countries in developing plans for reconciliation in the world’s many festering hot spots. Perhaps the judgment that the TRC succeeded is based on nothing more than the simple (and simplistic) observation that South Africa appears to have made a relatively peaceful and quite unexpected transition from the apartheid dictatorship to a reasonably democratic and stable regime. Some surely attribute South Africa’s transformation to its truth and reconciliation process. If a TRC “worked” in South Africa, perhaps it can work elsewhere. South Africans themselves are not so sanguine about the process. Many complain that the TRC exacerbated racial tensions in the country by exposing the misdeeds of both the apartheid government and its agents and the liberation forces. Some vehemently reject the conjecture that “truth” can somehow lead to reconciliation, claiming instead that uncovering the details about the horrific events of the past only embitters people, making them far less likely to be willing to coexist in a new democratic regime (see, for example, Biko 2000). Indeed, based on my casual observations of the South African media, complaints and condemnations of the truth and reconciliation process seem to far outnumber laudatory assessments. Social scientists must be more agnostic about the success of the truth and reconciliation process. Indeed, it is perhaps shocking to note how little systematic investigation has been conducted into the question of whether the truth and reconciliation process succeeded in its objectives. There are many who enumerate the objectives of the commission itself and document its activities, and there is no dearth of judges when it comes to evaluating the process (see, for example, Du Toit 2000), but no earlier research has treated the various components of the truth and reconciliation process as hypotheses subject to confirmation or disconfirmation through rigorous social science methods. To put it bluntly, we simply do not know even today whether (and to what degree) the truth and reconciliation process in South Africa succeeded in achieving any of its objectives. Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? 3 Of course, assessments of “success” depend mightily on the specification of the goals of the process. Though the TRC was charged with conducting several types of activities (for example, granting amnesty to gross human rights violators), my central, uncontroversial contention in this book is that the objective of the truth and reconciliation process was to produce reconciliation in South Africa. This may not seem like a very rigorous or helpful specification of the country’s aspirations, since “reconciliation” is one of the most ambiguous and abused words in the lexicon of South Africa. And others take a different tack, with some seeking to discover whether the process accurately discovered the truth of certain events (for example, Ignatieff 1996), others assessing whether the TRC maintained fidelity to the law that created it (see Jeffery 1999), and still others judging the process in terms of philosophical standards of justice (especially retributive justice; see Minow 1998).4 But no prior investigation has squarely and systematically attacked the big question: has truth led to reconciliation in South Africa? Answering that question is the objective of this book. This question is, without doubt, as complicated as it is important (or some might even say “intractable”). To begin, what is “truth”? What is “reconciliation”? Is it fanciful to think that such grand and amorphous concepts can be given rigorous empirical meaning? Can social science really say anything at all about a question as complicated as that of whether truth contributes to reconciliation? A central contention of this book is that truth and reconciliation are concepts that can be (and should be) measured and assessed using rigorous and systematic social science methods. Just because these concepts are complex and complicated does not mean that they are impenetrable, or that hypotheses such as the one that truth causes reconciliation cannot be investigated. My efforts on this score will surely not please everyone. But throughout this analysis I assign “reconciliation” and “truth” concrete and unambiguous conceptual and operational meaning. At a minimum, those who disagree with my approach will know exactly what it is that they disagree with. To assess the hypothesis that truth leads to reconciliation requires a careful plan of attack. The starting point in the analysis is 4 Overcoming Apartheid Free download from www.hsrcpress.ac.za the specification of the meaning of reconciliation. I contend that the construct refers to at least four specific and perhaps even independent subconcepts: • Interracial reconciliation—defined as the willingness of people of different races to trust each other, to reject stereotypes about those of other races, and generally to get along with each other • Political tolerance—the commitment of people to put up with each other, even those whose political ideas they thoroughly detest • Support for the principles (abstract and applied) of human rights— including the strict application of the rule of law and commitment to legal universalism • Legitimacy—in particular, the predisposition to recognize and accept the authority of the major political institutions of the New South Africa This may well not be an exhaustive definition of reconciliation— perhaps there are other important components of the concept— but most objections to this list would surely be concerned with the need to include additional aspects of reconciliation rather than the question of whether these particular dimensions are central to the concept. A reconciled South African is one who respects and trusts those of other races, who is tolerant of those with different political views, who supports the extension of human rights to all South Africans, and who extends legitimacy and respect to the major governing institutions of South Africa’s democracy. These are the dependent variables for this study, and each of these dimensions of reconciliation is considered in a separate chapter in the book. It should be obvious by this point that I treat reconciliation as an attribute of individual South Africans. Perhaps a more conventional viewpoint considers reconciliation as a characteristic of societies or groups. While not denying that the degree to which an aggregate is reconciled is an interesting question (as in analyzing change in levels of reconciliation in a nation-state over time), my contention is that any understanding of reconciliation profits from Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? 5 beginning with an examination of the beliefs, values, and attitudes of ordinary people. If reconciliation means groups getting along together, then obviously reconciliation requires that individual South Africans eschew racism and embrace tolerance. A polity may be more than the sum of the individuals living within its territory, but it is impossible to understand a society without first understanding individual citizens—and in this case, the degree to which they are “reconciled.” Why is reconciliation important? For a political scientist interested in whether South Africa will be able to consolidate its attempted democratic transition, this question has an easy answer: reconciliation is hypothesized to contribute to democratization. Briefly put, a successful liberal democracy requires a sustaining and reinforcing “political culture.” The beliefs, values, attitudes, and behaviors of ordinary citizens must be, at a minimum, not antithetical to the principles of democratic governance, and maximally they ought to favor and support the main institutions and processes of liberal democracy.5 The most obviously relevant component of my approach to reconciliation is political tolerance, a concept that has been thoroughly analyzed in South Africa and elsewhere (see, for example, Gibson and Gouws 2003). For a liberal democracy to flourish, people must be willing to put up with political differences, in sharp contrast to the time-honored history in South Africa of killing one another over political disagreements (as in KwaZulu Natal). I also contend, however, that democracies work best when people are vigilant about human rights, and when they are unwilling to sacrifice law and legal process for expedient solutions to social problems (such as crime and terrorism). Further, political institutions need some “slack”—a reservoir of goodwill—if they are to function. Institutions that are supported only when they produce favorable policy outputs tend to be weak and ineffective. With a reservoir of goodwill, institutions command the political capital to go against public opinion in the short term—as in, for instance, protecting human rights against an outraged majority. And even interracial reconciliation contributes to democracy, since democracy requires coalitions built on similar interests, and racism and its cousins are inimical to the formation of such coalitions. For a polit- Free download from www.hsrcpress.ac.za 6 Overcoming Apartheid ical scientist, reconciliation is not an end in itself; instead, reconciliation is valuable because it contributes to the likelihood that South Africa will consolidate its democratic transition. Indeed, South Africa’s truth and reconciliation process is actually a mini-theory about the process of democratization, including an implicit causal model of how the truth and reconciliation process would contribute to the consolidation of democracy in South Africa. The theory posits that: Amnesty → Truth → Reconciliation → Democratization That is, the framers of the TRC accepted the hypothesis that when gross human rights violators are granted amnesty, they will come forward and tell the truth about their deeds. If a condition of receiving amnesty for gross human rights violations is full disclosure, South Africa could learn something about the black holes in its past by making amnesty available. Many believe that amnesty did indeed produce specific evidence of past transgressions that would have never come to light otherwise. These truths were then aggregated into a collective memory about the past. In turn, understanding the past is hypothesized to contribute to reconciliation. Or to put the relationship somewhat differently, those who created the TRC assumed that understanding the truth about the struggle over apartheid was a necessary precondition to reconciliation. National unity and reconciliation could be achieved only, it was argued, if the truth about past violations became publicly known and acknowledged (see, for example, Truth and Reconciliation Commission 1998, vol. 1, ch. 4, p. 53). Truth might not automatically produce reconciliation, but without truth, reconciliation was thought to be highly unlikely. As Brandon Hamber and Richard Wilson (n.d.) put it: Thus a national process of uncovering and remembering the past is said to allow the country to develop a common and shared memory, and in so doing create a sense of unity and reconciliation for its people. By having this shared memory of the past, and a common iden- Does Truth Lead to Reconciliation? 7 Free download from www.hsrcpress.ac.za tity as a traumatised people, the country can, at least theoretically, move on to a future in which the same mistakes will not be repeated. Finally, reconciliation is seen as a necessary condition for successful democratization. The theory here is not complicated: unless South Africans can agree to get along and refrain from killing each other, democracy in the country will fail. Such an assumption seems entirely reasonable. This is a simple theory, although not simple to test. Many hope that the theory is empirically accurate, since South Africa has wagered a large part of its future on the veracity of these linkages. The single most important purpose of this book is to test empirically the core hypothesis in this theory—that is, that truth leads to reconciliation. I refer to this as the “truth → reconciliation hypothesis” throughout this book. The four empirical chapters in which I test this hypothesis—once for each of the subdimensions of the overall concept—are the heart of this book. One should not treat “truth” lightly, however, and I do not. Indeed, truth can be even more worrisome than reconciliation, especially since so many of us bridle (or should bridle) at even the intimation that “the truth”—official truth—exists. Whether one likes it or not, an explicit objective of the TRC was to produce a collective memory for South Africa. This is not just a chronicle of who did what to whom; instead, it is an authoritative description and analysis of the history of the country. Was apartheid a crime against humanity? Was the criminality of apartheid due to the missteps of a few rogue individuals, or was apartheid criminal by its very ideology and through its institutions? These are questions for which the TRC provided unambiguous and, by its accounting, definitive answers. My goal here is not to assess the historical accuracy of these claims but rather to determine the degree to which ordinary South Africans accept the truth as promulgated by the TRC—South Africa’s “collective memory.” When I consider the truth → reconciliation hypothesis, in every instance I am investigating the hypothesis that those South Africans who accept the truth as Free download from www.hsrcpress.ac.za 8 Overcoming Apartheid documented by the TRC are more likely to be reconciled. As I try to explicate more completely in chapter 3, “truth” here means the TRC’s truth, nothing more. Furthermore, factors other than truth can contribute to reconciliation. Indeed, social scientists have learned much about the various processes that undergird this hypothesis, and I would be foolish to ignore these important bodies of theory. For instance, one of the most venerable hypotheses of the literature on political psychology is that interracial contact contributes to interracial harmony. That hypothesis receives a great deal of scrutiny in this book. Other social science theories are carefully examined as well. My goal is to provide a comprehensive explanation of the variation in levels of individual reconciliation, as defined by the four dependent variables I identified earlier in this chapter. Thus, my motives in writing this book are twofold: First, I hope to address the extremely important policy questions posed by the truth and reconciliation process and, most particularly, the hypothesis that truth leads to reconciliation. I hope that the results of this research will inform efforts to establish truth commissions elsewhere in the world. Second, I hope to make a contribution to the social science of interpersonal and intergroup relations. Ultimately, reconciliation is about people getting along with and tolerating each other, and thus theories of political tolerance are directly relevant to this research. In addition to considering the standard hypotheses from the tolerance literature, I investigate the so-called contact hypothesis, as well as theories of collective memory and of the consequences of experiences with political repression. I realize that those who walk down the middle of the road often get hit by trucks traveling in both directions, but I try throughout this book to pursue both these policy and theoretical objectives. My hope is that those concerned with policy will come to appreciate the importance of theory, and that those mainly motivated by theory will see that it is important to try to address the policy issues at stake through rigorous social scientific inquiry. To provide the context for this analysis, I begin with a brief overview of the truth and reconciliation process in South Africa, Does Truth Lead to Reconciliation? 9 Free download from www.hsrcpress.ac.za focusing in particular on the objectives assigned to the TRC by the legislation that created it. Truth and Reconciliation in South Africa: History and Formal Objectives Ending apartheid in South Africa came at considerable cost to those who had long struggled against the oppressive system. In South Africa, in contrast to other nations emerging from a tyrannical past (for instance, Argentina and Uganda), the ancien régime was not defeated.6 This meant that the transition had to be brokered. One of the central issues in the talks over the transformation of the apartheid state was amnesty. The National Party and the leaders of other powerful white-dominated institutions (such as the security forces) made amnesty a nonnegotiable centerpiece of their demands (see Omar 1996). Without the promise of amnesty for the crimes (and criminals) of apartheid, the transition to democracy would have stalled, and the political violence that had been so widespread in the 1980s might have reemerged. The creation of the Truth and Reconciliation Commission, with the power to grant amnesty, was the price that the liberation forces had to pay to secure a peaceful transition to majority rule (Rwelamira 1996). The Truth and Reconciliation Commission was provided for by the “Postamble/Endnote” to the Interim Constitution of 1993 and enacted by the new Parliament in 1995 as the Promotion of National Unity and Reconciliation Act (no. 34, 1995). That statute called for the establishment of the commission, with separate committees on human rights violations, amnesty, and reparations and rehabilitation. The TRC began functioning shortly thereafter. The creation of the TRC was certainly controversial. Many parties, including Amnesty International, argued that international law and convention forbade granting amnesty for crimes against humanity, as well as for torture and similar offenses; the parties’ slogan was: “No amnesty, no amnesia, just justice” (quoted in Verwoerd 1997).7 Nonetheless, the South African Constitutional Court Free download from www.hsrcpress.ac.za 10 Overcoming Apartheid upheld the constitutionality of the act (Azanian Peoples Organization [AZAPO] and others v. President of the Republic of South Africa and others, CCT 117/96 [July 25, 1996]), and the TRC began functioning in 1995.8 The truth and reconciliation process was expected to last only two years. Instead, the TRC was in operation for six years. When President Thabo Mbeki dissolved the commission’s amnesty committee, effective May 31, 2001, his proclamation also revived the TRC for another six months for the purpose of preparing two supplementary volumes to the Final Report (issued in 1998).9 Some evidence suggests that the truth and reconciliation process was deeply unpopular among South Africans of every color. For instance, a survey conducted in mid-1998 by Business Day (a reasonably well respected South African newspaper) found that nearly two-thirds of the public believed that the truth and reconciliation process had harmed race relations in South Africa (see Business Day Reporter 1998; see also Theissen 1997; Theissen and Hamber 1998; Gibson and Gouws 1999; and Macdonald 2000). Critics charge that the process has been characterized by little remorse or penance among the perpetrators, that not all of the guilty came forward to admit their crimes (for example, former state president P. W. Botha), and generally that whites have been unwilling to accept responsibility for apartheid. A host of other criticisms have also been laid against the details of the process employed by the TRC (see, for example, Jeffery 1999). The TRC was established to achieve a general purpose as well as several specific objectives. According to the National Unity and Reconciliation Act, the goal “of the Commission shall be to promote national unity and reconciliation in a spirit of understanding which transcends the conflicts and divisions of the past” (sect. 3, 1). The specific means of achieving this goal were to include “establishing as complete a picture as possible of the causes, nature and extent of the gross violations of human rights which were committed during the period . . . including antecedents, circumstances, factors and context of such violations, as well as perspectives of the victims and the motives and perspectives of the persons Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? 11 responsible for the commission of the violations, by conducting investigations and holding hearings.” In calling for a “complete picture” of the past, the law specifically addressed the need to create a collective memory for South Africa. “A society cannot reconcile itself on the grounds of a divided memory. . . . Clearly, key aspects of the historical and ethical past must be put on the public record in such a manner that no one can in good faith deny the past. Without truth and acknowledgment, reconciliation is not possible” (Zalaquett 1997, 13). A second mandate for the TRC involved “facilitating the granting of amnesty to persons who make full disclosure of all the relevant facts related to acts associated with a political objective and comply with the requirement of this Act.” This objective obviously concerns the deeds of individual victims and perpetrators, but to many, over time, it came also to address the larger issue of apartheid itself. For instance, South Africa’s Human Rights Commission declared on July 31, 1993, that “the enormity of the crime of apartheid as a system of social engineering must be revealed in all its nakedness, including the distortions wrought upon some of those who, in their fight against this evil, lost their way and engaged [in] the very human rights violations so systematically practised by their oppressors” (quoted in Hay 1998, 55). The TRC was also charged with “establishing and making known the fate or whereabouts of victims and restoring the human and civil dignity of such victims by granting them an opportunity to relate their own accounts to the violations of which they are the victims, and by recommending reparation measures in respect of them.” The purpose identified in this section is that of achieving reconciliation by restoring lost dignity to victims as well as providing them with compensation. Finally, the TRC’s fourth objective was to produce a report addressing the first three issues and making recommendations on how to develop a political culture in South Africa that would be respectful of the human rights of all citizens. Thus, the most general goal of the truth and reconciliation process in South Africa has been to enhance the likelihood of recon- Free download from www.hsrcpress.ac.za 12 Overcoming Apartheid ciliation. Reconciliation was not given a great deal of specific content, although the framers of the process clearly sought to shape the views of individual South Africans—for instance, by getting them to accept the collective memory about the country’s past and to endorse an expansive definition of human rights. If reconciliation is to be turned into a concept amenable to empirical investigation, however, it must be given more specific and concrete meaning. I have already suggested that reconciliation is multidimensional, with four distinct aspects. It is useful now to consider these subdimensions more rigorously. The Meaning of “Reconciliation” Two themes dominate contemporary discussions of the truth and reconciliation process in South Africa. First, no one seems to know what “reconciliation” means. Mark Hay (1998, 13), for instance, calls “reconciliation” “one of the most abused words in recent history in South Africa.” Indeed, Max du Preez (2001, 13), like some others, has gone so far as to claim that, “of course, reconciliation is a concept that cannot be measured.” Du Preez apparently believes that this assertion is so self-evidently true that he judges it unnecessary to offer a defense of his point of view. Second, everyone is certain that reconciliation has failed in South Africa, or at least has not lived up to the expectations of most South Africans. People may not be able to define and measure the concept, but they seem to think they “know it when they see it”—or at least when they do not see it.10 But surely reconciliation means something. Indeed, the problem with the concept, according to some, is that it has too many meanings, not too few. For instance, Brandon Hamber and Hugo van der Merwe (1998) claim to have isolated five distinct ways in which reconciliation has been either implicitly or explicitly used. The problem with reconciliation is not that it is devoid of content; the problem is that reconciliation is such an intuitively accessible concept that everyone is able to imbue it with her or his own distinct understanding. Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? 13 But perhaps reconciliation is not such a difficult and complicated concept after all. A great deal of complexity can certainly be attached to the idea, but it is also possible to distill the concept down to a few simple and specific elements. Reconciliation is often discussed as a relationship, either between victims and perpetrators or between beneficiaries and the exploited. The TRC Final Report refers to the following types of relationships: individuals with themselves, between victims, between survivors and perpetrators, within families, between neighbors, between communities, within different institutions, between different generations, between racial and ethnic groups, between workers and management, and, “above all, between the beneficiaries of apartheid and those who have been disadvantaged by it” (Truth and Reconciliation Commission 1998, vol. 5, ch. 9, pp. 350–51). To make the concept empirically manageable, I must first be clear about who is being reconciled with whom or what. Discussions of reconciliation in South Africa typically refer to two distinct phenomena: dealing with the microtruth of what happened to specific loved ones, and dealing with the macrotruth about the nature of the struggle over apartheid. At the micro level, discussions often focus on the reconciliation of victims and perpetrators of gross human rights violations. This is the clearest meaning of the term “reconciliation,” and it has been the subject of wide media coverage in South Africa. Stories about the most profoundly injured victims (or their families) granting forgiveness to their evil tormentors are the stuff of which soap operas are made, and they captured the fancy of the South African public, at least for a while.11 When applied to victims and perpetrators, reconciliation typically means acceptance of blame, apology, and forgiveness. For many in South Africa, led by Desmond Tutu (see Tutu 1999), this type of reconciliation has deeply religious overtones, as in the third of the five definitions proposed by Hamber and van der Merwe (1998), which stresses a “strong religious ideology of reconciliation” based on identifying a “humanity” common to all groups in South Africa. Central to this definition is forgiveness, although Hamber and van der Merwe assert that “this perspective runs the risk of mistakenly equating forgiveness of past enemies with reconciliation.”12 Free download from www.hsrcpress.ac.za 14 Overcoming Apartheid The larger South African society is the context for the second meaning of the term—reconciliation between the races and, closely related, between those who profited from apartheid and those who were injured by it. Dan Markel (1999, 407) refers to this as “the public reconciliative relationship” and asserts: “The TRC hoped to cultivate a broad-swathed public reconciliative role among and with the various racial and ethnic groups of South Africa, so that social groups would learn the skills necessary to cope with the pain experienced as a group in the past.” This understanding of reconciliation has little to do with any specific human rights violation; rather, it involves coming to grips with (accepting responsibility and blame for) the subjugation of the black majority by the small white minority under apartheid. As Antjie Krog notes: “Reconciliation in this country is not between actual operators and victims, but between the beneficiaries (whites) and the exploited (blacks)” (quoted in Christie 2000, 147). This definition is similar to the first definition proposed by Hamber and van der Merwe (1998)—a “non-racial ideology of reconciliation,” which is basically a condition in which South Africans live together as “non-racial citizens within a harmoniously integrated social setting.” It is also similar to their second definition, which emphasizes “intercommunal understanding” and calls for “bridging the divide” between various distinct and generally separate racial communities. “From this perspective the TRC is considered to be a facilitator that can improve communication and mutual tolerance of diversity.” As Kadar Asmal (2000, 1226) notes: “If you never pulled a trigger nor held a smoking gun, but yet you benefited from the societal system defended by the violence—if all you did was loaf around a poolside in an opulent white apartheid suburb—you still needed to be involved in the process [of transition and reconciliation].” My concern in this book is thus with these broader sociopolitical aspects of the reconciliation of all South Africans, not just victims and perpetrators. Reconciliation may be thought of as a continuum describing the relationship between those who were masters and slaves under the old apartheid system, not just the relationship between those who were victims or perpetrators of gross human rights violations. Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? 15 In South Africa the groups that must reconcile are the four main racial groups in the country (see the appendix to this chapter)— whites, Africans, Colored people, and South Africans of Asian origin. The root cause of interracial alienation in South Africa was colonialism; the proximate cause was, of course, apartheid. The damage inflicted was inequality, loss of dignity, and untold violence and political repression. Whites treated Africans, Colored people,13 and South Africans of Asian origin as if they were inferior in nearly every sense, including in the political and legal domains, and blacks were even expelled from their own country to the socalled Bantustans. The essential condition for reconciliation is therefore that South Africans of every race accept all other South Africans as equals and treat them as equal, extending dignity, respect, and cocitizenship to them. The TRC Final Report (Truth and Reconciliation Commission 1998, vol. 9, ch. 5, p. 425) asserts that reconciliation requires the recognition that “we are all in the same boat—we simply need to understand each other better and be more respectful of each other’s culture.” The report also concludes: “Reconciliation is based on respect for our common humanity” (435). Hay (1998, 14) defines reconciliation as “the establishing or recovery of human dignity and humanity of every person, rooted in human rights, and the acceptance of this by the individual and the society.” When people talk about reconciliation, they often mean nothing more than the races getting along better—that is, a diminution of racial animosities. Accomplishing this aim requires that people come to interact with each other more (the breakdown of barriers across races) and communicate more, acts that in turn lead to greater understanding and perhaps acceptance and result in the appreciation and exaltation of the value of racial diversity and multiculturalism in the “Rainbow Nation.” Reconciliation also takes on three additional meanings beyond interracial reconciliation. The first is simply political tolerance, the willingness of South Africans to put up with their political foes.14 South Africans may not be required to like or agree with each other, but many have expected the truth and reconciliation process to contribute to a sort of relatively peaceful coexistence. In its minimalist version, this means putting up with those who hold differ- Free download from www.hsrcpress.ac.za 16 Overcoming Apartheid ent or even repugnant ideas and viewpoints. In its maximalist rendition, tolerance means embracing one’s former enemies, forgiving them, and perhaps even joining in political coalitions with them. A third aspect of reconciliation has to do with the development of a political culture in South Africa that is respectful of the human rights of all people. The creation of a human rights culture was one of the explicit goals of the TRC. The Final Report (Truth and Reconciliation Commission 1998, vol. 5, ch. 9, p. 435) asserts: “Reconciliation requires that all South Africans accept moral and political responsibility for nurturing a culture of human rights and democracy within which political and socio-economic conflicts are addressed both seriously and in a non-violent manner.” This conception is similar to the fourth definition of reconciliation proposed by Hamber and van der Merwe (1998). For South Africa’s nascent democracy to prosper, the political culture must be one in which the universalistic application of the rule of law—and the rejection of the arbitrary exercise of governmental authority and power—is deeply valued and respected. Finally, it is important as well to address the institutions that serve as the backbone of South Africa’s new democracy, since democracy is both a set of formal institutions and a set of cultural values.15 For instance, South Africans must come to tolerate each other, to be willing to countenance the expression of displeasing political ideas. But they must also come to support institutions that have the authoritative means of enforcing political tolerance as effective public policy. Just as the truth and reconciliation process sought to encourage respect for human rights in South Africa, it also implicitly sought support for the institutions charged with the protection of those human rights. If South Africans fail to extend legitimacy to the institutions of majority rule and the protection of minority rights, it would be difficult indeed to consider them reconciled with the nascent democratic system that has been implemented. To extend these institutions legitimacy is to accept at an elemental level South Africa’s multiracial system of democratic rule and reconcile with the new political dispensation in the country. Reconciliation requires that all South Africans recognize the Does Truth Lead to Reconciliation? 17 Free download from www.hsrcpress.ac.za legitimacy of the political institutions created after the fall of apartheid. Consequently, a “reconciled” South African is one who: • eschews racial stereotyping, treating people respectfully as individuals, not as members of a racial group; • • is tolerant of those with whom he or she disagrees; • subscribes to a set of beliefs about the universal application of human rights protections to all South African citizens; and recognizes the legitimacy of South Africa’s political institutions and is therefore predisposed to accept and acquiesce to their policy rulings. Thus, in this book I investigate reconciliation between people, among groups, with basic constitutional principles, and with the institutions essential to the new South African democracy. Understanding Variation in Reconciliation: Testing Theory and Hypotheses South Africans undoubtedly differ in the degree to which they hold reconciled attitudes. Since mine is a microlevel analysis, my overriding purpose is to account for the variability across individuals in these four aspects of reconciliation. Here I briefly introduce the theories that play a prominent role in my analysis. Theories of Collective Memory To speak of “truth” evokes the imagery of theories of collective memory. I contend in this book that the TRC attempted to assemble a collective memory of the apartheid past for South Africans, and this contention is unlikely to be controversial. What is perhaps novel in my analysis is the attempt to test the hypothesis that the degree of an individual South African’s participation in the collective memory forged by the truth and reconciliation process is re- 18 Overcoming Apartheid Free download from www.hsrcpress.ac.za lated to the degree to which that individual is reconciled. Thus, my findings have profound implications for those who would attempt large-scale attitude change through social persuasion, which is what the TRC sought to do—with at least a modicum of success. Interracial Contact Theory The contact hypothesis is one of the most venerable in political psychology, even if research has generated a variety of caveats to the simple idea that the more people interact, the more likely they are to accept each other. I rely heavily on the contact hypothesis as an explanation of the degree to which racial reconciliation has taken place in South Africa. Generally, racial isolation characterizes the country, but to the extent that people engage each other in circumstances of at least modest intimacy and a fragment of equality, reconciliation profits. Theories of Political Tolerance An impressive body of literature has emerged accounting for why some are tolerant of their political enemies and others are not. This literature has taught us, for example, about the crucial role that perceptions of threat play in shaping tolerance. In addition to investigating the conventional threat-centered explanations of intolerance, I focus here on the ways in which intolerance is connected to group identities. Intolerance is more than a matter of individuals not putting up with each other. It is also about intergroup conflict, and therefore social identity theory has much to say about political tolerance—a minimalist definition of reconciliation. Experiences with Political Repression Though no clear body of theory addresses the microlevel political consequences of having experienced political repression, I include this in my listing of theoretical approaches since it is so central to my analysis. Apartheid left deep scars on South Africa, and on South Africans, and individuals’ experiences under apartheid no doubt have shaped their understandings of both truth and reconciliation. But not all South Africans were directly harmed by apartheid; not even all Black South Africans were directly injured by the Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? 19 repressive system.16 One of the most unexpected findings from this research is that people fared so differently under apartheid. Consequently, the hypothesis that variation in experiences shapes levels of reconciliation is investigated across all four dimensions of reconciliation. Legitimacy Theory An important body of theory addresses the question of how institutions relate to the citizens they govern, and in particular how perceptions of institutional legitimacy contribute to compliance with unpopular political and legal decisions. I mobilize that body of theory in considering reconciliation with the new political dispensation in South Africa. To the extent that these new institutions suffer from a legitimacy deficit, they are unlikely to be effective agents of democratization. Thus, my hope is that readers will find this analysis rich in terms of both the theory it addresses and the significance of the overall research question: does truth contribute to reconciliation? The Plan of the Book This book is structured around seven chapters in which empirical evidence is adduced to consider the hypothesis that the truth and reconciliation process has contributed to a more reconciled South Africa. These chapters all draw on a survey of South African opinion that I conducted in 2000 and 2001. Chapter 2 is perhaps a somewhat unusual chapter for a book such as this, since it focuses on the independent variables used in the analysis rather than the dependent variables. I do this so as to set the context for my research on reconciliation, especially the context as defined by the relationship of contemporary South Africa to its apartheid past. Little rigorous research has been reported documenting how South Africans experienced apartheid, so I devote considerable effort to investigating how people believe they lived and their memories of the experiences they had under the old system. In general, many of my findings about apartheid are Free download from www.hsrcpress.ac.za 20 Overcoming Apartheid entirely unexpected, thereby justifying my examination of these results early in the book. Chapter 2 also discusses group identities. This portion of the chapter is considerably less surprising, since I document that identities are important and that most South Africans view groups as continuing to be important in South African politics. These and the other indicators described in this chapter set the stage for testing hypotheses about variability in individual-level reconciliation. In chapter 3, I analyze truth—or as I have noted, the truth as constructed by the truth and reconciliation process in South Africa. I begin the chapter by considering just what the TRC proclaimed about the country’s apartheid past. Then I examine the degree to which ordinary South Africans accept this truth as the collective memory of the country, and I consider whether the activities of the TRC itself contributed to a common understanding of the nature of the country’s apartheid past. My findings in this chapter are filled with unexpected nuggets, ranging from the widespread condemnation of apartheid among whites to the remarkable degrees to which Africans, Colored people, and those of Asian origin do not condemn apartheid. A multivariate analysis of truth acceptance reveals that group identities influence beliefs about the past, although not always in simple and direct ways. Perhaps the most important contribution of this chapter lies in documenting how collective memory can be rendered a rigorous theory and in suggesting how to analyze individual differences in acceptance of collective memories. The first reconciliation-dependent variable for my analysis is investigated in chapter 4. There I test a variety of hypotheses about the factors that have contributed to interracial reconciliation among South Africans. For instance, I discover that interracial reconciliation is heavily dependent on interracial contacts. Not all contact has a salutary effect on racial attitudes—for instance, simply working with people of a different race has no positive consequences— but when South Africans interact in conditions of relative intimacy and equality, racial animosities decline. Most important, I make the bold claim in this chapter that, at least under some circumstances and with some groups, truth (as promulgated by the TRC) does indeed contribute to reconciliation. This is perhaps the most Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? 21 important chapter in the book, since it addresses one of the most fundamental aspects of reconciliation—the question of whether South Africans of different races can live peacefully and respectfully with each other. In chapter 5, I analyze the second aspect of reconciliation—support for a human rights culture in South Africa, and in particular the universalistic rule of law. After discovering that support for the rule of law is not widespread in South Africa (although compared to such support in other established democracies, commitment to the rule of law in South Africa is not inordinately low), I report significant interracial differences in levels of commitment to this aspect of a human rights culture. Attitudes toward the rule of law are bound up in a larger set of beliefs about the rights and obligations of majorities and minorities in a liberal democracy, and these beliefs differ across the various racial groups in predictable ways. Finally, I produce evidence that the truth and reconciliation process may indeed have shaped (at least in part) the ways in which South Africans think about the rule of law and human rights. Chapter 6 continues my concern with intergroup relations by focusing on political intolerance. Political tolerance is the minimalist form of reconciliation because it requires nothing more than that South Africans put up with their political enemies. Beginning the analysis with well-trodden theories of tolerance, I evaluate the hypothesis that intolerance is a function of the threats that people perceive from their political enemies. The hypothesis is confirmed, although it is interesting to note that sociotropic threats (threats to the country, to the South African way of life) are considerably more powerful predictors of intolerance than egocentric threats (threats to the individual himself or herself) or perceptions of the political power of the group. I also find that threat perceptions have abated somewhat since an earlier similar study conducted in 1996. The most important contribution of this chapter is its investigation of the linkage between social identity theory and intolerance. Though the relationships are often complicated, the analysis establishes no significant connection between group identities and political intolerance. This finding represents an important challenge to Free download from www.hsrcpress.ac.za 22 Overcoming Apartheid those who see the roots of intergroup conflict in the group attachments of citizens. In some respects, the analysis I report in chapter 7 is a digression, since I do not directly consider one of the four dimensions of reconciliation. Instead, this chapter addresses the question of how to compensate for the inherent injustice of granting amnesty to those who admitted gross human rights violations. The TRC awarded amnesties to more than one thousand miscreants, thereby creating a “retributive justice deficit.” In this chapter, I report evidence that this justice deficit was in fact overcome by other aspects of the amnesty process. It will surprise few that distributive justice (compensation) can make up for some of the failure to achieve retributive justice. But the most important and interesting findings of this chapter are that two other forms of justice—restorative justice (apologies) and procedural justice (giving voice to the victims and their families)—contribute mightily to compensating for the inherent unfairness of granting amnesty to human rights violators. Though I explicitly do not claim that reconciliation requires the acceptance of the necessity of granting amnesties, I do conclude in this chapter that the amnesty component of the truth and reconciliation process did less damage to reconciliation than many believe, in large part because the process produced compensatory forms of justice. In chapter 8, the final empirical analysis of reconciliation, I investigate the legitimacy that South Africans accord to the political institutions of the New South Africa. My central contention here is that reconciliation requires that South Africa’s new Parliament and Constitutional Court be recognized as legitimate and authoritative and that, as a consequence, South Africans must accept the decisions of these institutions, especially the decisions of which they disapprove. This chapter relies heavily on legitimacy theory, which assigns a crucially important function to constitutional courts in liberal democracy: the need to overrule the actions of the majority when they infringe on the constitution, and especially on the rights of political minorities. In what is perhaps the most pessimistic portion of my analysis, I find that neither institution enjoys widespread and cross-racial legitimacy in contemporary South Africa. This lack of legitimacy is partly due to the failure of the Constitutional Court Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? 23 to establish a distinctly nonpolitical legal identity; as it is right now, views toward the court and the parliament are closely related and dependent as well on satisfaction with the short-term performance of the institution. Without a store of institutional legitimacy, South African institutions are dependent on pleasing their constituents if they are to gain acquiescence with their decisions. This does not bode well for the future, since these institutions must inevitably be able to make decisions that are displeasing to South Africans but pleasing to constitutional democracy when the two conflict. The last chapter returns to the overarching question of whether truth actually leads to reconciliation, based on the empirical results of the five analytical chapters. I draw conclusions about the role that the truth and reconciliation process has played in the consolidation of South Africa’s democratic transition. Though my most general conclusion is that the truth and reconciliation process has indeed contributed to reconciliation in South Africa, I nonetheless identify several ways in which the South African experience may not be generalizable to other transitional political systems. Ultimately, however, democracy in South Africa has been well served by the efforts of the TRC and the many people who tried to create a truth for the country that would reconcile South Africa’s many competing factions and interests. Concluding Comments This book’s look at reconciliation from the viewpoint of the individual citizen is both the primary strength and weakness of this research. Individuals constitute the building blocks of a political system, and it is difficult to imagine how a society could be reconciled without individual members of that society also being reconciled. Thus, learning about levels of reconciliation among South Africans and discovering something about the causes of such reconciliation are crucial tasks to which this book purports to make an important contribution. But not all questions of reconciliation can be understood in terms of the attributes of citizens. Groups are important, institutions are Free download from www.hsrcpress.ac.za 24 Overcoming Apartheid important, and some individuals (elites) are more important than others. Moreover, exogenous events (international terrorism, for instance) can play an enormous role in shaping how a society feels about difference and about those who are different. Thus, while my study here is concerned with a fundamentally important chapter in the larger book of South African reconciliation, the story I tell is a story of the beginning, not the end, of the process of reconciliation. Appendix: Race in South Africa Whatever one’s preferences, no one can write about South African politics without writing about race. Since race is such a salient part of the South African context—and since race is such a contentious concept—I offer here my understanding of the meaning of the concept.17 It is common in South Africa to divide the total population into four racial categories for the purposes of research or to explain the demographic realities and/or socioeconomic conditions in the country, and I follow this practice throughout the analysis reported in this book. As Wilmot James and Jeffrey Lever (2000, 44) note: “The use of these categories is unavoidable given the fixity that they have come to acquire both in popular consciousness and official business.” The use of these racial terminologies, however, differs from the way in which racial categorization may be understood in other societies. It is therefore important to understand the historical development of these categories, especially the legal boundaries imposed on racial groups by the apartheid government.18 The four racial groups are African, white, Colored, and South Africans of Asian origin (Indian). These groups are also often referred to as population groups, ethnic groups (although this term usually refers to African subcategories such as Xhosa or Zulu), or national groups. The African majority has been known by European settlers by different names over time, such as “native,” “Bantu,” or “Black,” and some of these terminologies were later formalized by apartheid legislation. The Africans were the original Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? 25 inhabitants of the area now called South Africa; descendants of Iron Age farmers, they speak different variants of the Bantu languages spoken in sub-Saharan Africa, east of Cameroon (James and Lever 2000, 44). Generally, I refer to these people as Africans or blacks. The white inhabitants of South Africa (formerly called Europeans) are descendants of Dutch, German, French (Huguenots who fled religious persecution in France), English, and other European and Jewish settlers. Though South Africa was colonized by the Dutch and the British in different historical periods, the British colonization caused English to become entrenched as a commonly spoken language. “Colored” is considered a mixed-race category, although as James and Lever (2000, 44) argue, it is actually a residual category of people of quite divergent descents. “Colored” refers to the children of intermarriages between whites, Khoi-Khoi (commonly referred to as “Hottentots”) and the San (commonly referred to as “Bushmen”), slaves from Malagasy and Southeast Asia (Malaysia), and Africans (Thompson and Prior 1982, 34). The Indian population came to South Africa as indentured laborers to work on the sugar plantations in Natal in the late nineteenth and early twentieth centuries. They came from different regions in the Indian subcontinent, adhered to different religions, and spoke different languages, and so, like Colored people, they are not a homogeneous group. I refer to these people as South Africans of Asian origin, despite the fact that some Colored people are technically of Asian origin. When the National Party came to power in 1948, it embarked on a legislative process aimed at securing white political power and keeping the white population group “pure.” The Population Registration Act, 30 of 1950, and its various amendments legislated that all citizens of South Africa be classified according to racial or ethnic origins. Racial origin was determined by the natural father’s classification. The policy was not consistently implemented, however, because when the father in a mixed-race marriage was African, the offspring was classified as African, but if the father was white, the offspring was classified as Colored (see Brookes 1968, 24). Addi- Free download from www.hsrcpress.ac.za 26 Overcoming Apartheid tional criteria were acceptance in the community and appearance; a 1967 amendment added descent (Thompson and Prior 1982, 36). Very often mixed-race families were split up owing to the hues of their skin, causing immense suffering (Horrel 1982, 2). The original act referred to the main groups as “white,” “colored” and “Native.” In 1951 the South African government replaced “Native” with “Bantu,” and in 1978 it officially changed the term to “Black.” The most commonly used term now for the original inhabitants of South Africa is “African,” while “Black” is often used inclusively to refer to everyone who is not white. (The term originated as a negative reaction to references to groups other than whites as “nonwhites.”) In this sense, “Black” is sometimes misleading, since it refers to Africans, Colored people, and those of Asian origin. (“Black” is rarely used in this way in this book.) The enforcement of the Population Registration Act was very important, since it was the foundation for the Group Areas Act, 41 of 1950 (legalizing separate neighborhoods for each racial group), and the Separate Amenities Act, 49 of 1953 (legalizing separate public facilities for the different racial groups).19 A direct response to the fixed racial categorization of the apartheid regime was the ideological endorsement of nonracialism by the African National Congress (ANC). This policy rejected race as a social construct and supported the underlying principle of equality for all, with appearance and descent playing no role. Yet the political and sociological realities created under apartheid—such as homogeneous neighborhoods and segregated schools—now coupled with political strategies such as affirmative action to undo past discrimination, still reinforce and politicize racial consciousness involving these specific categorizations (James and Lever 2000, 45). From the perspective of research on South Africa’s political culture(s), it could therefore be justifiably argued that the subjective experience of these racial categorizations, the class positions, and the sociological and historical realities of their members justify the general practice of reporting these results separately by these racial groupings. In earlier research (for example, Gibson and Gouws 2003; Gibson 2003), I have documented enormous differences across South Free download from www.hsrcpress.ac.za Does Truth Lead to Reconciliation? 27 Africa’s groups in terms of a wide variety of political attitudes. Consequently, it is essential that race be incorporated into the analyses in this book. To ignore race would be to fail to recognize that South African politics today continues to be shaped by its racist history. To incorporate race into this analysis is not to accept anything about apartheid but merely to acknowledge that apartheid shaped—and continues to shape—political reality in South Africa. Free download from www.hsrcpress.ac.za Chapter 2 Apartheid’s Legacy in Contemporary South Africa: Experiences, Attributes, and Attitudes of the Sample Apartheid left its mark on three fundamental dimensions of the South African political system: its value systems, its structure and its political culture. —Willie Esterhuyse, “Truth as a Trigger for Transformation” (2000, 148) A lingering “apartheid memory” continues to restrict the development of trust and allegiance in the new political dispensation and its institutions. —Johnny de Lange, “The Historical Context, Legal Origins, and Philosophical Foundation of the South African Trust and Reconciliation Commission” (2000, 29) The research on truth and reconciliation in South Africa reported in this book is based on a public opinion survey that I conducted in that country at the close of 2000 and beginning of 2001 (for technical details, see appendix A). The sample of over 3,700 South Africans we interviewed represents the entire South African population, with representative subsamples of each major racial, ethnic, and linguistic group in the country. Thus, it is entirely legitimate to draw inferences from the sample to the larger South African population of more than 43 million people. Few surveys have ever been conducted in South Africa that are of the technical quality of this one. Free download from www.hsrcpress.ac.za Apartheid’s Legacy 29 In the chapters that follow, I test a variety of hypotheses about the connections between truth and reconciliation. The central concern of this book is with the seemingly simple question of whether truth has contributed to reconciliation in South Africa. This question, however, is anything but simple if one wants to draw inferences about the causal connection between the two concepts. Truth may cause people to be reconciled, but complicating the causal flow, people with certain types of beliefs about South Africa’s past may be more likely to be reconciled in the first place. Thus, reciprocal causation could reasonably be expected to characterize this relationship. Discerning the flow of causality is one of the most demanding tasks facing empirical social scientists, and one can virtually never do better than drawing probabilistic conclusions about causation.1 Unfortunately for analysts, causality is of the utmost importance for understanding truth and reconciliation processes. It is important to acknowledge at the outset that causal inferences are never perfect, even in the most “scientific” research. The basic limitation is that alternative explanations for social and political phenomena abound, rendering it difficult if not impossible to control for all possible rival hypotheses. The problem of discerning causality is compounded when concepts are measured at the same point in time, as they typically are in cross-sectional surveys such as this. Social scientists are good at determining what goes with what. Determining what causes what, and how, is a far more arduous and dicey process. I shall have much more to say about the problem of causality throughout this analysis, since the question of what causes what has important implications for the analytical strategy of this book in general. I introduce the problem of causality at this point for two reasons. First, I want to encourage readers to think causally about all of the relationships considered in this analysis. Although social scientists are fond of asserting that “correlation does not imply causation” (things that go together do not necessarily cause each other), many of the linkages investigated here require some sort of conclusion about causality. Second, it is useful to introduce a variety of variables in this chapter that will be used as control variables in the substantive analyses throughout this book. These Free download from www.hsrcpress.ac.za 30 Overcoming Apartheid indicators, documented conceptually and operationally in this chapter, are used throughout my analysis as substantively interesting control variables. Consequently, I devote more space to presenting these variables than might otherwise be the case. Another, less theoretical purpose is also served by addressing these descriptive data in this chapter. For many reasons, little is known about the structure of public opinion in South Africa. This lack of knowledge flows in part from the history of apartheid, but it is also attributable in part to the feeble support for the social sciences in South Africa and in part to ideology. (When one “knows” something, it is often unnecessary to bother with empirical evidence.) Because so little is understood, it is useful to begin the empirical portion of this study with a description of the sample, and hence of the South African population. The picture that emerges throughout the chapter is often startling and entirely unexpected. Because the evidence is so unforeseen, I establish here a context for public opinion in South Africa that is essential for understanding the remainder of the book. Many readers will find the South Africa depicted here to be quite at odds with the South Africa of their imagination, or even with the South Africa they think they know. I begin this discussion with a brief overview of the apartheid system that dominated South Africa until 1994. A Brief History of Apartheid Apartheid—which may be translated as “separateness”—swept South Africa as a consequence of the electoral victory of the National Party in 1948. Of course, the introduction of apartheid to South Africa was not the first appearance of racism and racial subjugation in the country—from the moment the Dutch colonists landed in 1652, the Europeans sought to implement a system based on racial hierarchy—but apartheid was a codification of racism that the world had never seen before. Apartheid was manifest in a body of legislation defining racial groups and delineating many of the crucial aspects of people’s lives.2 A full chronology of the statutory basis of apartheid was produced by the Truth and Reconciliation Commission (1998, vol. 1, ch. 13, pp. 448–97). Free download from www.hsrcpress.ac.za Apartheid’s Legacy 31 As a system based on racial hierarchy, apartheid naturally found it necessary to construct racial categories for all South Africans: blacks (or Africans), whites, Colored people, and those of Indian origin. The Population Registration Act of 1950 established these four racial groups. For example: A White person is one who is in appearance obviously white—and not generally accepted as Coloured—or who is generally accepted as White—and is not obviously Non-White, provided that a person shall not be classified as a White person if one of his natural parents has been classified as a Coloured person or a Bantu. . . . A Bantu is a person who is, or is generally accepted as, a member of any aboriginal race or tribe of Africa. . . . [A] Coloured is a person who is not a white person or a Bantu. (Truth and Reconciliation Commission 1998, vol. 1, ch. 2, sect. 26, p. 30) One’s rights and responsibilities under the apartheid system were defined by one’s race, as established by law. Race was the defining characteristic of politics under the apartheid system established by the National Party. These categories, as noted in chapter 1, continue to dominate discussions of race in contemporary South Africa. The aim of apartheid was the total separation of blacks and whites. To achieve this goal, South Africa established the so-called Bantustans as areas where blacks were allowed to reside. Apartheid never sought the expulsion of Colored people and those of Asian origin from the territory of the country, and indeed later reforms of the system provided for political representation for these two groups. (In 1983 separate parliamentary chambers were established for whites, Colored people, and those of Indian origin.) Perhaps the ultimate degradation of apartheid was to deny blacks citizenship in the country of their heritage and birth. T. R. H. Davenport and Christopher Saunders (2000, 398) describe the “Afrikaner Nationalist dream state” as “an independent white-dominated republic, surrounded by a cluster of economically dependent and therefore politically impotent black client states.” As was shown by the truth and reconciliation process, apartheid also had a dark, extralegal, and illegal side. The state engaged in systematic political repression against the liberation forces (see de Free download from www.hsrcpress.ac.za 32 Overcoming Apartheid Kock 1998), with state-sponsored actions ranging from “dirty tricks” to the development of a vast system of informants, to assassinations and murders. The state also pursued a policy of “divide and conquer” against its adversaries through its (successful) efforts to foment political violence between the African National Congress (ANC) and the Inkatha Freedom Party (IFP). Apartheid is often defined as a system of laws, but in fact it was a conglomeration of legal and illegal means of separating blacks and whites and subjecting the former to subjugation and repression while providing vast subsidies to the small white minority. When did apartheid end? Answering this question is more difficult than it might seem. Some believe today that an “apartheid mentality” is still pervasive in the country and therefore apartheid has not yet ended. The formal end of the apartheid system is perhaps best signaled by the ascension to power of the ANC as a result of the first democratic elections ever held in South Africa (in April 1994). But the undoing of apartheid actually began in late 1989 (when, for example, public beaches were desegregated) and was accelerated with the unbanning of the ANC (February 1990) and the release of the ANC leader Nelson Mandela from prison shortly thereafter. A Demographic Profile of the Sample At the time of the survey, Statistics South Africa (as reported in Forgey et al. 2001, 49) estimated that the population of South Africa was 77.6 percent African, 8.7 percent Colored, 2.5 percent Indian or Asian, and 10.3 percent white.3 Thus, whites in South Africa are a small minority (smaller than the size of the African American population in the United States). South Africans of “minority” races constitute slightly less than one-fourth of the population. The design of this survey takes into account the distribution of race in South Africa. Specifically, because a simple sampling design based on random selection would not generate sufficient numbers of Colored people, people of Asian origin, and whites for statistical analysis, the survey includes oversamples of these groups. Appen- Free download from www.hsrcpress.ac.za Apartheid’s Legacy 33 dix A provides details on the sampling strategy, but I note here that the sample includes interviews with 2,004 Africans, 991 whites, 487 Colored people, and 245 South Africans of Asian origin. When I speak of the South African population, I of course weight the data appropriately so that the racial distribution in the sample approximates the population figures described in the last paragraph. Race is more than a technical complication in sampling. Most research—including our own (for example, Gibson 2003; Gibson and Gouws 2003)—reveals deep divisions in South Africa according to race, and it is therefore always reasonable to hypothesize that racial differences exist in the variables considered in this book. That hypothesis is resoundingly reinforced throughout this analysis, so most of the statistical results reported herein are reported separately by these four racial groups.4 I believe that race is a social construct, not a biological one (as Colored people so amply demonstrate),5 and therefore I do not treat race as some sort of determinant attribute of individuals.6 Consequently, my analysis often begins with a simple description of how, and to what degree, the different racial groups differ. But the important, and difficult, task is to determine why racial differences exist. For example, an oft-tested hypothesis is that racial differences can be explained by reference to differences in the way the apartheid regime treated the four racial groups. I also typically ask whether racial differences are in fact reflections of differences in group economic positions within South Africa. Thus, race is omnipresent in the analysis that follows, but race rarely occupies serious conceptual status in my models. Race is almost always a surrogate for some other variable representing the experiences or attitudes of members of the group. As the basic demographic figures reported here make plain, understanding race as a “black versus white” dichotomy is not particularly helpful. It is certainly true that the ideology of the liberation struggle joined Africans, Colored, and those of Asian origin in the single group “Blacks.”7 But whether these three groups share characteristics, experiences, and attitudes is an empirical question, and there are actually many good reasons to suspect that they may not be united. The analysis that follows does not prejudge the question 34 Overcoming Apartheid Free download from www.hsrcpress.ac.za Table 2.1 Demographic Attributes of South Africa’s Racial Groups Attribute African White Colored Asian Origin Gender Percentage female 50.1% 50.7% 49.3% 50.6% Age 18 to 30 years old 30 to 55 years old 56 years old and older 39.0 48.2 12.8 24.6 49.2 26.2 33.1 79.7 17.2 29.3 53.7 16.9 Level of education None or primary Secondary Tertiary 32.2 61.4 6.3 .2 59.5 40.3 23.2 64.3 12.5 22.4 67.8 9.8 Place of residence Percentage urban 40.3 84.3 59.5 84.1 Social class Upper or upper-middle Middle or nonmanual Manual or skilled, semiskilled Manual, unskilled or unemployed 3.8 14.7 39.6 41.9 23.1 60.0 14.4 2.4 8.0 22.0 35.7 34.3 11.0 31.8 43.7 13.5 Unemployment Percentage unemployed 34.7 3.9 20.3 15.1 8.4 17.4 12.3 12.2 Home language Afrikaans English Xhosa Zulu .1 .1 25.8 25.8 52.9 47.1 .0 .0 78.6 21.1 .0 .0 2.9 96.3 .0 .0 Literacy Percentage literate 42.9 97.0 75.8 79.6 Religiosity Percentage attending religious services more often than monthly 64.0 54.3 71.8 70.6 Social class—ownership of goods Refrigerator and/or freezer Electric floor polisher Vacuum cleaner Microwave oven Hi-fi music center 66.6 2.0 4.3 12.5 59.1 97.1 27.3 86.1 83.4 87.5 91.0 14.8 33.5 44.1 66.5 98.4 14.7 44.5 72.2 77.6 Opinion leadership Percentage opinion leaders Apartheid’s Legacy Table 2.1 (Continued ) Attribute Free download from www.hsrcpress.ac.za 35 Automatic washing machine Working telephone Television set Bank account Pension fund Automobile Percentage owning none of these African White Colored Asian Origin 6.8 18.7 73.8 39.9 15.4 9.7 10.6 88.8 81.5 95.1 94.3 68.2 85.0 .1 56.7 55.4 90.8 55.0 35.5 36.8 2.1 56.7 77.1 95.9 62.9 25.7 48.6 .8 Source: Author’s compilation from the 2001 Truth and Reconciliation Survey. of differences across these groups, and therefore I report most of the statistical results according to the four racial groups identified here. Indeed, it is perhaps instructive to begin this analysis with a consideration of the basic socioeconomic differences across these various groups. Table 2.1 reports a panoply of information about the attributes of those we interviewed. Rather than address every piece of information in the table directly (a daunting task), I use these data to draw some general conclusions. First, the sample clearly represents the diversity of the South African population. Unlike some earlier research, for instance, urban and rural residents are included in the survey. The diversity of home languages among whites and Colored people is noteworthy: the white portion of the sample is evenly divided between Englishand Afrikaans-speakers; most Colored respondents speak Afrikaans, although a sizable minority speak English as their home language. Note as well that the South African population is reasonably young, with between one-fourth and one-third of each racial group being thirty years old or younger. Consequently, a sizable portion of South Africans have never directly experienced the heavy boot of apartheid. Although it is difficult to identify a specific time at which apartheid ended, two dates mark important milestones: 1986, when the pass laws were abolished, and 1990, when Free download from www.hsrcpress.ac.za 36 Overcoming Apartheid the ANC was unbanned and Nelson Mandela was released from prison. If I (somewhat arbitrarily) split the difference in these figures and posit that apartheid ended in 1988, then 42.1 percent of the African respondents were eighteen years old or younger when it became defunct. One-third of the respondents were less than sixteen years old. Thus, upward of one-half of the black population in South Africa most likely never experienced apartheid directly, at least not as adults, the age when the full force of apartheid was felt. For a large percentage of Africans, these events took place in their childhood or early youth and were perhaps largely experienced by their parents, not by them. Third, the sample amply documents the vast economic inequality of South Africans of different races. Consider unemployment first: while more than one-third of the African respondents report being unemployed and looking for work, this is true of only 3.9 percent of the whites.8 The unemployment rate of Africans is twice that of Asian South Africans, and dramatically higher than that of Colored people. Class differences (as judged by the interviewer, and no doubt heavily reflecting employment status) are also staggering, with only 18.5 percent of the black respondents judged to be middle- or upper-class, compared to 83.1 percent of whites, 30.0 percent of Colored people, and 42.8 percent of the Asian respondents. Just what these class differences mean can be seen in the variables documenting the ownership of various goods. Many of the differences are enormous—85.0 percent of whites own an automobile, compared to 9.7 percent of Africans. Colored people and those of Asian origin appear to have a much higher economic standard of living than black South Africans. The data in this table provide stunning confirmation—if any is needed—of the vast economic inequality in South Africa. Inequalities in education and literacy are also startling. Fewer than one-half of the African respondents are judged to be literate to the extent that they could read with ease the showcards used in the interview.9 Nearly all white respondents are literate, as are large majorities of Colored people and those of Asian origin. Literacy rates reflect educational experiences: nearly one-third of the black respondents have little or no education, compared to 0.2 per- Free download from www.hsrcpress.ac.za Apartheid’s Legacy 37 cent of the white respondents. The black-white differences in these data astound, but so too do the differences between whites and the other two racial minorities, as well as the differences between Africans and Colored people and those of Asian origin. For instance, compared to Africans, nearly twice as many Colored people and those of Asian origin are judged to be literate. These data strongly indicate just how different are the characteristics and experiences of Africans, Colored people, and those of Asian origin, the three groups often thought of as one group—“Black” South Africans. These literacy rates have important implications for this survey. Though we tried to make our questions as simple as possible, the low level of literacy among some respondents surely caused them substantial difficulty in understanding some of the concepts about which they were queried. We anticipated this problem and made certain that virtually all of our questions had an explicit “don’t know” or “uncertain” response available to the respondents, so as to minimize guessing and the fabrication of answers to our queries. One important negative consequence of illiteracy, however, is that the reliability and validity of the measures I employ here most likely vary according to the literacy of the respondent. Unfortunately, the problem is exacerbated by the unequal distribution of literacy across the groups. We must be sensitive to this limitation throughout the analysis that follows.10 The important conclusions of this section of my analysis are that (1) racial differences exist in a wide variety of social and economic attributes, with the result that (2) vast inequality exists among these groups. These conclusions must be borne in mind throughout this analysis. I turn next to a consideration of South Africans’ experiences with apartheid. Perhaps these differ as well across the various racial groups in South Africa. Experiences Under Apartheid The experiences that South Africans had with the apartheid system provide one of the most important independent variables considered throughout the book. As noted earlier in this chapter in the 38 Overcoming Apartheid Free download from www.hsrcpress.ac.za Table 2.2 Racial Differences in Perceived Role Under Apartheid African White Colored Asian Origin Initial response Activist Hero Victor Collaborator Sellout Bystander Spectator Inactive opponent Victim Slave Beneficiary None Don’t know 5.4% 1.4 .7 1.5 .2 17.1 14.9 9.0 19.8 18.1 .4 .2 11.4 1.2% .2 .8 3.0 1.0 25.1 20.8 19.3 3.5 .0 6.0 .0 19.2 6.2% .2 1.5 1.1 .4 21.6 19.1 6.4 15.2 3.9 2.1 .2 22.1 2.0% .0 .4 .0 .8 23.3 14.3 6.1 38.8 2.0 1.6 .0 10.6 Any response Activist Hero Victor Collaborator Sellout Bystander Spectator Inactive opponent Victim Slave Beneficiary 7.4 2.0 1.8 3.0 .3 23.3 22.4 12.4 31.4 26.5 .6 1.7 .2 1.6 3.6 1.3 34.1 34.9 22.4 5.7 .3 8.4 7.5 .6 3.0 1.5 .9 29.8 26.1 8.4 20.6 6.0 2.6 2.0 .4 .8 .0 .8 33.1 23.3 7.8 43.3 3.3 2.0 Source: Author’s compilation from the 2001 Truth and Reconciliation Survey. Note: Entries are the percentage of all respondents of each race selecting the term. discussion of the age distribution of the sample, a large percentage of the respondents were only youngsters when apartheid nominally ended. Thus, determining how apartheid treated each respondent is an important task for this analysis. Table 2.2 reports our respondents’ replies to what is a deceptively simple question: “People use many different words to describe their relationship with apartheid in the past. Which of the following best describes your role under apartheid?” The closedend list consisted of eleven terms (reported in the table). These Free download from www.hsrcpress.ac.za Apartheid’s Legacy 39 roles were constructed on the basis of the replies of the participants in our focus groups (for a discussion of the focus groups, see appendix A). Up to two responses were recorded.11 The first portion of the table reports the initial self-characterization, while the second half of the table reports the percentages of respondents mentioning the term in either their first or second reply. The terms most commonly chosen by the Africans in our sample are “victim” and “slave,” selected by roughly 20 percent of the respondents as their initial characterization and by a majority of the respondents as one of their multiple responses. A second large group of Africans acknowledges that they were largely onlookers to apartheid (bystanders, spectators, or inactive opponents). It is perhaps surprising to note how few of the black respondents claim the term “activist”: only 7.4 percent selected this as one of their two responses. (Another tiny percentage selected “hero,” a possible synonym for activist.) Assuming that these respondents are being generous with themselves, the levels of self-characterized active opposition to apartheid seem remarkably low for what many perceive to have been a mass movement against apartheid. Whites too are willing to describe themselves as not much involved with apartheid, with “bystander,” “spectator,” and “inactive opponent” attracting a large percentage of the white respondents. Virtually none of the whites in the sample claims to have been an activist. Colored people are most likely to characterize themselves as bystanders (21.6 percent) or spectators (19.1 percent), with another 22.1 percent saying they do not know how to describe their role. A fairly small minority think of themselves as victims or slaves. These are important findings because they suggest that Colored people were especially disengaged from the struggle over apartheid. Being neither fully oppressed nor fully free, Colored people occupied an ambivalent position under the apartheid system. The principal term embraced by the Asian respondents, “victim,” was chosen by 38.8 percent of this group. This figure stands out as much larger than that for the other groups. South Africans of Asian origin may feel victimized by both apartheid and the struggle against it, although this cannot be clea...