Review Symposium Response

It is a great honor and pleasure to have my book as the subject of this review symposium and to benefit from the extremely thoughtful remarks of the group of scholars included here, each a major author on the theme of transitional justice. Each scholar brings a wealth of research and experience to the subject from a diverse perspective. The comments are too rich and varied for me to do justice to all of them in my response, but I hope to address some of the major points of each reviewer. In particular, I will address issues about (1) the policy scope of the book, (2) methods of comparison, (3) data concerns, and (4) processes of historical change. First, on policy scope, Bronwyn Anne Leebaw points out correctly that I do not intend the book to be a history or analysis of the entire field of transitional justice, but rather a comprehensive but focused account of a central transitional justice mechanism—prosecutions for individual criminal accountability. To do a history of the emergence and spread of this single transitional justice mechanism, as well as a systematic analysis of its effectiveness, was already a huge challenge. As Leslie Vinjamuri notes, I do not argue that prosecutions are more or less popular or effective than other transitional justice mechanisms, such as truth commissions, reparations, or amnesties. Leebaw persuasively reminds us that activists themselves have had long and continuing debates about the role and limits of legalism, something she has explored eloquently in her own work.1 These debates about the limits of legalism were often couched as debates about the advantages of alternative transitional justice mechanisms. While I did not make these debates a central focus on the book, I try to address them at times. Since it is impossible to present in the book even a fraction of the number of those who worked for accountability, the stories of a few actors stand in for the countless individuals and groups who work for justice. Thus I present skeptical arguments about justice by respected colleagues and friends like José Zalaquett or Ellen Lutz, in the

hope that these will stand in for broader debates in the human rights community. Ellen Lutz, to whom the book is dedicated, was often quite skeptical about trials, and I quote her as saying something that captures the concerns of many justice activists. "Whenever possible, justice for human rights crimes must be as accessible as possible to those who suffered most and that the best kind of justice leads to national acknowledgement of the wrongs that occurred and societal involvement in righting them" (131). I also briefly present in a number of places in the book Zalaquett's position in favor of truth commissions and reconciliation to represent innumerable others who hold this position. Many activists believed sincerely, and still believe today, that one could either have truth or justice, but not both-either peace or justice, but not both. I show that, although activists may believe these options to be mutually exclusive, our database and research reveal that they have not been so in practice, and I urge us to move beyond debates that present options in such dichotomous terms. Though much is made in these reviews of my "enthusiasm" for trials, I do not believe nor do I say at any point in the book that other transitional justice mechanisms are flawed or insignificant. Rather, this is a book that focuses on human rights prosecutions. At one point, I sum up my argument by saying that "Human rights prosecutions are only one of the many forces and pressures that can contribute to positive human rights change. They are not a panacea for human rights problems; they appear to be one form of sanction that can contribute to the institutional and political changes necessary to limit repression" (187, emphasis added). In this sense, it should be clear that I completely agree with the point that Jelena Subotic makes near the end of her remarks, "Human rights prosecutions are an important piece of this project but they should be just the start." Second, on regional scope, Vinjamuri is correct to signal the multiple sources of the justice cascade, something I try to underscore in chapter four in particular, when I talk about the two streams flowing into the cascade, as well as the long process of building the "hard-law" streambed that underpins the justice cascade and makes it more enduring. I have focused in particular on the "second" stream running into the justice cascade, that of domestic trials starting in Greece, Portugal, and Argentina, exactly because it is the part of the story that is least well known and documented. It will come as no surprise to any international relations or international law scholar that the first stream I say leads to the justice cascade started with the Nuremberg trials. But the role of domestic prosecutions around the world, and particularly early prosecutions in Latin America, is understudied and much less well known. Perhaps the most original research contribution of the book is bringing to light and documenting this "second stream," and it serves as an important counterbalance to those theorists who suggest that transitional justice initiatives have come primarily from the Global North and been imposed on unwilling countries of the Global South. One problem with the "cascade" metaphor is that people sometimes think this means that practices diffuse from the international to the local. But the book also stresses the possibility of bottom-up diffusion of transitional justice innovations from the global South. The protagonism of Latin America with regards to trials is not the only case of transitional justice innovation. As both Leebaw and Vinjamuri point out, South Africa has been an innovator of restorative justice mechanisms. Likewise, Eastern European countries innovated with lustration, and African and some Asian countries have been leaders in thinking about how traditional forms of justice can be adapted to address past atrocities. These are not the topic of this book, but I believe that one very promising line of future research would be to trace the origin and spread of other transitional justice mechanisms from other regions of the world, much as I have done to trace the importance of trials in Latin America.
Third, let me be clear about my argument with regard to "comparison to the ideal." I say that there are three main ways of evaluating the effectiveness of transitional justice: comparison to the ideal, counterfactual analysis, and empirical analysis. I argue that comparison to the ideal is an important form of ethical reasoning used by both activists and scholars and a powerful tool for promoting change in the world. But, I think it is useful to distinguish between comparison to the ideal and systematic empirical comparison. Because of problems I see with both comparison to the ideal and counterfactual analysis, I state that I "prefer systematic empirical comparison." I don't say others must share my preference, but I encourage those who engage in comparison to the ideal to be explicit about their yardstick of comparison. While an empirical comparison would compare countries using trials to countries that do not use trials, comparison to the ideal compares actual trials to our ideals and aspirations for justice. It is completely legitimate to compare actually existing trials to our aspirations for justice, but it is important to note that in such a comparison, existing trials will almost by definition fall short. The question is not if they fall short of our ideals, but how far short they fall. This is quite a different kind of yardstick (one with only negative numbers on it), and I think readers should be alerted to the kinds of yardsticks we are using to measure effectiveness. Second, different individuals have very different ideals of justice, and, if they are not explicit about their yardsticks, the readers might not understand the standard to which actual trials are being compared. The yardsticks I am using are not ideals about democracy or human rights but measures of these practices in actual countries. In response to Leebaw's comment, I would clarify that I do not have a different standard for those criticizing human rights institutions and those criticizing government. In both cases, I prefer empirical comparisons. I criticize the US failure to prosecute because I compare it to other real countries in the world, countries with more fragile political and judicial systems that have nevertheless been able to hold their own state officials accountable.
On questions of data and hypothesis testing, I would like to stress that this book is only a first step in an ongoing research agenda. In his remarks, David Mendeloff lays out an entire research agenda for the future of transitional justice researchers, much as he did in his influential article on the topic, "Truth-Seeking, Truth-Telling, and Postconflict Peacebuilding: Curb the Enthusiasm?" (2004). Mendeloff illustrates the hard questions that we need to continue to ask in transitional justice research. I plead guilty to not doing all these things in a single book. Over the next years, I hope to begin to take on some of the issues that Mendeloff raises in his comments. Mendeloff correctly suggests that we should not only compare the effectiveness of prosecutions to truth commissions, something Hun Joon Kim and I have done in our joint work, but also to judicial reform, security sector reform, support for human rights nongovernmental organizations (NGOs), formal oversight, creations of human rights commissions, etc. While Mendeloff is correct that it would be useful to compare prosecutions to these other activities, he is surely aware that global data sets for most of these activities do not yet exist. I am happy to say that the data Hun Joon Kim and I have used in our research has long been publicly available, 2 but more importantly, by the time this review is in print, the updated website of transitional justice data (prosecutions, truth commissions, and amnesties) that Leigh Payne and I have gathered with the support of a National Science Foundation grant and the entire Oxford-University of Minnesota team will be available at transitionaljusticedata.com. 3 Scholars will now be able not only to download the quantitative data that we have gathered but also to access the detailed qualitative data on which the quantitative data are based. Researchers will be able to know, for example, how many human rights prosecutions have taken place in the Democratic Republic of the Congo, the names and ranks of the accused, the charges against them, and (in cases where the trials have been completed) the outcome of the trials. This should address Mendeloff's "uneasiness" that the data are opaque and will allow researchers to see for themselves the kinds of prosecutions, truth commissions, and amnesties we are documenting. In particular, more research about the sequencing of transitional justice mechanisms should be possible, something Vinjamuri correctly suggests should be done.
We very much encourage researchers to continue to test our arguments. Vinjamuri suggests that I should be wary of the reactions to my claims, but I believe that the field will benefit from having this database available to test my arguments and those of others, and I welcome comments and suggestions about how the database and website can be improved. In addition, Leigh Payne and I are currently seeking additional funding, though it has not yet been secured, to expand this database and website to also include both quantitative and qualitative data on alternative accountability mechanisms, including reparations, lustration and vetting policies, traditional forms of justice, and civil trials for damages for past human rights violations. When and if such data is widely available, scholars will be able to address more satisfactorily many of the research questions and agendas posed by Mendeloff, Vinjamuri, and the other reviewers in this symposium.
While I applaud the ambitious research agendas proposed by the reviewers and am doing everything possible to help create and make publicly available the data to advance these agendas, I occasionally disagree with a few of their characterizations of my argument.
The main argument of the book in terms of the impact of prosecutions is that human rights prosecutions are associated with improvements in human rights practices "probably through a combination of deterrence and socialization." In other words, while I argue and present persuasive evidence that human rights prosecutions are correlated with improvements in human rights practices, my claims about deterrence are more preliminary. I recognize that it is much more difficult to determine exactly the mechanisms through which trials exert their effects. A deterrence argument does not require an assumption of perfect rationality, and I do not make such an assumption. Rather, deterrence is consistent with various understandings of bounded rationality. In the book, I recognize that we do not yet know the exact mechanisms through which trials exert their influence, though I believe that deterrence is part of the explanation. I continue to work on these issues with colleagues. For example, we have now recoded our database to determine the number of human rights prosecutions that result in convictions, and Hun Joon Kim and I presented a paper at the International Studies Association (ISA) 2011 meetings using this new data (Kim and Sikkink 2011). In it, we find that prosecutions that result in convictions have a greater effect than those that do not. This is ongoing evidence for the deterrence hypothesis, but given that trials which led to acquittals also have a positive effect, it does not rule out the importance of socialization as well.
In her remarks, Jelena Subotic also poses important research questions for the field. In particular, she urges us to look more at a series of variables that have not yet received sufficient attention from researchers, including factors about the domestic legal system, rule of law, judicial independence, corruption, etc, which could make it more or less likely that states will start human rights prosecutions. I completely agree with her suggestions that researchers should move in this direction. In an article with Geoff Dancy in the NYU Journal of International Law and Politics, for example, we address a number of the issues that Subotic raises in her review (Dancy and Sikkink 2012). We find that some domestic legal institutional variables such as judicial independence explain variation in human rights prosecutions across a full sample of all countries, but we do not find evidence that they explain variation in transitional countries. Other legal variables, such as British legal heritage, which we expected to influence the use of trials, had no relationship with the incidence of prosecutions. We found that an international factor-the ratification of specific human rights treaties with provisions for individual criminal prosecution-is a robust predictor of the use of trials in both sets of countries. In this same article, we also take up an issue that Subotic stresses in her comments and in her important book (Subotic 2009). We discuss different "pathways" through which countries adopt and use human rights prosecutions. Some countries, such as Serbia and Montenegro, use trials primarily because of external pressure, as Subotic would suggest. Others countries, like Argentina, use trials primarily as a result of intrinsic domestic demand, while in a third group, there are more interesting interactions between international institutions and domestic societies in the search for accountability. For example, Guatemalan courts have finally initiated a prosecution against former President General Rios Montt, the first time a Latin American country has ever prosecuted a former head of state for genocide. Although Guatemalan human rights and civil society organizations have worked for decades for such a trial, they were aided by external pressure from the US government as well as from foreign prosecutions in Spain. We believe that the third pathway is probably the most common, but more research is needed to examine the various interactions involved.
The final point I want to underscore is about the nature and dynamics of political change. A number of the reviewers suggest I believe that this process of demanding and securing justice was easy or without contestation. To the contrary, struggle and conflict are at the heart of the book. I have documented the struggles of activists from Athens to Buenos Aires to New York. The book starts with Uruguay in 1976, where no one can imagine accountability for past human rights violations, and ends with Uruguay in 2010, thirty-five years later, where President Bordaberry was finally convicted for his crimes. I had followed every convoluted twist in the difficult path to justice there and elsewhere. Human rights prosecutions around the globe have grown dramatically in the last 20 years, but in no case have they occurred smoothly and without delays, backlash, push-back, and unexpected legal maneuvers. But recognizing that delays are inherent in such processes is very different from advocating delays, something Vinjamuri has done in her work with Jack Snyder. The adopters of new human rights norms do not contract the ideas through the air like a virus but always through struggle, innovations, and often just plain luck. The justice cascade was not spontaneous nor was it the result of the natural evolution of law or global culture in the countries where the prosecutions occurred. At each step, delay has been the rule, not the exception. Only with the long-term view can the trend towards individual criminal accountability for human rights violations be discerned. I believe that over the long term scholars too will look back at this period as a watershed in the path towards accountability.
I want to thank Hun Joon Kim, the organizer of this review symposium, for his work and patience, and each of the reviewers for taking the time and effort to read the book so carefully and to produce their thoughtful and thought-provoking remarks.