In the decades after World War Two, most scholars working on genocide focused on particular cases, providing historically detailed descriptions of the causes and patterns of mass violence but rarely branching out beyond a specific case. The study of the Holocaust is typical of this; the vast majority of works on the Nazi genocide had little comparative dimension and instead examined the ways in which anti-Semitism and certain policies condemned disfavored minorities to persecution and extermination. These earlier works are particularly important because they gave us rich understandings of the origins, sequencing, and dynamics of mass violence, as well as the roles of dehumanizing cultural views and ideologies that facilitated extermination.Footnote ¹ Nevertheless, multi-case studies were the exception,Footnote ² and generally received little attention in the social sciences.

Much changed in the mid-nineties, when the massive bloodletting in Bosnia and Herzegovina and Rwanda led scholars and policy analysts to focus on other instances of mass atrocity. Today, genocide is receiving greater focus from scholars across the social sciences, and the field of genocide studies itself is growing rapidly: witness the founding of two comparative studies journals,Footnote ³ the growth of scholarly associations on genocide,Footnote ⁴ and the publication of numerous studies and analyses on conceptual clarification, necessary conditions, and the various patterns of genocidal violence. All of this has contributed to a deepening of our understanding of genocidal violence.

Nevertheless, while scholarship in genocide studies has come into its own, significant challenges remain. One of the most vexing is definitional: what, exactly, is meant by “genocide”? How does it differ from other forms of collective violence? The legal definition in the United Nations Genocide Convention ought to serve as a useful touchstone to orient analysis, but it is weakened by conceptual problems. Among other things, it rather arbitrarily delineates acceptable victim categories—ethnic, racial, national, and religious—while leaving out political, economic, and other groups. Furthermore, its contention that genocide requires the destruction of a group “in whole or in part” leaves unspecified the threshold required for mass killing to qualify as genocidal. Finally, it includes a variety of acts that can qualify as genocidal, from explicit killing to causing serious physical or mental harm to forcibly transferring children from one group to another, without specifying what these various policies have in common.Footnote ⁵

Because of these problems, scholars in the field have developed a variety of alternative definitions with the aim of giving the concept some coherence and boundaries.Footnote ⁶ Nevertheless, there is no consensus on its definition. Some scholars adopt a rather restricted view of what qualifies as genocide, focusing only on instances where extermination was driven by an explicit ideology of national purification and cleansing.Footnote ⁷ Others are less concerned with ideology as a bounding concept, and attempt to explain large-scale atrocity more generally, such as by focusing on the systematic physical destruction of groups, regardless of group identity or perpetrator motivation.Footnote ⁸ Yet others have produced a complex taxonomy of violence that includes urbicide, politicide, “ethnic cleansing,” murderous cleansing, and even auto-genocide to explain a variety of phenomena that share family resemblances, in Wittgensteinian terms, with one another and with the UN Convention's definition.

Martin Shaw's What is Genocide? brings new rigor and clarity to debates about the meaning of genocide. In this provocative and engaging book, Shaw draws on the work on Raphael Lemkin, the Polish jurist who coined the term “genocide” in the 1940s,Footnote ⁹ to develop a broad definition of genocide that avoids the pitfalls of strict legalism on the one hand and exclusive reliance on a single case on the other (typical of some scholars who largely equate genocide with the Holocaust, or otherwise treat the Nazi case as fundamentally different from all others).Footnote ¹⁰

For Shaw, genocide is not merely the physical destruction of a group, but includes the destruction of group identity as well. Thus, the destruction of its culture or the basic social and material conditions necessary for collective life can qualify as genocidal. “Killing, along with other forms of physical harm or cruelty, is the ultimate destructive means, which in the end trumps all others. But it is not the primary meaning of group destruction: that lies in the end of annihilating the way of life, social networks, institutions and values of the attacked community. This amounts to destroying the (real or putative) social power of the targeted groups” (Shaw, 33–34). Genocide, then, concerns the coordinated effort to destroy the foundations of group life.

Genocide thus serves as a “master concept” for various forms of violence aimed at eradicating the identity and life of a victim group (78). This has several analytical benefits, according to Shaw: without the overarching concept of genocide, we lose a coherent way of speaking about a variety of techniques that ultimately have the same end, namely, the elimination of the victim group. Furthermore, the master concept avoids adopting an arbitrary definition of victim groups typical of the UN Genocide Convention, which can lead to absurd distinctions between “legitimate” and “illegitimate” victims of genocide (common in debates over the Cambodian genocide, where different ethnic groups are classified as genocide victims according to their identity, and not the consequences of the violence they experienced).Footnote ¹¹ The more inclusive formulation put forth by Shaw avoids the “artificial conceptual fragmentation” typical of legalism (78).

Shaw is persuasive here. By defining genocide as a generic form of social action that includes a variety of techniques aimed at the destruction of a group, scholars are given a useful and inclusive formulation that can encapsulate numerous cases while maintaining conceptual coherence and rigor.

A second claim in What is Genocide? is perhaps more contentious. Drawing on his earlier work on degenerate warfare,Footnote ¹² Shaw contends that genocide is a form of war, although the target is a civilian population rather than an enemy army. “All groups or populations targeted in genocide are overwhelmingly civilian in character, and it is their civilian identity that makes their targeting genocidal” (113). The point here is that perpetrators reconceptualize civilian groups as military enemies, as well as political and economic opponents, thus “requiring” their destruction. As a degenerate form of warfare, genocide lacks the core rationality of conflicts between armies. Thus, counterinsurgency strategies, guerrilla wars, and other forms of conflict that reframe civilians as military actors are much more likely to become genocidal. Shaw's understanding of genocide as degenerate warfare shares something of the blurring of boundaries typical of the recent literature on “new” wars characterized by state fragmentation and civilian targeting, but with much more theoretical sophistication.Footnote ¹³

It may be argued that Shaw stretches the definition of war too far when he includes genocide under its umbrella, for the former normally includes a variety of norms, regulations, “rational” objectives, and clearly defined armed enemies, all features that are not present in genocidal violence. But this criticism would miss the point of his argument. While acknowledging that genocide “removes the last vestiges of rationality and legitimacy that cling to more conventional war” (112), he persuasively argues that genocide symbolically constructs civilians as the objects of techniques and strategies normally reserved for armed combatants, and thus the very nature of genocide has at its core a radicalized version of the logic of mass warfare: to destroy the social power (in this case, the collective identity) of the enemy. It is not surprising, therefore, that genocide often occurs against the background of war. Shaw's formulation serves two purposes: first, it provides an important critique of legalist definitions of war that assume a clear distinction between legitimate combat and illegitimate slaughter, highlighting how warfare often expands to include a variety of forms of attack on civilians; and second, it provides a sociologically rich articulation that emphasizes the complex social interactions between perpetrators, victims, and bystanders which is frequently lost in accounts that focus primarily on perpetrator intention (94–99). The result is a tightly argued book that challenges much of the thinking in genocide studies, and will no doubt become a standard text for scholars in the field. This is social theory in top form.

Ben Kiernan's and Mark Levene's books are contributions to the ongoing debates over the causes of genocide. This is one of the most fertile areas of inquiry in the field, and it has grown rapidly over the past decade or so. These books offer qualitative, multi-case studies that differ from more specific regional or single-case works,Footnote ¹⁴ or quantitative analyses.Footnote ¹⁵

Kiernan's massive Blood and Soil provides a sweeping account of extermination from Sparta through present-day Darfur. A major scholar of the Cambodian genocide,Footnote ¹⁶ Kiernan here offers a beautifully written study that is both global and historical, and devotes significant space to analyzing the conjuncture between imperial expansion and genocide. Kiernan uses the UN Genocide Convention definition of genocide, but applies it to a much broader range of cases than is typical in the literature. The range is impressive, particularly given the tendency of many recent studies to focus on the twentieth century and ignore earlier historical eras. He explores the dynamics of massacres in ancient Rome, historical Japan, Southeast Asia, the Americas, Ireland, Algeria, and sub-Saharan Africa (including the Herero and Nama in German South West Africa). The book also surveys recent large-scale exterminations in the Ottoman Empire, Nazi occupied Europe, China, the Soviet Union, Cambodia, and Rwanda. The numerous examples are treated with sensitivity to their unique dynamics, and Kiernan is careful to note variation in the scope of killing and the multiple motives of perpetrators.

Nevertheless, there are some broad elements common to his cases. For Kiernan, genocidal regimes (a) espoused a racist ideology, (b) were simultaneously involved in territorial expansion, (c) drew extensively on a cult of antiquity, and (d) idealized the life of the peasant (605). These four factors, which are largely focused on ideological representations of the “self” and “other,” place Blood and Soil in opposition to studies that emphasize elite rational behavior or the strategic use of violence to secure economic resources (though the discussion of imperial expansion deftly takes this latter issue into consideration). In this sense, Kiernan's book presents a historically grounded theory of ideology that marks an improvement over studies treating ideology in rather static terms.Footnote ¹⁷

Kiernan provides us with an important set of analytical concepts to study extermination, though it would be interesting to include a more systematic analysis of proximate conditions that explain genocidal outcomes rather than other types of political violence, such as targeted massacres or forced emigration or assimilation. Is there a particular sequence or pattern of radicalization that can be inferred across cases, which might contribute to genocide prevention? The four factors, while certainly common to virtually all of the cases covered in the book, nevertheless seem to over-predict genocide, given that radical ideologies, territorial expansion, and appeals to ancient traditions for legitimacy are more common than genocidal violence. The final factor—the idealization of the peasant life—may serve to give greater focus to the theory and avoid over-prediction, though the romanticization of the farmer and the land operates quite differently depending on the case: the destruction of indigenous communities by farmer/settlers in the Americas is quite different from the bizarre idolization of the peasant in Nazism or Ottoman Turkism. Greater specificity or parsing of the concepts thus might provide more analytical leverage and explain variation in political violence. Allowing for these concerns, Blood and Soil represents a major advancement in the comparative study of genocide that will demand the attention of all scholars of political violence.

Levene's enormous effort, Genocide in the Age of the Nation State, is one of the most audacious and impressive works on the topic to emerge in some time. The project will consist of four volumes when finished (two are currently in print and reviewed here) and attempts nothing less than re-imagining the consequences of the rise of the modern nation-state system.Footnote ¹⁸

Levene is skeptical of approaches that concentrate on particular cases without theorizing their linkages to one another and their place in the world system. There is a deeper dynamic at work, in his view, that is lost in studies that focus on domestic causes of extermination and that often implicitly assume that genocide results from authoritarian or totalitarian regimes that “deviate” from the proper path of liberal democratic modernization.Footnote ¹⁹ For Levene, this misses the fundamental political transformations of the past three hundred years that set the conditions for modern genocidal violence. His project, then, is “less concerned with detailing an exact chronological plot and more with developing a theoretical approach and geared towards introducing the reader to broad patterns, processes and inter-connections between what often may be quite unfamiliar instances of genocide” (Levene, 2:4).

Levene argues that “genocide occurs when a state, perceiving the integrity of its agenda to be threatened by an aggregate population—defined by the state as an organic collectivity, or series of collectivities—seeks to remedy the situation by the systematic, en masse physical elimination of that aggregate, in toto, or until it is no longer perceived to represent a threat” (1:35). He acknowledges that genocide has been a feature of human history probably since the beginning of collective life, but he differs from Kiernan's more uniform understanding of its causes in distinguishing between premodern and modern exterminations. Historically, empires would commit genocide against recalcitrant subject peoples, but this was largely a calculated strategy to ensure compliance and subservience. Empires were generally inclusive of various subnationalities provided they accepted imperial status hierarchies. Levene freely accepts that empires were racist and brutal, but imperial identity was not homogenous; it allowed for a variety of internal collective identities so long as they did not risk subverting imperial authority. Purity, in other words, was not at the center of imperial identity.

The transformation toward homogenous political identities began with the collapse of the Roman Empire and the development of new political entities in Western Europe during the Middle Ages. With Europe's fragmentation into small feudal states and the adoption of Christianity's vocabulary of “pure” and “impure,” “saved” and “fallen,” political conflicts were framed as religious crusades to eliminate heretics and schismatics. Both the Reformation and Counterreformation, combining political power with religious absolutism, emphasized uniformity and allowed for little tolerance of deviant groups (2:125). If the deviant would not convert, they had to be destroyed.

The emphasis on uniformity reached full maturation in the secular ideology of the modern nation state, with its focus on “the people” and hostility toward internal minorities. Consequently, for Levene modern genocide begins not with the colonial massacres in the Americas (though he notes the importance of these), but with the French Revolution and its exterminationist counterinsurgency program against royalists in the Vendée region (2:110). Here, the enemies of the state were not merely political opponents but an existential threat to the very foundations of the young republic, and there was no alternative but to eradicate them tout court.

These ideological genocides, with roots in the absolutism of the early European religious wars, replaced the drive for religious conformity with the secularized, utopian language of race, nation, and class. As new states sought to modernize and wrestle political and economic power from subnational and regional authorities, they invariably framed political conflict as existential struggles. Thus, the path from revolutionary France to the Young Turks to Nazi Germany—all radical modernizers—is short and brutal. As the Western nation-state model has come to replace other forms of political organization, most notably the multicultural dynastic empire, genocides have become more common, and will likely continue to occur as postcolonial states facing severe economic crises attempt to impose a strong sense of national identity and rapidly modernize to “catch up” with other, well-established states. This would explain, for example, the recent genocides in Cambodia, Rwanda, and Darfur. They have all been more than merely brutal wars against insurgents, for at their core they reflect an elite obsession to reshape the meaning of “the people” and destroy impure minorities.

Genocide is a nation-building project, resulting from the emergence of a new form of political organization and social imaginary. For Levene, however, it is the systemic crises faced by new nation states that provide the catalyst for genocidal violence. Extermination of “internal enemies” becomes more likely when newly emerging peripheral and semiperipheral states are confronted with economic pressures to integrate into the world market or face collapse, and thus turn to punishing internal minorities associated with external competitors (Armenians, Jews, “Vietnamese” Khmer, Tutsi). The struggle to maintain a uniform internal identity while securing access to the world economy, often through the export of one or two primary raw commodities like coffee or rice, is fundamentally destabilizing and directs already radicalized elites to attack vulnerable minority groups. The crucial point here for Levene is the systemic nature of genocidal violence, that is, its connection to the rise of a particular form of political organization and economic pressure. Genocide cannot be analyzed by looking solely at the internal dynamics of particular countries, but must take into account the broader political and economic structure that has its origins in Western Europe but today has spread around the globe.

Levene's project has much in common with Michael Mann's The Dark Side of Democracy, which argues that genocide is likely in societies where “the people” is understood in organicist, nonstratified terms.Footnote ²⁰ But Levene goes further than Mann in positing a broader structural change in state formation that is connected to transformations in global economic processes, and thus gives us a fascinating theory of genocide that ultimately reflects something of the structuralism typical of Immanuel Wallerstein's studies on global capitalism.Footnote ²¹ In this regard, Levene's work represents a much-needed addition to the literature, forcing scholars and analysts to consider more seriously broader historical developments that make genocide likely, and recognize the limitations of focusing on the country as the primary level of analysis.

Nevertheless, Levene's approach raises some questions. The first concerns its predictive capacity. The focus on macro historical processes seems to over-predict genocide. Given the ubiquity of the nation-state model today, shouldn't genocide be more common? Surely, political violence is evident enough, but genocide remains relatively rare. Another way to put this is the following: while radical transformations in state structures and acute economic instability may (and often do) result in severe violence, genocide is not the only or even the most likely result.Footnote ²² Naturally, this raises the question of what other proximate conditions or mechanisms can explain why genocide happens when and where it does. Levene's theory does an excellent job explaining the Armenian genocide, which occurred as the Ottoman Empire was collapsing and Young Turk modernizers were attempting to push the Empire into the future while contending with World War I and serious economic instability. But other cases seem a bit harder to explain; Rwanda experienced significant political and economic instability (and violence) during the forty years prior to 1994, so why did genocide occur then? Furthermore, there are many other cases in postcolonial Africa that have followed the crooked path of modernization sketched by Levene but did not pursue the extermination of minority groups, even if we use his definition of genocide. So, what contingent or immediate factors make genocide more likely than, say, targeted massacres, forced migrations, or other awful policies?

These questions about sequencing and timing point to a deeper methodological issue about what we should expect of theory more generally. Both Kiernan and Levene are historians, and while they discuss an enormous number of examples and posit some general patterns and commonalities across cases, they are wary of nomothetic approaches that could predict extermination in some scientifically robust way. Both are too sensitive to background historical conditions in particular cases to put much weight in abstract, predictive theory. For them, the conjunction of certain historically situated phenomena makes genocide probable, but it would be foolish to assume we can generate a scientific theory about causes. For Kiernan, the presence of ideology, colonialism, a cult of antiquity, and idealization of peasant life make genocide likely; for Levene, it is the rise of the nation-state system, the emergence of new conceptions of exclusivist collective identity, and contemporary pressures of political and economic modernization on young, fragile states. But neither thinker attempts to develop a full-fledged theory that can be analyzed for its predictive capacity.

Both books are likely to be more attractive to historians and qualitative social scientists than to political scientists or sociologists using quantitative methods, since neither author provides a hypothesis that is “testable” as such. I suspect some political scientists and sociologists working on genocide will find these accounts pitched at too general a level.Footnote ²³ How, for example, should we measure Kiernan's notion of radical ideology? What is the threshold of economic instability necessary for genocide in Levene's model? Presumably, one may argue, if these are the primary factors that lead to genocide, there should be some way of quantifying and measuring them. This line of criticism assumes that Kiernan and Levene's various explanatory factors can be modeled in some consistent and coherent way, and any variation due to particular national or regional histories can be bracketed for the purposes of testing the authors' respective theories. Thus, there ought to be some consistent way of measuring radical ideology, appeals to antiquity, or economic crises that could show us whether these factors are primary determinants of genocide.

Nevertheless, such a criticism would dismiss too quickly the methodological strengths of these works. Aside from the fact that quantitative early warning theories face a number of predictive limitations themselves,Footnote ²⁴ both authors view genocide in highly dynamic terms, where historical patterns as well as unforeseen contingencies play a pivotal role in shaping the likelihood of genocide. Their rich analyses are sensitive to the hermeneutic limitations of explaining and predicting events as complex as mass political murder, as well as the myriad ways in which perpetrators, victims, and bystanders react to and shape ongoing events. Through careful historical reconstructions of causal patterns and sensitivity to variation across cases, these analyses give significant depth to our knowledge of genocide, a depth that risks being lost in studies that attempt to bracket history from analysis.

I have already mentioned the problem of conceptual consistency in genocide studies; that is, the fact that very few scholars use the same operational definition of genocide, and therefore select different cases for analysis. Variation in case selection makes it difficult to compare alternative causal theories, since they have different objects of study. Levene, for example, analyzes the Vendée massacres, European colonial exterminations, Armenia, the Holocaust, Stalin's attacks against so-called Kulaks (or “wealthy” farmers), Cambodia, and Rwanda among others. Kiernan also presents an enormous list of cases, ranging from Sparta to Darfur. Other scholars, such as Manus Midlarsky, consider only Armenia, the Holocaust, and Rwanda as cases of genocide, while Cambodia is “politicide”; Eric Weitz discusses the Holocaust, the Soviet Kulaks, Cambodia, and Bosnia; Benjamin Valentino combines genocide and “mass killings” into an analysis that includes not only the Holocaust and Rwanda, but also the Guatemalan and Afghani civil wars and the Chinese communist revolution, among many others. Thus, it is difficult, if not impossible, to compare various causal explanations since they are essentially explaining different phenomena. Until scholars reach some degree of consensus on the definition of genocide, it will be difficult to compare theories in any rigorous manner.Footnote ²⁵ This is not a criticism of Levene or Kiernan, but rather a deeper problem in the field of genocide studies that makes further analytical advancement challenging.Footnote ²⁶ Perhaps Shaw's book will contribute to resolving some of these issues and provide a conceptual focal point for future research.

The previous works deal largely with the causes of genocide. The aftermath of genocide has also received significant scholarly attention since the establishment of the two UN tribunals in the mid-nineties: the International Criminal Tribunal for Rwanda (ICTR) and the International Criminal Tribunal for the Former Yugoslavia (ICTY). Much of this literature has focused on why Western states support war crimes tribunals, or the benefits and limitations of formal prosecutions to strengthening the rule of law.Footnote ²⁷ Remarkably, there have been very few book-length works examining the strategies court officials use to ensure cooperation from target states.Footnote ²⁸ Victor Peskin's International Justice in Rwanda and the Balkans fills this gap by providing a careful assessment of the complex dynamics surrounding what he calls “virtual trials,” or the efforts by tribunals to secure state cooperation for war crimes prosecutions. His goal is to explain why tribunal leaders fail or succeed in getting states to arrest suspects and provide needed witnesses and documents for trials. Often, these states are ruled by leaders who are partially complicit in atrocities (as in Serbia and Croatia) or by successor elites whose legitimacy rests on a historical narrative of violence that could be undermined if challenged in court (such as in Rwanda).

Peskin notes that Chief Prosecutors and their allies will employ a variety of incentives to secure support from noncompliant regimes, including conciliatory strategies such as postponing or sealing indictments, or timing them to minimize their disruptive effects domestically. Courts may also resort to shaming states publicly for noncooperation or use the court's moral status as defender of human rights to leverage influence over recalcitrant states. Of course, targeted states engage in their own political maneuverings, such as shaming courts for their shortcomings or biases (used to great effect by Rwanda against the ICTR). States may also argue that compliance with international tribunals would be too disruptive domestically, or that domestic judicial systems are a more appropriate venue for prosecutions.

Part I begins by detailing the major actors and main theoretical arguments, and presents the sources used in the book, which include over three hundred interviews in the Balkans, Rwanda, and at the ICTY and ICTR. Part II focuses on the ICTY's relatively successful efforts in using both conciliatory and confrontational strategies to cajole, coax, and shame the Serbian and Croatian governments to hand over documents and arrest indicted war criminals. Peskin convincingly shows that the Chief Prosecutor's success depended partly on the domestic conditions in target states. Serbia, for example, was (unsurprisingly) unwilling to cooperate during Slobodan Milošević's rule, but after his overthrow subsequent leaders slowly began providing support to the court, eventually handing over Milošević himself (Peskin, 80–83). Croatia under Franjo Tudjman was largely uncooperative, and in the period immediately after his death Croatia played a clever shell game of appearing to work with the tribunal while protecting its own war criminals. It is only in the last several years, as the possibility of joining the European Union emerged, that Croatia has agreed to cooperate with the court (137–40). Peskin's detailed and impressive reconstruction of the Balkan case is a much-needed addition to our understanding of the ICTY's politics.

Part III is devoted to Rwanda and the ICTR. Here, court prosecutors have had to contend with a government composed largely of the Tutsi, the main victims of the Rwandan genocide. The Tutsi-dominated government has been especially unwilling to work with the ICTR because of the court's alleged ineffectiveness, corruption, and arrogance toward domestic justice efforts in Rwanda. Rwandan President Paul Kagame sees the ICTR as little more than the West's way of expressing remorse for its failure to intervene during the genocide. Because Rwanda is generally perceived as a victim state (especially by the United States and Britain, who have until recently been strong supporters of Kagame), it has been more successful in generating international sympathy and opposing the ICTR's “soft power” of shaming (196–200). In fact, only France has consistently criticized and placed pressure on Rwanda, but France's own complicity with the genocidal Hutu regime has undercut its effectiveness.

For Peskin, chief prosecutors are just as much politicians as they are attorneys. They must establish and develop personal relations with influential leaders in target states and the international community, and generate favorable media coverage for their work. The most successful chief prosecutors, such as Carla del Ponte during her tenure with the ICTY, combined a clear understanding of the power of the media and domestic politics with an often relentless shaming of uncooperative leaders. Of course, Peskin is quick to point out the ICTY was less susceptible than the ICTR to charges of incompetence because of its better record with completing trials on time and its overall professionalism. Nevertheless, it is these political “virtual trials,” which consist of competing efforts to frame what justice is, that determine in large part the success of international tribunals. Peskin concludes that successful courts use a combination of conciliatory and adversarial techniques to secure cooperation, but sustained international pressure is important as well. While the conclusion is rather straightforward, Peskin's empirically rich analysis provides one of the most convincing accounts of why this is so.

Two criticisms can be raised of this otherwise very impressive work. In his assessment of the Tutsi Rwandan government, Peskin gives the impression that the ruling party is more united than it in fact is. While his discussion of the Balkans underscores the complex dynamics and disagreements among various elites, his treatment of Rwanda is at times less nuanced.Footnote ²⁹ Greater attention to the ruling party's internal differences could highlight what strategic openings might exist for ICTR officials, and whether and how these openings could be pursued. Even if there were no significant cleavages that could be transformed into support for the court, a more extended analysis of why this is so (along the lines of the analysis of the Balkans chapters) would be welcome.

A second theoretical point concerns the relationship between the three factors in his analysis. Which factor—if any—is the primary determinant of state cooperation with international courts: international intervention, national politics, or the court's soft power? It is not always clear whether one in particular is primary, or whether there are particular contexts in which certain combinations are more important than others. In the discussion of Serbia, international pressure seems to be the dominant factor (90), but the analysis of Croatia emphasizes the court's soft power (147–48). It may be that there have been too few cases to make broad assertions, and thus any theoretical advancements here will have to wait for similarly detailed studies of the hybrid courts in Sierra Leone and Cambodia, and the International Criminal Court. And of course the relationships between these latter tribunals and target countries vary significantly from the Rwanda and Balkans cases (a point Peskin notes in his concluding chapter), so any lessons will have to take into account some basic differences. Nevertheless, it would be helpful if Peskin confronted this theoretical question a bit more explicitly and systematically.

This final point only underscores the importance of Peskin's book, for he has advanced our understanding in a relatively neglected area of study. Future scholars of international criminal courts will have to contend with this very impressive work. Indeed, Peskin has the set the bar high for future comparative studies of tribunal politics.

The books reviewed here deserve attention not only from specialists working on genocide, but from scholars interested in the causes of political violence more broadly. The sophisticated theoretical frameworks presented by Kiernan and Levene, as well as the impressive conceptual critique offered by Shaw, show that the comparative study of genocide is generating more rigorous theoretical insights into these terrible events. Peskin's contribution will also likely make an important mark on postconflict studies, and deserves to be read by both scholars and practitioners of international justice.