Eight years ago, reviewing the state of the scholarly literature about the Rwandan genocide I observed that “[t]here is currently a major gap between the explanations developed by scholars—almost all situated at the macro level—and the nitty-gritty of the genocide: a series of acts of violence by some individuals against other individuals. We do not really know who killed who, where, why, and how. To be precise, we do know quite a bit on the ‘who’ issue, but not in a scientific manner. What we do know is written down by human rights experts; foremost the stunning work of Alison des Forges in Leave None to Tell the Story; but also the impressive early work by African Rights: Rwanda: Death and Despair. … However, this wealth of facts has not been connected to social science theorizing. There seem to be two parallel spheres: one of masses of descriptive micro-level data without scientific hypothesizing and theorizing; and one of theory-informed macro-level analysis” (“Reading the Rwandan Genocide,” International Studies Review 3 [Fall 2001]: 75–99).

Those days are now over, and the fine work under review here is one of the reasons why. In recent years, a new generation of scholars has gone into Rwanda's hills and prisons and talked to hundreds of people—victims, perpetrators, accused, and bystanders (although Fujii precisely argues that these categories are not remotely as distinct or clear as they appear, with ordinary Rwandans frequently having moved between all of them in the span of months, if not days)—in different locales. They do so supported by good social science theorizing, and the results are important not only for Rwanda but also for our understanding of mass violence in general.

In undertaking this kind of research, Lee Ann Fujii comes to some surprising conclusions: People who participated in the genocide, she argues, do not reveal a higher degree of ethnic hatred, did not subscribe to racist values more strongly, and indeed, in general did not particularly consider the Hutu–Tutsi divide a crucial part of daily life. “State sanctioned ethnicity,” as Fujii calls it, is simply not something that has much purchase in Rwandans' life. Other scholars who have done very similar research in Rwanda make the exact same observations; they include Scott Strauss (The Order of Genocide: Race, Power and War in Rwanda, 2006) and Omar McDoom (“The Micro-Politics of Mass Violence: Authority, Security, and Opportunity in Rwanda's Genocide,” Ph.D. thesis, London School of Economics, 2009). Neither ethnic fear nor ethnic hatred, then, can explain genocide (pp. 77, 181 ff.). Ethnicity may be what defines the target group of the genocide. But it is not automatically what causes the genocide (p. 180).

What, then, does explain why hundreds of thousands of Rwandans engaged in the wholesale slaughter of hundreds of thousands of other Rwandans who were frequently—as the title of this book states—their neighbors? Fujii's answer here is original, and while her work shares much with the work of Strauss and McDoom, it offers a distinctly different way of presenting the process by which people of all kinds, for very different reasons, became at least partly complicit in the genocidal machinery. Fujii uses the concept of a script—“rules and expectations for behavior”—and explains how such a script was enacted during a period of radical upheaval and war. According to Fujii, the script's writers are the genocidal masterminds, and their reason for writing this script are their desire to hold on to power at all cost, by presenting the civil war as an ethnic war, all Tutsi as the “enemy within,” and mass murder under their leadership as the only solution. In developing these themes, the author covers well-trodden scholarly ground.

Fujii goes further, however. For local power holders, she argues, “the script does not represent a set of instructions they must follow to the letter, but rather an opportunity to apply their own interpretations to the text” (pp. 12–13). This fits with Strauss's insistence that local (not national) politics is a key driver of the spread of the genocide. Local power holders adapt the centrally written script to local circumstances, their own interpretations, and their own political needs. But the degree of reinterpretation does not stop there. Indeed, as Fujii writes, “no director's vision is ever hegemonic. Realization of the director's ideas depends on the actors themselves—their skills, interests, and commitment level” (p. 13). This allows her to analyze the many different motives and levels of involvement that Rwandans had during the genocide. The keys here are local social ties and group dynamics, it is these that activate ethnic scripts, and people's participation in the genocide is very different depending on the people with whom they associate, when, where, why, and so on.

The usage of the concept of a script—and associated categories of directors, actors, and so on—is mostly metaphorical, but it is a nice one, and it allows Fujii to incorporate a wealth of empirical data that she has collected from her interviews. She provides a finely nuanced approach to deal with a social process that she sees as highly ambiguous and complex. The historical sections of the book—both the ones based on documentary evidence and those based on the interviews she conducted—reinforce this approach and demonstrate how the core concepts involved—including the very meaning of ethnicity itself—have been ambiguous and fluid over the course of many decades.

This is a fascinating and, I believe, largely correct account. The evil of genocide is so clear that it is hard for us to think of it as an ambiguous and fluid process. And yet at its very core, genocide is less about ethnic hatred than the diverse and local social dynamics that drive it forward.

I must confess that something about such an understanding leaves me unsatisfied. In almost every situation she depicts, Fujii goes out of her way to demonstrate that “state-sponsored ethnicity” has no real purchase in people's lives, and that ethnic distinctions are fluid, ambiguous, and unimportant. Yet over and over in Rwanda's recent history (1959, 1963, 1973, 1990–94), people have murdered each other along ethnic lines. Either Rwandans have a very high tolerance for murder (and rape, looting, destruction) or they are truly amazingly manipulable, and at the slightest opportunity or inducement to participate in violent ethnic scripts they comply. If ethnic identities are so unimportant, why are people repeatedly so willing to play violent roles in scripts that are based on extreme ethnic exclusion? And why are these scripts always so violently ethnic in the first place? To me at least, this question remains unanswered.

That said, this is an excellent book. It uses oral and written sources very well, is beautifully written and clearly argued, and is truly a pleasure to read (notwithstanding its grim subject). It intelligently describes and tries to explain the ambiguity of the genocide, a process that seems so clear (and evil) and yet, upon closer inspection, becomes far more blurry. Backed by other recent scholarship that comes to similar conclusions, Killing Neighbors is a book that poses tough questions and contains far-reaching implications for thinking about genocide and civil violence more generally.