Jonathan Fox's book makes a very important contribution to the ongoing debate on the impact of religion on domestic conflict. On the one hand, the author takes issue with Samuel Huntington's clash-of-civilizations thesis. As many others before him, Fox does not find much quantitative support for Huntington's bold expectations: We neither experience a major reorganization of international relations along civilizationally defined fault lines nor observe a significant increase in the number of violent disputes involving parties from different civilizations. On the other hand, he objects to those recent studies according to which the onset of civil wars is determined by political and economic factors, while religion is found to be largely irrelevant (see, for instance, James Fearon and David Laitin, “Ethnicity, Insurgency, and Civil War,” American Political Science Review 97 [no. 1, 2003]: 75–90, or Errol Henderson and David J. Singer, “Civil War in the Post-Colonial World, 1946–92,” Journal of Peace Research 37 [no. 3, 2000]: 275–99). By using data both from the Minorities at Risk Project (MAR) and the State Failure Project (SF), Religion, Civilization, and Civil War provides strong evidence that religious identity—whether the conflict involves two groups who belong to different religions or to different denominations—and, even more so, religious grievances do influence conflict escalation. This can be best studied when the crude distinctions of Huntington are questioned and religious diversity within civilizations is recognized.

The reader is presented with many interesting findings that cannot be exhaustively addressed in a single review. Most importantly, the author argues that conflicts involving religious issues became increasingly violent during the 1980s and that they are now even more violent than nonreligious disputes. This finding fits nicely with a recent study by Andrej Tusicisny (“Civilizational Conflicts: More Frequent, Longer, and Bloodier?” Journal of Peace Research 41 [no. 4, 2004]: 485–98). Although religion never appeared to be the primary cause of conflict, it always operated as an important intervening variable that, especially in combination with separatism, produced consistent results. Second, Fox found no religion or denomination to be particularly aggressive. Or to put it differently, the specific faith of the conflicting parties has no significant influence on the level of violence. This finding points to the important role of political elites within religions that more or less successfully seize upon preexisting traditions. Third, religious institutions matter. Fox found that groups with no formal religious institutions rebel less than do better-organized communities. However, it is not the groups with the most organized religious institutions—those with large-scale formal ecclesiastical networks—who use the most violence in their conflicts, but rather those with comparatively flat and nonhierarchical organizational structures. This observation supports similar findings of the Fundamentalism Project of the American Academy of Arts and Sciences. Fourth, political and military intervention by external powers appears to be more likely when the conflicting parties in the target state differ in their religious identities or when religious issues are at stake.

Finally, Fox reasonably speculates about conflicts involving religious issues that do not involve differences in religious identity. This is a very important step, since most studies searching for religious correlates of civil war still focus on Huntington's crude distinction of seven or eight civilizations. By ignoring religious diversity within civilizations and coding conflicts involving, for instance, secular governments and militant fundamentalists as intrareligious disputes, these studies artificially increase the number of observations of armed conflicts that do not involve religious differences. To illustrate this point, Fox found that during the 1990s, a growing proportion of conflicts within Islamic states involved religious issues and that these conflicts tended to become more and more violent. To consider these conflicts intrareligious would lead to an underestimation of the impact of religion on dispute escalation.

Repeatedly throughout the book, Fox stresses that his study constitutes the most thorough quantitative analysis of the impact of religion on domestic conflict to date. This claim certainly has some merits, and the author indisputably presents an authoritative analysis of MAR and SF data on religion and violence. However, some critical limitations of his study should be mentioned. First of all, the multivariate analyses of violent disputes concentrate on ethnic conflicts and are limited to the 1990s. Whether the patterns found for this particular type of conflict in this particular time period can be generalized has still to be investigated. Additionally, the author does not address explicitly scholarly work on civil wars presenting conflicting evidence to his claims. Most important in this regard, his study fails to include control variables that are widely used in other quantitative studies, such as a country's economic development or the dependency of violent disputes on past conflicts. Whether his findings will survive these tests is still an open question. Second, while Fox discovered a number of interesting empirical patterns, he is reluctant to address them theoretically. What does it mean for our understanding of civil wars that religion independently affects the escalation of conflicts, as the author pretends? Third and closely related to this point, he remains silent on the policy implications of his findings. While we have learned that the infusion of conflicts with religious issues significantly increases the probability of armed confrontations, we still do not know what follows from this information. Is there something that we can do to minimize the impact of religion on conflict behavior?

Overall, Religion, Civilization, and Civil War constitutes a significant addition to the literature on the impact of faith on domestic conflicts. It is clearly written, and the presentation of findings is well structured and accessible. The book should be required reading for all students with an interest in the topic, and it is a valuable resource for teaching graduate courses in conflict analysis.