David Rousseau‟s book, Democracy and War: Institutions, Norms, and the Evolution of International Conflict, provides a valuable addition to our understanding of democratic peace. His findings generate support for existing research while also presenting a more nuanced understanding of how norms and institutions may differ in their effects on states‟ propensities to engage in and raise the severity of conflict, both monadically (without respect to another state) and dyadically (when confronting another state). The book has six core questions that serve as the framework for the substantive chapters and are reviewed below. He tests these questions using both quantitative and qualitative empirical analysis. The quantitative analysis utilizes the International Crisis Behavior Data set.

Rousseau‟s first question is: are more democratic states less likely to initiate violence regardless of the regime type of their opponent, or are democracies more pacific only when dealing with other democracies? His findings support both the idea that democracies are monadically more peaceful (as they are less likely to initiate violence) as well as the existing democratic peace (examining pairs of democracies). The second question examines the impact democratic institutions may have on the escalation of nonmilitarized disputes into wars or militarized crises, which, he argues, has been neglected in scholarly research. His findings indicate that “the dyadic democratic peace exists at both the dispute and crisis stages of international conflict. However, the monadic constraint emerges only at the dispute stage.” He argues that the monadic constraint “evaporates as the opponents of democracies use or threaten the use of violence to resolve the conflict” (pp. 339-40).

Rousseau‟s third and fourth questions can be combined: has the almost exclusive focus on regime type led to a neglect of how domestic institutions can either constrain or encourage armed conflict? And are institutional structures or political norms more important in constraining democracies from using force? The evidence he presents strongly supports the institutional structure argument. Autocratic leaders who obtained power through violence were no more likely to resolve conflicts violently than those who achieved office through peaceful selection. Indeed he concludes that “no strong empirical support emerged for the normative argument across a wide variety of different tests.” (p. 340) The fifth question – does regime transformation, which is the process of becoming more or less democratic, influence decisions to resolve an ongoing dispute with force – is answered in the negative: “the analysis of disputes provides no support” for the claims that democratization or anocratization “increases the probability of military violence.” (p. 340) Finally, how does the evolution of conflict, institutions, and norms influence decisions to use force and the amount of violence in the system overall? Rousseau argues that the evolution of conflict has an important impact on the strength of domestic opposition and, therefore, the constraining power of institutions and needs to be further researched. (p. 341)

The case studies highlight some interesting points not easily discernable within the statistical analyses. Notable among these is the formation of unity cabinets by democratic polities in times of war. These “war” cabinets are much more likely to use force because the joint decision-making process makes any foreign policy failure much less likely to be exploited by the opposition which “minimizes one of the perceived risks of initiating violence.” (p. 341) Second, the case studies highlighted “a „surprising‟ number of democratic initiations against revolutionary regimes.” This suggests that the focus of previous studies of revolutionary regimes‟ propensities to engage in conflict may be misplaced if these states are often actually the targets rather than the instigators of aggression.

There is much to praise in Rousseau‟s work: the clarity of argument, the useful literature review, and the use of the ICB dataset, which provides more nuanced measures of conflict initiation and escalation. My concerns with the tests employed revolve around measures excluded. The first involves previous conflict; one would expect states that have a history of conflict to be more likely to engage in subsequent conflictual behavior. The second is the importance of trade and, to a lesser extent, foreign aid. There is a great deal of research on the effect that trade may have on conflict but this measure has also been excluded from the analyses. Indeed the importance of previous conflict, trade ties, and foreign aid were all highlighted in the case studies Rousseau used but not in his statistical analyses. The economic constraints facing leaders would seem to be particularly important when deciding to escalate a spat into something more serious or not, precisely the kinds of arguments that Rousseau is interested in.

M. Rodwan Abouharb is an Assistant Professor in the Department of Political Science at Louisiana State University.