Win, lose or draw: Domestic politics and war.

Abstract

This dissertation presents a quantitative test of a rational choice model of war outcomes. By explicitly demonstrating the links between the domestic politics of nation states and war outcomes, the dissertation investigates a serious flaw in the realist paradigm, namely the assertion that domestic politics are or should be largely irrelevant to the study of international relations. Starting from the rational choice perspective, the basic notion behind the model is that the two sides in a war are simultaneously trying to coerce each other. Each state is assumed to follow a cost-benefit or crude expected utility analysis process which is updated throughout the war. Four possible outcomes result from the combination of the two sides' preferences for fighting or quitting: Win, Draw, Stalemate, Lose. States are assumed to be more willing to continue fighting when faced with an expected net gain. States expecting losses will be more likely to quit. An econometric test of the theory using multinomial logit regression techniques follows the model development. Many of the variables shown to affect each sides' cost benefit analysis are effected by the decision makers' domestic political choices. Factors such as democratization, political repression, time and military strategy are covered in detail. Coverage of the effects of military capabilities, industrial capacity, population, surprise, and alliance contributions is also included in the empirical section. The quantitative findings are buttressed by a series of historical illustrations showing how domestic political factors such as ethnic, racial, and class based biases limit states' ability to project power. The illustrations are used to frame a concluding discussion of the merits of the rational choice paradigm versus the realist paradigm. The dissertation's conclusion and underlying theme is that as a normative and positivist framework, realism should no longer be viewed as the most useful paradigm within the international relations and security studies literature. Rather, the alternative approach, rational choice, should be viewed as more productive for future research in international relations and security studies.