Commitment and Counterinsurgency: Essays on Domestic Politics and Patterns of Violence in Wars of Occupation.

Abstract

The papers of this dissertation explore two dimensions of wars of occupation. The first paper concerns the outcomes of such wars. It proposes a theory of democratic accountability that links the level of casualties suffered by an occupying force to its leadership’s decision to withdraw from the contested territory. I argue that war casualties act as the cost that drives patterns of domestic support in democratic states. Political elites seeking electoral gain will give meaning to those casualties, frequently arguing that the benefits that might come with continued fighting will not justify those costs, and that such wars are not necessary for the state’s survival. I suggest that once these divisions occur at the elite level, it becomes very difficult to maintain support among the population. This leads to the finding that while democracies may not win or lose any less than autocracies, they consistently abandon them at significantly lower levels of casualties. In the second paper, I use an experimental survey to determine if news of American casualties or elite opinion drives attitudes about the war in Afghanistan. Existing experiments in this area typically portray elite support as split along traditional partisan lines, and support or opposition among the subjects is usually driven by their partisan attachments. But this approach rarely reflects reality, where elite foreign policy and security opinions frequently cut across partisan divisions. The findings reveal that news of American casualties has little independent effect on attitudes towards the Afghanistan conflict. Instead, overall support for the conflict only declines when elite opinions in response to those deaths cut across standard partisan divisions. The final paper pivots to a discussion of the dynamics of violence in these wars. Here, I examine whether the factors that were effective in reducing attacks against counterinsurgents in Iraq had the same effect on patterns of selective and indiscriminate violence against civilians. I find that greater American troop density in a given district is associated with declines in targeted attacks on Iraqi civilians, but this has no effect on the rate of indiscriminate attacks.