| My thesis focuses on the problem of civil wars, which are marked by high levels of violence and an extreme inability to bargain for peace. The literature on bargaining failure in civil wars has identified several reasons to explain this situation: asymmetric information, problem of credible commitment, and issue indivisibility. These studies, however, neglect to consider the legal/institutional environment where bargaining occurs. In contrast, I argue that this is a crucial factor in determining whether or not bargaining might be successful. In my thesis, I start off by formalizing the situation of conflict: this consists of two interdependent games, a War Game and a Bargaining Game, whose specifications incorporates the distinguishing features of civil wars: the nature of the actors involved, the way they fight (i.e. fighting without committing crimes and fighting by committing crimes), the type of government (democratic, non-democratic), etc. The key observation is that the full specification of these games (i.e., action available, payoffs, etc) depends on the legal regime in place. The study of how the War/Bargaining game varies with the legal regime in place allows one to compare the different regimes with respect to their ability of achieving the goals of peace and/or justice. I, then, apply these ideas to compare the relative performance of international criminal tribunals designed according to the principles of state sovereignty, human/cosmopolitan rights, and domestic tort litigation. A novel result is that the careful choice of the legal regime might substantially reduce the problems associated with the presence of asymmetric information in civil wars. I give an example of a situation where the domestic tort litigation model outperforms the other legal models, thus lending support to a thesis proposed by Anthony D'Amato (1994) during the civil war in the former Yugoslavia. |
