ABSTRACT
In recent years, political scientists have moved away from correlational analyses that try to maximize the amount of variance that is accounted for by a model and toward more reliance on identifying those factors that increase the probability of war. Such an effort has taken place primarily because of methodological considerations, but it also exhibits a sense that the search for correlates of war has been somewhat more intractable than had been anticipated. To map the probability of war, one should start with those factors that are presumed to be most fundamental. The theoretical explanation presented in The War Puzzle hypothesizes that territorial disputes are an underlying (as opposed to proximate) cause of war. If this explanation is true, then it should be expected that the presence of territorial disputes will increase the probability of war, and territorial disputes will have a higher probability of going to war than expected by chance.
