In this chapter, we have seen that war is ex post inefficient and can—but need not—arise because of uncertainty, commitment problems, or a dispute over an indivisible objective. We have also examined the central hypotheses of two major structural theories of the causes of war: (1) neorealism and (2) power transition theory. Neorealism’s claims that bipolarity promotes stability, that uncertainty provokes instability, that states routinely mimic one another, and that a balance of power fosters stability are logically problematic and unsupported by the historical record. The alternative structural explanation presented here, the power transition theory, is consistent with the historical facts with regard to some of its core predictions but not others. Neither theory seems to provide a sufficiently robust explanation of why or when war is likely.

Scientific progress in understanding how the world works is made by building progres¬sively on the ideas that we inherit and discarding those parts that clearly fail us in exchange for superior alternatives. Much valuable debate has been stimulated by structural theories, and many useful insights have been gleaned from these approaches. However, as we see in chapter 6, strategic game theory treatments of war provide alternative accounts that explain what structural theories explain and also explain things not accounted for by structural approaches such as neorealism or power transition theory.