Realist and other scholars commonly hold that rationally led states can and sometimes do fight when no peaceful bargains exist that both would prefer to war. Against this view, I show that under very broad conditions there will exist negotiated settlements that genuinely rational states would mutually prefer to a risky and costly fight. Popular rationalist and realist explanations for war fail either to address or to explain adequately what would prevent leaders from locating a less costly bargain. Essentially just two mechanisms can resolve this puzzle on strictly rationalist terms. The first turns on the fact that states have both private information about capabilities and resolve and the incentive to misrepresent it. The second turns on the fact that in specific strategic contexts states may be unable credibly to commit to uphold a mutually preferable bargain. Historical examples suggest that both mechanisms are empirically plausible.

Systemic theories of international politics are inadequate for explaining particular states' policies, and some neorealists reach for supplementary foreign-policy-level concepts. Yet these studies almost never provide the empirical evidence required by their motivational constructs. Available psychological studies rely too heavily on notions peculiar to the cold war—such as the image of the enemy. A new theory proposes four additional ideal-type images. Each image is likely to lead to a specified set of strategic behaviors. An application to dyadic relations across the Persian Gulf from 1977 through 1990 suggests that this theory can help account for otherwise puzzling behavior, and it illustrates a promising route toward a more sensitive interactionist international relations theory suited both to the former superpower relationship and to diverse others.

The extraordinary success of transnational anti-apartheid activists in generating great power sanctions against South Africa offers ample evidence that norms, independent of strategic and economic considerations, are an important factor in determining states' policies. The crucial role of a strengthened global norm of racial equality in motivating U.S. anti-apartheid sanctions illustrates the limitations of conventional international relations theories, which rely primarily on structural and material interest explanations, and supports theoretically derived constructivist claims. In particular, this case suggests that analysts should examine the role of global norms in defining states' interests, rather than viewing norms solely as external constraints on state behavior.

What accounts for the development of the Arab states system from the explosive mix of Arab nationalism and sovereignty to their simultaneous existence? To understand this development, one must first examine how institutions can shape the very interests and roles of states in such a manner as to encourage the development of relatively stable expectations and shared norms; that is, regional order. This approach illuminates how inter-Arab interactions and state formation processes led to the consolidation of sovereignty and a meaning of Arab nationalism that is consistent with sovereignty. Consequently, this region highlights how sovereignty—and its lack thereof—is consequential for understanding interstate dynamics, and how different meanings of the nation have different implications for security.

A growing literature in international relations concludes that democratic states pursue distinctive foreign policies. Specifically, democracies do not engage each other in war and only rarely engage each other in serious disputes short of war. Scholars have offered three basic explanations to support these findings. Each of the three invokes a different explanatory variable: norms, checks and balances, and trade. None of the three, however, provides a convincing explanation of the peace that is said to prevail between democratic polities: the distinction between norms and interests is unclear; substitutes for checks and balances exists in nondemocracies; and trade can deter conflict only under restrictive conditions.

Foreign aid often is interpreted as an international projection of domestic income redistribution mechanisms, and many authors suggest that differences between welfare states account for variations in donor behavior. A new understanding of the welfare state can improve traditional explanations of this linkage. Existing studies of the welfare–aid relationship use two welfare state indicators: domestic spending and partisan politics. We propose a third type of indicator—the institutional attributes of the welfare state—and demonstrate its relevance. The level of foreign aid provided by a country varies with social spending, but even more so with the degree to which its welfare state embodies socialist attributes. This finding helps explain how domestic political institutions influence the evolution of international cooperation and, specifically, how welfare principles institutionalized at the domestic level shape the participation of developed countries in the international aid regime.