• Choose a category
• All
• Classic-Fiction

By Root 729 0

we have done in this study, on middle-range theory.513 Highly general theories that attempt to formulate broad covering laws tend to have quite limited explanatory and predictive power. These broad generalizations often end up as probabilistic in character, with little indication of the conditions under which they hold. They are pitched at a level of abstraction that fails to give insightful explanation of foreign policy decisions or of interactions between states that lead to specific outcomes. Middle-range theories, on the other hand, are deliberately limited in their scope; they attempt to explain different subclasses of general phenomena. Middle-range theories attempt to formulate well-specified conditional generalizations of more limited scope. These features make them more useful for policymaking.514 For example, recent work has focused on the conditions under which power-sharing, peacekeeping, or partition are most effective in resolving ethnic conflicts.515

In our view, research that aims to produce policy-applicable knowledge and theory is not at all inconsistent with efforts to develop international relations theory. Rather, efforts to develop policy-relevant knowledge are indispensable for the further development and refinement of international relations theory. A similar position has been expressed for the development of political science as a whole by a number of scholars. Peter Ordeshook, a proponent of formal theory, has called on his colleagues to take more seriously the need to make their research more policy-relevant: “Until the ‘engineering’ component of the discipline assumes a central role, research—whether theoretical, empirical, or any combination of the two—will continue to generate an incoherent accumulation of theories, lemmas, correlations, and ‘facts.’”516 Similarly, in several unpublished papers, David Dessler has made similar observations regarding the need to infuse a “pragmatic dimension” into international relations theory.517

One example of the limited policy relevance of covering-law type generalizations is structural-realist theory, the dominant theory of international relations in American political science. These limitations have become increasingly evident in recent years and have triggered lively debates. 518 While structural-realist theory is certainly necessary, it remains insufficient by itself either for explaining foreign policy decisions and out-comes or for conducting foreign policy. Indeed, Kenneth Waltz himself has emphasized that his structural-realist theory is not a theory of foreign policy. He warned against expecting his theory to “explain the particular policies of states” and regarded it as an error “to mistake a theory of international politics for a theory of foreign policy.” Waltz regards structural realism as a theory of constraints on foreign policy rather than a theory of foreign policy: “… what it does explain are the constraints that confine all states.”519 In this important (though limited) sense, structural realist theory is indeed policy-relevant.

We are left, therefore, with a large vacuum in international relations theory that must be addressed if one wishes to develop better knowledge that will help explain and inform foreign policy. We believe it would be of limited value to try to develop a very general theory of foreign policy, or “statecraft,” as historians used to call it. More useful contributions are made by focusing specifically on each of the many generic problems encountered in the conduct of foreign policy—problems such as deterrence, coercive diplomacy, crisis management, war termination, preventive diplomacy, mediation, conciliation, cooperation, and so on. Theories that focus on such generic problems are examples of middle-range theory. Examples of middle-range structural-realist theories are present in the works of Thomas Christensen, Jack Snyder, and Randall Schweller, who formulate more contingent generalizations than does Kenneth Waltz about circumstances under which states balance, bandwagon, or buck-pass. These generalizations include