Getting the End of the Cold War Wrong

Correspondence Getting the End of the Cold War Wrong To the editors: Ted Hopf Gaddis has done the field of international relations theory a great service by pointing out that its practitioners were unable to predict the peaceful end of the Cold War. However, I believe that Gaddis has misdiagnosed the problem, and that therefore he has suggested solutions that are unlikely to produce better predictions for theories of international relations in the future. Gaddis argues that the stock of theories available to scholars was lacking in the kinds of theories that could have permitted the prediction of peaceful Soviet imperial collapse. Moreover, the methodologies implied by these theories were obstacles to understanding the complexities of international politics and Soviet domestic developments . His recommendation is that students of international relations must expand their theoretical repertoire to encompass non-linear causal relationships and insights from all the disciplines within world politics-behavioralism, structuralism, and evolutionism . Gaddis’ recommendations should be welcomed in any case, because, if followed, they would enrich the field of international relations. But I do not believe they address the problem for which he offers them, because the failure to predict how the Cold War ended was not a product of inadequate theorizing or constraining methodologies. Indeed, the problem is far more fundamental, because its roots are sociological. American social scientists did not look at the problem and then attack it with inappropriate theories and methods; they simply failed to look at the problem at all. And the reason they did not raise the now- obvious questions is that the research programs within which they worked did not have room for such “revolutionary ” science. A kind of intellectual hegemony was exercised by American academic institutions, foundations, and governmental agencies, a hegemony which deterred the investigation of questions such as how a peaceful imperial decline might occur within the Soviet empire. If what I argue is true, then the kinds of theoretical and methodological “improvements” suggested by Gaddis, while perhaps welcome on other grounds, will not affect the prospects for better predictions in the field of The author would like to thank Michael Desch, Matthew Evangelista, and Karl Mueller for their very helpful comments. He is grateful to Kevin Clarke for both research assistance and advice. Ted Hopf is Assistant Professor of Political Science at the University of Michigan. John Lewis Gaddis is Distinguished Professor of History and Director ojthe Contemporary History Institute at Ohio University. During the 1992-93 academic year, he served as Harmsworth Professor of American History at Oxford University. His most recent book is The United States and the End of the Cold War: Implications, Reconsiderations, Provocations (Oxford University Press, 2992). International Security, Vol. 18, No. 2 (Fall 1993), pp. 202-215 01993by the President and Fellows of Harvard College and the Massachusetts Institute of Technology. 202 Correspondence I 203 international relations in the future, because it is not methodology that is at fault here, but instead, the constraints of ”normal” science. GADDIS’S IDENTIFICATION OF THE PROBLEM There are several anomalies in Gaddis’s description of the problem. For example, Gaddis correctly points out that virtually nobody predicted the peaceful demise of the Soviet empire but, having argued that particular kinds of methods were to blame, Gaddis fails to ask why those who were not behavioralists or structuralists fared no better than those operating with the objectionable methodologies. His evidence, contrary to his argument, shows that predictive failure was distributed evenly across all theoretical approachesand tools; thus relying “upon stargazers, readers of entrails, and other ‘pre-scientific’ methods” (p. 18)’might not have guaranteed a significant advantage over more scientificmethods, Gaddis’s recommendation notwithstanding. Moreover, Gaddis (p. 10) uses the predictive failure of research programs to argue that these theories have failed because they did not predict the particular outcome in which he is interested, i.e., the peaceful end of the Cold War and the dissolution of the Soviet empire. Even Gaddis’s seemingly generous criterion for testing these theories, that they need make only “probabilistic“ predictions about the end of the Cold War (p. 18), misses the point. If none of these theories implied dependent variables that included either of these two events...