Case Studies and Theory Development in the Social Sciences - Alexander L. George [167]
Trust is usually regarded as being a necessary condition, though not a sufficient one, for states to cooperate.649 However, trust should not be viewed as a dichotomous attribute with complete trust being a necessary condition; rather, the amount of trust required for an agreement varies greatly.650 For example, states “must [also] have a shared interest in controlling their competition, adequate domestic support, and the ability to verify an agreement.”651
A “missed opportunity” for an agreement is defined as “a situation in which there was at least one alternative that parties to a conflict preferred or would have preferred to nonagreement.”652 To make the case that a missed opportunity existed “entails showing that both sides wanted an agreement, that history need not be completely rewritten to end up with a different outcome—in other words, that a plausible sequence of events could have led to an agreement.”653 This analytical standard is employed to guide the study of a variety of available data.
Larson examines five periods in which there was a major policy shift by one or both of the superpowers—a change she regards as being of critical importance for creating the possibility of a significant cooperative agreement. These were periods that had the potential for being “branching points” at which U.S.-Soviet relations could have taken or did take a different path. Although Larson compares cases of successful and ineffective cooperation, she makes it clear that they should not be viewed as independent of each other. Each leader’s efforts to improve relations drew on earlier experience. 654
Citing David Collier’s statement that “causal inferences about the impact of discrete events can be risky if one does not have an extended time series of observations,” Larson engages in extensive process-tracing of developments in each period.655 “Process-tracing,” she maintains, “is essential for uncovering the causal mechanism—in this case, cognitive processes [by the actors] of interpretation and inference.”656
Her case studies lead to an important finding: “Where the superpowers successfully reached cooperative agreements—the Limited Test Ban Treaty, the first Strategic Arms Limitations Treaty (SALT I), the Intermediate-Range Nuclear Forces (INF) Treaty—one side demonstrated its good intentions through several conciliatory actions, and it is difficult to envision how a cooperative outcome could have been achieved otherwise.”657 Here and elsewhere she documents the role of trust-building measures.
In cases of missed opportunities, she notes, “one must study non-events—things that did not happen… . To explain the causes of non-events, the analyst will have to vary initial conditions mentally … one should identify the critical turning points and consider whether alternative actions might have made a difference.”112
JOHN M. OWEN IV, LIBERAL PEACE, LIBERAL WAR: AMERICAN POLITICS AND INTERNATIONAL SECURITY. ITHACA: CORNELL UNIVERSITY PRESS, 1997.
The author cites a suggestion made by Joseph Nye that the democratic peace thesis needs “exploration via detailed case studies to look at what actually happened in particular instances.”658 This is, indeed, what Owen does in this study. He examines ten war-threatening crises involving the United States between the 1790s and the close of the nineteenth century “to see precisely what keeps liberal states at peace with one another and what leads them to war with illiberal states.”659
The need for small-n studies, Owen maintains, “stems from the requirements of establishing causality.” Large-N quantitative methods can establish correlations, and in such studies one can control for other variables “to see whether other possible causes can wholly