Practicing Coercion: Revisiting Successes and Failures Using Boolean Logic and Comparative Methods

• Published in: The Journal of conflict resolution Vol. 43; no. 6; pp. 840 - 871
• Main Author: Harvey, Frank P.
• Format: Journal Article
• Language: English
• Published: Ann Arbor, Mich Sage Publications 01.12.1999; University of Michigan, Department of Journalism
• Subjects: Behavior deterrence; Boolean data; Coercion; CONFLICT; Datasets; DETERRENCE; Diplomacy; Diplomatic relations; Empirical evidence; International politics; International relations; International relations theory; Military deterrence; Necessary conditions; Nuclear deterrence; Political theory; RATIONALITY; STRATEGY; Sufficient conditions; THEORY BUILDING OR THEORETICAL APPROACH
• ISSN: 0022-0027; 1552-8766
• DOI: 10.1177/0022002799043006008

When do coercive threats succeed or fail? Does success depend on a uniform set of necessary conditions or do core prerequisites vary depending on context? If prerequisites vary, are some combinations more likely than others to lead to success or provoke challenges? The evidence to date is far from definitive. Critics cite proof from comparative case studies that rational deterrence has failed as a strategy and theory, whereas proponents have compiled an equally impressive body of material to support the same theory. Because each side offers compelling reasons to accept its respective, often contradictory, interpretations of events in relevant crises, the debate over deterrence is not likely to be resolved solely with reference to historical records. This article reevaluates the "crucial" evidence put forward by Lebow and Stein in what has become the most widely cited critique of rational deterrence theory. Viewed through the prism of necessity and sufficiency, their data produce stronger support for the theory's core hypotheses than corresponding evidence compiled by proponents. The author develops this argument further by applying Boolean logic and set theory to data compiled by Huth and Russett and Lebow and Stein. Boolean minimization techniques are used to derive models from competing interpretations of behavior in these cases, which are then compared and evaluated in terms of the theory's predictions.