Abstract
The article distinguishes between deductive rationality models which explain the behaviour of large numbers of people placed in similar external circumstances, and explanations of a single action by one individual, individual rationality. The traditional justification for rationality models is that they offer good predictions. The main argument in this article is that individual rationality is testable and can serve to assess the empirical accuracy of the assumption of rationality in models. Three strategies to judge whether an individual action is rational — situational logic, motivational analysis and process-tracing — are used to analyse Henry Kissinger's decision in July 1975 to provide arms to Angola, an act which was intended to increase the value of the American reputation for toughness. The discussion is informed by Barry Nalebuff's game-theoretic analysis of rational beliefs concerning the value of reputation in deterrence. Finally, the article draws the implications of the empirical analysis of Kissinger's decision for the model of rational beliefs, concluding that such models can and should be evaluated on their assumptions rather than their predictions.
Get full access to this article
View all access options for this article.
