To Attack or Not to Accack: A Theory and Empirical Assessment of Extended Immediate Deterrence

Abstract

Specific theoretical conditions for the success or failure of extended immediate deterrence (EID) are identified based on assumptions about rational action. The resultant decision theoretic model yields four hypotheses which are tested against cases of EID between 1885 and 1970. The hypotheses are strongly supported, with about 77% of outcomes correctly predicted. In addition, the article distinguishes between two qualitatively different situations in which the model leads us to expect a defender`s policy of extended immediate deterrence to succeed or fail. The results are contrasted with Huth`s 1988 analysis. The formally derived hypotheses of this analysis fit the data as well as Huth`s probit model. The percentage of cortect predictions in the model here is slightly, but not statistically significantly, lower than that in Huth`s, and the model presented here is more parsimonious than Huth`s. Thus the model here seems to be a preferable construct of decision making during extended immediate detertence situations.