The focus of this final chapter is conventional deterrence and compellence theory. The purpose is to develop an argument in favour of an alternative testing strategy in this realm using protracted crises as the main source of empirical evidence. The chapter unfolds in four stages. Stage one briefly summarizes the key impediments to testing derived from chapter 2. Stage two introduces a different approach that recommends identifying separate deterrence and compellence encounters within a single foreign policy crisis, thus expanding the pool of evidence that would be appropriate for testing a wide range of theoretical propositions derived from theory. Stage three describes, in summary form, fourteen immediate deterrence/compellence exchanges between officials of the United States, NATO, and the United Nations and Bosnian Serb leaders that took place from April 1993 to September 1995; it considers whether the prerequisites for effective use of these coercive strategies were met, and it assesses whether the behaviour in these encounters was consistent with the theoretical predictions. Stage four addresses policy implications and the overall contributions of a protracted crisis approach to testing.
As described in chapter 2, the most prominent strategy used to produce evidence to evaluate rational deterrence theory recommends identifying cases of immediate deterrence, coding these cases as instances of success or failure, isolating conditions that were present (absent) during successes and absent (present) during failures and, based on these differences, drawing conclusions about why and how deterrence works. In the empirical domain, lack of correspondence