War and the fate of regimes: a comparative analysis

 Adequate? The authors put forward three hypotheses: Hypothesis 1. The chances of a domestically instigated violent regime change increase with the defeat in war. Hypothesis 2. Violent regime change is least likely for winning initiators; the likelihood of violent regime overthrow increases for winning targets, losing targets, and losing initiators in that order. Hypothesis 3. The chances of a violent regime change increase with the costs of war, irrespective of the nation’s war outcome or initial conditions. A regime is defined as the members of the political elite who make state decisions.The word regime is therefore a general word to mean any government. War refers to any confrontation between two governments where there is sufficient force and resistance to cause at least one thousand battle deaths. A regime change is defined as when the leader and his or her entire regime is removed from office by violent means. Violent means intends to cover revolutions, coup d’i?? tat, and the like but not removal by the opponent state.Time 2: (contracted): The period from one year before the end of the war until one year after its end. With Time 1 being taken as the germane and Time 2 as a check. The idea is to allow a long enough time after the war to allow related factors to fully develop, be recognised by the citizens and for them to form and act.The constrained time allows to exclude an uprising which may not be directly linked to the war. It may appear that just looking at regime changes is not enough and it may have been better to include large scale failed rebellions as well.This may lead to a better analysis of the chances of losing power, since failure of the rebellion still shows unrest at the regimes policy. This figure could then be broken down to the amount of successful regime changes. If as the authors suggest, regime change is an important aspect of foreign policy making, it would appear that this more complete figure could be more useful. The definitions of the main terms are on the whole exhaustive and mutually exclusive. However, the term ‘regime’ as defined by the authors is quite different from the general use of the word and as such could be easily misconstrued.The time periods described are always going to be open to question, it appears that the only way to remove this problem would be to analyse the cause of each regime change individually. This, it would seem is impractical, a great deal of resources would be needed plus there is not much doubt when saying that very few regime changes would have been documented well enough to warrant such analysis. e) What data were used? Were they appropriate? The data for this review is relatively simple. The war data is taken from the “Correlates of War” project reported in Small and Singer’s (1982) “Resort to Arms”.Beverly Hills: Sage. This sets out reports on; national involvement in all international wars between 1816 and 1975, the initiators, winners and losers. The data on violent regime changes is from two main sources, Banks’s (1971) “Cross-Polity Time-Series Data”. Cambridge: Massachusetts Institute of Technology Press. Secondly, Langer’s (1952) “Encyclopedia of World history”. Boston: Houghton Mifflin. This information was then compared to Spuler’s (1977) “Rulers and Governments of the World”. London: Bowler. Post-1965 was taken from “Facts on File”. Plus Bienen and Van deWalle’s (1991) “Of Time and Power”.Stanford: Stanford University Press. f) What analytical techniques were used? Were they clearly described? The first analytical technique used was simple mean average of war participating nations who experienced regime change. This was then compared to three random sample nation-years’ means. It showed that a violent regime change is twice as likely to occur in the extended time span after a war. For the first hypothesis, A cross tabulation showed that it is correct. For the second and third hypotheses, probit analysis is used for the ordinal data