Political Democracy and Militarized Conflict in Evolving Interstate Rivalries

In this paper I study the relationship between political democracy, interstate rivalry, and militarized interstate conflict. Past research has shown dyadic democracy to have a strong pacifying effect on relations between states; democracies have rarely if ever gone to war against each other. Much of the research in this area has been static and cross-sectional in nature, though, overlooking the history of relations between the states in question. Recent research on interstate rivalry has suggested that relations between adversaries tend to become increasingly hostile and conflictual as the adversaries accumulate a longer history of past conflict and rivalry. Particularly important for the present study, two adversaries are more likely to become involved in further episodes of future conflict in the future when they have accumulated a lengthier history of past conflict. This evolutionary approach to interstate conflict and rivalry suggests that the pacifying effects if dyadic democracy may decrease over time in ongoing interstate relationships, as the adversaries become involved in more confrontations and accumulate hostility and grievances toward each other from these multiple confrontations. The paper begins by laying out the general theoretical framework of an evolutionary approach to interstate conflict and rivalry, and by formulating a series of hypotheses on how an evolving interstate relationship should affect the relationship between democracy and conflict. These hypotheses are then tested statistically, using logistic regression analysis. The analyses suggest that dyadic democracy has not been entirely effective at preventing the recurrence of conflict between former adversaries, particularly for contiguous adversaries. These results suggest the need for further research on the introduction of democracy into an ongoing conflictual relationship, and for continued caution in applying the democratic peace proposition to policymaking.