Common interests or common polities? Reinterpreting the democratic peace

A long series of empirical papers has found evidence of a democratic peace. In this paper, we reexamine the data that support this consensus. Using standard sources, we find that it is only during the Cold War that violent disputes between democracies are relatively rare events. Before World War I, polity type and war are not significantly related, and less serious military disputes are marginally more likely between democracies. We investigate two explanations of these patterns. In the first, variation in patterns of common and conflicting interests between democratic states across time plays the critical role. In the second, evolution of democracies over time is the determining factor. Because we find stronger support for the first than for the second explanation, we suggest that it is common interests or their absence, not common polities, that explain dispute patterns.