2
A wide consensus has emerged in the field of international relations that
democratic regimes possess unique foreign policies. This distinction is most prominent
when democratic states interact with each other. A wide range of empirical research
confirms that democratic regimes are more likely to avoid the use of military force, and
particularly war, in their relations with each other. While the research on the presence of
democratic pacifism, or the claim that democratic states are less likely to go to war
against even autocratic regimes, is less robust, a number of authors maintain that the
pacifying power of the democratic does extend into this monadic realm (e.g. Ray 1995,
Benoit 1996, Rousseau et al 1996, Rousseau 2005).
For the most part, the trajectory of this literature has been one in which empirical
progress has spurred theoretical progress. Recently, intense theoretical development has
led researchers to move beyond an initial focus on electoral constraints and the
externalization of norms of conciliation and compromise found within democratic polities
to such causal mechanisms as the signaling capabilities provided by elections (Fearon
1994; Schultz 2001), the capacity of democratic regimes to make credible commitments
(Lipson 2004), the superior capacity of democratic regimes to fight only wars they can
win (Bueno de Mesquita et al 1999, 2003; Reiter and Stam 2002), learning on the part of
democratic regimes (Cederman 2001), and the level of regional democracy (Gleditsch
2002).
This paper argues that the connections between democracy and peace are much
less clear than literature has yet to acknowledge. Important differences in the behavior of
democratic states over time have been obscured by repeatedly testing democratic peace
hypotheses on samples aggregating data across the past two centuries. In particular, it
Convention