a1 Department of Government, Harvard University, Cambridge, Massachusetts, E-mail: [email protected]
a2 Department of Government, Harvard University, Cambridge, Massachusetts, E-mail: [email protected]
In a key finding in the democratic peace literature, Mansfield and Snyder argue that states with weak institutions undergoing incomplete transitions to democracy are more likely to initiate an external war than other types of states. We show that the empirical data do not support this claim. We find a dearth of observations where incomplete democratizers with weak institutions participated in war. Additionally, we find that the statistical relationship between incomplete democratization and war is entirely dependent on the dismemberment of the Ottoman Empire prior to World War I. We also find that the case selection in Mansfield and Snyder rarely involved incomplete democratizers with weak institutions. We therefore conclude that the finding that incomplete democratizers with weak institutions are more likely to initiate or participate in war is not supported by the empirical data.
We thank Edward Mansfield and Jack Snyder for generously sharing their data with us and for helpful comments. We also thank Lucy Barnes, Alex Downes, Taylor Fravel, Jeff Frieden, Michael Hiscox, Michael Horowitz, Dan Hopkins, Iain Johnston, Gary King, Jason Lyall, Siddharth Mohandas, Steve Rosen, Todd Sechser, Paul Staniland, Alex Weisiger, Ian Yohai, the participants of the Harvard Security Studies Workshop, two anonymous reviewers, and the editors of International Organization, Emanuel Adler and Louis Pauly, for their comments on earlier drafts.