Horowitz, Stam, and Ellis make excellent points in response to our recent book, War and Democratic Constraint. We appreciate the careful attention to the work, and do not dispute their observations so much as speculate on how future work might best address them.

To begin, they raise the interesting case of Germany, and specifically why that country contributed forces to ISAF, but not the Iraq coalition. Since our theory is about variations in media and electoral institutions, and these things did not meaningfully vary in Germany between 2001 and 2003, it is fair to point out that our theory cannot effectively explain this variation. This points to a broader issue—one that future work should address. That is, to make the issue tractable we model public opinion as 1) naturally more likely to be pacific than aggressive, and 2) non-responsive to the particulars of a specific conflict. Our analysis therefore rests on the assumption that, at least in aggregate, there are some conflicts that the public objects to but an executive would prefer to pursue anyway. We contend that in such cases citizens are only equipped to hold their leaders back when the correct institutions are in place. To be clear, the theory doesn’t require that there are no conflicts that a public would consent to—just that there are some that they would like to avoid and these are relatively equally distributed across states.

We do this out of concern for parsimony and the realities of available cross-national data. This is not, of course, a completely satisfying response, and more should be done to consider variability in public opinion about specific conflicts over both time and space. The particular history of the German case that Horowitz et al. raise may be helpful here. While Germany did contribute to ISAF, the public debate around this deployment was much more robust than it was for some other states. Moreover, it centered quite clearly on Germany’s NATO Article 5 obligations. This suggests that while the German system is more careful and selective in its engagements, it is able to deploy force when its interests and obligations warrant it.

This leads directly into Horowitz et al.'s compelling point about the ability of constrained democracies to respond effectively and promptly to real threats, and follow through once engaged. We would take this one step further—can constrained states effectively pursue their foreign policy interests when those interests involve the use of force? While more can certainly be done to nail down the mechanism, our theory implies that the answer is “yes” because the process entails selection. Constraint forces societies to do away with foreign policy adventures that are unnecessary or ill-advised, leaving those that are important, worth fighting for, and therefore a source of credible commitment in the eyes of adversaries. One could make the case that empowered publics apply equal pressure against all foreign policy engagements, and this could explain the difference in initiation that we see in Chapter 3. But it cannot help us understand the increase in credibility that emerges in Chapter 4, or the finding that citizens in states with effective foreign policy constraint have better access to political information and are more informed.

Finally, we acknowledge that we leave out the role of individual leaders. Even in the qualitative case studies, we tend to put these factors aside to hone in on our institutional mechanisms. It is, however, reasonable to argue that first-image factors would increase the model’s explanatory power, especially given the reviewers' own convincing work on the subject. Great orators might be able to convince reticent publics, risk takers might go at it alone even when the public puts up constraints that would stop their more personally cautious counterparts. This suggests that there may be important work to be done on the conditional relationships between the institutional forces that we raise and the personal attributes that animate Horowitz, Stam, and Ellis. Moreover, their LEAD dataset opens the possibility that some of this work could be done relatively easily. We hope future scholarship will take up this call.