I very much appreciate the review from Keren Yarhi-Milo and the opportunity to conduct this dialogue. Her insights are important and helpful, and I wish I had access to them before publication. If only this kind of sustained interaction occurred more often.

As to her critiques, I am afraid I am guilty as charged. I attempted no systematic analyses of the processes under consideration and tested no hypotheses. I am grateful that international relations (IR) remains a methodologically heterogenous field, because Psychology of a Superpower marshals no scientific support for its assertions. I never considered rigorous testing, because for these subjects I simply did not think any was possible, at least in convincing ways. How can one assemble data regarding unipolar psychology when we have only had one instance of it? It seems that one of the skills valued by today’s scholars is to find creative ways to quantify the unquantifiable in order to prove the unprovable. I suppose I could have tried to construct a dataset based on some sort of textual analysis of speeches or newspaper articles, or counted UN votes or militarized disputes, but I would have found it all terribly artificial and unconvincing. This is not to say that other scholars do not do such analysis very well, only that my subject did not lend itself to such methods. This book is saddled with my particular prejudices, I’m afraid, which are generally skeptical of the self-consciously scientific approach to security studies.

I often find case-study analysis to be even more problematic. Too often scholars develop theories based on their extensive, lifelong study of particular cases and then test the theory on those same cases. Though sold as tests of theory, case studies in IR are often merely extended defenses of the author’s argument. For these and other reasons, when reading IR books, I generally follow what I suspect is the standard pattern: I concentrate on the theory chapters up front and skim through case studies, reading in earnest again at the conclusion. All those hours of labor, all those interviews, archival research, and fieldwork go unappreciated by skeptics like me.

I wanted this book to be different, but that does not make it better. Professor Yarhi-Milo may well be right that I should have described in greater depth the psychological experiments and their results that provide so much of the support for the book’s points. I do describe a few here and there, but for the most part I merely report their findings. Earlier drafts of the book described more, but as I revised the manuscript those sections tended to bore me, so I assumed they would put all but the most determined readers to sleep. I suppose that adding more detail about the experiments, however, may well have strengthened the argument. If only Perspectives on Politics had commissioned this dialogue a bit earlier in the writing process, I could have incorporated this and her other thoughts into the book. As it is, we have to stick with the book in its present flawed form.