I am grateful to Ioana Emy Matesan for her thoughtful, probing, and generous review of Ordering Violence. She has identified two important issues in my project: how ideology and tactical imperatives relate to one another, and how to think about armed groups’ own agency.

First, Matesan asks for clarity on how ideological variables intersect with more fine-grained tactical imperatives. She is right that the specifics of this relationship vary across cases. This is an issue I wrestled with throughout the project. Ultimately, I put my theoretical emphasis on ideational processes of perceived threat and affinity, with tactical concerns distinctly secondary. In part, this reflects my judgment that so-called strategic or tactical explanations in the literature are quite saturated already, and so meaningful progress at this point requires a shift toward a different set of political questions.

However, there is no doubt in the empirical record that the actual relationship between ideas and local military/political incentives can be far more complex and variable. Matesan is correct that sequencing is often key, which aligns with the theoretical structure of the book: big-picture ideological politics set the general context, and then more immediate tactical incentives can “fill in” variation that is otherwise indeterminate.

That said, matters are not always so clear-cut. There are certainly cases in which tactical considerations are more important than my argument expects. That is fine, because disconfirming evidence reduces concerns about tautology and operationalization. I point to several such examples in the book.

Matesan’s concern is most far-reaching when ideology and tactics collapse into one another, with the worldviews of regimes or armed groups seeming to completely determine all their behavior. In retrospect, I wish I had been able to conceptualize more carefully the trade-off between what Matesan in her book calls “principle” and “pragmatism.” The reality is of a spectrum or distribution, rather than any single fixed prioritization. Measuring where states and groups lie on this spectrum ex ante is obviously enormously difficult, but the book would certainly have benefited from a more extended discussion of this kind of variation.

Second, I agree with Matesan that Ordering Violence focuses more on states than on armed groups, and that I tend to see more stability in governments’ than armed groups’ goals. Her own book does a much better job of explaining shifts over time in group goals than my state-centric account: The Violence Pendulum is where I would point those interested in this question.

In general, however, I view governments’ power advantages as putting greater pressure on nonstate actors to adjust their aims than vice versa. There are important exceptions to this generalization, to be sure, but it certainly applies to the bulk of empirical cases in my book. Although some armed groups do maintain an unyielding set of ideological commitments, many others must adapt or risk being destroyed or marginalized. The constraints on armed groups tend to bind more tightly, and thus limit their options, far more than those on governments.