I appreciate Christian Davenport’s thoughtful comments on my book and what he identifies as the strengths of the volume. In my response I focus on what he sees as five “limitations/problems” with the analysis. All but the first pertain to one chapter (ch, 4) of the book.

His first point concerns my finding that there is extensive variation regarding the conceptualization and measurement of peace in both the scholarly literature and the assumptions underpinning practice. The problem, in his view, is that “the reader is not provided with much insight regarding what they should do with all of this variation.” My purpose, however, was to demonstrate that among scholars and practitioners there is neither a consensus with respect to the characteristics of peace nor, as a result, on how to measure peace. My purpose was not to navigate through this variation but rather to offer what I consider to be principles that should inform a sound navigation, which I do in chapter 5.

Davenport’s second point concerns what he calls my “pivot” in chapter 4 to a statistical evaluation of peace duration, given what I argue elsewhere in the book about the limitations of a narrow conception of peace as the “mere” cessation of violence. “Why do this?” Davenport asks. For the simple reason that the nonrecurrence of violence is the sine qua non of peace, and it would seem to be important therefore to investigate the factors that contribute to peace stabilization, as I do (with Anke Hoeffler) in this chapter. The two perspectives are only incompatible if one reduces the meaning of peace to the nonrecurrence of violence, which I argue against.

Davenport also takes issue with the observation in this chapter that military victories result in a more durable peace than other conflict outcomes. This is a well-known empirical observation and is not meant to reflect any normative judgment about the quality of the peace that may emerge from military victories. Whether the quality of the peace that emerges “out of the barrel of a gun” is vitiated by its violent origins is an interesting question, but it is not a question that we explore (or need to explore) for purposes of our analysis in this particular chapter.

Davenport’s fourth point argues that we have not taken additional factors—specifically, “technologies of war” and “spatial diffusion”—into consideration when explaining peace stabilization. However, the literature that examines these factors (notably Stathis N. Kalyvas and Laia Balcells, “International System and Technologies of Rebellion: How the End of the Cold War Shaped Internal Conflict,” American Political Science Review, 104(3), 2010; Kyle Beardsley and Kristian Skrede Gleditsch, “Peacekeeping as Conflict Containment,” International Studies Review 17(1), 2015; Lise Morjé Howard and Alexandra Stark, “How Civil Wars End: The International System, Norms, and the Role of External Actors,” International Security 42(3), 2018) does so in relation to the onset, diffusion, and decline of violence, not the stabilization of peace. (I do not understand the point about “geographically contained entities”: the unit of the analysis is conflict, of which there may be several within one geographic entity.) Davenport is right, however, to point out the absence of robustness checks with regard to the statistical analysis in chapter 4. These were included in the report from which this chapter was drawn (Richard Caplan, Anke Hoeffler, and Henk-Jan Brinkman, An Analysis of Post-Conflict Stabilization: A Report for the Department of International Development of the United Kingdom, September 2015) but were omitted from the book in deference to the editor, as Davenport suggests they might have been.

Davenport’s final point concerns the absence in the book of any stated criteria for the selection of the cases for the qualitative analysis in chapter 4 that complements the statistical analysis, as well as the short summary manner in which the cases are presented. This is a fair criticism. The only defense I will offer is that the exercise was meant to be illustrative of an approach that combines large-N quantitative analysis with case study analysis in an iterative fashion, such that the quantitative analysis informs the case study analysis, and vice versa. Such an approach was possible because the case studies were specially produced for this project by expert analysts in tandem with the quantitative analysis. However, I agree that more detail regarding case selection could have been provided. I also accept that narratives, however detailed, may be contestable, which can lead to divergent judgments about the dynamics contributing to peace (or not), as well as the quality of peace, a point I make in chapter 5 in relation to differing assessments of postwar Eastern Slavonia.