Long-running civil wars blur the distinction between uniformed soldiers and civilians. Armed groups rely upon civilian cooperation for protection, recruitment, intelligence, as a tax base, and, if civilians stay loyal to the cause, as a postwar voting constituency. It is a short distance from these objective observations to a more cynical kind of realism. Perhaps there can be no neutrals or innocents in war. Perhaps it is invalid to treat civilians as victims or hostages and more analytically accurate to treat them as coproducers of violence (182). Rivalry and Revenge: The Politics of Violence during Civil War invites readers to view the Spanish Civil War as a social laboratory for what happens when these brutal and zero-sum variants of social theory are taken to their logical conclusion. Laia Balcells not only unblinkingly documents the horror but also rigorously tests theories about what motivated the killing of approximately 600,000 people. I have strong convictions that this book will stand the test of time as the primary text for teaching the Spanish Civil War in the disciplines of political science, history, and sociology.

Mass murder was justified as a tactic to win the war and to suppress political competition from one’s unconvertible ideological opponents. This drive to tip the demographic balance by brutally eliminating ideological enemies, often far behind the lines of contestation, can be statistically measured due to the unusually well-documented records of killing. The analytic differentiation of direct versus indirect killings and discriminate versus indiscriminate killings is critical. Direct killings were most likely to occur in areas where they would provide the most marginal benefit to the side producing the killing, where “base” civilian supporters of the two sides are evenly matched. The book really shines in Balcells’s nuanced treatment of reciprocal direct killing as a result of the psychosocial experience of wartime violence on combatants. Atrocity begets more atrocity. It could be possible to explain empirical patterns predicted by a sequence of shifting control patterns without reference to emotions, and their invocation is sure to alienate a stratum of political scientists and economists. In my view, however, placing the book in an academic lineage that runs through deviants like Roger Petersen, with a more nuanced view of human behavior than rational choice approaches allow, ought to increase its readership and long-term impact. Focus on the emotions that enable killing has an effect that is at once memorably humanizing, amenable to econometric testing, and utterly compelling.

Balcells presents the Spanish Civil War as an exemplar of conventional civil war, where an incumbent state with high capacity faces off against rebels in a situation of military parity in armaments and organizational capacity (for example, when a country’s military splits after a coup). Conventional civil wars are defined by maneuvers and confrontations between regular armies that face off across clear front lines. If a high-capacity state confronts insurgents unable to compete symmetrically, by contrast, an asymmetric irregular civil war is observed. Irregular wars are defined by guerrilla tactics, terrorism, and ambiguous territorial control. The hearts and minds of civilians living in contested zones are understood to be the resource being fought over, and anticipations of shifts in territorial control creates incentives to kill civilians. Over the last two decades, as political scientists have assisted policy missions in helping to collect intelligence from civilians (sometimes trying to make the counterinsurgency trains run on time) these have been the background scope conditions.  But not all civil wars are like this. In conventional wars, where the technology of warfighting approximates interstate war, civilians have the option of decamping for one side or the other of the line of control. Civilians are allowed to escape.

As an avid consumer of Balcells’s earlier scholarship, I came to Rivalry and Revenge convinced that the analytic distinction between the two types of wars is vitally important. I am less convinced than I was before that the Spanish Civil War fits the theory. The vast scale of the killing, encompassing every kind of cruelty, runs counter to simplification. For a purer example of conventional civil war, consider contemporary Ukraine. The fighting takes place between kontractniki, fighting with artillery across predictable front lines. Civilians can remove themselves as targets by exiting the conflict zones. About 14,000 people have died—around 2 percent of the deaths in the Spanish Civil War. Contra the litany of cruelty documented in Rivalry and Revenge, there has not been a case of a pregnant woman having the baby cut out her stomach, no mass rape, no mutilation of dead soldiers’ bodies, and not much torture. Celebration is premature, since the conflict could irregularize at any time, and Balcells’s research program provides many reasons to expect civilian victimization would be much worse in a hybridized conventional-irregular war.

What explains the huge gap in conflict intensity between Ukraine in this decade and Spain in the 1930s? Balcells’s text provides a few clues. One explanation is that the overall international environment in Europe today is more benign than it was in the 1930s. Russian speakers making pilgrimages to the front lines in Ukraine are performing their politics in a way that previous international brigades might have recognized. They are not actually threatened, however, by either Putinism or NATO in a way analogous to the existential threats posed by either Stalinism or Franco’s brand of fascism. Another change is that today’s civil war actors may have better tools at their disposal to identity, and then selectively discipline or disappear, focal points in the enemy’s ideological production network. Killing all the priests or all the activists is not necessary if one can selectively intervene to tag, track, and silence the most dangerous and prolific. In this vein, I hunger to know more about patterns of civilian victimization behind the lines in the Donbas, and more about the “shadow war” taking place behind the front lines in places like Kyïv. Ubiquitous social media penetration allows for real-time fact-checking of false news stories of atrocity, cauterizing spirals of defensive arming and allowing for faster ideological sorting. The suppression of tit-for-tat atrocity, if the Balcells account is correct, has likely had the secondary effect of arresting cycles of revenge. When Ukrainian and Russian researchers eventually gain access to state archives and can reconstruct Ukraine’s civil war with computer-assisted techniques, these speculations can be tested.

Balcells pairs primary sources with responsibly curated historiography from the secondary case literature, then organizes the quantitative data in a manner suitable for transparent empirical tests. Rivalry and Revenge should be widely assigned in graduate seminars as an example of what political science dissertations should aspire to if they have any desire to be taken seriously by historians. Moreover, students in the North American academy—and those in the Spanish, Basque, and Catalan academies—should demand Balcells’s replication data files and convince themselves that the hundred-page empirical core of the book (49–148) can survive empirical quibbles. (Does post-coup territorial control correlate with pre-coup ideology (60) or is it more accurate to say control is distributed unpredictably and randomly (62–63)?) Calling for rigorous replication and data exploration of these rich data is should not meant as an indictment of Balcells’s labor, but a celebration of it. Our guild would be strengthened if we could get more students to engage further with serious books on serious topics like this one. Our guild would be a better place with more scholars like Balcells in it.