During the last four decades, Colombia experienced the displacement of more than seven million of its people. Although the country is probably the worst performer in the world in this regard, it is certainly not alone: displacement is a global phenomenon. Abbey Steele’s book is about displacement in the midst of the Colombian civil war. She reads it as a form of “collective targeting” by armed groups. She asserts that such targeting takes place under two circumstances: when an armed group tries to take over a territory and when it has enough information about civilians’ political preferences to tell friends from foes. A major implication of Steele’s argument is that under certain circumstances violence and democracy can be inextricably linked: armed groups learn whom they should displace and where from through elections. Relatively free and genuine elections, it must be added; otherwise they would not be an adequate identification mechanism.

Steele’s book is a very strong contribution for many reasons. First, it places politics back into the analysis of civil wars. Wars are and have always been about many things. Politics is one of those things and a particularly important one, because, as Steele shows, it is the only domain that can provide the ways to translate the discourse of national “master narratives” of conflict into the discourse of regional and local grievances and vice versa. It is not, as Steele shows, that the master narratives of civil wars are any less “real” in any tangible sense than territorially grounded micro-conflicts; it is that both get linked through politics and political identities. As Steele carefully argues, loyalties can be very stable even in the presence of extreme violence. Indeed, it is difficult to understand displacement in Colombia without taking those loyalties into account. Precisely because of loyalty’s stickiness, armed groups use violence to put disloyal civilians and groups in their place. Note that including politics in the analysis of civil wars necessarily disrupts standard atomistic frameworks, where the only units of analysis to understand the logic of the victims are “individuals” and “communities” (defined as of individuals). Communities, at all territorial levels, are fractured along several fault lines. Politics provides the appropriate idioms to identify and negotiate these cleavages.

Second, Steele demonstrates a clear understanding that each form of victimization of civilians may have its own dynamics. She explains why homicidal violence is not always the proper tool for obliterating deeply grounded political identities: it may simply be too costly. The very nature of the political regime that provided the tool for the identification of civilian preferences—elections—could under certain circumstances increase prohibitively the costs of homicidal violence when and if they exceeded a threshold (James Ron, “Savage Restraint: Israel, Palestine and the Dialectics of Legal Repression,” Social Forces 4(1), 2000) while still allowing for massive victimization of the population. Here Steele’s concept of “collective targeting” reveals its usefulness. The Colombian paramilitaries—given the region that the author chose, Urabá, she is speaking mainly about counterinsurgents—did not have the same kind of fine-grained evidence that the author uses in her book: in particular, they could not pin down the preferences of the individuals who lived in the territory they were taking over. All they knew was that certain neighborhoods leaned toward the left and thus deserved to be punished—and then they chose the type of victimization that best matched their degree of information and their cost-benefit sheet. Because both the levels of information and the cost-benefit sheet are provided in good measure by the political regime, it is necessary to understand that regime and the way in which it works if we are to correctly interpret the victimization of civilians during civil wars. I deem this corollary very, very important.

In addition, Steele’s mixed-methods research design is simple and powerful. She discusses the country in general, but then zooms into Urabá, one of the most violent regions during the Colombian civil war. Therefore she has to demonstrate that in effect displacement affected people to the left more frequently than other citizens (in Urabá) and that elections played a fundamental identification role in displacement (in the country). Her methodological procedure is the following. First, she claims that displacement became a large-scale phenomenon only after local elections started to be held in Colombia (1988) and the political party born from the 1984 government–FARC agreements, the Patriotic Union (UP), participated in them. Second, she shows through historical and electoral analysis that the paramilitary effort to evict the FARC from Urabá and take control of the region triggered collective targeting and eventually a large wave of displacement. Third, she “zooms out” to test quantitatively if the levels of electoral support for the UP are associated with displacement. The R-squares she gets are quite impressive (pp. 190–96) and add strongly to the plausibility of her claims. They also allow her to further clarify conceptual issues discussed at the beginning of her book and to show that displacement is not only the result of a “humanitarian crisis” (caused, for example, by an increased risk of victimization, the disruption of the productive apparatus, etc.). For example, Steele’s regression demonstrates that displacement shows no clear association to homicide.

Steele is also open and candid about the scope conditions of her propositions. She does not engage in an exercise of spurious generalization: the argument is tailored to the Colombian war, although it can and does feed important comparative discussions. This focus necessarily leaves some interesting issues open. A first example regards the kind of group she is analyzing. The guerrillas also promoted displacement, not only because of what can be called “mechanical effects” (the humanitarian crisis spun by war), but also because they targeted specific sectors of the population, including political adversaries. Can we find something analogous for the FARC and the ELN—the main Colombian guerrillas—to what Steele found for the paramilitaries? If yes, what? If not, why?

A second example is the contrast between ethnic and political identities. Steele presents, and rightly so, the stability of political identities as one of her most important results. In contrast, some take the stability of ethnic identities for granted. I wonder if it would not be useful to pursue this lead a bit further. Ethnicity indeed has been reified in the midst of civil wars not only in the public space but also in academia (see Christopher Cramer and Paul Richards, “Violence and War in Agrarian Perspective,” Journal of Agrarian Change, 11(3), 2011); yet in many senses it is also a political resource and/or construct. For example, Scott Straus in Making and Unmaking of Nations: War, Leadership and Genocide in Modern Africa (2015) agrees with Steele that identification is a key aspect of civilian victimization—this time in the midst of ethnic cleansing—but at the same time finds that a politically created sense of danger is the main trigger of the genocidal dynamics he is studying. However, conflicts like the Colombian civil war have many distinctly ethnic dimensions. This is especially true for the type of crime—displacement—that Steele is investigating. I do not want to push the argument too far. Indeed, ethnicity is not or rarely is a political invention; indeed, it played a less salient role in Colombia than elsewhere. But to establish the separating line between “political” and “ethnic” wars (or identities) is not easy. How much does the stickiness of ethnic identities in wars in which the main identities are ethnic owe to politics?

Its capacity to rethink the politics of civil wars and to pose from that vantage point important comparative questions makes Steele’s book a powerful and relevant contribution for Colombian (Latin American) specialists and war scholars alike. The book’s most important and striking finding is the link between democracy and displacement. I found Steele’s analysis of the region of Urabá utterly convincing. On the other hand, even though the regression part—that is, the part that generalizes the result to the whole country—contributes to the plausibility of her proposition, it also raises some questions. Steele finds that the UP vote share is strongly associated with displacement during a relative short period in the 1990s. But the UP was an extremely small party, never capturing more than 5% of the vote nationally. Of course, as Leah Carroll (Violent Democratization: Social Movements, Elites, and Politics in Colombia’s Rural War Zones, 1984–2008, 2011) has shown, this low vote share did not make it irrelevant, especially in the face of its will and sometimes capacity to disrupt the local status quo. However, in many regions the mechanisms that triggered displacement apparently had nothing or little to do with the presence of the UP.

For example, in the north of the country a substantial portion of displacement was strongly associated with land dispossession (Jacobo Grajales, Gobernar en medio de la violencia. Estado y paramilitarismo en Colombia, 2017). Steele’s main independent variable is an average of UP vote share across several years, but if she had set up a panel, would she have gotten the same results? Being the target of murderous violence, the UP got smaller and smaller and eventually disappeared. In the period of study chosen by Steele, displacement was growing while the UP—which from the outset was small—was shrinking. There is something that does not match here, because the combination of both trends implies that a good deal of action that did not involve UP presence was taking place. Perhaps Steele’s regression did not include important variables. My first intuition is that, although displacement was not associated with homicides, as Steele shows, it might have been associated with massacres. Yet even though some emblematic massacres, especially in the 1980s, were directed against the UP in the 1990s, likely the majority had little or nothing to do with the UP presence (although they could certainly be highly political). In sum, it is possible that the strong link found by Steele between the UP vote and displacement is related to a third factor. One candidate would be the paramilitary expansionist drive, which took the paramilitaries to ever new territories that traditionally had not had their presence (and that thus were likely to be targeted as potentially dangerous).

Additionally, Steele finds that in Urabá two rural communities resisted paramilitary pressure particularly well. However, displacement nationally was mainly a rural phenomenon. This second mismatch suggests that the translation from the Urabá displacement experience to the national experience may demand the identification of additional mechanisms.

But these mechanisms are likely to involve politics and the political regime in several ways, precisely because of the reasons put forth by Abbey Steele. Her book puts politics back into the equation and provides a thoughtful, clear, and highly innovative reflection on the relationship between democracy and violence against civilians during civil wars.