To leverage the full range of observed variation in patterns of violence toward the development and testing of theories of political violence, scholars need adequate conceptual foundations: what should we mean by a pattern of violence on the part of an armed organization? Scholars often distinguish degrees or levels or types of violence across organizations and conflicts, but definitions and measures vary sharply. We argue that patterns of violence are not reducible in ways often assumed in the literature: lethal violence is not a good proxy for the overall pattern, and differences in patterns are not well captured in the binary “terror” versus “restraint.” To address these concerns, we provide a new conceptualization of political violence, defining an organization’s pattern of violence as the configuration of repertoire, targeting, frequency, and technique in which it regularly engages. This approach adds precision to the documentation and analysis of political violence, clarifies the evaluation of rival theories, and opens up new research questions. We demonstrate its utility through an analysis of violence against civilians in Colombia, drawing on an original database of massacres, judicial proceedings, and other sources, and show that the concept of “pattern” helps bring ideology and politics back into the analysis of organized violence.

The effort to build a political economy of war without politics is finding its limits. The question now is what comes next. How to put politics back in? This article compares systematically two non-state armed groups that participate in the Colombian conflict, the main guerrilla (FARC) and the paramilitary. It shows that despite their similar financial bases, they appear to exhibit systematic differences— regarding both their social composition and their internal/external behavior—and claims that the key to understanding them is the set of organizational devices that each group crafts in its process of survival and growth. All this suggests that a main tenet of the early political economy of war, that all non-state armed groups can be understood as being strategically identical, is flawed. It also poses a classificatory challenge.

Direct elite participation in civil wars remains unexplored terrain. It should be analytically telling, because it involves taking major risks and costs. Here, we consider the direct participation of one rural elite—big cattle ranchers—in the Colombian paramilitary saga. We claim that it was massive, locally specific, regulated by institutions, and riddled by permanent collective action issues. We focus on two important forms of direct participation: ranchers as leaders of paramilitary groups, and ranchers as promoters and beneficiaries of coercive land dispossession. This does not cover the full spectrum of potential elite participation in war, but it is a key starting point to sort out the ways in which extreme inequality is associated with political violence from above.

How important is ideology for the analysis of civil war? In contrast to literature that neglects ideology in its emphasis on structural variables or situational incentives, this article argues for the recognition of its essential role in the functioning of armed groups if they are to explain observed variation in armed group behavior. For example, sidelining ideology leaves major phenomena unexplained, including both mass killing and restraint in violence against civilians. Ideology is defined as a set of more or less systematic ideas that identify a constituency, the objectives pursued on behalf of that group, and a program of action (perhaps only vaguely defined). Ideology matters in two ways. First, it has instrumental value for armed groups, socializing combatants with heterogeneous motivations into a coherent group, dampening principal-agent problems, prioritizing competing goals, and coordinating external actors including civilians. Ideologies differ in the kind of institutions and strategies they prescribe for meeting these challenges and in the extent to which they do so. Yet this first approach is incomplete, as ideology has more than instrumental value. Members of some armed groups act on normative commitments in ways not reducible to instrumental reasoning, and some groups constrain their strategic choices for ideological reasons, often normative concerns prescribed by their ideology. Some groups, for example, engage in restraint, declining to use violence though it would have strategic benefit. The conclusion lays out a twin-fold research agenda: a ‘weak program’ that analyzes the instrumental adoption of ideology and a ‘strong program’ that explores normative commitments based on particular ideologies and on social preferences.

The Colombian conflict seems a typical instantiation of a “greedy war”and exhibits very strong links between criminal economic activities and rebel organizations. On the contrary, the author suggests that not even in Colombia does the “criminal rebels” thesis hold. On the other hand, the Colombian case shows that criminality and war mix in ways that escape a strictly economic interpretation of war.