Ramón y Cajal Associate Professor
Department of Social Sciences
Universidad Carlos III de Madrid
francisco.villamil@uc3m.es
Why does violence sometimes leave deep and lasting political legacies while in other cases its effects seem to vanish completely? Victims of the same conflict, living just miles apart, can end up with radically different long-term political identities. In some places, memories of wartime atrocities sustain decades of opposition. In others, silence prevails and no political trace remains. Despite a rapidly growing body of empirical research on the long-term consequences of political violence, we still lack a comprehensive explanation for this variation. Part of the reason is methodological: the emphasis on causal identification in contemporary social science has come at the cost of theoretical development.
Memories at War develops and test a theoretical framework to understand the legacies of violence. The book’s central question is when and why victimization produces long-term changes in political attitudes and behavior, and when it fails to do so. The answer challenges the default assumption in the literature: that violence automatically generates a counterreaction that is then transmitted across generations. Instead, the book argues that legacies are the product of active social processes, and that the local context in which those processes unfold is the key to understanding their direction and strength.
Violence does not leave legacies automatically. Between the original event and the outcome we observe years or decades later, a sequence of social mechanisms operate: framing, remembrance, and mobilization.
First, individuals exposed to violence must develop a political interpretation of events: who is to blame or what the violence means in relation to broader political divisions. This framing is the result of a social process, shaped by prior knowledge, local interactions, and the cultural tools available. Second, these individual interpretations must be shared, discussed, and consolidated into collective memories, which are group-based narratives of the past that keep events salient and politically meaningful. Collective memory is built in face-to-face interactions and can be actively constructed, contested, or suppressed. Third, collective memories must be mobilized into specific political responses, such as voting for a party, identifying with a social group, or participating in protests. Mobilization translates memory into action. Without it, collective memories might not produce any observable political outcome.
These three mechanisms explain how legacies can emerge. But they do not on their own explain why legacies take different forms in different places—or why they sometimes fail to emerge at all. The answer lies in the local ideological context: the political culture embedded in local social networks, defined by the presence of a sufficient number of ideologically-aligned individuals who share a worldview and interact frequently with one another. This local context determines how events are framed, whether and how collective memories are constructed, and which political responses are mobilized. Where a facilitating local ideological context exists, violence is likely to produce a backfiring effect. Where it is absent, individuals lack the shared tools to frame violence politically, collective memories are not formed or are actively suppressed, and legacies get lost. The theory also takes a broader approach and explores how this theory can be used to explain prewar-postwar trajectories at the macro level.
The argument is tested using evidence from two cases: the Spanish Civil War (1936–1939) and the Guatemalan Civil War (1960–1996). Both cases offer within-country variation in the local context, and in both the postwar period included active attempts by state authorities to prevent legacy formation: through repression in Spain and through co-optation and propaganda in Guatemala. This makes the local context observable, because it is what separates communities where legacies emerged from those where they did not.
In Spain, Francoist victimization increased long-term support for leftist parties, but only in municipalities where clandestine opposition networks survived the dictatorship. Where such networks were absent, memories of violence were depoliticized and silence prevailed. In Guatemala, state violence during the counterinsurgency campaigns of the late 1970s and early 1980s produced divergent outcomes: in communities previously exposed to prewar mobilization by Liberation Theology priests and rural cooperativist activists, violence backfired into higher support for the former rebels. In isolated communities without this prior exposure, the state’s propaganda campaign succeeded. Each case is examined using local-level quantitative data, archival and ethnographic sources, and oral history interviews.