Although much is known about the dynamics of civil wars, we still know little about the repertoires of violence that insurgents select. Some rebels spend their resources on military encounters with the state forces, whereas others combine guerrilla actions with terrorist bombings in main cities. I propose a detailed case study focusing on the spatial and temporal variation of violence produced by the Frente Farabundo Martí de Liberación Nacional (FMLN) in El Salvador, from 1979 to 1992. I draw on previous work to claim that the capacity of insurgents to seize and hold territory from the state’s grip has overwhelming effects on warfare dynamics. In this paper is tested whether tactics are also determined by territorial control. Whenever insurgents are able to control territory, they are better equipped to wage a guerrilla war. To the contrary, if insurgents do not liberate territory from the state’s hands, they must remain underground, and therefore they cannot but rely on attacks most people would identify as fully terrorist. The theoretical implication is straightforward: guerrilla warfare should be concentrated in the war zones, whereas terrorist attacks, such as IEDs, should thrive in urban areas under the incumbent’s control. I test this argument by analyzing the temporal and spatial distribution of FMLN attacks. I have created a dataset that includes the number of yearly attacks related to four tactics (guerrilla, bombing, assassination and kidnap) in any district in El Salvador from 1979 to 1992, together with demographic, geographic, political and economic covariates. The results strongly endorse the theory.
