A popular claim is often that religion is at the heart of violent conflicts today; and that religious conflicts are more intractable and bloody than other conflicts. This paper investigates recent civil conflicts in terms of their religious dimensions, in order to test the claim that religious wars are more severe than conflicts that do not involve a religious dimension. This paper answers to two challenges from the extant literature on religion and civil war. First, whereas most large-n studies have investigated religious conflict in terms of the identities of the actors involved or the relative strength of religious groups in terms of population size, religious conflicts are in this paper defined in terms of three dimensions: the actors involved, the issues at stake, and if religious rhetoric and symbolism is used for legitimizing uprisings and incite mobilization. Second, there is a discrepancy between theories of civil conflict and most systematic studies: the theories emphasize dyadic interaction, whereas the empirics most often focus on attributes of states and not the non-state antagonists. Using a dataset on recent civil conflicts I introduce variables that capture the three dimensions of defining religious conflict taking into account both the government and the rebel side. The analysis examines the extent to which the claim about bloody religious conflict is substantiated when using systematic data on battle-related deaths to capture conflict severity. I find that religion clearly has a bearing on several contemporary conflicts, but its overall significance for predicting battle casualties is limited.