Cultural and religious demography and violent Islamist groups in Africa

•Power relations of ethnic groups can incentivise the use of alternative identities.•Political dominance of ethnic groups is related to lower Islamist violence.•Political marginalisation of ethnic groups is related to higher Islamist violence.•The size of Muslim population is not related to levels of wider political violence.•Islamist and wider political violence occur in discrete political, demographic contexts.

This research explores the relationship between cultural demography and Islamist violence in Africa in a cross-national time series study. It argues that while religious demography can explain some aspects of Islamist violence, these explanations have to date been privileged over analyses which take into account the way institutional and political relations of the state incentivize and de-incentivize the salience of particular identities in collective action. This paper uses disaggregated conflict event data from the Armed Conflict Location & Event Dataset (ACLED) to test the relationships between religious group size, diversity, ethnicity and Islamist violence. The results highlight that approaches to explaining Islamist violence emphasising the cultural specificity of Islam as particularly prone to violence, and those focusing on competition between diverse identity groups as explanations for the rise of Islamist violence are misguided. Rather, ethnic political power relations emerge as important interacting factors in religious identity conflict, with Islamist violence as an example. The article makes an original contribution both empirically, by testing existing theories of Islamist violence on previously unanalysed data; and theoretically, by highlighting the importance of political marginalisation and strategic identity construction as explanations for violent Islamist activity.