Abstract:
Due to the growing number of intra-state conflicts, civil war studies gained momentum after the end of the Cold War. Scholars identified economic, institutional, and identity-related factors that increase the conflict propensity of states. These theoretical approaches contributed to a better understanding of conflict onset. However, it is often overlooked that in spite of determinants favouring violence, conflicts do not always turn violent, but are often waged non-violently. This is especially evident in the case of separatist conflicts in Sub-Saharan Africa that are relatively rare given the propitious context conditions. Hence, the question arises why some African self-determination movements choose violent rebellion, while others claim secession through non-violent collective action despite similar circumstances. Prevailing theories on violent conflict cannot explain this variation in conflict behaviour due to their overemphasis of structural factors. Therefore, this thesis proposes an alternative theoretical and methodological approach. It systematically compares the cases of a violent and a non-violent self-determination conflict (Casamance and Barotseland, respectively) by reference to the framing approach. Framing introduces an alternative perspective into conflict studies. It reveals how movements interpret and construct their environment and how this translates into specific strategies of mobilisation and – violent or non-violent – action. By doing so, it helps to identify micro-mechanisms explaining the (non-) escalation of violence, sheds light on the relevance of cultural, ideational, and emotional aspects for mobilisation and conflict escalation, and yields insights into internal dynamics of protest movements as well as their strategic interaction with their environment. Overall, the approach helps to gain in-depth knowledge concerning the dynamics causing collective violence and to refine our understanding of conflict onset.