This Working Paper analyses how violent conflict can enhance or reduce pre-existing forms of marginalisation and second, how new forms of marginalisation emerge as a result of violent conflict. To do so, we focus on the province of Tanganyika in the DRC, where the so-called ‘Twa-Bantu’ violent conflict has been disrupting the education sector since 2012, and secondarily on the province of Ituri, which has been affected by repeated armed conflicts since the 1990s. We use a mixed methods approach, combining quantitative data collection methods and several months of qualitative fieldwork. The study shows that the political marginalisation of ethno-territorial groups is key in understanding marginalisation from education in contexts of protracted conflict. Our results show that the Twa minority of Tanganyika has not only been more exposed to violence during the Twa-Bantu conflict, but also that exposure to violence has more severe effects on the Twa in terms of educational outcomes. We analyse key mechanisms, in particular spatial segregation, and the social segregation of schools along ethnic/identity lines. We also analyse the interaction between ethno-cultural marginalisation and economic, social and gender-related marginalisation.