This thesis analyses social dynamics in Masisi, a rural area of eastern Congo, and considers how the relationships on which people rely contribute to the (re)production of violence. The province of North Kivu is often called the epicentre of war; if this is the case, Masisi is its flashpoint. As in many areas of the world where endemic poverty and poor governance are compounded by environmental factors and insecurity, Masisi has endured decades of recurrent political violence. Indeed, in 2007 a controversial report estimated that in the decade between 1998-2008 5.7 million Congolese died of conflict-related causes (Coghlan et al., 2007). A decade later, despite peace accords and the presence of one of the United Nation's longest and largest peacekeeping mission, violence in eastern Congo continues. For millions of Congolese, navigating the risks of conflict in these difficult places has become a way of life. Despite recent appeals, how people live in such environments remains understudied. Moreover, we seldom consider how the elements that make life possible in such circumstances both interact with violence and feed processes that influence its recurrence. Yet, streams of scholarship across the social sciences agree that social connections and networks mould people's opportunities, help them weather adverse circumstances, and shape their lives. In addition, researchers of contentious politics show how apparently mundane social relations influence processes that make outbreaks of political violence more likely. However, despite these complementary insights, relational analyses of such interactions are infrequent. I draw on mixed-methods data (ethnographic, historical, and qualitative network methods, survey data, published and unpublished secondary sources and grey material) to build a relational analysis of Masisiens social relations and networks of support, as well as of their cumulative social outcomes. With the help of oral histories and secondary sources, I trace how people's needs, and the social, political, and economic context in which they are embedded, shaped these relationships and networks from the late 19th century to today. I show how these experiences led people to prioritise certain types of relations and networks, and moulded key institutions of governance and support (family, clans, associations). Whilst these relationships and networks help people get by, create new opportunities, meet aspirations, and navigate the unpredictability of their environment, it comes at a price. I argue that the very social relations and networks on which Masisiens depend generate network properties that interlace seemingly unrelated micro dynamics with macro drivers of conflict. In turn, these feed processes that encourage the recurrence of violence, whilst simultaneously constraining it. I posit that whilst local dynamics do not have the power to stop top-down outbreaks of political violence (as some local-level peacebuilding scholars argue), they can limit its impact. As, across the world, civil conflicts morph into uneasy continua of violent peace punctuated by flares of acute conflict, it becomes increasingly urgent that we not only improve our understanding of how people live and get ahead in such difficult places, but also explore how seemingly unrelated social dynamics can influence processes that maintain the status quo and make outbreaks of violence more likely.