This dissertation consists of three essays investigating the determinants of civil conflict from the aspects of history, geography, and external assistance, respectively. Each of the essays uses empirical economic techniques to analyze how these factors affect the spatial and temporal distributions of modern conflict events. I explore the long-run effects of slave trade on conflict in Africa. By using a geo-coded disaggregated dataset in sub-Saharan Africa and exploiting variation of slave trade intensity within countries, I find a robust positive relationship between slave exports from a region and current conflict prevalence. Evidence from an instrumental variable approach suggests that the relationship is causal. I then discuss the potential causal channels underlying the relationship, suggesting that slave trade is correlated with mistrust, local institution deterioration, and a weaker sense of national identity. I then examine how slave trade in history affects the relationship between weather shock and civil conflict. Exploiting weather variation during the growing season of the locally dominant crop, I find that adverse weather shock significantly increases the likelihood of conflict incidence, onset, and intensity. Furthermore, the effect of weather shock on the risk of civil conflict is substantially amplified by the exposure to slave trade. I also study the dynamic interdependences among foreign aid, development, and conflict. Although foreign aid is sensitively responsive to the conflict or development shock, its effects on reducing conflict and improving development are largely dependent on the wealth level and conflict proneness of the recipient country. I find that foreign aid only mitigates conflict in middle income developing countries, but can enhance the development of the poor and conflict-prone countries.
