We survey the recent literature on the mental health effects of conflict. We highlight the methodological challenges faced in this literature, which include the lack of validated mental health scales in a survey context, the difficulties in measuring individual exposure to conflict, and the issues related to making causal inferences from observed correlations. We illustrate how some of these issues can be overcome in a study of mental health in post-conflict Bosnia and Herzegovina. Mental health is measured using a clinically validated scale; conflict exposure is proxied by administrative data on war casualties instead of being self-reported. We find that there \nare no significant differences in overall mental health across areas which are affected by ethnic conflict to a greater or lesser degree.