Over half of the civil wars since 1945 have occurred in countries that experienced a civil war before. After the end of the Cold War, recurring civil wars have become the dominant type of armed conflict. This empirical reality has profound consequences for the academic study of armed conflict. While most of the academic literature on civil wars focuses on initial conflict onset, recurring civil wars have received increasing attention lately. Currently, the scholarly debate is shifting toward post conflict environments. A community of researchers tries to determine which factors are decisive for lasting peace after a conflict has ended. The diverse variety of factors analyzed includes power-sharing agreements, the presence of peacekeeping forces, economic development, and characteristics of the previous civil war such as conflict duration and severity. However, a comprehensive comparison of the determinants of initial civil war onset and of civil war recurrence based on quantitative data is missing so far. Why do some civil wars relapse while others do not? Are the causes of civil wars the same for conflicts in states with and without a history of internal conflict? And how should civil war research proceed to investigate how lasting peace after civil wars can be achieved? This paper sets out to answer these questions by introducing a quantitative model on the causes of civil war onset and recurrence. The model includes data to test the established greed and grievance approaches, the ethnic fractionalization thesis, and a variety of arguments from the post conflict environment literature. The results indicate that initial civil war onset and civil war recurrence are caused by different factors, and thereby underline the importance of research on post conflict situations.
