Modes of Democratic Breakdown and Violent Conflict

This paper asks whether different pathways of autocratization vary systematically in their propensity for violence. Using a new process-level dataset on democratic breakdowns, we distinguish breakdown processes as to whether they are incumbent-led and non-incumbent-led, unilateral versus negotiated and involve organized armed actors as well as the extent of mass mobilization, institutional manipulation, and violence. Because the data explicitly capture violence during the breakdown episode, the analysis evaluates whether some forms of democratic collapse are inherently more violent than others. Next, by combining the different features of the data, we study which breakdown processes are most likely to escalate into civil conflict under the new autocratic regime. The project’s contribution is to reconceptualize autocratization not as a uniform erosion of democracy, but as a set of distinct processes with sharply different conflict risks.