This paper investigates the interaction of protest and repression, drawing on Islamist protests and state repression in Tunisia and Algeria in the early 1990s. Putting the findings from large-n quantitative studies to the test in a case-centric design, it identifies serious shortcomings in current, largely static, approaches and proposes a shift towards a dynamic understanding of the relationship between protest and repression: Specific repertoires and practices of protest interact with and make more likely specific repressive responses (and vice-versa) in cycles of escalation or de-escalation. Building on this dynamic understanding, the paper specifies escalating and de- escalating practices and context conditions.
