Do Protest Bans Legitimize Police Repression? Experimental Evidence from Turkey

Research on repression-dissent nexus presents conflicting findings when it comes to effects of repression on mobilization. Existing literature demonstrates that while repression can quell protest mobilization, it can also lead to more people joining in the protests, ie. backlash mobilization. In explaining these dynamics, newer studies have emphasized the role of public opinion, underlining how the way outsiders perceive protests and their repression might matter when it comes to the repression-dissent nexus. They also emphasized that protest repression can be perceived differently and hence have differential effects depending on the identity of the outsiders. This research, however, has focused primarily on overt police violence and paid little attention to legal forms of repression or their interaction with physical coercion. In this paper, we address this gap by examining protest bans as a distinct and understudied form of legal repression and by analyzing how such bans shape public attitudes toward subsequent police intervention. We argue that protest bans, by rendering protest activity illegal prior to mobilization, can legitimize later police repression in the eyes of the broader public and thereby function as both repression and justification. As secondary hypotheses we argue that the effect of the ban varies by the issue of the protest and the partisan identity of the protestors. We test our argument using an original survey experiment conducted in Istanbul, Turkey, in 2025 (N = 2,102). Respondents were randomly assigned to read news-style vignettes describing protests on three different issues—cost of living, women’s rights, and education in Kurdish—either with or without a prior protest ban, followed by police intervention. Our findings provide strong support for our primary hypothesis: the presence of a protest ban significantly increases public approval of police repression and shifts preferences toward more coercive police responses. This effect is robust across experimental specifications and is particularly pronounced for protests over the economy and women’s rights. Consistent with our expectations, we find weaker and statistically insignificant effects for protests concerning education in Kurdish, where baseline support for repression is already high. Contrary to our expectations, we do not find systematic variation in treatment effects by partisan identity, despite large partisan differences in baseline attitudes toward repression. We do, however, observe stronger treatment effects among respondents with high predispositions to comply with state authority and among women. This paper contributes to scholarship on repression, legitimation, and public opinion by demonstrating how legal tools of repression shape public acceptance of police violence. It also advances theory building from an authoritarian context, highlighting the role of legality in structuring the social acceptability of state violence against protest.