How does information about repression affect protest propensity in an autocracy?

Despite substantive research interest, the effect of repression on political uprisings is still unclear, with results ranging from mobilizing to deterring to curve-linear effects.

Arguments are traditionally tested on a high level of aggregation, neglecting subnational variation, and most accounts leave open the question of how people gain knowledge about repression - a crucial precondition for any effect to work.

I argue that the effect of repression is heterogeneous because it is associated with spatially varying costs and threat perceptions, depending on the distance to the threat. I expect that while repression generally backfires, increasing the overall protest propensity in a country, it deters protests locally at the same time. Moreover, if repression at one protest site is to have an impact on mobilization elsewhere, information about repression must be communicated through a channel that reaches a large share of society.

Combining original geo-referenced protest and repression data on the day-municipality in the German Democratic Republic in 1989 with a content analysis of West German television broadcasts, I show that foreign media reports on repression fostered protest diffusion nationwide. At the same time, these reports deterred protests locally. Reports on repression at a specific protest location reduced the subsequent protest propensity in surrounding communities. Conversely, acts of repression without media coverage had no effect on the spread of protests.

The findings have important implications for the study of the repression-protest nexus and foreign media effects on regime stability.