Democratic Support, Protests, and Authoritarian Violence in Non-Democracies

Abstract

This thesis studies the effects of democratic support in non-democracies in relation to protest participation as well as violent/non-violent authoritarian responses to protests. Using Bayesian process-tracing, I investigate two democratic support-related hypotheses in case studies of democratic mass mobilisation. One case study centres on Georgia in 2003; the other centres on Armenia in 2018. The first "mobilising" hypothesis suggests that strong support for democracy on the individual level makes people more likely to join democratic protests. The mobilising hypothesis receives strong positive confirmation from both case studies. The second "moderating" hypothesis predicts that a high level of democratic support in non-democratic countries will make autocrats less likely to use violence against protesters. The findings for the moderating hypothesis are mixed: it is disconfirmed in Georgia's case, but there is insufficient evidence fully to confirm or disconfirm the hypothesis in the case of Armenia. Overall, the findings from the case studies show that democratic support plays a key role in motivating people to join democratic protests in non-democracies when the risks are high and the benefits are unclear.