Community and consent: Unarmed insurrections in non-democracies

This study explores popular challenges against the state through nonviolent means. Although previous research has started to examine the effect of these 'unarmed insurrections', the relationship between challenging the state apparatus (vertical legitimacy) and the state identity (horizontal legitimacy) has not been adequately addressed. We argue that unarmed insurrections are most likely to be successful when challenging the vertical, rather than the horizontal, legitimacy of the state. Studying data for 287 years of protests in 57 non-democratic countries during the period of 1946-2006, we find support for three implications of this proposition: 1) campaigns that demand governmental regime change are more successful than campaigns for territorial changes; 2) success is less likely when the identity of the insurgents and the government is split along ethnic lines; and 3) success is less likely when society is highly polarized along ethnic lines rather than being ethnically homogeneous. Thus, when the community is divided, the efforts to withdraw consent will be less effective. The study discusses the implications of these findings for policymakers and scholars interested in nonviolent strategic action.