Street-Level Autocrats: Individual Decisions with Collective Consequences

Led by Kristin M. Bakke

Oct 2018 – Dec 2021

Why do some pro-democracy movements succeed in overthrowing authoritarian regimes, while others do not? Why does the military sometimes 'shoot to kill' when faced with revolutionary movements, while in other cases they refuse to do so? And why are some post-revolutionary regimes more enduring than others? We propose that these questions are interlinked. The overthrow of an authoritarian regime is hard to explain without considering the army's decision to side with the revolutionary movement, just as it is difficult to explain the political trajectory of the new regime without considering the army's role during the revolution. 

Indeed, the loyalty of the security forces is vital for understanding what happens when large-scale social movements challenge the regime. When ordered to use force, members of the police and military do not always comply. If a sufficiently large share of the military and police decide to shirk or even desert and join the protesters, the days of the regime are likely to be numbered. The actions of these 'street-level autocrats' during the revolution also matters for the new regime's legitimacy—and, as such, long-term stability. People's perceptions of whose side the members of the security apparatus were on are likely to have an enduring impact on their trust in the security apparatus and the state as a security provider.

Using both qualitative and quantitative research methods, coupled with new and unique data, including surveys in Ukraine, Serbia, and Georgia, we aim to build a better understanding of regime change and stability. By disaggregating the actors that make up the state—from the incumbent government down to the individual member of the security apparatus—and examining how they interact with civil society, we aim to explain why some pro-democracy movements succeed only in the short run but not in the long run, while others successfully bring about democratic consolidation. 


​This project is funded by the Research Council of Norway (FRIPRO).

An error has occurred. This application may no longer respond until reloaded. Reload 🗙