Abstract
Are non-violent strategies conducive to loyalty shifts among security forces? It is often argued that non-violent campaigns are more likely to spur regime change compared to violent campaigns, partly because non-violent strategies encourage loyalty shifts among the security forces. While there seems to be a consensus about the campaign strategy-defection nexus, the few empirical research on the subject provide mixed and inconclusive results about this relationship. Furthermore, these works neglect crucial insights from the civil war literature, which emphasize how the cost of fighting, access to escape routes, and the organizational capacity of rebel groups can incentivize desertion and side-switching. We synthesize these different fields of
research and provide a comprehensive framework specifying the role of strategy. We demonstrate that non-violent strategies are linked to large-scale loyalty shifts, but we find no effect when using size-agnostic measures. Additionally, when assessing different types of disloyalty, we reveal that non-violent and violent strategies elicit distinct types of subversive actions. Security forces are more likely to shirk, voice loyalty to the campaign, or quit supporting the regime during non-violent campaigns, while desertion and side-switching are more common when the campaign is violent.