Under what circumstances are states able to successfully use violence, particularly lethal force, against social movements? Put another way, to what extent can contemporary European movements hope to achieve goals such as reversing austerity measures, ending neoliberalism or constructing a more real democracy without facing tanks and torture chambers? This paper explores the question of state legitimacy and the internal use of violence from a historical and theoretical rather than normative standpoint, distinguishing between the rhetoric of violence / non-violence and actual deaths. It starts from the relative decline in the use of lethal force by states against internal opposition which characterises some parts of Europe since the 1950s by comparison with earlier decades. It does this not in order to write an irenic narrative or to minimise the continuing reality of state violence on many levels, but rather to ask what prevents states from killing when entrenched interests face serious challenges to the status quo – from an activist perspective in which our ability to challenge the structures of power without being shot or tortured is an immediate, practical concern and not something that can be taken for granted. While social actors whose fundamental interests are threatened are regularly willing to kill in order to prevent social change, their ability successfully to deploy lethal violence is another matter and depends on the willingness of other actors to support them. The successful use of violence is not simply a matter of coercion but involves the successful construction of consent for its use among a hegemonic alliance. Social movements do not hold most of the cards in this respect, but the process of restricting the state’s ability to kill goes hand in hand with movements’ other concerns of constructing counter-hegemonic alliances for social change. This is important not only to contemporary movements but to all political actors who value the possibility of engaging in democratic struggles – most particularly in contexts where constraints on state violence are weaker. The paper draws on historical and contemporary examples and a range of European countries to situate its analysis.