In this paper, we develop a theory to predict the use of disruptive tactics in protest events. Using disruptive tactics entails a trade-off;although they can bring greater efficacy to attaining protest goals, they can also undermine the legitimacy of protest's claims. We argue that how protestors view this trade-off factors greatly into their choice of protest tactics. Drawing on a dataset of over 23,000 protest events that took place in the United States between 1960 and 1995, we find that protest events characterized by broadly resonating claims are more likely to employ tactics that are disruptive, but nonviolent. At the same time, these same events are also less likely to employ violent disruptive tactics. Moreover, when governmental entities are targeted, protests are much less likely to witness the use of both violent and non-violent disruptive tactics. We discuss the implications of our results for social movement theory and the dynamics of collective violence.