The UN has placed rhetorical emphasis on the prevention of armed conflict before it starts and has taken selective action toward that end. What determines where the UN gets involved? We examine UN preventive actions by focusing on UN Security Council (UNSC) resolutions in self-determination (SD) disputes. We argue that UN decisionmakers consider at least three factors when deciding where to target preventive action: the dispute’s conflict history, the potential for regional contagion, and the characteristics of the dispute. We further argue that the political dynamics of UNSC decisionmaking constrain the UN’s ability to pay attention to the third factor (the characteristics of the dispute). We test this argument using data on all UNSC resolutions comprising the authorization of diplomatic engagement, condemnation, the authorization of sanctions, and the deployment of force targeted toward SD disputes from 1960 to 2005. We find that the UN is much more likely to act in nonviolent disputes that have a history of violence and in disputes with a potential for regional contagion. The analysis shows that, while political barriers likely restrict the ability for the UNSC to act when dispute-level characteristics suggest armed conflict is more likely, the UN does act proactively to prevent violence, rather than just reactively responding to existing violence.