Codified at the 2005 United Nations World Summit, the doctrine of Responsibility to Protect articulates an ideal of international interventions motivated by compassion for victims and a desire to bring stability to hot-spots around the world. Despite this consensus, practitioners and scholars have debated the importance of unintended consequences stemming from the expectation of third-party intervention. We analyze how third-party intervention shapes the incentives to arm, negotiate settlements, and fight wars in a parsimonious game theoretic model. Among the unintended consequences we find: interventions that indiscriminately lower the destructiveness of war increase the probability of conflict and increasing the cost of arming makes destructive wars more likely. Other interventions, however, can have much more beneficial effects and our analysis highlights peace-enhancing forms of third-party intervention. From a welfare perspective, most interventions do not change the ex-ante loss from war, but do have distributional effects on the terms of peace. As a result R2P principles are hard to implement because natural forms of intervention create incentives that make them largely self-defeating.