The Responsibility to Protect (RtoP) has become a prominent feature in international debates about preventing and responding to genocide and mass atrocities. Since its adoption in 2005, it has been discussed in relation to a dozen major crises and been the subject of discussion at the UN Security Council and General Assembly. This article takes stock of the past five years and examines three questions about RtoP: What is its function? Is it a norm, and, if so, what sort? And what contribution has it made to the prevention of atrocities and protection of vulnerable populations? In relation to the first, it argues that RtoP is commonly conceptualized as fulfilling one of two functions (a framework for a policy agenda and a speech-act meant to generate the will to intervene), but that these two functions are incompatible. In relation to the second question, it argues that RtoP is best thought of as two sets of norms relating to the responsibilities of states to their own populations and international responsibilities. The first set are well defined and established, the second though are indeterminate and lack compliance-pull, limiting the extent to which RtoP can serve as a catalyst for action. This, the article argues, is reflected in RtoP’s track record thus far. RtoP has failed to generate additional political will in response to atrocity crimes but it has proven useful as both a diplomatic tool and as a policy lens.