A central challenge facing those attracted to consequentialism, and to other impartialist ethical views, is to do justice to our twin intuitions that our ethical theory should be impartial in some important sense but also that it should not be unrealistically demanding. I consider one seemingly attractive solution, advocated by Roger Crisp, which lies in proposing a dualism of practical reason, within which impartial reasons for action compete with partial reasons grounded in the interests of the moral agent. I argue that this proposal goes wrong in misconstruing the relationship between reasons and requirements, and develop an alternative framework, with a single source of reasons for action, which retains the strengths of Crisp's account,...