This article defends a claim about the conditions under which agents possess normative reasons for action. According to this claim, an agent has a normative reason to φ only if it’s psychologically possible for that reason to motivate the agent to φ. The claim is called‘Williams’s explanatory constraint,’since it’s drawn from Bernard Williams’s work on the topic of practical reason. A two-premise‘master argument’ for Williams’s explanatory constraint is put forward. First, an agent has a normative reason to φ only if that agent has the ability to φ for that reason. The second premise states that an agent has the ability to φ for a normative reason only if it’s psychologically possible for that reason to motivate the agent to φ. It is suggested that the ability to act for a normative reason involves the ability to act from a consequence-sensitive process of practical reasoning. Furthermore, normative reasons for action can motivate agents by being the objects of psychological states—particularly beliefs,desires, and intentions. In reply to the objection that normative reasons can never be objects of psychological states, I contend that all normative reasons are capable of being represented as the objects of psychological states.