Ronald Dworkin’s philosophy of law, in its mature version, is grounded in at least two central claims: first, a thesis about law and morality, which we might call the One-System Thesis; second, a thesis about how moral and legal propositions can be said to be true or false, which we might call the Interpretive Thesis. While the One-System Thesis holds that law and morality form a single system, the Interpretive Thesis makes two distinct claims: first, truth of interpretive propositions—such as moral and legal propositions—must be established from within the practice in which they figure; second, the soundness of an interpretive proposition is related to the purpose of the practice under consideration. Mark Greenberg’s Moral Impact Theory of Law accepts the One-System Thesis while rejecting the Interpretive Thesis. The Moral Impact Theory is a metaphysical theory of how moral facts rationally determine the content of the law. Its main contention is that the actions of legal institutions have an impact on the moral obligations people have in a polity, and the content of the law is made up of the moral obligations that result from the actions of such institutions. Greenberg assumes that moral facts pre-exist and have some metaphysical priority in relation to legal facts. Moral facts must be prior and independent from legal practice in order to play a part in the rational determination of the content of the law. The point of this paper is to offer a response to Greenberg. I argue that the One-System Thesis only should be supported if the Interpretive Thesis is correct, and that without the latter the former becomes an implausible version of natural law jurisprudence.