Naturalism is an ambiguous philosophical term that refers to one of two general philosophical theses: the metaphysical thesis that all facts, including moral facts, are natural facts, and the epistemological thesis that the methods of philosophical inquiry, including moral inquiry, are continuous with those of the empirical sciences. The metaphysical thesis has long been the default form of naturalism in ethics. The thesis of this dissertation is that, first, despite the convention, it the epistemological thesis—which I call "methodological naturalism"—that is genuinely naturalistic; and second, that existing forms of epistemological naturalism in contemporary ethics have largely failed to develop genuine continuity between moral inquiry and scientific inquiry and hence are not genuine forms of methodological naturalism.

The argument proceeds in two parts. First, I examine a growing trend in moral philosophy toward the use of scientific data, especially from cognitive science, psychology, and psychiatry, for resolving longstanding philosophical disputes about the nature of morality and moral agency. This empirical strategy is widely regarded as naturalistic. I argue that many of these approaches fail to satisfy a fundamental requirement on genuine naturalism, namely the development of continuity with scientific inquiry, either because they wrongly take the metaphysical thesis rather than the methodological thesis as primary, or because despite rightly privileging the latter they nevertheless seek to resolve traditional philosophical disputes that are founded on an outmoded psychological framework, namely the framework of commonsense folk psychology. The traditional philosophical disputes in question are those that jointly constitute what Michael Smith (1994) calls the "Moral Problem," the central organizing problem in metaethics. I focus in particular on the dispute surrounding the so-called Humean theory of moral motivation, according to which an agent's being motivated to act morally requires not just a belief about what is morally right but also a desire to act in accordance with that belief.

The second part of this dissertation applies naturalistic methodology to the Humeanism dispute. I synthesize data ranging from cellular-level neuroscience to cognitive science, social neuroscience, and neuroeconomics to sketch a philosophically informed scientific model of moral cognition and motivation. I argue that this model is in tension with the commonsense psychological framework presupposed by the Moral Problem and by the Humeanism dispute in particular. Central to this argument is the idea that cellular level data on decision-making, which is too often regarded as philosophically irrelevant, can be brought to bear directly on the kind of data from cognitive science that is frequently cited in moral philosophy. The model of moral judgment, decision, and motivation that results, I argue, is a much more promising starting point for philosophical inquiry into moral agency than the framework of commonsense folk psychology. Genuinely naturalistic approaches to ethics must aim to develop real continuity between philosophical inquiry into morality and scientific inquiry into human agency. This requires that we use the best available scientific data to vet the empirical assumptions at work in traditional philosophical theories rather than to merely use that data to adjudicate between them.