I defend a pluralist version of metaethical constitutivism. Constitutivism says that practical norms are grounded in the nature of rational agency. Pluralist constitutivism says that practical norms for an agent are grounded in her kind of rational agency. Pluralism, I argue, avoids a standard problem for constitutivism that I call the “minimal agency problem.” The problem is that for any given practical norm, there is a possible “minimal” rational agent for whom that norm is not a constitutive standard of rational agency. If that is right, then there are no constitutive standards of rational agency as such. For pluralist constitutivists, the possibility of minimal agents poses no problem: even if there are no constitutive standards of rational agency as such, there might be constitutive standards of our kind of agency. I then respond to two as-yet-unanswered challenges for pluralism. First, can we draw a principled distinction between kinds of agency? I answer yes: we can do so by appealing to forms of reasons explanations. Second, do the standard arguments for constitutivism also support pluralism? Again, I answer yes: suitably modified, they do.

Purely instrumental agents can reason about how to realize their ends, but not about which ends to pursue. They can do one thing in order to do another but cannot choose their final ends for reasons. Some have argued that such agents are impossible, and that the success of ‘moral constitutivism’ depends on their impossibility. Moral constitutivists hope to ground moral norms in the nature of rational agency as such. But if purely instrumental agents are possible, then rational agency is too thin to ground moral norms. I argue that purely instrumental agents are possible, but their possibility by itself does not refute moral constitutivism. Moral constitutivists might embrace a pluralist form of constitutivism, according to which practical norms for an agent are grounded in the nature of her kind of agency, not in the nature of rational agency as such. This ‘pluralist’ variant of constitutivism, however, faces problems of its own. I leave open whether pluralism fares any better overall than extant versions of moral constitutivism.

Duties to oneself are central to Kant’s moral thought. Indeed, in his Lectures on Ethics, he claims that they “take first place, and are the most important of all” (27:341). Despite this, Kant is not clear about what they are or why they are ‘the most important.’ What is it for a duty to be owed to oneself? And in what sense do such duties ‘take first place’? I answer these questions: a duty to oneself is a self-legislated duty, and they ‘take first place’ in that all ethical duties are, fundamentally, duties to oneself. Based on a close reading of Kant’s “Apparent Antinomy” regarding duties to oneself in The Doctrine of Virtue §§1–3, I argue that Kant holds that (1) a duty is owed to its legislator, and (2) all duties are self-legislated. I then argue that this interpretation is further supported by Kant’s understanding of the Categorical Imperative as a “principle of autonomy” in the Groundwork (4:433). I conclude by offering an interpretation of the Categorical Imperative on which even though all duties are ultimately owed to oneself, there remains an important sense in which some duties are owed to others as well.

We articulate and defend the most central claims of contemporary Kantian moral theory. We also explain some of the most important internal disagreements in the field, contrasting two approaches to Kantian ethics: Kantian Constructivism and Kantian Realism. We connect the former to Kant’s Formula of Universal Law and the latter to his Formula of Humanity. We end by discussing applications of the Formula of Humanity in normative ethics.