One of the defining features of this alleged moral constructivism is its aspiration to account for objectivity without robust ontological commitments. Thus, the anti- realist contention that objective values do not possess validity independent of our beliefs, conceptions

Kant famously insisted that “the idea of the will of every rational being as a universally legislative will” is the supreme principle of morality. Recent interpreters have taken this emphasis on the self-legislation of the moral law as evidence that Kant endorsed a distinctively constructivist conception of morality according to which the moral law is a positive law, created by us. But a closer historical examination suggests otherwise. Kant developed his conception of legislation in the context of his opposition to theological voluntarist accounts of morality and his engagement with conceptions of obligation found in his Wolffian predecessors. In order to defend important claims about the necessity and immediacy of moral obligation, Kant drew and refined a distinction between the legislation and authorship of the moral law in a way that precludes standard theological voluntarist theories and presents an obstacle to recent constructivist interpretations. A correct understanding of Kant’s development and use of this distinction reveals that his conception of legislation leaves little room for constructivist moral anti-realism. Kant famously insisted that the autonomy of the will, “the idea of the will of every rational being as a universally legislating will”, is the supreme, in fact the only, principle of morality [G 4: 431]. 1 As he explained, {258} According to this principle, all maxims are repudiated that cannot accord with the will’s own universal legislation. Thus, the will is not merely subject to the law, rather it is subject [to it] in such a way that it must also be regarded as self-legislating and precisely on this account, above all, as subject to the law (of which it can consider itself author). [G 4: 431] This principle led Kant to his third formulation of the categorical imperative, the so-called “formula of autonomy”: “Act only on maxims through which the will could at the same time regard itself as universally legislating.” And the principle forms the core of Kant’s conception of human dignity: the idea “of a rational being who obeys no law other than that which he at the same time legislates himself.” [G 4: 434] Recently, it has become popular to associate this “Legislation Thesis” and Kant’s conception of “autonomy as self-legislation” with what has become known as “moral constructivism”. 2 In A Theory of Justice, John Rawls famously suggested that in moral philosophy “Kant’s main aim” was “to deepen and to justify Rousseau’s idea that liberty is acting in accordance with a law that we give to ourselves.” 3 This aim, Rawls has suggested, led Kant to develop “a distinctive method of ethics” which “is the leading historical example of a constructivist doctrine”. 4 One of the defining features of this alleged moral constructivism is its aspiration to account for objectivity without robust ontological commitments. Thus, the antirealist contention that objective values do not possess validity independent of our beliefs, conceptions or activities; values are, instead, as Rawls puts it, constituted by the activity, actual or ideal, of practical (human) reason itself. [... Kant’s constructivism] goes to the very existence and constitution of the order of values. [...] The intuitionist’s independently given order of values is part of the transcendental realism Kant takes his transcendental idealism to oppose. 5 {259} Along similar lines, Christine Korsgaard has claimed that, on Kant’s view, “values are not discovered by intuition to be ‘out there’ in the world. [...] values are created by human beings [...] constructed {260} by a procedure, the procedure of making laws for ourselves.” 6 “The Kantian laws of autonomy are positive laws”; moral laws exist and possess their normative authority “because we legislate them.” 7 Andrews Reath has recently suggested that such a constructivist interpretation of Kant’s conception of self-legislation is supported by an analysis of the “ordinary” concept of legislation. 8 Our ordinary concept of legislation, Reath suggests, involves the idea of a sovereign legislator, a person with the authority to make a law, one who has control and discretion over the content of the law, and whose legislative action creates new reasons for his subjects to act in the specified way. Thus, Kant’s emphasis on the self-legislation of the moral law is naturally taken as a claim that the moral law is a positive law, thereby constituting apparently decisive evidence that Kant was a moral constructivist. The moral law, or, more generally, all values and normative standards for action, are the products of human rational construction, not part of a reality which is prior to our beliefs or constructions, a reality within which reason could discover them. 9 In spite of the influence and apparent plausibility of such a constructivist interpretation of Kant’s conception of self-legislation, Kant’s moral theory contains an overlooked distinction between the legislation and the authorship of the moral law: while Kant maintains that the moral law is legislated by rational agents, he also insists that the moral law has no author. In this paper I will argue that a correct understanding of the development and use of this distinction within Kant’s moral philosophy clarifies Kant’s conception of legislation, a conception which leaves little room for constructivist talk of “sovereign discre{261} tion”, “making positive law”, “creating” or “constituting values”, or dependence upon “our activity” or “conceptions.” 10 There may be grounds, independent of his conception of self-legislation, to consider Kant a proponent of moral constructivism. It might be thought, for example, that Kant’s famous talk of a “Copernican revolution” in theoretical philosophy 11 implies global constructivist anti-realism or, short of that, that it implies moral antirealism because of the allegedly general limitations which Kant’s transcendental idealism imposes upon the validity of synthetic a priori judgments, of which the categorical imperative is supposedly one. While I believe such suggestions are mistaken, a satisfactory resolution of these issues is well beyond the scope of this paper.12 However, if the argument of this paper is correct, any constructivist moral anti-realism in Kant would have to depend upon such allegedly general features of transcendental idealism, not specifically moral considerations surrounding his conception of self-legislation, sovereignty and moral autonomy, a result of significance in its own right. The argument in this paper will proceed as follows. In Section I, I argue against a tempting “short” argument for moral constructivism. Section II offers an account of Kant’s conception of the legislation of {262} the moral law in the light of Kant’s critique of theological voluntarism and the development of his distinction between the legislation and authorship of the moral law. It is argued that this critique appears to preclude straightforward moral constructivism as well as theological voluntarism. Section III rebuts recent attempts by Reath to articulate a more sophisticated version of constructivism that might seem to avoid this conclusion. I conclude with a final section, Section IV, devoted to a brief articulation of the significance of this non-constructivist conception of self-legislation to Kant’s moral philosophy.