A third version of the categorical imperative follows in Kant’s text almost immediately upon the four examples of duties derived from the second version, the formula of humanity. This new formula, emphasizing the way we may be supposed to have acquired our subjection to moral requirements, is called the formula of autonomy. It says that we are subject to the requirement of our maxims’ conformity to universal law through our own free will. That is what Kant indicated in the quotation above, referring to the will’s “giving the law to itself” (Kant 1997, 39/4:431).
Why now a third version? It looks at first as this new version of the categorical imperative has been introduced for merely systematic reasons—in order to round up the previously introduced pair to a complete triad. Kant does not bother this time to derive the same four duties, as he had in illustrating how to apply the formula of universal law and the formula of humanity. And it is hard to see how the formula of autonomy would be applied in moral judgment any differently from the canonical version of the categorical imperative, the formula of universal law. Apparently, the only innovation introduced by the third version is that, within its formula, it refers implicitly to a pure will. Kant explains that it is able “to indicate in the imperative itself the renunciation of all [empirical] interest in volition from duty...as the specific mark distinguishing categorical from hypothetical imperatives” (Kant 1997, 40/4:431-2). But it requires some subtle reasoning to see how it does this. The idea is that if the rational being is supposed to give the law to himself, through his own will, then in doing so he cannot be beholden to any empirical interests, or any theological commitments. For this would mean that, ultimately, these determine the requirements of morality; and it would mean, too, that his interest in adhering to those requirements is only indirect: morality is fundamentally a form of prudence; or a matter of piety. In each of these cases, especially the case of prudence, the moral law ends up commanding with a hypothetical imperative rather than one that is categorical: You ought always to do the right thing because, in the end, that is in your best interest—which means that nothing can be the right thing to do unless it is in your best interest. Though this point may not seem especially interesting, Kant is interested in it as a way of developing his thesis that all previously expounded moral principles exhibit a similar defect, and only his system, based on a categorical imperative, accurately reflects the nature of morality. In the text he will shortly make this thesis more explicit, in terms of the distinction between autonomy and heteronomy.
The derivation of the formula of autonomy. The premises by which we arrive at a version of the categorical imperative referencing autonomy—the self-legislation of moral requirements—are the previous two versions. Kant presents the argument as follows:
the ground of all practical lawgiving lies (in accordance with the first principle) objectively in the rule and the form of universality which makes it fit to be a universal law (possibly a law of nature); subjectively, however, it lies in the end; but the subject of all ends is every rational being as an end in itself (in accordance with the second principle); from this there follows now the third practical principle of the will, as supreme limiting condition of its harmony with universal practical reason, the idea of the will of every rational being as a will giving universal law. (Kant 1997, 39/4:431)
In order to see how we arrive at the formula of autonomy at the conclusion of this derivation it is necessary to focus on the second premise, the formula of humanity. Here the important concept of an “end” implies legislation, or lawgiving. It is possible to see an end, like a law, as something making it necessary to will an action—either as a means of bringing about its existence, or in order not to bring about its non-existence. If every rational being is an end in the latter sense, as implied by the formula of humanity, then every rational being is lawgiving. Here we do not conceive of something else—like God—legislating that all rational beings are to be treated always as ends, on account of their intrinsic value. Instead, the idea is that each rational being, simply in virtue of being an end in itself, makes necessary (legislates) that it be treated always as an end. The legislation given by each is also universal, applying to itself as well as to every other rational being. Hence, the third version of the categorical imperative commands not acting on any maxim that could not qualify as giving universal law.
What the principle of autonomy introduces. It is hard to see any real difference between the new, third version of the categorical imperative and its first version, emphasizing conformity to universal law. In the first version we are told that an action is morally permissible provided its maxim can be willed as a universal law. Here we are told that it is permissible provided the maxim can qualify as giving universal law. As Aune remarks: “This difference is not really great, for willing that something be a universal law seems to be an act of legislation itself” (Aune 1979, 86). Still, with the introduction of the formula of autonomy comes a crucial part of Kant’s overall argument in the Groundwork. Recall how the idea of conformity to universal law implies that actions are ruled out as immoral if their maxims are in some way contradictory upon universalization. That may suggest merely that morality is a system of objective practical reason—that as rational beings we are subject to a requirement of conformity to universal law that is legislated by impersonal “reason.” One might agree with this and still be tempted to ask, “so why should I always obey reason?” (Lehrer 2003). The key innovation of the third formula is to suggest something about the obligating force of practical reason’s moral laws. It is to suggest, as the quotation above explicitly says, that we are subject to its laws through an act of our own will. This insight provides, in turn, a basis for conceiving of a community of legislators as “the kingdom of ends,” a concept that will be useful for Kant’s argument for the validity of the categorical imperative, in Groundwork III.
Aune, Bruce (1979), Kant’s Theory of Morals, (Princeton, NJ: Princeton University Press).
Lehrer, Keith (2003), “Reason and Autonomy,” Social Philosophy & Policy 20: 177-98.
Kant, Immanuel (1997), Groundwork of the Metaphysics of Morals, trans. Mary Gregor
(Cambridge: Cambridge University Press).