In his Groundwork of the Metaphysics of Morals, Immanuel Kant crafts an elegant solution to the free will problem that divorces rational will from the apparently deterministic framework of reality. By distinguishing between objective reality and its appearances, Kant proposes a dual membership status that shares rational beings between the realm of objects in themselves (i.e. the noumenal realm) and that of objects as they appear (i.e. the phenomenal realm.)

Part of his goal in this division is to find a way to free the will from any notion of extrinsic causation that would relegate it to the position of mere effect. It is vital for him to prove that the will contains its own principle of causation in order to salvage the individual agent’s responsibility for her actions, which he obviously views as a prerequisite for the existence of morality.

I will explain how Kant presents his case for the authority of moral law over human beings through his appeal that all rational beings have autonomous will. I will then show how this necessitates his noumenal–phenomenal distinction. Finally, I will argue why Kant’s proposal fails to solve the free will problem and put forward one possible human implication of his moral universe.

The Kantian Proposal

Kant holds that the will must be autonomous because it is the principle of causation for rational beings. If its actions were only the effects of the laws of nature, then it could not be considered free. A will requires autonomy to be free—it cannot be subject in any way to laws that exist outside of it.

That is not to say, however, that it must not be subject to any laws at all. Quite the opposite, in fact, as a will that is not subject to laws is therefore incapable of participating in causality, neither subject to it as an effect nor as the cause of any effects. This would be absurd, as the will itself is a cause and would be no will at all were it not able to bring about any effects. Instead, Kant claims, the will must be autonomous—it must contain its own causal principle, which it gives to itself. Since the moral law has already been shown through the categorical imperative to be an end in and of itself, Kant concludes that the free will is analytically connected to the moral law, going as far as to say, “a free will and a will under moral laws are one and the same” (4:447).

Because Kant has shown that moral law governs the free will, then if we have free will, moral law must govern us. Kant adds that, insofar as we are rational, we act “under the idea of freedom” (4:448), so that when we perform rational acts, we say that we chose this act, or that we could have chosen that other if we had wanted to will it. Because our rational choices presuppose freedom, we see ourselves as being governed by the laws of freedom, and the laws of freedom are the moral laws. As rational beings, then, we should see our free choices as being governed by moral law.

The Veneer of Appearances

For Kant, the concept of morality necessarily follows from the idea of the free will, insofar as the will is free if and only if it operates under the moral laws that it gives to itself. This would be a circular argument, were it not for Kant’s shift to a discussion on his two-aspect conception of the world: the noumenal realm, which is the way we can think of the world as it is in itself, and the phenomenal realm, which is the way we can think of the world as it appears to us.

Through our senses, we participate in the phenomenal realm where we are passive recipients of the world’s appearances. But we are capable, through reason, of considering the world apart from its appearances, which allows us to think of it as noumenal. A very important point regarding this noumenal perspective, however: though such a perspective may give us for recognizing the existence of a noumenal realm, there is still no way for us to know anything about it, that is, we have no faculties through which we can things in themselves apart from their appearances—it is not a part of our knowable world.

One’s dual membership to both perspectives allows the rational being to regard himself from either so that he can “cognize laws for the use of his powers and consequently for all of his actions” (4:452). When we consider ourselves noumenally, we can think of ourselves operating under the autonomous laws of reason—in other words, we can understand ourselves as the free authors of our own choices, insofar as we are rational beings. When we view ourselves phenomenally, however, we see ourselves operating heteronomously, subject to the natural laws of cause and effect that govern the sensible realm—from this perspective, we are determined.

Understanding what it means for something to be a thing in itself as opposed to its appearances, Kant gives primacy to the noumenal realm, stating, “the world of understanding contains the ground of the world of sense and so too its laws” (4:453).

If there were no phenomenal realm and we were exclusively rational, then moral law would command our actions as a natural law, as the moral law governs our rational will. However, in the phenomenal realm, the moral law takes the form of an “ought” statement, obligating us to act from reason, but it is a practical impossibility to entirely avoid the irrational needs and that come into play in this realm.

This is Kant’s solution to the problem of free will: that when we act apart from the moral laws we’ve given to ourselves insofar as we are rational agents, we act heteronomously in accord with the principle of causality that governs the phenomenal realm. through the external causation of the sensible realm. But when we act in accordance with our moral laws, we act autonomously, through our own free will.

The Irrelevance of Kantian Freedom

Though it is a convincing argument for free will amid the existence of causation, it is a somewhat disturbing conclusion. If our only hope of freedom exists in the noumenal realm, and exists in our reason, which is itself governed by the so-called laws of freedom through which we universally command for ourselves the same laws, then our freedom only exists in what we ought to do, which itself is what we actually do in the noumenal realm.  So whenever we choose to perform some action other than that which we ought to perform, we are in fact not free but are imprisoned in the causality of the phenomenological realm, the realm of choices as they appear (i.e. not choices.) Either we do as the noumenal realm commands or we do what the phenomenal world orders, trapped in a prison with two cells and an open cell gate between them—and we are supposed to somehow feel comforted by the fact that one of the cells was a cell that we built for ourselves?

Kant claims that the moral law is what we will for ourselves, so that our autonomy is at the heart of our freedom. But he partitions the self into its determined and rational iterations, which leaves our noumenal reason void not merely of appearances, but of that which resembles the whole individual. He has given very good reasons for us to believe in free will, but at the cost of the individual—not merely the individual as a rational being, but as a human being.

Even if we accept Kant’s epistemic framework, we may still regard ourselves as co-mingled beings, beings that are whole only when considered both in terms of our appearances and our reason. We should be no less satisfied to have our noumena freed from our phenomena than if our heads were freed from our torsos. To say there is a faculty of the human mind that is free while the rest of her mind is determined is to make a proposition of no practical consequence to the individual as taken whole.

Furthermore, by introducing the two-point perspective, Kant has assigned freedom to our reason and determinism to our appearances, but he does not go into any satisfactory explanation of why exactly our motivation from reason sometimes or always fails to will moral law in the phenomenal realm. Without an adequate explanation, one is left with a gap between the noumenal and phenomenal perspectives of self, with one absolutely determined and the other absolutely free, but no sense of to what extent which principle is at play or where the choice resides as to whether to act from reason or else from inclination.

In the end, one must wonder what good Kant’s solution to the free will problem is when it steals from us our sense of empirical identity, dismissing such caused desires as mere appearances and then taking that which we can’t see and subjugating it to a rational prison of our own construction.

The Groundwork seems laced throughout with the faulty notion that once individuals understand every desire, inclination or need they have is determined; that they might somehow no longer find such attachments palatable in the face of free moral law. When the laws of freedom are only those laws that we would follow when we act from reason, and when more often than not we seem to understand ourselves as existing apart from such freedom, perhaps there is more to be said for what brings value to the human experience than what is merely free or rational.