Summary
Kant begins the final section by examining more closely the relationship between the freedom of the will and its autonomy. In the natural, perceivable world, he says, everything is subject to causation. But the will is beyond the natural world, and so it determines itself.
This definition of freedom is negative—it is freedom from causation, and so it is not hugely helpful. Freedom from the law of causation would hardly be a basis for morality, because we have defined morality as acting in accordance with a law. Natural necessity is just the heteronomy of natural causes (I ate a tuna sandwich, so my stomach hurts, so I feel too lazy to help my brother move into his apartment.) The solution to this problem, as Kant showed in the last section, is for the will to act as though it is a law unto itself; that way, the will can act in accordance with a law and still be free. This, of course, is nothing other than the categorical imperative.
Therefore, if we suppose that the will is free, then the concept of morality—acting in accordance with a self-legislated, universal law—necessarily follows. Freedom is the concept in which the absolute goodness and the lawfulness of the will can be reconciled.
We also have to presuppose that freedom is possible for all rational beings, whatever else these may be, and not just humans, because morality applies to us only insofar as we are rational. All rational beings are necessarily free, because to act rationally in a practical respect means to act in accordance with the laws of reason, which reason can only give to itself. For reason to act other than according to the dictates of reason would be for it to act irrationally. Thus, to act rationally and to legislate universal laws for oneself—that is, to act freely—are one and the same.
That does not necessarily prove that we are such beings, and that we must be moral. Again, why should I be moral? I can’t say that I have an interest in being moral, because that would cut against the categorical imperative. For purely rational beings, the “ought” of the categorical imperative would be automatically fulfilled, since the ought comes from that being itself. But for divided beings, like us, who have both a rational and a physical side, the "ought" remains, simply, an ought.
So, in the idea of freedom, we have only found the moral law again, but we can’t prove its objective necessity. But we still can’t answer the question of why we have to follow the categorical imperative. We can say that we accept, as human beings, our worthiness to be happy, without at the same time taking the desire to be happy as the binding condition of our action.
One possible way forward is to think of ourselves as divided beings, with one foot in each of two different worlds. Beyond the perceivable world, we are free, our own efficient causes, because we are beyond the chain of causality which applies onto the perceivable world (since it is a category that the mind applies to visual perceptions). The one is the world of sense; the other the world of understanding. Reason, Kant argues, is the means by which we distinguish the one from the other; it is pure self-activity. In the one world, we are purely heteronomous, in the other world, purely autonomous.
Kant resolves the two by pointing out, as he did in the Critique of Pure Reason, that these worlds are not exactly distinct. The world of understanding is the condition of possibility for the world of sense; it gives it order and shape, like a glass does to water. If we were purely rational beings, the categorical imperative would simply be a will—we would follow it all the time. But the ought, which tells us we should do something—but that we could not do it, because we have sensory inclinations as well—is the unique nature of the human. Nonetheless, because the sensory world seems to confirm that everything has a cause, freedom remains an idea of reason—something we can’t have sensory knowledge of, and therefore can’t truly know that we have.
This now leaves us with the problem of how a self-determined cause from the world of understanding (the categorical imperative) can cause something in the world of sense, without itself being caused, and thus no longer free. Kant concedes that he cannot solve this problem. He argues that he has successfully shown that the categorical imperative is the only basis for legitimately moral action. But he cannot show that we must act according to the categorical imperative without completely subjugating man’s existence in the world of understanding, which is free, to the world of appearances, which, because it functions on the principle of causation, means we always have an incentive to act one way or there other. We cannot deny that our mind poses us this ought. But when we consider it, we must experience awe at its ultimate incomprehensibility.
Analysis
Considering the boldness of the argumentation in the first two sections, it might come as a surprise to the reader that Kant closes the Groundwork with something of a shoulder shrug. His goal in the final section is to demonstrate why we must act morally. After sketching out the problem as he sees it, Kant concludes that we cannot legitimately come to this conclusion, and the Groundwork ends.
To give some perspective to this problem, we might glance back to the Critique of Pure Reason. In the Critique, Kant divided the human mind into the part of the mind that thought conceptually, which he called the understanding, and the part of the mind that absorbed sensory experience. We know things by combining the two forms of knowledge. While we can’t truly know anything beyond what we can sense, objective knowledge nonetheless exists, by applying categories of the understanding, like causality, to sensory impressions.
With this discovery, Kant set about tackling four “antinomies.” These were philosophical problems that couldn’t be answered empirically, and where logical thinking yielded two equally legitimate outcomes. One of these was the question of causation. Does everything in the world have a cause? Or are some things “free,” in the sense of causing themselves?
Kant’s answer to this question was that everything that we perceive does have a cause, because causation is a category that the human mind uses to link sensory impressions. But there are things, not in our experience, of which we are nonetheless aware. One of these is moral obligation. Moral obligation is hypothetical—we know that we ought to do something, and this knowledge doesn’t come from our senses. Therefore, moral obligation is “free” in the sense of being uncaused. We obligate ourselves.
Because this observation came towards the end of the Critique, Kant did not really have to deal with its implications. But here the problem becomes obvious—how can something in our mind affect a change in the sensory world, as moral action must, and not enter the machinery of nature, where everything has a cause? Then it would not be free, in the sense of self-caused, and cannot be moral. Kant, like many philosophers who split the mind and the body, cannot solve this problem, though he insists that the categorical imperative is still the only legitimate form for moral judgment.
Following the philosopher Pierre Hadot, one possible solution is simply to hold onto the categorical imperative as a “technique of the self,” a kind of thought experiment that is intended to gradually bring about a change in disposition of character in the person who thinks it. Here, the categorical imperative would be like a set of goggles through which we could see ourselves and others as beings of dignity and worth, and question the extent to which the thought and authority of others was determining our thought. Kant’s own solution would be to turn to the realm of aesthetic experience, which, as he would argue, was an actual sensory experience of freedom.
