Introduction to Kant's Practical Philosophy, Part 1

This is part one of a two-part series. In this part I discuss practical philosophy in general and introduce Kant's practical philosophy in relation to it. In part 2 I discuss Kant's moral philosophy in light of the issues discussed in the first part.

Kant's moral philosophy has come to occupy a strange place in philosophical pedagogy. It's the area of his thought most commonly taught in intro philosophy courses and most commonly known outside of academia, but in these contexts it is almost always represented independently of Kant's theoretical philosophy, and even of his broader practical philosophy. This is understandable, because Kant is a difficult thinker, and his moral philosophy is probably the easiest area of his thought for a new student of his to get a grip on. Nevertheless, the detachment of his moral philosophy from the rest of his thought is regrettable, as it leads to a superficial and inaccurate understanding of his moral philosophy, which Kant himself saw as forming part of an organic whole along with his theoretical thought.

That said, I won't attempt the gargantuan task of investigating that unity here. What I will try to do is provide an introduction to Kant's practical philosophy that is accessible to those who have never studied him before, without being distorted or superficial. In particular, I would like to present it in a way that is relatively faithful to Kant's own arguments, so that readers can evaluate it rationally rather than by appeal to “intuition,” which is the prevailing standard in most contemporary Anglophone philosophy, where it is put to uses, including in the evaluation of Kant's moral philosophy, which Kant himself explicitly and insightfully condemned.

Kant's practical philosophy evolved throughout his career, but its first mature statement came in the Groundwork of the Metaphysics of Morals, published in 1785. This text is divided into three sections, corresponding to a division which Kant made use of throughout his work. In the first two sections Kant makes use of what is called the analytic method to draw out the implications of what he calls “common moral cognition.” In other words, he analyzes commonly held moral concepts in order to understand their structure and their relations to other concepts, without making any claims about whether there is anything in the world which instantiates these concepts. Therefore, these first two sections don't make essential use of any substantive knowledge about the world. Rather, they depend only on knowledge of commonly held concepts. Kant's goal in these sections is to make explicit and to clarify the moral philosophy which he believes everyone already holds, albeit in a confused and inchoate form.

In the third section Kant turns to the synthetic method, which means that he introduces substantive claims about the world—claims which, unlike analytic claims, are true not merely by definition—in order to justify the common moral philosophy he has just elucidated. His goal here is to rule out the possibility, left open at the end of the first two sections, that common moral beliefs—or indeed, any moral beliefs at all—are mere “chimeras of the brain,” that is, concepts without objects or delusions. Moral error theorists, also called moral skeptics or nihilists, insist that this is the case, but by the end of section three Kant believes he has proven them wrong.

Throughout all three sections, both analytic and synthetic, Kant argues a priori, that is, without drawing on propositions that can only be known as a result of experience. This is easy enough to see with regard to the analytic sections. You might think that Kant must implicitly be invoking experience of people's moral beliefs in order to justify his claim to be investigating “common moral cognition,” but he sees himself as analyzing concepts that all persons possess simply in virtue of being persons, rather than making a generalization based on experience. What may seem harder to understand is how the third, synthetic section can also be a priori. How can a substantive claim about the world whose truth depends on something more than the definitions of the terms involved by knowable independently of experience? This is one of the central questions of Kant's philosophy, but it is not one that he addresses explicitly in his practical philosophy, so we won't go into it here. If you want to learn more about Kant's justification of synthetic a priori knowledge, take a look at my introduction to Kant here.

Now that we understand the basic structure of the Groundwork, let's take a look at the text itself. The best way to understand the text is not to follow its exact order, but rather to reconstruct Kant's argument from its foundations. I will, however, follow the analytic-synthetic order described above.

The first thing to understand is the basic framework within which Kant conducts his investigation. This is not an uncontroversial issue, but I believe that the best way to understand Kant's practical philosophy, as well as his philosophy as a whole, is to see it as presupposing that all human activity, theoretical as well as practical, philosophical as well as nonphilosophical, is carried out from an inescapable first-personal perspective. In other words, all human activity must begin with and remain oriented and constrained by the features and limitations of the human mind, considered from the perspective of a subject who (very roughly speaking) is such a mind. In asking any question, philosophical or otherwise, a person must start with “I”—what should I think?, what should I do?—and whatever answer he or she arrives at must be one that can be adopted from that perspective—I should think x, I should do y.

This may sound like a trivial thesis, but it is far more controversial than it may seem, and understanding it is essential to understanding Kant. As I have stated it, it is vague enough that it could probably be interpreted in a form that would make it acceptable to most contemporary philosophers, excepting only the most radical skeptics about subjectivity, that is, those who don't believe that subjects or persons exist in any sense at all. However, in the sense in which Kant holds it, it is a distinctly minority position. Most contemporary Anglophone philosophy presupposes an essentially third-personal perspective, and considers the first-personal perspective, if at all, only at the end of inquiry, when it comes time to issue prescriptions—one should believe x, one should do y. What's at issue is the application of a conclusion arrived at third-personally to the demand by a subject or group of subjects for action-guiding advice. Even in this context the judgement is only trivially first-personal, as I have suggested by using “one” rather than “I.” It is not essentially guided or constrained by any feature of the subject considered first-personally; rather, it is mechanically derived from a conclusion which takes no account of the first-personal perspective.

To see how this works in practice, consider one standard argument for utilitarianism, which is the moral theory that holds that only well-being is intrinsically good (good in itself rather than as a means to an end), and that the fundamental moral imperative is to maximize well-being. This argument invites us to consider our intuitions about the goodness of various things that people take to be good. Take money, for instance. People don't value money for itself, the argument claims, but only as a means to happiness or well-being. If acquiring more money didn't contribute to well-being, people would no longer want to do it. Therefore, money is only instrumentally good, that is, it is only good as a means to well-being. The same can be said for any other thing we value other than well-being. Therefore, well-being is the only good which is intrinsically valuable, and all other goods are good merely to the extent that they contribute to it. For that reason, one ought always to act so as to maximize well-being.

This is a simplified version of the argument, but it is good enough for our purposes. As you can see, it begins with a third-personal premise about the things that people value, then analyzes the way in which they value those things in order to reach the conclusion that they value them only as means to the end of well-being. The argument might be recast in terms of what “we” value or what the reader values, but this would be a merely verbal change, as the argument would still not depend on any features of the subject other than the structure of his or her desires, which is in principle just as accessible from a third-personal as from a first-personal perspective. Therefore, the argument is essentially third-personal, and only first-personal in the trivial sense that it results in an action-guiding conclusion which can be applied from a first-personal perspective.

As I have said, most of contemporary Anglophone philosophy is conducted in this manner, that is, from an essentially third-personal perspective, drawing on the intuitions of the reader or the philosophical community. The problem with this perspective is that it tends to lead to either dogmatism or skepticism, because it fails to address one of the fundamental problems of philosophy: the problem of normativity. I alluded to this problem above in discussing Kant's turn to the synthetic method in section III of the Groundwork, which was necessary in order to prove that morality isn't a mere “chimera of the brain.” In other words, Kant's challenge is to justify the belief that morality has legitimate authority over us. More generally, the problem of normativity is to determine whether not only morality but reason in general has any legitimate authority over us. The question is, do we have any reason to believe or do anything at all, and if so, why?

Put this way, the question may seem trivial. It may seem obvious that we have reason to do all sorts of things, including most basically whatever we want to do, assuming there is no countervailing reason not to do it. However, this response begs the question by assuming that desire itself is normative. This assumption is so common in both philosophy and common discourse as to be almost invisible, but it is not in any way obvious or exempt from the need for justification. The problem of normativity is not that of justifying the authority of any particular reason by showing how it can be reduced to some other reason-giving source, such as desire, but rather that of showing how any claim of reason whatsoever can be justified.

Still, it may be hard to understand why the problem of normativity is a genuine problem. Almost all of us believe that we have reason to do at least some things, and there's no obvious reason to believe that we're mistaken, so why should we doubt our commonsense belief in normativity? There are a variety of ways to motivate the problem, some of which presuppose some degree of commitment to Kantianism, but there is also a motivation that should be compelling to non-Kantians, at least those who endorse what has been called the “modern scientific worldview.”

The modern scientific worldview is a loose collection of views including the following: nontheism, or the view that gods do not exist or at least that we have no reason to believe in them; physicalism, or the view that everything that exists is physical or in some sense reducible to something physical; and naturalism, or the view that science is the paradigm method of knowledge acquisition, and its results are therefore generally to be believed and its methods imitated in other domains whenever appropriate. The modern scientific worldview or something like it is increasingly hegemonic among philosophers and nonphilosophers alike, and yet it does not seem, at least superficially, to leave any room for normativity. In ancient and medieval philosophy, god was often identified as the source of normativity, on the ground that the will of a supreme being is authoritative at least for finite beings. This view is called metaethical divine command theory (metaethical because it concerns the foundations of ethics or moral philosophy). However, the nontheism of the modern scientific worldview excludes divine command theory. If no gods exist, then of course they cannot be the source of normativity.

Other philosophers, from antiquity through the present, have thought that normativity arises from properties such as pleasure or pain, goodness or badness, or rightness or wrongness, which supposedly exist independently of any mind or subject at all, whether human or divine. This view is called metaethical realism. However, the modern scientific worldview leaves no room for realism, as the properties in question, with the possible exception of pleasure and pain, do not seem to be physical, and in any case science does not make essential reference to any of them. Psychology makes reference to pleasure and pain, of course, but not as normative properties. Nothing in the content of scientific theories supports the idea that there is any way the universe “ought” to be, and indeed it seems as though this is not the kind of belief that science could ever reach even in principle, since normative properties are not empirically confirmable, at least not in anything like the way in which other scientific properties are.

If neither divine command theory nor realism can justify belief in normativity, what can? A third theory which rose to prominence in the early modern era is called voluntarism, because it holds that normativity arises from human acts of will or volition. Unlike realism, this theory does not posit the existence of mind-independent normative properties, but rather locates the source of normativity in the will, a prima facie plausible place to locate it. The assumption which I mentioned above in motivating the problem of normativity—that desire is unproblematically normative—is a kind of voluntarism, and its pervasiveness testifies to the plausibility of the theory. What's more, unlike divine command theory, voluntarism does not posit the existence of any suprahuman or divine will, so it is compatible with nontheism.

As it is usually formulated, however, voluntarism has similar problems to those of the the desire-satisfaction theory already discussed, as well as more loosely to those that undermine realism. Although it may be plausible to suppose that the will is the source of normativity, this intuition does not suffice in place of an argument. Furthermore, voluntarism, again as it is usually formulated, does not really fare any better than realism when evaluated by physicalism and naturalism. For one thing, the will itself is not obviously a physical thing or an essential concept in any science, though there may be theories of the will which make it physicalistically and naturalistically acceptable.

More importantly, however, regardless of the status of the will itself, the normative properties which supposedly arise as a result of its acts are not any more acceptable to the modern scientific worldview than those which moral realism posits as existing mind independently. Even if it is more plausible in the eyes of common sense to locate such properties in the will, this does not render them any more empirically verifiable or otherwise acceptable to science, let alone necessary to it. Again, psychology and the other social sciences theorize the will in some sense—they describe the structure and content of people's desires and preferences, they investigate what people believe about rationally and morality and how these beliefs affect their actions—but all of this is a purely descriptive enterprise which makes no essential reference to the purported normativity of the phenomena in question. Most people certainly believe in normativity, including moral normativity, but from the perspective of science their beliefs may perfectly well be “chimeras of the brain.” This does not entail that normativity positively does not exist from the perspective of science, only that science provides no evidence for its existence, which leaves it without support from the perspective of the modern scientific worldview.

Things do not look good, then, for voluntarism as it is normally formulated. However, as you may have guessed, Kant offers a kind of voluntarism which is different from the normal formulations, precisely because it is formulated from the first-personal rather than the third-personal perspective. This is where the radical nature of the first-personal perspective emerges. Remember that in introducing this perspective I characterized it by saying that all enquiry conducted from the first-personal perspective must begin with and remain constrained by features of the human subject, considered from a first-person point of view. This means that no other perspective, including the modern scientific worldview, can supplant the first-personal perspective. The modern scientific worldview and other third-personal perspectives may perfectly well be added on to the first-personal perspective, but they can never become fundamental, and if they conflict with the first-personal perspective they cannot be endorsed, at least not in any simple or obvious manner.

Now, this does not mean that Kant is a theist in any traditional sense, and it certainly does not mean that he absolutely rejects the authority of science. It only means that he rejects the claims of science to be absolute or fundamental, that is, to supplant the first-personal perspective. Scientific investigation plays an essential role in Kant's philosophy, but like all other human activity, it can only be conducted from the first-personal perspective and in accordance with whatever commitments, beliefs, and norms this perspective entails. The investigation of just what these commitments, beliefs, and norms are will be the subject of the next post, which will lead us into Kant's moral philosophy proper.