The Groundwork of the Metaphysics of Morals is a text in what we now call “normative ethics” as well as in what we now call “meta-ethics.” Kant’s compound aim in the text is to search for and to establish the supreme principle of morality. He recognizes that the latter part of this aim belongs to the metaphysics of ethics (or morals), or as we would say today, to metaethics.
Realism and anti-realism in ethics. In metaethics a “moral realist” is someone who thinks not only that morality has an objective authority, but that its authority depends on something that actually exists. An “anti-realist” in metaethics believes that the basis of ethics is not something out there, but something dependent in some way upon human thinking, sentiment or desire. Anti-realists run the risk—if they consider it a risk—that ethics might not always be the same. Human opinion and feeling, in general, might change, or vary from age to age, or culture to culture. Realists can usually count on the stability or universality of ethical principles, but they cannot easily explain the way morality seems to matter. J.L. Mackie, a 20th-century philosopher, is known for what is called the anti-realist argument from “queerness.” He assumed that moral features of conduct are matters of intrinsic concern to human beings, and he pointed out that real things are not matters of intrinsic concern. We might care about them, or we might not. We could answer every question we have about them from a scientific point of view, but it would be a separate question why anyone should care about them. If something real, existing independently of human thinking or feeling were the object of the kind of human concern prompted by moral properties of conduct or character, it would be very queer (Mackie 1977, 38). Simply by existing it would command everyone’s respect; and recognizing its existence would include caring about it. That is why it seemed to Mackie that moral realism is mistaken. If morality or its basis exists, as the realists say, then it would be very queer among existing things. Upon knowing that it exists and what it is like it would not be a separate question why anyone should care about it.
Existence as an end in itself. In the Groundwork Kant is clearly a moral realist. After explaining what a categorical imperative would command, if there is one, and after illustrating how to apply its command to derive a set of duties, he raises the question whether a categorical imperative exists. By this he means, whether the formula of universal law is a valid law for human conduct. He will not answer this question definitely, at least not yet, but he gives a reply something like this: if it is a valid law, then it must be based on something that exists. Kant understood exactly the problem for moral realism that Mackie would point to a few centuries later—the problem that would come to be referred to in terms of “queerness.” That is why he thought that the existing basis for the validity of the categorical imperative has to be something that is at least unique among existing things. So he proposed, as in the quotation above, that it would be something that exists as an end in itself, and for that reason has “absolute worth” (Kant 1997, 36/4:428).
Teleology and the order of ends. Kant has appealed to teleological ideas and ways of thinking several times already in the Groundwork. The initial argument about the end of practical reason rested on a teleological premise, as did the argument in the suicide example. The new concept introduced here, of an existing end in itself, comes from teleology. An end in natural teleology is that for the sake of which something else exists, or should exist. An end in itself is a special sort of end, since it is not something that exists, or should exist, for the sake of anything further. Traditional natural teleology, following Aristotle, posited ends or purposes for which natural objects exist, and toward which they adapt themselves in their growth, development, or regular movement. An object’s natural end or purpose was supposed to be what it exists in order to bring about. Kant’s notion of an existing end in itself is a little surprising, therefore; because as already existing it is not something to be brought about (see Kant 1997, 44-5/4:437). Still, the relationship between the existing end in itself and the (existing) means for sustaining its existence remains understandable. Here, also, the point about “queerness” arises. An existing thing—the end in itself—has a special, lawlike relation to other existing things, which are its means. The latter are “meant for” the former; the former is what the latter are “about” (see also Kant 2000, 297-303/5:429-36). The relation of each end in itself to every another can be described in a similar way: each is also a means for the existence of the others. As a whole they therefore belong to a community that Kant will later refer to as a “realm” or “kingdom” of ends (Kant 1997, 41/4:433).
Rational nature as an end in itself. If human beings and rational nature in general (whether human or not) exist as ends in themselves, then anything destructive of their existence as ends is contrary to their purpose. Using them merely as means, for other purposes, is therefore inconsistent with their status as existing ends in themselves. By applying the same point about “queerness” made just above, it can be seen now how certain real things would be objects of intrinsic concern to human beings. It would not be possible to understand these real entities fully without recognizing them as objects of respect, or of “moral importance.” They, and they alone, Kant claimed, could be the real basis of a valid categorical imperative. We know, of course, that members of the human species exist. But Kant will have to go beyond this commonplace observation to establish that humanity exists as an end in itself. In other words, in order to offer a satisfactory reply to the argument from queerness he must go beyond “supposing” that rational beings exist as ends in themselves, and find some way to convince his readers that they actually do. Although many readers are convinced of this already, there are plenty of skeptics. One pretty good line of argument he might give could be adapted from the earlier discussion of the end of practical reason. Assuming we are rational beings, capable of being influenced by reasons in our thinking about what to do, he can ask: what would be the point of our being capable of that? It does not make sense that our rationality is a means to some further purpose—as if nature has made us rational in order to bring about something else. Rationality would not be a very efficient means to something else: a natural instinct would be far more effective. Compare how effective our reproductive instincts are for perpetuating the human species. Imagine that we did not have those desires and associated passions. Imagine that nature gave us only reasons to procreate, instead. The human race would surely not have lasted more than a few generations. The point is that we are not rational beings in order to bring about some further purpose(s). From an efficiency standpoint that would not make sense. Therefore, our rationality, our existence as rational beings, seems to be an end in itself. This is not exactly the argument Kant will eventually offer; but it is compelling nevertheless.