... Kant produced a short book destined to become the main text for the study of his ethics, the Foundations of the Metaphysics of Morals (1785). The purpose of this work is 'the search for and establishment of the supreme principle of morality' (G 392/8). His plan was then to write a Metaphysics of Morals. In Kant's terminology, a 'critique' investigates the legitimacy of applying pure rational principles and their concepts to objects; while a 'metaphysics' sets forth those principles and their implications. The Third Section of the Foundations contains a deduction of the moral law and so is a critique of practical reason; at the time he wrote this work, he thought that that would be sufficient. Later he saw that the moral law could be validated in a different way, and the Critique of Practical Reason was the result. But we must turn to the Foundations and the Metaphysics of Morals to get the substance of Kant's ethics, for in the second Critique the problem of validating the moral law and showing how it fits into his system supplants Kant's interest in its formulation and application.

Kant's method in the First Section of the Foundations is analytic: he uses examples in order to analyze our ordinary conception of a good will and to arrive at a formulation of the principle on which such a will acts. A good will is easily distinguished from one that acts from an indirect inclination, doing the right thing merely as a means to some ulterior end, a 'selfish purpose'. The difficult thing is to distinguish a good will from a will that has a 'direct inclination' to do something that is (as it happens) right (G 397ff/13ff). For instance, there are people 'so sympathetically constituted that without any motive of vanity or selfishness they find an inner satisfaction in spreading joy, and rejoice in the contentment of others which they have made possible' (G 398/14). Having a natural inclination to do what coincides with duty is not the same thing as acting from duty, so for clarity we must contrast this case with one where the duty is done without natural inclination. Take someone whose mind is 'clouded by a sorrow of his own which extinguished all sympathy with the lot of others' or one who is 'by temperament cold and indifferent to the sufferings of others' (G 398/14). If such a person is nevertheless beneficent, it must be from a good will. What is the principle on which he or she acts? We see, first, that such a person does his or her duty just because it is his or her duty. Furthermore, we see that what makes him or her do it -- and so what makes it his or her duty -- is not simply its purpose. For the naturally sympathetic person and the unsympathetic but beneficent person both have the same purpose, helping others, although one has this purpose because of a direct inclination and the other has it from duty. Both are contrasted with the selfish man who does the right thing for an ulterior purpose, such as fear of punishment or hope of reward.

Duty, then, is not a matter of having certain purposes. If we remove all purposes -- all material -- from the will, what is left is the formal principle of the will. The formal principle of duty is just that it is duty -- that it is law. The essential character of law is universality. Therefore, the person who acts from duty attends to the universality of his/her principle. He or she only acts on a maxim that he or she could will to be a universal law (G 402/18). In this way Kant moves from the idea that a good will is one that acts from duty to a principle that can be used to tell us what our duties are.20,

In the Second Section Kant reaches the same point by another route: the investigation of rational action. 'Everything in nature works according to laws. Only a rational being has the capacity of acting according to the conception of laws, that is, according to principles' (G 412/29). The principle that you give to yourself, that you act on, Kant calls a 'maxim'. Your maxim must contain your reason for action: it must say what you are going to do, and why. If your maxim is one that it is rational to act on, it meets certain tests, commands of reason expressed in imperatives. Your action must be a means to your end, and (unless it is morally required) your end must be consistent with your happiness. These tests are embodied in the two kinds of hypothetical imperatives, those of skill and prudence. But there is also an imperative that tells us what we must do, regardless of our private purposes. This is the moral or categorical imperative, and because it is independent of all material, we know that 'there is nothing remaining in it except the universality of law as such to which the maxim of the action should conform' (G 421/39). So from the very idea of a categorical imperative we can tell that it says: 'Act only according to that maxim by which you can at the same time will that it should become a universal law' (G 421/39; C2 27126).

But how can you tell whether you are able to will your maxim as a universal law? On Kant's view, it is a matter of what you can will without contradiction. This is important, for it helps to secure the categorical character of the results -- any agent who applies the contradiction test should get the same result, regardless of his/her private interests. To determine whether you can will your maxim at the same time as its universalisation without contradiction, you envision trying to will your maxim in a world in which the maxim is universalised -- in which it is a law of nature. You are to 'Ask yourself whether, if the action which you propose should take place by a law of nature of which you yourself were a part, you could regard it as possible through your will' (C2 69/72). Contradiction may arise in two ways: if the maxim cannot even be conceived as a law of nature without contradiction, it is contrary to strict or perfect duty; if it can be conceived but could not be willed without contradiction, it is contrary to broad or imperfect duty (G 424/41 - 42).

The best example of the first sort of contradiction concerns a man whose maxim is to make a false promise in order to get some money, which he knows he will be unable to repay. To see whether this can be willed as a universal law, we imagine a world in which this is, so to speak, the standard procedure for getting ready money -- it is a law of nature that anyone who needs money tries to get it this way. Then we imagine the agent trying to will to act on his maxim in that world. Kant tells us that this gives rise to a contradiction because such universalisation would make 'the promise itself and the end to be accomplished by it impossible; no one would believe what was promised to him but would only laugh at any such assertion as vain pretense' (G 422/40). It is important to notice the sense in which this is a contradiction. Kant's view, as we saw earlier, is that hypothetical imperatives are analytic, because they express a relation of conceptual containment. The negation of an analytic statement is a contradiction. The man in the example derives his maxim from a hypothetical imperative: 'If you want some ready money, you ought to make a false promise'. This imperative is derived from a causal 'law' -- that false promising is a means to getting ready money -- combined with the analytic principle that whoever wills the end wills the means. The causal 'law' in question, however, turns out to be no law at all, because false promising could not be the universal method of getting ready money. The efficacy of a lying promise depends on the fact that it is exceptional, for people believe promises only because they are normally made in good faith, and lend money on the basis of them only because they believe them. In willing the universalisation of his maxim, the deceitful promiser wills a world in which promises of this kind are not normally in good faith and therefore will not be accepted. This means that they will not be a means to getting ready money, and that the hypothetical imperative from which the deceitful promiser derives his own maxim will be falsified. This is where we get the contradiction: the lying promiser who attempts to will the universalisation of his maxim wills the denial of the analytical principle on which he himself proposes to act, and the denial of an analytical principle is a contradiction. Later critics claim that undermining the efficacy of promises is only a contradiction if promises are themselves necessary. But Kant's point in the example is more modest than that; it is not intended to establish that promises are necessary. Promises are necessary for the man in the example, because he proposes to use a promise as the means to his own end. This is why Kant says that he cannot will his maxim and its universalisation At the same time. Whenever you propose to perform an action whose efficacy depends on its exceptional character, you get a contradiction of this kind.

The other kind of contradiction arises when you attempt to will the universalisation of some policy which would undermine the will's efficacy more generally. For instance, if you try to will a universal policy of neglecting talents and powers, you contradict your will because these serve you for 'all sorts of possible purposes' (G 423/41). If you try to will a universal policy of not helping others, you contradict your will because you yourself, as a finite rational being, are often in need of assistance. Kant is not offering an egoistic reason for an actual agreement here. Imagining yourself in a world without assistance is a thought experiment to determine whether you can will your maxim as a universal law. The duty of helping others holds even if you do not in fact get any assistance from anyone else, or have any real hope that you will.

... a new formulation of the imperative, the Formula of Humanity: 'Act so that you treat humanity, whether in your own person or in that of another, always as an end and never as a means only' (G 429/47).

Kant...treats the same set of examples he used earlier, showing how the immoral maxims involve a violation of the unconditional value of humanity. Violations of perfect duty occur when the power of rational choice definitive of humanity is made subordinate to other, merely conditional goods. A suicide, for instance, treats his/her own humanity as something he/she can throw away for the sake of his/ her comfort (G 429/47; MM 422­423/82­84). Anyone who uses deceptive or coercive methods to undermine the freedom of choice and action exercised by others also violates perfect duty. The Iying promiser uses the lender as a mere means because he tricks him into giving away his money rather than allowing him to choose whether or not to do so. He thus treats his having the money, a conditional good, as if it were more important than the other's humanity. Coercion (except to protect rights) and deception are unjustifiable no matter what end they serve, for a good end is an object of every rational will, and reason is 'just the verdict of free citizens' (C2 62/6465; C1 A739­740 B767/593).

Although humanity is not a purpose to be achieved, we can act in a way that expresses a positive value for it, and imperfect duty is violated when we do not. We ought to realise our humanity by developing our talents and powers, our rational capacities. We ought to acknowledge that others are sources of value by treating their chosen ends as good, and pursuing their happiness as they see it (MM 388/46). All human activities and pursuits are to be regarded as good as long as everyone can in principle agree to them. 'This principle of humanity and of every rational creature as an end in itself is the supreme limiting condition on the freedom of the actions of each man' (G 430­431/49). The same idea is implicit in the Formula of Universal Law: for your reason to be sufficient, it need only be universalisable. Adoption of humanity as the unconditional end leads to the conduct which the Formula of Universal Law prescribes.