We can distinguish between those political philosophers who are concerned to carry the original Kantian project further, like Wolfgang Kersting, Otfried Höffe, Ernest Weinrib and Fernando Teson, and those contemporary political philosophers who have given up the original project but seek to draw inspiration from Kant's thinking. Two political philosophers who belong to this latter trend are Habermas and Rawls.

At the core of Kant's Groundwork of the Metaphysics of Morals lies his ‘derivation’ of the categorical imperative: his attempt to establish that, if there is a supreme principle of morality, then it is this imperative. Kant's argument for this claim is one of the most puzzling in his corpus. The received view, championed by Aune and Allison, is that there is a fundamental gap in the argument, which Kant elides by means of a simple but deadly confusion, thus robbing the argument of all validity. We will here contest the received view, as well as Korsgaard's alternative interpretation of the argument. In place of these positions we will offer a reconstruction of the derivation which reveals its coherence and force. We will show that it illuminates some interesting grounds for rejecting certain candidates, including a utilitarian principle, for status as the supreme principle of morality. While certainly not free of all defects, the argument will be shown to be far more powerful and interesting than it has commonly been held to be.

In this paper, I shall argue that the most moderate and balanced way to view Kant's transcendental philosophy is as a species of psychological investigation analogous to Hume's, but refounded on a doctrine of pure (a priori) sensibility, such as Hume never allowed himself (and may never even have thought of). This might seem to fly in the face of what many interpreters of Kant deem conventional wisdom: that the burden of proof is on one who claims that psychology is essential to transcendental philosophy. On this view, there is to be found in Kant ‘a more austere strictly transcendental philosophy’, which needs to be carefully distinguished from the psychological doctrines in which it is enmeshed; and they would insist on being convinced of the contrary before abandoning a position that, in their eyes, is the most moderate and balanced an interpreter of Kant can adopt. My purpose in this two-part essay is to urge them to think again. For while there can be no question of Kant's opposition to empiricism, it is equally certain that his praise for Hume was never freer or more unreserved than in respect of the latter's psycho-genetic approach to cognition. So, rather than supposing that Kant ipso facto rejected solutions to philosophical problems grounded on psychology when he rejected Hume's empiricism, it seems to me that the more moderate and balanced interpretive approach is to begin by supposing that Kant's transcendental philosophy is a species of philosophical psychology in the same mould as Hume's, differing from it only by virtue of involving a priori syntheses of a manifold of a priori intuition.

In the preface to the second edition of the Critique of Pure Reason Kant explicitly states that his motivation for writing this work is to make room for faith or the practical employment of reason (Bxxv, xxx). How does Kant accomplish this? The topics of God and the immortality of the soul do not arise until the conclusion of the antinomies. How does Kant get from the desire to make room for faith to its fulfilment in the latter parts of the first Critique? A common response to this question is a discussion of the constitutive and regulative employment of the ideas of reason. It is this distinction that sustains Kant's attempt at reconciling empirical knowledge and moral discourse. The constitutive and regulative analysis, however, has its roots deep within the initial stages of the Critique of Pure Reason. It is, in fact, the mathematical/dynamical distinction, which Kant introduces early in the analytic, that makes possible the constitutive/regulative distinction. Not only has the mathematical/ dynamical distinction itself been disregarded, but the relation between the mathematical/dynamical and the constitutive/ regulative has been almost universally ignored by commentators. If a commentator does mention the mathematical/dynamical distinction, it is usually in a dismissive tone. Walsh, for example, calls the distinction ‘hard to interpret’ and ignores it for the rest of his commentary.

In an interview with Philippe Nemo, Emmanuel Levinas makes a very revealing comment about what he was trying to accomplish in his ethical philosophy. In response to a question about the ‘starting-point’ of his ethics, Levinas protests: ‘My task does not consist in constructing ethics; I only try to find its meaning … One can without doubt construct an ethics in function of what I have just said [in describing his philosophy up to this point in the interview], but this is not my own theme’ (El 95–9/90).

One third of the transcendental dialectic in Kant's Critique of Pure Reason is devoted to demolishing the pseudo-science of rational psychology. In this part of his work Kant attacks the idea that there is an ultimate subject of experience — the ‘I’ or Self — which can only be investigated and understood intellectually. The belief that such a study is possible is natural to human reason; but it is based on demonstrable error. Kant tries to exorcize our minds from falling prey to this mistake.

One of the more striking developments in contemporary philosophic discussions about morality has been the rise of anti-theory — the rejection of moral theories as ‘unnecessary, undesirable, and/or impossible’. Among those associated with this view have been Bernard Williams, John McDowell, Edmund Pincoffs and James Wallace.