In the Critique of Fure Reason Kant distinguishes two sorts of conditions of knowledge. First, there are the space and time of pure intuition, introduced in the Transcendental Aesthetic. They are grounded in our dependence on a special sort of perceptual (or imaginative) field for the location of objects. Second, there are pure concepts of the understanding, or categories, introduced in the Analytic. In one respect these are grounded in the logical function of the understanding in judgements, introduced in the first chapter of the analytic of concepts: Clue to the Discovery of All Pure Concepts of the Understanding; in another respect, they are grounded in transcendental unity of apperception, which is introduced in the second chapter: Deduction of the Pure Concepts of the Understanding, or Transcendental Deduction. I shall be concerned with the latter and its contribution to the overall strategy of argument in the analytic. Within the Analytic, Kant distinguishes between an Analytic of Concepts and an Analytic of Principles (also called ‘transcendental doctrine of judgement’ [A137/B176]). This corresponds to a traditional distinction between a doctrine of concepts (or understanding) and judgement. It is arguable that Kant's theory of concepts undermines this distinction. However, I shall not deal with that general issue, but with a more specific issue related to the first two chapters of the Analytic of Principles: the Schematism and the System of Principles. (The third chapter, on phenomena and noumena, is basically an appendix.)

At the end of §40 of the Critique of Judgement, after a discussion of the sensus communis and its connection with taste, Kant writes:

If we could assume that the mere universal communicability as such of our feeling must already carry with it an interest for us (something we are, however, not justified in inferring from the character of a merely reflective power of judgment), then we could explain how it is that we require from everyone as a duty, as it were (gleichsam), the feeling in a judgment of taste. (5: 296; 162)

Kant argues that we must regard the will from two mutually exclusive standpoints. One is the natural standpoint, according to which the will is determined entirely by natural causes in conformity with natural law. The other is the standpoint of freedom, according to which the will transcends the laws of nature and is free to determine itself in conformity with its own law. Kant's idea is that a complete account of the will necessarily involves both standpoints, shifting between the two. Thus it is sometimes called the two-standpoints account of the will.

Unlike that of most liberal thinkers, Kant's theory of punishment is unabashedly retributive. For classical liberals punishment is justified only by the harms it can prevent, not by any allegedly intrinsic good served by making the guilty suffer. Here Hobbes' blunt insistence that the aim of punishment ‘is not a revenge, but terror’ is prototypical in substance, if not in style. Hobbes, Locke, Mill, Bentham and Beccaria, for all their differences, agree that punishment must look to future good rather than to avenging past wrongs. This attitude on the part of classic liberalism toward retribution is not surprising, given its association with the kind of theocratic politics liberalism arose to combat.

Henry Allison over the years has produced important work on Kant's transcendental idealism of the objects of our empirical knowledge as well as on Kant's conception and defence of the freedom of rational agency. He has done so both in two major books (Kant's Transcendental Idealism, 1983; and Kant's Theory of Freedom, 1990) and in a string of articles. Most recently, his continuing refinements of a number of interrelated themes prominent in his two books, together with discussions of some other issues, have provided the material for the publication of a collection of papers, Idealism and Freedom) Ten of the twelve essays included in this collection have already appeared elsewhere, but it is gratifying to have them, together with two additions, all available in their present form. Allison's debates, both with himself and with his critics, help increase our understanding of his views and hence of Kant's as well.

There are two familiar strategic approaches to Kant's Critique of Judgement which commentators have not always found easy to combine. One would regard the work as fitting snugly into Kant's enterprise as the keystone that absorbs the forces of his theoretical and practical philosophies, uniting them and itself into a single sound structure. That Kant saw it this way is obvious from his Introduction to the Critique. But the other approach has sometimes seemed more fruitful: start with the Analytic of the Beautiful and take it not as the completion but as the beginning of something, a treatise which, in offering possibly the fullest and most rigorous account of the autonomy of aesthetic judgement, plays a foundational role in the discipline now known as aesthetics. On this approach questions concerning morality and the unity of Kant's philosophy can be set aside for attention later, if ever.