In her new book, Patricia Kitcher supports Onora O'Neill's view that the categorical imperative is the highest principle of both practical and theoretical reason. I claim that neither O'Neill's original interpretation nor Kitcher's additional evidence in favour of it are convincing. At its core, this misconception of Kant's position consists in the identification of self-referential critique of reason with the concept of autonomy. It will be shown that the ‘common principle’ (Kant) of both practical and theoretical reason is not the categorical imperative, but the reflective power of judgement, as Kant claims in the Critique of the Power of Judgement.

I argue that Patricia Kitcher's Kant-inspired account of self-consciousness overintellectualizes the requirements for rational cognition. Kitcher claims that a person can only believe something on the ground of another belief if she is able to recognize the grounding belief as grounding the first belief and as one of her own. I criticize this claim by arguing that (i) someone can believe something for a certain reason without recognizing this reason as a reason (the possibility of unreflected reasons), and that (ii) she can recognize something as a reason for something else without being able to self-ascribe either her original belief or the belief that grounds it (the possibility of reflected but not self-conscious reasons).

In chapter 15 of Kant's Thinker, Patricia Kitcher claims that we can treat Kant as ‘our contemporary’, and that his theory of apperception ‘offers “new” and plausible perspectives on issues of considerable recent interest’. I question this with respect to two of her four chosen topics. First, I address her attempt to show that Kant's theory of apperceptive self-knowledge is immune to sceptical doubts of the sort Barry Stroud presents. Second, I turn to her argument that this theory is superior to current accounts of the special authority of self-knowledge. Over and above specific weaknesses, it seems that Kitcher's considerations generally lack sufficient reflection on how philosophical arguments of the past can be relevant to current agendas.

The article deals with Kant's theory of the self in Patricia Kitcher's Kant's Thinker in three respects: (1) I argue that it is doubtful whether accompanying representations with the ‘I think’ as such yields a principle for the categories since it does not require any strong kind of connection between them. (2) I discuss textual evidence for and against Kitcher's attempt to make sense of Kant's claim that the ‘I think’ requires the continued existence of cognizers per se. (3) I ask whether Kitcher's understanding of Kant's positive theory of the self leans towards minimal substantialism or towards functionalism.