- Possible Experience: Understanding Kant's Critique of Pure Reason
This is an important book. It deals with an issue key to any comprehensive interpretation of Kant's first Critique, namelythe problem of idealism. Collins argues, contra many commentators, that Kant was not an idealist, and that interpreted properly, many of the passages that suggest Kant was either an idealist or torn between realism and idealism show that his philosophy is consistently anti-idealist. As such, the book provides a much-needed revisionist exposition of the problem. The book is succinct and not overwrought with a scholarly apparatus. It is divided into fifteen chapters. Some of them are quite short, but each is devoted to a separate issue having an impact on the question of Kant's idealism. The author is quite knowledgeable and sheds a good deal of light on difficult issues in the first Critique.
Collins first provides a clear account of what Kant understood as idealism. For Kant the idealism of Berkeley—absolute idealism—and the idealism of Locke and Hume—problematic idealism, were inextricably hooked up with Cartesian assumptions, namely, that we have direct access only to the contents of our minds. If this is accepted, then it follows that either all there is are these ideas (absolute idealism) or that the inference from our ideas to the outside world is problematic (problematic idealism). Collins argues persuasively that it was Kant's objective in the first Critique to repudiate this Cartesian conception of consciousness. He also argues that Kant was largely successful in realizing this goal.
Collins provides a close analysis of problematic concepts, arguments, and passages in the first Critique that have been thought to prove Kant's idealism, beginning by discussing the difference between subjectivism and idealism, and pointing out that for Kant, all aspects of our experience are affected by our constitutional make-up. Our experience is therefore always subjective, but subjectivism does not entail idealism. That my experiences will always be colored by constitutional features allowing me to be receptive to experience does not entail that what I have access to is inner and mental. In order to sustain the force of this distinction, Collins details the many problems the concept of a representation presents for a non-idealistic reading of Kant. Not least of these are Kant's assertion in the subjective deduction (A99) that representations are "modifications of the mind," and that appearances are "only a play of representations," that in the end come down to a "determination of inner sense" (A101). Having thus set out the problem by picking out texts most lending themselves to idealist readings, Collins sets out to reinterpret these passages in light of non-Cartesian assumptions that will yield a non-idealist Kant. For instance, he defends Kant's claim at 104A that "an object must be thought of only as something in general = X, since we have nothing outside our knowledge which we could set against it as corresponding to it." He argues that we cannot compare a representation with that which it represents in order to see if that representation is adequate to it or not, since having representations is what constitutes our only access to empirical objects. Moreover, while representations are mental items constituting our access to objects, strictly speaking it is not they we are conscious of but of outer objects through them. [End Page 130]
Collins does a good job of distinguishing Kant's project concerning the "how possible?" questions from foundationalist epistemologies that begin with indubitable propositions based on direct access to the contents of the mind. Other strengths of the book include a lengthy discussion of the differences between outer and inner sense in which Collins shows that our access to objects of inner sense is not perceptual. For Kant the form of intuition exclusively reserved for what is given to us through perception is that of outer sense. Since material must be given to us through sensibility in order for us to have knowledge of empirical objects, it makes little sense...