The Transcendental Turn: The Foundations of Kant's Idealism (review)

494 JOURNAL OF THE HISTORY OF PHILOSOPHY 96:3 JULY 1988 Moltke S. Gram. The Transcendental Turn: The Foundations ofKant's Idealism. Gainesville: University Presses of Florida, 1985 . Pp. xii + ~6o. $~5.oo. This was the last book published by Professor Gram before his premature death. I wish I could write that in it he had left behind a noble memorial. Unfortunately, I cannot. The Transcendental Turn is a tendentious book. Gram postulates that there is a welldefined conception of a relation of affection as contrasted to causation in Kant. He then interprets this distinction as intended to ground the distinction between appearances and things in themselves, and he criticizes what he classifies as the two traditional interpretations of this distinction for their failure to observe the difference between affection and causation. But the supposition that the distinction between appearances and things in themselves is intended to follow from an antecedently established distinction between affection and causation surely perverts Kant's own order of inference, in which the latter distinction is clearly introduced in an (ill-starred) attempt to accommodate the natural and traditional causal theory of perception to an antecedent denial of spatial and temporal properties to things as they really are, which denial is founded on independent arguments. And this failure to understand Kant's own grounds for his transcendental idealism then undermines Gram's own interpretation of the distinction between appearances and things in themselves, for this interpretation--which Gram can only distinguish from one of the two main traditional interpretations by radically misrepresenting the latter--is utterly incompatible with the key premise of Kant's most fundamental argument for the non-spatiality and non-temporality of things as they are in themselves. Gram begins by distinguishing two traditions in the interpretation of transcendental idealism, which he dubs the "Two Worlds Theory" (TWT) and the "Two Descriptions Theory" (TDT). TWT is the view that things in themselves and appearances are numerically distinct objects, the former lacking the spatio-temporal characteristics by which things are presented to our forms of intuition and the latter possessing them; TDT is the view that the concepts of appearances and things in themselves connote two different sorts of descriptions of objects, the former referring to the characteristics that objects actually present to our sort of sensibility in this world and the latter referring instead to characteristics that objects might present to other forms of intuition in other, possible worlds. The main argument of the book, which occupies the first two chapters ("Double Affection" and "Things in Themselves (I)") is then that neither of these interpretations is compatible with the distinction between affection and causation which is supposed to be the premise for the distinction between appearances and things in themselves. The notion of affection is supposed to be defined by the "Affection Condition"--that a thing in itself must somehow "generate an appearance by standing in some relation to the forms under which we apprehend appearances" (4u)-and the "Cognitivity Condition" (which would, however, be more helpfully called the noncognitivity condition)--the requirement that the objects satisfying the affection condition not be presented themselves by the sensory awareness which they effect in us. Unfortunately, Gram entirely omits any explanation of why Kant would adopt this (non-) cognitivity condition unless he had already argued for the non-spatiality and BOOK REVIEWS 495 non-temporality of things in themselves. But this does not deter him as he goes on to argue that TWT must be rejected because its numerically distinct appearances and things in themselves cannot satisfy the affection condition, while TDT, since it really just replicates ordinary causal relations between objects and perceptions in other possible worlds, cannot satisfy the (non-) cognitivity condition. In each case, the flaw in the interpretation is supposed to derive from a failure to distinguish between affection and causation: TWT rejects causation between things in themselves and appearances but recognizes no alternative to satisfy the affection condition (i.e., affection ); and TDT, instead of explaining affection as an alternative to causation, simply replicates causation in the actual world with further instances of causation in possible worlds. Further chapters of the book then trace out the problematic effects of...