Being Realistic about Kant’s Idealism (Translated by M. Rouba)

This paper deals with the question of whether Kant's transcendental idealism allows for an explanation of the a posteriori aspects of mental content by the properties of empirical objects. I first show that a phenomenalist interpretation has severe problems with assuming that we perceive an object as being red or as being cubical partly because the perceived object is red and cubical, and then present an interpretation that allows us to save the realistic intuition behind these claims. According to this interpretation, Kantian phenomenal properties are understood as response-dependent properties of extra-mental objects that also have to have some response-independent (in-itself-) properties. I show that this interpretation is well supported by Kant's remarks about the transcendental object in the A-edition of the first Critique and that it also makes intelligible why Kant took explanations of mental content by means of empirical properties to imply an explanation by means of noumenal properties without thereby violating his own doctrine of noumenal ignorance. This not only allows us to establish a realistic reading of Kant idealism but also to discern the true kernel in Adickes’ infamous talk about Kant's theory of double affection.