Abstract

In his Refutation of Idealism, Kant attempts to show that what Cartesian idealism takes to be most certain is only possible on the condition of that which Cartesian idealism takes to be least certain . Despite widespread interest in this argument, most 20th century Anglophone commentators think it unsound. I provide textual and philosophical evidence that these criticisms depend chiefly on either of two common interpretive errors in determining the reference of the pivotal expression "something permanent in perception" employed in the second premise of this argument. Different groups of commentators coalesce around each error: ;Group I includes those who think the permanent in perception is a perceptual continuant, either permanent or relatively permanent , to which the mind has unmediated access through its faculty of sensibility whenever it is aware of its own empirical existence in time. I argue that for Kant, in fact, the mind's access to the PIP is indirect, mediated by outer appearances which are themselves immediate but transitory representations with a necessary, nonimmediate relation to that which is permanent in outer sense. ;Group II includes those who think the PIP is either 'an enduring framework of space' , or a pure formal intuition of space . These interpretations are incompatible with Kant's oft-repeated claim that space is a product of the mind's faculty of sensibility, is no less ideal than time, and cannot be an actual representation . The PIP is needed to provide what the form of inner sense cannot, i.e., an objective representation of time in the field of appearances. But neither a pure formal intuition of space nor an enduring framework of space can do this job because, according to the Transcendental Aesthetic, space is strictly subjective. ;These criticisms are byproducts of a new interpretation of the Refutation, the elaboration and defense of which is the central theme of my work. On this new interpretation, Kant's "something permanent in perception" is not any particular empirical object, nor a noumenal object, nor even a pure object of intuition, but rather what Kant means by 'Nature': a qualitative unity of dynamically interrelated material substances which, taken together, yield a real individual which can only be cognized indirectly. In contrast to other interpreters, I claim the PIP is a set of unified, necessary relationships among material substances inferred from sensations which together constitute objective time through acts of mind without which cognition and experience would be impossible. ;Whether the resulting Refutation of Idealism is sound is not settled in this work. In the last chapter I attempt to consider reasonable challenges and replies, and hope that in any case I have helped to clarify the setting within which this well-known argument is evaluated