Reconstruction from recollection and the refutation of idealism: A Kantian theme in theAufbau

Abstract

We know from Carnap's autobiography that he was strongly influenced in his early writings by Kant's transcendental idealism of space and time, especially in the former's dissertation on space. By the time he wrote the Au?au, however, Carnap tells us that his use of "the neutral language of symbolic logic" had helped him realize that various sides of traditional philosophical controversies, such as realism and idealism, "were merely modes of speech, not formulations of positions".1 Upon reformulation of a symbolic definition of a philosophical idea as a rule of operation for a constructive procedure, says Carnap, it became "applicable by anybody, be it Kant's transcendental subject or a com puting machine".2 This does not mean, however, that Kant's problems no longer had a place in the Au?au, or even that Kant's own solutions to them cannot be fruitfully compared with his more constructive ap proach to them. On the contrary, Carnap himself provided interesting leads for such a comparison, most notably in Section 83 on interpreting "the basic relations as categories", though he says it may be omitted. But the bulk of the literature on the Kantian background of Carnap's early work up through the Au?au has focused rather on the neo Kantian influence exerted on him by his own contemporaries.3 The purpose of this paper is to lay out a direct comparison of the respective approaches and solutions of the Critique and the Au?au to problems centered on what Kant called the refutation of idealism, which involve in both cases underlying problems about time.4 My claim is that one cannot only recognize a treatment of these problems in the Au?au, but one can also see in it clear signs of philosophical progress when measured against Kant's treatment. I also argue that the significance of some of the Au?au's claims for Kant's problem was obscured by the fact that Carnap had not yet clearly distinguished implicit from explicit definitions.