KANT'S THEORY OF KNOWLEDGE KANT'S THEORY OF KNOWLEDGE Selected Papers from the Third International Kant Congress Edited by LEWIS WHITE BECK D. REIDEL PUBLISHING COMPANY DORDRECHT-HOLLAND / BOSTON-U .S.A. Library of Congress Catalog Card Number 74-83006 ISBN-13:978-90-277-0S29-7 e-ISBN -13:978-94-01 0-2294-1 DOl: 10.1007/978-94-010-2294-1 Published by D. Reidel Publishing Company, P.O. Box 17, Dordrecht, Holland Sold and distributed in the U.S.A., Canada, and Mexico by D. Reidel Publishing Company, Inc. 306 Dartmouth Street, Boston, Mass. 02116, U.S.A. All Rights Reserved Copyright © 1974 by D. Reidel Publishing Company, Dordrecht, Holland Softcover reprint of the hardcover 1s t edition 1974 No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher TABLE OF CONTENTS PREFACE VII PART I Aspects of Kant's Method in the Theory of Knowledge EVA SCHAPER / Are Transcendental Deductions Impossible? 3 MARTIN J. SCOTT-TAGGART / The Ptolemaic Counter- Revolution 12 PART II Linguistic and Transcendental Themes K.-O. APEL / From Kant to Peirce: The Semiotical Transformation of Transcendental Logic 23 JUSTUS HARTNACK / B 132 Revisited 38 JOSEF SIMON / Phenomena and Noumena: On the Use and Meaning of the Categories 45 PART III Analytic and Synthetic Judgments RICHARD AQUILA / Concepts, Objects and the Analytic in Kant 55 KONRAD CRAMER / Non-Pure Synthetic A Priori Judgments in the Critique of Pure Reason 62 GERHARD KNAUSS / Extensional and Intensional Interpretation of Synthetic Propositions A Priori 71 R. M. MAR TIN / On Kant, Frege, Analyticity and the Theory of Reference 77 PART IV Space IVOR LECLERC / The Meaning of 'Space' in Kant 87 VI T ABLE OF CONTENTS ROBER T PALTER / Absolute Space and Absolute Motion in Kant's Critical Philosophy 95 ROBER TO TORRETTI / On the Subjectivity of Objective Space 111 PART V Causality and the Laws of Nature HENR Y E. ALLISON / Transcendental Affinity - Kant's Answer to Hume 119 GERD BUCHDAHL / The Conception of Lawlikeness in Kant's Philosophy of Science 128 RALPH C. S. WALKER / The Status of Kant's Theory of Matter 151 PART VI The Thing in Itself PETER KRAUSSER / Kant's Theory of the Structure of Empirical Scientific Inquiry and Two Implied Postulates Regarding Things in Themselves 159 RALF MEERBOTE / The Unknowability of Things in Themselves 166 NICHOLAS RESCHER / Noumenal Causality 175 PART VII Kant and Some Modern Critics J. N. FINDLA Y / Kant and Anglo-Saxon Criticism 187 MARGARET D. WILSON / On Kant and the Refutation of Subjectivism 208 PREFACE The Third International Kant Congress met in Rochester, New York, March 30 to April 4, 1970. The Proceedings, published by D. Reidel Publishing Company in 1972, contained 76 complete papers and 30 ab stracts in three languages. Since this large volume covered many phases of Kant's philosophy from a wide variety of standpoints, it is unlikely that the entire contents of it will be of interest to anyone philosopher. I have therefore selected from that volume the 20 papers that seem to me to be most likely to be of interest to English-speaking philosophers who are, to use a fairly vague description, in the 'analytical tradition'. The topics treated here are those which are most relevant to current philosoph ical debate in the theory of knowledge, philosophy of mind, and the philosophy of science. The division of papers under the seven principal topics, however, is in some respects a little arbitrary. I hope this little volume, published 250 years after Kant's birth, will show philosophers who are not already convinced that Kant is one of the most contemporary of the great philosophers of the past. I believe that the efforts of the authors of the papers will show that there can be genuine Kantian contributions towards the solution of problems that have fre quently been handled in opposition to, or obliviousness of, the eighteenth century philosopher who did more than anyone else to formulate the problems which still worry philosophers in the analytic tradition. LEWIS WHITE BECK PART I ASPECTS OF KANT'S METHOD IN THE THEORY OF KNOWLEDGE EVA SCHAPER ARE TRANSCENDENTAL DEDUCTIONS IMPOSSIBLE? Professor Kornerl has recently argued very persuasively that transcen dental deductions are logically impossible. If this were so, the detail of Kant's Deduction need no longer occupy us. Such optimism I believe to be misplaced. Transcendental arguments exhibit the necessary presuppositions with out which something we say, or want to be able to say, cannot be said at all. Such arguments include, but need not be restricted to, arguments eliciting the preconditions of conceptualizing experience in the way it is conceptualized by us, the necessary preconditions of empirical inquiry as we understand it. This formulation may not seem very close to Kant's definition of the term 'transcendental', but I am assuming what I cannot argue in detail now that Kant's central question "How are synthetic a priori propositions possible?" requires at least an answer to the question "What are the necessary conditions (if any) of our being able to speak intelligibly about the world of our experience?", and that it is this re quirement which gives Kant's question its primary significance. When Kant insists, e.g., that the notion of describing experience presupposes distinctions which are prior to any specific descriptive statement, he is arguing transcendentally; in particular, when he distinguishes two basic kinds of prior conditions, space-time and the categories, he is claiming that a conceptual scheme adequate to the requirements of empirical knowledge must enable us to do at least two things: to individuate and to attribute. If all transcendental arguments exhibited the conditions of concept ualizing experience as it is in fact conceptualized by us, they might be thought to be relatively unproblematic. Some of Kant's transcendental arguments, however, make much stronger claims than those just outlined. These arguments deal not merely with the conditions of making the empirical claims we do make, but rather with the conditions of making any intelligible claim at all about any conceivable kind of experience. Kant argued for this stronger claim in what he considered his most 4 EVA SCHAPER important transcendental argument, the Transcendental Deduction. A transcendental deduction, then, is an argument which shows, or purports to show, not only what the necessary features of a conceptual scheme are which underpin (,makes possible') a given structure of experience, but also that the conceptual scheme thus exhibited is based on specific principles without which we could not think coherently about experience at all. It is this larger claim which is problematic in a way the weaker transcendental arguments are not. Korner expresses something like the distinction I have sketched in this way: a transcendental deduction is "a logically sound demonstration of the reasons why a particular categorial scheme is not only in fact, but also necessarily employed, in differentiating a region of experience" (318- 319). A deduction must then satisfy two conditions. (1) It must show that a categorial scheme is "established," or can be established, i.e. that it has or can have application, and (2) that it is unique. According to Korner the first condition can be met, the second cannot logically be met. It follows that transcendental deductions, as Korner defines them, are impossible. Now though Korner's distinction between "establishing a scheme" and "proving its uniqueness" reflects the distinction I drew, it nevertheless leads to a misrepresentation of the relation between the two claims in Kant's case, and this gives a distorted view of what can be salvaged from the Critique if, according to Korner, we disallow transcen dental deductions. But first I want to ask whether Korner, given his own definition, does in fact prove that transcendental deductions are impos sible. If we are to entertain the possibility of the uniqueness of a particular categorial scheme we must, of course, at least be able to ascertain that it has application, or, as Korner puts it, we must be able to establish the scheme. Statements about (a region of) experience presuppose that we have the means of differentiating within that experience between objects and their properties and relations. This is to have what Korner calls a "method of differentiation." Such a method belongs to a categorial scheme if and only if among the concepts exhibited by the scheme are, firstly, some which are constitutive of it, i.e. tell us what is to count as an object of experience. (This is Korner's condition of comprehensive appli cability.) And, secondly, some which are individuating for those objects, i.e. tell us the criteria by which in general one object is to be distinguished ARE TRANSCENDENTAL DEDUCTIONS IMPOSSIBLE? 5 from another. (This is Korner's condition of exhaustive individuation.) A scheme is then said to be established when it has been shown that a method of prior differentiation belongs to it. In terms of these distinctions, in order to demonstrate the uniqueness of a categorial scheme, we would have to show that every way of differen tiating experience belongs to the scheme for which we are claiming unique ness. It is this which Korner says is impossible. His strategy is to show that all the abstractly possible methods of proving uniqueness must logically fail. In fact, and rightly ignoring any vague appeals to "other methods, e.g. some mystical insight or some special logic" (321), he thinks there can only be three such methods: (1) A comparison of the scheme with undifferentiated experience. (2) A comparison of the scheme with possible competitors. (3) An examination of the internal constitution of the scheme. The first method is, of course, not one which Kant would have thought possible, nor one which could in general plausibly suggest itself. For quite apart from the question of uniqueness, it requires that we should be able to think first of a pure experience untouched by any differentiation, and no sense can be attached to such a supposition. What I want to point out, however, is that it follows from Korner's definition of a categorial scheme and how it is established that this method is incoherent. For we cannot show uniqueness without establishing the scheme, and we cannot do that unless some method of prior differentiation belongs to it. Now this is a feature of Korner's discussion of the second method of proof also. It follows immediately from his description of the second method - com paring a scheme with possible competitors and finding the latter wanting that this method is also bound to fail because it is self-contradictory. It requires anyone who uses it to admit before the argument begins that what he is trying to prove is false, otherwise he could not even try to prove it. The reason is simply that a possible competitor is a scheme which could be established by some method of differentiation and which is yet not identical with the scheme whose uniqueness is supposed to be demonstrable. This argument prejudges the issue, however, and in a way which makes suspect any attempt to show uniqueness of something on grounds which involve the consideration of other claimants - a procedure which we do not normally regard as vicious. It is true that transcendental deductions