Transcendental arguments and the concept of self-refutation

Abstract

The overall aim of this thesis is to find out whether any version of the transcendental method could be made to yield conclusions which are both factual and indubitably certain. In other words, could any sense be given to the notion of an a priori-synthetic principle, and if so, could such a principle be arrived at by means of a transcendental deduction? An affirmative answer is given to both these questions. Part I. “The search for a transcendental proof “. I try to show that although Kant recognizes at least two kinds of transcendental argument - transcendental deductions and transcendental expositions - only transcendental deductions are capable of giving rise to principles which are, in some sense, a priori-synthetic. I defend the concept of an a priori-synthetic principle against the objections of radical pragmatists (like Quine) on the one hand, and traditional empiricists (like Russell) on the other. I criticize Kant's reasons for thinking that transcendental deductions, as he conceives them, logically guarantee the truth of their (synthetic) conclusions, and I explain how the structure of his arguments would have to be changed in order to achieve this. Main conclusion: The initial premise of a transcendental deduction must be self-guaranteeing in the sense that any attempt to deny it would be self-refuting. Part II. 'The concept of self-refutation. I examine a wide range of self-refuting statements, and produce ac comprehensive system of classification. This is designed to show why some forms of self-refutation can be used as-the basis of a transcendental deduction, but not others.I point out that some of the things which have been said about self-refuting statements would seem to suggest that transcendental deductions based on any form of self-refutation are impossible, but that these suggestions are unfounded. Main conclusion: A. transcendental deduction may be described as an argument of the transcendental form in which the initial premise is operationally self-guaranteeing. Part III. 'The composition of a transcendental deduction. I consider what other requirements an argument must satisfy in order to count as a transcendental deduction. I argue that it should be based on a Cartesian-style premise and should be presuppositional rather than implicational in form. This leads to a comparison of the transcendental and Cartesian methods. I explain that the only way in which ultimate principles or axioms can be justified, on other than pragmatic grounds, is by some form of transcendental deduction. Main conclusion: Our revised notion of a transcendental deduction has much in common with Kant's original conception, although his claim that this method of argument is entirely different from any hitherto conceived' is not altogether justified. Show more