The Disappearing ‘We’

Abstract

There is a model of transcendental arguments with which we are all familiar, which I shall call anti-sceptical. A transcendental argument for a conclusion X, on the anti-sceptical model, proceeds by arguing that for a condition Y to be possible, X must be the case. Since the value of a transcendental argument is thought to consist in its ability to combat scepticism, Y should be a condition the sceptic must accept and X a condition he calls into doubt. Some transcendental arguments let Y be the condition of speaking a language and then argue that the sceptic's very ability to state his doubts about X show that X must be the case. The strongest form of transcendental argument is thought to let Y be self-conscious experience. For no interesting sceptic can deny that we have such a mental life; so if the transcendental argument is valid, it is thought, the sceptic is genuinely undermined. The paradigm of a transcendental argument is thought to be Kant's Transcendental Deduction of the categories. However, had Kant thought that the Transcendental Deduction merely showed that for experience to be possible it must conform to the categories, he would have considered his argument a failure.' Indeed, before he even mentions the need for a Transcendental Deduction, Kant has already argued that all our thinking must conform to the categories. Kant argues, in the Analytic of Concepts, chapter one, that every act of the understanding is a judgement and every judgement must employ its associated category. So if self-conscious experience involves any thinking, it will have to employ the categories. The Transcendental Deduction, by contrast, aims to show that we are entitled to employ the concepts which Kant has already argued we must employ in any thinking. It is, of course, possible to see that the Transcendental Deduction is concerned with the legitimation of