In the previous lecture, I surveyed the question that Wittgenstein posed concerning the possibility of a logically private language – that is, a language that no one but the speaker can, logically, understand. I tried to make clear that, although the question seems, at first glance, to be bizarre and of no obvious interest at all, it is in fact exceedingly important. For a large part of the philosophical thinking of the past three and a half centuries, unwittingly presupposed that the languages we all speak are private languages. The words of our language, it was thought, refer to our own immediate private experiences – which are something that only we can have and that only we can know. And I pointed out that mainstream ideas of contemporary cognitive neuroscience, psychology and theoretical linguistics are likewise tacitly committed to this conception of language and experience. The idea of a logically private language can be represented as a three-legged stool. The three legs are:

First, the experience is logically privately owned: Only I can have my experience. I cannot have your experience and you cannot have mine. Experiences are logically inalienable possessions.

Second, the experiences are privately known. Only I really know of my experiences. For only I have access to them by means of introspection. I have privileged access to my own experiences. But others can judge of my experience only on the basis of my behaviour.

Third, the mental mechanism whereby words are linked to the experiences which they name is private ostensive definition.

This is the three-legged stool of a private language. If the legs are sound, the stool will stand. If so, then it follows that the language each of us speaks is a logically private one. That will face us with the intolerable paradox that no one else can possibly understand us. For the words of the language are defined by reference to samples that no one else can apprehend.

Finally, I drew attention to the characteristic attempt to find a way out of this paradox by invoking the distinction between numerical and qualitative identity. You will remember that this distinction between two uses of the expression ‘the same’ or ‘the identical’ applies clearly to things, such as chairs or cars.