Published online by Cambridge University Press: 25 February 2009
I have appropriated the terms ‘descriptive’ and ‘revisionary’ metaphysics from P.F. Strawson's Individuals. In the Introduction to that work he draws a broad general distinction between two types of metaphysics. Descriptive metaphysics is concerned to ‘describe the actual structure of our thought about the world’ while revisionary metaphysics is ‘concerned to produce a better structure’. They also differ in that revisionary metaphysics requires justification of some sort whereas descriptive metaphysics does not. Strawson makes this point when he says, ‘Revisionary metaphysics is at the service of descriptive metaphysics’. Thus, the descriptivist has the sober, scientific task of elucidating our extant conceptual schema while the revisionist has the speculative, slightly literary job of inventing a new conceptual schema.
page 108 note 1 The first question which seems to arise here concerns the meaning of the term ‘unanalysable’. On at least one standard account of analysability (G. E. Moore) there is an entailment relation between ‘X is unanalysable’ and ‘X is indefinable’. Thus, ‘Yellow is inanalysable’ entails ‘The word “Yellow” is indefinable’. This seems to be adequate for simple qualitites such as colours, as there does not seem to be any clear answer to such a question as, ‘What sort of thing is yellowness?’ It is difficult to imagine an answer to this question which would constitute an analysis of ‘yellowness’, and yet the same does no seem true of ‘persons’. Prima facie the word ‘person’ does seem definable, i.e. analysable, in a way in which the word ‘yellow’ does not. That is, one can think of a whole range of answers to the question, ‘What sort of thing is a person?’ whereas one cannot seems to think of any answers to the question, ‘What sort of thing is yellowness?’ The point here is that the sense in which ‘persons’ is said to be unanalysable is different, or seems different, from the sense in which any other concept is said to be unanalysable, and it is not clear whether a new concept of analysability is being introduced or whether it is just illegitimate to entertain answers to the question, ‘What sorts of things are persons?’ if such an answer is intended to be an analysis of ‘persons’. This is a crucial point because with ‘persons’ we are dealing with a concept which does in fact appear to be analysable, even at the level of ordinary language, into minds and bodies.
page 118 note 1 I wish to thank Mr I. M. Crombie for his many helpful discussions of the mind-body problem.
