To: Graham C.
From: Geoffrey Klempner
Subject: Locke's attack on the theory of innate ideas
Date: 19th january 2009 11:53

Dear Graham,

Thank you for your email of 23 December, with your essay for the University of London Modern Philosophy Descartes et. al. paper, in response to the question, ''Locke's attack on innate ideas and innate knowledge does not seriously damage any theory that a competent philosopher would wish to maintain.' Do you agree?'

I have taken you at your word and postponed my reply until this week. I hope you had an enjoyable time in Indonesia.

This is a knowledgeable and well-argued essay, which makes the case for a stronger conclusion from that stated in the examination question, namely, that there is no interesting or coherent thesis of innatism worth taking the trouble to argue for, or against. Your contemporary example of Chomsky's claims regarding innate knowledge of depth grammar, and your observation that the question 'how much' different human languages have in common is analogous to the question, 'How long is a piece of string?' leads to the strongly sceptical conclusion that philosophers who have argued the toss over innateness have basically been wasting their time.

Is that true? It wouldn't be the first time that an issue which seemed highly contentious dissolved under analysis in this way.

One thing that makes me pause before reaching this conclusion is the very strong empirical (though admittedly at present difficult to test) claim entailed by Chomsky's theory, that intelligent alien creatures who do not share our evolutionary heritage would not be capable of learning a human language (unless, by happy cosmic accident, they evolved the very same deep grammar on their home planet).

I have actually had the chance to argue with philosophers who have 'gone over' to Chomsky; their standard line is that *if you look at the empirical evidence* of the speed at which infants learn linguistic structures, no other explanation is possible. On the other side of the argument is the Davidsonian view that the only interesting 'structure' within human language is first-order predicate logic. The problem is, you can't engage with the argument at a general level: you have to go down into the details, something which the majority of philosophers working in the philosophy of logic and language have been loathe to do (I include myself).

One issue, then, is whether it is conceivable that there could be such alien creatures. I don't know of any argument that it is not conceivable (it is difficult to rule out a priori every possible way in which intelligent creatures might be prevented from learning a human language). On the other hand, suppose that it is conceivable: what does that show?

As you argue, the question of whether or not the 'disposition to form a concept' counts as 'having' that concept is no less obscure or fuzzy than the question to what extent all human languages have a common structure.

One thing that doesn't help (a point you could have made) is that Leibniz's analogy of Hercules 'in' the block of marble is seriously misleading in at least one respect: it treats concepts as analogous to pictures or representations. One starts of thinking that 'you either have the picture in your mind or you don't.' If you have the picture then you are able to recognize objects in the world which correspond to it. If you don't, then you have to look at the objects first and then form the concept. Then a third alternative is suggested: if you have the mere 'disposition' to form a picture, looking at corresponding objects 'triggers' the concept and makes it active. -- This makes it all sound utterly trivial.

Apart from the fact that concepts are not pictures, another relevant observation is that concepts are not employed or acquired individually, but in complex structures. This is a point made by Peter Geach in his attack on abstractionism (in 'Mental Acts' Routledge Library of Philosophy and Psychology). You can't have the concept of red unless you have the concepts of colour, reflective surface, light etc. An infant who has merely learned to say 'red' in more or less appropriate circumstances does not thereby prove that they possess the concept of red.

Yet another angle would be Kant's argument in the transcendental deduction regarding the a priori necessity for the concepts of substance and cause. This is a good an example as any of a concept which cannot be regarded as analogous to a picture (although, Kant confusingly talks of 'schemata' of the imagination in an attempt to explain something which doesn't really need explanation). Objective thought is impossible without causal thinking and the identification of spatiotemporal continuants: that's all Kant has to say.

One issue where, possibly, more argument is needed concerns Descartes' claim that he has an innate idea of God (from which he ingeniously derives an argument for the existence of an actual God who caused him to have the idea). Descartes is here appealing to anyone who thinks that they comprehend what is meant by, e.g. the 'infinity' of God's attributes. Descartes' argument does not make any claim about what human beings universally assent to: it is in a sense ad hominem, addressed to the reader alone. The continental philosopher Emmanuel Levinas has argued that my concept of 'the other' is, in a sense, analogous to this insofar as an information gained from experience could not be sufficient on its own to justify belief in the 'reality' of other minds.

This would be innateness with a vengeance. We cannot help but believe (apart from those unfortunate human beings who are classed as 'psychopaths') that other persons are real. Yet arguably there is no empirical evidence that one could point to which would justify that claim, rather than the lesser claim that the world is inhabited by creatures whose behaviour can be predicted and manipulated using a tool called 'language'.

All the best,

Geoffrey