To: Manuel R.
From: Geoffrey Klempner
Subject: Locke's theory of ideas
Date: 10th April 2009 11:28

Dear Manuel,

Thank you for your email of 16 March, with your latest essay towards the Associate Award, entitled, 'On the Validity of Locke's Theory of Ideas.'

I apologize for the delay in responding, owing to my recent bereavement (see Philosophy Pathways Issue 142).

There are a lot of things about this essay which I like. You are clear about your objective, and you express yourself simply and clearly. A student following your exposition of Locke would be able to see exactly how you understand the main elements of Locke's theory of ideas, and what your criticisms are.

Also, it must be said, we are in agreement that there is something fundamentally wrong with Locke's theory.

However, we seem to disagree about what error Locke is making, and how this should be corrected. The fact that there is a disagreement is not a criticism in itself: as you note, commentators have argued over Locke ever since the publication of his essay. However, I would like to explore possible reasons for this disagreement, to see whether in fact this might be resolved.

For Locke, 'idea' is the name for something in the mind. From our modern perspective, this seems a rather casual approach, because it ignores the fundamental distinction between mental events and mental contents.

A mental event is individuated by its relation to other mental events and to objective time. My noticing that there is a red tomato on the table is a mental event. So is the thought that just came to me, how nice it would be to have a salad Nicoise for lunch. So is the memory which followed immediately afterwards, that I forgot to buy any eggs this week. Here we have three events, all occurring in the mind, which might plausibly be taken to form a causal series. That's what events do: they cause other events.

Let's get back to the tomato. I noticed that the tomato was ripe, because it was red, and therefore suitable for a salad. My recognition that the tomato is red involved a mental content, 'red'. A mental content copies or 'represents'. The mental content or 'idea' red is 'simple' because unanalysable into more basic elements. To 'possess' a simple idea in this sense is to have a recognitional capacity, the ability to judge or say 'red' whenever there is a mental event of perceiving something which is red.

But what exactly is a 'mental content'? You are right to contrast Locke's approach with the scholastic tradition. However, Locke would say in his defence that in his theory there is no need for the elaborate apparatus you describe. The concept of 'concept', at the fundamental level, is fully accounted for by the notion of a 'recognitional capacity'. Whenever a recognitional capacity is exercised, there is a mental event (the event of recognition) and a mental content (the concept involved in the exercise of that recognitional capacity).

So what would Locke make of the objection, 'which even a first-year student of philosophy can recognise', that his theory fails to 'bridge the gap' between sensation and the formation of universals?

Surely, Locke would say, the onus is on you to explain what is so difficult about the idea of mind as the capacity to make mental copies of received data, and use this copying capacity to recognize similarities and differences between different data. This is 'red' that is 'not-red'. If asked how the mind is able to do this, Locke would no doubt give a similar answer to the one about perception, which you cite in your essay. Receptivity (perception) and the exercise of recognitional capacities are what we start with; no more explanation can be given, or indeed can be demanded.

However, at this point we come to what I think is a serious difficulty for Locke. Your discussion of non-human animals illustrates this. A dog can recognize objects which it perceives, and act on the information so gained. What it cannot do is form thoughts about other times, or generalisations of the form, 'All X are Y.' Locke would explain the difference by saying that dogs do not have a 'power of understanding'. They can only react. Something vital is missing from their mental apparatus. But what can this be?

The answer a contemporary philosopher would give is that dogs do not have the capacity for language. It is only in language that thoughts about the past or the future, or about generalisations, can be framed. (For the classic text, see Jonathan Bennett 'Rationality' 1964.)

It is also on the question of language that I would frame my fundamental criticism of Locke's theory. On Locke's account, it would be perfectly possible for a solitary human being, cast away on a desert island at birth, to 'form ideas' of the things which he/she perceives -- yellow, blue, green, white, sand, sea, trees, white -- then using their capacity for understanding, form more complex thoughts. In Locke's universe, in fact, we are all such 'solitary' beings, using words for the common objects of our perceptions, while none of us can ever know how things are in the mind of any other language user. We call the same thing 'leaf' but I can never know whether or not the colour you and I describe as 'green' is the 'same' or different in our respective consciousnesses.

The classic text (which I may have mentioned before) is Wittgenstein's 'Philosophical Investigations', and in particular the paragraphs where he outlines his argument against the possibility of a 'private language'. In short: Locke's theory of ideas implies the possibility of a private language. This is a defect which can't be fixed. However, not all of the things that Locke says are affected by this criticism. That is why philosophers continue to debate Lockean themes today.

All the best,

Geoffrey