In his article ‘Words Don’t Mean, Minds Do!’ in this issue of THINK, Terence Moore argues that words, in and of themselves, are intrinsically meaningless. On Moore’s view, meanings are something located somewhere else – somewhere necessarily inaccessible to others. Meanings reside in our private minds. As Moore puts it: ‘The words that do the triggering are public. The meanings we create for those words are unavoidably private and mobile.’

Moore is here endorsing a view about meaning that’s associated particularly with the philosopher John Locke.

In this short response to Moore’s article, I am going to suggest that Locke gets a great deal wrong about meaning, and that the later work of Wittgenstein is particularly helpful in revealing what Locke gets wrong. In particular, I am going to argue that we should reject the view that meaning is located in some mysterious private realm.

Of course there’s something right in what Locke has to say about meaning. Words really are intrinsically lifeless and inert. These squiggles you see on the page before you have no meaning in and of themselves. And of course it’s also true that we – beings with minds, beings capable of using and understanding them – are what give such sounds and squiggles meaning. It’s also true that we sometimes use terms with different meanings, resulting in misunderstanding. I can happily sign up to all this.

However, Locke has much more to say about meaning. According to Locke, the mind is a sort of inner world. At first the mind is entirely empty – Locke famously calls it a ‘tabula rasa’ or blank slate. It’s gradually furnished with what Locke calls Ideas (I use a capital ‘I’ to indicate I am using the word in this special, Lockean way). For example, I might look at a poppy and so come to have a visual impression - an Idea of red. Or I might touch some ice and thereby come to have another sensory experience - an Idea of cold. In this way, my initially empty mind comes to be filled with a rich and varied collection of Ideas.

On Locke’s view, these Ideas are necessarily inaccessible to others. Of course, two subjects might have the same kind of Idea. We might both have an experience of the exact same shade of red, say. But because there are two of us subjects, so there are two Ideas. Just as you and I might own cars that are indistinguishable locked away in our respective garages, neverthess the number of cars is two. Similarly, though our ideas may be exactly similar, they are not (to use the jargon) numerically identical – the number of Ideas experienced is still two. No one else can enter my mind and have this particular experience of red along with me.

According to Locke, words obtain their meaning by standing for Ideas. He says:

The meaning of the word ‘red’, for example, is just the Idea of red. Locke says about the Ideas that humankind possesses that:

So how do we learn the meanings of words, on Locke’s view? We learn the meanings by being given the relevant experiences and told the word, so that we come to correlate them. For example, I might show you a poppy, causing you to have an Idea of red, and say ‘That colour is red’. The connection between the public word and its private meaning is thereby forged.

To summarize: on Locke’s view, words are public, while Ideas, and thus meanings, are private. We can never access the mind of another to check whether the Ideas they associate with a word are the same as those that we associate with a word. On Locke’s view, for example, the colour experiences I have might be inverted compared to yours (a bit like looking at a colour negative). In which case, the Idea I associate with ‘red’ is the Idea you associate with ‘green’, and vice versa. On Locke’s view, this is a possibility that cannot be ruled out. We may use words with quite different meanings and so misunderstand each other. This sort of misunderstanding was a concern for Locke. He thought that much philosophical confusion was caused by our not using our public signs with the same private meanings. If we associate different meaning with our words, then we may end up arguing at cross purposes.

Moore suggests that the way we create meanings for ourselves is by ‘filtering public words through little-understood private processes in our own Minds’ (my italics). To what sort of mysterious private processes is Moore here referring?

On Locke’s view, the reason why I mean something when I say ‘It’s hot today’ whereas a parrot, uttering the same public words, does not, lies in what’s going on privately, behind the scenes. My use of those words is accompanied by an inner process – an inner parade of private Ideas (the Ideas I associate with the words ‘hot’ and ‘today’). While the parrot might utter the same public words, it lacks this private inner accompaniment, and so means nothing by what it says.

On Locke’s view, understanding a word also involves a private process. How, on Locke’s view, am I able to understand what ‘red’ means? For example, how am I to know which flower to pick if someone asks me to pick the ‘red’ one? Locke’s explanation is that I have previously associated the word ‘red’ with a certain Idea, which I now store in my memory (my memory being a vast storehouse of Ideas). On hearing the word ‘red’, I pull out the right Idea, which I can then use as a template or sample, comparing the flowers I see with it until I get a match. You may not be conscious of having engaged in such a private looking up process, but perhaps that’s because the process is so quick and habitual that you no longer pay it any attention.

To summarise again: on Locke’s view, meanings are private accompaniments to the public use of language. Meanings are Ideas – and because our Ideas are necessarily inaccessible to others, so are our meanings.

On Wittgenstein’s view, the picture of meaning and understanding sketched out above is fundamentally confused. He says:

Wittgenstein attacks this picture of private meanings in a number of ways. Perhaps the best-known way is via his famous Private Language Argument. Notice that, on the above picture, we all speak a ‘private language’ in the sense that we all use words to refer to what, necessarily, can only be known to us, the speaker. Wittgenstein famously argues against the possibility of such a language in section 258 of his Philosophical Investigations. I previously outlined Wittgenstein’s argument in ‘A Short Introduction to Wittgenstein’s Private Language Argument’ (THINK 69). If interested, you can read that later.

What I am going to do here is point to just one of the various other arguments Wittgenstein offers against this tempting Lockean picture of language as involving private meanings.

One way on which we determine what words mean is by reference to samples. A paint manufacturer may provide charts with samples of all the many different colour paints they produce. If I want to know which of several different rooms is painted with the colour they call ‘Appalachia’, I can pull out my paint chart, look up the colour sample for ‘Appalachia’, and then go through the rooms comparing them with my sample until I get a match.

As we saw, Locke seems to think we do something similar when asked to pick a red flower. I go to my storehouse of Ideas, find the Idea I previously associated with the word ‘red’, and then compare the flowers with that Idea until I get a match.

But notice a crucial difference between these two explanations of how I’m able to pick out the right colour in response to a word. A paint chart is something objective – something that continues to exist even while I’m not experiencing it. I can write ‘Appalachia’ on a piece of coloured card, put it in a drawer, and then pull it out again later when I need it. But now consider Ideas. They are subjective, not objective. In particular, they exist only while they are experienced. They’re like pains in this respect – just as you can’t have an unfelt pain, so you can’t have an unexperienced Idea. But this entails that an Idea is not something that can be labelled and filed away for future reference, as Locke suggests. If an Idea were something objective – like a piece of card - there would be no problem labelling it and putting it into storage. But it’s not objective. Nothing gets ‘filed away’. Rather, it vanishes.

So, later, when you want to know what ‘red’ means, how are you going to pull up the right Idea or memory sample: the one you earlier labelled ‘red’? You will need to be able to identify, among all the previous experiences you seem to have had, the red one. But identifying which one is red is the very ability this internal looking up process is supposed to explain. You were asked to pick a red flower. It’s suggested you do this by comparing the flowers to an inner object – a red Idea. But now you need to know which inner object is the red one. We have gone round in a circle. We have not explained how you are able to identify things as red; rather, we have presupposed that ability.

So how are you able to identify something as being red, be it a flower, an Idea, or anything else? It’s tempting to say – well, you just can do that because you remember what ‘red’ means. But if that’s what we say about your ability to identify the red Idea, why don’t we just say that about your ability to identify the red flower? What’s the point of introducing all the private machinery? As Wittgenstein points out, when asked to pick a red flower, you don’t usually compare it to anything:

If, in order to imagine – conjure up an Idea of – a red patch, you need first to conjure up an Idea of a red patch, then you’ll need to conjure up a red patch to conjure up the second Idea of a red patch, and so on ad infinitum. The Lockean explanation generates a vicious regress.

As I say, what’s happened here is that we have attempted to explain how you are able to understand a meaningful sign only by presupposing that you already have that ability. What Locke offers is merely the illusion of an explanation of how we are able to apply signs correctly.

Incidentally, it’s worth stressing that Wittgenstein doesn’t deny that there’s stuff happening in our brains that causally underpins our public use of language. Of course there is. A scientific investigation might even make those physical goings-on public. What Wittgenstein rejects is the view that the essence of meaning and understanding resides in some necessarily private domain.

True, I use ‘hot’ meaningfully and the parrot does not. So in what does the difference between me and the parrot consist, then, if not in something essentially private? Roughly speaking, on Wittgenstein’s view, it consists in what we are able to do. For example, if asked, I can explain what the expression ‘happy’ means. I can point to examples. I can use the expression appropriately. I can also use these words to construct many other different sentences. Parrots, on the other hand, can do none of these things.

The revolution in thinking about meaning brought about by Wittgenstein’s later work lies in this shift in focus from what goes on privately ‘inside’ to our publicly observable abilities. Meaning isn’t ‘hidden’. It lies on the surface, in the use to which we put our words and other signs. On Wittgenstein’s view, to grasp the meaning of a word is not to have correlated it with some mysterious inner object, but, roughly speaking, to know how it’s used.

None of this is to say that Terence Moore is not entirely correct to remind us that words are intrinsically meaningless, and that we can end up using the same word with quite different meanings, resulting in considerable confusion. Moore’s advice to keep tabs on such potential confusions is good advice. What I have rejected here is the Lockean view that the meanings of public words are necessarily private and inaccessible to others, as they would be if Locke’s views about meaning were actually correct. I suggest, along with Wittgenstein, that the meaning of public language is essentially public, not private.