Saturday, 29 October 2011

Meaning is Use Part I: Argument

While writing “Wittgenstein's Toolkit” it became clear to me that I have unfinished business with the notion of meaning as use. Several aspects of it still puzzle me. First off, I’m intrigued and slightly uneasy about the flow of Wittgenstein’s argument on this subject. In the course of the opening 43 sections his position (or, at least, his prima facie position) gradually appears to shift. I suspect this is intentional, but it can lead to a lot of head-scratching – indeed, each time I reach the end of §43 I wonder whether Wittgenstein really is claiming that meaning is use at all. I think we’d better consult the text.

The first mention of use comes in §1 when, after the “five red apples” scenario, the interlocutor asks: “But what is the meaning of the word ‘five’?” And the reply comes: “No such thing was in question here, only how the word ‘five’ is used.” Considered literally, this suggests that “meaning” and “use” have nothing to do with each other. Of course, Wittgenstein is being disingenuous; he’s preventing us (via the interlocutor) from dashing down a blind ally, and – for reasons yet to be explained – keeping us focused on use. Still, it’s a long way from “use, not meaning” to “use is meaning”.

That journey begins at §5 when he explains the role of his primitive language-games:

[….] the general concept of the meaning of a word surrounds the working of language with a haze which makes clear vision impossible. – It disperses the fog if we study the phenomena of language in primitive kinds of use in which one can clearly survey the purpose and functioning of the words.

Here, then, we have a connection in terms of helpfulness: when considering meaning it is helpful (demystifying) to look at use. Obviously, there’s still no claim that meaning is use, but the two have, so to speak, inched closer together.

Next, at §§8-9, Wittgenstein outlines an extended version of the builder’s game (introduced in §2) involving number-words, colour samples and indexicals. This prompts the following question: “Now what do the words of this language signify?” (§10). “Signify” is a slippery word in this context. For example, is it synonymous with “mean” or “indicate” or “denote” or all three? And is there a nod here to its technical use in semiotics? But such questions aside, there seems to be a clear link with the interlocutor’s question in §1. They both suggest an underlying impulse to correlate a word with a thing in order to explain meaning. This time, however, Wittgenstein’s reply is markedly different: “How is what they signify supposed to come out other than in the kind of use they have?” Here, it seems, the connection has shifted again: use is not merely helpful when considering meaning, it is essential.

The reply is provocative because it’s natural to assume that, when grasping signification, an object is what we really need rather than an account of use. To put it in Russellian terms, we want acquaintance but instead we’re being offered description as a more pertinent alternative. Actually, this is the start of Wittgenstein’s discussion of ostensive definition – a discussion which will rumble on intermittently over the next 30 or so sections. I am not going to address its specific points in this post, but it is important to note that a) his attack on the philosophical account of ostensive definition is an attack on the link between object and meaning, and b) it is indicative of Wittgenstein’s general approach that his substantive arguments in this area are almost all negative. He seems less concerned with proving that use has the right of succession than with killing off its rivals to the crown.

The importance of use is next suggested in §§19-20 during a discussion of how the command “Slab!” might be meant either as a single word or an elliptical version of “bring me a slab”. Here Wittgenstein is glancing at a second pretender to the throne: meaning as a mental object or activity. The temptation is to think that the key difference between the two ways someone might mean “Slab!” lies in “something different going on in him when he pronounces it” (§20). Wittgenstein argues against this, and at the end of the section suggests the strongest connection so far between meaning and use: “doesn’t their having the same sense consist in their having the same use?” As with “signify”, “sense” ought to be approached with caution; it doesn’t necessarily mean “meaning”. Assuming even a rough equivalence, however, we are confronted with the idea that use is not just an essential part of considering meaning; meaning consists in use. This is getting mighty close to saying that meaning is use. Note, however, that neither the comment at §10 nor the one at §20 are presented categorically. They are framed as questions. The onus is put on us to decide (and this, of course, is Wittgenstein’s therapeutic method in action). Nevertheless, it’s becoming progressively harder to miss the direction he wants us to go in.

So far meaning has been considered in three different guises: straightforwardly as “meaning” itself, as “signifying” and as “sense”. Each time, use has been implicated in the process, and each time its role has become more central: use helps clarify meaning (§5); only use shows what words signify (§10); and sense consists in use (§20). At §37, however, Wittgenstein turns his attention to naming (although this is prefigured at §15), and at §39 he introduces a pronoun in the guise of Siegfried’s magic sword “Nothung” (NB: previous translations of the Investigations substituted “Excalibur”). Here, it seems, we are tackling the word/object connection in its strongest possible manifestation. Surely, no matter how things function elsewhere, the meaning of a pronoun is the object it names? And this, of course, is precisely what Wittgenstein denies.

The discussion hinges upon making clear the hardness of the “is” involved in the statement “the object is the meaning”. This is an identity-claim, so it’s not like “grass is green” but like “Johnny Rotten is John Lydon”. (Note how “green is grass” is nonsense, but “John Lydon is Johnny Rotten” is not. Likewise, the nouns in “the object is the meaning” are interchangeable.) Wittgenstein’s task, then, is to force us to confront the full implications of this claim. To this end he points out that if the object (or person) is the meaning of the pronoun then the word must cease to have meaning if the object is destroyed (or dies) because the meaning itself is thereby destroyed. Yet clearly this doesn’t (usually) happen. Why not? In answering this, he first considers (§41) a situation where it would be right to say that the object’s destruction renders the word meaningless. If a named tool in the builder’s game is broken then it could be said that the symbol for that tool is meaningless, but only because the symbol no longer achieves anything. It has no use because no convention has been established about what to do in such a case. As soon as a convention is put in place (eg, builder B shakes his head whenever he’s asked for a broken object) then the symbol continues to function. It has a meaning because it has a use. In the same way, for us names have meaning in phrases such as "Jones is dead" because we have a whole series of linguistic conventions relating to talking about the deceased. In fact, such is the primacy of use over object as the grounds for meaning that it is even possible to imagine a meaningful name that signifies no object whatsoever (see §42). All the name needs is a role in the language-game. This, I think, amounts to a devastating attack on the attempt to equate “meaning” with “object”. The final stronghold has been breached.

During §§1-42 the object/meaning correlation is doggedly tracked through various manifestations, backed into a corner and exposed as empty. Surely now (we think) Wittgenstein can stop skirting round the issue and categorically assert that meaning is use? Instead, we get this:

For a large class of cases of the employment of the word “meaning” – though not for all – this word can be explained in this way: the meaning of a word is its use in the language.

And the meaning of a name is sometimes explained by pointing to its bearer.
Philosophical Investigations §43

Taken by itself, “the meaning of a word is its use in the language” is exactly the categorical statement we were hoping for. Unfortunately, it comes at the end of a passage which seems expressly designed to undercut the potency of its message and perhaps mangle it beyond recognition. Why, for example, has Wittgenstein suddenly moved from what meaning is (or isn’t) to how the word “meaning” is explained? What is the implication of the caveat “though not for all”? If some explanations of meaning don’t involve use, is it because the object is the meaning in such cases? But that would directly contradict the arguments outlined above! So if the explanation doesn’t involve use and the meaning isn’t the object then what the hell is meaning in such situations? And, anyway, which situations are we actually talking about? Which explanations of meaning fall under “though not for all”? And, while we’re listing our grievances, what are we supposed to make of the second paragraph of §43? It seems to fly in the face of the important distinction Wittgenstein makes at §40 between the meaning of a name and its bearer. Surely we don’t explain meaning by pointing at the bearer; we explain who the bearer is?

Once we’ve resisted the urge to throw our copy of the Investigations across the room, we might gather our litany of complaints into one central question: in what sense (if any) does Wittgenstein actually say that meaning is use? He spends a lot of time presenting powerful arguments to show what meaning isn’t, and all the while nudges us closer and closer to the idea that use is the only thing that fits the bill. But then, at the crucial moment, he seems to pull back. Or, rather, he both does and he doesn’t. He both says the words and retracts them at the same time. So instead of a straightforward declaration we get a caveat within a caveat that leaves us with more questions than answers.

How are we to make sense of this? I think the key lies in Wittgenstein’s descriptive approach to philosophy. Before we get to that, however, we need to consider an issue I side-stepped earlier: the role of ostensive definition. And that will be the subject of the next post.#

Meaning is Use Part II


  1. brilliant post, but what's with the new cliffhanging blogging style? And there is even a hint that you will only reveal your answers in the post after next. What is this EastEnders? :-)

  2. I know, I know - I nearly ended it with the words "same bat-time, same bat-channel".

    I never like the posts to go on for too long - mine are usually about 4 pages in Word and that's already fairly lengthy. And sometimes I'm conscious of compromising on depth in order to keep things tolerably brief. So I thought I'd have a go at tackling meaning/use over several posts to see if that provided the best of both worlds.

    At the moment I'm trying to understand the sense (if any) in which ostensive definition gives the use of a word. It's making my head hurt.

    1. Once you realize that the way to account for language is not to construct an inventory of objects that the words of language supposedly stand for, but to look at how the words of language work in human contexts (anthropologically, so to say), then it follows that the meanings of words will be explained by their contexts of use. Hence "meaning is use." It is a move from theoretical mentalism to observational holism. (Such holism you find it Goethe, except that Wittgenstein has no use for Goethe's noumenon [if the picture of Wittgenstein's decapitating Goethe's holism and applying _that_ to language to create a wonderful new kind of account of meaning is repellent, don't use it].)

  3. Hi petruchio - yes, that, of course is the broad thrust of the argument. Or, at least, it is usually represented in that sort of way. But if you look at the text itself things are not quite so straightforward. The Devil, as they say, is in the details.

  4. To point at the bearer of a name is to explain the name's use.

  5. Yes, given the right context for the action. We'll get to that in the next post!