Meaning is use

"Meaning is use" is the short form of a remark in Wittgenstein's Philosophical Investigations: for a large class of cases, the meaning of a word is its use in the language. To understand a word is not to have an object or an image in mind that it stands for. It is to know how to use it: when to say it, what follows from it, what people do with it.

It sounds almost too simple. But it overturns a picture of language that is very old and very natural, the picture in which every word is a name and its meaning is the thing it names.

Die Bedeutung eines Wortes ist sein Gebrauch in der Sprache.

The meaning of a word is its use in the language.

PI §43

In the animation above, the same word, "Water!", turns up in three scenes. A thirsty hiker says it, and it is a request. Someone shouts it as a friend is about to step into a flooded cellar, and it is a warning. A pupil says it in a chemistry lesson when asked what H₂O is, and it is a description. The sound is identical. What the word does is completely different each time. Then a toolbox opens: hammer, pliers, saw, screwdriver, ruler, glue pot, nails. That image comes straight from Wittgenstein.

Denk an die Werkzeuge in einem Werkzeugkasten: es ist da ein Hammer, eine Zange, eine Säge, ein Schraubenzieher, ein Maßstab, ein Leimtopf, Leim, Nägel und Schrauben.

Think of the tools in a toolbox: there is a hammer, pliers, a saw, a screwdriver, a ruler, a glue pot, glue, nails and screws.

PI §11

The functions of words, he goes on, are as diverse as the functions of these objects. We are misled because words look alike when we hear or read them, just as all the handles in a locomotive cab look alike, though one is a crank, one a switch, one a brake lever and one a pump handle (PI §12).

What view is Wittgenstein arguing against?

The Investigations open with a passage from Augustine's Confessions in which he describes learning to speak: grown-ups named objects and pointed at them, and he came to understand which words signified which things. Wittgenstein sees in it a whole picture of language: words name objects, sentences are combinations of names, and the meaning of a word is the object it stands for (PI §1).

That picture fits some words reasonably well, like "table" or "Mary". It fits "five", "red", "not", "hello" or "ouch" very badly. It was also, in a more refined form, close to the picture theory of his own Tractatus. Wittgenstein immediately gives a counterexample. Someone goes shopping with a slip that says "five red apples". The shopkeeper opens the drawer marked "apples", finds the colour red on a chart, and counts up to five, taking an apple at each number. Nowhere is the meaning of "five" an object. What matters is what the shopkeeper does with the word.

Meaning is not the bearer

A clear way to see the difference: when Mr Smith dies, we say that the bearer of the name has died, not that the meaning of the name has died. If the name had lost its meaning, the sentence "Mr Smith has died" would make no sense (PI §40). So the meaning of a word cannot simply be the thing it refers to.

That is why Wittgenstein adds a qualification in §43. He does not say that use is the meaning in all cases, and he notes that the meaning of a name is sometimes explained by pointing to its bearer. Pointing is fine. It just works only against the background of a practice in which we already know what kind of word is being explained.

How do you find out what a word means?

Wittgenstein's practical advice is short:

Wie ein Wort funktioniert, kann man nicht erraten. Man muß seine Anwendung ansehen und daraus lernen.

One cannot guess how a word functions. One has to look at its application and learn from that.

PI §340

Dictionaries are a kind of summary of use. But we learn most words the way children do: by taking part in language games where the word does some work. You learn "sorry" by apologising and being apologised to, not by being shown a sorry. And you learn the use of "know", "mean" or "understand" in exactly the same way, which is why philosophers who try to define them in the abstract so often go wrong.

Is it a theory of meaning?

Here interpreters disagree. Some philosophers, such as Paul Horwich, have developed a full "use theory of meaning" inspired by Wittgenstein. Others, such as Peter Hacker, insist that Wittgenstein was not offering a theory at all. He says explicitly that philosophy may not advance any theory and should replace explanation with description (PI §109). On that reading "meaning is use" is a reminder, not a thesis: when you are puzzled about meaning, stop looking for a hidden object and look at how the word is used.

Either way, the idea has had a long life. It fed Oxford ordinary language philosophy and J.L. Austin's speech act theory. A distant cousin of it appears in modern linguistics and computing, where the meaning of a word is modelled by the company it keeps in large amounts of text. Whether that is really what Wittgenstein meant is doubtful, since for him use means use in human activities, not just patterns of words. The Stanford Encyclopedia entry on theories of meaning places his view among the alternatives. For his philosophy as a whole, see the Stanford Encyclopedia of Philosophy and the Internet Encyclopedia of Philosophy.

A test you can run today

Pick a word that causes friction at work or at home, like "urgent", "done" or "fair". Instead of asking what it really means, collect three situations in which people actually use it. You will often find that two people are not disagreeing about facts at all. They are using the same word as two different tools.