St. Augustine describes the process of learning language as associating names with objects. This picture of language suggests that every word has a meaning and that sentences are sets of names. Such a picture of language disregards the different kinds of words. For instance, imagine a language-game where someone writes "five red apples," and shows it to a grocer. The grocer makes different uses of each of the three words in filling out the order: "apple" directs him to a particular fruit, "red" leads him compare the fruit with a color, and "five" tells him how many fruits to count out. We haven't explained how the grocer knows that "red" is a color or what the meaning of "five" is; we have simply described how these words are used. It is not yet clea r what such explanations would look like.
Augustine's picture of language may seem adequate for a simpler language than our own. We can imagine, for instance, a language between a builder and his assistant that consists only of the words "block!" "pillar!" "slab!" and "beam!": the builder shouts one of these orders and his assistant fetches the object he names. Children learn by ostensive teaching what these different words name and are trained to obey and give orders. We should not say that the ostensive teaching creates an understanding of the words in question, because it can only be said to do so in the wider context of the training: teaching and understanding cannot be examined in isolation.
In section 8, Wittgenstein considers a complication of this language-game, introducing number-words, color-words, and "this" and "there." We can teach these words by pointing and speaking, but the manner of teaching will be very different depending on th e kind of word being taught.
We may be tempted to generalize about all words, since they all look alike, but such generalizations are useless. We might say, "every word in language signifies something," but what we say they signify depends on what sort of distinction we are making. W e might say "one" signifies a number if we want to distinguish it from "slab!" or we might say it signifies the first cardinal number if there is some confusion as to the order of number-words. All questions of what a word signifies come back to how that word is used, and words are used in unlike ways.
These language-games are not incomplete. They simply have a smaller scope than ordinary language. Every language embodies a particular form of life that is complete in itself.
Wittgenstein asks if "slab!" in this language means what it does in ours. We can shout the order "slab!" but we might say in our language this means, "bring me a slab," whereas the words "bring me a" do not exist in the builder's language. However, how d o we know that "slab!" means, "bring me a slab"? How do we mean the four-word sentence while uttering the one-word command? We only mean, "bring me a slab," as opposed to other sentences that involve the word "slab": "hand me a slab," "bring me two slab s," etc. This contrast exists in our language not because of any mental act of meaning, but because our language contains the possibility of different forms of expressions. If "slab!" means the same as "bring me a slab," this is not due to any mental act of comparison, but simply because the two expressions are used in the same way.
The choice to open the Investigations with a quotation from Augustine is unusual, both because Augustine is not a philosopher of language, and because the selected quote is not representative of Augustine's position on language. In the context of t he Confessions, from which this quotation is lifted, Augustine is simply describing how he came to learn language, and is not attempting any theoretical explanation of how learning takes place.
Stanley Cavell and Warren Goldfarb have suggested a reading of the opening sections of the Investigations that focuses precisely on this contrast between Augustine's straightforward description of learning language and the subsequent "picture of la nguage" that Wittgenstein reads into it. On the descriptive level, what Augustine says is perfectly correct: young children learn words like "tummy" and "kitty" because parents point at objects and say what they mean. It is only when we take this descrip tion to be a theoretical explanation of how language in general is learned that we come up with the troublesome picture of language where words are names of objects and sentences are complexes of names. Theoretical explanation differs from description in that it infers something general about language from an isolated example.
Wittgenstein is not directly challenging Augustine. Instead, he is showing what is wrong with the way that Augustine reasons. This "picture of language" is not held by any philosopher, nor does it take a sharp mind to recognize what is wrong with it: word s like "the" and "under" are not the names of objects, and not all words are alike. Here, as throughout the Investigations, Wittgenstein is not attacking a particular philosophical position, nor is he trying to advance any philosophical theses of h is own. Rather, he uses simplified examples such as the theoretical reading of Augustine to show us the dangers of isolating and inferring general conclusions from particular aspects of how we speak and think. No serious philosopher holds Augustine's view , but most philosophical theorizing shares this tendency to move too hastily from a particular example to a general conclusion.
The example of Augustine and the subsequent language-games highlight our temptation to think of language as essentially connecting words with things. The language-game of section 2 is meant to be a paradigmatic case: if any language has a direct connectio n between names and things named, then this language does. But once we strip a language down to just four words, it is far from clear that these words are names of objects. In our language, we might call "slab" the name of an object because we can contras t the word with prepositions or names of colors, and we can say something like, "this is the word in the sentence that directs our attention to that object."
But in the simple language-game of the builders, there is no such complex machinery in place. Wittgenstein shows in sections 19–20 that our word "slab" is very different from the builders' word "slab." This difference has everything to do with the language that has been built up around this word. Wittgenstein is suggesting that words get their sense from the language that surrounds them and we cannot fruitfully isolate particular words and talk about their connection with the world. The Augustinian picture of language shares with more complex theories of language (including the view Wittgenstein himself held in the Tractatus) the tendency to extract certain parts of language from their larger context and to infer a certain relationship betwe en language and the world based on this isolated study.
Wittgenstein introduces language-games to counter this tendency. Rather than study what all language-games have in common, Wittgenstein shows us how much language-games can differ.
The point of this exercise is to show us that the connection between words and things only makes sense within the wider context of a language. At the end of section 6, Wittgenstein makes an analogy with a brake lever. The lever connected to the rod only w orks as a brake if the rod and lever are attached to the other mechanisms that make up a car. A rod and lever on their own are not a brake, nor much of anything else. Similarly, to say that language connects words with things only makes sense given the co mplex mechanisms of language that are already in place. We cannot take the word-thing connection to be fundamental to, or prior to, the workings of language.
Augustine's description works fine as a description because, within a language-using framework, we can certainly claim that certain words name certain things, and that we can learn by pointing and speaking which words name which things. The trouble with c onstruing Augustine's description as a theoretical account of language is that this construal adopts a position outside of language; that is, it does not take the existence of language for granted, but rather tries to build a model of how language works f rom the ground up. Unacknowledged assumptions are bound to creep into this model because there is no position that is genuinely outside of language for us to adopt.