1. Wittgenstein begins by quoting Augustine on how language is learned. There is an association between words and objects: words name the objects, and in tern the object is the meaning of the word. This technique of associating words and things works best for common nouns referring to things ('computer,' 'desk') and proper nouns, which also refer to things ('Ludwig Wittgenstein,' 'Immanuel Kant'). It works less well for nouns which are supposed to stand for actions ('golfing,' 'prosecuting') and properties ('good,' 'pious'). Every other word has to "take care of itself."
Next Wittgenstein considers complexes of words, e.g. 'five red apples.' According to the picture just given, each of the words has a meaning, so that a person using the complex expression would consult the meaning of each one separately and put them together. The word 'apple' means objects in a drawer, which the person opens. The word 'red' means a color on a chart, which the person notes. The person counts up to five, and for each number picks from the drawer an apple with the color found on the chart. But what does 'five' mean? Wittgenstein says that it does not matter what that word means; all that matters is that the person can use the word effectively by counting. Also, how does the person know where to look on the chart for the meaning of 'red'? Again, this is something that does not have to be explained, so long as the word is used effectively.
2. The view of language which associates words with meanings describes only how a small part of language functions. Wittgenstein sets up an example of a "primitive language" which functions in this way. A builder (A) and an assistant (B) communicate by use of the words 'block,' 'pillar,' 'slab,' and 'beam.' Assistant B learns the language in the way described by Augustine, by associating objects with the words. Then B responds to the words by taking to the builder the object corresponding to the word that is called out.
3. The limited scope of Augustine's description of language is compared to a definition of 'game' in terms of moving pieces on a board according to rules.' Not all games fit this description, though some do. Similarly, only a narrow part of language conforms to Augustine's description.
6. It is possible to think of the primitive language described in section 2. as the whole language of A and B, or even of a larger group. "The children are brought up to perform these actions, to use these words as they doso, and to react in this way to the words of others."
An important part of teaching the language is the act of "ostention," or pointing out of the objects which correspond to the words. How is the association between the word and thing thus established? A traditional theory is that the hearing of the word evokes an image in the mind of the hearer. This is not the purpose of the use of langauge (since the purpose is to facilitate the movement of the stones), but it may help in carrying out the purpose.
Ostensive teaching alone does not determine how we understand a word. If understanding is shown by carrying out an activity, such as taking the stone to the builder, then it is more than merely associating the word with an object. If B were taught to do something else with the stone upon hearing the word, then B would understand the word differently.
Wittgenstein attempts to illustrate this point by an analogy. A part of a machine, e.g. a lever, is what it is because of the way it functions in the context of the machine. Apart from that context, "it may be anything, or nothing."
7. Part of the process of using the language of section 2. is that A calls out the word 'slab' and B acts on that call by taking a certain piece of stone to the builder. In learning the language, B names the object, uttering the word 'slab' when the teacher points to a certain piece of stone. Also, the learner repeats the word after the teacher. These two activities are "processes resembling language."
It is also possible to think of the above-described activity of language as a game, and a primitive language will sometimes be referred to as a language-game., as are the two processes just mentioned: naming the stones and repeating words.
Wittgenstein will also use the term 'language game' to refer to the "whole, consisting of the language and the actions into which it is woven."
8. Wittgenstein now extends the primitive language to include items similar to those mentioned in section 1. One kind will be counters, such as letters of the alphabet. Another will be the words 'there' and 'this' which are used when pointing. Finally, in the building area there will be a number of color samples" Here is a sample order given by the builder A:
d -- slab -- there.
The first dash indicates an activity of the builder, pointing to a color sample, and the second indicates another activity, that of pointing to a place in the building. So the assistant B picks up as many stones as is indicated by 'd' of the color pointed to and with the shape assocated with 'slab' and takes them to the place in the building to which the builder pointed.
13. To say that every word in a language signifies something is not to say anything until we say what we are trying to distinguish as being signifying or non-signifying. One possibility is that the words in the language of section 8. are distinguish from "nonsense" words as in a Lewis Carroll poem.
14. In trying to say what a word signifies, we run into a problem. If we say that one word is significant because it indicates something else, we still must say what that something else signifies. Wittgenstein makes an analogy between the signification of words and the function of tools. Just as we say that all words in a language signify something, we say that all tools change something. A hammer changes the position of a nail, for example. Then we must ask what the position of the nail changes, and the answer is that it makes the box more solid. Wittgenstein seems to think that this process is not helpful in our understanding of the function of tools. "Would anything be gained by this assimilation of expressions?"
18. Perhaps it will be objected that we cannot learn anything from the languages in sections 2. and 8. because the only activity consists in carrying out orders. Isn't such a language incomplete? But even natural languages can be said to be incomplete. Language is always growing, like a town. The older part is a "maze" while the newer part (containing the precise languages of science and mathematics) is like a planned suburb.
19. Here Wittgenstein introduces another fundamental expression, by stating that to imagine a language is to imagine a form of life. The language of section 2. is about giving and taking orders. Another might consist of questions and expressions for answering affirmatively and negatively.
Wittgenstein asks whether the call "Slab!" is a word or a sentence. It might be called either, perhaps a degenerate sentence, or a limiting case of a sentence. It is difficult to compare the use of 'Slab!' in the language of sentence 2. with the use of the same sentence in English. In English, we say it means 'Bring me a slab!' But that sentence of English is not meaningful in the simple language.
21. Now Wittgenstein adds to the language-game a new activity: reporting in answer to an order. The assistant might report "Five slabs" to indicate the number of slabs in a pile. There is an ambiguity here, since the words are the same as in the order "Five slabs!" The difference is "the part which uttering these words plays in a language-game." The only real difference is in the application of the words, for although we would ordinarily imagine other differences, such as differences in tone, such differences are not necessary.
23. This section generalizes the discussion of the earlier ones. Wittgenstein claims that there are many kinds of sentences, each corresponding to the relevant activity. Asserting, questioning and commanding (corresponding to declarative, interrogative and imperative sentences) are but a few uses of sentences. Futher, languages change, adding new uses from time to time, and forgetting others.
The expression 'language-game' indicates that "the speaking of language is part of an activity, or a form of life."
Here is a list of some of the many linguistic activities. Some have thought there is only one essential activity of language (as did Wittgenstein himself, in his early book Tractatus Logico-Philosophicus), but there are many.
46. Some philosphers, such as Plato, believed that names signify "simples," such as triangle itself. These are primary elements out of which everything else is composed. It impossible to describe the thing named, only to name it. Wittgenstein quotes Plato's dialogue Theaetetus as a source of this view.
Other philosophers have taken up this line. Bertrand Russell, the most famous philosopher of the first part of this century, had held that names pick out individuals which an only be called "this" and "that." Wittgenstein himself had a similar view.
47. Philosophers have looked to simples because they want to say what makes up composites. They seek to answer the question, "What are the simple constituents of x"for things like chairs or visual images. Wittgenstein claims that simplicity is purely relative. "It makes no sense at all to speak absolutely of the 'simple parts of a chair'" because there is always something simpler relative to any level we reach. The bits of wood that make up the chair are composed of molecules, and the molecules which make up the bits of wood are composed of atoms, etc. So Wittgenstein rejects the philosophical question of the component parts of things by claiming that any answer is relative.
65. Here Wittgenstein admits that his early work was a failure. In the Tractatus Logico-Philosophicus, he had searched for the essence of language, for what unifies all the activities found in all language games. This search for the "general form of all propositions and of language" had given him great difficulty, and now he recognizes that he could not in principle complete it.
The task initiated by Socrates and Plato, to find the common element, is declared impossible when directed at language. But though they have no one thing in common, they are still related in many different ways.
66. An example of the relationship of the various "linguistic" activities can be found by asking what a game is. There is no common element among all games, but there are "similarities, relationships, and a whole series of them at that." Card games have some things in common with board games, which in turn have some things in common with ball games. Not all games are amusing, and not all involve winning and losing. What counts as skill and luck varies among them.
What we have is "a complicated network of similarities overlapping and criss-crossing: sometimes oveall simlarities, sometimes similarities of detail."
67. These similarities are best described as family resemblances, in the manner in which members of families resemble one another without sharing any one unique feature in common.
70. It might be objected that if we let go of rigid meanings, we don't understand what we say when we use a term. But Wittgenstein points out that even common language does not presuppose rigid meanings. If I say, "The ground is covered with plants," I may be unable to give a precise specification of what a plant is, yet still understand what I am talking about. It is enough to be able to give a rough picture.
71. It seems that a concept like "game" is a fuzzy one, but it is still meaningful: a fuzzy photograph is still a photograph. And a sharp photograph is not necessarily the best for every purpose.
It is also meaningful to give vague orders, such as, "Stand more or less over there." One does not pick out a specific spot. A game is of this sort. One does not find in examples of games a common element. I give examples and show how they can be extended by analogy, which information can then be used by one who hears it to understand what a game is; and this is how I play the language-game with respect to the word 'game.'
75. Can we know what a game is without being able to say it? Perhaps in that case we are in possession of a hidden, unformulated, definition. But Wittgenstein replies by claiming that giving examples, showing how they can be extended by analogy, excluding some things, etc. completely expresses his concept of a game.
81. Now the discussion turns to the attempts made by Wittgenstein and others to interpret language on the model of formal logic, which has fixed meanings and rules. This attempt is still made by philosophers and linguists, but Wittgenstein thinks it is a mistake. In the first place, it is difficult to capture the vagueness of natural language with fixed rules. Further, we should not think of natural language as an approximation to formal logic. Then we think of the formal as being ideal, as if our natural, informal, language were inferior to it, "as if it took the logician to show people at last what a proper sentence looked like."
Wittgenstein claims to have risen to a higher perspective on what is being done by the formalizers of language. Once one "has attained greater clarity about the concepts of understanding, meaning, and thinking" (as Wittgenstein apparently thought he had), one can recognize why someone would try to formalize language.
89. The role of logic is now questioned. It is widely considered to be the most fundamental of all the sciences, seeking the nature of all things, whether they exist or are merely possible. "It takes its rise, not from an interest in the facts of nature, nor from a need to grasp causal connections: but from an urge to undestand the basis, or essence, of everything empirical." Logicians do not seek to uncover new facts, but to explain what is already "in plain view."
But they run up against a problem, for we know how things are when we are not asked, but cannot explain them when asked (as Augustine had said about time). Wittgenstein states that in this case we must remind ourselves of something which we have forgotten.
90. Logic should not try to penetrate phenomena, to get to their inner essence. Instead, it should seek broadly to show what is possible, by reminding ourselves about the kind of statement we make about phenomena.
When we look at the kind of statements we have made (e.g. about time) we investigate grammar. The point is to clear away misunderstanding which results from the use of words. Sometimes this can be done by "analysis," which is a kind of "taking apart" of the expressions we use.
91. Does analysis finally reveal the essences for which philosophers have searched? In making our expressions more exact, do we move toward an ideal final state of complete exactness?
98. Perhaps not. There is nothing intrinsically wrong with the use of vague sentences, such that exact sentences are superior. They have a perfect order in themselves.
99. Still, if we leave something open in explaining the meaning of the sentence, it seems as if we have failed completely, as if we had claimed to have locked a man in a room, when one door is open. Wittgenstein asks whether it is true that an enclosure with a hole is as good as none.
100. Another objection is that if there is vagueness in the rules, one has no game at all. But this only means that it is not a perfect game. To say it is not a game at all is to be dazzled by an ideal.
108. If the structure of natural language is that of a family resemblance, what is the role of logic? It would seem to disappear if its rigorous rules are relaxed.
But this is again mistaken. Logic is concerned with ordinary sentences, and these sentences are a natural phenomenon in space and time, as the early sections indicated. There are different ways of talking about words, in terms of meaning and in terms of use, just as we can talk about a chess-piece in terms of the rules of chess and its physical properties. Both must be included.
109. Wittgenstein now states that the only activity of his investigation is description, and explanation is given up. We are mistaken in seeking for philosophical explanations. The best we can do is to find in language the sources of our search for "new information" (e.g. about Platonic forms) in the way language works. "Philosophy is a battle against the bewitchment of our intelligence by means of language."
119. Philosophy attempts to take us beyond the limits of language, and in so doing it bumps itself. The right response is to uncover the nonesense which results. "These bumps make us see the value of the discovery."