Speaking as social action

priorities and goals in interpreting human behavior. Speech act theorists start from the assumption that “language is action” but do not question their own notion of “action.” They assume that “action” itself is a universal dimension of human existence that does not need further analysis. Thus, in analyzing directives, the issue is “what are the rules we need to assume to explain how a person can get another to do something?” The question of who is doing what for whom and why is not entertained. Such an issue would be seen as outside the domain of the theory.

Ethnographers, on the other hand, believe it is important to extend the philosophical characterization of “action” to include the notion of person implicit in such a characterization and the relationship between language use and local theories of truth, authority, and responsibility. This means that for ethnographers interpretive analysis of words as deeds is different partly because the notion of context is different. For a linguistic anthropologist, as suggested by Lindstrom (1992: 104)

... contextual analysis begins ... by asking what kinds of talk can be heard and understood, and what kinds of talk cannot. Are all participants qualified to speak and to speak the truth? Can talk carry all meanings?

These are far more complex and far-reaching questions than those usually addressed by speech act theorists. Should we then conclude that any kind of interface between philosophers and anthropologists is destined to be ill-fated? Not necessarily. There have been attempts within western philosophy to sketch a theory of language as action that is closer in spirit to that practiced by most linguistic anthropologists. One such theory is the one developed by Wittgenstein in the 1930s and 1940s after his return to Cambridge.

7.4Language games as units of analysis

In his later writings, Wittgenstein often invoked the metaphor of games for talking about how people use and understand language.

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

Grammar describes the use of words in the language.

So it has somewhat the same relation to the language description of a game, the rules of a game, have to the game.

(Wittgenstein [1933 ca.] 1974: 60)

Wittgenstein’s analogy between language and games has often been taken too literally. Searle (1969: 43), for example, contends that the analogy does not work because when one makes a move in a game like chess one is not said to mean

236

7.4 Language games as units of analysis

anything by that move.20 Phrased in this way the parallelism does not work precisely because playing a game of chess and speaking are two different activities – Wittgenstein might have been the first one to admit this. We must look for what the metaphor points to rather than for what is obviously different between the two activities.21 What Wittgenstein is proposing with the chess metaphor is that understanding a word in a sentence is like understanding a move in a game. Part of this knowledge is what psychologists call procedural knowledge (the knowinghow) (see chapter 2) but it goes beyond that. We get an understanding of how a word is used by matching it with other words and other contexts and by projecting its impact on future words and utterances just as we project a move of chess against past and future moves. The game metaphor also implies a differentiated understanding among users. An expert chess player understands a move differently from a novice or someone who has never played the game.22 Similarly, not everyone understands a word or an utterance in the same way. There are many different domains or contexts (read “games”) for language use. Not everyone is the same in terms of the ability to act within a certain domain. Whereas Austin’s and Searle’s goals of finding a finite set of conventions and conditions give the impression of a universally shared linguistic knowledge, in reality different speakers, even neighbors or close friends, can have quite different understanding of the same linguistic expressions. I remember telling an artist friend that I had bought a Fender electric guitar. “What color?” he asked. “White,” I said. When I later took it out of the case, he looked at it and, with a disappointed look on his face, complained “You said white! This is ivory!” The difference between our linguistic characterization of the color of the guitar implied, as Wittgenstein

20“Characteristically, when one speaks one means something by what one says; and what one says, the string of sounds that one emits, is characteristically said to have a meaning. Here, incidentally, is another point at which our analogy between performing speech acts and playing games breaks down. The pieces in a game like chess are not characteristically said to have a meaning, and, furthermore, when one makes a move one is not characteristically said to mean anything by that move” (Searle 1969: 42–43).

21Searle’s reading of Wittgenstein seems vulnerable to the same kind of criticism presented by Tambaiah regarding Malinowski’s misunderstanding of magical spells (see section 7.1).

22“When a man who knows the game watches a game of chess, the experience he has when a move is made usually differs from that of someone else watching without understanding the game. (It differs too from that of a man who doesn’t even know that it’s a game.) We can also say that it’s the knowledge of the rules of chess which makes the difference between the two spectators, and so too that it’s the knowledge of the rules which makes the first spectator have the particular experience he has. But this experience is not the knowledge of the rules. Yet we are inclined to call them both ‘understanding’ ” (Wittgenstein 1974: 49–50). On the difference between experts and others, see also Putnam (1975), who proposes a theory of meaning based on the idea of a “division of labor” among speakers, with experts knowing what common people do not need to bother with.

237

Speaking as social action

would have said, a different “form of life” (“... And to imagine a language means to imagine a form of life” [Wittgenstein 1958: 8]). Colors and color differentiations mean something different to a painter. They are part of different forms of life.

Wittgenstein’s point is not only that to know how to use a word (or any kind of linguistic expression) means to know the kinds of things we can do with it – a piece of chess can move only in limited ways but there are countless new situations in which we can use it and in each case there is a new “meaning” – but also that there is a particular kind of existence that a use implies.23 That is why he wrote, “If a lion could talk, we could not understand him” (Wittgenstein 1958: 223).

This view of language has important consequences for how one should write a grammar of a language. To write a grammar of a language means to describe what people do with certain expressions (see section 7.2.2). As we shall see in chapter 8, an analysis of sentence structures as parts of interactional sequences comes close to the kind of analysis envisioned but never fully realized by Wittgenstein.

Given his recurrent use of the game metaphor, it should not be surprising that the closest thing Wittgenstein ever came to what we might call a unit of analysis is his notion of language game,24 which he first introduced in The Blue Book and is amply used in his subsequent manuscripts:

I shall in future again and again draw your attention to what I shall call language-games. These are ways of using signs simpler than those in which we use the signs of our highly complicated everyday language. Language-games are the forms of language with which a child begins to make use of words. The study of language-games is the study of primitive forms of language or primitive languages. If we want to study the problems of truth and falsehood, of the agreement and disagreement of propositions with reality, of the nature of assertion, assumption and question, we shall with great advantage look at primitive forms of language in which these forms of thinking appear without the confusing background of highly complicated processes of thought. When we look at such simple forms of language the mental mist which seems to enshroud our

23The theory of gender differences by Maltz and Borker (1982) follows a similar logic: men and women use language differently because boys and girls learn to use language in different contexts, in other words they have been socialized differently, or, in Wittgenstein’s terms, they use the same words but experienced different “forms of life.” A similar view is held by Tannen (1990).

24For a discussion of the development of the notion of language game in Wittgenstein’s writings, see Baker and Hacker (1985: 47–56).

238

7.4 Language games as units of analysis

ordinary use of language disappears. We see activities, reactions, which are clear-cut and transparent. On the other hand we recognize in these simple processes forms of language not separated by a break from our more complicated ones. We see that we can build up the complicated forms from the primitive ones by gradually adding new forms. (Wittgenstein 1960: 17)

The notion of “language game” is thus a working notion, it is not a category like “speech act” or “illocutionary act” and is not something that is out there in the phenomenological world of speaking. It is only an instrument for analysis, a heuristic device, which is used to first isolate “primitive” cases (“primitive” here means “simple” and does not have an evolutionary connotation). Only once we become expert at analyzing these simpler cases, we can graduate to looking at more complex ones. Simplicity is the only concession that Wittgenstein seems to make to traditional scientific methods. Otherwise, the emphasis here as elsewhere in Wittgenstein’s teaching is on the importance of observation and description. We must resist the scientistic drive to make quick generalizations. This drive leads us to confusion because it is based on the wrong assumption that things that have the same name will necessarily share a common set of characteristics. We must instead cultivate and enjoy the practice of description of particular cases. It is an investigation of particular cases that will clear the confusion brought about by wrong ways of thinking about language such as the tendency to conceive of meaning as a mental image shared by everyone. The metaphor of the “game” is used to stress that different uses of language are like different games, namely, that they may share some features but they need not. Just like we might call “games” a number of activities that do not share the same basic features or rules, upon inspection we might find that language activities might not always share the same set of properties.

It should be clear by now that Wittgenstein uses the notion of language game to argue some of the main points of his view of meaning and interpretation. These points include the idea that connecting words with objects cannot be the basic method for acquiring a language and the observation that the same word or sentence can acquire different meanings depending on the activity within which it is used. But Wittgenstein also uses language games to argue against the idea that the meaning of a linguistic expression is just in someone’s head. Through the concept of language game, he invites us to look at the context of what speakers do with words and for this reason constitutes an insight into what linguistic anthropologists are interested in. At the beginning of Philosophical Investigations, for instance, Wittgenstein gives the example of a situation in which a builder is working with an assistant. The assistant has to pass the proper stone to the

239

Speaking as social action

builder in the order in which the latter requests them. In this context, the builder’s use of simple nouns like block, pillar, slab, beam must be understood as an order, that is, an instruction for the assistant. Linguists have often suggested that to account for how a single word can, in certain contexts, be understood as a command, we must assume that the word, e.g. slab!, stands for an entire sentence, e.g. something like give me a slab! This is a process of “deletion” that grammarians call ellipsis (the same process that accounts for how expressions such as I do or me too can be interpreted as some related but different version of what has just been said). Wittgenstein argues that the analysis of single-word sentences as elliptical – i.e. as missing something – is unnecessary and leads to absurdities. The force of slab! as an order is not only in the linguistic form – which may or may not be pronounced with a particular type of intonation – but also in the activity that is being performed.

The sentence is “elliptical,” not because it leaves out something that we think when we utter it, but because it is shortened – in comparison with a particular paradigm of our grammar.

(Wittgenstein 1958: 10)

In other words, even the explanation of the meaning of a single word as a shortened version of a longer expression is a language game, the language game played by grammarians! There is nothing wrong with such a language game, of course, but it is only one of the many possible ones in providing an interpretation of slab! in the context described above. The same type of analysis can be applied to the use of ostensive definitions (“chair” means “this” – while pointing to a chair). Ostensive definitions too can be used to explain the meaning of words and sentences but they must be understood as part of specific language games such as the routines used in foreign language classrooms. The teacher points to the blackboard and says blackboard (if he is teaching English) or lavagna (if he is teaching Italian). This is a perfectly legitimate way of teaching words and meanings, but it has a restricted range of uses and, according to Wittgenstein, is by no means more basic than other uses of language. Think for instance of the familiar routine when the teacher points to himself and says My name is John and then goes around the room asking each student what is your name? The successful accomplishment of this speech act depends on the students’ success at conforming to the rules and expectations implicit in the teacher’s actions. Beyond the fact that the teacher’s question must be understood as a request for information and hence as requiring a linguistic performance on the part of each student, there are a number of cultureand context-specific assumptions that are implicitly at work, a crucial one being the criteria for what constitutes an appropriate answer. The students, for instance, must come up with something that satisfies

240

7.4 Language games as units of analysis

the requirements of the English word name in the particular context of the classroom. The teacher’s answer to his own question provides a model to be followed (My name is John), but such a model is not an instruction that can be universally followed; it does not contain all the possible ways in which the rule could be satisfied (and hence all the ways in which it could not be satisfied). Students, for instance, must decide which of their several names or nicknames should be provided in the allowed slot. In my case, for instance, I would have to decide whether to give Alessandro or Sandro as an answer. But in fact even a decision of this sort does not exhaust the possible alternatives available in the context. Some students might interpret the model offered by my name is John as suggesting that they should provide an English name. This is how Yosef becomes Joseph and

Gianni becomes Johnny. In my case, I would then have a much larger set of possibilities including Alexander, Alex, Sandy.25 Choices of this sort provide resources for locating one’s teacher and classmates within different networks of acquaintances and may implicitly constitute a particular stance with respect to one’s identity in a foreign country – not a simple task for the students and certainly something of a magnitude that most English-as-a-Second-Language teachers might not be prepared to deal with. Finally, the activity of exchanging names in a classroom does not easily transfer to other situations, when goals or participants differ. Thus, if a student is stopped by the police and has no identification with him, the model my name is John will not do. To get a first sense of the different meanings this utterance might acquire it is sufficient to start imagining it said by different people: a student, a teacher, a waiter, a doctor, a prostitute. In each case, we could build a simple language game within which my name is John constitutes a different move and hence affords different following moves. More generally, speaking is an activity that involves particular forms of cooperation among participants in an interaction.

Here the term “language-game” is meant to bring into prominence the fact that speaking of language is part of an activity, or of a form of life. (Philosophical Investigations, § 23)

The notion of language game is appealing to ethnographers, who must make sense of linguistic interpretations that do not follow the western grammarian’s model of providing glosses of words. For instance, Rumsey (1990) uses the notion of language game to explain the unexpected answer he received from a Ngarinyin man (in northwestern Australia) when he asked him for the meaning

25As we go deeper into this analysis we realize that the issue of what constitutes an appropriate or acceptable answer to the question what is your name? is nothing but the condition for generating all the names that could constitute a natural class with “John.” See Sacks (1972).

241

Speaking as social action

of baba. This is a term that Rumsey believed to be an address term and in fact later identified as a vocative kin term for mamingi “my mother’s father,” “my mother’s brother’s son,” and so on. The man, however, did not provide the kind of paraphrase expected by the anthropologist. He said, instead, that baba meant something “like a jannjuli [‘give me’], give me tobacco, or thing like that.”

What he was giving me was obviously not what we would think of as the sense, or possible reference term, but rather, a locution that makes explicit the pragmatic function of this term of address within a typical context of use – mamingi being someone from whom I am entitled to demand things. Of course it was possible in time for this man to learn my language game of glossing on the basis of referential function as distinct from other pragmatic ones, just as it was possible in time for me to come to a better understanding of his. But in order to do so, both of us had to put aside our everyday, commonsense way of talking about language. (Rumsey 1990: 353)

If speaking a language is part of an activity, to provide metalinguistic statements is also part of an activity and one that follows local theories (or “ideologies”) about the relationship between words and the world (Schieffelin, Woolard, and Kroskrity 1997; Silverstein 1979; Woolard and Schieffelin 1994). The notion of language game allows fieldworkers to deal with different interpretive strategies without giving up on the idea that there is order (or a logic) behind the apparently strange answers they receive. As units of analysis, language games assume that language is a set of unbounded and yet manageable (and learnable) set of cultural practices. There are however two sorts of criticisms that are leveled at the notion of language game as a unit of analysis:

(i)Language game is such a general category that it is difficult to see where it would not apply. It would include very simple as well as very complex uses of language. How do we distinguish between them? How do we know where a language game starts and where it ends?

(ii)The notion of language game, with its implicit rejection of a “core” meaning of linguistic expressions, makes it impossible to make generalizations about language structure and language use.

The first criticism could be answered by saying that, as discussed earlier, Wittgenstein thought of language games as rather simple kinds of speech activities. The study of such simple activities is a prerequisite for the study of the more complex real-life situations. What Wittgenstein’s theory needs is a better way of defining the boundaries of such situations. As long as we continue to create our own examples and imaginary situations, we are never going to know whether simplicity is in the situation or in the eyes of the observer. Wittgenstein’s method

242