Performative Pragmatics

Teacher's Guide

Chapter 2
Histories of Linguistics

The Roots of Performativism in History and Intuition

The histories I trace in this chapter make it clear that the conflict between what I am calling performative and constative linguistics is actually part of a much larger scholarly clash between Kantian and post-Kantian constructivism and the older and better-established objectivist tradition. The rhetorical goal, of course, is to legitimate performative linguistics by historicizing it—tracing its scholarly roots back two hundred years, exploring the different scholarly traditions that have grown out of those roots, especially in the twentieth century. Without this sort of history, performative linguistics might just be a bit of clever iconoclasm, a subversive fantasy that I'm intent on hurling in the faces of my "establishment" colleagues.

But of course no matter how it is historicized, so long as the objectivist/constativist tradition remains dominant in our culture, performative linguistics will remain counterintuitive for most students—"students" both in the narrow sense of people who take our classes and in the broader sense of people who study language. I assume you're reading and possibly using this textbook because you feel a certain sympathy for a performative approach; if so, you're probably already aware of what a tiny counterhegemonic minority you're in.

And I've set up this book to minimize students' "intuitive" (hegemonic) rejection of or resistance to performative ideas, starting with the seemingly harmless and pluralistic language-as-drama metaphor in Chapter 1 and doing a dense scholarly history of performative linguistics here in Chapter 2. But the resistance will likely always be there, at least in certain students, the ones who expect objectivity and truth in university classrooms.

The other side of this coin, of course, is that a performative sense of people performing actions with words is not too far removed from everybody's commonsensical understanding of how language is used in actual speech situations. Everybody knows that the same words can mean lots of different things, depending on when and how and to whom you say them. Everybody knows that you have to read past the "obvious" or "surface" meaning of somebody else's words to what they are "really" saying. In this sense it is the constative approach that feels alien to students, that constative linguists have to push hard to inculcate.

The conflict comes out of a widely accepted division of intellectual labor: you think commonsensically, intuitively ("any old which way"), outside the classroom, scientifically ("constatively") inside the classroom. The university, students assume, is for scientific thought. When they walk into a class on language, they expect to be taught analytical methods for the study of objective reality—and performative methods don't seem to be "analytical" enough, airtight enough, rigid enough, and seem to be (and are) based on a thoroughly anti-objective approach to language, starting with the assumption that language has no objective existence and is (constantly re)constructed dialogically in the heads of the groups of people who use it. That doesn't seem to be what a university classroom is for. That isn't what professors are supposed to teach: uncertainty, rough guesses, half-hearted expansions of intuitive distinctions into analytical models.

It bothers students that you can never really know for sure what speech act someone is performing, or what s/he is implicating. It bothers them that any utterance at all can work performatively—that there isn't a clear-cut distinction between performative and constative utterances. It bothers them that I don't give them a single comprehensive tabulation of speech-act types, don't favor one of the several I offer, and then even encourage them to make their own. This seems a cavalier attitude toward analysis, toward objectivity, toward intellectual activity in general (which as I say they tend to associate with science).

It typically takes me the first month of class to answer their qualms satisfactorily and ease them onto the path of exploring what they already intuitively know: that people do things with words, and it is often difficult to impossible to determine exactly what they're doing; that it's possible to analyze something fruitfully without imposing a false "scientific" or "objective" form on it; that most of the distinctions we make in class they already know how to make already, in their everyday conversations with friends, and that all they're doing now is becoming more aware of doing so.

Given the relative success I've had with students appealing to this "intuitive" or "commonsensical" feel for the material in this book, in fact, and their fairly stiff resistance to the theoretical and methodological angle, it may even be wise not to teach this history chapter—to have them skim it, say, so they have a vague sense that there is a scholarly history behind all this but don't have to face the overwhelming onslaught on their scientific assumptions that it represents. I wrote this book over two semesters of teaching the course, and didn't write this history chapter until halfway through the second semester, and have found that students responded more openly to a performative approach those two semesters than they have later, when I've assigned the history chapter as well. But then that may just be me.


1. My naming of a certain approach to the study of language "performative" is a performative utterance. Other people have performed similar approaches as performative before me; I'm not the first. But I am, still, performing them as what I take them already to be, performing them as what other people have performed as, bringing into being what already is. I am not, in other words, bringing something out of nothing; I am merely organizing existing approaches in a new way, getting people to see them in a new way.

As to whether this is a bad thing, that depends on your assumptions about human knowledge. If you believe that objectivity is the ideal, then the notion of making people see things in a new way will seem vaguely silly. Why not get them to see things the right way? Why is newness so important? From a post-Kantian, social-constructivist, performative point of view, there is no right way; there are only socially constructed ways, knowledge that is already accepted by almost everybody and knowledge that is just being argued for.

2. From a constative viewpoint, Chomsky is playing a dangerous game: opening up a perspective on language that takes human variability into consideration. If the key to a linguistic methodology is the speaker's creativity in transforming basic sentence structures in a potentially infinite number of ways, how is it possible to say anything about language? (Anything objective, that is: anything generally true.) If speakers are so creative, they'll overstep the bounds of any rules we care to set up, any structures we claim to find.

Chomsky successfully flirts with this danger by setting up the importance of the speaker's creativity but then idealizing the speaker, turning the speaker into an imaginary abstraction that doesn't have the power to deviate sharply (or actually at all) from Chomsky's objectivist descriptions.

This successfully contained danger was a big part of Chomsky's popularity, I think. He wasn't just playing it safe! He came very close to opening up a big can of worms, and people found that attractive.

And what is happening now is that Chomsky's followers are pushing past his idealization of the speaker to look at the creativity of real speakers—and that is in fact making Chomskyan linguistics increasingly performative (specifically, cognitive).

3. Constative linguistics is not Christian in any dogmatic sense, of course. A constative linguist would accept no phrase from the Apostle's Creed as a statement of his or her linguistic method. It is Christian rather in the "civilizational" sense: since Christianity has for two thousand years been the dominant religion in Western civilization, it has given significant shape to the acceptable ways we have of knowing things. For many centuries, in fact—until, say, the sixteenth century—scholarly objectivity meant recognizing the importance of God having created the universe in a rule-governed way. That was an essential part of Western thought. That specifically theistic requirement has now dropped away, and scholars are allowed to be objective without tracing their objectivity back to God; but we still have an abiding faith in the objectivity that came from believing that a single rational God created the universe, and that the universe therefore makes rational sense and it is our duty to see that sense as clearly and accurately as possible. That belief is originally Platonic, but it comes to us through Christianity, through the incorporation of Platonic thought into Christian theology by Paul and Augustine and others, so the specific form we inherit is saturated with Christianity—even if it lacks a belief in God or a specific salvation history.

4. Who knows, maybe this is a "learning style" thing, a personality difference. Some people seem to take to a Kantian approach like fish to water; others keep on rejecting it. Those who accept it find it a great relief, a solution to intellectual problems that had been troubling them; those who reject it just can't see any value in it at all. Get students talking about these differences.

5. The problem arises within constative pragmatics from Charles Morris's original definition of the field: pragmatics is the study of the relations between signs and interpreters. What interpreters? That means people. In fact, if we are scholars attempting to interpret speech acts in snatches of conversation overheard on a bus or found in a corpus, it means us. We are the interpreters that relate with the signs.

But that poses a huge problem for constative linguistics, because if it all depends on us, then it's subjective. Suppose someone says, "But how do you know that that's a behabitive and not, say, a directive?" Can you say simply "I know because I'm an interpreter and I'm relating to that sign"? No. Not if you're a constative pragmatician. That's the obvious answer, but it isn't an objective answer (what if I'm wrong?), so it isn't permissible. The constative pragmatician has to find some other way of establishing the precise nature of the speech act being performed; s/he can't rely on her or his own intuitions.

The ideal pragmatic methodology for a constative linguist would be one where the underlying structures of the speech act forced people (including pragmaticians in the act of studying speech acts) to interpret them in certain ways. There needs to be a rule-governed machine "in" the speech acts that "operates" people mechanically, like robots, so there is no wiggle room, no human "play" in the system. Then the people don't really matter. Then they can't disturb the proper stable functioning of abstract structure, so important for objectivity.

And constative pragmatics has by and large been an ever more complex quest for these machines: for not only the deep structures of speech acts but (most importantly) for the visible or audible "markers" of deep structure that not only signal to us what is going on pragmatically but in some important sense trigger our conversational or interpretive reactions. If we don't actively and creatively respond to speech acts but are activated by speech acts, or by their conversational markers, then our interpretations of speech acts are not subjective but objective, hence allowable in a constative linguistics.

The easy answer to all this, of course, is that it's all about people. Remove this ban on individual speakers' and listeners' subjectivity, allow pragmatics to revolve around people doing things with words, and the structural complexities and paradoxes that have plagued constative pragmatics all disappear. Then of course you need a new methodological rationale: you can't just reject the ideal of objectivity without offering anything to take its place. This book explores the performative replacement for objectivity.


1. a. A constative pragmatician would probably want to define "conversation" in some rigid formalistic way that would make it possible to count the number of conversations Josh has exactly. The interesting thing about this scene for a performative pragmatician is precisely the breakdown of easy boundaries between "conversation" and "no conversation." Does a conversation require that he be talking to someone? Well, the baseball announcer is talking to his audience; but Josh isn't a real announcer and the audience is imaginary. Does that count? Who knows! But it's interesting to hash out—and that's the main thing here, to get students talking about the boundaries.

b. Obviously, the magic in the movie that makes Josh's fantasy of being "big" come true isn't realistic: although Josh's wish, "I wanna be big," does work as a performative utterance and makes him big (though not till some time late that night, when he's asleep), performative utterances in our world don't have that power. Still, the power of words (conceived performatively) to create a feeling of reality—especially a social feeling, as when Josh and Billy work together to create fantasy worlds—is very real.

c. The performative imagination continually attempts to transform the world, creatively, imaginatively; the world resists those transformations. This sets up a kind of dialogue or tension between the "thinginess" of the world and the transformative power of the imagination. Even the movie's magic, in fact, which does transform the thirteen-year-old Josh into an adult man played by Tom Hanks, doesn't make him a complete adult: he has no (new) memories of the "lost" years since he was thirteen; he still misses his mother, and can't imagine having sex with the female colleague who becomes his sort-of girlfriend. The world resists the transformation. And the performative imagination has to adapt to that resistance, often incorporating it into a new imaginative construct.

More: Teacher's Guide index

General: Structure | Pedagogy | Dialogues

Chapter 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12

Last Modified: