Performative Pragmatics

Teacher's Guide

Chapter 12
Politeness and Face

The Consequences of Ignoring the Body

This chapter presents two different models of politeness (Leech's based on maxims, Brown/Levinson's based on the management of face-threatening acts) and three different models of face (Brown/Levinson's as image, Goffman's as value, and mine as the somatic exchange of value). The structure of the chapter takes us from the focus of Part III, implicatures (Grice's maxims), through Brown and Levinson and Goffman on face, to the focus of Part IV, the body—specifically, my own somatic revision of Brown and Levinson's and Goffman's theories of face. Both Brown/Levinson and Goffman nod toward the body in their recognition of the importance of feelings in the economy of face; but Brown and Levinson basically recognize it once and then forget about it for the rest of their book, and Goffman makes feelings separate from and secondary to face, an ancillary phenomenon that somehow belatedly "gets attached" to face.

My argument is that the body and its feelings (and external displays of feelings) are intrinsic to face, the very basis of the social exchange of face-as-value. This is, after all, the origin of our folk notion of "losing face": your facial expression shows that you have lost some sort of coherent connection to the character you were displaying, your "face falls," or crumples, or bursts into laughter or tears, etc. It's possible, of course, to "lose face" while maintaining a poker face; but if you keep your face stoic enough in the "face" of a major face-loss, that face-loss can be minimized, the damage to your social value controlled. And of course it's possible to imagine yourself losing face-as-value, and, worrying about that, feeling anxiety about that, lose face-as-body-language (display your worry or anxiety on your face and in your other body movements), and as a result let your secrets be revealed and lose actual face-as-value.

Everybody knows, of course, that this happens. We've all experienced it countless times. Neurologists and anthropologists and social psychologists have studied it thoroughly, documented it in literally hundreds of books and thousands of articles (and some of them are references in the chapter's "suggestions for further reading"). But the prevailing scholarly paradigms in the humanities are still mentalist, focused on what we do with our minds, to the exclusion of bodily things like feelings and body language. Anything to do with the body is by default considered "not my field"—something for people in other disciplines to study. Hence Leech's quite attractive and persuasive study of politeness without a single reference to how people feel, or how they display their feelings in their body language. It's common sense that we know when we've hurt someone's feelings or been rude by looking at their faces, reading their feelings there! But the mentalist scholarly paradigm coaches us to leave such commonsensical matters out of our theorizing, so we do. Hence also Brown and Levinson's passing reference to feeling early in their book and subsequent reduction of politeness and face to formal (and quasi-mathematical) categories. Hence even Goffman's relegation of feeling to secondary status, as a mere ancillary indicator of face, nothing intrinsic to it.

And of course, as I've said elsewhere in these notes, many scholars of language still feel utterly skeptical about the somatics of language, inclined to pooh-pooh this notion of a somatic exchange of value as fancy theorizing with no basis in reality. In other words, maybe it's overstating things a bit to say that everybody knows that this happens. Not everybody does, clearly. And many people who do recognize that they may feel bad when saying certain things are unwilling to agree that their bad feeling is somehow integrally connected with the saying: for them, as for Goffman, it's a secondary phenomenon, without much significance for the study of language or other human social behavior.

But consider the fact that we often have the sense of knowing what someone means. Especially someone we know really well, someone we've lived with for years. You say "Are you still mad at me?" and s/he replies slowly, with long-drawn-out rising intonation and partially averted eyes, "Nooooo ..." You know instantly: s/he's still mad at you. You know it. It's a violation of the maxim of quality, and thus in Gricean terms should be impossible for you to figure out, but you figure it out anyway. How? Based on body language. You can not only see it; you can feel it. You say "What's wrong? What did I do?" and s/he replies tersely, or even with a minimal attempt at neutrality, "Nothing." And you think, oh man, what on earth did I do? Why is s/he so pissed off? This is not, as I say, what Grice would implicature. Your significant other may be implicating something—yes, I'm angry, but you're going to have to figure it out on your own, work out in your own head what you did and then find just the right way to apologize for it, I'm not going to help you, you bastard—but is laying no implicatural pathway for your interpretive activity. So you have to go away and think through everything that you have done recently that might be provoking this reaction, and figure it out. But your first intuitive reaction is crystal-clear: s/he's angry. S/he's lying about there being nothing wrong.

This is the great advantage of taking body language into consideration in the study of language. This is, in fact, the huge back-breaking problem with constative models of language, at least for pragmatic phenomena: 93% of our communication is somatic, and a constative pragmatician is going to ignore all that and try to model communicative success (and failure) based solely on the remaining 7%. Hence the incredible complexity of constative pragmatics. Look at Levinson (2000), for instance, which builds on two decades of constative pragmatic theorizing. Even he says in his introduction that "half a dozen years interposed between the first and second drafts of this manuscript, and this time revealed to me the passing nature of many of the formalisms and formal theories to which the material in this book could be related" (2000: 1). So he pared many of them away—and the book is still rebarbative in the extreme. This is because Levinson's constative pragmatic paradigm forces him (and all of the constativist colleagues whose pragmatic theorizing he draws on) to ignore the obvious: that we often read people's meaning through their body language, based on long experience with their and others' body language. Strategically (self-)denied access to this somatic explanatory model, they have to generate vast conceptual frameworks carried somehow (often utterly unconvincingly) "by the structure of utterances, given the structure of the language, and not by virtue of the particular contexts of utterance" (Levinson 2000: 1)—including body-language context.

Constative phonetics and syntax seem a reasonable enterprise—though there are phonetic and syntactic irregularities and slippages that give constativists gray hairs, and make them stretch their conceptual ingenuity to the breaking point, trying to come up with plausible constative models for structural anomalies ("exceptions"). Constative semantics as a study of dictionaries and phrase books is also sort of defensible. Abstract (almost) any linguistic phenomenon far enough away from actual use and constative approaches become not only feasible, but quite useful. But pragmatics is about use. This makes constative pragmatics an almost unbearably hopeless enterprise.


1. If face-as-image is a picture (a two-dimensional representation of a face, say) and face-as-value is an abstract mathematical quantity (high or low), obviously it's possible to study them, and to manipulate them in actual social encounters, with sheer analytical thought, ignoring the body and its feelings. Until very recently, of course, the body was almost never taken into consideration in studies of human social behavior—certainly not in studies of language, which were thought to be purely mental. Even now the thought that speech and other human social activities might be somehow significantly shaped by the body seems wildly counterintuitive to many. And this is one reason it's so important to stress body function in discussions of human communication: because the scholarly tradition makes it so easy to ignore it. The other main reason, of course, is that a performative pragmatics is based on performance, which is always done in the body. What the body is doing while we perform in it, therefore, is going to be of major significance.

2. Of course we can do things that have an impact on how people see us. The real question is how effective such unilateral social action ever is. Since face depends on how others see us, to gain face we can't just do something good; we have to get others to think that we did something good. And it's hard to convince people to think a certain way—especially when we want them to think well of us. One problem is that we tend to think well of people who do good things without really trying. That means that trying to make someone think well of you will tend to be perceived as a form of boasting, and will usually have the opposite effect. A far bigger problem, though, is that face is so very relational, diffused through an entire social network. It's not just that I'm going to be unwilling to think well of you if that means thinking poorly of myself (if you're so great I must just be an ordinary mortal)—that I'm at least as attached to the value of my own face as I am to the value of yours—it's also that I have similar attachments to the face of other people as well. A gain in anybody's face sends ripples through the entire face-network.

3. The connection between feelings (autonomic responses, somatic markers) and body language is physiologically true, of course, but it hasn't been studied much, and it is talked about even less, so that students may be inclined to resist it. "I never have any of those responses! It's probably just some fancy theory." In my experience women, trained by society to be more aware of feelings and social relations, are much more receptive to the theory of somatic exchange than men. A sociolinguist friend of mine, a very smart man whose work I admire enormously, once spent an hour trying to convince me that my whole theory was crap—or, as he put it, "metaphorical," based on an extension of the phrase "I feel" to some imaginary body function. Meanwhile, his girlfriend kept shaking her head at his obstinacy: "That's what you always say! No feelings! Can't you see it's true?"

4. The notion that our bodies store what we learn in the autonomic nervous system (especially the amygdala and ventro-medial prefrontal cortex) and reminds us of what we know through autonomic responses ("somatic markers") comes from Antonio Damasio, one of the world's leading neurological researchers. He has only been arguing that position himself since the early 1990s, and has met with some resistance from other neurologists; but his view is rapidly taking hold, winning converts. It is controversial and far from a widely accepted fact, in other words; but it is also far from a crackpot theory. The notion that somatic markers are our main channel of social regulation is one I developed out of Nietzsche, Freud, Burke, and Foucault back in Robinson (1991), and applied to Damasio (1994) in Robinson (2003). It has since become a central part of his argument in Damasio (2003).


1. a. In Brown and Levinson's terms, lateness restricts another person's freedom of action, and thus threatens negative face; apologizing for lateness would thus be negative politeness. Apologizing takes responsibility for a face-threatening act and thus puts you in the wrong, makes you look bad, and thus threatens positive face; reassuring a person that there's no need to apologize would be positive politeness. The joke in the man's response to "I'm twenty minutes late myself," of course, is that she means she's late coming to the restaurant, but the man takes her to mean her period is late and she might be pregnant: "Oh, let's not start worrying about that yet!" Instead of taking her confession of her own lateness to be reassurance (a response to his apology) and thus positive politeness, the man takes it to be a statement of worry, possibly even an admission of a possible threat to his freedom of action (in patriarchal tradition a woman's pregnancy "ties a man down" by obliging him to marry her or pay child support if they aren't married, and to suffer the "loss" of his wife and possibly even change diapers if they are). His response, then, would be reassurance that there's no need to worry about possible negative face-threats (yet: twenty minutes is of course an absurdly short time to "be late" for a period), and thus negative politeness. But then the whole joke is, as Graham Chapman would say as the Colonel, unutterably silly.

b. This is, of course, just the beginning of the sketch, when things are just getting started: the face-threats to the customers posed by the restaurant staff's attempts to manage the dirty-fork face-threat to the customers are here just beginning to escalate, and are so far barely noticeable. But the face-threat posed by the waiter's slight overreaction to the dirty fork is that it makes the customer seem like an imperious and demanding person—or, as Mungo the Chef will later say, a "vicious, heartless bastard." The dirty fork threatens the restaurant's face, but it is a very tiny threat that should properly be "made to go away"—nobody should pay too much attention to it. A waiter would normally just quietly replace the fork, possibly say with a slight show of embarrassment "Sorry about that, sir," and that would be the end of it. That way the restaurant's face-threat is quietly managed (eliminated) and no additional threat is created for the customer's face. By making a big fuss about it, the restaurant in this sketch implicitly puts the customer in the wrong, makes him look bad, threatens his positive face: as a result of his "petty, vicious, heartless quibbling," as Mungo the Chef will later put it, the head waiter will be disturbed (in this first section), the entire washing-up staff will be fired (in the next), the manager will sink into despair (in the third), and finally, in the last segment, the manager will commit suicide, the head waiter's war wound will incapacitate him, and Mungo the Chef will go homicidally insane.

2. a. The deeper threat, of course, is that the restaurant is portraying the customer as making a huge fuss over nothing, and thus making him look bad in their eyes, and in other customers' eyes. This does point to the relational nature of face: afraid that the customer will feel bad because he has been given a dirty fork, and thus lose positive face, and as a result will think badly of this fancy restaurant, causing the restaurant to lose positive face, the serving staff go overboard in their "polite" (face-threat-managing) response; by working too hard to protect the customer's face, the serving staff destroy the restaurant's face, which in turn poses a serious threat to the customer's face, because he is made to look like an ogre who is intent on making the restaurant look bad. The presumed audience for this imaginary "being made to look like a ogre" is the other restaurant clientele, of course; significantly, though, in the sketch nobody looks up or over at the table where the action is happening: people yelling, committing suicide, suffering physical breakdowns, brandishing cleavers, being tackled, etc. Clearly the customer's fear of losing face in front of the other customers in the restaurant is imagined—but no less powerful for that.

b. In the head waiter's imagination, the dirty fork becomes the cause of huge horrible bad things: the restaurant loses its sterling reputation (and five stars), customers shun it, it loses money, people are let go, he himself loses his job—indeed he may be afraid of losing his job even if the restaurant doesn't go under: he may be afraid he'll be held responsible for the dirty fork and be sacked as the "fall guy." His imagination is humorously out of control, of course; and one of the things his out-of-control imagination does is to focus all blame for these imagined calamities not on the people who didn't wash the fork properly (in his mind they're already sacked) but on the fork itself, which thus becomes dirty, filthy, nasty, smelly, etc.

3. a. It is, of course, "his" restaurant. He may not be the owner, but he's the manager, the person with ultimate responsibility for the restaurant's success. Any blemish on the restaurant's reputation is a blemish on his own—a mark of his personal failure. Because he feels that the restaurant as a whole has threatened the customers' face, and he personally as the managerial embodiment of the restaurant has as well, he offers an explanation of the restaurant's collective "bad patch" and his own personal hard times ("I haven't been too well recently") as an apology for the failure and the face-threat that failure presumably poses to the customers. Structurally, in fact, his apology is very similar to the man's apology to the woman for being late: there is a threat to someone else's face which must be managed by an apology. But see the answer to b for differences between the two.

b. The significant differences between the man's apology to the woman and the manager's apology to the man are (a) that the man doesn't go on and on about the reasons for his lateness (bad traffic, rain, car broke down, etc.), which might have led the woman to suspect he was actually late for some less excusable reason (he was cheating on her with another woman), and is babbling on out of nervousness; and (b) that the man is intimately involved with the woman and might be expected to provide some sort of deeply personal explanation for his actions had he done something terrible, and the restaurant manager is a total stranger and would normally be expected to keep his explanations minimal and impersonal. And we do experience unwelcome and inappropriate confidences and confessions as face-threatening: they give us too much social power over a stranger, put us in the uncomfortable position of being able to do too much damage to a person we don't know and have no reason or desire to damage or help.

4. a. This sketch is a humorous exaggeration of Goffman's (1967: 10) apophthegm that "while [our] social face can be [our] most personal possession and the center of [our] security and pleasure, it is only on loan to [us] from society; it will be withdrawn unless [we] conduct [ourselves] in a way that is worthy of it." Face, in other words, is precious and precarious: it matters enormously to each individual but is "regulated" (often, it seems, capriciously or whimsically) by the group. The tiniest little deviation from socially appropriate behavior can threaten it; and a social ripple effect can escalate threats until they become disastrous. The ripple effect in this sketch stems from the restaurant staff's oversensitivity to face-threats, exaggerated worry about the threats the dirty fork poses directly to the customers' and indirectly to the restaurant's face; their extreme worry about face-loss causes them to escalate the incident to calamitous proportions.

b. Structurally, in terms of the management of face-threats in general (leaving intensity out of the equation), "Never kill a customer" is identical to "Never give a customer a dirty fork." It does manage a face-threat, obviously: if Mungo the Chef really had killed the customer (or a waiter had gone ahead and given the customer the dirty fork), the customer would have lost face—literally, in Mungo's case, since his cleaver is aimed directly for the customer's face. But also, and more important, it isn't the sort of thing a head waiter would want to say to a member of his staff in the presence of a customer, and the customer's physical presence when such an instruction was given would tend to threaten everyone's face: the reprimanded person's, because it makes him look like an idiot in front of the customer; the customer's, because he thus becomes the cause of the reprimanded person's face-loss; and the reprimander's, because he is revealed in front of the customer as someone so little in control of his staff that he has to reprimand them in public. As the management of face-threats go, of course, "Never kill a customer" is just slightly more intense than "Never give a customer a dirty fork"!

More: Teacher's Guide index

General: Structure | Pedagogy | Dialogues

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

Last Modified: