Speech Acts and Literary Theory

Sandy Petrey

• Publisher: Routledge
• Year: 2016
• ISBN:
  9781134983735
• Edition: 1
• Topic: Languages & Linguistics
• Subtopic: Linguistics
• Index: Languages & Linguistics

This book, first published in 1990, combines an introduction to speech-act theory as developed by J. L. Austin with a survey of critical essays that have adapted Austin's thought for literary analysis. Speech-act theory emphasizes the social reality created when speakers agree that their language is performative - Austin's term for utterances like: "we hereby declare" or "I promise" that produce rather than describe what they name. In contrast to formal linguistics, speech-act theory insists on language's active prominence in the organization of collective life. The first section of the text concentrates on Austin's determination to situate language in society by demonstrating the social conventions manifest in language. The second and third parts of the book discuss literary critics' responses to speech-act theory's socialisation of language, which have both opened new understandings of textuality in general and stimulated new interpretations of individual works. This book will be of interest to students of linguistics and literary theory.

Part II
Applications

4
Austin and Searle Together and Apart

"How in the world do I handle John Searle?" has been running through my mind as a nagging as well as vaguely poetic question ever since I first had the idea of writing this book. There was no hesitation over whether to include him. His work, constantly cited by literary critics as well as philosophers, has towering stature among scholars interested in speech acts. My problem was how to include him, for his great contributions to spreading Austin's ideas through the academic community have from the beginning been inseparable from explicit and implicit departures from Austin that impede adaptation of speech-act theory to literary purposes.

So I wrote the first part of this book without mentioning Searle even though there were many points where his work deserved (positive or negative) introduction. My knowledge of his most significant arguments, those that repeat and refine Austin's work, was invaluable as I organized my own exposition of How to Do Things with Words in the first part of this book. I credit him here instead of there because earlier introduction of his developments of Austin would have meant introducing criticisms as well, and those criticisms are irrelevant to the foundation Austin laid.

The points on which Austin and Searle disagree, and the scholars who share my opinion that their disagreements should almost always be resolved in Austin's favor, will acquire sharper focus as this book proceeds. At present, I want to consider what students of literature have taken from Searle's work and how he has himself addressed literary topics. Although Searle has suggested that literary critics always get it wrong, his own work gets it right often enough to deserve a chapter in this study of speech acts in literary criticism.

Searle's most serviceable refinement of Austin is his systematization of speech-act types, which substitutes a five-part schema for the complex (not to say chaotic) classification Austin tentatively put forward in the last of the lectures collected in How to Do Things with Words. Almost all critics who have looked at the different kinds of illocution represented in literature have taken Searle's schema over Austin's list, for the advantages are multiple. Searle's taxonomy (1979; pp. 12-20) posits five varieties of illocution. The constative becomes the "assertive," and Austin's multiple categories of the performative are reduced to four: directives (I order, I beg), commissives (I promise, we pledge), expressives (I apologize, Thanks a lot) and declarations (the whole set of institutionally based exercises of verbal authority already extensively discussed here).

Searle systematically lays out his version of illocutionary analysis in Speech Acts: An Essay in the Philosophy of Language, which appeared in 1969. For students of literature, the thorniest aspect of Speech Acts is its proclivity for the abstract notation of logical analysis—Searle represents the general form of illocution not by discussing society and conventions but by writing "F(p)," for example—and for positing the simplified context on which this notation depends. Although Searle has forcefully rejected the extra-contextual formalization favored by transformational grammarians like Jerrold Katz, he has also implicitly agreed with Katz that speech-act theory should follow the lead of the natural sciences and assume impossibilities like a frictionless plane to explain the actualities of the world. This is from Searle's introduction to his description of the promise, which his Speech Acts takes to represent all illocutions.

I am going to deal only with a simple and idealized case. This method, one of constructing idealized models, is analogous to the sort of theory construction that goes on in most sciences, e.g., the construction of economic models, or accounts of the solar system which treat planets as points. Without abstraction and idealization there is no systematization. (56)

Despite Austin's refusal of the simple situations envisaged in logical theory, Searle takes them as the starting point for his explanation of how to do things with words.

Since literature produces situations that logical theory customarily ignores, literary critics have with reason been leery of programmatic devotion to abstraction, idealization, and systematization. Fortunately, though, Searle also provides concrete principles of broad value for literary criticism. He shares Austin's conviction that to study language as form without attending to the form's function within human lives is a self-defeating futility. Even though he takes construction of economic models as an example to be emulated, for example, Searle also sharply criticizes economists who classify a community's currency without noticing that money helps the community's members get through the day (17). Despite his commitment to the sometimes hermetic discourse of analytic philosophy, despite his reservations about literary critics' ability to handle speech-act paradigms, Searle's vision of language in the world holds important lessons for textual analysis.

One of the pivotal lessons in Speech Acts is the distinction between brute facts and institutional facts, a needed expansion of the idea Austin expressed as the difference between kicking a ball and kicking a goal. Brute facts, wholly independent of conventions, are best exemplified by physical realities: what balls do when a foot hits them with a certain momentum, what happens to water at a certain temperature. Institutional facts are on the contrary conventionally determined. "They are indeed facts; but their existence, unlike the existence of brute facts, presupposes the existence of human institutions" (Searle 1969; p. 51). When a ball scores a goal, the brute fact of the momentum imparted to it by a foot is of a different order from the institutional fact that it changes the relative standing of the two sides in the game.

Searle's other examples of institutional facts include a couple's marriage, a defendant's conviction, the result of a baseball game, passage of a Congressional act. In each, something real takes its reality solely from the socio-institutional dynamics prevailing at a given time and place. Institutional facts are in no way less substantial or important than brute facts. It's a brute fact that George Bush has two big toes, an institutional fact that he was elected President of the United States of America. For the world at large, probably even for George Bush, the institutional fact is easily the more consequential. That illocutionary force is an institutional fact in no way diminishes it.

The category of institutional facts is closely bound up with what Searle calls "constitutive rules," those that can't be violated under penalty of abolishing the activity to which they apply. Constitutive rules are opposed to "regulative rules," which may govern but do not constitute a form of behavior. When two teams from the National Football League meet on a Sunday afternoon, the home team has first choice of what color jersey to wear. That rule is regulative; nothing about the game of football would be different if the visiting team got to choose jerseys. The rules governing what counts as a valid play, however, are constitutive. Ignore or alter them and the game is no longer what it was.

Constitutive rules are institutional rather than brute facts, and their importance shows that brute facts alone are woefully insufficient. Without constitutive rules, we would not only lose the games we play, we would undermine the lives we lead. Rules give us social existence as well as touchdowns, the ability to coexist as well as the chance to checkmate. The things they constitute are necessary to our own survival as well as to that of our games.

Searle uses concepts like constitutive rules to repudiate inherited ideas about language's primary responsibility to describe the brute facts of the extra-linguistic world. His dismissal of the descriptive fallacy is decisive: "These are the only two plausible ways of applying the theory of descriptions to all kinds of illocutionary acts. Neither works. The theory should, therefore, be abandoned" (162). When institutional facts are at issue, as they always are with human beings, we get nowhere by proceeding as if brute facts were all.

Like Austin, Searle shows that many standard philosophical principles become inapplicable when we come to see the distinction between language in the world and language in the abstract. Since Wittgenstein, a tautological assertion like "Either it's raining or it's not" has been generally considered void. Because tautologies are always true, no tautology says anything worth mentioning; because none says anything, all tautologies are equivalent and interchangeable. Searle contends that "nothing could be further from the truth" (124) than to assume that tautological identity exempts us from considering the existential variety inseparable from actual use of language, including tautological language.

His evidence is the clear distinction—the obvious lack of equivalence and interchangeability—between two other tautological sentences, "Either he's a Communist or he isn't " and "Either he's a Fascist or he isn't." While it's true that no assertion is made in one or the other (both are true about every "he"), the first tautology suggests the possibility of left-wing extremism, the second that of right-wing extremism. Each therefore performs the speech act of suggestion, and few politicians would be indifferent to the suggestion made.

This contrast between tautologies is convincing because the Fascist and Communist examples introduce the speech-act matrix of sociohistorical specificity in a way Wittgenstein's raining example fails to do; the facts of Fascism and Communism are clearly institutional rather than brute, and their rhetorical impact no less clearly depends on the conventions Austin codified in Rule A.l. Interestingly, though, Searle does nothing with this feature of his argument, which is typical of his general strategy to pay little attention to the social specificity of illocution. An accepted conventional procedure with conventional effect was for Austin conceptually as well as alpha-numerically first. In Searle, it's less important.

As we saw, How to Do Things with Words deduces all the other rules for performative force from Rule A.1, Searle's Speech Acts takes over the idea of a principal condition that "determines the others" (69), but now what is called the "essential rule" in every illocution is what the particular utterance "counts as" (63, 66—67). "I'll be there at seven" counts as a promise if it commits me to be there at the time I said, but it might also count as a question if I say it in such a way that it leads you to tell me what time you're coming, or as a request if what I'm after is to get you to come at the same time. Although Searle's essential rule is patently an institutional fact and as such depends on conventions, he chooses not to foreground conventional determination with anything like Austin's assertiveness.

I see this difference as related to Searle's decision to direct attention to individual performance rather than collective production of speech acts. It's apparent throughout most of Searle's writings that what an utterance counts as is not some absolute given but rather the effect of a community establishing and enforcing its particular form of coexistence. The concluding words of Speech Acts make the point strongly: "speaking a language—as has been the main theme of this book—consists of performing speech acts according to rules, and there is no separating those speech acts from the commitments which form essential parts of them" (198). The rules and commitments accepted whenever we speak come from the conventions that apply because we always speak in definite sociohistorical circumstances.

Nevertheless, Searle's emphasis on what an utterance counts as over the conventions through which it comes to count has the effect of devaluing the Austinian dialectic between illocutionary force and social identity. This change in emphasis leads to curious instances of completely asocial performatives. For example, while discussing declarations in "A Taxonomy of Illocutionary Acts," Searle admits a divine exception to the requirement that only people with definite positions in social institutions can declare war, declare a couple married, or produce any of the other reality-transforming utterances institutions make possible: "there are supernatural declarations. When, e.g., God says 'Let there be light' that is a declaration" (1979; p. 18). To the contrary, divine beings are totally incapable of performative speech, which is accomplished solely by the protocols organizing human communities. God stands outside those communities, conventions are radically inapplicable to Him or Her, the norms of social interaction fail to reach heaven. "Let there be light" and a human sentence that also does what it says have nothing else in common.

Confusion of the illocutionary and the supernatural is consistent with Searle's general move away from the set toward society in How to Do Things with Words. God's declarative power is part of an argument leading to this distancing of Searle's version of speech-act theory from Austin's: "Austin sometimes talks as if all performatives (and in the general theory, all illocutionary acts) required an extra-linguistic institution, but this is plainly not the case" (18). If God can do without institutional support, the passage implies, so can analysis of the specifically human force called illocution.

Austin's suggestions are valid or not depending on our understanding of "institution." If the word designates a specific organization with full bureaucratic credentials, then Searle is correct to distinguish between declarations like those producing a law or a state of siege from more homespun performatives like those making a bet or an apology; the institutional investiture required for a declaration to be felicitous isn't necessary for an apology to be made.

But if we understand institutions in the broader sense—in just the sense brought forward in Searle's category of institutional facts—then it plainly is the case that the way "Austin sometimes talks" represents the way Austin always thinks. In the looser sense, extra-linguistic institutions permeate not only bureaucratic organization but all the protocols—all the institutional facts—establishing and preserving a social formation. Institutions can well be understood as neither more nor less than the sum of the conventions that invariably make speech act.

Searle's peremptory announcement that Austin is "plainly" wrong about extra-linguistic institutions may indicate discomfort with "extra-linguistic" as well as "institution." Austin and Searle certainly share an understanding of speech acts as deriving from social organization rather than from qualities inherent to language. Yet the extralinguistic almost always attains greater prominence in Austin. The difference between Rule A.1 in How to Do Things with Words and the essential rule in Speech Acts is that between the prior social reality necessary for speech to act and the present definition of the act performed. In one case, we must consider a conventional interaction as the first step toward validly apprehending a linguistic utterance; in the other, we begin with the utterance. The distinction is subtle but consequential. Despite Austin and Searle's fundamental agreement, they assign different value to the collective construction of illocutionary felicity.

Searle's diminished concern with society has strong impact on his definition of the problematics appropriate to a speech-act consideration of literature. The essay called "The Logical Status of Fictional Discourse" follows the lead of a brief discussion in Speech Acts by taking the primary question posed by the intersection of illocution and literature to be not how literature is related to society but how fiction is different from lies. Comparing a passage from the New York Times to a passage from Iris Murdoch's novel The Red and the Green, Searle points out that both passages consist principally of assertions. Yet neither the essential rule for assertions nor any derivative rules apply to the novel, whereas all apply to the newspaper. Searle's goal is to explain the "logical status" of this apparent breakdown of the constitutive rules for asserting felicitously.

To reach that goal, Searle appropriates the banal idea that authors are "pretending" to say things. Fiction consists of pretending without any wish to be taken seriously, and in Searle's view production of what looks like lies but isn't results from fiction's articulation with a set of conventions that "suspend the normal requirements" (66) imposed by the rules applying to assertions elsewhere. Searle calls the two sets of conventions, one committing the speaker to be truthful, the other removing that commitment, "vertical" and "horizontal." His argument is that normal, vertical conventions continue to function as the source of meaning even though the fictional, horizontal sort insulates the author from responsibility for what's being said. "What distinguishes fiction from lies is the existence of a separate set of conventions which enables the author to go through the motions of making statements which he knows to be not true even though he has no intention to deceive" (67).

This set of pretending conventions allows us to distinguish logically between statements like "Sherlock Holmes lived in London" and "Sherlock Holmes had a blonde wife." In one sense, neither assertion is true, for there never was a Sherlock Holmes. But in the sense made possible by recognizing specifically fictional conventions, it's true that Sherlock Holmes lived in London and false that he had a wife. We can refer to fictional worlds for the same reason we can refer to actual worlds, because illocutionary understanding gives us the capacity to do so. It's of no special interest that the enabling conventions are not identical; what matters is that they are indeed enabling.

Yet I have real problems understanding how this will help me as a reader of fiction. Assuming that there do exist pretense conventions analogous to conventions for promising and declaring, how do we internalize and apply them? While reading a parti...

Table of contents

• Cover
• Half Title
• Title
• Copyright
• Original Title
• Dedication
• Contents
• Acknowledgements
• I Beginnings
• II Applications
• III Challenges
• Bibliography
• Index