Total Speech: An Integrational Linguistic Approach to Language (Post-Contemporary Interventions) - Softcover

Toolan, Michael

Softcover

ISBN 10: 0822317907 ISBN 13: 9780822317906

Publisher: Duke University Press, 1996

Synopsis

Units, rules, codes, systems: this is how most linguists study language. Integrationalists such as Michael Toolan, however, focus instead on how language functions in seamless tandem with the rest of human activity. In Total Speech, Toolan provides a clear and comprehensive account of integrationalism, a major new theory of language that declines to accept that text and context, language and world, are distinct and stable categories. At the same time, Toolan extends the integrationalist argument and calls for a radical change in contemporary theorizing about language and communication.
In every foundational area of linguistics—from literal meaning and metaphor to the nature of repetition to the status of linguistic rules—Toolan advances fascinating and provocative criticisms of received linguistic assumptions. Drawing inspiration from the writings of language theorist Roy Harris, Toolan brings the integrationalist perspective to bear on legal cases, the reception of Salman Rushdie, poetry, and the language of children. Toolan demonstrates that the embeddedness of language and the situation-sensitive mutability of meaning reveal language as a tool for re-fashioning and renewal.
Total Speech breaks free of standard linguistics’ fascinated attraction with “cognitive blueprints” and quasi-algorithmic processing to characterize language anew. Toolan’s reflections on the essence of language, including his important discussion of intention, have strong implications for students and scholars of discourse analysis, literature, the law, anthropology, philosophy of language, communication theory, and cognitive science, as well as linguistics.

About this title

About the Author

Michael Toolan is Associate Professor of English at the University of Washington.

From the Back Cover

""Total Speech "provides an entirely original discussion of some of the most central and controversial issues in the theory and philosophy of language. This book is not just a synopsis of the integrationalist school of thought, it is a major contribution to the development of that thought and to its emergence as a major force in American and European thinking about language."--Talbot Taylor, College of William and Mary

Excerpt. © Reprinted by permission. All rights reserved.

Total Speech

An Integrational Linguistic Approach to Language

By Michael Toolan

Duke University Press

Copyright © 1996 Duke University Press
All rights reserved.

ISBN: 978-0-8223-1790-6

Contents

Acknowledgments, 
Introduction, 
1 On Inscribed or Literal Meaning, 
2 Metaphor, 
3 Intentionality and Coming into Language, 
4 Further Principles of Integrational Linguistics, or, On Not Losing Sight of the Language User, 
5 Relevance in Theory and Practice, 
6 Repetition, 
7 Rules, 
References, 
Index, 

CHAPTER 1

On Inscribed or Literal Meaning

It might seem that the thesis that there is no such thing as literal meaning is a limited one, of interest largely to linguists and philosophers of language; but in fact it is a thesis whose implications are almost boundless, for they extend to the very underpinnings of the universe as it is understood by persons of a certain cast of mind. – Stanley Fish, Doing What Comes Naturally

It is crucial to distinguish between what a sentence means, that is, its literal sentence meaning, and what the speaker means in the utterance of the sentence. We know the meaning of a sentence as soon as we know the meanings of the elements and the rules for combining them. – J. Searle, "Literary Theory and Its Discontents"

Preliminaries

This chapter, together with the one following, discusses literalness, metaphoricality, and figurative language: what they are and how they emerge, are understood, and are related. Although the first chapter is "on" literal meaning while the second is on metaphor, they are not on these subjects respectively; that is to say, chapter headings notwithstanding, literalness and metaphoricality are treated not as essential kinds or strictly separable phenomena but as names that are artifacts of convenience. The categorization separating literal meaning from metaphor, and the substantial delimitation of one in relation to the other, is a deeply embedded convenience of literate Western culture, a culture within which contemporary theorizing about language has its place. Nor is the term convenience intended to be denigratory: its close kinship to more authoritative terms such as purpose, relevance, and intention needs acknowledging. All are touchstones in the larger argument that is to emerge. The larger argument is that language should be viewed essentially as other-oriented situated behavior: but other oriented not at the expense of self-interest but by way of calculated pursuit of reasonable self-orientedness.

The focus on literal meaning here, and on metaphor in the next chapter, reflects a prevalent classification of utterance interpretation in such fields as pragmatics and psycholinguistics. But my treatment is intended to be unitary, classing the literal and the metaphoric as aspects of a single phenomenon of sense making in language: the two chapters should be read as one, with the subsuming title and topic of "Figuration," for it will be argued that conceptualizations in terms of literalness versus metaphoricality are themselves essentially a form of figuration: the ascribing of "literal meaning" is a kind of troping, and metalinguistic exercises of separating out literal sheep from metaphoric goats (or wolves) is a parallel "figuring things out."

In this chapter, I argue that, although a conceptualization of literal meaning as the basic, determinate, and context-free meaning of words and sentences is necessary for standard linguistic treatments of the semantics and pragmatics of a language, in practice no such domain of context-free meaning exists. The standard linguistic procedure is to concede that some kind of "background" is a necessary frame for literal meaning but also to represent that background as neutral and inconsequential. I review and critique some recent discussions that attempt to maintain an account of literal meaning as "context-transcending" meaning. I argue, on the contrary, that literal meaning is itself a highly contextualized notion, that it is a cultural and ideological construct very much designed to characterize some language practices as orderly, authorized, and authoritative (and others as not so); it is therefore well suited to and reflective of societal interests in literacy, order, and authority.

In the introduction to an influential collection of articles in pragmatics (Searle, Kiefer, and Bierwisch 1980), the editors compare and contrast different treatments of the term pragmatics, and notions (of denotation, sense, and use of linguistic expressions) surrounding it, in three analytic traditions: formal philosophy, linguistic semantics, and ordinary language philosophy. Despite major differences, they feel able to conclude: "In all three traditions something like a notion of literal meaning is essential, and some contrast between literal meaning and speaker's utterance meaning seems essential to any account of language. Speaker's utterance meaning may differ from literal meaning in a variety of ways. Speaker's meaning may include literal meaning but go beyond it, as in the case of indirect speech acts, or it may depart from it, as in the case of metaphor, or it may be the opposite of it, as in the case of irony" (p. xi).

Literal meaning is essential and foundational, then, in this account. At the same time, it is not a plane of meanings that is context free or interpretable without reference to a background. The literal meaning of an expression is never meaning in a "zero context," Searle argues in the same volume, but rather always is meaning relative to background assumptions that cannot themselves be part of the expression's meaning. Searle proceeds to discuss a series of sentences in which the "same" verb, cut, is used literally and yet determines different sets of truth conditions in each case (cutting the grass, the cake, the cloth, etc.):

The reason that the same semantic content, "cut," determines different sets of truth conditions in these different sentences... derives not from any ambiguity of a semantic kind, but rather from the fact that as members of our culture we bring to bear on the literal utterance and understanding of a sentence a whole background of information about how nature works and how our culture works. A background of practices, institutions, facts of nature, regularities, and ways of doing things are assumed by speakers and hearers when one of these sentences is uttered or understood.... My knowledge that cutting grass is a different sort of business from cutting cakes is part of this larger system of knowledge. (Searle 1980: 227)

By this postulation of a required "background" in the interpretation of a literal sentence, Searle appears able to protect the notion of the invariance of literal meaning. The literal meaning of cut is said not to change: only the truth conditions (and, shaping these, the background assumptions) do.

However, if we adopt Searle's arguments for the necessity of background assumptions, certain issues remain unclear. The first of these concerns the literal meaning of cut: nowhere does Searle offer a formulation as to what this is, although he insists on its identity across the five sentences he discusses. His own characterization of cut as "a physical separation by means of the pressure of some more or less sharp instrument" he dismisses as "very misleading." The same example, of cutting the grass (vs. cutting the cake) is used in Searle (1994a: 640), where again it is claimed that the literal meaning of cut is univocal and determinate – "We do not have different definitions of the word 'cut,' corresponding to these two occurrences" – and that we secure correct understanding of such utterances by interpreting them against "a whole cultural and biological Background," which is of a very different order than literal meaning and linguistic factors. Evidently Searle would not accept the idea that there are certain established commonest senses of the verb cut (such as "to crop or shorten," "to divide into parts," etc.), along the lines that a detailed dictionary reflects. Rather, there is one, literal, all-purpose meaning of cut – still not spelled out in Searle (1994a) – and then a depth of "background" contextualizing that jointly enables us to understand or reject particular utterances in which the word is used. But if the literal meaning of cut is such a feeble guide to the interpretation of the word in contexts of use, one is prompted to ask why – and whether – it is "crucial" to separate out literal or sentence meaning at all. It is not clear that the exercise is crucial for actual users of a language. In what way would communication be impaired if users began with the assumption that every word was associated and associatable with a range of senses, from common to rare, and that some selection from among these was to be expected in every case of situated utterance? "Sentences have to have standing, conventional sentence meanings in order that we can use them to talk with," writes Searle (1994a: 646), but what is the justification, other than the pressure for an elegant simplicity of modeling, for assuming that such standing meanings are single – that any utterance will have just one "standing sentence meaning" associatable with it? If we question that unificationism, we are questioning also the inflated status of the notion literal meaning. It is no accident that Searle's (1994a) "reiteration" defends the literal or sentence meaning versus speaker meaning distinction alongside a series of other ones (principally, type vs. token; sentence vs. utterance; use vs. mention). Each of these binary distinctions entails the others; they are a mutually justifying network of assumptions. The very first rejoinder from an advocate like Searle to a critic of any one of these notions (e.g., literal meaning) is that the critic has neglected or misunderstood one of the related orthodox distinctions, such as that between types and tokens. But for radical critics such as integrationists the point is rather that a quite different view of types and tokens, too, is necessary.

The second point not fully resolved concerns the means of identification of the intended background assumptions determining an utterance's meaning: What might these means be, and how are they used by a hearer? But an outline of standard linguistic views of literal meaning and utterance meaning is desirable before these issues are pursued more fully.

Literal, Conventional, and Utterance Meaning

An informal characterization of literal meaning might be the following:

The conventional meanings of words of a language and the meanings of sentences in that language, where any sentence meaning is derived from a complex synthesis of the meanings of its composite words.

Such a characterization seeks to distinguish literal meaning from conventional sentence meaning. The distinction is clear in the contrast between two interpretations of the sentence

John got the sack.

The conventional interpretation, sensitive to the usual idiomatic use of get the sack, might offer the paraphrase "John was fired." But the literal meaning purports to express more "basic" or foundational meanings, those residing first in a sentence's component words, and so would produce the paraphrase "John fetched/received the (nonrigid) container." One assumption is that idiomatic meanings are some kind of overlay of the literal meaning, such that a conventional sense, attached to a fixed sequence of words taken as an unanalyzed unit, supplants the literal interpretation of the component words. Idiomatic conventional meanings thus occupy an in-between area, being neither literal nor fully metaphoric.

The definition of literal word and sentence meanings above amounts to an appeal, in determining meanings, to the information enshrined in a reliable grammar and dictionary of a language. But such an appeal may often yield several possible literal meanings for a sentence. Consider the following:

Can you pass the salt?

The literal meaning of the sentence may vary considerably, given the multiple conventional meanings of can and pass, as should be evident if we imagine the sentence used in a chemistry laboratory or in a medical interview with a patient, besides the dinner-table setting. Thus, polysemy too (in addition to idiomaticity) overlays literal meaning. It is important to note, however, that even in the last-mentioned context of use, at the dinner table, the literal paraphrase will be "Are you able (is it possible for you) to convey the salt?" that is, something rather different from its probable conveyed or speaker meaning, which would not be an inquiry about ability or possibility at all but rather a polite request. Thus literal meanings and conveyed (or speaker or utterance) meanings are in principle quite different from each other. Indeed, much contemporary linguistic pragmatics purports to explain how language users seemingly proceed inferentially from decontextualized literal meanings of sentences to their conveyed meanings in particular contexts of utterance.

Even thus situated, however, the characterization of literal meaning given above retains many problems. The definition refers to meaning synthesis, a variant term for the projection, combination, or predication analysis that semanticists have postulated as the necessary means for capturing the semantic representation of sentences. But just how such a synthesis might take place in a language user's mind – and even the assumption that it does at all – remains highly controversial. Before pursuing the issues involved, a more formal definition of literal meaning may now be in order, to set beside the earlier informal one:

Literal meaning is always language-specific. The meaning of an utterance of a sentence s of language L is said to be literal if it is only composed of the meanings of the words and phrases in s in accordance with the syntactic conventions in L. It is, however, not always clear what the meaning of the words and phrases in s actually is, because the words may have different meanings, and because their meaning often depends on the context c of the utterance. For example, the meaning of indexical expressions such as I, here, now in an utterance, or the meaning of anaphoric expressions such as he, then, that in an utterance depends on c. Even words such as enough,hut, otherwise, big, can, and many others, have a context-dependent meaning, i.e., their meaning includes a context variable x. (Wunderlich 1980: 298)

This formulation introduces the crucial issue of context dependency. The determination of an utterer's meaning in using a particular sentence can be seen to rest, typically and heavily, on principles or phenomena that are in a sense both semantic and pragmatic. The expressions are grammaticized, distinctions built into the language (hence semantic); but their specific sense and reference can be determined only in relation to some concrete context. Within the latter area fall (at least) deixis and anaphora and, probably, presupposition. But there are in addition other relatively systematic principles at work, which it is arguable we use in the determination of utterance meaning: within this area lie proposals concerning conversational implicature and cooperativeness and speech act theory (or, as a recent counterproposal, relevance theory) and the designs of talk identified by conversational analysts.

Searle on Literal Meaning Relative to a Background

In fact Wunderlich's description of literal meaning would seem to leave very few sentences sufficiently context free to have a purely literal meaning. While that might suit antiliteralists, it is clearly not what Searle and other speech act theorists have in mind when they speak of literal meaning. They might complain that, in writing of "the meaning of indexicals" and similarly, Wunderlich has not distinguished sentence from utterance: there is a specifiable literal meaning of words such as I, here, and enough, and without this the derivations and determinations of particular context-bound referents and values for these words, in actual utterances, would be impossible. Or, as Searle has put it, literal meaning as such is not affected by the particular reference of an indexical or the truth or otherwise of a particular statement in context: the truth conditions may vary, but the literal meaning cannot.

As indicated above, for Searle literal meaning is not entirely context free. He has emphasized a certain "contextualizedness" of literal meaning: "The notion of the literal meaning of a sentence only has application relative to a set of background assumptions, and furthermore these background assumptions are not all and could not all be realized in the semantic structure of the sentence in the way that presuppositions and indexically dependent elements of the sentence's truth conditions are realized in the semantic structure of the sentence" (Searle 1979: 120). But this necessary background is hardly of the kind stipulated as required by thoroughgoing contextualists. For example, Searle has subjected the sentence The cat is on the mat to detailed scrutiny. He imagines the cat and mat floating freely in outer space; or, on earth, suspended separately but contiguously from wires; or with the mat poking stiffly out of the floor and a drugged cat perched on the elevated end of it. It transpires that whether it is literally true to say that the cat is on the mat in any such case depends on context-specific assumptions as to what might constitute being on the mat. Searle offers such outlandish contexts in order to advance his thesis that the literal meaning of a sentence and its associated determinate truth conditions have application only relative to potentially variable background assumptions that are quite different from the familiar semantic/pragmatic parameters such as indexicality, change of meaning, and presupposition. Furthermore, these background assumptions are too numerous to have a place in the semantic structure of the sentence, and, besides, the very statement of those assumptions, in literal terms, relies on yet other background assumptions in order to be intelligible.

(Continues...)

Excerpted from Total Speech by Michael Toolan. Copyright © 1996 Duke University Press. Excerpted by permission of Duke University Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Edition details

• Publisher: Duke University Press
• Publication date: 1996
• Language: English
• ISBN 10: 0822317907
• ISBN 13: 9780822317906
• Binding: Paperback
• Number of pages: 352