University of Southern California Dissertations and Theses

Rhetoricians On Language And Meaning: An Ordinary Language Philosophy Critique

(USC Thesis Other)

Rhetoricians On Language And Meaning: An Ordinary Language Philosophy Critique

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content I 70-23,187 STEWART, John Robert, 1941- RHETORICIANS ON LANGUAGE AND MEANING: AN ORDINARY LANGUAGE PHILOSOPHY CRITIQUE. University of Southern California, Ph.D., 1970 Speech University Microfilms, A X E R O X Company, Ann Arbor, Michigan © Copyright by JOHN ROBERT STEWART 1970 RHETORICIANS ON LANGUAGE AND MEANING: AN ORDINARY LANGUAGE PHILOSOPHY CRITIQUE S by John Robert Stewart A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Speech) January 1970 UNIVERSITY O F S O U T H E R N CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 0 0 0 0 7 This dissertation, written by under the direction of his..... Dissertation Com mittee, and approved by all its members, has been presented to and accepted by The Gradu ate School, in partial fulfillment of require ments for the degree of D O C T O R OF P H I L O S O P H Y Joim.JR.okfirt.Siswaxt. Dean DISSERTATION COMMITTEE . CONTENTS \ Chapter Page I. INTRODUCTION ................................. 1 Statement of Problem Limitations of Study Method of Study Review of Literature Preview of Dissertation II. RHETORICIANS AND THE NATURE OF LANGUAGE AND MEANING............................... 17 The Nature of Language Language Is Fundamentally a System of Symbols Apparently Non-symbolic Words Are Also Symbolic The Nature of Meaning The Referential Theory The Ideational Theory The Behavioral Theory Other Approaches to Language and Meaning A Literary Approach Existential and Phenomenological Approaches Approaches in Philosophy and Rhetoric Summary III. ORDINARY LANGUAGE PHILOSOPHERS AND THE NATURE OF LANGUAGE AND MEANING............. 84 Classical and Medieval Backgrounds Nineteenth-Century Backgrounds Modern Backgrounds Russell and Moore Wittgenstein and Logical Positivism The Common Thread: Words Are Fundamentally Names "Wittgenstein II": Meaning Is Use in the Language ii Chapter Page Post-Wittgenstein Ordinary Language Viewpoints Language Is Not a Calculus Meaning Is Not Naming Generalizations about Language Are Not Valid Language-using Is Ordinary Behavior Ordinary Language Method Summary IV. LANGUAGE AND MEANING: SPEECH BEHAVIOR AND U S E ................................... 159 Weaknesses in the Symbol Metaphor Confusion among Symbols Difficulties with Non-symbolic Terms Symbols and the Theory of Abstraction Weaknesses in the Theories of Meaning The Referential Theory The Ideational Theory The Behavioral Theory A Reason for the Symbol Metaphor: The Speech-Language Dichotomy An Ordinary Language Approach to Speech and Language Speech Is More Fundamental Than Language Speech Is Behavior and Language Is Systematic The Problem of Meaning Meaning-in-general Meaningful Speech Behavior Summary V. SUMMARY AND CONCLUSIONS..................... 229 Implications Suggestions for Future Research A LIST OF WORKS CONSULTED........................... 252 iii CHAPTER I INTRODUCTION Presuppositions about language and linguistic meaning are essential to the foundation of speech-communication theory and pedagogy. If, for instance, a rhetorician, whether theorist, teacher, or critic, views language as "the instrumentality by which speakers and writers embody their ideas . . . ,1,1 his understanding of communication probably rests on the assumption that language is something which happens after thought, and the task of the communicator is to generate appropriate ideas and then translate them into words. If, however, a rhetorician sees language as a spe cial kind of psychological stimulus and meaning as a type of response, he is likely to reduce communication to "the per formance of a set of behaviors" or "the transmission or reception of messages."2 And if a rhetorician holds that language is a logical system governed by invariable rules, his communication analysis, evaluation, and instruction will XA. Craig Baird, Rhetoric: A Philosophical Inquiry (New York, 1965), p. 14'1. ~ 2David K. Berio, The Process of Communication (New York, 1960), p. 168. 1 2 emphasize the application of appropriate formulas instead of the study of how discourse works in a given situation. Be cause presuppositions about the nature of language and mean ing are crucial to one's concept of speech-communication/ there is a continuing need to reexamine them in light of the latest and most useful philosophical analysis available. This need is particularly true of speech-communication pedagogy. Authors of speech journal articles and of public speaking, argumentation, and persuasion texts, hereafter referred to as rhetoricians, have based their approach to language and linguistic meaning on the conclusions of semanticists, general semanticists, and a few psychologists. This reliance has led them to the following presuppositions: 1. Language is fundamentally a system of symbols. 2. Words are fundamentally names. 3. The meaning of a word is the object, idea, or behavioral response it represents. These presuppositions are unsound according to the views of ordinary language philosophers. Philosophers such as Ludwig Wittgenstein, Gilbert Ryle, J. L. Austin, P. F. Strawson, and William P. Alston contend: 1. Language is not a unified, consistent system of any kind. 2. Uses of words are almost infinitely various. 3. Meaning is a function of linguistic use. These views are assumed in this study to be the most valid 3 of contemporary concepts of language and meaning and will form the criteria by which the assumptions of rhetoricians will be evaluated. Statement' of Problem Three questions were posed in this dissertation: 1. How do rhetoricians in the field of speech- communication view language and meaning? 2. How do ordinary language philosophers view language and meaning? 3. What conclusions may be drawn from an evaluation of the assumptions of rhetoricians in light of the views of ordinary language philosophers? Limitations of Study Two limitations were imposed on this study. First, the review of works by rhetoricians was not exhaustive. It included selected speech-communication textbooks, and jour nal articles and book reviews published in the eight major regional and national journals since 1953. In that year the first classic work in ordinary language philosophy, Ludwig Wittgenstein's Philosophical Investigations, was published, and the ordinary language approach began in earnest to influence western philosophical thought. The review included one hundred twenty-eight articles,3 eighteen "best 3Since the index and bibliographical resources avail able for periodicals were not entirely satisfactory, the author examined each issue of the journals involved. Thus 4 selling" basic speech-communication texts five additional 403 issues were reviewed: the sixty-five issues of Southern Speech Journal published between March, 1953, and Spring, 1969; the forty-nine issues of Central States Speech Journal published between March, 1953, and Spring, 1969; the sixty- five issues of Western Speech published between January, 1953, and Winter, 1969; the sixty-six issues of the Quar terly Journal of Speech published between February, 1953, and February, 1969; the sixty-five issues of the Speech reacher published between January, 1953, and January, 1969; the sixty-six issues of Speech Monographs published between March, 1953, and March, 1969; the twenty-three issues of roday's Speech published between February, 1961, and Novem ber, 1966; and the four issues of Philosophy and Rhetoric published between January, 1968, and Fall, 1968. The table of contents of each issue was checked care fully, both for relevant articles and for important book reviews, and each possibly germane part of the issue was examined. 4 A list of "best-selling" texts was provided by Mrs. Mary Andersen, whose dissertation, in progress at the Uni versity of Michigan, is titled, "An Analysis of the Treat ment of Ethics in Selected, Contemporary College Public Speaking Textbooks." The list included the following texts: Martin P. Andersen, Wesley Lewis, and James Murray, The Speaker and His Audience (New York, 1964); A. Craig Baird and Franklin H. Knower, General Speech, 3rd ed. (New York, 1963); Virgil L. Baker and Ralph T. Eubanks, Speech in Per sonal and Public Affairs (New York, 1965); William Norwood Brigance, Speech: Its Techniques and Disciplines in a Free Society, 2nd ed. (New York, 1961); Donald C. Bryant and Karl R. Wallace, Fundamentals of Public Speaking, 3rd ed. (New York, I960); Milton Dickens, Speech: Dynamic Communi cation, 2nd ed. (New York, 1963); Giles Wilkeson Gray and Waldo W. Braden, Public Speaking: Principles and Practice, 2nd ed. (New York^ 1963); Kenneth G. Hance, David C. Ralph, and Milton J. Wiksell, Principles of Speaking (Belmont, calif., 1962); James H. McBurney and Ernest J. Wrage, Guide to Good Speech, 2nd ed. (Englewood Cliffs, 1960); Alan H. Monroe and Douglas Ehninger, Principles and Types of Speech, 6th ed. (Glenview, 111., 1967); Robert T. Oliver and Rupert L. Cortright, Effective Speech, 4th ed. (New York, 1961); Raymond S. Ross, Speech Communication: Fundamentals and Practice (Englewood Cliffs, 1965); Alma Johnson Sarett, Lew Sarett, and William Trufant Foster, Basic Principles of Speech, 4th ed. (Boston, 1966); Raymond ( T . Smith, Principles pf Public Speaking (New York, 1958); Otis M. Walter and Robert L. Scott, Thinking and Speaking: .A Guide to Tntelli- laent Oral Communication. 2nd ed. (New York, 1968); Andrew 5 Dasic texts,5 four persuasion texts,6 five argumentation Thomas Weaver and Ordean Gerhard Ness, The Fundamentals and j ’ orms of Speech, 2nd ed. (New York, 1963); Carl H. Weaver, Speaking in Public (New York, 1966); and John F. Wilson and 3arrollC. Arnold, Public Speaking As a Liberal Art (Boston, L964). The more recent editions of Hance, Ralph, and wiksell (1969) and Wilson and Arnold (1968) were used. Mrs. Andersen justified the list in these words: "The Dooks were selected by the following method: on January 6, L967, major book publishers were sent questionnaires requesting a listing of their most recent first, second, and third best-sellers in the area of fundamentals of public speaking. The eighteen texts chosen were the results of answers from the letters. From three of the larger publish ing companies assumed to have wider sales, the second as well as the first book stipulated was selected for analysis. Thus the books studied are believed, for the most part, to oe widely used today in the beginning public speaking classes throughout the country. . . . With the exception of one textbook, all of the books selected for study have been oub-lished during 1960-1968. The books are designed to be ased in the college fundamentals of public speaking classes. They represent a variety of approaches toward communication and persuasion. All recent textbooks could not be included oecause of the time factor involved; nor is such exhaustive ness necessary." (From a personal letter from Mrs. Ander sen, dated December 15, 1968). 5A. Craig Baird, and F. H. Knower, Essentials of Gen eral Speech, 3rd ed. (New York, 1968); Jane Blankenship, Public Speaking: A Rhetorical Perspective (Englewood Cliffs, 1966); Charles T. Brown and Charles Van Riper, Speech and Man (Englewood Cliffs, 1966); E. C. Buehler and Wil A. Linkugel, Speech: A First Course (New York, 1962); Edem, Speech Communication: A First Course (New York, L969) ; and James C. McCroskey, An Introduction to Rhetorical Communication: The Theory and Practice of Public Speaking (Englewood Cliffs, 1968). 6Wallace C. Fotheringham, Perspectives on Persuasion (Boston, 1966); Patrick 0. Marsh, Persuasive Speaking: The ory Models Practice (New York, 1967); Wayne C. Minnick, The' Art of Persuasion, 2nd ed. (Boston, 1968); Thomas M. Schei- 3el, Persuasive Speaking (Glendiew, 111., 1962). 6 texts,7 one beginning communication text,8 one style text,9 and fifteen works that rhetoricians consistently cited in their accounts of language and meaning.10 Additional books and articles consulted were found to be irrelevant to this 1 study. Second, the examination of these works was limited to their explicit or implicit comments about the nature of 7Glenn R. Capp and Thelma Robuck Capp, Principles of Argumentation and Debate (Englewood Cliffs, 1965); Argumen tation and Debate, rev. ed., ed. James H. McBath (New York, 1963); Gerald R. Miller and Thomas R. Nilsen, Perspectives on Argumentation (Chicago, 1966); Glen E. Mills, Reason in Controversy: An Introduction to General Argumentation (Bos ton, 1964); Russel Windes and Arthur Hastings, Argumentation and Advocacy (New York, 1965). eBerlo, The Process of Communication. 9Jane Blankenship, A Sense of Style: An Introduction to Style for the Public Speaker (Belmont, Calif., 1968). 10Stuart Chase, The Tyranny of Words (New York, 1938); Idem, Power of Words (New York, 1954); John C. Condon, Jr., Semantics and Communication (New York, 1966); Rudolf Flesch, The Art of Plain Talk (New York, 1946); S. I. Hayakawa, Lan guage in Thought and~Action (New York, 1944); Idem, Symbol, Status, and Personality (New York, 1963); Wendell Johnson, People in Quandries: The Semantics of Personal Adjustment (New York, 1946); Alfred Korzybski, Science and Sanity: An Introduction to non-Aristotelian Systems and General Seman tics, 3rd ed. (Lakeville, Conn., 1948); Susanne K. Langer, Philosophy in a New Key: A Study in the Symbolism of Rea son, Rite and Art, 2nd ed. (New York, 1351); Irving J. Lee, Language Habits in Human Affairs: An Introduction to Gen eral Semantics (New York, 1941); C. K. Ogden and I. A. Richards, Meaning of Meaning; A Study of the Influence of Language on Thought and of the Science of Symbolism, 8th ed. (New York, 1946); Charles E. Osgood, Method and Theory in Experimental Psychology (New York, 1953); I. A. Richards, The Philosophy of Rhetoric (New York, 1965); Stephen Ull- mann, The Principles of Semantics, 2nd ed. (New York, 1967); and The Use and Misuse of Language, ed. S. I. Hayakawa (Greenwich, Conn., 1962). 7 language and meaning. This study was a critique of philo sophical presuppositions about language and meaning from the perspective of ordinary language philosophy; it is not a complaint about how speech-communication is taught. Method of Study The method of this study was philosophical and criti cal. By philosophy was meant "the systematic critique of presuppositions."11 Rhetoricians' presuppositions about the nature of language and meaning, expressed in textbooks and journal articles, constituted the object of criticism. Ordinary language philosophers provided the criteria in terms of which the rhetoricians' assumptions were evaluated. It might be argued that pedagogical works, especially beginning textbooks, are an inappropriate place to look for presuppositions about any subject matter. However, the con trary would appear to be the case; every account of language and meaning is necessarily based on assumptions that are. philosophical in nature, whether expressed or implied. Those assumptions, as previously stated, help determine one's entire approach to speech-communication. Furthermore, textbook writers are responsible for the soundness of their assumptions as well as for their public expression and exemplification of principles of speech-communication. The 1xThis widely-accepted definition of philosophy is uti lized, e.g., in Maurice Natanson, "The Limits of Rhetoric," Quarterly Journal of Speech, XLI (April 1955), 127-132. ------------------------------------------------------------------------- ff fact that speech-communication textbooks have almost totally ignored ordinary language philosophy is difficult to explain, unless rhetoricians believe that ordinary language philosophy is unsound or irrelevant to their interests. It seems more likely that they have preferred to follow prece dent rather than thoroughly avail themselves of the best that can be known about the nature of language and meaning. In short, insofar as textbooks rest on philosophical assump tions and are public statements, they are invitations to critical assessment. This study assumes the validity of criteria derived from the ordinary language philosophers1 approach to lan guage and meaning. That assumption seems warranted for several reasons. First, the ordinary language "movement" is highly regarded by British and American philosophers. Anthony Flew recently observed that it "has gained momentum until now it dominates the philosophy faculties of Oxford, Cambridge, and London, is powerfully represented elsewhere in the United Kingdom, and even has outposts overseas . . . ."12 Many historians of philosophy contend that the movement is responsible for a "revolution in philosophy."13 12Anthony G. N. Flew, Introduction, Logic and Language, First Series, ed. A. G. N. Fiew (Oxford, 1968) , p. IT 13See D. F. Pears, "Logical Atomism: Russell and Witt genstein," in The Revolution in Philosophy (London, 1957), p. 41; Richard Rorty, Introduction, The Linguistic Turn, ed. Richard Rorty (Chicago, 1967), p. 1; G. J. Warnock, English Philosophy since 1900 (New York, 1966), p. 104.______________ In addition, the ordinary language approach has resulted in a significant improvement in the understanding of "how we do things with words." For instance, the work of these philos ophers suggests that the view that language is a system of symbols is extremely oversimplified. Ordinary language phi losophers have also demonstrated why "it cannot be the case that to say a word has a certain meaning is to say it refers to something."14 Moreover, they have shown why "we cannot explain the meaning of a word by postulating undefined ide,.n to which it refers."15 Their success led J. L. Austin to suggest that the movement not only constitutes a revolution in philosophy, but also "if anyone wishes to call it the greatest and most salutary in its history, this is not, if you come to think of it, a large claim."16 Both the ordi nary language philosophers' methods of analysis and their conclusions about language and meaning offer significant insights to every language scholar. Review of Literature Since the accounts of language and meaning published ir both speech-communication literature and ordinary language 14William P. Alston, Philosophy of Language (Englewood Cliffs, 1964), p. 16. lsRobert F. Terwilliger, Meaning and Mind: A Study in the Psychology of Language (New York, 1968) , p. 151. Ital ics added. 16J. L. Austin, How to Do Things with Words, ed. J. 0. Urmson (New York, 1965), pp. 3-4. 10 philosophy literature are outlined in Chapters II and III, this review is concerned with two observations. First, only four references to ordinary language phi losophy were found in speech-communication literature. John B. Newman, relying principally on the writings of Charles Morris and Rudolf Carnap, argued in the Quarterly Journal of Speech that the ordinary language approach would never replace the semantic study of language, chiefly because of the futility of trying to talk about ordinary language using ordinary language.17 Beginning with the assumption that one cannot talk about a language except in another language, Newman argued that because ordinary lan guage philosophers discussed everyday language in everyday language, they were methodologically limited. He focused neither on the nature of language nor on meaning. John Searle, a philosophy professor at Berkeley and frequent con tributor to philosophical journals, introduced and partially developed an aspect of J. L. Austin's approach to linguistic meaning in a chapter in Human Communication Theory.18 Searle also did not deal specifically with the nature of language or, strictly speaking, with the nature of meaning. The third article, Alexander Sesonske's contribution to the 17,,The Semantic Analysis of Ordinary Language," Quar terly Journal of Speech, XLIV (December 1963), 410-41^1 18,lHuman Communication Theory and the Philosophy of Language: Some Remarks," Human Communication Theory, ed. Frank E. X. Dance (New York, 1967), pp. llb-liJy. 11 first issue of Philosophy and Rhetoric, is discussed in detail in part three of Chapter II. The final article was reprinted from the Saturday Evening Post in the Central States Speech Journal.19 Written by the philosopher Morton White, this essay was an introduction to the ordinary lan guage philosophers. After characterizing their main thrust as an examination of language use, White noted that the father of the movement, Ludwig Wittgenstein, "said plainly and insistently that the main subject matter of the philoso pher was the behavior of human beings trying to communicate with one another ..." (p. 191). White also asserted that the analysis of ordinary language could change philosophy from "empty speculation about 'being as such1" to a "deeper understanding of man gained from a study of his ways of com municating with his fellow men" (p. 191). The second observation is that the philosophical mate rial used in this study came primarily from writings by the five ordinary language philosophers previously noted and from comments in philosophical publications about those views. The principal works consulted were the following: Alston, William P. Philosophy of Language. Englewood Cliffs: Prentice-Hall, 1964. This textbook includes a discussion of the nature of language, an account of Alston's version of an ordinary 19"New Horizons in Philosophy," Central States Speech Journal, XII (Spring 1961), 188-196. T2 language approach to meaning, and a treatment of linguistic vagueness and metaphor. Alston's analyses of the weaknesses of classical theories of meaning was heavily relied on in Chapter IV of this dissertation. Austin, J. L. Philosophical Papers, ed. J. 0. Urmson and G. J. Warnock. Oxford: Clarendon Press, 1961. Austin's Papers includes all of the works he published before his premature death in 1960. Especially important to this study were his essays, "The Meaning of a Word," and "Performative Utterances." Austin, J. L. How to Do Things with Words, ed. J. L. Urm son. New York: Oxford University Press, 1965. This volume is a reproduction of the William James Lec tures delivered by Austin at Harvard University in 1955. Austin had given similar lectures several times at Oxford and elsewhere, and this account was created from his most recent notes. The lectures trace his concept of "speech acts" from inception through incomplete development to par tial rejection. Ayer, A. J., et al. The Revolution in Philosophy. London: Macmillan and Co., Ltd., 1957. In this work Ayer, Ryle, Strawson, and five other authors present a non-technical description of the philo sophical background of ordinary language philosophy. The book was especially germane to the first part of Chapter 113 of this study. Caton, Charles E., ed. Philosophy and Ordinary Language. Urbana: University of Illinois Press, 1963. 13 Caton's brief but informative introduction and his col lection of essays by Austin, Ryle, Searle, Strawson, and others were very helpful. Especially important were Ryle's essays, "Ordinary Language," and "The Theory of Meaning." Passmore, John. A Hundred Years of Philosophy, rev. ed. New York: Basic Books, 1966. This history of philosophy from John Stuart Mill to the Existentialists and Phenomenologists was invaluable back ground reading for Chapter III. Passmore's account of "Wittgenstein and Ordinary Language Philosophy" was particu larly useful. Ryle, Gilbert. Dilemmas. Cambridge: Cambridge University Press, 1966. The Tarner Lectures, which Ryle gave in 1953, have been reproduced in this volume. The lecture "Informal and Formal Logic" was extremely important, and was described in detail in Chapter III of this study. Strawson, P. F. Introduction to Logical Theory. London: Methuen and Co., Ltd., 1952. The explanation of "the logic of language" in Chapters I and VIII of Strawson's work was relied on in Chapters III and IV of this study for much of the account of the informal logical rules governing speech behavior. Wittgenstein, Ludwig. The Blue and Brown Books. New York: Harper and Row, 1958. The notes from which Wittgenstein lectured at Cambridge in the early 1930's were collected into these two manu scripts. As preliminary studies for his Philosophical ------------------------------------------------------------------------ Investigations, they are necessary reading for every student of Wittgenstein. Wittgenstein, Ludwig. Philosophical Investigations, trans. G. E. M. Anscombe. Oxford: Basil Blackwell, 1953. This work is the basic ordinary language philosophy manuscript. In it Wittgenstein introduced the approach and outlined the subject matter that philosophers of this tradi tion have largely followed to the present day. It is unquestionably the single most important work in British philosophy published in this century. Preview of Dissertation Chapter II describes approaches to language and meaning in current speech-communication literature. In part one the view that language is fundamentally a system of symbols is set forth in detail. The conclusions that words are essen tially names and that Korzybski's theory of abstraction can account for the nature of apparently non-symbolic words are also recounted. In part two, the three theories of meaning adopted by rhetoricians are described. Chapter III briefly characterizes the history of phi losophical interest in language that led up to the ordinary language "movement." In addition, it describes four funda mental conclusions that mark the ordinary language approach to language and meaning: (1) Language is not a calculus. (2) Since language is not mathematically regular, words do not function in any single way; specifically, meaning is not -------------------------------------------------------T5 simply reference, and words are not simply names. (3) Vir tually all generalizations about language are oversimplified and hence misleading. (4) Language can most accurately be viewed as a kind of ordinary human behavior. In addition, the ordinary language methodology is described, an approach that focuses on analysis of the "informal logic of the func tionings of ordinary expressions." Chapter IV is a critique of the rhetoricians' views in light of contentions of the ordinary language philosophers. In part one the assumption that language is a system of symbols and the three representational theories of meaning consistent with that assumption are evaluated and rejected. In part two it is argued that rhetoricians were led to their mistaken views of language and meaning by asking two incor rect first questions: instead of "What is the nature of language?" they should have asked "What is the nature of speech?" And instead of "What is meaning?" they should have asked "How is speech behavior meaningful?" The latter two questions, based on an ordinary language perspective, pro vide the foundation for an alternate view of the nature of speech that emphasizes its ordinary, oral, and rule-governec aspects, and an alternate view of the nature of meaning that: approaches meaning as a process rather than a product. Chapter V summarizes the exposition of rhetoricians' views of language and meaning from Chapter II, the ordinary language philosophers' contentions about language and ------------------------------------------------------------------------------------------------------------------------------------------------ -J-g- meaning from Chapter III, and the critique of the rhetori cians' views from the perspective of the ordinary language philosophers. It concludes with an examination of the sig nificance and implications of the alternate views of speech and meaning and suggestions for future research. CHAPTER II RHETORICIANS AND THE NATURE OF LANGUAGE AND MEANING Few rhetoricians treat language or linguistic meaning in detail either in textbooks or journal articles. Occa sionally the beginning paragraphs of a basic text mention that language is the medium common to all communication, and briefly outline the author's analysis of its nature and function. Discussions of style sometimes include references to language and meaning and suggest how the student could utilize knowledge of them in preparing and delivering an effective speech. Most often, however, rhetoricians' views ef language and meaning are implicit in their accounts and only emerge in passing references and brief explanations. The Nature of Language Rhetoricians agree to the following conclusions about the nature of language: (1) Language is fundamentally a system of symbols, and, because symbols stand for or name something else, words are fundamentally names. (2) Those words that apparently do not stand for anything can either be shown to be symbolic in terms of the theory of abstrac- 17 ------------------------------------------------------------------------ ra tion or can be ignored because they are not primarily used to communicate. Language Is Fundamentally a System of Symbols The contention that language is symbolic is consistent throughout the literature. Rhetoricians argue, for instance, that when we learned to speak, our first babblings were expressive signs, not symbols. . . . Only gradually did we learn that words stand for but are not part of persons, objects, events. After a while . . . the process of human symbolizing had been established and our world was never the same again.1 As a baby learns to talk, his "word becomes a symbol, a representation"; he masters "the simple names of the objects that surround him and . . . their referents . . . ."2 Rhetoricians also support the contention in their treatments of language use. The mature speaker, they main tain, learns that language is systematic verbal symbolism . . . we may say that language has to do with verbal symbolism (words) , that language has to do with system (pattern and order) , and that oral lan guage has to do with sound. When someone speaks he uses symbols, system, and sound to say something . . . .3 The speaker also discovers that "words should accurately xAlma Johnson Sarett, Lew Sarett, and William Trufant Foster, Basic Principles of Speech, 4th ed. (Boston, 1966), p. 179. Cf. Martin P. Andersen, Wesley Lewis, and James Murray, The Speaker and His Audience (New York, 1964), p. S. 2Charles T. Brown and Charles Van Riper, Speech and Mar. (Englewood Cliffs, 1966), p. 15. 3Jane Blankenship, Public Speaking; A Rhetorical Per spective (Englewood Cliffs, 1966), p. 104. 19 mirror the reality they are intended to symbolize. You must achieve a high degree of correspondence between the symbol and the referent . . . Most rhetoricians agree that "a symbol is something that stands for something else."5 They also concur that, since words are symbols, they too "stand for something else." As Dickens puts it, A symbol is anything that represents something else. . . . When two or more people, however, through mutual agreement or common usage attribute the same or similar meanings to given sounds, gestures, writings, or even smoke signals, then language symbols have been created.6 In short, rhetoricians hold that words are symbols which are "conventionally agreed upon to represent certain things [and] to classify things,"7 which "direct attention to spe cific differentiating aspects of the thing they symbolize,"8 and which represent and substitute for "objects, events, 1 > A. Craig Baird and Franklin H. Knower, General Speech, 3rd ed. (New York, 1963), p. 138. Cf. Wallace C. Fothering- ham, Perspectives on Persuasion (Boston, 1966), pp. 54 and 58; James H. McBurney and Ernest J. Wrage, Guide to Good Speech, 2nd ed. (Englewood Cliffs, 1960), p. 12. sBrown and Van Riper, p. 100. 6Milton Dickens, Speech; Dynamic Communication, 2nd ed. (New York, 1963), p. 371. Cf. Raymond G. smitn, Princi ples of Public Speaking (New York, 1958), p. 29. 7Raymond S. Ross, Speech Communication: Fundamentals and Practice, 3rd ed. (Englewood Cliffs, 1965), p. 3b. eAndersen, Lewis, and Murray, p. 188. 20 axperiences, and concepts."9 As Dickens explains, consider the concept symbolized by the word football. You can think it, read it, write it, hear it, or say it. Whichever you do, the word itself is a symbol, and the symbol stands for the same thing, whether it is thought, read, written, heard or said.10 Most rhetoricians also agree that to say language is essentially symbolic is to say that words are fundamentally names; "the first function of a language is to name, to 9A. Craig Baird and F. H. Knower, Essentials of Gen eral Speech, 3rd ed. (New York, 1968), pi T2W. 10Dickens, p. 205. Cf. Baird and Knower, General Speech, p. 138; Blankenship, p. 104; E. C. Buehler and tfil A. Linkugel, Speech: A First Course (New York, 1962), p. 239; Kenneth G. Hance, David C. Ralph, and Milton J. Wiksell, Principles of Speaking (Belmont, Calif., 1969), p. 213; Patrick O. Marsh, Persuasive Speaking: Theory Mod els Practice (New York, 1967) , p. 258; McBurney and Wrage, p. 162; Marie Hochmuth [Nichols], "I. A. Richards and the 'New Rhetoric,1" Quarterly Jourril of Speech, XLIV (February 1958), 3; Henry Nelson Wieman and Otis M. Walter, "Toward an analysis of Ethics for Rhetoric," Quarterly Journal of Speech, XLII (October 1957), 266; Loren Reid, "The Disci pline of Speech," Speech Teacher, XVI (January 1967), 1-10; C. Merton Babcock, "The Importance of Perspective in Commun ication," Central States Speech journal, VI (Fall 1954), 3- 6; Wilbur E. Moore, "I Disagree," in "A Controversial Sympo sium: The Basic Function of a Speech Department Is Develop ing Speech Skills," Central States Speech Journal, X (Autumn 1958) , 50-56; Donald W. Dedman, "Cnticizxng Student Speeches: Philosophy and Principles," Central States Speech Journal, XVIII (November 1967), 276-284; Elwood Murray, "Streamlining the Speech Edifice: I," Western Speech, XXIII (Fall 1959), 197-202; idem, "The Laboratory for Semantic Compatibility at the University of Denver," Western Speech, XXXI (Fall 1967), 274-280; John B. Newman and Milton W. Horowitz, "Speaking and Writing," Today's Speech, XIII (February 1965), 2-4+; Richard M. Rothman, "Language and Thought Re-visited," Today's Speech, XIV (April 1966), 11- 12+; Edwin Benjamin Black, "A Consideration of the Rhetori cal Causes of Breakdown in Discussion," Speech Monographs, XXII (March 1955), 15-19; Paul J. Rosenthal, "The Concept of Ethos and the Structure of Persuasion," Speech Monographs, XXXIII (June 1966), 114-126. 21 point to, or in some way to indicate the concepts we have acquired . . . ."11 In addition, "when we objectively rea son our way to a conclusion, we manipulate words as the lames of objects in order to save ourselves the trouble of manipulating the objects themselves."12 Baker and Eubanks summarize this concept: With the words of language man was able to get a grip on his environment. He learned to name things— to utter sounds that served as symbolic "stand-ins" for the objects and events about him. . . . With the use of spoken symbols that could serve as stand-ins for objects, events, and feelings, man became a rational animal.13 None of the rhetoricians claim to have originated the view that language is fundamentally a system of symbols, or that words are essentially names. The principal sources for their symbolic account of language are Susanne Langer, Alfred Korzybski, and the general semanticists who expanded, simplified, and applied Korzybski's "Non-Aristotelian" system.14 In Philosophy in a New Key, Langer draws on the lxMarsh, p. 258. Cf. Andersen, Lewis, and Murray, p. 86. 12Alan H. Monroe and Douglas Ehninger, Principles and Types of Speech, 6th ed. (Glenview, 1967), p. 21. 13Virgil L. Baker and Ralph T. Eubanks, Speech in Per sonal and Public Affairs (New York, 1965), p. Cf. Sarett, Sarett, and Foster, p. 80; and Raymond G. Smith, "Motivation and Communication-Theory," Central States Speech Journal, XV (May 1964), 97-98. 1‘ *An analysis of all footnotes, suggested additional readings, bibliographical references, and index citations of the textbooks' accounts of language and meaning produced the — 22 accounts of many authors, including Whitehead and Russell, Ogden and Richards, and Wittgenstein, all of whom had tried to invest verbal language with the precision of mathematical symbolism. Their studies of the human use of symbol sys tems, she asserts, have led to a new understanding of man's fundamental nature. It is this: animals can learn by trial and error that certain phenomena are signs of certain others. But "man, unlike all other animals, uses 'signs' not only to indicate things but also to represent them."15 following results: SOURCES CITED IN SPEECH TEXT AUTHORS' ACCOUNTS OF LANGUAGE AND MEANING Rank Author Times Cited 1 S. I. Hayakawa 41 2 Stuart Chase 33 3 Irving Lee 25 4 Susanne Langer 24 5 Wendell Johnson 23 6 Charles Osgood 23 7 C. K. Ogden and I. A. Richards 20 8 David Berio 16 9 Rudolf Flesch 16 10 Alfred Korzybski Others— Edward Sapir, Charles Morris, Stephen Ullmann, Joshua Whatmough, Benjamin Whorf, Anatol Rapoport, Ken 13 neth Burke 27 15Susanne K. Langer, Philosophy in a New Key: A Study in the Symbolism of Reason, Rite and Art, 2nd ed. (New York, 1951), p. 3 7 . ____________ 23 The signs that humans characteristically use do not just announce things, they remind us of them. They serve . . . to let us develop a characteristic atti tude toward objects in absentia/ which is called "thinking of" or "referring towhat is not here. "Signs" used in this capacity are not symptoms of things, but symbols. (p. 37) Langer asserts that when the mind creates symbols it is responding to a characteristically human need, the need to symbolize. This basic need, which certainly is obvious only in man, is the need of symbolization. The symbol-making function is one of man's primary activities, like eating, looking, or moving about. It is the fundamental process of his mind, and goes on all the time. (p. 45) Symbolization, Langer argues, is the essential act of mind, and subsumes what is normally called thought. The brain is always "actively translating experiences into symbols, in fulfillment of a basic need to do so."16 Korzybski also begins his account of language with an explanation of its symbolic nature. He defines a symbol as a sign which stands for something. Any sign is not neces sarily a symbol. If it stands for something, it becomes a symbol for this something. If it does not stand for some thing, then it becomes not a symbol but a meaningless sign .... Before a noise [etc.] may become a symbol, something must exist for the symbol to symbolize. 7 The notion that words are fundamentally names can also 16Langer, p. 46. Cf. Walter Coutu, "An Operational Definition of Meaning," Quarterly Journal of Speech, XLVIII (February 1962), 59-64. 17Alfred Korzybski, Science and Sanity; An Introduc tion to non-Aristotelian Systems and General Semantics, 3rd ed. (Lakeville, Conn., 1948), pp. 78-79. 24 be found in Korzybski's work: All languages are composed of two kinds of words: (1) of names for all the somethings on the unspeakable level, be they external objects [etc.] or internal feelings, which admittedly are not words, and (2) of relational terms, which express the actual, or desired, or any other rela- tions between the unspeakable entities of the objective level. (p. 250) According to this view, statements and propositions, whether mathematical -or otherwise, are made up of names connected by relations. The general semanticists, who are primarily concerned with developing and applying Korzybski's insights, maintain his emphases.18 As Irving Lee summarizes, Words may be thought of as signs which name that for which they are signs: table is the name of an object, red of a quality, run of an activity, over of a relation .... Thus, words may be consideredas pointers, indicators, forms of representation, which are intendedto correspond to anything whatsoever that may exist, that may be expe rienced, or that anyone may want to talk about. Or put another way, words may be used for the almost endless nam ing of the inexhaustible electronic events, objects, per- sons, situations, relations, etc., observed outside-our- skins, along with the sensations, feelings, beliefs, opin ions, values, tensions, affective states, etc., expe rienced inside-our-skins.19 Expressed in syllogistic form, the basic ideas of rhet oricians' views of language would read: 10See, e.g., S. I. Hayakawa, Language in Thought and Action (New York, 1944), p. 25. Cf. John C. Condon, Jr., Semantics and Communication (New York, 1966), pp. 6 and 30, "naming is the basis of language." 19Irving J. Lee, Language Habits in Human Affairs: An Introduction to General Semantics (New York, 1941) , pp. 15- T5T Cf. Michael M. Osborn, "The Evaluation of the Theory oJ: Metaphor in Rhetoric," Western Speech, XXXI (Spring 1967), 122; and Fotheringham, pp. 58-59. ------------------------------------------------------------75 Symbols stand for something else. Words are symbols. Therefore, words stand for something else. When they develop this point of view, many rhetoricians agree that the "standing-for" relationship referred to in the syllogism may be thought of as a naming relationship. Thus the corresponding syllogism would read: Symbols are names of something else. Words are symbols. Therefore, words are names of something else. Apparently Non-symbolic flords Are Also Symbolic The account of language outlined in these two syllo gisms cannot adequately describe all the speaking that peo ple normally do; that fact is clear both to rhetoricians and to the theorists whose accounts of language they cite. The reason that the symbolic account is inadequate is that at least three kinds of commonly-used words apparently do not symbolize or name any sort of referent. For example, although it may be clear to many that "table" is the name of an object, and perhaps even that "red" is the name of a guality, "run" of an activity, and "over" of a relation, what ."something else" is there for abstract terms to name— 3uch as "democracy," "truth," or "semantics"? Exclamatory atterances such as "Ouch!" and "Damn!" also present prob lems. In addition, syncategorematic words, i.e., those that philosophers claim have no meaning unless paired with a word that refers to a category of reality, are also prob- 26 lematic.20 What "things" can words like "the," "for," and "of" be said to name? Rhetoricians and general semanticists argue that most of these apparently non-symbolic terms can be shown to be symbolic with the help of the theory of abstraction. According to the theory, although some words may not directly or obviously name a referent, such words are abstractions or abstractions of abstractions, or so on, and they eventually do refer indirectly to some objective real ity. Raymond Ross outlines the process of abstraction in these words: Abstracting is a process of thinking in which we selec tively leave out details about concrete or real things • • • • The process of abstraction may be classified into levels. For example, Bambi may be considered a first- order verbal abstraction, that is, Bambi is a very specif ic form and kind of deer. A deer in turn is a special kind of animal and so on. As we move from lower- to higher-order abstractions, we tend to consider fewer and fewer details of the specific or original object. Another way of looking at abstracting is to consider first-hand 20William P. Alston explains the meaning of "syncate- gorematic" this way: "This term was introduced by medieval logicians to apply to words like conjunctions, which were regarded as not standing for anything and so as not having meaning 'in isolation.1 These were the linguistic units that were left over after one had gone through everything that could be assigned to Aristotle's ten 'Categories,' a classification of terms made by Aristotle. Thus, the rem nants were terms that were used only with (syn— categore- matic) the categories." (Philosophy of Language [Englewood Cliffs, 1964] , p. 14.) 27 observations as facts, but facts which may never be described in an absolutely complete way.2 The primary source for rhetoricians' treatments of the process of abstraction is Korzybski. He.begins his analysis ay postulating that whatever we say is based on something more fundamental than language: As we have already seen, when we make any proposition whatsoever we involve creeds, or metaphysics, which are embodied silently as structural assumptions and in our undefined terms. The use of terms not definable in simpler terms at a given date is inherent and seemingly unavoidable.2 2 According to Korzybski, "there is no escape from the fact that we must start with undefined terms which express silent, structural creeds or metaphysics" (p. 373). How ever, he argues, by following the "structural metaphysics" of modern science, we can avoid all the problems created by "undefined terms." Whereas the old "Aristotelian" system of language was distorted by the fallacy of identity, under tfhich the word "dog" became identified with the thing dog, and with "elementalism," or the dichotomizing of "emotion" and "intellect" (pp. 31ff.), the new "non-Aristotelian," 21R o s s , p. 41. C f . Giles Wilkeson Gray and Waldo W. Braden, Public Speaking: Principles and Practice, 2nd ed. (New York, 1963), pp. 460-461? Smith, Principles, ' pp. 136- 137; Marsh, p. 258; Thomas M. Scheidel, Persuasive Speaking (Glenview, 111., 1962), p. 35; Baird and Knower, General Speech, p. 156; Sarett, Sarett, and Foster, p. 86; William Norwood Brigance, Speech: Its Techniques and Disciplines in a Free Society, 2nd ed. (New York, 1961), pp. 21-24? Dick- ens, p. 216; and Andersen, Lewis, and Murray, p. 100. 22Korzybski, pp. 371-372. 28 "scientific" language would not be limited by the "is of identity," and would be "non-el" in structure. For Korzybski, the new language accurately reflects reality as science helps us understand it. The simplest object, e.g., a pencil lying on a table, looks different to various people viewing it from different positions. That phenomenon occurs because no person is able to perceive all the stimuli inherent in the pencil, and thus each abstracts from the total configuration of stimuli those perceptions determined by his experience, ability, context, etc. This process, based on the specifity of response to stimuli, is called "non-allness" (p. 375). Since no one, even with the lelp of instruments, can perceive all of anything, "we abstract whatever we and the instruments can; then we sum marize; and, finally, we generalize, by which we mean the process of abstracting carried further" (p. 377). In this way we arrive at abstractions of different orders. Korzyb- ski's "structural differential" (pp. 386-411) is a pictorial ievice to illustrate the process of abstraction. The Important point is that for Korzybski, the ultimate basis of perception, determined by his "structural metaphysic1 1 of science, is the "unspeakable entity"— the partially-perceiv- able, ineffable substance "on the objective level" that is at the basis of every word. S. I. Hayakawa's explanation of the process of abstrac- bion is widely known, primarily because of his use of 29 'Bessie the cow," and, a term Korzybski used, "the abstrac tion ladder."23 Hayakawa contends that all objects consist of "atoms, electrons, etc., according to present-day scien tific inference. "2 * * Since humans cannot perceive at that Lowest level of reality, their nervous systems abstract from "the totality that constitutes the process-object" to arrive at the second level of abstraction, the object of our sense- perception. The word that we assign to that object, i.e., Its name, is at the third level of abstraction. From that point, succeeding levels of abstraction are all verbal: from the ineffable "un-speakable entity" of electrons making up a cow, we abstract to the cow we perceive. We further abstract to the word "cow," and subsequently to words like "livestock," "farm assets," "assets," and, perhaps, "wealth." At this point Bessie's abstraction ladder has been scaled; "wealth" is obviously a highly abstract term. The basic assumption that is reflected in this part of Haya- kawa's analysis of language is that, as Korzybski asserted, some sort of objects lie at the base of all our words. At the atomic level the objects cannot be perceived; at the object level they are ineffable; at the verbal level they 23See Hayakawa, L a n g u a g e in Thought and Action, espe cially pp. 167-185. . 2''Hayakawa, p. 169. ------------------------------------------------------------------------3tr are named.25 In this sense, then, general semanticists and their followers in speech-communication agree that abstract words are also symbolic. A term like "democracy" or "semantics" is not itself a name of any object/ but is a higher-order abstraction of a term that is in turn an abstraction of an abstraction . . . and ultimately, of an object. If the lan- guage-user forces himself to back down the abstraction ladder, Korzybski, Hayakawa, and virtually all of the rhet oricians considered here agree, he will eventually reach the objective level, where words are clearly names of "refer ents," "designata," or "objects-in-reality." The other classes of apparently non-symbolic words are handled in one of two ways: the authors either ignore them completely or offer some brief explanation consistent with the symbol metaphor and explain that such words are not their primary concern. For instance, Baker and Eubanks briefly discuss speech that is not designed for public hear ing, which they call "covert expression." It may take many forms: exclamations such as "Ouchl" when we hurt ourselves, a sigh, or a yawn .... We are not concerned with speech as covert expression in this course, since such speech is not necessarily a communicative act 25Cf. Stuart Chase, The Tyranny of Words (New York, 1938) , pp. 8 and 101.___________________________________ 31 consciously planned for others to hear and react or respond to. Syncategorematic words are treated in similar ways. As Wendell Johnson asserts, there is a sense, of course, in which such words [as "of," and "now"] can be given factual reference, but we under stand them mainly as words which we use in order to con struct sentences or phrases. They are fasteners and hinges in the language chain, as it were. We use them primarily to express relations between other words, and so ultimately between facts.2 7 Johnson also contends that words like "now" and "yet" are "time words" that express relations between events, and "stand for whatever we experience as the feelings of longer or shorter durations or as 'now' and 'then'" (p. 130). According to Blankenship, such words are "signals." "A signal is something in one word that tells us something about another one."28 For instance, "the," "les," "die," and "los" are all articles that signal certain predictable relationships with the nouns they accompany. In short, virtually all rhetoricians considered here recognize that the symbol metaphor creates problems with the explanation of everyday language use. However, they main tain their basic assumption that words are symbols or names 26Baker and Eubanks, p. 72. Cf. James C. McCroskey, Ar. Introduction to Rhetorical Communication: The Theory and Practice of Public Speaking (Englewood Cliffs, 1368) , p. 21. 27Wendell Johnson, People in Quandries: The Semantics of Personal Adjustment (New York, 1946), p. 129. 28Blankenship, p. 109. 32 of things. Accordingly, they conclude that abstract words indirectly name the objects at the base of the abstraction ladder, exclamations represent or name internal emotional states, and syncategorematic words name relations between words or events. The assumption that language is fundamen tally a system of symbols— the symbol metaphor— also under lies rhetoricians' views of the nature of meaning, which is the subject of part two. The Nature of Meaning Just as few rhetoricians discuss the nature of lan guage, few of them thoroughly examine the subject of mean ing. Although meaning is mentioned in almost every text,29 and in many journal articles,30 it is seldom analyzed in any 29E.g., Baird and Knower, General Speech, Chapter 10, "Language and Semantics"; Sarett, Sarett, and Foster, Chap ter 4, "Effective Speech Communicates the Meanings Intended by the Speaker"; Ross, Chapter 3, "Language: Meaning and Use." Brief treatments can be found in Eubanks and Baker, Chapter 16, "Language in Speech"; McBurney and Wrage, Chap ter 12, "Language and Style"; and Smith, Chapter 7, "Lan guage for Gaining Response." 30E.g., Glenn R. Capp, "General Semantics for the Debater," Southern Speech Journal, XIX (May 1954), 294-303; Ralph T. Eubanks, "Nihilism and the Problem of a Worthy Rhetoric," Southern Speech Journal, XXXIII (Spring 1968), 187-199; Walter Goldschmidt, "Language and Culture: A Reply," Quarterly Journal of Speech, XLI (October 1953), 279-283; Margaret Gorman, "A Critique of General Semantics," Western Speech, XXXI (Winter 1967), 44-50; Charles V. Har- tung, "The Scope of Linguistic Study," Quarterly Journal of Speech, L (February 1964), 1-12; John B. Newman, "Sympto matic Signals in Speaking and Writing," Western Speech, XXXIII (Winter 1960), 40-44; Carl H. Weaver and Gary L. Weaver, "Information Theory and the Measurement of Meaning," Speech Monographs, XXXII (November 1965), 435-447. 33 detail. When the subject of meaning is explored, the dis cussion is generally consistent with the view of language outlined in the previous section. That is, rhetoricians agree that language is essentially a system of symbols, and they also agree with the correlate of that position, that meaning involves whatever linguistic symbols represent. Thus theories of meaning found in current speech-communica- tion literature tend to be representational theories, i.e., they postulate that meaning involves a symbol and something represented or symbolized. Rhetoricians develop three approaches to meaning from their common assumption that language is symbolic: (1) lin guistic meaning is in the object to which the word-symbol refers, or in the relationship between the symbol and the object; (2) the meaning of a language symbol is the idea the symbol represents; (3) although language symbols do have referents, meaning is not to be found in these referents, but in the behavioral responses that individuals make to the referents. These three theories of meaning will be called respectively (1) referential, (2) ideational, and (3) behav ioral. The Referential Theory The simpler version of the referential theory of mean ing posits that the meaning of a word is the object it represents. As typically expressed in a beginning speech- . 34 * V communication text, word meanings are in their referents. The unique fact about language symbols is that they have no meaning in themselves alone: their real meaning is in the thing to which they refer or the event for which they stand.3 In other words, language has meaning for you and those around you solely on the basis of the associations which have been built up between the sound patterns and the things for which they have come to stand. Meaning for you is thus based entirely on your own individual experiences in associating symbol, that is word or phrase, with object.32 A referential theorist determines a word's meaning in roughly the following way: in a given situation, the name "Robert" might not have any meaning for someone, because there are so many people called Robert. However, if the referent or person to whom the name refers is indicated, the meaning is then clear: ". . .if 'Robert E. Lee' or 'Robert Burns' were specified, who Robert is— that is, what the wore. 'Robert' means— becomes more obvious."33 The more complex version of the referential theory of meaning holds that "meaning resides in the relationship between a word and the object (event or value) which it 3lAndersen, Lewis, and Murray, pp. 89-90. 32Gray and Braden, p. 456. Cf. David Rynin, "Seman tics," Western Speech, XX (Winter 1956), 37; Condon, pp. 207-208 and p. 227; Fotheringham, p. 59; Smith, Princi ples, p. 129; Ross, p. 47; and Baird and Knower, General Speech, p. 138. 33Blankenship, p. 108. 331 represents or re-presents."34 Baird and Knower indicate their agreement with both the symbol metaphor and this ver sion of the referential theory when they write, Words are a class of symbols by which objects, experi ences, ideas, feelings, and'emotions are represented. Thus words are to be understood as representing and substituting for the objects, events, experiences, and conceptions that give rise to the initial meaningful association with the stimulus. These sources are the referents. We apparently cannot philosophically discern these referents, but at least we assume their reality as the original source of stimulation through our common sense and experience. The referent leads to the thought or reference and in turn to the word itself, the symbol. Thus we note the word-thought-referent relationship.35 Rhetoricians often develop their versions of the refer ential theory by explaining denotation and connotation. For referential theorists, the denotational meaning of a word is the object or referent which it represents, whereas the sym bol's connotation is affected by the background and emo tional experiences of its user: The denotative meaning of a word is its exact literal meaning, its objective reality, the physical, chemical, or structural characteristics of whatever the word stands for, its functional uses, or its relationships. . . . Denotative meanings are objective, logical, impersonal anc relatively uniform. They are the meanings which are "out side a person's head." This literal meaning of a word is sometimes called its "extensional" meaning .... In addition to what words denote in the world of objective reality, they also have connotative meanings, 3^Blankenship, p. 105. Cf. Jane Blankenship, A Sense of Style: An Introduction to Style for the Public Speaker (Belmont, Calif., 1968), p. 19. 35Baird and Knower, Essentials, p. 126. Cf. Kenneth Frandsen, "Semantic Compatibility in an Interpersonal Com munication Laboratory," Western Speech, XXXI CSttxnmer 1967), 211. 36 often referred to as "intensional" meanings. These are highly personalized; they vary from person to person • • • • Connotative meanings are also described as "emotive, subjec tive, personal, and intensional."37 Some rhetoricians maintain that a word's denotative meaning is "the relationship between the word and the object it stands for," whereas "connotative meaning resides in the relationship between the object, the words, and the speaker/ listener."38 For instance, The word "pencil" may cause one person to visualize a red pencil; another a green one; a third a short, much used pencil; and a fourth a long, unsharpened one. But all would generally agree impersonally that "pencil" equals "that which writes."39 In other words, the denotational meaning of "pencil" is "that which writes." However, many, if not most, words elicit personal emotive responses beyond the objective meanings of the words. . . . Thus, although denotative meaning resides in the relationship « between the object and the word which represents it, con notative meaning resides in the relationship between the object, the word, and the speaker/listener. 0 Smith's summary of denotation and connotation illus- 36Andersen, Lewis, and Murray, pp. 94-96. 37Buehler and Linkugel, p. 244. Cf. John F. Wilson anc[ Carroll C. Arnold, Public Speaking as a Liberal Art (Bos ton, 1964), p. 300. "Blankenship, Public Speaking, p. 106. 39Blankenship, Style, p. 21. 40Blankenship, Style, p. 21. Cf. David K. Berio, The Process of Communica£IonT An introduction to Theory an? Practice (New York, i960), p. 192 and p. 269. 37 trates the relationship between these concepts and the referential theory of meaning: When a speaker utters a word, he expresses two types of meaning: he expresses the symbol of the object as well as his emotional attitude toward the object. Likewise he stirs up two types of meanihg in the mind of his listener: he stirs up the image of the object and, in addition, whatever emotional associations the listener has made either with the object or the word. The listener, then, may grasp three meanings. First, whatever object the sym bol means to him; second, whatever emotional associations he, himself, may have with the object; and third, the speaker's attitude or emotional feeling toward the ob ject.**1 . In other words, both denotative and connotative meaning depend, in the last analysis, on the fact that words refer to objects. Thus an account of denotation or extension and connotation or intension is generally part of the referen tial theory of meaning..**2 The primary sources for referential treatments of mean ing are Alfred Korzybski and his general semantics dis ciples, Hayakawa, Chase, Lee, and Johnson.1+3 Korzybski out- ^Smith, Principles, p. 204. **2See Baird and Knower, General Speech, pp. 153-154; Brigance, p. 209; Robert T. Oliver and Rupert L. Cortright, Effective Speech, 4th ed. (New York, 1961), p. 312; Sarett, Sarett, and Foster, pp. 190-191; Andrew Thomas Weaver and Ordean Gerhard Ness, The Fundamentals and Forms of Speech, rev. ed. (New York, 19(>3) , p. 185; Goldschmidt, p. 282; and Raymond G. Smith, "Development of a Semantic Differential for Use with Speech Related Concepts," Speech Monographs, XXVI (November 1959), 264. **3One should not confuse a semantic analysis of meaninc; with the referential view of the general semanticists. Although semanticists and general semanticists both accept a referential theory of meaning, they do not approach the study of language in exactly the same way. Whereas semanti-- 38 lines his referential theory in the following way: The explanation is quite simple. We start with the nega tive K premise that words are not the un-speakable objec tive level, such as the actual objects outside of our skin and our personal feelings inside our skin. It follows that the only link between the objective and the verbal world is exclusively structural, necessitating the conclu sion that the only content of all "knowledge” is struc tural. Now structure can be considered as a complex of relations, and ultimately as multi-dimensional order. From this point of view, all language can be considered as names either for un-speakable entities on the objective level, be it things or feelings, or as names for relations. **k Korzybski apparently means that since words are not "things" — either objects or emotions— the only link between things and the verbal world is structural. The structure that con nects things and the verbal world is composed of various kinds and combinations of relations. Thus language either names the things of the objective world or names the rela tions that bind things with the verbal world. He uses the primarily concerned with social and personal problems caused by faulty ("unsane" was Korzybski's term) language behavior. General semanticists contend that a clearer understanding of the relationships between words and what they symbolize is therapeutic. Korzybski writes that, while the disciplines called "semantics," "signifies," and "semasiology," are mainly interested in words per se, he is interested in evaluation or a general theory of values. "General seman tics is not any 'philosophy,' or 'psychology,' or 'logic,' in the ordinary sense," Korzybski argues. "It is a new extensional discipline which explains and trains us how to use our nervous systems most efficiently." (Science and Sanity, p. xi.) Because of their emphasis on evaluation and therapy, general semanticists are interested.less in general theories of meaning than in the explanation of insights and techniques to clarify linguistic meaning in specific circum stances. Korzybski, p. 20. 39 axample, "Smith kicks Brown," in which he points out that nost of the elements of the reality that that statement sym- aolize are "objective," including Smith, his leg, "some part :>f the anatomy of Brown," and Brown. However, "kicks" does not name any object, but instead names a relation (pp. 152- 153) . Korzybski's explicit identification of words as names is fundamental to his theory of meaning, and crucial to his entire "non-Aristotelian" system. He argues consistently that his views of structure and order are great improvements over old approaches. He maintains that the "usual proced ure" for analyzing language is structured from words to the world: "first we have our structurally 'preconceived' doc trines and languages? next, we observe the structure of the world; and then we try to force the observed facts into the linguistic structural patterns" (p. 219). However, his approach to language corrects the old errors. . . . in the new way, we start with silent observations, and search empirically for structure? next, we invent verbal structures similar to them; and, finally, we see what can be said about the situation, and so test the lan guage. Experience shows that the old habits of labels first, objects next, instead of the structurally natural order of objects first, labels next, is semantically per nicious and harmful. . . . The semantic structural rever sal of the unnatural reversed order is crucial for sanity, (p. 219) Korzybski develops his referential approach in his dis cussion of "extension" and "intension." As previously noted, he argues throughout Science and Sanity that humans 40 can never perceive all of any "object" (p. 93). Thus he views our perceptions as abstractions, our words as abstrac tions of abstractions, and so on. However, he also main tains strongly that one should keep his language as close to the "objective, un-speakable level" as is humanly possible. In order to do that, one has to adopt an extensional rather than an intensional orientation. Korzybski's preference for extension is related to his concept of order. He asserts that the "natural survival order" of mankind "starts with absolute individuals," moves to lower abstractions first, and then to higher abstractions. Accordingly, it is easy to show that the extensional attitude is the only one which is in accordance with the survival order and nervous structure, and that the intensional attitude is the reversal of natural order, and, therefore, must involve non-survival or pathological s.r. [semantic response]. (p. 173) The advantage of verbal extension is that it "recognizes the uniqueness, with corresponding one-value, of the individual by giving each individual a unique name, and so makes con fusion impossible" (p. 179). One could apply the principle of "giving each individ ual a unique name" in the following way: The [Aristotelian] "apple" was a name for a verbal inten sional definition; in a non-A system we manufacture indef initely many names for the indefinitely many objective and different "apples" by subscripts, "apple/1, "apple2", "apple/', [etc.], supplementing the subscript with the date; thus, in "apple-j^ Feb.23,1933" * • • • tP- 465) In addition, an extensional orientation necessitates the use* of "etc." in a great many statements to indicate that words 41 can never represent objects completely. Thus the exten sional attitude is one method used to avoid the damaging effects of both high-order abstractions and intensional def initions. If words are consistently viewed as names for unique, 1 1 one-valued" individuals, the vagaries of abstract terms and emotionally-colored definitions can be replaced by the concrete specificity of language with a one-to-one cor relation with the physical world. Korzybski's theory of abstraction is also important to his referential account of meaning in another sense. He maintains that since both words and objects are abstractions of different levels, and since what one person abstracts from a given set of stimuli is private, i.e., different fron what any other person would abstract from the same set of stimuli, "an individual, A, cannot know what B abstracts, unless B tells him, and so the 'meaning' of a word must be given by a definition" (p. 20). Consequently, a dictionary could provide all necessary meanings were it not for the fact that dictionaries define words in terms of other words, so that, if we enquire about the "meaning" of a word, we find that it depends on the "meaning" of other words used in defin ing it, and that the eventual new relations posited between them ultimately depend on the [multiordinal] mean-• ings of the undefined terms, which, at a given period, cannot be elucidated any further. (p. 21) Korzybski's concept of multiordinality, like his con cepts of extension and intension, is also related to his theory of abstraction. Although he asserts that the only way to find meanings is through definition, he argues that if one tries to express a word's meaning he will ultimately fail for one of two reasons. (1) He will reach a point where he can only "define in circles," e.g., defining "space" by "length" and "length" by "space," and will soon realize that he has "reached the bottom and the foundation of all non-elementalistic meanings— the meanings of unde fined terms, which we 'know' somehow, but cannot tell" (p. 74). Even though the "meaning" in this case is inef fable, it is nonetheless accurate.^ (2) He will discover that the term he is trying to define is multiordinal, i.e., it has "different meanings in general, depending on the order of abstraction" (p. 74). He will fail in the second place because "without the level of abstraction being speci fied, a m.o. term is only ambiguous; its use involves shift ing meanings . . ." (p. 74). Korzybski's three fundamental assertions about the nature of meaning are: (1) Although words clearly refer to things, the Aristotelian law of identity does not apply to ^Korzybski writes, "On the lowest level of our analy sis, when we explore the objective level (the unspeakable feelings in this case), we must try to define every 'mean ing' as a conscious feeling of actual or assumed, or wished etc. relations which pertains to first order objective entitj.es, psychological included, and which can be evaluated by personal, varied, and racial— again un-speakable first order— psychophysiological effects." (p. 23) ------------------------------------------------------------------------ o language; the word is not the thing. (2) Every language user must be aware that he necessarily and naturally deals continually in abstractions; consequently/ (a) everyone must adopt an extensional orientation/ to maintain close contact between his language and "the objective level/" or (b) at the very least, everyone in each communication situation must be aware of the level of abstraction at which each per son is talking. (3) Many of our most important and most often-used terms, including "yes," "no," "true," "false," "relation," "number," "name," "definition," "abstraction," "structure," "to know," "to speak," "to love," "meaning," etc., are multiordinal terms. They are "ambiguous or infin- ityvvalued, in general, and . . . each has a definite mean ing, or one value, only and exclusively in a given context, when the order of abstraction can be definitely indicated" (p. 433). Korzybski's referential approach to meaning is clari fied, developed, and occasionally altered by his influential followers, S. I. Hayakawa, Stuart Chase, Wendell Johnson, and Irving Lee. Like Korzybski, Hayakawa stresses the notion that "the symbol is not the thing symbolized; the word is not the thing; the map is not the territory it stands for."116 He emphasizes the role verbal context plays in determining meaning (p. 57) and notes that, because of the influence of changing verbal contexts, "no word ever has ^Hayakawa, p. 31. ------------------------------------------------------------in exactly the same meaning twice" (p. 60). Whereas Korzybski virtually ignores the kind of lan guage Ogden and Richards call "emotive," Hayakawa treats "the language of affective communication" in some detail. For instance, his discussion of connotation distinguishes between "informative connotation," or a word's "socially agreed upon 'impersonal' meanings" (p. 83), and "affective connotations," or "the aura of personal feelings [a word] arouses, as, for example, 'pig': Ughl Dirty, evil-smelling creatures, wallowing in filthy sties,1 and so on" (p. 84). In addition, he devotes a chapter to the view that while science is the most exact kind of reporting, literature is the most exact expression of feelings.**7 He discusses sev eral literary and rhetorical devices, including metaphor, simile, allusion, irony, pathos, humor, and "the affective use of facts," and notes that literature is not written to serve as an accurate map of actual territories, but to express feelings. He suggests that the greatest value of literature lies in its ability to bring "civilizing influ ences to bear upon our savage wills" (p. 136). Stuart Chase, another prolific and influential inter preter of Korzybski, makes few changes in the original ideas ^Hayakawa, Chapter 8, "The Language of Affective Com munication." 45 Df general semantics.1 *8 In People in Quandaries Wendell Johnson expresses his referential view of meaning at the beginning of his chapter sailed "The Language of Science and Sanity." There is a cardinal principle in terms of which language is used scientifically: It must be used meaningfully. The statements made must refer directly or indirectly (by means of interrelated definitions) to something in the realm of experience. It is not enough that they refer to something for the speaker and they also refer to something for the listener. What is required is that they refer to approximately the same thing for both the speaker and the listener. . . . The degree to which communication occurs depends precisely upon the degree to which the words represent the same thing for the listener that they do for the speaker. (pp. 50-51) Johnson agrees with Korzybski regarding multiordinal terms, i.e., he notes that they have different meanings at differ ent levels of abstraction. He also stresses that "the word Ls not the thing," and that extensionalization, or the process of insuring adequate verbal evaluation by contin ually testing one's assumptions against non-verbal experi ence, is one of the most important skills of the well- adjusted language-user. Because of his interest in personal adjustment, Johnson devotes the second half of his book to "practical devices and techniques" for the solution of a wide variety of personal problems.^ 9 Irving J. Lee's Language Habits in Human Affairs: An Introduction to General Semantics is also Korzybskian. 1 *8Chase, Tyranny, pp. 8 and 84; Power, pp. 11-21, 16, L22ff., 139, and 146-147. lf9See, e.g., p. 234, "Semantic Relaxation." Since, according to Lee, the goal of general semantics is to establish "adequate language-fact relationships," and the nature of language .is that "words may be thought of as signs which name that for which they are signs," his approach to meaning is that words, when correctly used, refer to or name "non-verbal facts" (pp. 15-20). As he notes, ". . . 'the meaning' of a word in use is roughly synonymous with what a word is used to represent on any particular occasion" (p. 36). In sum, rhetoricians who adopt the referential theory of meaning borrow heavily from accounts by Korzybski and his followers, who argue (1) that the meaning of words can be discovered by realizing that they name the objects that all words, including abstractions, ultimately refer to; (2) that the two kinds of meaning are extensional or denotative and intensional or connotative meaning; (3) that denotative meaning is clearer and enables the language-user to avoid more "unsanity" than connotative meaning; (4) that "ambigu ous or infinity-valued" meanings of multiordinal terms can be specified only when their level of abstraction is pre cisely determined; and finally (5) that the fact that all words have referents makes it difficult to remember the primary lesson of general semantics; namely, "the word is not the thing." The Ideational Theory The referential theory of meaning holds that words_____ 3T refer to objects, and their meaning is either in the objects to which they refer or involved with the relationships between objects and words. The ideational theory of mean ing, on the other hand, posits that words represent ideas, and that the meaning of a word is the idea or conception that the word symbolizes or "calls up." In speech-communication literature the theory is often expressed as part of a discussion of style. For example, Wilson and Arnold note, The ways in which a speaker symbolizes thought reveal his capacities to discriminate among meanings, to conceive ideas clearly, to represent them precisely. . . . Words, figures of speech, and images alone and in combination reflect the minds of men so well that the study of spoken style is eminently humane. Speeches consist of ideas converted into words .... Whenever we say anything orally acoustic symbols stimulate the audience, and stand for our ideas .... Words are symbols for meanings since they stand for ideas.5 0 The ideational theory sometimes also provides the foundation for complex discussions of meaning. In his article "Motivation and Communication Theory," Raymond Smith postulates that a human "stores incoming signals in only two forms— as concepts, and as need-want bonds," both of which "may be viewed as units of meaning." That is, "sensory sig nals indicating the properties of objects are received by 50Wilson and Arnold, pp. 274 and 299. Cf. Blankenship, Style, p. 3; Blankenship, Public Speaking, pp. 3 and 6. 48 the organism and are taken to mean such gualities."51 Smith asserts that humans perceive objects in sets of related "informational elements" that coalesce into need-want bonds or concepts of the objects. Words are the names or "tag elements" of the concepts or need-want bonds. "I am saying that every symbol in every language represents an underlying set of elements which constitutes the concept [i.e., idea] of the person, thing, event, or relationship" (pp. 97-98). In some cases the ideational theory is only partly accepted. For example, Wallace Fotheringham divides lin guistic signs into signals and symbols, and explains the meaning of signals with the help of the referential the ory.52 However, symbolic meaning, he asserts, involves a relationship between the symbol, a conception or idea, and the objective referent, "all associated in such a way that the conception it arouses fits the reality it stands for" (p. 59). > Some authors call the "ideas" or "concepts" that con stitute meanings by other names. Citing Hubert Alexander's Meaning in Language,5 3 Thomas Scheidel begins with the assumption that man imposes order on nature by constructing categories with the assistance of language. Words are used 51Central States Speech Journal, XV (May 1964), 97. 52See Fotheringham, p. 59. 53(New York, 1966). 49 to label or name the categories we form. One of the primary functions of language is to assign labels or names to these groupings or sets .... The important common elements in the set are combined and nar rowed, and it is this narrowed "meaning" of the category to which the label applies. . . . The labels applied to our categories provide a code with which we can manipulate symbolically the groupings into which we have analyzed our universe.5 * * In other words, for Scheidel, a word is essentially a name for the mental category into which a human organizes percep tions, and the meaning of the word is the mental category that the word symbolizes.55 The following statement is a typical summary of the ideational theory: Language is more than a medium of communication, more thar. a sequence of orderly events. It is a symbolic experi ence. And symbolic experience is synonymous with meaning. And to a large extent, though perhaps not completely, meaning is synonymous with idea or thought. If these statements are true, we can say that a speechmaker employs; language to transmit ideas to others. Indeed, speeches are built on ideas. Language and gesture are simply the vehicles which carry the ideas.5 6 The primary source for rhetoricians' accounts of the ideational theory is Susanne Langer. As previously noted, Langer separates signs and symbols; the former only indicate 5ItScheidel, pp. 34-35. 55Cf. Baird and Knower, General Speech, p. 153; Brigance, p. 299; Dickens, pp. 16-17; Monroe and Ehninger, p. 17; Marsh, p. 257; and John B. Newman, "Symptomatic Sig nals," p. 40. 56Donald C. Bryant and Karl R. Wallace, Fundamentals of Public Speaking, 3rd ed. (New York, 1960), p. 26. Cf. idem, Fundamentals of Public Speaking, 4th ed. (New York, 1969), p. 23. 50 objects, while the latter can represent them. Her theory may be viewed as ideational chiefly because of her approach to the meaning of symbols. She asserts that symbols do not represent objects, but are vehicles for the conception of objects .... In talking about things we have conceptions of them, not the things themselves? and it is the conceptions, not the things, that symbols directly "mean."37 In other words, while signs announce or point to objects, symbols lead one to conceive of— to have conceptions or ideas of— their objects. For Langer there are thus four necessary aspects of the meaning relationship: a subject, a symbol, a conception or idea, and an object. Her explanation of the difference between "concept" and "conception" clarifies her ideational approach. She explains that each individual "idea" of an object shares certain characteristics with other "ideas" of that object, and "that which all adequate conceptions of an object must have in common, is the concept of the object" (p. 70). Thus a concept of something is made up of all accurate, individ ualized ideas of it. Understanding is possible, she main tains, to the extent that people share similar concepts of c thing: Probably no two people see anything just alike. Their sense organs differ, their attention and imagery and feel ings differ so that they cannot be supposed to have S7Langer, p. 61. The relationships between Langer's and De Saussure's analyses are outlined in Hartung's, "The Scope of Linguistic Study." ------------------------------------------------------- 5T identical impressions. But if their respective concep tions of a thing, (or event, or person, etc.) embody the same concept, they will understand each other. (p. 70) Langer relates concept and conception to meaning in the following way: A concept is all that a symbol really conveys. But just as quickly as the concept is symbolized to us, our own imagination dresses it up in a private, personal concep tion, which we can distinguish from the communicable pub lic concept only by a process of abstraction. (p. 70) She reemphasizes her ideational theory of meaning in a 1960 Quarterly Journal of Speech article. Although she notes that language may be used "to announce one's presence, to greet people, to warn, to threaten, to express pain or joy," she maintains that the great step from anthropoid to anthropos, animal to man, was taken when the vocal organs were moved to regis ter the occurrence of an image, and stirred an equivalent occurrence in another brain, and the two creatures referred to the same thing. At that point, the vocal habit that had long served for communication assumed the function of communication. To evoke ideas in each other's minds. not in the course of action, but of emotion and memory— that is in reflection— is to communicate about something, and that is what no animals do.58 In Philosophy in a New Key she summarizes her view of meaning in the following words: . . . the main lines of logical structure in all meaning- relations are . . . the correlation of signs with their meanings by a selective mental process; the correlation of symbols with concepts and concepts with things, which gives rise to a "short-cut" relation between names and things, known as denotation; and the assignment of elab orately patterned symbols to certain analogues in experi 58Susanne K. Langer, "The Origins of Speech and Its Communicative Function," Quarterly Journal of Speech, XLVI (April 1960), 133. 5 2 - ence, the basis of all interpretation and thought. These are, essentially, the relationships we use in weaving the intricate web of meaning which is the real fabric of human life. (p. 75) In brief, Langer's theory of meaning resembles the referential theorists' in all crucial respects but one: the neaning of a word is not in the object or thing it refers to Dr in the relation between word and object, but in the idea Dr conception that the word refers to or names. The Behavioral Theory Those rhetoricians who subscribe to the behavioral the ory of meaning generally accept the assumption that language Ls a system of symbols, and thus that a word stands for some sort of referent. However, they contend that a word's mean ing is not to be found in its referent— whether object or Idea— or in the relationship between the symbol and what it symbolizes. Instead, they maintain that the meaning of a word is in the behavioral response that it elicits in those who perceive it. Weaver and Ness outline the behavioral theory in these words: Meaning always consists of responses and, therefore, it exists only in persons, not in things or symbols. Meaning is a subtle fusion of responses to present stimulation with the residues of responses to past stimulation, touched off through the mechanism of conditioning. Mean ing is never something done to us; it always is something we ourselves do. . . . A symbol may be defined as a response which is substi tutable for another response. We find ourselves looking at a small, furry animal and we say "cat"? our pronouncing the word, either aloud or silently to ourselves, is sub stituted for the responses of fondling the cat, thrusting the cat away from us, picking up the cat— or any one of a hundred other responses we may previously have made to the 53 visual stimulus, cat.59 Another view of the behavioral theory is offered by Walter Coutu. After developing his presuppositions about the symbolic nature of language, he discusses the meaning of linguistic signals and symbols. Linguistic signals, he asserts, denote sensory realities, whereas linguistic sym bols can represent something "even when it is perceptually absent or doesn't even exist."60 He explains, "both signals and symbols are used as units of coding systems to structure central responses specifying relationships between their referents and the organism making the response" (60). He calls meaning "mental response," and diagrams the cognitive process of sign manipulation in the following way: S 0 >rMs ( ^R, where S = Stimulus, M = meaning, and R = response. He asserts that "to mean" refers to making a coded central process (rMs) specifying what the self is to do about a referent— how to perceive it, how to think about it, what to name it, how to inter pret it, how to relate oneself to it, what to do about it — in short, how to behave in relation to it at all levels of behavior in that particular situational context.61 Coutu's account clearly illustrates the main characteristics of the behavioral theory of meaning: Words do have refer 59Weaver and Ness, pp. 184 and 189. Cf. Bryant and Wallace, pp. 39-41; Dickens, pp. 372-373; Scheidel, p. 35. S0Coutu, "An Operational Definition of Meaning," p. 60. 61Coutu, p. 62. Cf. McBurney and Wrage, p. 12; and Carl H. Weaver and Gary L. Weaver, "Information Theory and the Measurement of Meaning," p. 435. ents, but the meaning of a word is not in the referent but in the response one makes to it, "at all levels of behavior in that particular situational context." Those rhetoricians who adopt the behavioral analysis of meaning rely principally on the accounts of Ogden and Richards, Stephen Ullmann, and Charles Osgood. Early in Meaning of Meaning, Ogden and Richards divide language usage into "referential" and "emotive," the latter of which they identify with "popular or primitive speech."62 They focus on referential usage, "which for all reflective, intellec tual use of language should be paramount" (p. 10). Ogden and Richards define the three crucial parts of the meaning situation as symbol, thought or reference, and referent. The triangle they use to illustrate the relation ships between these three aspects of meaning is probably the most familiar figure in the literature: THOUGHT REFERENT SYMBOL Ogden and Richards explain the figure this way: 62C. K. Ogden and I. A. Richards, Meaning of Meaning: h Study of the Influence of Language on Thought and of the Science of Symbolism, 8th ed. (New York, 194b), p. 10. ^ Between a thought and a symbol causal relations hold. When we speak, the symbolism we employ is caused partly by the reference we are making and partly by social and psy chological factors— the purpose for which we are making the reference, the proposed effect of our symbols on other persons, and our own attitude .... Between the Thought and the Referent there is also a relation? more or less direct (as when we think about or attend to a coloured surface we see), or indirect (as when we "think of" or "refer to" Napoleon), in which case there may be a very long chain of sign-situations intervening between the act and its referent? word— historian— contem porary record— eye-witness— referent (Napoleon). Between the symbol and the referent there is no rele vant relation other than the indirect one, which consists in its being used by someone to stand for a referent. (pp. 10-11) The authors stress the arbitrary nature of the symbol- referent relationship: "The fundamental and most prolific fallacy is, in other words, that the base of the triangle given above is filled in" (p. 15). Like Korzybski and his followers, Ogden and Richards emphasize that clear communi cation cannot take place whenever anyone forgets that "the word is not the thing." However, unlike Korzybski, Ogden and Richards use the idea to introduce their contextual the ory of signs. They point out first that the fact that word-symbols occur in a natural context along with "things" originally gave rise to the primitive idea that there was some "magic bond" between words and objects. Such a mistaken analysis of language encouraged the view that, when a person believes, for instance, that he is alive, he is in a direct relation of some unique kind to an entity which is neither temporal nor spatial, but which is called "the proposition 56 'that I am alive'" (p. 49). Ogden and Richards' goal is to complete an analysis of thought and language that does away with "mysterious entities" like "the proposition 'that I am alive,'" and substitutes a purely causal explanation of thinking. Their causal explanation of thinking, which is also the aasis of their behavioral analysis of meaning, goes roughly like this: when stimuli of any kind impinge on a human organism, they cause the organism to adapt or respond to their presence. The affects in the organism caused by that adaptive response are never completely lost by the organism. Ogden and Richards call that "residual trace of an adapta tion made by the organism to a stimulus" an engram. Engrams tend to "group" in the organism in "engram complexes." These groups of related residual traces of prior stimuli form contexts. The relation between these engram-complexes, or contexts, and words is the following: . . . when a context has affected us in the past the recurrence of merely a part of the context will cause us to react in the way in which we reacted before. A sign is always a stimulus similar to some part of an original stimulus and sufficient to call up the engram formed by that stimulus. (p. 49) In other words, sets of experienced responses congeal some how into contexts, and the type of meaning Ogden and Richards discuss here is caused by a sign reminding us of some part of that context. The reminding act, in effect, causes us to recall the rest of the context, so that we 57 ultimately have a total meaning that is a memory— "residual trace"— of the past experienced responses. In such cases the referents are the objects or experiences that originally provided the stimuli, the references or thoughts are the responses caused by the referents, in the form of engram complexes, and meaning relations hold between the symbol anc the thought and between the thought and the referent, but never between the symbol and the referent. The authors' summary near the end of Chapter IX, "The Meaning of Meaning," is worth reproducing in full: With these considerations before us we can now understand the peculiarities of Symbols with their two-fold "meaning" for speaker and hearer. A symbol as we have defined it . symbolizes an act of reference; that is to say, among its causes in the speaker, together no doubt with desires to record and to communicate, and with attitudes assumed towards hearers, are acts of referring. Thus a symbol becomes when uttered, in virtue of being so caused, a sigr. to a hearer of an act of reference. But this act, except where difficulty in understanding occurs, is of little interest in itself, and the symbol is usually taken as a sign of what it stands for, namely that to which the reference which it symbolizes refers. When this interpre tation is successful it follows that the hearer makes a reference similar in all relevant respects to that made by the speaker. It is this which gives symbols their pecu liarity as signs. Thus a language transaction or a com munication may be defined as a use of symbols in such a way that acts of reference occur in a hearer which are similar m all relevant respects to those which are sym bolized by them in the speaker.63 In other words, a word has meaning because it is a sign, of a mental and to some extent behavioral response of an indi vidual to a referent. Communication occurs when two people 63Ogden and Richards, pp. 205-206. Italics added. ------------------------------------------------------------------------ ’ Sg respond to a word or symbol in ways that are "similar in all relevant respects." Stephen Ullmann's The Principles of Semantics, which has been called "a most comprehensive introduction to modern thought on the subject,"64 brings Ogden's and Richards' behavioral theory up to date without changing their funda mental view of the nature of language and meaning. Like Ogden and Richards, Ullman identifies the three basic terms of symbolization as the sign, defined as "a stimulus similai to some part of an original stimulus and sufficient to call up the engram formed by that stimulus,"65 the symbol, iden tified as "those signs which men use to communicate with one another," and the engram, "the residual trace of an adapta tion made by the organism to a stimulus" (p. 27). He con tinues, "Language symbols are, then, engrams, capable of being actualised. Once actualised they will function in the: same way as any other symbol or indeed sign" (p. 28). In a section called "A Functional Analysis of Meaning," Ullman identifies "the nucleus of the semantic situation" as "this 'bi-polar relation' between the two prima facie components, the sign and the thing signified— that which means and that which is meant" (pp. 68-69) . 6^John B. Newman, (rev. of The Principles of Semart- tics), Quarterly Journal of Speech, XLIV (October 1968), 1237 ----------------------- b5Stephen Ullmann, The Principles of Semantics, 2nd ed (New York, 1957), p. 53. 5 -g ; - Ullmann noticed an ambiguity between Ogden's and 5 Richards' referent and reference, which he resolves by dis tinguishing between a word's name and its sense. The former is the word's appearance to the ear, as it were, "the acous tic shape of the word, the string of phonemes of which it is composed" (p. 69). The word's name has two aspects: a "virtual" one in the language of a given community— English, French, etc.— and an "actualised" one in the speech of a given person. "Similarly, the mental content called up by the name will be termed the sense of the word . . . with the proviso that it refers to mental content, not to the refer ent itself" (p. 69). In light of this distinction, "the functional definition of meaning can . . . be formulated in the following terms: Meaning is a reciprocal relation between name and sense, which enables them to call up one another" (p. 70). Charles Osgood's theory of meaning is twenty years more recent than Ogden's and Richards' but only slightly more popular with rhetoricians.66 Osgood's approach to meaning is based on the same idea that is fundamental to his learn ing theory and, in an important sense, to his total psycho logical system: the mediation hypothesis. Osgood reports that the basic idea for the hypothesis came from a paper 66See note 14. read by C. L. Hull in 1930,67 in which Hull postulated that humans engage in certain behavioral acts "whose sole func tion is to serve as stimuli for other acts" (p. 395). The fact that these acts exist, Hull speculated, enables one to explain how humans can react to objects that are not physi cally present. His insight suggested that such reactions to absent or non-apparent objects are caused not by the objects, but by what Osgood calls a mediation process. Osgood explains the process in the following way: some complexes of stimuli (objects) elicit responses directly, i.e., humans react to them at least in part because of the objects' sensory presence. Examples of these complexes, which Osgood calls "stimulus-objects," include electric shock, food objects, and a rubber ball, which, when placed in someone's hand, "elicits grasping movements of the fin gers, perhaps certain fixational movements of the eyes, and throwing motions" (p. 396). Some sensory-elicited reac tions, however, can occur when the sense stimulation itself is absent. Osgood calls these responses "detachable." These "detachable" responses are the fundamental "stuff" of mediation. When they are "detached" from response com plexes, they "set up pure stimulus acts and, according to this hypothesis, makes possible mediation in general" (p. 396). $7Charles E. Osgood, Method and Theory in Experimental Psychology (New York, 1953), p. 395. & T - Since any response an organism makes produces self stimulation, the "detachable” responses involved in the mediation process can act as stimuli to the organism. The aehavior they elicit, Osgood postulates, is always part of the behavior that was originally associated with the stimu lus object, but which subsequently became detached from it.S8 He stresses the fact that the mediation process associated with the sign of an object must include some por tion of the behavior originally experienced toward the object itself. "... the sign 'means' or 'refers to' a particular object because it elicits in the organism employ ing it part of the same behavior which the object itself elicits."69 The relationship between the mediation hypothesis and Osgood's theory of meaning becomes clear when he discusses "symbolic processes."70 He first defines symbolization as a mediation process, and then asserts that the fact that a word can stimulate responses that were, in part, originally stimulated by an object, illustrates the symbolic nature of words. That is, words can stand for part of the behavior that originally could be stimulated only by the presence of 68This aspect of his theory was schematized previously by Coutu in the diagram: S<--- >rMs<:— *R. That is, a’ stimulus causes a mediation response in the organism, which, in turn, becomes a stimulus for an eventual behavioral response. (Coutu, 60-62). 690sgood, p. 412. 70See Osgood, part-IV, pp. 601-727..____________________ 6 ' 2 some object? thus, words can serve as symbols. Osgood also asserts that his theory can explain the meaning of both connotative and denotative signs. For exam ple, he considers the "largely connotative sign" "spider." He explains that the stimulus-object— "the actual visual pattern of hairly-legged insect body"— first elicits a pat tern of behavior (p. 696). In his own case, the behavior includes "a heavy loading of autonomic 'fear1 activity" (p. 696). Then, "through short circuiting, 'detachable' portions of this total behavior to the spider-object (par ticularly 'anxiety') become conditioned to the sign, the word SPIDER" (p. 696). This mediating reaction produces a pattern of self-stimulation which in turn elicits a variety of overt behaviors, from saying "Ugh!" and shivering, to running away. The denotative meaning of the word "hammer" can be explained in a similar way. Initially the object itself stimulates a pattern of behavior, parts of which become detached from the object and are elicited by the wore alone. In the case of denotation, reduction of the behav ioral responses to those directly associated with the stimulus-object is more complete than with connotational words. That is, the meaning of "hammer" is more delimited to its use as an object, and less to someone's affective responses to it. Osgood differentiates his theory of meaning from other behavioral theories in the following way. He notes that the - g3; "naive application of Pavlovian conditioning principles by early behaviorists such as Watson" had led to the theory that signs have meaning because they elicit behavior that is the same as the organism originally made toward the object (p. 692). When weaknesses in that theory were made appar ent, Charles Morris postulated that signs did not have mean ing because they evoked similar responses, but because they evoked similar dispositions to respond (p. 693). According to Morris' approach, "any pattern of stimulation which is not the object becomes a sign of the object if it produces in an organism a 'disposition' to make any of the responses previously elicited by the object" (p. 693). Osgood's approach is different from that of both of his behavioral predecessors. As already mentioned, he stipulates that the sign "has meaning" not because it evokes all the responses originally made to the object, nor because it evokes a dis position to respond as if to the object, but because it evokes part of the original actual response to the stimulus-- object. In Osgood's words, a pattern of stimulation which is not the object is a sign of the object if it evokes in an organism a mediat ing reaction, this (a) being some fractional part of the total behavior elicited by the object and (b) producing distinctive self-stimulation that mediates responses which would not occur without the previous association of nonobject and object patterns of stimulation. (p. 696) Osgood explains that his use of "object" is not restricted to things that are normally referred to by that word. At one point he writes, "the reader should not be 64 confused by the term 'object' here— 'salty taste in the mouth' is as much a stimulus-object in this context as is 'rubber ball in the hand'" (p. 396). Later, he notes, This is necessary since although one usually thinks of "objects" as being the things denoted by signs, actually any and all patterns of stimulation— a gust of cold, northerly wind against the face, the mass of sensations we call a "bellyache," the pattern of sensations that accom pany being rained upon— may be "objects" at this level of discourse. (p. 691) In sum, Ogden and Richards, Ullmann, and Osgood all offer behavioral theories of meaning. For Ogden and Richards and Ullmann, the meaning of a word is roughly the mental response an individual makes to a symbol. For Os good, words have meaning by virtue of their evoking a response in the organism that is, in part, identical to the response caused by the object that the words originally represented.71 Regardless of which of the three representational theo ries of meaning the rhetorician adopts— the referential, ideational, or behavioral— he usually clarifies his concep tion of the relationships between all the variables of lan guage and meaning— words, ideas, language, speech, thoughts, symbols, etc.— by describing the total communication pro cess. For example, Ross' account of communication illus trates his acceptance of the symbol metaphor and the idea- 71David Berio utilizes Osgood's analysis of meaning in his discussion of communication. See Berio, The Process of Communi c at ion, Chapters 7 and 8. & 5 tional theory of meaning. Let us assume we have a message (which might be referred to as an idea or concept or meaning) which we wish to con vey to another person. Our brain or thinking apparatus now sorts through our storehouse of knowledge, experience, feelings, and previous training to cefine and select the precise meaning we are seeking. Before it is transmitted it goes through a phase which we may think of as a lan guage-attaching or coding event; that is, the refined idea is now encoded or put into symbols, which we commonly think of as language. Gesture, facial expression, and tone of voice may also be considered as symbols or codes. . . . This code is then transmitted, and it has meaning tc you, the sender. . . . The receiver now proceeds to decode the signal or at least attempts to recode it. . . . The listener proceeds, figuratively, to sort and select mean ings from his storehouse of knowledge, experience, and training until he has created in his own mind a replica oi the images and ideas contained in the mind of the sender. The idea, concept, or meaning in the mind of the listener is therefore heavily dependent upon, if not restricted to, the knowledge and experience he can bring to bear on the code. . . . We shall thus define communication as a process involving the sorting, selecting, and sendXng of symbols in such a way as to help a listener recreate in his own mind the meaning contained in the mind of the com municator .7 2 ~~~ In short, according to Ross, ideas or concepts are coded into symbols by the brain, transmitted as some kind of sig nal, and then decoded from their symbolic form into other images and ideas that, as nearly as possible, replicate those in the mind of the sender. Brigance's seven-step explanation is consistent with the behavioral theory of meaning: 1. You start with a thought in your mind. 2. This thought is coded into a series of impulses. . . . known as a neurogram. 3. This neurogram is coded further into phonetic symbols by muscular movements of the tongue, lips, face, diaphragm, etc. XI These phonetic symbols 72Ross, pp. 3-4 5=6 . . . are coded still further into sound waves. . . . 5. The sound waves strike the listener's ear~drums. . . . In this process the sound waves are decoded again into phonetic symbols. 6. These decoded phonetic symbols are sent in the form of impulses along the nerve fibers to the brain . . . and compose a neurogram. 7. In the listen er's brain this neurogram is decoded further into a thought. This last step is the semantic stage, wherein the various brain assemblages translate the incoming sym bols into meaning. The amount of meaning in each instance will depend: (a) on the distinctness and clarity of incoming symbols, and (b) on the variety of brain connec tions, built-in and temporary, that are available for use.7 3 McCroskey outlines a similar process in three steps. First, the communication source must "conceive the idea to be communicated, determine its intent toward the receiver, and select the meaning which it hopes to stimulate the receiver to create in his mind." Then the already-conceived idea is encoded into "a message appropriate for transmission to a receiver." This process includes three "essential" parts: creation of a message, adaptation of the message to the intended receiver, and transmission of the message via some channel to a receiver. The third step is the "decoding process," which involves four parts: "hearing-seeing"; "interpretation . . . the act of the receiver in determining what he believes the source meant by the message trans mitted"; "evaluation," or the source determining the meaning of the message for himself; and "response," eithei: overt or covert or both.7* * 73Brigance, pp. 23-24. 7**McCroskey, pp. 24-27. Cf. Andersen, Lewis, and Murray, p- 27; Waynp, C. Minnick. fffre Art of Persuasion, 2nd 67 In sum, both rhetoricians and the language scholars whom they cite accept the assumption that language is funda mentally a system of symbols. Although there is less agree ment among rhetoricians regarding the nature and function of meaning, the three approaches to meaning that they adopt are representational, i.e., consistent with their belief in the symbol metaphor. Those who accept the referential theory of meaning hold that symbols have meaning because they refer to or name objects or relationships between words and refer ents. Those adopting the ideational theory of meaning pos tulate that the meaning of a word is the idea that it repre sents. And those accepting the behavioral theory of meaning assert that, although words do generally have referents, their meaning is not in the objects or events they refer to, but in the responses they elicit in those who perceive them. Other Approaches to Language and Meaning Three approaches to language and meaning can be found in recent speech-communication literature that, to varying ed. (Boston, 1968), p. 25; W. M. Parrish, "'Getting the Meaning1 in Interpretation," Southern Speech Journal, XXXIII (Spring 1968), 179; Baird and Knower, General Speech, p. 153; Dickens, pp. 16-17; Gray and Braden, p. 456; Hance, Ralph, and Wiksell, p. 21ff.; McCroskey, p. 21; Glen E. Mills, Reason in Controversy: An Introduction to General Argumentation (Boston, 1964), p. 230; Ross, p. 3; Weaverand Ness, p. 184; John B. Newman, "The Semantic Analysis of Ordinary Language," Quarterly Journal of Speech, XLIV (December 1963), 414; and Carl H. Weaver, "Semantic Distance between Students and Teachers and Its Effect upon Learning," Speech Monographs, XXVI (November 1959), 273. &8; iegrees, do not seem to fit into the above classifications. These accounts are briefly dealt with here, in order to make this review of the literature complete. Generally, the three approaches do not deviate significantly from the major assumptions of speech-communication scholars, i.e., that Language is fundamentally a system of symbols, and that meaning involves symbols and whatever they symbolize. h Literary Approach I. A. Richards' volume of six lectures, The Philosophy cf Rhetoric, includes what might appear to be a unique approach to linguistic meaning. Viewing rhetoric as "a study of misunderstanding and its remedies,"7 5 Richards pegins by rejecting the view "that words just have their meanings and that what a discourse does is to be explained as a composition of its bricks" (p. 9). In place of what he calls the "Proper Meaning Superstition,1 1 Richards proposes a contextual theory of meaning. He notes that, as humans, we are responsive to many kinds of stimuli, but that we never respond to a stimulus except in terms of how we have been influenced by other, related stimuli in the past. Even if we experience a completely new stimulus, he argues, "effects from more or less similar happenings in the past would come in to give our response its character and this as far as it 75I. A. Richards, The Philosophy of Rhetoric (New York, 1965), p. 3. 69 went would be meaning" (p. 30). Words operating as stimuli affect us in precisely that way; consequently, their meaning is their delegated efficacy. That is, they act as "substi tutes exerting the powers of what is not there. They do this as other signs do it, though in more complex fashions, through their contexts."7 6 Richards' notion of "context" is crucial to his theory. As he explains, "context" is generally a name for a whole cluster of events that recur together— including the required conditions as well as whatever we . may pick out as cause or effect. But the modes of causal recurrence on which meaning depends are peculiar through that delegated efficacy I have been talking about. In these contexts one item— typically a word— takes over the duties of parts which can be omitted from the recurrence. There is thus an abridgement of the context .... When this abridgement happens, what the sign or word— the item with the delegated powers— means is the missing parts of the context. (p. 34) Richards notes that, to ask how this abridgement happens, i.e., how a word comes to stand for an absent part of a con text, is to "come up against the limits of knowledge at once. No one knows" (p. 34). However, he maintains, "it is enough for our purposes to say that what a word means is the missing parts of the contexts from which it draws its dele gated efficacy" (p. 35). Richards admits that his theory would be difficult to use, for instance, to compare the meaning of two words. "The office of the theorem is much more negative than posi- 76Richards, p. 32. Italics added. 70 tive," he notes; "preeminently what the theorem would dis courage, is our habit of behaving as though, if a passage means one thing it cannot at the same time mean another and an incompatible thing."77 If taken as an independent, distinct theory of meaning, Richards' views in The Philosophy of Rhetoric do not com fortably fit the pattern established in parts one and two above. However, as Marie Hochmuth Nichols illustrates, Richards' 1936 theory cannot be viewed apart from his 1923 theory.78 Compare, for instance, Richards' view that the meaning of a word is "the missing parts of the contexts from which it draws its delegated efficacy," to Ogden and Richards' assertion that the meaning of a word is caused by a sign acting as "a stimulus similar to some part of an original stimulus and sufficient to call up the engram formed by that stimulus." In The Meaning of Meaning "engram complexes" form contexts; in The Philosophy of Rhetoric a context is "a whole cluster of events that recur thgether." In both cases, however, meaning is fundamentally viewed as some missing part of a context that the symbol causes to be remembered. The later book is not explicitly dependent on 77Richards, pp. 37-38. For another brief account of Richards' theory, see Walter R. Fisher, "The Importance of Style in Systems of Rhetoric," Southern Speech Journal, XXVII (Spring 1962), 173-182. 78Hochmuth [Nichols], "I. A. Richards and the 'New' Rhetoric," pp. 1-3. 71 the symbol-reference-referent triangle that forms the foun dation for the earlier analysis. However, all three ele ments can still be identified in the later work: the refer- snt is the original stimulus the mind receives; the refer- ance is the response to the stimulus in terms of previous, related stimuli; and the symbol is the word. Perhaps the principal difference between the two books is that, whereas the authors of Meaning of Meaning distin guish between "emotive" and "referential" utterances and devote virtually all of their attention to "referential" Language, Richards distinguishes between "scientific" and / • "poetic" discourse but stresses the understanding of liter- ary ("religious and poetic") language. In other words, one reason Richards' later treatment of meaning occasionally appears almost to contradict his earlier one is that he is talking about the meanings of significantly different "kinds" of language. The fact that The Philosophy of Rhet- cric, in spite of its title and first lecture, focuses almost exclusively on poetic language is probably the prin cipal reason that it has not often been cited by rhetori- * 7 9 cxans. 79Richards' book was only referenced twice in the text- cooks consulted. See also Osborn and Ehninger, "The Meta phor in Public Address." Their treatment of metaphor, while citing Richards as a source, does not use his terminology, in spite of the fact that the authors focus directly on "the interaction of two thoughts, or interpretants, one of which springs from ..." etc. 72 Existential and Phenomenologica1 Approaches Although a few articles in the recent literature have explored speech-communication from an existential point of view, there is at this writing no clearly articulated approach to rhetoric based on "existentialism."80 However, 9ome comments in the articles that have appeared suggest what sort of view of the nature of language might underlie 3uch an approach, were it to be developed. In his essay, "Speech in the Existential Situation," Henry Nelson Wieman stresses the desirability of communica tion becoming what he calls "creative interchange."81 In Pieman's view, the profound interpersonal understanding that creative interchange implies is beyond the ability of lan guage to produce. The meaning of much communication, he argues, depends on more than the language employed: Words are used, but the meaning conveyed by these words is not at all derived from the verbal context. It is derived from the existential situation, including the unconscious expressiveness and the unconscious interpretation of this expressiveness on the part of the persons in the eoMost historians of philosophy emphasize the impor tance of considering existentialists instead of "existen tialism, " because of the wide variety of viewpoints within the generally anti-Platonic, value-centered, humanistic approach general to most existentialist writers. See, e.g., James Collins, The Existentialists (Chicago, 1952), p. xiii? and John Passmore, "Existentialism and Phenomenology," in A Hundred Years of Philosophy, rev. ed. (New York, 1966), pp. 476-5167-------- 81Quarterly Journal of Speech, XLVII (April 1961), 150. 73 situation.82 A. Craig Baird attacks existentialist philosophers as "anti-intellectual" in a 1962 article.83 In answer to Baird, Myrvin P. Christopherson argues that existentialist philosophers can make several potential contributions to a philosophy of speech. Christopherson stresses the humanis tic viewpoint of existential thought, and suggests that an existential analysis of man rejects absolute categories, emphasizes the importance of decision-making, and offers implicit suggestions for identifying premises for rhetorical enthymemes.0 4 In an article based on his dissertation, Raymond E. Anderson outlines "Kierkegaard's Theory of Communication."85 Although Anderson does not deal with Kierkegaard's view of the nature of language, his stress on the Danish philoso pher's fondness for "indirect discourse" and "edifying dis course" suggest some parallels with the view of language Wieman outlines. According to Anderson, Kierkegaard empha sized the importance of a system of values, and the crucial 02Wieman, p. 155. Wieman developed similar ideas in a shorter article, "The Philosophical Significance of Speech," Central States Speech Journal, XII (Spring 1961), 170-175. 83"Speech and the 'New' Philosophies," Central States Speech Journal, XIII (Autumn 1962), 241-246. 8H"Speech and the 'New' Philosophies Revisited," Cen tral States Speech Journal, XIV (February 1963), 10. 85Speech Monographs, XXX (March 1963), 1-14. 74 role the existential situation plays in one's life. The fifth article that focuses on existentialism is Robert L. Scott's "Some Implications of Existentialism for Rhetoric."86 Scott stresses the existentialist belief that all that is relevant to man is the "existential instant," the here-and-now: "what matters . . . is not the meaning of life in general but rather the specific meaning of a per son's life at a given moment."87 He argues that an existen tial orientation is much more applicable to man's communica tion as described in concrete studies than is either the neo-Platonic or the logical positivist point of view on which most scholars in speech base their professional philosophy. Three other articles deal with phenomenological approaches to communication theory,88 discussion,89 and rhetorical criticism.9 0 Bemis and Phillips point out that 86Central States Speech Journal, XV (November 1964), 267-284. 87Scott, p. 270, citing Viktor Frankl, Man* s Search for leaning: An Introduction to Logotherapy (New York, 1963), pp.T.M-105.------- ----------------- 88James L. Bemis and Gerald M. Phillips, "A Phenomeno logical Approach to Communication Theory," Speech Teacher, XIII (November 1965), 262-269. 89Remo P. Fausti and Arno H. Luker, "A Phenomenological Approach to Discussion," Speech Teacher, XIV (January 1965), 19-24. • 90Richard B. Gregg, "A Phenomenologically-Oriented Approach to Rhetorical Criticism," Central States Speech Journal, XVII (May 1966), 83-90. 75 meaning is the central concern of a phenomenological approach to communication theory: "By definition, communi cation in phenomenological terms means "interchange of mean ing.'"91 However, they treat meaning holistically, as a function of much more than the language involved: "Societal differences determine individual differences, so that mean ings may be similar or may vary regardless of the accuracy of information transmission" (p. 263). In this respect the phenomenological approach resembles the existential one. The authors' main goal is to justify their view of a subjec tive study of communication, as an alternative to what they regard as the predominant, logical positivist-oriented approach. Their argument is basically a humanistic one: "The chief criticism of the mathematical approach to human communication lies in the precision of the machine versus the uncertainty of humans" (p. 266). Their view of lan guage, however, is based on the approaches outlined in parts I and II above: Human communication is based on a mental process that takes the sensations of the body and turns them into responsive verbalizations. The process includes sensa tion, perception, image formation, symbolization, and lan guage. . . . Meaning and interpretation are placed on com munications by persons concerned, and become dependent on both phenomenal environment in general and phenomenal self specifically. (p. 267). Gregg's phenomenological approach to rhetorical criticism is similarly based. His concept of image relies partially on 9 1 Bemis and Phillips, p. 263. 76 Susanne Langer, who suggests that the act of forming mental images which become symbols determines one's linguistic expression. According to Gregg, communication is founded on internal symbolic images, and images comprise one's symbolic reality. Thus the rhetorical critic should try to discover, define, and describe the images which are active within the confines of the rhetorical act.92 Phenomenological or existential approaches are clearly not yet well-developed options in the rhetorician's intel lectual repertoire. In the future these philosophical alternatives might revolutionize the field's approach to many of its subjects. However, at this writing, rhetori cians who have identified themselves with either of these approaches have not yet made a significant break with the traditional view of the nature of language outlined in part one of this chapter. Approaches' in Philosophy and Rhetoric Three essays recently published in Philosophy and Rhet oric come closer than any other material in the speech- communication literature to either defining or manifesting an approach to language that differs significantly from that outlined above. However, none of the three completely develops a new approach to language and meaning. 92Gregg, p. 83. Gregg was also influenced by Kenneth Boulding's analysis of "image." ------------------------------------------------------------77 Alexander Sesonske's essay, "Saying, Being and Freedom of Speech," appears in the first issue of Philosophy and Rhetoric. Sesonske's agreement with the ordinary language philosophers' approach is apparent as he focuses on instances in which language is not used directly to communi cate, but is a "constitutive act." Sesonske argues that when one hits his thumb with a hammer and yells, "Damnl" or when one sings while walking alone in the rain, the purpose of his use of language is not to communicate; "rather the speech acts are part of the com plex organic states we call irritation and joy."93 That is, in those cases language partially constitutes the "complex organic state" itself. He notes that such speech acts have often been called "expressive," since "Damnl" was assumed to express irritation, and a song to express joy. But this suggests that irritation and joy are completely inner states, only causally connected with the utterances that express them. But I think it more accurate to say that uttering the imprecation "Damnl" is a way of being irritated . . . and that singing is a way of being joyous. These linguistic acts are not mere external accompaniments or consequences of an inner state; they are themselves constituents of the states in question. (p. 26) He points out in a footnote that the relationship in such cases between the verbal act and the person's state of being is not a relationship between a cause and its effect but between a part of the whole. That is, "Damnl" is not caused by some "inner affective state"; it is part of the 93Sesonske, p. 26. 78 state of being annoyed. In the remainder of the article, Sesonske contends that such constitutive uses of language are quite commonplace, and that there are important implications of his conclusion for cases involving freedom of speech. However, he seems to embrace both the assumption that language is sometimes not used to name or refer to or express anything, and the assumptions common to the symbol metaphor and the referen tial theory of meaning. He asserts, "words like anger, fear, astonishment, joy, embarrassment, do not refer simply to feelings but to complex states involving many components which vary from person to person and occasion to occasion" (26). It would seem that, if Sesonske were to be consis tent, he would not speak of "words like anger, fear," etc., referring to anything, whether "complex states" or simply "feelings." In some cases such words may be used in state ments describing (and thus, in a sense, 'referring to') an individual's behavior. But in utterances like "I'm utterly astonishedl" or "I fear himl" or "Joy!" the words take on definite constitutive roles, and his analysis of "what" they refer to loses its relevance. Carroll Arnold's essay, "Oral Rhetoric, Rhetoric, and Literature," explores distinctions among those three forms of discourse. Arnold's view of language stresses the active nature of language usage, but does not wholly abandon the symbol metaphor. For instance, he writes, 79 The interplay among rhetorical speaker, listener, and rhetorical discourse is therefore much more complex than is conceivable within the now popular sender-channel- receiver-feedback communication model derived from princi ples of electrical circuitry. One who speaks rhetorically chooses to inaugurate and to try to sustain until attain ment of a purpose a series of events in human relations.9 * * The logical extension of that view could lead to a replace ment of communication analysis in terms of the variables suggested by electrical circuitry with analysis in terms of the nature and function of the complex acts that Arnold calls the "series of events in human relations." However, Arnold does not develop his argument along those lines. Instead he combines his insights with the tra ditional symbolic view of language, in such statements as [the speaker's] verbal and physical behaviors merge to form a flow of symbolic activity representing . . . the rhetorical speaker's entire physical and psychological organization. . . . (p. 199) Lloyd Bitzer's article, "The Rhetorical Situation," is also clearly not a traditional analysis. There are paral lels between his view of rhetoric and some implications of the ordinary language philosophers' view of language devel oped in Chapter III of this study. However, because it unquestionably is not his purpose, Bitzer does not indicate either that his view of rhetoric might be the manifestation of a significantly different approach to language, or that one's approach to non-rhetorical forms of discourse could be substantially altered by the presuppositions underlying his 9* * Philosophy and Rhetoric, I (Fall 1968), 200. 80 analysis. Bitzer's main contention is that "rhetorical discourse . . . obtain[s] its character-as-rhetorical from the situa tion which generates it."95 He develops that contention in several ways, for instance defining rhetoric as "a mode of altering reality, not by the direct application of energy to objects, but by the creation of discourse which changes reality through the mediation of thought and action" (p. 4). Of primary interest, however, is Bitzer's definition of the rhetorical situation: Let us regard rhetorical situation as a natural context of persons, events, objects, relations, and an exigence which strongly invites utterance; this invited utterance partic ipates naturally in the situation, is in many instances necessary to the completion of situational activity, and by means of its participation with situation obtains its meaning and its rhetorical character. . . . So controlling is situation that we should consider it the very ground of rhetorical activity .... (p. 5) Like Arnold, Bitzer stresses the active nature of rhetorical language use. His analysis of rhetoric is in terms of the exigences and constraints that call rhetoric into being, and aow those factors determine the nature and function of rhet orical discourse. The centrality of situation, Bitzer argues, is such that rhetorical utterance is not only called into being by it, but in many cases the utterance itself is a necessary part of the situation (see Sesonske above), and obtains both its rhetorical character and its meaning by 95Philosophy and Rhetoric, I (Winter 1968), 3. 81 nature of its situational role. Bitzer is apparently sug gesting that, instead of viewing utterances as symbols of objects, mental states, or responses, one could profitably view them as parts of living, interpersonal complexes, i.e., as acts in human situations. Summary This chapter was designed to delineate the views of language and linguistic meaning held by rhetoricians and the principal sources upon which they relied. An extensive search of recent speech-communication literature resulted in the following account: Language is fundamentally a system of symbols. Gener ally, symbols are things that stand for something else. Thus, to say that language is fundamentally symbolic is to say that words are fundamentally things that stand for other things. Most of the time the way in which words stand for other things is to serve as their names. This analysis also applies to abstract and syncategorematic words. Although such words may not directly or obviously name some referent, they are always abstractions, or abstractions of abstrac tions, or so on, and they eventually refer indirectly to some objective reality. Meaning is fundamentally representation, i.e., involves the word symbol and whatever it symbolizes. Some rhetori- ri-jaryct png-hnl ate>_ that -the meaning of a word is its referent; others contend that meaning is a function of the relation ship between the word and its referent. Both referential approaches distinguish between denotational and connota- tional meaning. Ideational theorists hold that the meaning of a word is the idea ("concept," "need-want bond," "concep tion," "category") that it symbolizes or represents. Behav ioral theorists focus on human response to linguistic symbol 3timuli, and generally hold that the meaning of a word is the reaction it causes in the person who perceives it. finally, some rhetoricians adopt the ideational theory of meaning for some terms and offer a behavioral or referential oxplanation of other kinds of meaning. The "minority views" of language and meaning are founded on the same assumptions as those previously out lined. I. A. Richards' account of meaning looks different primarily because he is interested more in literary (poetic and religious) language than in referential language. How- aver, his analysis is nonetheless consistent with the Ogden- Richards approach. Attempts to apply parts of existential and phenomenological thought to speech-communication hold promise for the future. However, most authors who mention language and meaning in the essays cited rely on the tradi tional analyses. Three essays recently published in Philos ophy and Rhetoric come closer than anything else in the literature to either defining or manifesting an approach to Language that differs significantly from the traditional---- — 83 one. Alexander Sesonske clearly bases his approach on ordinary language philosophy. The implications of his brief comments about language are potentially very important. For instance, if language is, as he writes, sometimes used "constitutively," then it is a mistake to characterize it as universally "symbolic." Carroll Arnold's article also approaches language as active rather than always symbolic. Lloyd Bitzer's article is an attempt to suggest a new view of rhetoric. Bitzer's approach also breaks with the symbol metaphor tradition, and does so more completely and more consistently than either Sesonske or Arnold. However, Bitzer does not develop the implications of his approach to all forms of discourse. Chapter III introduces a group of modern language philosophers which has developed some of those implications. CHAPTER III ORDINARY LANGUAGE PHILOSOPHERS AND THE NATURE OP LANGUAGE AND MEANING Ordinary language philosophers form one recent trend in what is usually called ’ ’analytic" or "linguistic" philoso phy, a heterogeneous movement "with affinities to the work of many classical philosophers, but flourishing since around the turn of the century."1 Linguistic philosophy is gener ally characterized by the view that philosophical problems nay be solved (or dissolved) either by reforming language or oy understanding more about the language we presently use.2 Ordinary language philosophers, including Ludwig Wittgen stein, Gilbert Ryle, J. L. Austin, P. F. Strawson, William P. Alston, and several others, are identified by the way they take special and systematic account of the role ordi nary language plays in the creation and resolution of philo sophical problems. Their approach to philosophy is "ordi nary language" in two senses: 1Charles E. Caton, Introduction, Philosophy and Ordi nary Language, ed. Charles E. Caton (Urbana, 1963), p. v. 2Richard Rorty, Introduction, The Linguistic Turn: Recent Essays in Philosophical Method, ed. Richard Rorty (Chicago, 1967), p. 3. ____________ 84 _______ 85 . . . first, in contrast with the formalised3 writings of symbolic logicians, [ordinary language] philosophers dis cuss logical issues in an informal way, without recourse to special invented languages, and secondly they believe that a consideration of "what we ordinarily say" is at least a useful preliminary to the discussion of philosoph ical problems. But these points of agreement cover a great many disagreements about the precise importance of formalised logics, and the extent to which the detailed investigation of usages is itself of philosophical inter est. " Those ordinary language philosophers who closely follow Wittgenstein emphasize the therapeutic effect that a careful analysis of everyday expressions can have on the statements and other forms of utterances in which perennial philosophi cal questions are couched. A member of this group might argue, for instance, that the age-old metaphysical question, "What is Reality?" can be permanently laid to rest ("dis solved") by a careful analysis of the question itself and of the various expressions we ordinarily use to talk about real things and unreal things. On the other hand, ordinary lan guage philosophers whose work follows J. L. Austin's are interested in the analysis of various ordinary uses of words and expressions in order to discover insights about our employment of language itself. Philosophers of both groups are convinced that the exacting analysis of everyday 3The spelling and punctuation in all direct quotations from British and Australian sources has been left as it was in the original. "John Passmore, "Wittgenstein and Ordinary Language Philosophy," A Hundred Years of Philosophy, rev. ed. (New York, 1966) , p. 488. 86 language, done in simple, clear terms and striving for com plete understanding, is a philosophical undertaking which promises to provide solutions both to current problems and to those that have confounded philosophers for centuries. This chapter begins with a brief account of the philo sophical background of the ordinary language movement, from pre-Aristotelian philosophical interest in language to a consideration of logical positivism, a movement that imme diately preceded ordinary language. Part II details the ordinary language philosophers' views of the nature of lan guage and meaning. Classical and Medieval Backgrounds Twentieth-century developments in British philosophy are termed "revolutionary" by many commentators.5 For exam ple Von Wright describes the ordinary language approach as "entirely outside any philosophical tradition and without literary sources of influence."6 In a sense this conclusion is accurate. When Wittgenstein, Ryle, Strawson, Austin, and other "linguistic analysts" contend that all traditional philosophy has been plagued by misunderstandings about lan- 5See, e.g., D. F. Pears, "Logical Atomism: Russell and Wittgenstein," The Revolution in Philosophy (London, 1957), p. 41; J. L. Austin, How to Do Things with Words, ed* J. 0. Urmson (New York,. 1965); and Rorty, pp. 2-4. $Georg Henrik Von Wright, "Biographical Sketch," Norman Malcom, Ludwig Wittgenstein: A Memoir (London, 1958), p. 15. ^ guage and that no philosophical progress can be realized until we gain a clear understanding of the informal logic of everyday discourse, they seem to have broken completely with the past. Yet, in another sense, the thread of philosophi cal concern with language runs virtually unbroken through the history of western thought. Anthony Quinton outlines the earliest concerns of lin guistic philosophers: In the ancient world philosophical interest in language centered on the problem of its origin. The issue was characteristically posed in terms of a conflict between nature (Heraclitus) and convention (Democritus). Plato's Cratylus is the chief memorial to this debate. Grammar, as we understand it, was invented by the Sophists, Protag oras' classification of parts of speech, tenses and moods being the school's leading achievement here. The Soph ists' scepticism, their critical and sometimes cynical attitude to forms of argument, prepared the ground for Aristotle and the systematic study of logic. Although Aristotle was not the first Western philoso pher to deal with language, he was the first to offer a com prehensive linguistic analysis. Much of his writing, in works as diverse as the Rhetoric, Metaphysics, Politics, Topics, On the Soul, and the Prior and Posterior Analytics, deals with language as manifesed in its three forms or arts: logic, rhetoric, and poetic. The Categories includes an analysis of homonyms, synonyms, derivatives, simple and composite expressions, predication, and several other topics 7Anthony Quinton, "Linguistic Analysis," Philosophy in the Mid-Century, ed. Raymond Klibansky, Vol. II (Firenze, 1961), 152-153. 88 common to modern philologists, grammarians, linguists, semanticists, and linguistic philosophers.8 Ordinary lan guage philosophers also have some affinity to the Nicho- machean Ethics, in which Aristotle investigates the concepts cf Virtue, Happiness, etc., in part by examining nthe gen eral moral judgment of his age and country, as evidenced by proverbs and quotations, by the forms of language, and by the accepted terms of praise and censure."9 Richard McKeon suggests some relationships between rhetoric and the other arts of language when he writes The art of rhetoric, which is the study of means of per suasion, is therefore closely related to ethics and poli tics; and the manner of their relation may be seen, as was true of the relation of logic to theoretic sciences, by the derivative meanings attached to the word by the extension of "discourse" to the rational processes perti nent to the function of speech.10 In brief, Aristotle established the foundation for subse quent philosophical treatments of language. Like many other theorists he viewed language as a system of symbols that expresses thought by referring to things, but he adapted this fundamental view to his various subjects. In some 8Aristotle, The Works of Aristotle, trans. W. D. Ross, Vol. I, "Categoriae" (Oxford, 1928). 9Aristotle, The Nicomachean Ethics, trans. N. Rackham (London, 1926), p. xix. See, e.g., Book VI, part v, section 2, or VI., v, 7. Cf., Antony G. N. Flew, "Philosophy and Language," Philosophical Quarterly, V (January 1955), 23. l0Richard McKeon, "Aristotle's Conception of Language and the Art of Language: Conclusion," Ciassical Philology, XLII (January 1947), 33-34. 89 cases, he treated language as an "instrument of knowledge and control felative to natural processes and things"; with other subjects it was a "medium of communication and under standing relative to men"; in still other instances, Aris totle dealt with language as "a form of edification and pleasure relative to human products."11 The scope of his treatment has not yet been equaled. In the late middle ages the English philosophers Roger Bacon, Duns Scotus, and William of Ockham all took a criti cal interest in language; and in the seventeenth century Francis Bacon, Thomas Hobbes, and John Locke made signifi cant contributions to the philosophical explication of the nature of language. Bacon focused on the uncritical employ ment of language in his discussion of the idols of the marketplace and treated rhetoric as one of the two fundamen tal intellectual arts, instrumental in the use of reason to guide conduct.12 In addition to introducing Aristotle's Rhetoric to Englishmen, Thomas Hobbes based his logical sys tem on the idea that logic deals fundamentally with language.13 John Locke has been referred to as sharing with 11Richard McKeon, "Aristotle's Conception of Language and the Art of Language: Part I," Classical Philology, XLI (October 1946), 200. 12Quinton, 153. See also Karl L. Wallace, Francis Bacon on Communication and Rhetoric (Chapel Hill"]! 1943) , p. 86. 13Quinton, 153. Cf. Passmore, p. 24. ------------------------------------------------------------------------ 5U Descartes and Bacon the distinction of "having established the modern attitude of mind" and, more importantly, as being "the first writer to express that attitude in words that tfould give it the earliest of its modern vocabularies."14 An understanding of the nature of language still held by many writers can be traced to Books III and IV of Locke's Essay Concerning Humane Under Standing.15 In Book III Locke calls language "the great instrument, and common Tye of Society" (Ch. 1, para. 1). According to his "Doctrine of Signs," the mind makes use of words both for understanding and "conveying its Knowledge to others." As Locke explains, Since the Things, the Mind contemplates, are none of them, besides it self, present to the Understanding, 'tis neces sary that something else, as a Sign or Representation of the thing it considers, should be present to it: and these are Ideas. And because the Ideas of one Man's Mind cannot immediately be laid open to the view of another; nor be themselves laid up any where, but in the Memory, which is apt to let them go and lose them: therefore to communicate our Ideas one to another, as well as record them for our own use, Signs of our Ideas are also neces sary. . . . The Consideration then of Ideas and Words, as the great Instruments of Knowledge, make no despicable part of their Contemplation, who would take a view of humane Knowledge in the whole Extent of it. And, perhaps; if they were distinctly weighed, and duly considered, they would afford us another sort of Logick and Critick, than what we have been hitherto acquainted with. (IV, 20, 4) In this passage Locke reintroduced a theory of the nature of language (words are names of ideas) that has been ^Wilbur Samuel Howell, "John Locke and the New Rhet oric," Quarterly Journal of Speech, LIII (December 1967), 321. 15John Locke, An Essay Concerning Humane Under Stand ing (in four books; London, 1690). 91 the critical focus of the current philosophical "revolu tion," and he foresaw over two centuries ago the new sort of "Logick and Critick" that the philosophers with which this chapter primarily deals have been working to produce. Nineteenth-Century Backgrounds The primary philosophical strains or "styles of think ing" that characterize the two current schools of language philosophy can be identified first in the late nineteenth- century writings of Francis Herbert Bradley and Gottlob Frege. Bradley was fundamentally a metaphysician and thus outside the British empiricist tradition of Locke, Mill, and others; he was much more akin to Hegel than Hume. Bradley's writings are historically significant because of (1) his rejection of the psychologism that Locke introduced to English philosophy and (2) his belief that Reality is an indivisible whole. Subsequent philosophers have almost uni versally accepted the former idea and rejected the latter.16 Locke and most other philosophers of his time viewed philosophy as the study of the foundations of human knowl edge; knowledge was that which was or could be in the human mind. For Locke the contents of the mind were ideas, "actual mental phenomena, psychical facts, something that occurred like a sensation or emotion" (p. 14). Bradley con 16R. A. Wolheim, "F. H. Bradley," The Revolution in Philosophy, p. 12. 92 tended that thinking was not "having" images or ideas, but using ideas to refer to other objects. Since ideas ..could only be manipulated, as it were, through language, the study of thinking necessarily became the study of propositions. Accordingly, Bradley argued, psychology might study what ideas are, but philosophy or logic studies what they mean (p. 14). Thus, Bradley is important to this account because he rejected Locke's psychologism and changed the focus of philosophy from "ideas" or "images" to language, and because he accepted Locke1s view of the nature of language— that words are fundamentally names we use to refer to things. Frege represents a significantly different viewpoint from that of Bradley. His fundamental interest was the phi losophy of mathematics; he was specifically interested in reducing arithmetic to logic, "the bedrock of the whole edi fice of mathematics."17 Frege worked toward his goal by trying to develop an ideal language, one which would mirror the structure of facts without distortion. His aim was similar to Leibniz', who earlier had looked forward to a time when men who found themselves disagreeing could pick up their pens and say, "Calculemus"— "Let us calculate"— because they knew that their disagreements could be removed by strict attention to the rules of their calculus-like 17W . C. Kneale, "Gottlob Frege and Mathematical Logic," The Revolution in Philosophy, p. 31. 93 scientific language.18 Frege recognized that this ideal language would have to be composed of the most basic consti tuents of thought— logical atoms. He also maintained that one could only arrive at these atoms via a thoroughgoing process of reductive analysis, beginning with familiar con cepts and working toward the "ultimate simples." Frege pro duced a work in 1893, called Grundesetze der Arithmetik (Fundamental Laws of Arithmetic), that he thought achieved his goal. However, a contradiction was discovered in the work by a young Englishman named Bertrand Russell. Although he failed to realize his ambition, Frege established a pro gram of analysis and a concept of an ideal language that have strongly influenced— either positively or negatively— all subsequent philosophers of language. Until early in the twentieth century, no one had tried to discover Locke's "new sort of Logick and Critick." So long as philosophers were convinced they could investigate thought directly, by examining essences, feelings, and ideas, they were not primarily interested in language. But when scholars systematically probed these concepts— when Bradley began emphasizing the importance of language, and when people like Russell began wondering what sort of thing an "idea" was and just what the philosopher-psychologists meant when they said that thinking was simply the manipula 10Kneale, p. 39. 94 tion of these abstract entities— the subject matter of phi losophy changed. At that point philosophers began to con sider language as the activity through which thought could be empirically observed and, it seemed, completely under stood. The concern with language almost immediately became a concern with meaning or, more specifically, with the clarification of meaning.19 Ryle asserts that "the story of twentieth-century philosophy is very largely the story of this notion of sense or meaning. . . . Meanings are just what, in different ways, philosophy and logic are ex officio about."2 0 Just as the common concern of modern language philoso phy is meaning, its common method is analysis. In spite of their disagreements, all the philosophers being considered here concur that they should properly focus on the analysis of language. However, they disagree about what constitutes analysis. Some theorists, usually identified as formal analysts or formalists, try to abstract statements into their basic constituents and to express them in an ideal language where the rigidity of formal logic can remove all ambiguity. Informal analysts strive more for explication 19Garth Hallett, S.J., Wittgenstein's Definition of Meaning as Use (New York, 1967), pp. 1-2. Cf. Max Black, Introduction, Philosophical Analysis:' A Collection of Essays, ed. Max Black (Ithaca, 1950), p. 14. 20Gilbert Ryle, Introduction, The Revolution in Philos ophy, p. 8. 95 than reduction, analyzing different uses of an expression and of words and expressions related to it in an effort to surround all the possible meanings rather than restrict their understanding to a mathematically precise definition. For both types of analysts, however, "philosophy is not a discipline coordinate with or comprehensive of the special sciences: it is an activity of clarification directed on to the concepts and methods we employ in these sciences and, indeed, in all forms of thought."21 Modern Backgrounds Russell and Moore22 Bertrand Russell, in the context of this chapter, was 21Quinton, 146. . 22Beginning with Bertrand Russell's Logical Atomism, the history of British language philosophy sometimes is divided into "schools." Philosophical historians almost unanimously concur that, with the exception of the Logical Positivists in the Vienna Circle, twentieth-century language philosophers should not be thought of in terms of "rival schools and groups, with rival heroes and leaders." (G. J. Warnock, The Revolution in Philosophy, p. 124. Cf. Morris Weitz, "Oxford Philosophy," Philosophical Review, LXII [1953], 187; Flew, "Philosophy of Language," p. 21; and Rorty, p. 12.) However, as J. 0. Urmson points out, "It would be hypersensitive to allow the fact of individual variety of view to obscure the fact that a general pattern of thought can be discerned, which it is helpful to examine both as a preliminary to the more detailed study of the views of indi vidual philosophers and as a background against which the present can be better understood." (Philosophical Analysis: Its Development Between the Two World'Wars [Oxford, 195b], p_Tx7J One further caveat: the philosophical development that is being traced in this chapter is almost exclusively limited to the philosophers' views of language. In most 96 the primary modern proponent of formal analysis. His pro gram was called Logical Atomism: it was logical in that/ whereas others believed philosophers could analyze ideas, Russell maintained they must analyze propositions; it was atomistic in that its goal was the analysis of thought into its ultimate, simple elements. Although violently opposed to Bradley's metaphysics, Russell was metaphysical in his own way. His chief philosophical point in focusing on lan guage was that it was the best way to approach reality. He believed that the only effective way to arrive at the ulti mate constituents of reality was "first to construct the outlines of a logically adequate language and then to read off the structure of reality from that."23 His technique was aimed at stripping off the "surface complexities of the world" so to arrive at and isolate the "last residue in analysis."2* * The atoms that made up this residue were, for cases only a minor part of each theorist's program is pre sented. The resulting distortion obscures the fact that the several theories are in many ways not parts of any single historical development. For instance, Bradley and Russell disagreed fundamentally in their metaphysical theories; Moore saw metaphysics from a third point of view that was significantly different from the other two. For the pur poses of this chapter, differences of this sort are ignored, so that we may focus on how each philosopher viewed lan guage . 23William P. Alston, "Ordinary Language Philosophy," Readings in Twentieth-Century Philosophy, ed. William P. Alston and George Nakhnikian (New York, 1963), p. 296. 2l *G. J. Warnock, English Philosophy Since 1900 (New York, 1966), pp. 22-23. _______ ------------------------------------------------------------57 Russell, the world's ultimate constituents. Russell wanted to analyze facts, not things. Facts, he maintained, are stated in propositions, which in turn are composed of words. Some words are simple; others are com plex. A proposition composed of only a proper name and a simple predicate is an "atomic" one, and the fact such a proposition states is an "atomic" fact.25 His final goal was an ideal language that could state all such atomic facts; he was led toward it by two interrelated assumptions. The first assumption was that there is often a discrep ancy between the apparent logical form of a proposition and its true logical form, i.e., between the appearance of a statement and what analysis will reveal are its actual logi cal constituents. Consider, for instance, the statement, "The present king of France is bald." It cannot be a propo sition because it does not state a fact; it cannot be either true or false because there is no king of France for it to refer to. But the statement seems to make sense. What has happened, Russell argued, is that the apparent grammatical form is obscuring the true logical form of that statement. By analyzing it, Russell would translate the statement into something like, "There is a being 'x', about which it may be predicated that it is both bald and the present king of France." With its true logical form thus revealed, one 2SWarnock, p. 23. could see that the proposition is false. By following this tind of program, Russell believed he could create a logi cally adequate, completely structured, wholly unambiguous, ideal language, in which there would never be any discrep ancy between a statement's apparent and its true logical form. , r' The second assumption was that the meaning of a word is the thing it names. According to D. F. Pears, Russell's approach rests on the assumption that the meaning of a word is the thing designated by it. Once this is granted, then, since Russell thought that we must be able to learn the meanings of some names of things separately and in complete isola tion from everything else, it follows that there must be absolutely simple particulars.2 6 Russell believed there are two general sorts of objects that words name: particular things and general things, the - latter of which are qualities and relations. After analyz ing statements, he maintained, one can discover which of the two sorts of things each word is the name of. One of the goals of his analysis was to reduce the number of kinds of entities that must be assumed to exist in order for proposi tions to be meaningful. His ideal language would be com posed only of references to sense data, or the "atoms" of perception, and predicates: "It is only names of sense data, then, which will be allowed in the atomic sentences of the ideal language, together with predicates signifying 26Pears, p. 53. 99 properties of sense data or relations between them.”27 Russell's program failed, partly because he did not consistently follow his own advice, and partly because "the logical atomists, like other metaphysicians, were seeking clarity and order, and where they failed to discover clarity and order, they invented them."28 G. E. Moore was the first important "informal" analyst. It is more difficult to characterize his philosophy than to recognize the influence of his work. Moore was an analyst, but his system of analysis was different from Russell's in both method and goal. He believed philosophical difficul ties arise mainly because of "the attempt to answer ques tions, without first discovering precisely what question it is which you desire to answer. . . . Often what moral phi losophers have before their minds is not one question but several"; what they need to do is "the work of analysis and distinction," which, though fruitful, is extremely diffi cult.29 For instance, in his classic article, "The Refuta tion of Idealism,"3 0 Moore identified the basic assumption of the idealists as "esse = percipi,1 1 i.e., "to be is to be perceived." He analyzed that assumption, and found all 27Alston, "Bertrand Russell," p. 293. 28Pears, p. 46. 2 9G. E. Moore, Principia Ethica, preface; cited in G. A. Paul, "G. E. Moore: Analysis, Common Usage, and Com mon Sense," The Revolution in Philosophy, p. 58. 30Mind, XII (1903). 100 three terms to have several meanings. For example, the copula "is" can mean (1) that esse and percipi are synony mous, as in a definition; (2) that, although the two terms are not identical, one includes the other as part of its meaning; or (3) that, whenever there is being there also is perception. He settled on the third interpretation; the remainder of his article was an effort to drive a wedge cetween the necessary connection between esse and percipi. Moore's analysis was an attempt to clarify ordinary concepts aot by translating them into an ideal language, but by sxplicating what they mean in ordinary language. Another of Moore's contributions was his "common sense" approach to philosophical problems, i.e., his ability to simplify the use of an abstract term by familiar concrete substitutes. For instance, he began his analysis of the idealist claim, "the entire universe is spiritual," with the remark, "Chairs and tables, and mountains seem to be very different from us; but when the universe is declared to be spiritual, it is certainly meant to assert that they are far more like us than we think."31 Similarly, he pointed out that idealist philosophers who have voiced doubt about the existence of any world external to them have "intended me to 31Moore, Philosophical Studies, p. 1; cited in Paul, ?. 62. 101 loubt that here 1^ hold up my hand in front of me."32 In 3hort, Moore refused to doubt what he regarded were absolute truisms. It seemed to him that other philosophers sometimes denied what every sane man knew quite well to be true, and foore tried patiently to persuade them that they ought not to do that. Moore also emphasized "the merits of the common tongue."33 He insisted that the language used by the ordi nary person should not lightly be declared vague, ambiguous, or paradoxical. He believed that whenever philosophical problems could be traced to ordinary language, their solu tion could not be found by translating the offending state ments into some idealized linguistic calculus, but by care fully examining them to discover "their correct analysis." Most of Moore's writing was devoted to spirited attacks of what he thought were examples of fuzzy thinking by other philosophers. Yet, as Max Black noted, "careful examination of his essays shows again and again that he fails, or rather loes not try, to arguey on crucial issues he seems to attack lis opponents by vehement affirmation and reiteration."3 * * En defense of Moore's approach, it might be observed that he was consistently trying to show that there is nothing better bo do with common sense truisms than to assert them: "If an 32Moore, Proof of an External World; cited in Paul, p. 63. 33Pears, p. 67. 3l*Black, p. 7. 102 inveterate metaphysician is consistent enough to deny them— to say it is not certain that he sees a hand . . . --argument is not going to help."35 In sum, Russell introduced a program of formal language analysis designed to create an ideal language in which all words named either particular things or general things. Koore concentrated on the informal analysis of ordinary lan guage based on the rejection of Cartesian skepticism and the "common sense" belief in what every sane man knows quite well to be true. Wittgenstein and Logical Positivism Ludwig Wittgenstein's thought has dominated both the formal and the informal strains of twentieth-century lan guage philosophy. Wittgenstein's philosophy, like his life, san be neatly divided into two phases, even though, like most neat divisions, this one sometimes oversimplifies and distorts. As a young man he developed an interest in the philosophical foundations of mathematics, which soon became an obsession that dominated his life. He studied with Russell for eighteen months at Cambridge, but did not enjoy the relaxed academic atmosphere. He wrote his first major philosophical work, Tractatus Logico-Philosophicus, while serving in the Austrian army in World War I, and during his internment as a prisoner of war in Italy. He published the 35Black, p. 7. 103 Tractatus in 1921 and, sincerely believing he had solved all possible philosophical problems, retired from academic life. Wittgenstein gave away a large inheritance, enrolled in a teacher's college in Vienna, and taught school in lower Austria for six years. Continually dissatisfied with him self, he gave up teaching, became a monastery gardener, and then spent two years in Vienna. Conversations with Austrian and English philosophers convinced him there were weaknesses in his work, so he returned to Cambridge and to philosophy in 1929. He spent seventeen years at Cambridge, teaching, living in stark simplicity, and writing his second major work, Philosophical Investigations. By 1947 he had become so depressed that he resigned and moved to the west coast of Ireland, where he became a legend for his ability to tame wild birds. He traveled restlessly between Cambridge, Vienna, America, and Ireland until he died of cancer in April, 1951.36 His Philosophical Investigations was not published until 1953. Those who maintain there are "two Wittgensteins" char acterize them in the following way: "Wittgenstein I" was the author of the Tractatus. He believed ordinary language was an inadequate medium for the solution of philosophical 36Ved Mehta, Fly and the Fly-Bottle; Encounters with British Intellectuals (Baltimore, 1965), pp. 81-88; Alston, pp. 4^5-505; and Von Wright in Malcom, pp. 1-22. 104 problems, and that a logically perfect language was needed to replace it. He agreed with Russell that there exist, in a sense, nothing but simple, elementary facts, and that these facts can be expressed in elementary propositions. He added the principle that elementary propositions have mean ing by virtue of being "pictures of facts"; "the essential relation in all meaning was that of 'picturing.'"37 "Wittgenstein II" was the author of the Philosophical Investigations. He believed "ordinary language is all right," and that philosophical problems arise not because our language is faulty but because philosophers misconstrue it or are "bewitched" by it. He explicitly rejected both the monolithic simplicity of his approach in the Tractatus and its attempts to make philosophy scientific. "Wittgen stein II" emphasized the immense diversity of the ways words are used, and the variety of purposes served by language. He repudiated the search for essentials and perfect defini tions, which he claimed were "neither possible nor desir able." 38 He focused instead on the careful examination of concrete instances of the use of words, in an attempt to clarify ordinary concepts and thus "dissolve" philosophical problems. Regardless of the adequacy of this bifurcation of his 37G. J. Warnock, "The Philosophy of Wittgenstein," Philosophy in the Mid-Century, 203. 38Passmore, p. 432. _ 105 life and writings, Wittgenstein "can be identified without hesitation as the most powerful single influence upon recent and contemporary philosophy in Britain, and to some degree also in America."39 His most influential works have been of great importance to two often contradictory contemporary philosophical movements. The Tractatus was fundamental to Logical Positivism; the Philosophical Investigations pro vided the foundation for ordinary language philosophy.k0 The philosophical movement known as Logical Positivism originated with the Vienna Circle, one of the few clearly demarcated "schools" in contemporary philosophical history. The Circle formed in 1922 around Moritz Schlick, when he occupied a chair of philosophy at Vienna University. Mem bership included the philosophers Schlick, Friedrich Wais- mann, Rudolf Carnap, Otto Neurath, Herbert Feigl, and Victor Kraft and mathematicians Hans Hahn, Karl Menger, and Kurt "vjrarnock, "The Philosophy of Wittgenstein," p. 203 . ‘ t0This statement is admittedly an oversimplification. There were views that may still be identified as logical positivist that cannot be attributed to the Tractatus. Even more importantly, Wittgenstein had influenced many students long before the publication of his Investigations. Moore's notes from sessions with Wittgenstein, and the surrepti- tiously circulated Blue and Brown Books were means for the dissemination of Wittgenstein's later ideas, even though he strenuously objected to their publication. In addition, as has already been noted, not all so-called ordinary language philosophers emphasize the therapeutic nature of language analysis that chiefly occupied Wittgenstein and character ized his program. However, the statement is, I think, accurate and helpful enough to stand, given these qualifi cations . 106 Sodel. **1 Wittgenstein was never officially a member of the group, but he was a close friend of Schlick and Waismann, and the Tractatus Logico-Bhllosophicus "to a considerable axtent . . . set the pattern which, at least in its early lays, the Vienna Circle followed."**2 Quinton suggested that bhe group split into two "wings," a number of "fairly ortho dox Wittgenstenians," and "a more adventurous left wing."1 *3 3ut the group as a whole was much more noted for its simi larities than for its differences. The Circle took over an established philosophical journal, renamed it Erkenntnis, and used it to publicize their views. By 1940, "the Vienna circle, as a movement, was a thing of the past. So, in a way, was logical positivism. But many of its ideas live an."1 *1 * Logical positivist tenets endure partly because of their iconoclasm. For instance, the Vienna Circle is still identified by its extreme distaste for traditional metaphys ics. The Circle viewed metaphysical statements not as false, but as nonsensical. This position was based in turn on the major assumption that grounded positivist thought: the principle of verification. The verification principle has been expressed in vari ous ways by several different writers. A common phrasing ^A. J. Ayer, "The Vienna Circle," The Revolution in Philosophy, p. 70. /*2Ayer, p. 70. 1 *3Quinton, p. 157. ^Ayer, p. 73. 107 is, "the meaning of any statement is shown in the method of its verification," assuming that verification must always at least terminate in empirical observation or sense-experi- ence.1 *5 Another version goes, "the meaning of a statement is the possible fact it depicts; if this fact obtains, the statement is true."1 *6 In either case, since no statements of ethics, religion, or metaphysics can possibly depict facts or be empirically verified, all such statements are adjudged meaningless. Proponents of the principle stress that it does not rule out the possibility of interesting uses of language other than verifiable statements; it just postulates that those uses are not capable of asserting facts. Critics of the verification principle or of logical positivism in general have stressed three weaknesses in the Vienna Circle's program. First, application of the princi ple raises serious questions about apparently obvious events. If verification can only occur when there is a pos sibility of empirical observation or sense-experience, how can one ever succeed in meaning something about the past? Similarly, can it really be that any two people can ever mean or understand the same thing? Verification for each must occur by virtue of his own sense-experiences. Since I * >8Warnock, English Philosophy Since 1900, p. 30. ■^Quinton, p. 156. 108 cannot "have another's pain," or his happiness, can we ever share a meaningful event? And if what any statement really neans must be systematically different for each person who might hear or read, speak or write it, how do we succeed in communicating with each other? **7 A second major difficulty with the principle of verifi cation was raised by Wittgenstein in the Tractatus, and led tiim to conclude that work with the idea that nothing con tained in it could be said to have any meaning, i.e., his entire treatise was necessarily meaningless. The problem was that the positivists held that statements could be either of (not about) facts or of the relations between lin guistic expressions. However, statements that purported to relate linguistic expressions to facts would belong to neither class, and hence were disallowed. Since all the statements in the Tractatus— and in the rest of positivist writing— purported to be linguistic expressions about existent facts, none of them could have any meaning. Hence, "Whereof one cannot speak, thereof must one remain silent." A third, related difficulty that positivists encoun tered was the problem of verifying the verification princi ple itself. If statements could only have meaning by virtue of factual or sense-experience verification, what evidence could be adduced to establish the "meaningful-ness" of the l *7Warnock, English Philosophy Since 1900, p. 33. 109 verification principle? A. J. Ayer, a current proponent of the positivist point of view, admitted that the principle is not verifiable. However, he argued, it was not meant to be. It was put forward as a defini tion, not as an empirical statement of fact .... It purports to lay down the conditions which actually govern our acceptance . . . of common sense and scientific state ments, the statements which we take as describing the world . . . .'*e Beyond that, Ayer did not rebut the charge. The positivist program also included the typically formalist drive for an ideal language. Positivists agreed with Russell that ordinary, natural languages embody and sncourage the confusions of ordinary thought, and only deliberately constructed languages can avoid them. This aspect of positivist doctrine is still actively followed by contemporary symbolic logicians, whose ultimate goal is the creation of an ideal language, i.e., one that is free from all the blemishes of ordinary language, including vagueness, ambiguity, emotive meaning, pointless irregularities in grammar, and discrepancy between the apparent logical form and the real logical form of an expression.l f 9 In short, logical positivists believed that "science gives us our knowledge of the world.”50 Thus the role of lt8Ayer, p. 75. For another account of this same prob lem, see Urmson, Philosophical Analysis, pp. 168-171. * * 9Roulon Wells, "Philosophy of Language," Philosophy in the Mid Century, p. 139. S0Ayer, p. 78. -----------------------------------------------------------------------m r the philosopher was to analyze and clarify the concepts which are used in scientific language. Philosophy was to oecome the logic of science, analyzing and refining the statements of science to insure their correspondence with empirical fact. "The result of philosophizing is not to establish a set of philosophical propositions, but to make other propositions clear."51 The Common Thread; Words Are Fundamentally Names The view of language shared by all of the philosophers from Aristotle to "Wittgenstein I" is that words are essen tially names of things, and hence utterances are strings of names, each name referring to some sort of entity. John Locke did not originate the doctrine, but he clearly expli cated it in 1690. A human, he wrote, uses words "as signs of internal conceptions . . . as marks for the Ideas within his own Mind. . . .1,52 For Locke, words were essentially names of ideas. John Stuart Mill developed the same view of words in his System of Logic.53 "Following Hobbes' lead," he "took it for granted that all words, or nearly all words, are names."51* Mill noticed, however, that two different 51Ayer, p. 79. s*Locke, III, 1,1. 53John Stuart Mill, A System of Logic, Ratiocinative and Inductive, 8th ed. (New York, 1874). sl*Gilbert Ryle, "The Theory of Meaning," Caton, p. 131. Ill descriptive phrases may both fit the same thing or person, so that words cannot strictly be equated with what they describe. He met this difficulty with his theory of conno tation and denotation, in which he maintained that most words do two things at once. They denote the things or per sons they are the names of, and they also connote, or "sig nify the simple and complex attributes by possessing which the thing or person denoted is fitted by the descrip tions."55 In short, words either name the things or people they refer to, and/or they name "essential attributes," i.e., attributes the person or thing possesses that make it fit the description the words give it. Gilbert Ryle sug gested that Mill's theory of meaning "set the questions, and in large measure, determined the answers" for thinkers as diverse as Brentano, Meinong, Husserl, Bradley, Jevons, Venn, Frege, James, Peirce, Moore, and Russell.56 Bertrand Russell spent a great deal of effort trying to strengthen the most obvious weaknesses of the view that words are fundamentally names. As already noted, Russell's entire theory rested on the assumption that the meaning of a word is the thing designated by it. However, Russell recog nized that inventing mystical entities for descriptions like "the round square" to name was not a satisfactory way to construct an adequate theory of meaning. In order to sal- 55Ryle, "The Theory of Meaning," p. 135. _____56Pyl "The Theory of-Meaningp.. 13Q. . _______________ 112 /age the words-are-names principle, he created his theory 3f descriptions. He argued that the grammatical form of a phrase like "the round square" leads one to believe it 3hould have a subject, but that by analyzing the phrase, its true logical form would be revealed, in which its words would name entities. The final, correct logical form Russell gave for the phrase, "The round square does not sxist," was: "It is not true that there is an entity c such that the propositional function 'x is square and round' is true, if x is c and otherwise false."57 Other troublesome words like "Pegasus," which could have no referent to name, Russell viewed as truncated or abbreviated descriptions, s.g., for the descriptive phrase, "the winged horse captured oy Bellerophon."58 In sum, Russell recognized more clearly than had anyone before him the difficulties inherent in the assumption that all words name things. But rather than reject the assumption, he worked to overcome the awkwardness Df the theory by devising explanations for all the confound ing examples he could discover. The assumption was modified in Wittgenstein's Tracta tus . Wittgenstein was familiar enough with the weaknesses 57Quoted in Justus Hartnack, Wittgens tein and Modern Philosophy, trans. Maurice Cranston (New York, 1965) , p. 11(>, no reference given. 58Leonard Linsky, Introduction, Semantics and the Phi losophy of Language, ed. Leonard Linslcy (Urbana, 1952) , p. 6. --------------------------------------------------- 113 in Russell's analysis not to accept it at face value, but his own theory that language essentially pictures reality failed to break completely with the words-are-names tradi tion. Wittgenstein expressed his "picture theory" in this way: All language is made up of propositions, and all propositions are logical pictures in which "one name stands for one thing, and another for another thing, and they are connected together. And so the whole, like a living pic ture, presents the atomic fact."59 He maintained that lan guage does not picture objects but the combination of objects that constitute a fact. For example, if there is an object, and it is red, then that it is red is a fact. Lan guage pictures that fact, which is in turn made up of objects ('red' and 'object') that have names.60 The theory that words are names is far from dead. Rudolf Carnap's 1947 publication, Meaning and Necessity: A Study in Semantics and Modal Logic, perpetuates the view. The theory has also been propounded recently by the semanti- cists Alfred Tarski and C. I. Lewis. Tarski defined seman tics as a discipline which "deals with certain relations :>etween expressions of a language and the objects (or 59Ludwig Wittgenstein, Tractatus Logico-Philosophicus, with an introduction by.Bertrand Russell (London, 1922), 4.0311. 60See Hartnack, p. 19. n r 'states of affairs') 'referred to' by those expressions."61 C. I. Lewis accepted Charles S. Peirce's notion that "the essentials of the meaning-situation are found wherever there is anything which, for some mind, stands as a sign of some thing else . . . ,"62 Lewis also wrote of meaning in terms of "denotation or extension" and "connotation or intension." He defined the denotation of a term as "the class of all actual or existent things to which that term correctly applies" (p. 52). His explanation of connotation indicated how completely that concept also conforms to the assumption that words are names: If nothing would be correctly namable by "T" unless it should also be namable by "Ax", by "A2"/ and . . . and by "An:, and if anything namable by the compound term "Ax and A2 and . . . and An" would also be namable by "T" then this compound term, or any which is synonymous with it, specifies the connotation of "T" and may be said to have the same connotation as "T." (p. 52) The words-are-names theory is a common thread running through the fabric of the philosophy of language from pre- Aristotelian Greece to 1929. In 1929 Ludwig Wittgenstein returned to Cambridge and began to work out the ideas that were expressed in his lecture notes, published as the Blue Book and the Brown Book,6 3 and his classic work, Philosoph- 61"The Semantic Conception of Truth," Semantics and the Philosophy-of Language, p. 17. 62C. I. Lewis, "The Modes of Meaning," Semantics and the Philosophy of Language, p. 50. 63New York, 1958. 115 * tca 1 Investigations.6* * Wittgenstein's rejection of the pic ture theory he had developed in the Tractatus, and of its underlying assumption that words are essentially names is the fundamental difference between "Wittgenstein I" and "Wittgenstein II." The belief that words are not simply lames is also the fundamental change ordinary language phi losophers have made in the traditional view of the nature of Language. Their rejection of the words-are-names theory will be developed in part II of this chapter. "Wittgenstein II": Meaning Is Use in.the Language Ludwig Wittgenstein's two conflicting views of the lature and function of language consisted, in a sense, of the detailed explication of two metaphors that struck him with uncommon force. According to Von Wright's account, Wittgenstein's first insight came while he was serving with the Austrian army.65 He was reading a newspaper in a trench one day when a diagram of the sequence of events in a car accident caught his eye. As he studied it, he became aware that the diagram of the accident stood for a possible pat tern of occurrences in reality, i.e., he saw a correspon dence between the parts of the drawing and specific objects or events in the world. He saw a similar correspondence between the parts of a sentence and elements of the world, 6‘ "Oxford, 1958. 6 5Von Wright, in Malcom, pp. 7-8. 116 and thus developed the analogy that a proposition is a kind of picture. This was the picture theory which formed the basis of Wittgenstein's view of language in the Tractatus. More than fifteen years later, "as suddenly as a sketch of a car accident had inspired the ideas in the Tractatus, so a gesture of an Italian friend destroyed them."6 6 In 1933, Wittgenstein and an economist friend, Piero Sraffa, were arguing over the ideas of the Tractatus. As Malcom recorded it, One day (they were riding, I think, on a train), when Wittgenstein was insisting that a proposition and that which it describes must have the same "logical form," the same "logical multiplicity," Sraffa made a gesture, famil iar to Neapolitans as meaning something like disgust or contempt, of brushing the underneath of his chin with an outward sweep of the finger-tips of one hand. And he asked: "What is the logical form of that?" Sraffa's example produced in Wittgenstein the feeling that there was an absurdity in the insistence that a proposition and what it describes must have the same "form." This broke the hold on him of the conception that a proposition must literally be a "picture" of the reality it describes.67 A third incident furnished Wittgenstein with the meta phor that replaced the picture theory Sraffa1s gesture had obliterated. One day when Wittgenstein was passing a field where a football game was in progress the thought first struck him that in language we play games with words. A central idea of his philosophy [in the Investigations], the notion of a "language game," apparently had its genesis in this incident.6 Much of Wittgenstein's Blue Book and Brown Book, and approx imately the first one hundred sections of his Investiga- 66Mehta, p. 85. 67Malcom, p. 69. 68Malcom, p. 65. 117 bions, are devoted to refuting the words-are-names theory that had reached its culmination in his Tractatus, and in establishing his concept of language-games. The first paragraph of the Investigations is a Latin quotation from St. Augustine's.Confessions, which, according to Wittgenstein, presents "a particular picture of the assence of human language. It is this: the individual tfords in language name objects— sentences are combinations of such names."69 Wittgenstein criticizes that viewpoint for several reasons. First, he suggests that even if one considers the archetypical example of a word naming a thing — the case of a proper noun— the theory is incorrect. It is important to note that the word "meaning" is being used illicitly if it is used to signify the thing that "corresponds" to the word. That is to confound the mean ing of a name with the bearer of the name. When Mr. N. N. dies one says that the bearer of the name dies, not that the meaning dies. And it would be nonsensical to say that, for if the name ceased to have meaning it would make no sense to say "Mr. N. N. is dead." (para. 40) As he subsequently points out, it is incorrect to hold that proper names have any sort of "meaning." The writer's younger daughter Lisa announced recently that she had named a new doll "Calabera." "That's interesting," her mother commented, "What does 'Calabera' mean?" After only a brief pause Lisa replied, "'Calabera' doesn't mean anything. It just tells who she is." 69Ludwig Wittgenstein, Philosophical Investigations (Oxford, 1953), paragraph 1. All references to the Investi gations will be to the numbered paragraphs. ------------------------------------ — ---------------------------------------------------------------------------- IT 8 Wittgenstein argues that it is equally evident that words other than proper nouns also do not function as names. "Think of exclamations alone, with their completely differ ent functions. Water1 Away! Owl Help! Fine! No! Are you inclined still to call these words 'names of objects?"' (para. 27). He also considers the words "block," "slab," and "pil lar," all of which traditional philosophers would agree have meaning by virtue of their being names of specific objects. However, Wittgenstein asks, if a carpenter or mason were to call out any of those words to his helper on a construction project, would he be using the words just to name objects? Wittgenstein argues that much more would be meant: "When I call 'Slab!1, then what I want is, that he should bring me a slab" (para. 19). In other situations the single word could mean "Hand me a slab," "Bring him a slab," "Bring two slabs," etc. "Think of the tools in a tool-box," Wittgenstein writes: "there is a hammer, pliers, a saw, a screw-driver, a rule, a glue-pot, glue, nails and screws— The functions of words are as diverse as the functions of these objects" (para. 11). Thus, he argues, it is a fundamental mistake to assume words are always (or just) names of objects; some times, in some senses they refer to entities. But in other sases they serve entirely different purposes. The important fact the old theory ignored is that words have meaning only when people are employing them in specific contexts. Malcom phrases the point/ which he calls "espe cially noteworthy" and a summary of "a good deal of his phi losophy," as, "an expression has meaning only in the stream of life."70 Wittgenstein views language fundamentally as "a form of. life,"71 as an instrument for the accomplishment of various human ends; words are thus tools which are used to do various kinds of jobs. The use of a given expression is a function of the mode of activity or the form of life that makes up the context it occurs in. Wittgenstein calls these "forms of life" "language-games." Just as he argues is the case with all words, Wittgenstein's term "language-game" does not have just one meaning. He suggests the breadth of his concept by his examples, which include giving orders and obeying them, describing the appearance of an object, or giving its measurements, reporting an event, framing and testing a hypothesis, play-acting, guessing riddles, trans lating from one language into another, asking, thanking, cursing, greeting, praying, and lying (para. 23, 50). In each instance of a language-game with which Wittgenstein deals, he stresses the vitality of language. He emphasizes that "speaking and writing belong to intercourse with other people. The signs get their life there, and that is why the 70Malcom, p. 93. Italics added. 7Wittgenstein, Investigations, para. 19. 120 language is not just a mechanism.1,72 His rejection of the words-are-names theory of meaning, his belief that language is a form of life, and his confi dence that what words mean is a function of how people use them, lead him to substitute this view for the traditional one: "For a large class of cases— though not for all— in which we employ the word 'meaning' it can be defined thus: the meaning of a word is its use in the language."73 Post-Wittgenstein Ordinary Language Viewpoints Although the Philosophical Investigations was not pub lished until 1953, Wittgenstein's influence on his students was apparent from the time he began lecturing at Cambridge in 1929. The extent of his ability to affect others is indicated by a remark made by one of his most famous pupils, John Wisdom. In the first footnote to his article, "Philo sophical Perplexity,"71* which has been called the "first full-blown announcement of the new philosophy to the outside world,"7 5 Wisdom writes, Wittgenstein has not read this over-compressed paper and I 72Ludwig Wittgenstein, The Blue and Brown Books (New York, 1958), p. xv. 73Wittgenstein, Investigations, para. 43. 7**John Wisdom, "Philosophical Perplexity," Proceedings of the Aristotelian Society, n.s. XXXVII (1936-1937), 71-88. 75Alston, p. 508. 121 warn people against supposing It a closer Imitation of Wittgenstein than it is. On the other hand I can hardly exaggerate the debt I owe to him and how much of the good in this work is his— not only in the treatment of this philosophical difficulty and that but in the matter of how to do philosophy. As far as possible I have put a W against examples I owe to him. It must not be assumed that they are used in a way he would approve. It would be incorrect to assume that all current ordinary language philosophy is based strictly on Wittgenstein? how ever, most of what has appeared under that general heading since 1940 can be viewed as a development of what was expressed or implied in his teaching and writing between 1930 and 1951.76 The current ordinary language philosophers' view of the nature and functions of language can best be understood by examining four related assumptions that are developed from Wittgenstein's preliminary conclusions: (1) Language is not a calculus? there is no mathematically "ideal" language. (2) Since language is not mathematically regular or precise, words do not function in any single way? specifically, 76As mentioned in the introduction, although the ordi nary language philosophers share a common view of the nature of language, there are significant differences in what they regard as the proper task of philosophy. Generally, those closer to Wittgenstein, including Wisdom, Anscombe, Malcom, Bouwsma, Ambrose, and others, stress the therapeutic poten tial of the analysis of ordinary language. That is, they believe recurrent problems of traditional philosophy can be solved with new insights into language. The other group, which includes Austin, Ryle, Strawson, Warnock, Urmson, Flew, and others, emphasizes the systematic study of the informal logic of ordinary language for its own sake— for what such study will teach us about language-using. (See Quinton, 176) . 122 meaning is not simply reference, and words are not simply names. (3) Not only are the above generalizations about language incorrect; virtually all generalizations are. (4) Language can most accurately and profitably be viewed as a form of ordinary human behavior; specific instances of language employment can best be seen as "language games," or "speech acts." Language Is Not a Calculus Ordinary language philosophers share the view that lan guage does not naturally and cannot accurately resemble a calculus, and that programs such as those of Russell, the logical positivists, and the symbolic logicians, which attempt to discover or create an ideal, mathematically pre cise language, are neither possible nor desirable. To Russell, the logical positivists, and the symbolic logi cians , the suggestion that ordinary language is ideal sounds absurd. After all, the main reason these philosophers orig inally turned to language analysis was that the inability of natural language to support a clear formulation of philo sophical problems necessitated the search for a mathemati cally precise way of stating them. Ideal language philoso phers hold that it is essential to construct a language that is an alternative to ordinary language, in order to clarify and systematize philosophy. Ordinary language philosophers maintain that it has not been ordinary language which has 123 saused the difficulties that have been experienced with tra il tional philosophical problems, but the positivists' and others' use of ordinary language. Ideal language philoso phers, they argue, have not used natural language; they have formulated problems in what looks like ordinary language, out have in fact misused the language, for example, by using terms jargonistically while relying on the ordinary connota tions of these terms. As Rorty summarizes, If Ordinary language Philosophy had an explicit program (which it does not), it might run something like this: we shall show that any argument designed to demonstrate that common sense (or the conjunction of common sense and sci ence) produces problems which it cannot answer by itself (and which therefore must be answered by philosophers, if by anyone), is an argument which uses terms in unusual ways. If philosophers would use words as the plain man uses them, they would not be able to raise such prob lems . 7 7 In the Blue Book Wittgenstein severely criticizes the movement toward an ideal language that had occupied Locke, Heinong, Frege, Russell, and the author of the Tractatus. ■ r' At one point he~writes, We . . . constantly compare language with a calculus pro ceeding according to ,exact rules. This is a very one-sided way of looking at language. In practice we very rarely use language as such a calcu lus. For not only do we not think of the rules of usage— of definitions, etc.— while using language, but when we are asked to give such rules, in most cases we aren't able to do so. We are unable clearly to circumscribe the con cepts we use; not because we don't know their real defini tion, but because there is no real "definition" to them. To suppose that there must be would be like supposing that whenever children play with a ball they play a game 77Rorty, p. 12. 124 according to strict rules.78 Two of the earliest and most influential articles fol lowing the new approach, John Wisdom's "Philosophical Per plexity," and Gilbert Ryle's "Systematically Misleading Expressions,"79 are founded on the rejection of the idea that language resembles a calculus. In a paper called "The leaning of a Word," J. L. Austin expresses the rejection this way: . . . Our new working-model, the supposed "ideal" lan guage, is in many ways a most inadequate model of any actual language: its careful separation of syntactics from semantics, its lists of explicitly formulated rules and conventions, and its careful delimitation of their spheres of operation— all are misleading. An actual lan guage has few, if any, explicit conventions, no sharp lim its to the spheres of operation of rules, no rigid separa tion of what is syntactical and what is semantical.80 Gilbert Ryle echoes Austin's view when he attacks for mal logicians who assume that "the logic of every-day state ments and even the logic of the statements of scientists, lawyers, historians and bridge players" can be adequately represented by the formulas of formal logic.81 He admits that the so-called logical constants do have "their sched uled logical powers." However, "the non-formal expressions 78Wittgenstein, The Blue and Brown Books, p. 25. 79Proceedings of the Aristotelian Society, n.s., XXXII (1931-1932), 139-176. ” " Philosophical Papers, ed. J. 0. Urmson and G. J. Warnock (Oxford, 1961), p. 35. 81"Ordinary Language" in Caton, pp. 108-127. 125 both of everyday discourse and of technical discourse have their own unscheduled logical powers, and these are not reducible without remainder to those of the carefully wired marionettes of formal logic."8 2 In an exhaustive treatment of "Linguistic Analysis," Anthony Quinton summarizes the opposing views in the follow ing words: Where formalists regard modern formal logic as the essence of language, as the wholly rational skeleton that is obscured by the historically and practically determined surface complexities of language, the philosophers of lan guage see it as a distortion or caricature of the rules of language as it is, a distortion produced by demands for systematicness and extensionality, the gravitational pull of a mathematical ideal. Language, say the formalists, ought ideally to be a calculus. It is actually, and fortunately, nothing of the sort, reply their opponents.83 In other words, "neither Aristotelian nor Russellian rules give the exact logic of any expression of ordinary language; for ordinary language has no exact logic."8l t Meaning Is Not Naming Ordinary language philosophers also agree that since language is not mathematically regular, words do not func tion in any single way; specifically, meaning is not simply reference, and words are not simply names. Their character istic view of meaning is most comprehensively expressed in 82Ryle, "Ordinary Language," p. 125. 83Quinton, p. 180. ^P. F. Strawson, "On Referring," Mind, LIX (July 1950), 320-344. 126 J. L. Austin's essay, "The Meaning of a Word."85 He begins with an important caveat: "What alone has meaning is a sen tence." To say a word or phrase "has a meaning" is to say that "there are sentences in which it occurs which 'have meanings': and to know the meaning which the word or phrase has, is to know the meanings of sentences in which it occurs."8 6 Austin admits that people quite commonly discuss ques tions like "What is-the meaning of the word 'pikestaff', 'handsaw', etc." They answer such questions in various ways, perhaps talking of the word's syntactics, i.e., how it is used in the English language; or sometimes "demonstrating the semantics" of the word by imagining or exploring situa tions in which it would naturally occur. However, they encounter difficulties when they move from questions like the above to the general question, "What is the meaning of a word?" That question actually asks, "What is the meaning of a- word-in-general?" or "of any word?" Austin comments, "Now if we pause even for a moment to reflect, this is a per fectly absurd question to be trying to ask."87 One can only answer questions like that when they are about some partic- 8 5Philosophicai Papers (London, 1961). 86Austin, "The Meaning of a Word," p. 24. Cf. note 80, "An expression has meaning only in the stream of life." 87Austin, "The Meaning of a Word, p. 26. 127 alar word. That form of the question is an example of what Austin calls "the fallacy of asking about 'Nothing-in-par- ticular1 which is a practice decried by the plain man, but by the philosopher called 'generalizing' and regarded with some complacency" (p. 26). There are many instances of this fallacy in philosophy, e.g., asking "What is reality?" or "What is truth?" The logical extension of this program is to ask, "What is anything?" In the past when philosophers have been misled into asking questions like these and have found no suitable answers, they have created all sorts of entities out of which they construct responses, entities like "ideas," "images," "classes of similar sensa," "sense- data," etc., "all of which are equally spurious answers to a pseudo-question" (p. 27). We are led astray, Austin suggests, by two related mis takes. First, we have difficulty avoiding the assumption that all words are names. Second, after we have analyzed a sentence like "The State owns this land," we feel inclined to ask of our analysis, "What in it is 'The State?'" Or, after having described a university— its campus, buildings, student body, goals, administrative system, faculty, etc.— we sometimes want to ask, "What, in all that, is 'the uni versity,7'" We are led astray because we continually expect to discover meaning "in" some entity, i.e., "referent," fldesignatum," etc. Yet, "it is clear that there is no sim ple and handy appendage of a word called 'the meaning of 128 (the word) "x".'" (p. 30). Inherent in Austin's analysis of meaning are three interrelated contentions that characterize the ordinary lan guage approach to meaning: (1) meaning is not reference or referring; (2) words are not names; (3) meaning is use. The attack on the assumption that meaning is reference proceeds in this manner: Traditional philosophers argue that a word is like an ostensive definition, i.e., in order to teach a child the meaning of the word "pencil," one points to a pen cil and says the word. However, as Wittgenstein asks, how i is it that the child knows precisely what one is pointing to or defining? That is, how does he know specifically to what "pencil" refers— the size, shape, eraser, wood, or color? The answer is, the child must learn that the word "pencil" is not used as a size-word, shape-word, etc. In order to learn the meaning of "pencil" he needs to know more than what object it names. Hallett puts it this way: . . . when we learn the meaning of a name we learn not only that it is a name for this, but that it is a name. We learn not only what it refers to but that it refers. We learn not only that it refers to this, but that it refers. We learn not only that it may refer to this indi vidual thing but that it may refer to other things of the same sort (or that it may not).8 8 He concludes that "meaning therefore, is not a thing referred to but the use of the word to refer to such a 88Hallett, p. 91. Cf. Gilbert Ryle, rev. of Meaning and Necessity. Philosophy, XXIV (1949), 69-71.______________ 129 thing."09 Similarly, P. F. Strawson, in his evaluation of Russell's theory of descriptions,90 argues that since the same statement may be made at different times, its truth or falsity is dependent not on its referents but on its use— whether or not it is used to make a true or false assertion. "'Mentioning,' or 'referring', is not something an expres sion does? it is something that someone can use an expres sion to do" (p. 326). The second ordinary language contention, that words are not names, follows from the view that words do not merely refer. Gilbert Ryle christened the traditional view the "Fido-Fido fallacy."91 The fallacy consists in the traditional belief that to ask What does the expres sion 'E' mean? is to ask, To what does 'E' stand in the relation in which 'Fido' stands to Fido? [that is] The significance of any expression is the thing, process, per son or entity of which the expression is the proper name, (p. 69) Ryle has devoted much of his writing to the destruction of .this point of view. He has made much of the point Austin mentions, that, when philosophers cannot find an obvious referent for troublesome words to name, they create various extra-linguistic correlates. Problems arise when utterances like "he took a stick" become the model for "he took a walk," "a nap," or "his time," and philosophers busy them- 89Hallett, p. 71. 90"On Referring." 91Ryle, rev. of Meaning and,Necessity. 130 selves creating entities that are "his walk," "his nap," and "his time." As a result, not only Fido and London, but also centaurs, round squares, the present King of France, the class of albino Cypriots, the first moment of time, and the non-existence of a first moment of time must all be credited with some sort of reality. They must be else we could not say true or false things of them.9 2 Ryle's influential article, "Systematically Misleading Expressions," is basically an explication of this viewpoint. By "misleading" he means expressions that are "couched in a syntactical form improper to the facts recorded and proper to facts of quite another logical form than the facts recorded."93 In other words, the sorts of statements he examines appear to name entities that do not exist. For instance, the second type of expression Ryle cites is the statement about universals, such as "Virtue has its own reward." Ryle's point,is that such statements lead the phi losopher (not the common man, who is much too sensible) to look for some thing that he can call "virtue." Ryle con cludes that all such statements are misleading, because "they all suggest the existence of new sorts of objects . . . they are all temptations to us to 'multiply entities'" (p. 165). J. L. Austin calls the same sort of mistake the 92Ryle, "The Theory of Meaning," p. 140. 93Ryle, "Systematically Misleading Expressions," p. 143; 131 "descriptive fallacy." He notes that "for too long" philos ophers assumed that all utterances were statements and that "the business of a 'statement' can only be to 'describe.'"91* The current questioning of that view arose, according to Austin, when the strict application of the principle of verification resulted in many apparent statements being labeled "pseudo-statements." That is, when the logical positivists began contending that all ethical, religious, poetic, and metaphysical statements are nonsense, philoso phers began wondering what would lead them to come to such an extraordinary conclusion. The resulting investigations (carried on chiefly by Austin himself) have uncovered other sorts of utterances that cannot be classified as "descrip tive statements." Austin's main point is that neither these other sorts of statements or the words in them name any thing. As Ryle summarizes, "learning the meaning of an expres sion is more like learning a piece of drill than like coming across a previously unencountered object. It is learning to operate correctly with an expression and with any other expression equivalent to it."95 Ryle, Strawson, and Austin all emphasize that the use of an expression is not some additional substance; it is not 9‘ 'Austin, How To Do Things With Words, p. 1. 95Ryle, "The Theory of Meaning," p. 146. 132 a non-physical or non-mental object. The reason it is not that sort of object is not that it is either a physical or a mental object, but that it is not an object of any sort. Thus, to the extent that philosophy is the study of meanings (or the clarification of meaning), its subject matter is linguistic, not psychological or Platonic. However, as Ryle points out, the kind of linguistic problems philosophy deals with are quite unlike any of the problems of philology, grammar, phonetics, rhetoric, prosody, etc., since they are prob lems about the logic of the functionings of expressions. Such problems are so widely different from e.g., philolog- ical problems, that speaking of them as linguistic prob lems is, at the moment, as Wittgenstein foresaw, mislead ing people as far in one direction as speaking of them as problems about Meanings or Concepts or Propositions had been misleading them in the other direction. The diffi culty is to steer between the Scylla of a Platonistic and the Charybdis of a lexicographical account of the business of philosophy and logic.9 e The use or employment of language can thus be viewed as "the [informal) logic of the functionings of expressions." To study meaning is to study that informal logic. How the ordinary language philosophers study the informal logic of the functionings of expressions will be described shortly. But first, one final assumption about what one cannot do when studying language must be explained. 96Ryle, "The Theory of Meaning," p. 152. Italics added; m generalizations about Language axe Not valid Ordinary language philosophers maintain that not only are the generalizations about referring and naming incor rect, but virtually all generalizations about language unnecessarily distort its true nature. Wittgenstein argues that attempts to arrive at generalizations about language or the "essence" of a word's meaning can never be wholly suc cessful, because actual language usage has no sharp bounda ries. He states, The idea that in order to get clear about the meaning of a general term one had to find the common element in all its applications has shackled philosophical investigation; for it has not only led to no result, but also made the phi losopher dismiss as irrelevant the concrete cases, which alone could have helped him to understand the usage of the general term.97 To illustrate his point, Wittgenstein considers the case of "the meaning" of "expecting." What is it, he asks, to "expect someone from 4:00 to 4:30"? He notes that the term does not refer to one process or state of mind going on throughout that interval but to a great many different activities and states of mind. For instance, if the person I am expecting is coming to tea, what happens may be this: At four o'clock I look at my diary and see the ndme "B" against to-day's date? I prepare tea for two; I think for a moment "does B smoke?" and put out cigarettes; towards 4.30, I begin to feel impatient; I imagine B as he will look when he comes into my room. All this is called "expecting B from 4 to 4.30." (p. 20) 97Wittgenstein, The Blue and Brown Books, pp. 19-20. 134 since there is no single "similarity" among the myriad of different states of mind and physical processes that make up bhis particular instance of "expecting," there is even less zhance that different cases of "expecting" could be "simi lar" in any consistent way. There are, however, common features that some aspects of some cases of "expecting" share and others do not. The nature of these common fea tures led Wittgenstein to call them "family resemblances." He explains his concept of "family resemblances" in this way: Consider for example the proceedings that we call "games". I mean board-games, card-games, ball-games, Olympic games, and so on. What is common to them all?— Don't say: "There must be something common, or they would not be called ' games'1 1 — but look and see whether there is any thing common to all. — For if you look at them you will not see something that is common to all, but similarities, relationships, and a whole series of them at that. To repeat: don't think, but lookl— Look for example at board-games, with their multifarious relationships. Now pass to card-games; here you find many correspondences with the first group, but many common features drop out, and others appear. When we pass next to ball-games, much that is common is retained, but much is lost. — Are they all 'amusing'? Compare chess with noughts and crosses. Or is there always winning and losing, or competition between players? Think of patience. In ball games there is winning and losing; but when a child throws his ball at the wall and catches it again, this feature has disap peared. Look at the parts played by skill and luck; and at the difference between skill in chess and skill in tennis. Think now of games like ring-a-ring-a-roses; here is the element of amusement, but how many other character istic features have disappeared! And we can go through the many, many other groups of games in the same way; can see how similarities crop up and disappear. And the result of this examination is: we see a com plicated network of similarities overlapping and criss crossing: sometimes overall similarities, sometimes simi larities of detail. I can think of no better expression to characterize 135 these similarities than "family resemblances"? for the various resemblances between members of a family: build, features, colour of eyes, gait, temperament, etc. etc. overlap and criss-cross in the same way. — And I shall say: 'games' form a family.98 Following Wittgenstein's advice, J. L. Austin deline ates seven of "the more obvious cases where the reasons for 'calling different sorts of things by the same name' are not to be dismissed lightly as 'similarity.'"99 He asks what is similar among the expressions "healthy body," "healthy com plexion," and "healthy exercise." He also considers the concept of "pleasure," instances of which "not merely resemble each other in being pleasant, but also differ pre cisely in the way in which they are pleasant" (p. 41). In another case Austin deals with words like "youth," "love," and "truth," which may in one instance be interpreted as the name of a substance, in another a quality, and in a third a relation. Austin's final example is the sense in which one talks of a cricket bat, a cricket ball, and a cricket umpire. Each is called by the same name apparently because each has its own special part to play in the activity called cricketing. But one cannot say that cricket simply means "used in cricket," because "used in cricket" cannot be explained except by explaining the special part each plays in cricketing (p. 41). Whereas Wittgenstein's aim is. to 98Wittgenstein, Investigations, para. 66-67. "Austin, "The Meaning of a Word," p. 39. 136 warn philosophers against generalizing, Austin wants to show how important it is "to have a thorough knowledge of the different reasons for which we call different things by the same name, before we can embark confidently on an enquiry" into "the meaning" of some general term (p. 42). Wittgenstein also rejects the idea that communication involves someone else grasping the sense of one's words— the notion that all communication is similar in that the listen er or reader takes into his own mind, as it were, the thoughts or feelings the speaker or writer has while he is communicating. As he points out, one cannot even begin to discover for himself any consistent or common experiences that could be shared. He suggests, "Ask yourself: 'When I said, "Give me an apple and a pear and leave the room," had I the same feeling when I pronounced the two words "and"?'"100 As Hallett comments, "When you have such diffi culties knowing what you yourself experienced, how can you possibly discover what another person experienced? Under standing certainly does not consist in conjectures about another person's feelings or personal experiences."101 On the contrary, "understanding," like "meaning," does not consist of anything that accompanies some utterances and does not accompany others; it is rather a function of the 100Wittgenstein, The Blue and Brown Books, p. 79. 10hallett, p. 65. ----------------------------------------------------------------- 137 use of expressions. When one says he "understands" an expression, he does not mean that he possesses some set of ideas, experiences, feelings, etc., that replicate what the speaker or writer possessed when he composed or uttered the expression. Rather, as Wittgenstein points out, "we refer oy the phrase, 'understanding a word' not necessarily to that which happens while we are saying or hearing [an ex pression] , but to the whole environment of the event of saying it."102 Speaking with understanding differs from speaking without understanding, but not in that the former is accompanied by something which is lacking in the second sase. "Understand" is one of a large class of terms like "hope" and "expect" which refer not to any single event, but to "a pattern which recurs, with different variations, in the weave of our life."103 L artaU a g e-using Is Ordinary Behavior The use of language, in short, is not as the tradi tional philosophers represented it. There is no one pattern f to be revealed, no single account to be offered, no small set of definite, calculus-like rules. On the contrary, the forms and uses of language are inexhaustibly flexible and various; speaking is not like a game, but a whole family of 102Wittgenstein, The Blue and Brown Books, p. 157. 10 3Wittgenstein, Investigations, para. 174. 138 games, and the rules for these games, their purposes, and the methods of play are almost endlessly diverse. In addi tion, speaking is not unique or strange; it is one of the ordinary things we do— "as much a part of our natural history as walking, eating, drinking, playing."10 * J. L. Austin was responsible for developing the insight that speaking is behavior, not just "symbolic" behavior. As previously noted, Austin christened the logical positivist view of language the "descriptive fallacy," because it fal laciously assumed that statements served only to describe objects or report events, and thus were all necessarily either true or false. Austin pointed out that, especially with utterances like "I do" in a marriage ceremony, or "I christen thee" at a ship launching, the important question about the utterance is not what is described by it, but what is done. Since an action is being performed in saying such sentences, Austin called them "performatives"; "To issue such an utterance is to perform the action— an action, per haps, which one scarcely could perform, at least with so much precision, in any other way."105 Austin attempted to clarify his concept of performa tives first by exploring ways to criticize them. He noted that, since performatives are neither true or false, some 10 “ 'Wittgenstein, Investigations, para. 25. 105J. L. Austin, "Performative-Constative," trans. G. J. Warnock, Caton, p. 22. 139 ather criterion must be found to determine their effective ness. Since they are acts, it should be appropriate to criticize them in terms of whether or not they "fail to come aff in special ways." He wrote of these "failures to come aff" as "infelicities" or in terms of the utterances being "unhappy."106 He initially classified infelicities into three types: "inappropriatenesses," "insincerities," and "breaches of commitment." The utterance, "I divorce youl" tfould be an inappropriate attempt at a performative act in vestern Christian culture, because there is no appropriately applicable custom. If one said to a friend, "I congratulate you," and did not mean it, his performative-attempt could be judged insincere. And if one said, "I promise to . . . ." and failed to live up to his promise, the utterance could be described as failing because of a breach of commitment. In addition Austin investigated methods of identifying performative utterances by distinguishing them from other types of speech acts. Grammatical criteria did not prove satisfactory: although the first person indicative active, "I promise," is quite clearly performative, and the second ar third person passive, "Passengers are hereby warned that . . . ," is also often performative, the distinctions are not always clear. For instance, "DogI" may take the place af the entire performative, "I hereby warn you that a dog is x0SAustin, "Performative Utterances," Papers, p. 224. 140 about to . . . Consequently Austin rejected all gram matical classifications. He noted, We often find cases in which there is an obvious pure per formative utterance and obvious other utterances connected with it which are not performative but descriptive, but on the other hand a good many in between where we're not quite sure which they are. On some occasions . . . they seem positively to revel in ambiguity. (p. 234) His attempts to identify performatives by virtue of their lack of conventional truth or falsity were also largely unsuccessful. Although most explicit performatives could not be said to be either true or false, there were some exceptions. For instance, when a baseball umpire says "Outl" the expression is performative in an important sense; nonetheless, it seems that his utterance does have something like the duty to be true or false to the situation. In addition, utterances like the philosophically interesting, "The cat is on the mat but I don't believe it is," are not performative in the sense that "I do" is, but they can none theless most effectively be criticized in terms of their infelicity or unhappiness, rather than in terms of any claims to truth or falsity. Further complicating his efforts to distinguish performatives, Austin recognized that most descriptive statements can also be viewed as performa tive. We see then that stating something is performing an act just as much as is giving an order or giving a warning; and we see, on the other hand, that, when we give an order or a warning or a piece of advice, there is a question about how this is related to fact which is not perhaps so very different from the kind of question that arises when we discuss how a statement is related to fact. (p. 238) 141 As he worked through his performative-constative dis tinction, Austin came to realize that his analysis was undoubtedly partial, i.e., that there might be many more than just two kinds of speech acts. However, he also recog nized that the approach he had pioneered held promise for all language scholars. He commented, . . . setting out from a pretty vague distinction between the straightforward utterance (stating) and the act (or dering) , we meet, as we go along, a number of difficulties which lead us appreciably to modify our original analysis, and not to go on seeing inside language just two types of "speech acts": and this leads us to reconsider in its entirety our conception of language, which may emergefrom the test a good deal the worse for wear, without our yet being in a position to formulate a theory embracing every kind of "speech act."107 Austin's book, How to Do Things with Words., is devoted to detailing his conception of speech acts. In it he alters some aspects of his analysis: infelicities are divided into "misfires," "flaws," and "hitches"; performatives are separated into "explicit"— first person indicative active and obvious to most observers— and "primitive"— performative only by virtue of the situation in which they are uttered. In a fresh effort at clarifying his classification he introduces three new terms. Statements are called "locu- tionary acts"; utterances in which an action was performed in saying something are termed "illocutionary acts"; and utterances in which an action was performed b£ saying some 107Austin, "Performative-Constative," p. 34. Italics added. 142 thing are called "perlocutionary acts." If Austin's division is viewed as classifying different types of utterances, the following are examples of each speech act: the statement "Los Angeles is in California" would be considered a locutionary act. The most important questions about it relate to its meaning and its truth or falsity. The request, "May I borrow your car?" could be considered an illocutionary act. In making the utterance, the speaker performed the act of requesting; his utterance could most effectively be criticized in terms of its force, i.e., whether it was an appropriate, efficient, clear, etc., request. The act of persuading a man to vote for X b£ say ing, "Y is corrupt" could be termed a perlocutionary act. Perlocutionary acts are most appropriately judged in terms of the effect they produce. In other words, illocutionary acts may be assessed regardless of their effect; their force is at least as important as their meaning. Perlocutionary acts are those that produce a specific effect. For instance, while informing and ordering could be called illo- outionary acts, persuading, deterring, and misleading, each of which depends for its success on the production of a certain effect, are perlocutionary acts. Whereas a locu- iionary act has meaning, an illocutionary act has force, and a perlocutionary act achieves an effect. Unfortunately, Austin also experienced problems devel oping this tripartite division. His conclusion to How to Do ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- J - f J rhings with Words was similar to that cited above. He main tained that the philosopher's attention should be focused on "the total speech act in the total speech situation" (p. 147) . Even though we may not yet be able to formulate a theory embracing every kind of speech act, the initial insight that speaking involves a wide variety of human actions— not just symbolizing and not just symbolic actions — leads us, as Austin suggested, to "reconsider in its en tirety our conception of language." Ordinary Language Method The main interest of this chapter is the ordinary lan guage philosophers' view of the nature of language and mean ing, not how they employ these insights about language as they "do philosophy." However, a brief treatment of the method of ordinary language philosophy will reveal some of the relationships among the four concepts discussed above, and clarify some terms used in, and corollaries of, the main assumptions cited.100 The ordinary language philosophers' common starting point is the assumption that language can most accurately and profitably be viewed as a form of ordinary human behav ior, and that specific instances of language employment can aest be seen as "language-games," or "speech-acts." The loeAgain, the assumption that there is one method is invalid. However, there is sufficient family resemblance to warrant a brief discussion. 144 subject of ordinary language investigations could be called "the informal logic of the functionings of expressions."109 Two concepts essential to their subject need clarification before ordinary language methodology is discussed: (1) what is meant by "ordinary" in phrases like "ordinary language" or "the use of ordinary language?" and (2) what is meant by "informal logic?" Austin provides one of the clearest justifications for concentrating on ordinary language. He points out that in doing philosophy, as in living, words are our tools, and "as a minimum, we should use clean tools: we should know what we mean and what we do not, and we must forearm ourselves against the traps language sets us."110 In addition we need bo be able to examine words in various real and contrived situations, in order to "realize their inadequacies and arbitrariness" (p. 130). Finally, Austin notes, and more hopefully, our common stock of words embodies all the distinctions men have found worth drawing, and the c connexions they have found worth making, in the lifetimes of many generations: these surely are likely to be more numerous, more sound, since they have stood up to the long test of the survival of the fittest, and more subtle, at least in all ordinary and reasonably practical matters, than any that you or I are likely to think up in our arm chairs of an afternoon— the most favoured alternative method. (p. 130) En a later series of lectures Austin adds, "the distinctions smbodied in our vast and, for the most part, relatively 10 ^See note 96. 1 Austin, "A Plea for Excuses," Papers, p. 129. ___ 145 ancient stock of ordinary words are neither few nor always yery obvious, and almost never just arbitrary. . . ,"lll: Ryle suggests that "ordinary" can be understood by con trasting it.with terms like "esoteric," "technical," or 3ometimes "archaic." It means "common," "current," "collo quial," "vernacular," "natural," "prosaic," and "on the tongue of Everyman," and is used in contrast with expres sions or languages that only a few people know how to use, s.g., the technical terms or symbolisms of lawyers, theolo gians, economists, philosophers, cartographers, and symbolic logicians. Ryle admits that "the edges of 'ordinary' are slurred," but maintains that "usually we are in no doubt whether a diction does or does not belong to ordinary par lance";112 consider, for instance, "isotope" versus "bread," or "transfinite cardinal" versus "eleven." In another sense, Ryle continues, "Ordinary" contrasts with "non-stock" or "non-standard." "Non-stock" uses of words include meta phorical, hyperbolical, poetical, and deliberately restricted uses. He comments that, although some people, especially critical philosophers, take pains to forget what a stock use of a word is, they have no trouble when they are teaching children or foreigners how to use it, or consulting 11lJ. L. Austin, Sense and Sensibilia, reconstructed from the manuscript notes by G. J. Warnock (New York, 1964), p. 63. 112Ryle, "Ordinary Language," p. 109. 146 dictionaries. Philosophers study the ordinary uses of language secause some of the knottiest of perennial philosophical problems are created by the existence of "logical tangles" in common thought and discourse, e.g., in the concepts of sause, evidence, knowledge, mistake, ought, can, know, etc. Ryle notes that the emphasis in "ordinary use of language" should be on use, in order to help bring out "the important £act that the enquiry is an enquiry not into the other fea tures or properties of the word or coin or pair of boots, aut only into what is done with it. . ." (p. 113). Ryle also differentiates between "use" and "usage." "A usage is a custom, practice, fashion or vogue .... There cannot be a misusage any more than there can be a mis- custom or a misvogue. The methods of discovering linguistic usages are the methods of philologists" (p. 116). In an earlier article, Baier treated this differentiation as "the ordinary use" versus "an ordinary use." Baier pointed out that statistics can pinpoint "the" ordinary use (or the "usage"), while "an" ordinary use is one of the uses that isn't out-of-the-way or extraordinary.113 Any native speaker or consistent user of a language can determine what ordinary use is in almost any given case; no special expertise or empirical sophistication is needed. 113K, Baier, "The Ordinary Use of Words," Proceedings of the Aristotelian Society, n.s., LII (1951-1952), 50. 147 Baier maintained, "Philosophers and also the common man are in as good a position for knowing the ordinary use of words as the philologists and sociologists. . . . We could not speak of one language if there were not a stock of common words, each with one use or an established set of uses.1,1 As Ryle puts it, objects like tennis rackets may be wielded with greater or less skill, but coins, stamps, buttons, shoelaces, and separate words offer no scope for talent. "Either a person knows or he does not know how to use and how not to misuse them."115 Prose and argument can be better or worse, but the best lawyer or philologist does not know the use of "rabbit" or "and" any better than the man on the street. In sum, if I know the "ordinary use" or meaning of a word or phrase, I know something like a body of unwritten rules, or some thing like an unwritten code or general recipe. I have learned to use the word correctly in an unlimited variety of different settings. . . . I have learned to put it to its work anywhen and anywhere, if there is work for it to do. 116 The nature of "informal logic" is as crucial to the ordinary language approach as the nature of "ordinary use." These philosophers reject the application of formal logic to ll**Baier, p. 52. ll6Ryle, "Ordinary Language," p. 117. ll6Ryle, "Ordinary Language," p. 120. r?8 language, because they do not believe language resembles a calculus. As already noted, Wittgenstein maintains that the philosopher must view language in terms of family resem blances instead of a structure of well-defined classes and sub-classes. He must reject his "preconceived idea of crys talline purity" and examine language as it is used in spe cific situations.117 P. F. Strawson, called the "leading figure" in the battle between informalists and formalists,110 notes that he has no objection to the construction of formal systems, because they are useful in appraising "context-free dis course," as exemplified in mathematics or physics. However, he argues, formal logic needs to be supplemented by a logic of everyday discourse, for it is incapable of coping with the complexities of ordinary speech. For example, "the for mal logician cannot deal effectively with arguments which depend on temporal relationships or are otherwise 'tied' to specific places and times."1; 10 Strawson suggests an "ordi nary language logic" could begin by asking such questions as "What are the conditions under which we use such-and-such an expression or class of expressions?" He admits it would not be as elegant or systematic as formal logic, but argues it can still "provide a field of intellectual activity unsur- 117Wittgenstein, Investigations, para. 108. 118Passmore, p. 470. ll9Passmore, p. 472. 149 passed in richness, complexity and the power to absorb.".120 Gilbert Ryle discusses relationships between formal and informal logic at length in his collection of lectures called Dilemmas. 12 1 Ryle suggests that the two kinds of logic figured prominently in a "boundary dispute" "which has fairly recently broken out between certain philosophers and certain philosophically-minded logicians" (p. 111). He notes that Aristotle established the foundations for both kinds of thinking; formal logic originated with the Aris totelian theory of the syllogism, and was developed along lines resembling Euclidean geometry. However, Aristotle's development of the theories of pleasure, perception, and moral responsibility, and his debates with Plato and Socra tes were different: the debate does not take the shape of a chain of theorems, nor do the arguments used in that debate admit of nota- tional codification. Whether a given philosophical argu ment is valid or fallacious is, in general, itself a debatable question. (pp. 111-112) However, Ryle emphasizes, though formal logic evolved into a system quite different from the informal logic employed by most philosophers, "the operations characteristic of Formal Logic exercise a detectable, if minor, control over the operations characteristic of philosophy" (p. 112). Ryle views the relationship between formal and informal 120Passmore, p. 472. Passmore based his account on Strawson's Introduction to Logical Theory (London, 1952). . 12 Cambridge, 1966. 150 logic in terms of an analogy between a soldier fighting on a battlefield and drilling on a parade-ground. Although the tforst thing a soldier could do in a battle situation would be to operate as if he were on the parade-ground, his responses under fire are related to his parade-ground train ing. Specifically, "it is not the stereotyped motions of drill, but its standards of perfection of control which are transmitted from the parade-ground to the battlefield" (p. 112) . This relationship is seldom recognized by either phi losophers or logicians. Aristotelian formal or symbolic logic, supplemented by the additions of Megarian and Stoic logicians, has grown into a well-developed, independent sci ence. As a result, logicians are now beginning to tell phi losophers that they should stop trying to solve problems by their ineffectual exercises in improvisation, trial-and- error, and armchair speculation, and should use the precise, scientific methods of logic. Philosophers, on the other hand, are apt to reply that no problem of any philosophical importance could possibly be reduced to fit the logician's machinery without being distorted beyond recognition. "... questions which can be decided by calculation are different, toto caelo different, from the problems that per plex" (p. 114). Ryle points out that neither party is entirely right in this dispute, but that "both are more nearly right than the appeasers who try to blend the opera- ----------------------; ------------------------------------ 33X tions of the one party with the operations of the other" (p. 114). Formal logicians have concentrated their investigation on ranges of inference that involve notions of "all," "some," "not," "and," "or," "if," etc. The rules they have discovered apply regardless of concrete subject matter, i.e., "all" and "some" can be followed by "cows," "men," or "gods" and any logical inferences that may properly be drawn remain unchanged. The success formal logicians have had in codifying certain crucial inference patterns involving these topic-neutral expressions has led others to suppose that any piece of valid reasoning can be reduced somehow to one of the scheduled formal logical patterns, and, conversely, that every fallacious piece of reasoning can be reduced to one of the categorized logical howlers. On this view, formal logic can be neatly distinguished from philosophy; it maps the inference-powers of logical constants while philosophy con centrates on subject-matter concepts. However, Ryle points out, that distinction quickly breaks down. First, topic-neutrality alone cannot qualify an expression as a logical constant. There are many expres sions in English and other European languages that can qualify as being topic-neutral, but not as being logical constants. For instance, the concepts elucidated by each of the following terms is the same, regardless of its use in any given expression: "several," "most," "few," "three," 152 "half," "although," "because," "perhaps," "may," "after," "while," "before," and "hosts of other conjunctions, parti cles, prepositions, pronouns, adverbs, etc." (p. 116). Thus formal logicians quite properly limit their attention to only some of the many topic-neutral expressions. In addi tion, the formal logician does not actually deal with the "and," "not," and "some" familiar to most civilian speakers. For instance, "and" in ordinary language can carry a tempo ral notion, i.e., if one believes that "she took arsenic and fell ill," he will reject the rumor that "she fell ill and took arsenic." Yet, as used by the formal logician, the force of "and" is such that the latter sentence is regarded as an absolute paraphrase of the former. Similarly, the formal logician's use of "if" is much more limited than that of the man on the street. In brief, formal logic operates only with some topic-neutral expressions, and only with some senses of those expressions. With a clear conception of the nature of formal logic, one can understand in what sense both the logician's and the philosopher's inquiries can appropriately be termed "logi cal," i.e., one can understand the nature of "informal logic." Even though the logician and the philosopher are interested in different problems, and even though neither is improperly doing what the other is doing properly, both the philosopher and the formal logician operate according to logical considerations, just as both the battlefield soldier 153 and the parade-ground soldier operate according to military considerations. Ryle explains this point with a detailed analogy that is simpler to describe than reproduce. He suggests that one can classify economic goods used in commerce into (1) abso lutely commodity-neutral coins and other legal tender; (2) virtually commodity-neutral baskets, sacks, brown paper, 3tring, etc; and (3) all other goods, which may be used for carter and trading, the value of which depends on the nature cf the goods and demand for them, not on "the terms of their cfficial charters." Ryle suggests that, in linguistic dealings, we use a variety of topic-neutral words, some of which function like coins and currency notes and some of which function rather like baskets, string, etc. Societal pressure for such topic-neutral expressions increases as the need increases for universally acceptable and understandable linguistic decisions. That is, when lawbreakers have to be tried, contracts have to be entered into and enforced, wit nesses have to be cross-examined, laws have to be drafted, and when theorists have to consider in detail the strengths and weaknesses of their own and one another's theories, then topic-neutral expressions are needed. Ryle continues, The conscript expressions actually used by Formal Logi cians, together with the methodically designed expressions of mathematics, correspond in many respects with a legal tender. A sentence with one or more "logical words" in it, is a sentence with one or more price-tickets on it. Other topic-neutral words, inflections, etc., correspond more closely with the paper, string, sacks and pitchers 154 with which we go to and return from the market. (p. 121) But most of the terms of everyday and even technical discourse are not topic-neutral; "their barter-values are not stamped upon their faces." They can be used in reason ing, but there is no way to reduce them into logical con stants, or to construct a logical system from them. Such terms have their logical powers, but their force cannot be read off some official charter, because there is no charter. Their logical powers can only be extracted by examining the dealings people transact with them. Thus formal logic is to the informal logician or phi losopher what Euclidean geometry is to the cartographer. In Dther words, the cartographer finds no Euclidean straight hedgerows or Euclidean plane meadows. Yet he could not map the sinuous hedgerows that he finds or the undulating meadows save against the ideally regular boundaries and levels in terms of which alone he can calculate out the relative positions and heights of the natural objects which he is to record from the visual observations that he makes. (p. 123) 3ut, just as cartography cannot be reduced to geometry or fighting to drill or trading to the balancing of accounts, so the handling of philosophical problems cannot be reduced bo the derivation or application of theorems about logical sonstants. The philosopher is perforce doing what might be called "informal Logic," and the suggestion that his problems, his results or his procedures should or could be formal ized is as wildly astray as would be the corresponding suggestions about the soldier, the cartographer, and the trader. (p. 124) In short, as Ryle points out, all discourse is not and can not be reduced to formal logical discourse. Ordinary speech, like other human activities, does not consistently conform to a detailed system of immutable laws. Its logic is informal, and when we study ordinary speech we study the "informal logic of the functionings of expressions." Based on this understanding of "ordinary use" and "informal logic," the characteristic ordinary language approach is to view uses of language as forms of human behavior that constitute one or a family of "speech acts" which explicate some concept. Their specific method has been simplified and explained in John Wilson's Thinking with Concepts.122 It basically involves three steps. First, the philosopher focuses on some philosophically interesting con cept or language use. His interest may be aroused because the concept has confounded other philosophers, or because, in his judgment, it needs clarification on general princi ples.129 In the former case, he may choose to examine, as Ryle has ably done, the concept of "mind."121' In the lat ter, he may focus, as Austin has, on the nature of state- 12Cambridge, 1966. 123See note 76. 12HThe Concept of Mind (New York, 1949). 156 ments in language versus utterances that are not state ments . 12 6 Second, he examines, in as much detail as possible, instances of the use of expressions relevant to the concept le is studying. Ryle examined how we talk of our own mind and of the minds of others; he examined expressions about thought and thinking and perception that are used in ordi- ciary speech. He traced the history of statements about minds, dissecting the psychological terminology that has grown up around the concept of mind. Wilson suggests that this step might also include a consideration of contrary cases, borderline cases, invented cases, and the practical results or results in the language of different views of the concept.i26 The philosopher is thus engaged in "analysis of the informal logic" of the expressions that define the concept. His third step is to indicate what important insights le has gained from his analysis, i.e., what perplexing phi losophical questions have been solved— or "dissolved"— or what previously undiscovered nuances of meaning have been found in ordinary use. The ordinary language philosopher considers only exam- 126See, e.g., Austin, "Performative-Constative"; "Per formative Utterances," in Papers, pp. 220-239; and How to Do Things with Words. . l26Wilson, pp. 23-29. 157 ?les of expressions as they normally appear in everyday dis course. He assiduously avoids removing expressions like "ought," "truth," or "mind" from the uses in which they ordinarily occur. The uses— including specific instances of use, plus the conventions, habits, rules, and traditions that determine ordinary employment— are the language-games or speech-acts.127 Summary This chapter identified the ordinary language philoso phers and outlined their views of the nature of language and meaning. In part one philosophical concern with language was traced from pre-Aristotelian Greece to the logical posi tivists. Aristotle offered the first comprehensive analysis of language, based on the assumptions that language is a 12’ Morris Weitz briefly describes "the procedure of these philosophers" in the following way: "There are certain ordinary ways of employing certain ordinary expressions. Traditional philosophers have offered certain descriptions of those modes of employment and have either accepted them or, because of difficulties, recom mended new uses. This has obviously been the case with the standard uses of "the" followed by a singular noun, and with "see," "know," and the like. Now, what these Oxford philos ophers say is that insufficient attention has been paid to the logical characteristics of these ordinary ways of using these ordinary expressions. Look again to these logical characteristics. This is all their appeal to ordinary lan guage amounts to. . . . the role of ordinary language in the whole of philosophical activity, they hold, is to provide the setting of the logical elucidation of certain ordinary expressions; and it becomes the crucial testing ground of any proposed elucidation, including their own. (Weitz, , l Oxford Philosophy," p. 229). 158 3ystem of symbols and words are fundamentally names. With out questioning the fundamental Aristotelian assumptions, Locke reemphasized the theory that words represent ideas. 3radley turned British philosophy from the study of Lockean Ideas to the study of language. Frege emphasized the role of analysis in the creation of an ideal language. Russell sophisticated formal linguistic analysis and Moore intro duced an informal "common sense" approach to language and neaning. The logical positivists carried formal analysis to Its extreme, rejecting traditional metaphysics, and using reductive analysis to move toward an ideal language in which all words would function as verifiable names. Part two presented the ordinary language philosophers' sriews of language and meaning. "Wittgenstein II," the author of the Philosophical Investigations, influenced both nost of the ordinary language philosophers' questions and their method of searching for answers. Their new approach to language is based on a rejection of the theory that words are simply names, and is developed in four related basic assumptions: (1) language is not a calculus, but is almost iimitlessly complex. (2) Words are not simply names; mean ing is not simply reference. (3) Virtually all generaliza tions about language are inaccurate. (4) The use of lan guage is ordinary human behavior. Finally, an explanation of "ordinary language method" clarified what is meant by "ordinary use" and "informal logic." CHAPTER IV LANGUAGE AND MEANING: SPEECH BEHAVIOR AND USE The argument of this dissertation is that rhetoricians' views of the nature of language and the nature of meaning are unsound given the ordinary language philosophers' con clusions about language and meaning. Rhetoricians were led to invalid views because they relied on language scholars, mainly semanticists and general semanticists, who began their analyses by asking two incorrect first questions. Instead of "What is the nature of language?" the appropriate first question should have been "What is the nature of speech behavior?" And instead of "What is meaning?" they should have initially asked some form of the question, "How is speech behavior meaningful?" This argument is supported in part one of this chapter by a critical examination of the symbol metaphor and the three representational theories of meaning outlined in Chapter II, from the perspective of ordinary language phi losophy. Part two presents the contention that the most important insight one can gain from an ordinary language philosophy approach to language and meaning is that instead of focusing initially on language, the language scholar 159 160 should begin his inquiries with speech. He should do so for two reasons: (1) His first questions should be about the most fundamental aspect of his subject matter, and speech is more fundamental than language. (2) If he begins by study ing language systems instead of speech behavior, he is liable to make oversimplified generalizations about human communication. Speech behavior is defined as meaningful luman oral behavior governed by informal logical rules; lan guage is defined as an abstract system consisting of verbal codes and the informal logical rules of speech behavior. In addition, an alternate view of the nature of meaning is presented, based on the conclusions about speech behavior. Weaknesses in the Symbol Metaphor Rhetoricians agree that language is fundamentally a system of symbols. Three weaknesses in this position can be identified: (1) it is not consistently clear which lin guistic entities, i.e., spoken words, written words, phrases, utterances, etc., allegedly function as symbols. (2) The symbol metaphor cannot by itself account for the nature of non-symbolic terms. (3) When an attempt is made to account for such terms, the theory of abstraction is employed, and this theory is based on the incorrect assump tion that all words ultimately represent objects. Confusion Among Symbols Rhetoricians often were not consistent about which 161 elements of language were symbolic, i.e., whether written words, spoken words, sounds, phrases, sentences, utterances, or expressions functioned as symbols. At one point it was asserted that "sounds, gestures, writings, or even smoke signals" could be the language entities that "represent something else."1 In another case symbols were identified as "sound patterns . . . that is word or phrase."2 Not only does this imprecision make the symbol metaphor confusing, but it also calls attention to several of its weaknesses. For instance, Walter Coutu's account of meaning is weakened by his failure to distinguish between words and sentences. When Coutu divides signs into signals and sym bols, he notes that whether a sign is signalic or symbolic often depends on the context in which it is used. "Hickory nut" can be symbolic if it is used when no nuts are present. But if someone says, "Bring me a hickory nut" in the k presence of one or more hickory nuts, "the utterance is sig nalic. "3 However, Coutu does not clarify whether the word "nut," the noun and modifier, "hickory nut," or the request, "Bring me a hickory nut" is the linguistic entity that 1Milton Dickens, Speech: Dynamic Communication (New York, 1963), p. 371. 2Giles Wilkeson Gray and Waldo W. Braden, Public Speak ing: Principles and Practice, 2nd ed. (New York, 196>3), p. 456. . 3Walter Coutu, "An Operational Definition of Meaning," Quarterly Journal of Speech, XLVIII (February 1962) , 61. 162 could be signalic or symbolic. When he writes, "the utter ance is signalic," Coutu suggests that the sentence signaled the presence of a nut. Earlier he implied that words, not sentences, were signs. If one accepts the conclusion that words are signs, then only "nut" (or perhaps "hickory nut") could be signalic or symbolic (depending on the context). One might then ask what role "bring" and "me" and "a" play in the utterance. If, however, one assumes that the complete sentence is the signal or symbol, then he might ask whether the sentence "Bring me that hickory nut," uttered in exactly the same context as Coutu's "Bring me a hickory nut," would be sig nalic or symbolic in a different way or of a different nut. A related difficulty was the confusion of oral symbols with written symbols. After a discussion of Kenneth Burke's view of language, Jane Blankenship accepts Joshua What- mough's definition of language as "systematic verbal symbol ism. 1 1 l f She notes: "And since this book is about spoken language, we shall add: language is oral systematic verbal symbolism . . . 'oral' refers to the spoken rather than the written word" (p. 20). Blankenship does not distinguish between oral and written words except to suggest that the former are "a series of sound arranged in a specific pattern **Jane Blankenship, A Sense of Style: An Introduction to Style for the Public Speaker (Belmont, Calif., 1968), p. 19, citing Joshua Whatmough, Language: A Modern Synthe sis (New York, 1956), p. 20. 163 of rhythm and intonation" (p. 20). Her subsequent analysis deals with the meaning of single, written words— "orange," "pencil," etc., indicating that they are the entities that serve as symbols. Difficulties with this analysis arise because of the differences between spoken and written words. When one greets someone with the sounds represented by /hdU&rju / (howareyou?), has he said one or three spoken words? Is the wedding vow /cildu./ one or two "spoken" words? Does each symbolize some separate "thing?" Does the utterance itself symbolize something? Or are the written words, "I do," the entities that are symbolic? As part two of this chapter illustrates, a clear understanding of the nature of and dif ferences between oral, acoustic speech behavior and the abstractions from that behavior that are lenguage is a pre requisite for any valid account of language and meaning. Developments of the view that language is symbolic have often failed to make this essential distinction. Hence the first weakness of the symbol metaphor is that it does not consistently explain which language entities function as symbols. Difficulties with Non-symbolic Terms The second weakness of the symbolic view of language is that it cannot by itself account for the nature of many apparently non-symbolic terms. Rhetoricians and others Who 164 subscribe to the symbol metaphor have sometimes recognized the difficulties it creates with these terms, and have been notably unsuccessful in circumventing them. However, as was indicated in Chapter II, they continue to cling to the assumption that language is a system of symbols. For instance, the general semanticist Wendell Johnson recognized that terms such as "of" and "now" had no obvious referents, although he maintained that "there is a sense . . . in which such words can be given factual reference . . . . ”5 Rather than explain that "sense," he maintained that such terms are "fasteners and hinges in the language chain ..." (p. 129). Nords like "now" and "yet" he called "time words," which "express relations between events"(p. 130). There are several weaknesses in this attempt to account for non-symbolic terms. Johnson apparently failed to notice that without referents of some sort, his "hinge" and "time" words could not be consistent with his symbolic conception of language. In addition, he made no attempt to indicate the size of these classes of words, i.e., one wonders how many "hinge" or "fastener" words there are, and what differ entiates a "hinge" from a "time" word. For instance, "and," "because," and most other conjunctions are probably clear cases of "hinge" words. However, what about "while?" In the sentence, "He climbed the tree while she stood watch," 5Wendell Johnson, People in Quandries: The Semantics of Personal Adjustment (New York, 1946), pi 129. 165 does "while" function as a hinge between the two independent clauses, or is it a "time" word indicating that both actions occurred simultaneously? The symbolic analysis provides no way of answering this question— or many others one might raise. Some symbol theorists call apparently non-symbolic words "names of relations," rather than "hinge" or "time" words. However, others note the contradiction between that position and most theorists’ vehement warnings against reification.6 Some non-symbolic terms are also called "signals," because they allegedly tell us something about another word that occurs within the sentence. For instance, Blankenship argues that "the," "les," "die," and "los" all signal that a noun is coming; similarly, some verbs ending in "s" signal that the corresponding noun is singular.7 The principal weakness of this explanation is that it is oversimplified. For instance, one might ask whether "whereas" is a signal word, and if so, what it signals? Sometimes it indicates subordination; in other cases it begins the clauses of a 6See Johnson, p. 76ff., where he scornfully labels entities created to serve as solutions to unsoluble prob lems, "plogglies." 7Jane Blankenship, Public Speaking: A Rhetorical Per spective (Englewood Cliffs, 1966), p. 109. formal resolution.8 "About" presents similar problems. When used to mean "approximately," it can perhaps "signal" some numerical range (if indeed such a thing can be "sig nalled" in this sense). However, what about sentences in which it is used as it is in this one? What does "about" .signal there? Are signal words only signal words some of the time? Is there any consistent way to tell when they are? In another case words were divided into "lexical" or "content" words and "function" or "empty" words. Nouns, pronouns, adjectives, adverbs, and verbs went in the former category, while prepositions, conjunctions, "and so on" went in the latter.9 However, one must ask in what sense prepo sitions, conjunctions "and so on" could be said to be "empty" or free of content? Certainly it is not suggested that they are "empty" of meaning; one would then have to maintain that "I arrived early and knocked on the door," means exactly the same as "I arrived early but knocked on the door." Perhaps the distinction is meant to indicate such words are "empty" of reference; if that is the case, either the class is much larger than often is indicated, or the symbol theorist will have to be prepared to identify 8The writer is reminded of the probably apocryphal story of Edmund Burke, who realized clearly that he was not cut out to be a poet when, after careful deliberation, he began his first attempt at poesey with "Whereas . . . ." 9Blankenship, Style, p. 31. 167 referents for all sorts of apparently non-symbolic adjec tives, adverbs, and verbs, along with a perplexing group of pronouns such as "he" and "our," and a similarly puzzling series of nouns, including "semantics," "power," and "honor." On the other hand, perhaps the vague "and so on" was meant to leave the classification indeterminately large. Other authors experience similar problems with utter ances like "Ouchl" They call them, e.g., "covert expres sions," and conclude that, since they are not meant for com munication, they can be omitted from a treatment of lan guage.10 It has already been suggested that Sesonske's approach to such utterances is demonstrably more satisfac tory. Suffice it to say here that it is difficult to accept the notion that one's "Ouchl" while in the dentist's chair serves no communicative function, especially given the dentist's intense concentration on one's reaction to his drilling. In such cases where interjections do communicate, one must also ask how they would fit into the conception of language as a system of symbols. In sum, rhetoricians failed consistently to apply the view that language is a system of symbols, because they were unable to make the symbol metaphor adequately account for the nature of apparently non-symbolic terms. 10Virgil L. Baker and Ralph T. Eubanks, Speech in Per sonal and Public Affairs (New York, 1965) , p. 72. ----------------------------------------------------------- TFET Symbols and the Theory of Abstraction When they encountered non-symbolic terms, rhetoricians used the theory of abstraction to explain such terms in a symbolic context. The assumption underlying the theory is that all terms ultimately refer to objects. Relying princi pally on Korzybski's and Hayakawa's analyses, rhetoricians assumed that, although, e.g., "wealth" may have no obvious referent, when one traces it back down the abstraction lad der, he will discover that its referent is ultimately, in the proper context, "the cow known to science." The major shortcoming of the theory is identified by William Alston: "It is a defect in such a theory that no one has ever made a plausible case for the possibility of defining all meaningful words in the language in terms of the lowest level."11 In other words, the theory of abstrac tion— whether called the "structural differential" or the "ladder of abstraction"— has been postulated, accepted, and repeated, without having been extensively tested. The reason for that oversight is that the theory can only successfully be applied to a small number of words. It is of no help in explaining the meaning of the vast majority of "abstract" terms. For instance, several writers agreed that "semantics" is an example of an abstract word; one ^William P. Alston, Philosophy of Language (Englewood Cliffs, 1964), p. 68. 169 3hould thus be able to find its referent by tracing "seman tics" through successively less abstract related terms. If one follows Hayakawa's technique, first with his "Bessie," one might proceed in the following ways Beginning at a higher abstraction, one might ask, "What are the members of the class, 'farm assets?'" The obvious answers are machin- ary, land, livestock, etc. Any one of those terms is thus one step "down" the abstraction ladder. If one then asks, "What are the members of the class 'livestock?'" the answers would be cows, pigs, horses, etc. Asking the same question about the class of cows would bring one to the individual cow, "Bessie," and then, according to Hayakawa, to "the cow we perceive," and finally to "the cow known to science." If one asks the same question about "semantics," how ever, he would not find his way down the ladder so clearly narked. He may be able to identify a step ujo the ladder from "semantics," e.g., to Charles Morris' "semiology," which, in Morris' system, was made up of semantics, syntac tics, and pragmatics. However, what sort of answer can one give to the question, "What are the members of the class 'semantics?'" Perhaps the proper question is rather, "What are the constituents of 'semantics?'" i.e., "What is 'seman tics' made up of?" Relying on the definitions of semantics offered in Chapter II by Ullmann, Tarski, and Lewis, one might conclude that since semantics is the study of the relationships between words and things, its constituents are 170 words, things, and conclusions about their relationships. It is difficult to tell how to proceed beyond this point. However, (1) either one has reached the last perceivable level, and "words" and/or "things" are the referents of the word "semantics," or (2) it is necessary to probe further, to discover that only some kinds of words or things are referents for "semantics." In either case, the general semanticist would argue that there is still at least one un- perceivable "process level" below the lowest perceivable rung of the abstraction ladder, and that the "process" or "mad dance of electrons" at that level is the ultimate referent for "semantics." What sort of "process" that might ce is extremely difficult to say. Apparently the theory of abstraction does not apply to all "abstract" terms.12 However, perhaps some of the other terms Korzybski identified as "multiordinal" can be successfully analyzed with the help of the theory of abstraction. According to Korzybski they could? after listing some of the "endless 12Sarett, Sarett, and Foster, in a somewhat confusing analysis of "semantics," suggest that one can discover how "abstract symbols and the ideas they embody" can be con nected with "the external reality we can observe through our senses" by understanding the metaphorical nature of abstract conceptions. After examining the metaphorical nature of several words "pointing to parts of the body," e.g., "lion- hearted," "bloodthirsty," "hearten," etc., they point out that "semantics, the study of meaning, has its origin in semainein, the "Greek verb meaning ’to show."' From this analysisthey somehow conclude that "the symbols that embody abstract ideas have their roots in the external world that can be seen and heard." (Alma Johnson Sarett, Lew Sarett, and William Trufant Foster, Basic Principles of Speech, 4th ad. [New York, 1966], pp. 186-187.) ______ 171 array of the most important m.o. terms we have," he pointed out that such terms are ambiguous or infinity-valued, in general, and . . . each has a definite meaning, or one value, only and exclusively in a given context, when the order of abstraction can be definitely indicated.13 Consider one word Korzybski explicitly called "multi ordinal" (p. 433), the word "yes." The search for a refer ent for "yes" could conceivably lead to something like "an affective state or feeling of affirmation." However, such an entity could hardly serve as the referent for "yes" in the sentence, "The vote was 14 yes and 19 no," or the sen tence, "He is little more than a 'yes' man." In addition, finding a referent for each word in the elliptical response, "Yes, but . . . ." would require being able to separate the alleged affective state of affirmation from the accompanying alleged affective state of hesitation or qualification. And not only does this account ignore the difficulty of identi fying anything one might appropriately call "an affective state" of any sort, but it does not even ask how one would find the "un-speakable, objective level" at the very bottom of the abstraction ladder— in this case, the "affective state known to science." Apparently the theory of abstrac tion is inapplicable both to nouns like "semantics" and adverbs like "yes." Without further belaboring the point, 13Alfred Korzybski, Science and Sanity: An Introduc tion to Non-Aristotelian Systems and General Semantics, 3rd ed. (Lakeville, Conn., 1948), p. 433. Italics added. -----------------------------------------------------------------------1777 Lt is suggested that the theory will also fail to lead to the referents of most other "abstract" terms, including adjectives, prepositions, and conjunctions. According to Alston, the most strenuous efforts to apply the theory of abstraction have been made with theoret ical terms in science, and "even for such relatively low- level terms as 1 electric charge,1 'specific gravity,1 'habit,' and 'intelligence,' empiricists have by now admitted that such definitions cannot be provided."14 The reason the theory of abstraction cannot consistent ly be applied is that the assumption underlying it is false: all words are not uniformly symbolic— directly or indirectly — of "things," objective, emotional, atomic, or whatever. Weaknesses in the Theories of Meaning Rhetoricians adopted three theories of meaning, each of which was based on the assumption that language is funda- nentally a system of symbols. Their dependence on the sym- d o I metaphor meant that all of the theories were representa tional . That is, if one accepts the assumption that words are symbols, then it follows that they somehow represent, or stand for, or name something— some object, referent, desig nation, etc. When discussing the meaning of words, an adher- ant to the symbol metaphor can scarcely ignore the "things" that words are allegedly symbols of. Thus meaning must be ^Alston, p. 68. 173 explained in terms of what the words represent, ideas about what is represented/ or responses or dispositions to respond to what is represented. These approaches respectively char acterize the theories of meaning outlined in Chapter II. If one accepts the ordinary language philosophy view that/ since language is not a calculus/ words do not function in any one way, one must reject the representational theories of meaning: words do not just represent— objects/ ideas, or responses. Moreover, each of the representational theories has its own characteristic weaknesses. The Referential Theory The referential theory postulates that the meaning of a word is its referent or the relationship between the word and its referent. The theory probably attracts many adher ents because it seems so simple and clear. The word "chair" seems obviously to refer to what I am now sitting on; its meaning is the object supporting me. "House" also refers to what surrounds me, "city" to where I live, "car" to what I drive, and so on. The relationship between words and their referents, in short, seems obvious and one-to-one. However, examination of the referential theory from an ordinary lan guage perspective reveals that its simplicity and clarity are only apparent, because (1) it cannot adequately account for the meaning of nouns, pronouns, or proper nouns; (2) it cannot account for the meaning of verbs, adverbs, conjunc- tions, or prepositions; (3) words cannot, by themselves, refer to anything; and (4) in the cases where words are related to objects, it cannot be shown that they refer to them. In the first place, the theory cannot account for the meaning of nouns. If a noun's meaning is held to be the object or event to which it refers, then there must be a distinguishable object or event to serve as each noun's meaning. This is simply not the case, as Russell and other philosophers have shown. Although the expressions "the evening star" and "the morning star" have different mean ings, they refer to the same celestial entity. Similarly, many expressions may properly be used to refer to the same person or thing. "The Nobel Peace Prize winner assassinated in 1968," "Leader of the Montgomery bus boycott," and "Rev. Dr. Martin Luther King," unquestionably do not have the same meaning; yet, they refer to the same person or "referent." The converse is also true; some nouns or pronouns may tiave the same meaning but different referents. "I," "you," "here," and "this" are words that have constant meanings when used correctly. Yet, to the extent that they refer, their referent changes with each different instance of their use. If the meaning of "this" were its referent, no one could ever fully know its meaning, because it can be used to designate a virtually infinite number of things. Blankenship attempts to apply the referential theory to proper nouns when she contends that, while it may be diffi- -------------------------------------------------- ITS cult to identify the meaning of "Robert," if the person to whom the word refers is made clear, "who Robert is— that is, what the word 'Robert' means— becomes more obvious."15 As Wittgenstein noted, to say that "Robert E. Lee" means the Confederate general who surrendered at Appomatox is to con fuse meaning with the bearer of a name. It is to say that the meaning of "Robert E. Lee" was the son of "Light Horse Harry" Lee, and that the meaning of "Robert E. Lee" won the first battle of Bull Run. It is also to say that the mean ing of "Robert E. Lee" has been dead for over a hundred /ears. The referential theory of meaning is also difficult to apply to nouns like "pencil," which are often said to name a class or category of objects. For instance, Blankenship asserts that "pencil" refers to "that which writes."16 Blankenship's account is clearly oversimplified; "that which writes" can refer to an astonishingly large number of non pencil writing instruments. However, the problems of apply ing the referential theory of meaning to "class terms" are more Complex than might be apparent by an examination of Blankenship's assertion. Consider the word "house." The meaning of "house" cannot be its referent, because there is no &)}e thing or set of things that "house" always refers to. 15Blankenship, Public Speaking, p. 108. 16Blankenship, Style, p. 21. 176 The word can be used to talk about a wide variety of objects, from a toy dwelling constructed for a child's dolls, through all manner of family residences, to the forty-story headquarters of a large industrial concern.17 This fact suggests, as Wittgenstein and others have pointed out, that its meaning is a function not of its referent or its referring, but its use. It is perhaps even clearer that the referential theory cannot successfully account for the meaning of words other than nouns. For instance, a referential theorist might assert that the meaning of the verb "run" is the activity that the word refers to. But that explanation cannot ade quately account for utterances like "Don't let the water run too long," or "Run the flag up the flagpole," or "These nylons run easily." It is almost impossible to determine the number of separate activities and states of being that would have to be distinguishable in order for each instance of "run" to have some identifiable thing to refer to. Attempts to treat syncategorematic expressions referen- tially have led some theorists to conclude that "and" stands for a conjunctive function and "or" for a disjunctive func tion. As Alston notes, the difficulty with this point of view is 17The latter example may stretch the point somewhat; yet intra-plant publications are usually called "house organs," and restricted meetings "in-house conferences." --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --T77 that there is no way of explaining what a "conjunctive function" is, except by saying, for example, that it is what we are asserting to hold between the fact that it is raining and the fact that the sun is shining when we say "It is raining and the sun is shining." And this means that we cannot identify a "conjunctive function" except by reference to the way we use "and" and equivalent expres sions. Thus, we have not really gotten at an indepen dently specifiable referent for "and" .... In other words, to say that "and" stands for a "conjunctive func tion" is just to talk in a misleading way about the kind of function "and" has in sentences. No real extralinguis- tic reference has been demonstrated. Thus, there seems to be no escape from the conclusion that such expressions as conjunctions stand in no semantically interesting rela tions to extralinguistic entities.18 When scholars attempt to apply the referential theory to conjunctions, adverbs, adjectives, and even some nouns, they are forced to contend that these words can be said to refer to or be names of some kinds of things. The effort to determine what sorts of entities referents are often leads them to untenable conclusions, such as that "quick" is the name of a quality, or "slowly" refers to "how he walked." Such assertions run counter to the semanticists1 and general semanticists' advice against reification and hypostatiza- tion,19 in that they lead one to believe that "how he walked" is some thing that can serve as a word's referent. One is reminded of Ryle's complaint that not only Fido and London, but also centaurs, round squares, the present King of France, the class of albino Cypriots, the first moment of time, and the non-existence of a first moment of time must all be credited with some 18Alston, p. 18. 19See, e.g., S. I. Hayakawa, Symbol, Status, and Per sonality (New York, 1963), pp. 114-116. 178 sort of reality.. 2 0 In brief, the referential theory of meaning cannot ade quately explain a large number of linguistically meaningful words without making claims that are anathema to the same persons who support the theory. That is, either the theory must not be used when explaining the meaning of adjectives, adverbs, conjunctions, prepositions, or interjections, or one must be willing to give "referent" status to a wide variety of pseudo-entities, few of which can be identified or described, and some of which cannot even be shown to be extralinguistic. The third major weakness of the referential theory is that words cannot refer, by themselves, to specific parts of complex entities. Robert Terwilliger notes this deceptively simple principle of language analysis when he writes, "an essential characteristic of language is that it is used."21 That is, words can only be used by someone in some situation to refer to something. When I say to my daughter, "Stay in the house until noon," the word "house" can be said to refer to this dwelling we now live in. But that is not to say that the word "house" means this dwelling because it refers to it, but rather to say that I did at that particular time . 20Gilbert Ryle, "The Theory of Meaning," in Philosophy and Ordinary Language, ed. Charles E. Caton (Urbana, 1963), p. 146. . 21Robert P. Terwilliger, Meaning and Mind: A Study in the Psychology of Language (New York, 1968), p. 148. 179 ase the word "house” to refer to this dwelling. Its meaning Ls thus a function of the use of the word, not the word's alleged referent. Terwilliger points out that a word like "table" has meaning precisely because it has no referent, aut "can function in a large number of conceivable situa tions, and because [it] is not determined by any specific aneV (p. 150). As Wittgenstein notes, it is a mistake to assume that a shild learns "the meaning" of an apparent name like "chair" ay learning what object the word refers or points to. When we learn a name, we learn not only that it is a name for something, but that it is a name. We learn not only what it refers to but that it refers. We learn not only that it refers to this but that it refers .... The meaning, therefore, is not a thing referred to but the use of a word to refer to such a thing..22 The referential theory is also inadequate because those words which clearly stand in some relation to an entity or 3et of entities or "referents" cannot be said to refer to bhose entities or "referents." For instance, although one may contend that the word "pencil" is related to a class of Dbjects which can be identified and described, it cannot be shown that "pencil" refers to that class of objects, because if one wishes to talk about that class of objects, he cannot use "pencil" to do it. If one wants to say, for instance, 22Garth Hallett, S.J., Wittgenstein's Definition of pieaning as Use (New York, 1967), p. 9l. ~ 180 that the class of pencils is very large, he cannot say, "Pencil is very large." The word "pencil" alone cannot be used to refer to the class of objects it allegedly names.23 Similarly, even if one assumes that adjectives, are sometimes related to what could be called qualities or characteris tics, and verbs are related to kinds of behavior or activ ity, these types of words also cannot be said to refer to their alleged referents. If a characteristic could be iden tified, for example, as "the quality of beauty," one could not refer to it simply by using the word "beautiful," i.e., if he wished to say that the quality of beauty varied from culture to culture, he could not say, "Beautiful varies from culture to culture." And as Alston writes, "I would not say that what I just did belongs to the class of acts of running by uttering the sentence, "What I just did belongs to run." 2 * * In brief, sometimes, in some situations, some words may be used by some people to refer to some things. However, since referring is only one linguistic job among others, and is assigned only to some types of expressions in some situa tions, "no account of meaning that presupposes that all meaningful units refer to something can be correct." In other words, "it cannot be the case that to say a word has a 23See Alston, p. 1 6 . ____________ 2* * Alston, p. 15. -----------------------------------------------------------------------TffT oertalh meaning Is to say it refers to some thing. 2 5 When rhetoricians develop their accounts of the refer ential theory of meaning with treatments of connotative and denotative meaning, they also manifest confusion between logical connotation and literary ambiguity. They view deno tative meaning as a word's "exact literal meaning, its abjective reality, the physical, chemical, or structural characteristics of whatever the word stands for,"26 often stressing the idea that denotative meaning is "a sign-object relationship."27 Connotation, on the other hand, involves "emotive," "subjective," and "personal" meaning,28 or is described as "the relationship between the object, the words, and the speaker/listener."29 However, the distinc tion rhetoricians make between denotation and connotation is Doth inconsistent with that made by the semanticists upon whom they base much of their approach to language, and onworkable. The semanticist, C. I. Lewis, defines denotation in 25Alston, p. 16. 26Martin Andersen, Wesley Lewis, and James Murray, The Speaker and His Audience (New York, 1964), p. 94. . 27David K. Berio, The Process of Communication: An Introduction to Theory and Practice (New York, 1960), ? . 192. . 20Andersen, Lewis, and Murray, p.-96. Cf. John F. Wil son and Carroll C. Arnold, Public Speaking as a Liberal Art, 2nd ed. (Boston, 1968), p. 306. 29Blankenship, Public Speaking, p. 106._________________ 182 much the same way as do the rhetoricians: "the denotation of a term is the class of all actual or existent things to which that term correctly applies."30 However, his defini tion of connotation differs greatly from their approach: If nothing would be correctly namable by "T" unless it should also be namable by "Ai", "A2", and ... and by "An", and if anything namable by the compound term "Ai and A2 and ... and An" would also be namable by "T" then this compound term, or any which is synonymous with it, speci fies the connotation of "T" and may be said to have the same connotation as "T".31 Alston's version of the same analysis of connotation is somewhat clearer: W connotes the property, P [if] the possession of P by something is a necessary and sufficient condition of W being correctly asserted of it (that is, of its belonging to the denotation of W).3 2 In other words, "person" connotes the property "human," because if something possesses the property of human-ness, it is a necessary and sufficient condition of one asserting that it is a person. The primary reason for these widely divergent approaches to connotation seems to be that semanticists are interested in a logically adequate definition of connotation that is explicitly consistent with the symbol metaphor and their account of denotation, whereas speech scholars seem more interested in accounting for what might be called the 30"The Modes of Meaning," Semantics and the Philosophy of Language, ed. L. Linsky (Urbana, 1952), p. 52. 31Lewis, p. 53. 32Alston, p. 17. 183 literary ambiguity of many words' meanings. Unfortunately, the rhetoricians' literary account of connotation is not specifically designed to follow logically from the semantic account of denotation. Thus rhetoricians are forced, in order to maintain some semblance of consistency with the symbol metaphor, to write of denotation as "sign-object1 ' meaning and connotation as "sign-object-person" meaning. The major problem with that approach is that it makes the connotation-denotation distinction unworkable by render ing the concept of denotative meaning virtually useless. That is, since rhetoricians also maintain that meaning is to oe found in people rather than in things, there can be no meaning until.some relationship exists between a person and the other aspects of the meaning situation. Thus denotative meaning, which allegedly is a function only of the "sign- object" relationship, by definition cannot exist. The rhetoricians' approach to denotation and connota tion has also led them to argue that all words have connota- tive meaning to "go along with" whatever denotative meaning might be isolated. That point of view raises many questions about the connotative meaning of words like "five" or "sub tract." Although such words may cause some purely emotional response in some persons at-some times, to assert that all words have what the speech scholars called connotative mean ing is at least problematic. Most of the weaknesses in the referential theory of -------- I W neaning are summarized by William Alston: The referential theory is based on an important insight— that language is used to talk about things outside (as well as inside) language, and that the suitability of an expression for such talk is somehow crucial for its having the meaning it has. But in the referential theory, this insight is ruined through over-simplification. The essen tial connection of language with "the world," with what is talked about, is represented as a piecemeal correlation of meaningful linguistic units with distinguishable compo nents of the world. What the preceding discussion has shown is that the connection is not so simple as that. Speech does not consist of producing a sequence of labels, each of which is attached to something in "the world." Some of the meaningful components of the sentences we use to talk about the world can be connected in semantically important ways to distinguishable components of the world, but others cannot. Hence, we must look elsewhere for an account of what it is for an expression to have meaning, remembering that the account must be framed in such a way as to give due weight to the fact that language is somehow connected with the world. (p. 19) The Ideational Theory Rhetoricians' accounts of the ideational theory of meaning are generally consistent with Susanne Langer's approach, i.e., that meaning is not to be found in objects ar designata, but in ideas: "It is the conceptions, not the things that symbols directly 'mean.'"33 As Alston points Dut, the ideational theory is in the background when people think of language as an "instrument for the communication of thought," or when they define a sentence as a "chain of 33Susanne K. Langer, Philosophy in a New Key: A Study in the Symbolism of Reason, Rite and Art, 2nd ed. (New York, 1951), p.61. 185 words expressing a complete thought."3* * According to this theory, language is the publicly observable indication of private ideas that are coursing through our minds. The ideational theory is inadequate because it suffers from all the shortcomings of the referential approach, because it is unworkable, and because it cannot account for the meaning of even those words that are related to mental images. As explained in Chapter II, the essential differ ence between the referential and ideational accounts of. meaning is that the referential postulates that meaning is in a word's objective referent, whereas the ideational holds that the meaning of a word is the idea it symbolizes or represents, i.e., the ideational theorist views all refer ents as ideas. Thus the ideational theory inherits the weaknesses of the referential. The ideational theory is unworkable first, because, as Terwilliger pointed out,. the theory has 5ne logical drawback— namely, that it can explain everything. No matter what word we dredge up, if we ask what its meaning is or to what it refers, we can say "the idea or concept of X." All words can be accounted for in this manner, since all that one is really doing here is saying that the word has meaning, and that we know already. We cannot explain the meaning of a word by postulating undefined ideas to which it refers; to do so is to merely restate the fact that it has meaning.35 In other words, the concept of "idea" is so vague that one 3lfFor examples of this point of view, see, e.g., Dick ens, pp. 16-17, and Gray and Braden, p. 456. 35Terwilliger, p. 151. 186 can use it to explain virtually any kind of "meaning." Ter williger might also have pointed out that "the meaning" of flags and other patriotic symbols, geometric forms, and color combinations could equally comfortably be explained by reference to the "ideas" of "fatherland," "solidity," or "heat and cold." Additional difficulties arise when one attempts to apply the ideational theory of meaning to specific utter ances. For instance, consider the sentence, "I have a dream, that one day this nation will rise up and live out the true meaning of its creed . . . ." When one utters that sentence aloud with his mind on what he is saying, it is very doubtful that there is a distinguishable idea in his mind that corresponds to each linguistically meaningful unit in it. Is there, for instance, an idea of "that," "will," "live," (or "live out"), and "true" that is present as each word is pronounced? And even if there is an idea of "will," is it recognizable as the same idea that is in one’s mind whenever he utters "will" in that sense? Could one call up the idea if the word were not present, or does one at least know what it would be like to do so? In other words, is the idea something identifiable and producable apart from the word? Does one ever get the idea of "will" when he utters other words— "until," "mackerel," "procrastinate?" As Alston points out, finding answers to such perplex ing questions is not the only problem faced by those who 187 would apply the ideational theory. What is disturbing about these questions is not that they have one answer rather than another, but that we do not know how to go about answering them. What are we supposed to look for by way of an idea of ["will"]? How can we tell whether we have it in our mind or not? Just what am I supposed to try for when I try to call it up out of con text? The real difficulty is that we are unable to spot "ideas" as we should have to in order to test the idea tional theory.36 In other words, the ideational theory is unworkable, not because ideas cannot be verified by the senses, but because one cannot say how to go about identifying ideas, or how one would know when he had identified one. As Wittgenstein sug gests, ask yourself, "When I said 'Give me an apple and a pear and leave the room,1 had I the same feeling [idea, con cept, etc.] when I produced the two words 'and?'"37 Wittgenstein's suggestion also illustrates why the ide ational theory cannot account for communication or mutual understanding. Communication cannot be a matter of the com municatee creating an idea or image that is "similar in all relevant respects" to that of the communicator because, "when you have such difficulties knowing what you yourself experienced, how can you possibly discover what another per son experienced?"38 The third major weakness of the ideational approach is 36Alston, p. 24. 37Ludwig Wittgenstein, The Blue and Brown Books (New York, 1958), p. 79. 3 0Hallett, p. 65. _ -------------------------- J g g that it does not satisfactorily explain even the meaning of words that have an obvious connection with mental images, e.g., "dog," "truck," and "typewriter." Insofar as one's use of "dog" is accompanied by a mental image, the image is by no means the same each time he uses that word in the same sense. At one time it may be the image of a dog he once owned, at another of a dog he would like to own, at another the image of a dog sitting, at another the image of a dog standing, etc. Conversely, one can have the same mental image for several different words with quite different mean ings. For instance, the image of a standing English bulldog night accompany one's utterance— or hearing— of the words "bulldog," "dog," "pet," "watchdog," or "animal." In sum, the ideational theory is subject to all the weaknesses of the referential theory. In addition, it is unworkable, chiefly because it posits the existence of a 3lass of entities that virtually defy explanation. More over, it cannot explain even the meaning of image- or idea- celated words. ' Che Behavioral Theory Rhetoricians who subscribe to the behavioral theory of meaning contend that the meaning of a word is the response :hat it elicits in those who perceive it. The theory is attractive, especially to scientifically-oriented students of language, because it appears to avoid many of the short- oomings of the explicitly referential and ideational________ — m approaches by focusing on publicly observable aspects of speech behavior and on speech behavior in communication situations. That is, the behavioral approach apparently does not rely for an explanation of meaning on the identifi cation of mysterious entities such as the referents of "of" or the conception of "toward." Instead, it seems to utilize the fact that linguistic meaning is to a large extent a mat ter of widespread agreement about how words are used by peo ple to communicate. Moreover, the success psychologists have enjoyed in explaining other forms of behavior in terms of observable stimuli and responses naturally gives rise to the hope of being able to give the same kind of treatment to verbal behavior. The simplest version of the behavioral theory postu lates that the meaning of a word is the response it stimu lates in the hearer. This approach is weak because, like the ideational theory, it requires that there be features of any given response to a word that are common and peculiar to all responses made to that word when used in a given sense. That is, whenever one says "play" in the sense of playing a game, this version of the behavioral theory requires that the responses to "play" must be common to all uses of "play" in that sense. In addition, it requires that the responses be peculiar to the use of that word, i.e., not common to the use of another word, or to another use of "play." These conditions are obviously not satisfied by the use of "play," 190 or many other terms. This version of the theory is inade quate, in short, because the meaning of a word is not con sistently dependent on the way people react to it. However, perhaps it would do more justice to the behav ioral theory to consider responses to complete utterances, rather than to single words. Imperatives seem to hold the most promise, since they explicitly require responses, and thus offer the greatest opportunity to observe and correlate them. But the meaning of an imperative also cannot simply oe the response it elicits. Consider, for example, the parent-to-chiId imperative, "Sit up straight and eat your *1 • . . . . t dinner." That demand could conceivably elicit a wide vari ety of responses, ranging from no response, through explicit refusal to comply, demand for justification, justification for noncompliance, and change of subject, to compliance. £et it is clearly not accurate to say that the imperative tiad a different meaning in each case, depending on the' response it elicited. Meaning is apparently not simply oehavioral response. Charles Morris attempted to avoid some of the difficul ties of the simple response analysis by defining meaning in terms of "dispositions to respond" rather than overt, observable responses. According to his approach, the mean ing of "sit up and eat your dinner" is not the overt reac tion, but the disposition or proclivity to react inwardly or outwardly to the command. Alston notes that Morris' modifi- 191 cation of the theory does not solve all of the problems of the behavioral approach, because (1) there are many sen tences, such as "Mozart wrote Idoffieneo at the age of 25," that can hardly be said to produce any semantically impor tant dispositions beyond perhaps the one to say "twenty- five" if asked, "At what age did Mozart write Idomeneo?" (2) A statement like "Your son is ill," will produce the disposition, for example, to go to one's son, only if one understands the speaker, believes the speaker, does not already know of his son's condition, has not already taken care of his son, etc. In short, one may have to make a "disposition" indefinitely complicated before there is any plausibility in supposing that it is regularly produced by the utterance of, and thus could be the meaning of, a given sentence. Alston points out that Morris also saddled him self with the fundamental assumption of the referential the ory, that every meaningful expression is a "sign" of some thing .3 9 The most sophisticated version of the behavioral theory that speech scholars cite is Charles Osgood's. The key to Osgood's approach, as noted in Chapter II, is his mediation hypothesis, which postulates that organisms do not always respond directly to sensory stimuli, but sometimes behave in response to stimulation caused by a mediating response to an "Alston, pp. 28-30. 192 original stimulus. Osgood stresses the notion that the mediation process associated with the sign of an object must include some portion of the behavior originally experienced toward the object itself. ". . . the sign 'means' or 'refers to' a particular object because it elicits in the organism employing it part of the same behavior which the object itself elicits."1 '0 Osgood's account of meaning in terms of the mediation hypothesis is difficult to accept or apply chiefly because of his use of the word "object." On the one hand he writes of "objects" as if they are common "things" like bats and balls and balloons: a pattern of stimulation which is not the object is a sign of the object if it evokes in an organism a mediating reaction, this (a) being some fractional part of the total behavior elicited by the object and (b) producing distinc tive self-stimulation that mediates responses which would not occur without the previous association of nonobject and object patterns of stimulation. (p. 696) On the other hand, Osgood maintains that "objects" can be "actually any and all patterns of stimulation" from a fire place (p. 291), to "a gust of cold, northerly wind against the face" (p. 291). Osgood's utilization of "objects" makes his approach representational, because he asserts that words are signs of these objects. He contends that these word-signs have mean ing insofar as they elicit in someone part of the behavior lf0Charles E. Osgood, Method and Theory in Experimental Psychology (New York, 1953), p. 412. 193 he originally experienced toward the object ("pattern of stimulation”) that the word is a sign of. However, problems arise with meaningful utterances not as clearly associated with physical objects as is Osgood's example, "spider." For instance, consider the phrase, "Fourscore and seven years ago . . . ." Unless Osgood were to contend that those words have no meaning, he would have to identify and describe the "object" that "fourscore" is a sign of, so that one could tell if his meaning for "fourscore" were indeed part of the behavior the "pattern of stimulation" "fourscore" originally elicited in him. One would also have to identify a "pattern of stimulation" for "seven," "years," and "ago." It appears that Osgood's theory might explain the meaning of many nouns and some pronouns. But it is difficult to tell where even to look for the "patterns of stimulation" allegedly signi fied by words like "whereas," "divide," "really," "charac terize," or "psychology." One might accuse Osgood of "trying to have it both ways." He apparently uses "objects" in a conventional sense in order to make his explanation of meaning as clear and reality-oriented as that of the explicit referential theor ists. But he attempts to avoid the shortcomings of the referential theory by making his "objects" "patterns of stimulation." In spite of his attempts, however, even his account of meaning in these terms cannot explain the meaning of many linguistically important entities, because he too is 194 concerned exclusively with responses, and thus his theory can be criticized on the same grounds as the other behav ioral accounts noted above. Alston summarizes his evaluation of the behavioral approach in these words: Like the others, the behavioral theory is based on an insight that it perverts through oversimplification. Just as the meaningful use of language has something to do with reference to the "world," and just as in some way we do express and communicate our thoughts in using language, so it is also a significant fact that units of language get their meaning through being used by people, through the fact that they are involved in various sorts of behavior. Behavioral theories err in conceiving this behavioral involvement in oversimplified terms. They suppose that a word or sentence has a certain meaning by virtue of being involved, as response and/or stimulus, in stimulus- response connections that are basically similar, except for complexity, to a simple reflex like the knee jerk. Unfortunately, no such connections have ever been found, except for those that are obviously not determinative of meaning, like the fact that a sudden loud utterance of "Look out!" typically elicits a start. Some more adequate characterization of linguistic behavior is called for i i * f 1 t • • i The presupposition that language is a system of symbols is an inadequate foundation for a consistent, viable approach to the nature of language. The symbol metaphor does not explain which linguistic entities can serve as sym bols, and it cannot successfully account for the nature of apparently non-symbolic terms. In addition, the assumption grounding the crucial theory of abstraction is unsound: all words are not, directly or indirectly, names of objects. 1Alston, p. 31. 195 The referential theory of meaning is unsound because sach meaningful word does not "have" its own extralinguistic referent, because words themselves can only be used to refer, and because when a word is related to an object or class of objects, it cannot be said to refer to them. In addition, rhetoricians' accounts of connotation and denota tion manifest a confusion between those concepts and what might be called literary ambiguity. Besides suffering from the inadequacies of the referential theory, the ideational theory is unworkable because it depends on one's ability to identify and describe a discrete "idea," "conception," etc., for each linguistically meaningful unit of discourse, and it cannot account for the meaning of image-related words. The behavioral theory is also inadequate chiefly because it limits meaning to response. Utterances do not have differ ent meanings just because different people react or are dis posed to react to them in different ways. Moreover, Os good's account of this theory depends on the existence of "patterns of stimulation," which are as difficult to iden tify and describe as "ideas." A Reason for the Symbol Metaphor: The Speech-Language Dichotomy Before suggesting an alternative to the view that lan guage is fundamentally a system of symbols, it would be helpful to attempt to determine why the symbol metaphor has been so universally approved. Although ordinary language 196 philosophers have not made the point, it would seem that one reason why rhetoricians have relied on the symbol metaphor and representational theories of meaning is that they have depended on scholars who focused their study on language rather than speech. In order to understand (1) why vir tually all language scholars have begun their analyses with the question, "What is the nature of language?" and (2) why they should begin with the question, "What is the nature of speech?" one must first understand the distinction between speech and language. The distinction is formalized by Ferdinand de Saussure in his Cours de Linguistique Gdnerale.1 *2 De Saussure dif ferentiates between three objects of study: Parole, lan- gage, and langue. By parole, which is translated "speak ing," De Saussure means individual idiosyncratic speech; by langage, or "speech," he means the language activity of a linguistic community; and by langue, or "language," he means the system made up of speech and the rules for its use. De Saussure maintains that individual acts of speaking (parole) cannot be studied separately, because they cannot be orga nized into a unified system without being grossly distorted. However, he admits that parole is fundamental to both langage and langue: . . . speaking is necessary for the establishment of lan ded. Ch. Bally et A. Sechehaye avec la collaboration de A. Riedlinger (Paris, 1960). 197 guage and historically its actuality always comes first. . . . We learn our mother language by listening to others; only after countless experiences is it deposited in our brain. . . . Speaking is what causes language to evolve.1 *3 De Saussure explains langage/ or "speech/" in these words: Taken as a whole, speech is many-sided and heterogeneous; straddling several areas simultaneously— physical, physi ological, and psychological— it belongs both to the indi vidual and to society; we cannot put it into any category of human facts, for we cannot discover its unity. (p. 9) Be emphasizes that langage cannot be systematically studied D e c a u s e , like p a r o l e , i t i s t o o h e t e r o g e n e o u s t o be u n i f i e d . Langue, or "language," is the most important of the three divisions; it denotes the system of language, "both a social product of the faculty of speech (langage) and a col lection of necessary conventions that have been adopted by a social body to permit individuals to exercise that faculty" (p. 9). De Saussure views langue as a well-defined object in the heterogeneous mass of speech facts. He sees langue as external to an individual, and existing only by virtue of a kind of social contract. Because of its unified nature, it can be conveniently studied separate from parole or langage. In addition, "linguistic signs are tangible; it is possible to reduce them to conventional written symbols, whereas it would be impossible to provide detailed photo graphs of acts of speaking. . ." (p. 15). lfSDe Saussure, Course in General Linguistics, ed. C. Bally and A. Sechehaye in collaboration with A. Riedlinger, trans. Wade Baskin (London, 1960), p. 18. 198 Langue exists in the form of a sum of impressions depos ited in the brain of each member of a community, almost like a dictionary of which identical copies have been distributed to each individual. . . (p. 18) De Saussure asserts that language should be the primary object of study. He writes, "from the Very outset we must put both feet on the ground of language and use language as the norm of all other manifestations of speech" (p. 9) . He maintains that those scholars who believe that parole should be studied rather than langue are mistaken. Although it may appear that langue is conventional or learned, whereas parole is natural and therefore of primary importance, De Saussure contends that the reverse may in fact be true. He argues that since our original Latin word for articulated speech referred to individual units of a systematic se quence, "we can say that what is natural to mankind is not oral speech but the faculty of constructing a language, i.e.,. a system of distinct signs corresponding to distinct ideas" (p. 10). Otto Jespersen generally accepts De Saussure1s analy sis, but concentrates his attention on two of De Saussure's three areas of study: language and speech. Jespersen notes that the English phonetician, Harold Palmer, makes essen tially the same point as De Saussure when he calls speech "the sum of the mental and physical activities involved when * one person communicates to another. . . ," and language "the sum of the conventions adopted and systematized by a social- 199 ized mass of users of [speech] in order to ensure common intelligibility. "I f ‘ ' Jespersen also reinforces De Saussure's and Palmer's emphases on the study of language rather than speech in the following way: "... the task set to lin guistic science . . . is the study of the language, while, by comparison with this, speech is for philologists some thing accidental and secondary or subordinate." **5 . . . A prominent American ordinary language philosopher, William Alston, similarly distinguishes between speech and language. He notes, we should keep in mind the often repeated, but seldom con sistently observed, distinction between language and speech. Speech comprises the totality of verbal behavior that goes on in a community; whereas language is the abstract system of identifiable elements and the rules of their combinations, which is exemplified in this behavior and which is discovered by an analysis of the behavior.1 '6 In short, De Saussure, Jespersen, and Alston agree fun damentally on the distinction between language and speech. ‘ '''Otto Jespersen, Mankind, Nation, and Individual from a [linguistic Point of View (Bloomington, Indiana, 1964), p. 13. ^5Jespersen, p. 12. ^Alston, pp. 60-61. In an encylopedia article, Alston wrote of language: it is "not a kind of activity, process, or social interaction; still less is it an aggregate of actions, sounds, or other concrete phenomena. It is, rather a system of abstract elements." In the same place he de scribed speech as "the totality of verbal activity in a com munity," and later defined language as the abstract system of sound types and the rules of their combination, which is exemplified in verbal activity and which is discoverable through an analysis of this activity." (Alston, "Language," in The Encyclopedia of Philosophy, ed. Paul Edwards [New Jfork, 1967], IV, 385. 2__ They also concur that one should focus his attention on lan guage . And they agree that language is fundamentally a sys tem of symbols. As Alston expresses it, It should be clear that language belongs somewhere within the category of symbols, in Peirce's sense of the term. Language is often defined as a system of symbols, and this can be accepted as a summary statement.I f 7 An Ordinary Language Approach to Speech and Language In this writer's judgment, the most important insight one can gain from an ordinary language philosophy .approach to language and meaning is this: instead of focusing ini tially on language, as did De Saussure, Jespersen, Alston, and their semanticist, general semanticist, and rhetorician colleagues, the language scholar should begin his inquiries by focusing on speech. He should do so for two reasons: (1) His first questions should be about the most fundamental aspect of his subject matter, and speech is more fundamental than language. (2) If he begins by studying language sys tems instead of speech behavior, he is liable to make over simplified generalizations about human communication. Speech Is More Fundamental Than Language If theories of rhetoric, like theories of semantics and general semantics, are viewed as philosophical systems, ^Alston, p. 59. See also p. 98, where he bases his analysis of metaphor on Peirce's'approach. 201 having a set of basic assumptions and a means by which the assumptions can be consummated,^8 then the rhetorician, semanticist, general semanticist, and philosopher of lan guage must initially be dedicated to the systematic critique of presuppositions. **9 That is, it is crucial how each the orist decides where to begin his inquiry, i.e., what ques tions to ask in order to establish and clarify fundamental assumptions— "starting points" in Otis Walter's words.50 The most fundamental assumptions underlying all the various aspects of language and its use are assumptions, not about the nature of language, but about the nature of speech. Speech may be viewed as a kind of ordinary human activ ity or behavior, which often takes place in interpersonal situations and which, like similar kinds of ordinary human aehavior, usually follows some imperfectly-defined set of Informal logical rules. Obviously many kinds of human oehavior fit this description; speech behavior differs from them in only one respect: instances of speech behavior are Instances in which humans make meaningful utterances. Speech behavior is limited to humans because, although **8Walter R. Fisher, "The Importance of Style in Systems of Rhetoric," Southern Speech Journal, XXVII (Spring 1962), L73-182. H9See Chapter I, note 11. 50Otis M. Walter, "On Views of Rhetoric, Whether Con servative or Progressive," Quarterly Journal of Speech, XLIX (December 1963), 373. 202 porpoises, chimpanzes, etc., may engage in various forms of speech-like activity, what they do is not generally called "speech" in the sense that human speaking is. What is implied by requiring that speech behavior be meaningful is clarified below. Here it is perhaps enough to note that the requirement does not limit speech behavior to interpersonal situations. A human utterance "to oneself" can be as mean ingful as discourse with another. The use of "utterance" in the description is meant to indicate that the behavior is primarily oral, i.e., is largely made up of the sounds recognized as words, and those that affect the rate, volume, quality, intensity, etc., of oral speaking. Gestures, changes of posture, and facial expressions can supplement speech behavior, but a strong downward motion of a clenched fist is not itself an instance of such behavior. When some one utters the words / bol/ , / / , or V£Yi tol / , tiis speaking is an instance of speech or speech behavior. On the other hand, the words "ball," "growl," or "The trees are very tall," are language entities. In other words, language may be viewed, as Alston suggested, as "an abstract system of identifiable elements and the rules of their combinations." Verbal codes are the "identifiable elements" Alston wrote of, i.e., words and forms of written punctuation. The "rules for the combinations" of the iden tifiable elements are the informal logical rules of lan guage. These linguistic rules are "informal" in two senses: 203 First, as Ryle explained in the lecture discussed in Chapter III,s 1 formal logic limits the validity of related asser tions by means of highly systematized, "commodity neutral" rules, whereas informal logic deals with topic-related expressions whose logical powers can only be extracted by examining the dealings people transact with them. All dis course is not and cannot be reduced to formal logical dis course. Ordinary words used in ordinary speaking have "their logical powers," but not because of any context-free linguistic fiat. How words can be used, what can be said, what makes sense and what does not, are functions not of formal logic, but of "the informal logic of the functioning of expressions."5 2 Second, linguistic rules are "informal" in that every one has license to violate them if he can show a point in doing so.53 We can "give meaning" to utterances that would appear meaningless or even self-contradictory by themselves. For instance, the sentence, "I am and I am not," seems utterly self-contradictory. Yet, the question, "Are you a sxGilbert Ryle, Dilemmas (Cambridge, 1966), pp. 111- 129. 52Strawson has also commented on the differences between context-free formal logic and the context-bound con cepts of ordinary language. See, e.g., John Passmore, A Hundred Years of Phi-losophy, rev. ed. (New York, 1966), pp. 4Vlff. 53P. F. Strawson, introduction to logical Theory (Lon- don, 1966), p. 230.__________________________________ _______ 204 supporter of the governor?" might appropriately and meaning fully be answered by that utterance. As Strawson notes, "If we do this, we invite the question, 'What do you mean?'? and if we can explain what we mean, or show the point of saying what we say, then we have not contradicted ourselves."51* We also commonly speak of figurative uses of language, and of the creative writer's "poetic license" to break recognized linguistic rules for aesthetic effect. Yet it is often difficult to determine when some rule is, in fact, being broken. When e. e. cummings writes, "the sweet small clumsy feet of april came into the ragged meadow of my soul," he is clearly utilizing his privilege to break lin guistic rules for poetic effect. But what about speaking of "the leg of a table?" As Richards points out, such a state ment is metaphorical— although he called it a "dead" meta phor.55 Can we clearly say in such instances which lin guistic rules are applicable, which are being observed, and which broken? As Strawson notes, this fluidity in our rules, and this imprecision in the distinctions they involve, are things we must be aware of if we aim at a realistic study of the logic of ordinary speech. But though they make such a study more compli cated and less tidy than the study of formal systems, they do not make it impossible. In a way, the awareness of them makes it easier; for if we realize that we are at best describing only the standard and typical uses of certain kinds of expressions, we shall be less 5HStrawson, p. 7. 55I. A. Richards, The Philosophy of Rhetoric (New York, 1965), p. 117. 205 disconcerted by untypical cases.5 6 In sum, the relationship between speech and language being drawn here is similar to that Katz suggests when he writes, "by speech we understand an observable phenomenon with physiological, behavioral, and acoustic aspects and by language we understand a certain mental reality underlying observable speech phenomena."57 More specifically, speech may be defined as 'meaningful' human oral behavior governed by informal logical rules. Language is an abstract system con sisting of verbal codes and the informal logical rules that govern speech behavior. Speech is more fundamental than language because speech is behavior whereas language is an abstraction from that behavior. As noted in Chapter III, both Wittgenstein's treatment of language games and Austin's concept of speech acts stresses the idea that speaking involves human activity or behavior. One assumption of the symbol theory, on the other hand, is that speaking is something that usually takes the place of acting. Rhetoricians, semanticists, and general semanticists commonly begin their investigations by focusing on codified words, i.e., the words that partially constitute language, words that can be written down, etc., rather than words-being-used or speech behavior. Since they erroneously 56Strawson, p. 231. 57Jerrold J. Katz, The Philosophy of Language (New York, 1966), p. 116. — — 206 assume that words stand for something that the speaker would have done, thought, believed, etc., had he not spoken instead, words are commonly treated as convenient surrogates for actions. This point of view is expressed in a beginning speech text in the following way: "By repeatedly hearing certain vocal sound patterns in close association with things, actions, events, or qualities while you were learning to speak, you learned to use those patterns yourself as substi tutes for the nonverbal occurrences."5 8 The idea is also expressed in another context. The authors of several argu mentation texts point out that speech plays a crucial role in a democratic society, where citizens are continually faced with the choice of resolving controversy by "either talking it out or fighting it out."59 Rational delibera tion, discussion, and debate have always been preferred seGray and Braden, p. 456. Italics added. Cf. Alan H. Monroe and Douglas Ehninger, Principles and Types of Speech, 6th ed. (Glenview, 111., 1967), pp. 2 and 21; Andrew Thomas Weaver and Ordean Gerhard Ness, The Fundamentals and Forms jf Speech, rev. ed. (New York, 1963), p. 11; Sarett, Sarett, and Foster, p. 80; and Raymond S. Ross, Speech Communica tion: Fundamentals and Practice (Englewood Cliffs, 1965), ?. 36. 59See, e.g., James H. McBurney and Glen E. Mills, Argumentation and Debate, 2nd ed. (New York, 1964), chapter 1; Glen E. Mills, Reason in Controversy: An Introduction to general Argumentation (BBston, 1964), chapter 1, especially ?p. 12-13; Eugene R.Moulton, The Dynamics of Debate (New JTork, 1966), chapter 1. Cf. Eubanks and Baker, chapter 1, aspecially p. 5; and A. Craig Baird and Franklin H. Knower, Sssentials of General Speech, 3rd ed. (New York, 1968), shapter 1. -----------------------------------------------------------------------ZffT over their alternatives, because they only involve symbols of— or convenient substitutes for— things, rather than the things (refusals, confrontations, submissions, etc.) them selves. This point of view ignores the fact that speaking is itself a kind of action or behavior, and many times is the most important action occurring in a given situation. To cite only one example, the dichotomy between talking it out or fighting it out is mythical; some of the worst fights are "purely verbal." Wittgenstein writes, "To imagine a language means to imagine a form of life."60 He emphasizes that "speaking and writing belong to intercourse with other people."61 And as Austin concurs, to issue some utterances "is to perform the action— an action, perhaps, which one scarcely could per form, at least with so much precision, in any other way."62 Although the detailed explication of the Wittgenstenian- Austinian notion of speech as fundamentally act is yet to be made, the ordinary language view suggests that it is more profitable to ask about the utterance, "The sun is shining today," "What was he doing when he said that?" than "What did his words symbolize or stand for?" While final agree- 60Ludwig Wittgenstein, Philosophical Investigations (Oxford, 1953), para. 19. 6xWittgenstein, The Blue and Brown Books, p. xv. 62J. L. Austin, "Performative-Constative," trans. G. J. Warnock, in Caton, p. 22. 208 nent about whether the speaker was stating, praying, entreating, etc., may not immediately be reached, I submit that the effort to reach agreement will be much more enlightening— and heuristic— than trying to discover and establish the characteristics of any of the following pseudo-entities: the-fact-that-the-sun-is-shining, the proposition, "the sun is shining is true," the quality of "shining," the affective state of understanding-that-the- sun-is-shining, or the disposition-to-respond-with-under- standing-that-the-sun-is-shining. Whereas speech is behavior, language is an abstraction from speech behavior. To abstract from something is "to separate [it] in mental conception; to consider [it] apart from the material embodiment, or from particular instances.116 3 Both the verbal codes and the informal logi cal rules of language are abstractions from speech behavior. Words and punctuation are abstractions from speech behavior in the sense that they are used to reproduce the behavior so that it may be considered apart from any particular instance of someone behaving. The word "the" is thus an abstraction from the spoken / Ba./ or /Bi/ . The words and punctuation of the sentence, "Go home, Bill." are similarly used to repro duce the speaking / goxt hdiro (pause) bil / in order that it may be considered apart from any specific instance of its 63The Oxford English Dictionary, I (Oxford, 1933), 42. 209 utterance. The informal logical rules of language are also abstractions from speech behavior, i.e., they characterize ways of speech-behaving apart from any specific instance of speech behavior. For example, the rule statement "use 'an' instead of 'a' before a word beginning with a vowel" charac terizes a unit of acceptable speech behavior apart from any specific instance of that behavior, i.e., it prescriptively generalizes about speech behavior. Linguistic rules are abstractions from speech behavior, in the same sense that social mores are abstractions from other kinds of interper sonal behavior, i.e., rules of speech behavior come into existence.when elements of the behavior count as right or wrong among fluent speakers of the language.61* This aspect of these rules was described by Kurt Baier when he wrote, "it is only because people generally adopt certain encourag ing attitudes to conformers and discouraging attitudes to non-conformers that we can say that a group has these rules, that these rules are rules of that group."65 As Richard 6Richard Hertz, "Rules and Language: A Philosophical Study of Linguistic Communication," Unpub. diss. (Univ. of Pittsburgh, 1967). Professor Hertz liberally employed the term "linguis tic community" in his study. After extended discussion with him, I have substituted "people who speak the language fluently" as my interpretation of what he meant by "linguis tic community." 65The Moral Point of View (New York, 1965), p. 69. Cited in Hertz, “The Existence of Linguistic Ryles," unpub. ms., p. 3. 210 Hertz writes, A statement of a rule of language will imply that certain modes of speech or of the written use of the language count as correct or incorrect in a linguistic community. That there should be such corresponding regularities of usage serves as an empirical constraint upon the adequacy of any proposed rule-formulation. Statements of rules of language will be judged adequate only on the condition that there are such corresponding regularities of usage in the community.66 Since they are abstracted from speech behavior, these rules must not only reflect regularities of linguistic use, but also be maintained and supported by fluent speakers of the language. That is, fluent speakers support and maintain the rules that are in force by being disposed to react nega tively to infractions of those rules. When an ungrammatical sentence is uttered, other than as an example, quotation, figurative usage, etc., a fluent speaker will tend to cor rect the deviant usage, point out errors to the violator, or otherwise manifest awareness of the deviance. Thus speech is more fundamental than language because speech is behavior whereas language is an abstraction from that behavior. Speech Is Behavior and Language Is Systematic The second reason the language scholar should begin his inquiry with speech instead of language is that if he begins by studying language systems instead of speech,behavior, he 66Hertz, "The Existence of Linguistic Rules," p. 8. --------------------- 2TI is liable to make oversimplified generalizations about human communication. As previously noted, De Saussure argues that the student should focus on language rather than speech, because language is systematic whereas speech is irregular and heterogeneous. For De Saussure the structural unity of langue is one of its most important assets. At one point he asserts that, since the faculty of articulating words can only be exercised with the help of the instrument of lan guage, "it is therefore not fanciful to say that it is langue which gives unity to langage."6 7 De Saussure's focus on the study of language systems led him to the oversimplified notion that what he was study ing was fundamentally a system of symbols. He even argued that the language system "gives unity to speech." Yet most contemporary language scholars complain about the lack of structure or unity in ordinary speech. The struggle to create an ideal language, outlined in part in Chapter III, has been caused in large measure by this lack of unity and consistency in everyday speech behavior. Furthermore, ordinary language philosophers point out that the language scholar should not approach his subject matter with the preconceived idea that he will find mathe matical structure and sharp boundaries between concepts. As Wittgenstein illustrated, he should concentrate his 6 7Pe Saussure, trans. the writer, p. 77.________________ 212 attention on concrete cases of speech behavior. Wittgen stein replaced the search for unity that occupied De Saussure with his concept of family resemblances, which explained that two terms or concepts could be related by "a complicated network of similarities, sometimes overall simi larities, sometimes similarities of detail."68 Austin and other ordinary language philosophers have developed this less-rigid view of the nature of their subject to the point at which we should no longer feel bound, as De Saussure and many others were, to the fallacious search for unity in lan guage . In brief, language scholars usually begin their inquiries by focusing on language. Since they view language as a system— of "signs," "signals," "symbols," etc.— they do not hesitate to draw conclusions that reflect the systematic nature of their subject matter, e.g., "all words are names"; "all words represent ideas"; "all communication involves a sharing of ideas"; "our responses are our meanings; our meanings are our responses." Were they to concentrate on the virtually limitless variety of speech behavior instead of language, they might be much less eager to characterize human behavior with such, sweeping generalizations. I have contended that the most important insight one can gain from an ordinary language philosophy approach to language and meaning is that the language scholar's first 68wittgenstein. Investigations. paras. 66-67.__________ 213 question should not be "What is the nature of language?" but "What is the nature of speech?" Two arguments have sup ported that contention, in the first place, the language scholar's initial questions must inquire into the most fun damental aspect of his subject matter, and speech is more fundamental than language, because speech is behavior, whereas the verbal codes and informal logical rules of lan guage are abstractions from that behavior. In addition, when he begins by studying language systems instead of speech behavior, the language scholar is often led to make oversimplified generalizations about his subject matter. The Problem of Meaning Fundamental changes in the nature of the language scholar's initial subject matter necessitate changes in the rest of his analysis. In.this case, the change in initial subject matter from language to speech results in a change in approach to meaning, from a representational theory to a viewpoint that emphasizes that "meaning" is process rather than product. The direction of and justification for this change can best be understood, first by analyzing the nature of what is termed meaning-in-general, and then by indicating the nature of meaningful speech behavior in terms of that analysis. Meaning-in-general Many non-linguistic objects and occurrences may be said 214 to be meaningful in a clear and uncomplicated sense. The wail of a siren means an ambulance, fire engine, etc., is on an emergency run; certain cries of some animals mean danger; the display of a flag or insignia or trademark can mean that a person is patriotic or a member of a fraternal organiza tion, or that a product is made in Germany. Nonverbal aspects of speech behavior may also be meaningful. The utterance, "That was a wise move," said in a sarcastic tone of voice may mean just the opposite of "what it literally says." In a certain situation, a member of a group who insists that a plan of action is "just fine with me" can mean that he would rather adopt an alternative plan; a child's cry, depending on its tone, volume, etc., can mean that the child is tired, angry, frustrated, or injured. In short, if the possibilities for linguistic meaning seem vir tually limitless, the possibilities for non-linguistic and nonverbal meaning seem even more so. Nonetheless, all these instances of meaningfulness share one very general common characteristic. Dallas Willard explains the characteristic in these words: The problems which are often discussed in terms of "mean ing" all seem to me to arise, in one way or another, out of that everyday occurrence of our looking at or hearing one particular kind of thing . . . and, because of what we then see or hear, thinking of something altogether other than it.69 "Dallas Albert Willard, "Meaning and Universals in Husserl's Logische Untersuchungen," unpub. diss. (Univ. of Wisconsin, 1§64), p. 1. _____________ ________ 215 In other words, some things— words, objects, sounds, etc.— can in some situations carry a person conscious of them beyond them, i.e., lead him to think of something else. To a visitor from the East, a single.wire fence on a Colorado range may be just a single wire fence. Although he may be conscious of the fence, and even of the single wire, he may not be led to think of anything else because of it— he may not "go beyond" it. To a Colorado farmer or rancher, on the other hand, the single wire may mean that the fence is electric, that it is designed to enclose sheep or cattle but not horses, that the ranch he is on is relatively well- equipped , etc . The fence "has meaning" for the rancher because it leads him beyond the fence itself. Similarly, a childless person may simply be conscious of an infant's crying, whereas its mother or another parent can tell whether the cry means that the child is tired, angry, frus trated, or injured. Likewise one may be aware that a totally unfamiliar word or a word in a language totally foreign to him is a linguistic utterance, but unless he knows something about "what the word means," he is not conscious of anything beyond the word itself. In brief, when an object, sound, word, sentence, etc., is meaningful, the person conscious of it is in some way able to "go beyond" it. Elbailp is meaningless at first reading. The reader is conscious of the letters and perhaps even of their "word-like" arrangement. Since the letters 216 occur in a context of verbal discourse, the reader probably assumes they make up a word, and perhaps tries various strategies to make a word out of them— to give them meaning. But until they are meaningful to him, he is stopped with the letters themselves; he does not go beyond them. Only when he is told to reverse the sequence of the letters, or dis covers on his own to do so, can he go beyond the letters to understand "pliable." If this account of the general nature of meaning is correct, then two possible questions about meaningful speech behavior follow from it. If meaning involves the ability to "go beyond" what is meaningful, then an analysis of meaning could start either by asking "Where?" or by asking "How?" one "goes beyond." Some version of the former question has been asked by virtually all past and many present language scholars. By asking, in effect, "Going beyond where?" or "Going beyond to what?" their attention has been focused on discovering entities of some sort that figure prominently in all meaning situations. They have concentrated on looking for things, in the broadest sense, that whatever is viewed as meaningful can lead to. In other wo.:ds, because language scholars have Igiegun by studying the abstract system of lan guage rather than the activities of speech behavior, and because they subsequently have asked where these entities lead, they have virtually all approached language as sym bolic and meaning as involving words representing whatever 217 entities they allegedly symbolize. Wittgenstein recognized the dangers inherent in this approach when he wrote, The questions "What is length?," "What is meaning?," "What is the number one?" etc., produce in us a mental cramp. We feel that we can't point to anything in reply.to them and yet ought to point to something. (We are up against one of the great sources of philosophical bewilderment: a substantive makes us look for a thing that corresponds to 11.) As previously noted, it is the contention of this dis sertation that language scholars have begun their analyses with two incorrect first questions. Instead.of "What is the nature of language?" they should have asked, "What is the nature of speech behavior?" In addition, instead of "What is meaning?" they should have asked some form of the ques tion, "How is speech behavior meaningful?" or, as Willard asked, "What role do the words which I now see or hear play in bringing me to think what I now think, and what is it about these words which makes this role a possible one for them?"71 In other words, I am suggesting that the study of meaning must shift from a search for the referents, desig- nata, ideas, responses, or dispositions to respond that "are" meaning, to a study of how speech is used by people meaningfully, i.e., must shift from a search for the product that allegedly is meaning to a study of the process that meaning is. 70Wittgenstein, The Blue and Brown Books, p. 1. 7lWillard, p. 4. 218 Meaningful Speech Behavior Initially the new approach indicates that words do not "have meaning." Speech behavior can be meaningful or mean ingless, which is to say that one can talk of "the meaning" of a sentence (including, of course, a one-word or ellip tical sentence), but not of a word. As Austin noted, it appears that the-sense in which a word or phrase "has meaning" is derivative from the sense in which a sentence "has a meaning": to say a word or phrase "has a meaning" is to say that there are sentences in which it occurs which "have meanings": and to know the meaning which the word or phrase has is to know the meaning of sentences in which it occurs.72 One can answer the question, "What is the meaning of 'muggy?'" by trying to describe what mugginess is and is not, by giving examples of sentences one might use "muggy" in, or perhaps by getting the questioner to imagine situa tions in which "muggy" might be properly employed. However, those tactics are attempts to answer the question, "What-is- the-meaning-of (the word) 'muggy?'" not "What is the-mean- ing-of-(the-word)-'muggy?'" That is, the tactics for answering the question are used to indicate the conditions under which "muggy" can be meaningful, they are not used to try to isolate something that is "the-meaning-of-(the-word)- 'muggy.'" Under the symbolic view of language, inquiries about 72J. L. Austin, "The Meaning of a , Word," in Philosophi- cal Papers, ed. J. 0. Urmson and G. J. Warnock (Oxford, rS'ST), p .“24. _____________________ ___________ 219 meaning were understood as inquiries about what words sym bolized. So the question, "What is the meaning of (the word) 'muggy?'" would be answered, depending on the theory of meaning held by the respondent, by "the concept of muggi ness," or "the class of sensa of which it is correct to say 'this is muggy,'" or "the responses (or dispositions to respond) made by an organism to the (mediated) stimulus caused by the symbol, 'muggy.'" Under the view suggested here, not only are the mis takes inherent in the above analysis avoided, but the ques tioner also understands the folly of proceeding from a ques tion about the meaning of a given word in a given situation to a question like, "What is meaning?" which is equivalent to asking, "What is the-meaning-of-anything?" That is, the question, "What-is-the-meaning-of (the word) 'meaning?'" has an answer, although it is obviously neither clear nor simple. But the question, "What is 'meaning?'" has no answer at all. In brief, "there is no simple and handy appendage of a word called 'the meaning of (the word) ii jjii i ii 7 3 Given an understanding of the account of meaning sug gested here, the language scholar should recognize that meaningful speech behavior can only be understood by (1) viewing speaking as fundamentally an activity, and 73Austin, "The Meaning of a Word," p. 30. ------------------------------------- zxo (2) studying the entire situation the speech behavior occurs in. One does not speak instead of acting; when he is speak ing he is acting. Since speaking is acting, "the meaning" of speech behavior must involve an account of the act taking place. The politician's hyperbole is not taken to mean what it literally does, because it is an act in the context of other hyperbole-producing political acts. Similarly, "the meaning" of any speech behavior must be determined in the light of what is being done in the situation— what the speaker is doing in speaking and b£ speaking, how what he is doing relates to what others have done before him, how what he does affects his hearers, etc. Like physical behavior, speech behavior can be almost infinitely varied and can have widely different purposes, aimo, and goals. I may raise my right arm in order to stretch it, gain someone's attention, pick an apple, or threaten someone, depending on the situation and my pur poses. Similarly, an act of speaking is not just used to "communicate thoughts" or "exchange ideas" but may (also) be an act of issuing an order, sealing a bargain, christening a ship— or a baby— , worshipping a god, or fortifying one's courage. Thus "the meaning" of speech behavior cannot be just one sort of "thing" or explained in just one way. Meanings in one's speech behavior are as varied as the pur poses of one's physical behavior. 221 J. L. Austin's attempt to classify the two or three basic kinds of speech acts may have failed because of his implicit assumption that natural languages will eventually be comprehensively and systematically accounted for by a carefully structured universal theory of language. That is, Austin's failure may have been due to his belief that ordi nary speech behavior would lend itself to the kind of sys tematization that he envisioned. He apparently overlooked the fact that, because to speak is to behave, what any given unit of speech behavior is.doing is so variable that the same words said in the same tone of voice by the same person to the same person could have significantly different mean ings, depending only on one or two aspects of the situations in which they are uttered. Consider, for instance, the utterance, "I'm finished," said to a classmate on graduation day and to the same person four months later when the speaker has received his army induction papers. In short, any consideration of meaningful speech behav ior must begin with the understanding that speaking is acting and thus (a) it must be judged in terms of its suc cess, how it is affected by previous acts, its effect on subsequent acts, etc., and (b) it is virtually infinitely varied. A brief consideration of how children learn to speak illustrates the validity of viewing speech behavior as fundamentally an activity. With the guidance of elders and 222 ay imitation they learn how to use sounds to do things. They learn, perhaps from television's Romper Room, to per form the act of saluting the flag by assuming a certain posture and uttering certain sounds. At first the sounds are meaningful only as parts of the ceremony, and the child is satisfied with, e.g., "I pledge a legions to the flag of the Unita States of America, and to the public for richard stands. ..." As he is corrected and he begins to under stand "what" the noises mean, he is able to use them in ether situations. He learns, for instance, that "United States of America" is the name of the country he lives in, and, although he often confuses his.country with his state and city, he has begun to learn how to use "United States of America," that is, he has begun to learn "its meaning." Similarly, children learn "the meaning" of "NoI" or "I love you" or "Sit down and eat your dinner," not by discov ering what these utterances stand for or symbolize, but by understanding what their elders are doing when they say them. A child learns "the meaning" of "NoI" when he learns that his parent is warning him, i.e., what he will do if the child persists in his present behavior, not when he learns that "No!" symbolizes an affective state of negation or a conception of negativity in the mind of his parent. In short, the child learns words by finding out how they are used; he understands words by understanding what can be done or is being done with them. He learns from the beginning _ 3 that speaking is a kind of ordinary human activity, and that speech behavior is meaningful by virtue of the activity it figures in. The language scholar should also recognize that mean ingful speech behavior can only be understood by studying the entire situation the behavior occurs in. Both Wittgen stein and Austin stress the importance of the total situa tion to "the meaning" of any given unit of speech behavior. Wittgenstein notes that "A word has meaning.only in a con text," and emphasizes that the context is a living one, and need not include words.7t As Hallett points out, to say that the context involves both verbal and nonverbal aspects of the situation does not yet suggest its full complexity. The context of speech behavior is in a sense determined by the entire culture it occurs in: "Only in the stream of life does a word have meaning."75 Austin makes the same point when he writes, "the total speech act in the total speech situation is the only actual phenomenon which, in the last resort, we are engaged in elucidating."76 In order to understand meaningful speech behavior, one must understand that the total context surrounding the speech act will affect its meaning. An important part of the context or situation is 7* * Hallett, p. 102. 75Hallett, p. 102. 76J. L. Austin, How To .Do Things with Words, ed. J. 0. Urmson (New York, 1965), p. 147. JT4 created by the existence and application of the informal logical rules previously discussed. One can mean the oppo site. of what he says because of the commonly-understood informal logical rules involving "kidding" or sarcasm. Con versely, one may mean literally what he says sarcastically, for instance, because of the flexibility of the rules and tiis right to "break" them. I have every right to understand that Chicago's John Hancock building is the tallest office cuilding in the United States, if you tell me so with au thority, because I understand enough about the informal logical rules of speaking to understand you. Conversely, you may mean that the building is the tallest of its type, or the tallest west of Cleveland, but only if you either rephrase your comment to abide by the rules that determine its meaning, or justify your noncompliance with the appro priate rules. In short, the informal logical rules of speech behavior play an important role in explicating the speech context that determines meaning. The relationships between meaning and informal logical rules can also clarify the nature of meaninglessness. Gen erally, an utterance is meaningless if the person conscious of it cannot somehow go beyond the utterance. Noncorapliance with rules of speech behavior is one way to prevent an utterance from being meaningful. "Twelve he logic tickled juxtapose" obeys virtually no rules of use and is hence meaningless. The reader is stopped by the words, and does 225 not go beyond them. On the other hand, as Strawson notes, one may concoct an explanation of the meaning of "twelve he logic tickled juxtapose" that would, in effect, explain why certain rules were broken or would show how other rules were in fact being obeyed. Then the utterance could conceivably oe meaningful. Like Strawson, Wittgenstein stressed this flexibility of meaning. According to Hallett, he refused to draw sharp boundaries around what could be meaningful and what could not. He said of meaning what he said of games: "How is the concept of a game bounded? What still counts as a game and what no longer does? Can you give a boundary? No. You can draw one, for none so far has been drawn." "One might say that the concept 'game' is a concept with blurred edges." Yet "are we to say that we do not really attach any meaning to this word, because we are not equipped with rules for every possible application of it?" "Many words in this sense then don't have a strict mean ing. But this is not a defect. To think it is would be like saying that the light of my reading lamp is no real light at all because it has no sharp boundary.77 In sum, the language analyst should begin his study of meaning with some version of the question, "How is speech behavior meaningful?" rather than with the question, "What is meaning?" He should understand that speech behavior is an activity, and thus (a) must be judged in terms of its success, effects, etc., and (b) is virtually indefinitely varied. In addition, he should remember that speech behav ior always occurs in a situation, and thus that any analysis 77Hallett, p. 109. 226 of "the meaning" of speech behavior must take into account the total situation, including its physical and human aspects, and the informal logical rules that are applicable. Summary In this chapter rhetoricians’ views of language and meaning were found to be unsound when viewed from the per spective of ordinary language philosophy. In part one, three weaknesses were found in the view that language is fundamentally a system of symbols: (1) it does not explain which language elements can serve as symbols; (2) it cannot successfully account for the nature of apparently non-sym- colic terms? and (3) the assumption grounding the theory of abstraction, which is used to adapt the symbol metaphor to non-symbolic terms, is unsound: all words are not names of cbjects. The referential theory of meaning was adjudged unsound because each meaningful word does not have its own extralinguistic referent, because words themselves can only be used to refer, and because when a word is related to an object or class of objects, it cannot be said to refer to them. Besides suffering from the inadequacies of the referential theory, the ideational theory was unworkable because it depends on one's ability to identify and describe a discrete "idea" for each linguistically meaningful unit of discourse, and because it cannot account for the meaning of image-related words. The behavioral theory was found inade- 227 guate because it limits meaning to response and because utterances do not have different meanings just because dif ferent people react or are disposed to react to them in dif ferent ways. Moreover, Osgood's account of this theory iepends on the existence of "patterns of stimulation," which are as difficult to identify and describe as "ideas." Part two argued that rhetoricians and other language scholars had been led to their mistaken views of language and meaning by asking two inappropriate first questions about their subject matter. First, it was asserted that the most important insight one can gain from an ordinary lan guage philosophy approach to language and meaning is that the scholar's first question should not be "What is the mature of language?" but "What is the nature of speech?" rwo arguments supported that assertion. The scholar's first guestion must inquire into the most fundamental aspect of tiis subject matter, and speech is more fundamental than language in the sense that speech is behavior and language is an abstraction from that behavior. In addition, when he begins by studying language systems instead of speech behav ior, the language scholar is often led to make oversimpli fied generalizations about his subject matter. The focus on speech was then employed in an analysis of meaning. Meaning-in-general was viewed as a function of whatever is meaningful enabling the person conscious of it to "go beyond" the thing itself, i.e., enabling him to think 228 d£ something else because of it. It was argued that rhet oricians and other language scholars had been led astray by bheir proclivity to ask of meaning-in-general, in effect, "Going beyond where?" or "Going beyond to what?" instead of "How does it lead beyond?" and had thus limited their efforts to the inherently impossible task of identifying and olassifying the non-existent "things" they believed words symbolize. Under the alternate approach to meaning, the Language analyst would (1) not expect to find some "simple and handy appendage of a word" that he could accurately call "the meaning of the word"; (2) consistently view speech aehavior as an activity, and consequently (a) evaluate it in terms of its success or efficiency, rather than in terms of tiow effectively it "symbolized" something, and (b) expect and even welcome the virtually limitless variety of possible speech behaviors; and (3) never try to understand meaningful speech behavior except in terms. of the total context it Dccurs in, including the personal and situational variables, and the informal logical rules "in effect" in each circum stance. CHAPTER V SUMMARY AND CONCLUSIONS This study was based on the assumption that presupposi tions about language and linguistic meaning are essential to the foundation of speech-communication theory and pedagogy. It focused on the presuppositions of two groups: rhetori cians, defined as authors of speech journal articles and of public speaking, argumentation, and persuasion texts, and the ordinary language philosophers Ludwig Wittgenstein, Silbert Ryle, J. L. Austin, P. F. Strawson, and William P. Alston. The study attempted to answer three questions: 1. How do rhetoricians in the field of speech-communi cation view language and meaning? 2. How do the ordinary language philosophers view lan guage and meaning? 3. What conclusions may be drawn from an evaluation of the assumptions of rhetoricians in light of the views of ordinary language philosophers? An extensive review of selected speech-communication textbooks and over four hundred issues of speech journals published since 1953 revealed that rhetoricians view lan guage in the following way: Language is fundamentally a 229 230 system of symbols. Since symbols are generally "anything that represents something else,"1 linguistic symbols repre sent other things. These symbols represent other things chiefly by serving as their names. Many words do not seem to name or represent anything. However, the theory of abstraction illustrates how most of these apparently non- symbolic terms do indirectly represent or name some object. h few non-symbolic terms, e.g., interjections, may be dis missed because they are not primarily used to communicate. Rhetoricians tend to base their work on three theories of meaning: the referential, ideational, and behavioral. Each is a representational theory in that it is based on the assumption that the function of linguistic entities, i.e., words, phrases, sentences, etc., is to represent other things, and these other things figure prominently in what the entities mean. One referential theory held that "word meanings are in their referents";2 a more complex version posited that "meaning resides in the relationship between a word and the object it represents."3 Both versions were developed, in part, by explanations of denotative meaning, which was held to be "exact, literal, and objective" or IMilton Dickens, Speech: Dynamic Communication (New York, 1963), p. 371. . 2Martin Andersen, Wesley Lewis, and James Murray, The Speaker and His Audience (New York, 1964), pp. 89-:90. 3Jane Blankenship, Public Speaking: A Rhetorical Per spective (Englewood Cliffs, 1966), p. 105. 231 "sign-object" meaning, and connotative meaning, which was viewed as "subjective, emotive, and personal" or involving sign, object, and person. The.ideational theory of meaning held that "words are symbols for meanings since they stand for ideas., , , f The mental entities words represent were also called "concepts," "conceptions," "need-want bonds," and "categories." There were several versions of the behavioral theory; all held that although words have referents, meaning is a function of behavioral response to linguistic stimuli. Sometimes the theory was expressed simply; "Our responses are our meanings; our meanings are our responses.115 Other expressions of the theory were more complex; "'to mean' refers to making a coded central process (rMs) specifying what the self is to do about a referent."6 An account of the history of linguistic philosophy emphasized that philosophers from Aristotle to the logical positivists had viewed language as a highly structured sys tem of symbols and words as fundamentally names. Current ordinary language philosophers contend that language is not as systematic as many previous philosophers thought. Since language is not mathematically regular, it is a mistake to ^John P. Wilson and Carroll C. Arnold, Public Speaking as a Liberal Art (Boston, 1964), p. 299. sDonald C. Bryant and Karl R. Wallace, Fundament als of Public Speaking, 3rd ed. (New York, 1960), p. 39. 6Walter Coutu, "An Operational Definition of Meaning," Quarterly Journal of Speech, XLVIII (February 1962), 62. 232 assume that words function in any single way; words are not simply names; meaning is not simply a matter of referring. In addition, virtually all generalizations about.language are oversimplified and hence misleading. The use of ordi nary language can thus most accurately and profitably be viewed as a kind of human behavior; the form and uses of language are inexhaustibly.flexible and various. When they "do philosophy," ordinary language philoso phers systematically, examine "the informal logic of the functionings of ordinary expressions." By "ordinary" they mean "common," or "natural," or "on the tongue of Every man."7 The "informal logic" of these expressions limits what can or cannot be meaningfully said in ordinary language by means of flexible, contextrrelated rules that may be violated with cause. The informal logical powers of ordi nary terms "can only be extracted by examining the dealings people transact with them."8 The rhetoricians' concepts of the nature of language were then evaluated from the perspective of ordinary lan guage philosophy. The assumption that language is funda mentally a system of symbols was found to be unsound for three reasons. First, the symbol metaphor does not consis- 7Gilbert Ryle, "Ordinary Language," Philosophy and Or dinary Language, ed. Charles E. Caton (Urbana, 1963), pp. 108-109. 8Gilbert Ryle, "Formal and Informal Logic," Dilemmas (Cambridge, 1966), p. 123. 233 tently explain which linguistic entities allegedly function as symbols. Sometimes it is argued that words are symbolic; in other cases statements, utterances, or sound patterns are held to represent "other things." Besides this inconsis tency, the symbol metaphor cannot account for the nature of many apparently non-symbolic terms. Even the proponents of this assumption often experience difficulties explaining the nature of words like "when," "of," "honor," and "semantics" in terms of the symbol metaphor. In order to account for the nature of these apparently non-symbolic terms, most rhetoricians turn to the theory of abstraction. The third reason for the rejection of this.view is that the abstrac tion theory is unsound and unworkable, because it is based on the incorrect assumption that all words ultimately represent objects. The referential theory of meaning is also unsound when viewed from the perspective of ordinary language philosophy. The notion that the meaning of a word is what it refers to cannot be valid, because many words with different meanings may refer to the same thing, and many things may serve as "referents" for the same word. If the meaning of a proper noun were actually the person to whom it refers, the meaning would die with the person. As for verbs, it is impossible to determine the number of entities that would have to be created in order for a verb like "run" to have something to refer to each time it is used in a different sense. Syn- 234 categorematic terms can only be explained under this theory ay creating entities like "conjunctive functions," which cannot be shown to be extralinguistic. Even if such enti ties were accepted, the words that could be said to be related to them cannot be said to refer to them, i.e., one cannot refer to "the class of pencils" simply by saying "pencil." From another point of view, words themselves can not refer; they can only be used, in some instances, to do so. Finally, rhetoricians' accounts of connotation and denotation manifested a confusion between logical connota tion and literary ambiguity. The ideational theory of meaning, which is also held by many rhetoricians, cannot successfully account for lin guistic meaning for three reasons. In the first place it suffers from all the shortcomings of the referential theory. In addition, the term "idea" is vague enough to allow the theory to be used to explain everything: "No matter what word we dredge up, if we ask what its meaning is or to what it refers, we can say, 'the idea or concept of X.'"9 The theory is also virtually impossible to apply, because it depends on one's ability to identify an "idea" that corre sponds with each linguistically meaningful entity one utters or hears. The behavioral theory of meaning is mistaken, because 9Robert F. Terwilliger, Meaning and Mind: Ti Study in the Psychology of Language (New York, 1968), p. l5l. 235 it also posits that words are signs or symbols, and thus inherits the weaknesses of the referential account. In addition, the contention that meaning is response is invalid because, for instance, the meaning of the imperative "sit down" does not depend on the dozens of potentially different responses different people in various situations might make to it. Charles Morris' substitution of "dispositions to respond" for "responses" complicates the theory, but does not make it any more valid. . ..Finally, Charles Osgood's account of meaning in terms of.the mediation hypothesis posits the existence of "patterns of stimulation," which are potentially more difficult to isolate, identify, and de scribe than "ideas." The most important insight one can gain from an ordi nary language philosophy approach to language and meaning is this: instead of beginning his inquiry by studying lan guage, the language scholar should begin by focusing on speech. He should do so for two reasons. First, if the ories of rhetoric, like theories of semantics and general semantics, are viewed as philosophical systems, having a set of basic assumptions and a means by which the assumptions can be consummated, then the rhetorician, semanticist, gen eral semanticist, and philosopher of language must initially be dedicated to the systematic critique of fundamental pre suppositions. The most fundamental presuppositions underly ing all the aspects of language and its use are assumptions 236 about speech, not language. Speech is more fundamental than language in the sense that speech is meaningful human oral oehavior. governed by informal logical rules, whereas lan guage is a system of verbal codes and the informal logical rules that govern speech behavior, which is abstracted from 3peech. Second, when the language scholar begins his study with language systems rather than speech.behavior, he is liable to make oversimplified generalizations about human communication. The change in initial subject matter from language to speech resulted in a change in approach to meaning from a representational theory to a viewpoint that meaning is a process rather than a product. Linguistic meaning was explained by beginning with what was termed meaning-in- general, which is a function of whatever is meaningful enabling the person conscious of it to "go beyond" the thing itself, i.e., enabling him to think of something else because of it. According to this account, an object, sound, word, etc., is meaningful to someone if, when he perceives it he is led to think of something altogether different from the meaningful object itself. The second inappropriate question occurred when rhetoricians and other language scholars asked of meaning-in-general, in effect, "What is thought of?" instead of "How does the meaningful object make someone think of something else?" The inappropriate ques tion forced language scholars to undertake the impossible 237 task of identifying and classifying the non-existent "whats" — referents, designata, etc.— they believed words symbolize or lead people to think of. Under the approach to meaning that began with a "How" question, language scholars would (1) not expect to find some "simple and handy appendage of a word" that is its meaning. (2) Consistently view speech behavior as an activity, and consequently (a) evaluate it in terms of its success or efficiency, rather than in terms of how effectively it "symbolized" something, and (b) expect and even welcome the virtually limitless variety of possible speech behaviors. (3) Never try to understand meaningful speech behavior except in terms of the total context it occurs in, including the personal and situational variables, and the informal logical rules in effect in each circum stance. Implications This study demonstrated the need for rhetoricians to recognize that, when they treat the subjects of language and meaning in both scholarly and pedagogical works, they are dealing with philosophical matters and thus should at least be aware of ordinary language philosophy. Their works, being both philosophical and public, are open to critical assessment. One would expect such works to be able to with stand appropriate criticism. However., rhetoricians, as defined in this study, are almost completely unaware of 238 ardinary language analyses of language and meaning and appear to ignore the need to adopt a consistent approach to these subjects. The contention that rhetoricians ought to be aware of advances in knowledge made in related fields has often been made. A Speech Teacher editor counseled his subscribers to "read what is being written relating to our subject area by athers than those active in our subject field,"10 and H. H. Perritt wrote, "we must read the [linguistics] publications; there is little evidence in our own literature that many of as are doing so."11 The same could be said of the litera ture of ordinary language philosophy. Eight of the ten sources rhetoricians cited most often in their accounts of language and meaning were published before the ordinary lan guage approach was widespread. Heavy reliance is still placed on Ogden's and Richards' work, first published in 1923, Korzybski's, first published in 1933, and the treat ments of general semantics, first published between 1938 and 1946. Many of the rhetoricians' accounts of language and meaning were also internally inconsistent. Those who 10Donald H. Ecroyd, "Book Reviews," Speech Teacher, VII (March 1958), 348. 1XH. Hardy Perritt, "Linguistics: A Lambent Glance," Southern Speech Journal, . XXVI (Summer 1961), 280. Cf. Samuel L. Becker, "Research on Emotional and Logical Proofs," Southern Speech Journal, XXVIII (Spring 1963), 198. : ------------------------------------- 2T? recognized the fundamental importance,of language and mean ing also often realized that the symbol metaphor and repre sentational approach to meaning could not adequately account for the wide variety of possible speech behaviors. Thus they often presented a brief account of language and meaning from several points of view, citing, for instance, Korzyb- ski, Osgood, Ogden and Richards, Hayakawa, Berio, Langer, Ullmann, and others, in an apparent effort to surround their subject. Unfortunately, the result was, in some instances, the worst kind of eclecticism, a potpourri of unrelated the ories, some trivial and some profound. As P.F. Strawson commented, there is nothing morally wrong with.inconsistency: . . . a man who contradicts himself may have succeeded in exercising, his vocal chords. But from the point of view of imparting information, of communicating facts (or falsehoods) it is as if he had never opened his mouth. He utters words, but does not say anything. . . .He arouses expectations which he does not fulfil. . . .1Z A more desirable goal would be the consistent development of the most valid assumptions available about one's subject matter. If the rhetorician would practice identifying assump tions and consistently developing rhetorical and pedagogical theories from them, he would also be more able to recognize the relevance to speech-communication of developments in 12P. P. Strawson, Introduction to Logical Theory (Lon- .don, 1953), pp. 2-3. 240 other areas of study. For instance, the implications of existentialism and phenomenology for rhetorical theory and pedagogy might well be clarified if the rhetorician under stood the presuppositions about man, reality, value, lan guage, and meaning that underlie those philosophies. Simi lar knowledge about B. F. Skinner's presuppositions about verbal behavior could enable the rhetorician to determine more validly and effectively the extent to which Skinner's theories might be applied to speech-communication. A clear understanding of Kenneth Burke's assumptions about man, society, language, and meaning.is necessary in order to com prehend his approaches to rhetoric and poetic. In brief, a more philosophical orientation, i.e., one more concerned with the systematic critique of presuppositions, would enable the rhetorician to develop sounder theory and peda gogy and to utilize more effectively the insights of schol ars in other fields. The views of speech behavior and meaning presented in Chapter IV have many potentially significant implications for rhetorical theory and.pedagogy? only a few will be men tioned here. In the first place, the rhetorical theorist's attention should be initially focused on "the informal logic of the functioning of expressions," or more specifically on the informal logical rules that govern the functioning of expressions. Descriptive linguists also focus on linguistic rules, but from a significantly different point of view. 241 These linguists view the rule-governed aspect of lan guage as its most pervasive and significant characteristic. According to Katz, the linguist's theory of language con sists of three sub-theories, each of which is discussed in berms of the linguistic rules it explains: The phonological component is a statement of the rules by which a speaker deals with the speech sounds of his lan guage; the syntactic component is a statement of the rules by which he organizes such sounds into sentential struc tures; and the semantic component is a statement of the rules by which he interprets sentences as meaningful messages.13 Chomsky's linguistic system is also based on.transforma tional rules, an understanding of which facilitates the inscription of semantically significant grammatical rela tions that underlie the phonetic form of sentences.1 * * Descriptive linguists even define communication in terms of linguistic rules. As Katz asserts, "It seems necessary to * conclude that speakers of a natural language communicate with each other in their language because each possesses essentially the same system of rules."15 Although the account presented in this study also emphasizes that rules are basic to the nature of speech 13Jerrold J. Katz, Philosophy of Language (New York, 1960), p. 111. Also in Katz, "The Relevance of Linguistics to Philosophy," The Journal of Philosophy, LXII (October 21, 1965),,219. 1HSee Katz, "Relevancy," p. 596; and Noam Chomsky, Aspects of the Theory of Syntax (Cambridge, Mass., 1957). 15Philosophy, p. 102. 242 behavior, the view of such rules presented here is funda mentally different from that offered by descriptive lin guists. The basic point of disagreement, as Katz ably points out,16 is on the question of how systematic speech behavior actually is. Katz apparently believes that a care fully structured, universal code of linguistic rules will eventually be developed, which will enable one to explain all aspects of all kinds of speaking in all linguistic com munities and situations. The view presented here is not the alternative Katz suggests, i.e., that "natural languages are highly unstructured and unsystematic conglomerations of ver bal constructions." However, it is argued that human speech Dehavior cannot be wholly codified, and thus that all aspects of all possible uses of ordinary speech will never ce adequately represented in a single, monolithic theoreti cal system.17 In other words, this study suggests that the theorist should focus on rules, but on the informal rules of speech behavior, not on the descriptive linguist's rules of language. A second implication concerns the role in rhetorical theory of the "rhetorical (communication, speech) situa tion." Rhetoricians have generally noted the important 16Philosophy, p. 16ff. 17Garth Hallett discusses Wittgenstein's view of the sxtent to which language is actually systematic in Wittgen stein's Definition of Meaning As Use (New York, 1967*)” p. 79ff. 1 1 _________________________ 243 effect the "occasion" has on communication. However, the tendency has been to emphasize how situational variables can affect the prepared message, i.e., how one should adapt to the situation. If speaking is viewed fundamentally as ordi nary behavior, as this study suggests, then the total, liv ing communication context or situation will be viewed as that which not just affects by determines speech behavior. Lloyd Bitzer makes a similar point when he writes, "rhet orical discourse, I shall argue, does obtain its character- as-rhetorical from the situation which generates it."18 The constituents of Bitzer's "situation" are "persons, events, objects, relations, and an exigence which strongly invites utterance."19 The account presented here emphasizes the informal logical rules in effect (which might be part of Bitzer's "relations"), and attempts to outline their nature and function. "Situation" in this study is made up of peo ple behaving ordinarily by uttering meaningful statements in accordance with informal logical rules. Insofar as they are made up of interrelated acts, and insofar as they are governed by informal logical rules, speech situations might profitably be viewed as "speech games." The term "game" is not meant to imply that these sets of rule-governed acts are trivial or insignificant, any 18Lloyd F. Bitzer, "The Rhetorical Situation," Philos ophy and Rhetoric, I (January 1968), 3. 19Bitzer, p. 5. 244 more than are the games of Berne20 or Rapoport.21 "Speech games" should be subject to identification and description, or at least to general characterization. That is, any "communication situation" should be describable by identifying the behavior and the rules that make up the speech game. The theorist should find that most speech games are not at all unusual; they are in many ways similar to Wittgenstein's description: We can easily imagine people amusing themselves in a field by.playing with a ball so as to start various existing games, but playing many without finishing them and in between throwing the ball aimlessly into the air, chasing one another with the ball and bombarding one another for a j oke and so on . . . . . . ► is there not also the case, where we play and— make up the rules as we go along? And there is even one . where we alter them— as we go along.22 The communicator should be as completely aware of the speech games being played as possible. His preparation for communication might include a review of the speech games he characteristically plays in situations similar to the one he is preparing for (gained, perhaps, from sessions utilizing some of the techniques of sensitivity training), and an analysis of the speech games the other communicators have . 20See, e.g., his Transactional Analysis in Psychothera py (New York, 1961). . 21See, e.g., his Fights, Games, and Debates (Ann Arbor, 1960); and Two Person Game Theory: The Essential Ideas (Ann Arbor, 19667^ % . 22Ludwig Wittgenstein, Philosophical Investigations (Oxford, 1953), para. 83. 245 played and are likely to play in this situation. His preparation should go beyond the traditional concerns of "audience analysis" to the point where he is able phenomeno- logically to put himself in the place of the other players of the game, i.e., to understand their point of view, "move" options, etc., without evaluating them.23 J. L. Aranguren sees communication as a competitive game in which each opponent has to put himself in the place of the other so as to decide what to do himself.. He must imagine his adversary attacking at pne point, or pretending to do so and then really attacking in another, and then try to defend himself adequately at the same time anticipating his actions. For only by putting himself in his oppo nent's place can he anticipate him and in due course make a fitting reply,. 24 Speech game analysis could also shed light on the pro cess of persuasion. If one intends to persuade other com municators, he might well begin by planning his communica tion so he and the others play, as nearly as possible, identical speech games. An attorney, for instance, should base his preparation in part on his knowledge of the speech games that are played in a court of law. Different persua sion games, requiring different arguments and different proofs are played, e.g., in the college debate tournament., the classroom, and the office. . - 23Carl R. Rogers and F. J. Roethlisberger, "Barriers and Gateways to Communication,1 1 Harvard Business Review, XXX (July-August, 1952), 46-53. 2‘ 'Human Communication, trans. Frances Partridge (New York, 1967), p. 16. 246 Lawrence Rosenfield recently suggested that the game image "needs to be taken seriously as highlighting signifi cant variables and relationships which demand investiga tion," and that it is "the most insightful model of Aris totelian rhetoric I have yet encountered.1,25 In this writer's .opinion, viewing communication situations as speech games, can perhaps redefine some old rhetorical concepts and generate new information about and/or insight into communi cation. The distinction between speech and language drawn in this study, i.e., between meaningful human oral utterance governed by informal logical rules, and the system of lin guistic entities and rules for their combination abstracted from speech behavior, suggests that spoken persuasion might profitably be viewed apart from written persuasion. Differ ences between speech behavior and language suggest that efforts to persuade by speaking are different in-kind from efforts to persuade in language. This distinction might also provide the foundation for an improved differentiation between oral and written style that takes into account more than just the relative number of contractions and personal pronouns, length of sentences, etc. More fundamental dif ferences, are suggested involving, e.g., varying degrees of • . 25Lawrence W. Rosenfield, "A Game Model of Human Com munication," '.'What Rhetoric. (Communication Theory) Is Appro priate for Contemporary Speech Communication?" ed. David H. Smith (Minneapolis, 1969), p. 28. 247 systematization and differences in applicability of informal logical rules. The views of speech -behavior and meaning presented in Chapter IV also have implications for speech-communication pedagogy. The concept of speech games has import in the classroom, where few students.ordinarily consider.the multi tude of factors that can. affect a speech game and how it is "played." Briefly, a unit or course organized around this principle might have three goals: (1) to familiarize the student with relevant communication variables and their effects on speech games; (2) to develop the student's abil ity, to perceive the games and roles he commonly plays, and to self-reflect on his role and the progress of the game he is in at any given instant; and (3) to develop the student's ability to recognize speech games that are being played by others, and to assess how he can most effectively "join in." Such a class could also utilize the technique of role-play ing, to illustrate how various role and situation permuta tions affect a speech game. Perhaps even more importantly, the real situation that is the class itself could be con stantly analyzed and used in evaluating student communica tion . Several changes might also be made in beginning text books. More texts might begin, as do Ray E. Nadeau's26 and . 26Ray E. Nadeau, A Basic Rhetoric of Speech-Gommunica-, tion (Reading, Mass., 1^69) 248 James C. McCroskey's,27 with chapters dealing with the nature of language, meaning, and communication. However, the treatments of these subjects hopefully would not be cased only on the accounts of semanticists and general semanticists. As already mentioned, text authors might also recognize the desirability of establishing a sound philo sophical position and developing a consistent treatment of speech-communication from it.28 In addition, if more textbook authors were familiar with ordinary language methodology, they might be less likely to use the metaphysically problematic word "idea" as freely as they now do.29 They would also be less likely to assert that the denotative meaning of "pencil" is "that which writes"; that the "meaning" of the name "Robert E. Lee" is the Confederate general who surrendered at Appomat tox; that "language and gesture are simply the vehicles 27James C. McCroskey, An Introduction to Rhetorical lommunication (Englewood Cliffs, 1968). 28For instance, Virgil L. Baker and Ralph T. Eubanks' Speech in Personal and Public Affairs (New York, 1965), one cf the few texts following this pattern, begins with "the assumption that the art of oral discourse is the basic civilizing art of man," develops it along the lines of Richard Weaver's "advisory rhetoric," and explicates it in chapters dealing with, e.g., "Speech as a Civilizing Force," and "Kinds of Actualizing Detail." 29In a sample of five of the twenty-three basic texts examined for this study, fifteen chapter headings included the word "idea," eJg., as "Creating Ideas," "Discovering Ideas," "Managing Ideas," "Forming Ideas," "Adjusting Ideas to Audiences," "Delivering Ideas," etc. . • » 249 which carry our ideas"; or that "our meanings are our responses; our responses are our meanings." The speech behavior-language distinction and speech game analysis also have implications for the theory and practice of rhetorical criticism. If the former is valid, the critic should be able to specify how evaluating a speech that he experienced when it was originally uttered differs from evaluating one he has only read and read about. A speech game assessment of a rhetorical act would necessarily include an examination of all of the factors identified, for instance, by Rosenfield (Speaker, Message, Environment, and Critic),30 plus a consideration of the informal logical rules operative in the rhetorical situation. Anything less sould not be considered a full criticism of the speech aehavior. In brief, the implications of adopting an ordinary lan guage philosophy view of language and meaning are poten tially far-reaching. If the rhetorician's view of the aature of language and meaning is fundamental to everything ae writes and teaches about speech-communication, then a change in this view must result, as Austin pointed out, in a change in the rest of his philosophical— and rhetorical— position. ■ 30Lawrence Rosenfield, "The Anatomy of Critical Dis- [?p,urm>£r4S]g^^^ , 50-70,.,, - - ..- - _ . . . . - 2501 Suggestions for Future Research 1. What is the relationship between speech behavior and language, i.e., how valid are the views outlined in Chapter IV of this study? a. Will the distinction stand additional philosophi cal analysis? b. Are there empirical procedures which will validate the distinction? 2. What is the nature and function of the informal logi cal rules of speech behavior? a. How do they come into being? b. To what extent are exceptions to the rules limited? c. How are such rules similar to and different from transformational-generative grammatical rules? 3. What is the basic unit of speech behavior? a. Can it be utilized in a new approach to content analysis? b. Can it be utilized by the rhetorical critic? 4. What are the parameters of the "speech game?" a. What situational variables influence speech behav ior? b. How do they influence it? c. How can speech games be identified and described? 5. What vocabulary most effectively expresses the in sights gained from viewing meaning as a process r iw"''-•»vrr.‘‘ * 251 instead of a product? 6. What are the relationships between an ordinary lan guage approach to.language and meaning and approaches suggested by: a. Eric Berne's Transactional Psychiatry? b. Anatol Rapoport's game theory? c. Kenneth Burke's view of language and meaning?31 (1) How is Burke's method of analysis related to the ordinary language philosopher's? (2) What relationships exist between Burke's view of language as symbolic action and the speech game analysis outlined in this study? (3) How is Burke's analysis related to J. L. Austin's views? 252 A LIST OF WORKS CONSULTED Books Alston, William P. Philosophy of Language. Englewood Cliffs: Prentice-Hall, 1964. , and George Nakhnikian, eds. Readings in Twen tieth-Century Philosophy. New YorJcl The Free Press, 1963. Andersen, Martin'P., Wesley Lewis, and James Murray. The Speaker and His Audience. New York: Harper and Row , T ^ ' 4' . ---■ --- Aranguren, J; L. Human Communication, trans. Frances Part ridge v NeiFYoBc7"'''HMcG^ , 1967. Aristotle. TheNicomachean Ethics, trans. H. Rackham. Lon don: William Heinemann, 1926. _. The Works of Aristotle, trans. W. D. Ross. Oxford: Clarendon Press, 1928. Austin, J. L. How To.Do Things with Words, ed. J. 0. Urm- son. New York: Oxford University Press, 1955. / . Philosophical Papers, ed. J. 0. Urmson and Gv J;:Warnockv Oxford: Clarendon Press, 1961. . Sense and Sensibilia, reconstructed from manuscript notes;by G.: J> Warnock. New York: Oxford University Press, 1964. Ayer, A. J., etal. The Revolutionin Philosophy. London: Macmillan and Co., 1957. : ; Baird, A. Craigs Rhetoric: A Philosophical Inquiry. New York: Th^RonaldPress, 19<>3. “ , andFranklin H. Knower. General Speech: An Introductionj 3rd ed. New York: McGraw-Hill, 1963, 253 257 Baird, A. Craig. Essentials of General Speech, 3rd ed. New York: McGraw-Hill, 1968. Bake*# Virgil L., and Ralph T. Eubanks. Speech in Personal and Public Affairs* New York: David McKay, 1965. Berne, Eric« Transactional Analysis in Psychotherapy. New York: Grove Press, 1961. . ■ . . ; . ; ■ : ■ . . Black, Max. Philosophical Analysis: A Collection of Essays. Ithaca, New York: Cornell University Press, 1950. Blankenship, Jane. Public Speaking: A Rhetorical Perspec tive. Englewood Cliffs: Prentice-Hall, 1966. . A Sense of Style: An Introduction to Style for the Public Speaker. Belmont, Calif.: Dickenson Publishing Co., 1968. Brigance, William Norwood. Speech: Its Techniques and Dis ciplines in a Free Society, 2nd ed. New York: Appleton-Century-Crofts, 1961. Brown, Charles T., and Charles Van Riper. Speech and Man. Englewood Cliffs: Prentice-Hall, 1966. Bryant, Donald C., and Karl R. Wallace. Fundamentals' of Public Speaking, 3rd ed. New York: Appleton- Century-Crof ts, 1960. ___________. Fundamentals of Public Speaking, 4th ed. New York: Appleton-Century-Crofts, 1969. Buehl^r, E. Christian, and Wil A. Linkugel. Speech: A First Course. New York: Harper and Brothers, 1962. . Speech Conaunication: & First Course. New York: . Harper and: Row, 1969. Zapp, Glenn R., and TJhelma Robuck Capp. Principles of Argu mentation andqDebate. Englewood cliffs: Prentice- Hall, 196S. f Caton, Charles E., ed. Philosophy and.Ordinary Language. Urbana, 111.: Universitv^of^jVLI^Lnors Press^>1963. Berio, David K. The Process of Communication: An Introduc tlon to Theory and Practlo" “ “ 9 ^ ? hart, and winston,l960. /8 ) 255 Chappell, V. C., ed. ^OrdinaryogLanguage^^Essays-ran^Philo-7. . •v/toi.sophicaloMe.thod. 0 " ' 0 Eriglewood Cliffs: Prentice-Hall, 1964. Chase, Stuart. The Tyranny of Words. New Yorks Harcourt, Brace, and Co., 1938. . Power, of Words. New Yorks Harcourt, Brace, and Co., 1954. T ; Condon, John C., Jr. Semantics and£Communication. New York: The Macmillan Co., 1^66. De Saussure, Ferdinand. Cours de Linguistique Gdndrale, pubiie par Ch. Bally et A. Sechehaye avec la col-- laboration de A. Riedlinger. Pariss Payot, 1960. Course in General Linguistics, ed. Charles Bally and Albert Sechehaye in collaboration with Albert Reidlinger, trans. Wade Baskin. Londons Peter Owen Ltd., 1960. Dickens, Milton. Speechs Dynamic Communication, 2nd ed. New Yorks Harcourt, Brace, and World, 1963. Flesch, Rudolf. The Art of Plain Talk. New Yorks Harper and Bros., 1946. Flew, Anthony G. N., ed. Logic and Languages First Series, Oxfords Basil Blackwell, 1961. Logic and Languages Second Series. Oxford, Basil Blackwell, 1966. Fodor, Jerry, and Jerrold J. Katz. The Structure of Lan guages . Readings in the Philosophy of Language. Englewood cliffss Prentice-Hall, 1964. . . . . . . ■ Fotheringham, Wallace C. Perspectives on Persuasion. Bos- tons. Allyn and Bacon, 1966. Sellner, Ernest. Words and Things: A Critical Account of Linguistic Philosophy and a Study in Ideology. Lon dons Victor Go1lanez Ltd., 1959. r ““' Gray, Giles Wilkeson, and Waldo W. Braden. Public Speakings Principles and Practice, 2nd ed. New Yorks Harper and Row, 1963. Hallett, Garth, S. J. Wittgenstein's Definition of Meaning as Use. New Yorks Fordham University Press, 1967. 256 Hartnack, Justus. Wittgenstein and Modern Philosophy, . ^ 0 trans. Maurice Cranston. «New, York: New York^'Uni- 1 : 6 ver'sity^ Press f *1965. Hance, Kenneth G., David C. Ralph, and Milton J. Wiksell. Principles of Speaking, 2nd ed. Belmont, Calif.: Wadsworth Publishing Co., 1969. Hayakawa, S. I. Language,in.Thought and Action. New York: Harcourt, Brace, and Co., 1949. : : Symbol, Status', and Personality. New York: Harcourt, Brace, and' World, 1463. , ed. The Use and-Misuse of Language. Green wich , Conn.: Fawcett Publications, i$bb. Jespersen, Otto;. Mankind, Nation and Individual from a Lin guistic Point of; View. Bloomington, Indiana: Indi ana University Press, 1946. Johnson, Wendell. People in Quandaries: The Semantics of Personal Adjustment. New York: Harper and Bros., 19- 46. ---- Katz, JerroldJ. ^ The Philosophy of Language. New York: Harper and Row, 1966. Korzybski, Alfred. Science and Sanity: An Introduction to Non-Aristotelian Systems and General Semantics, 3rd ed. Lakeville, Conn.: The International Non-Aris- totelian Library Publishing Co., 1948. Langer, Susanne K. Philosophy in a New Key: A Study in the Symbolism of Reason, Rifce*' and Art, 2nd ed. New fork: The New American Library, 1951. Lee, Irving J. Language Habits.in Human Affairs: An intro duction to General Semantics. New York: Harper and Brothers, I94i~ • Linsky, Leonard, ed> Semantics and the Philosophy of Lan- e. Urbana, 111.: University of Illinois Press, Locke, John. An Essay Concerning Humane Under Standing, in four books. London: Elizabeth Holt, for Thomas Basset, 1690. Malcom, Norman. Ludwig Wittgenstein: A Memoir. London: Oxford University Press, 1958. £5 / tfarsh, Patrick p., r > aPersuaslVe^Speaking; Theory/ Models, 4) -Practice. New York: Harper and Row, 1967. McBath, James H., ed. Argumentation and Debate, rev. ed. New York: Holt, Rinehart, and Winston, 1963. McBurney, James I!., and Ernest J. Wr age• Guide to Good Speech, 2nd ed. Englewood Cliffs: Prentice-Hall, flcCroskey, James C. An Introduction to Rhetorical Communi cation: The Theory and Practice of Public Speaking. Englewood cliffs: Prentice-Hall, 1968. Mehta> Ved. Fly and the FlyBottle: Encounters with Brit- ish Intellectuals. Baltimore: Penguin Books, 1965. Miller, Gerald R., and Thomas R. Nilsen, eds. Perspectives on Argumentation. Chicago: Scott, Foresman and Co., 1966. Mills, Glen E. Reason in Controversy: An Introduction to General Argumentation. Boston: Allyn and Bacon, T5&T.----- ------------ Minnick, Wayne C. The Art of Persuasion, 2nd ed. Boston: Houghton Mifflin Co., 1968. Monroe, Alan H., and Douglas Ehninger. Principles and Types of Speech, 6th ed. Glenview, 111.: Scott, Fores man, and Co., 1967. Moulton, Eugene R. The Dynamics of Debate. New York: Har court, Brace, and World, Inc., 1966. Ogden, C. K., and I;.A. Richards. The Meaning of Meaning: A Study of the Influence of.Language upon Thought and of the Science of Symbolism, 8th ed. New York: Harcourt, Brace, and world, i946.; Oliver, Robert T., and Rupert L. Cortright. Effective Speecht 4th ed. New York:- Holt, Rinehart, and Winston, 1961. Osgood, Charles E. Method and Theory, in Experimental Psy chology. New York: Oxford University Press, 1963. Passmore, John. A Hundred Years of Philosophy, rev. ed. New York:" Basic Books, 1966. Pitcher, George. The Philosophy of Wittgenstein. Englewood Cliffs: Prentice-Hall. 1964. 238 Pole, David. The Later Philosophy of Wittgenstein. London: The Athlone Press, 1958. RichardsA. The Philosophy of Rhetoric. New York: Oxforcl University Press, 1965. r* o ®oS> o 0 © Rorty, Richard, ed. The Linguistic Turn: -Recent^Essays^in° - Philosophical Method.' Chicago: The University of Chicago Press, 1967. loss, Raymond S. Speech Communication: Fundamentals and Practice. Englewood Cliffs: Prentice-Hall, 1965. Ryle, Gilbert. The Concept of Mind. New York: Barnes and Noble, 1949. “ .Dilemmas. Cambridge:. The University Press, ------ 1556. ------- Sarett, Alma Johnson,.Lew Sarett, and William Trufant Pos ter. Basic Principles of Speech, 4th ed. Boston: Houghton Mifflin Co., 1966. Scheidel, Thomas M. Persuasive Speaking. Glenview, 111.: Scott, Foresman,: and Co. , '1967. Smith, Raymond G. Principles_of Public Speaking. New York: Ronald Press Co., 1958. Strawson, P. F. Introduction to Logical Theory. London: Methuen and Co,. |^Ltdi, 1952. — Terwilliger, Robert F. Meaning and Mind: A Study in the Psychology of Language. New York: Oxford Univer sity Press> 1 1 Ullmann. Stephen. The Principles of Semantics, 2nd ed. London: Basil Blackwell, 1957. Urmson, J. 0; Philosophical Analysis: Its Development Between the Two World Warsi Oxford: The Clarendon Press, 1956. Walter, Otis M., and Robert L. Scott. Thinking and Speak ing, 2nd ed. New York: The Macmillan Co., 1968. Warnock, G. J. English Philosophy Since 1900. New York: G. J. English Philosophy Sin Oxford University Press, 1966. Weaver, Andrew Thomas, and Ordean Gerhard Ness.- The Funda mentals and Forms of Speech, rev. ed. New York: ______ Odyssey Press, 1963. * __________ 259 Wilson, John. Thinking with Concepts. Cambridge: The Uni versity Press, 1966. i^ilson, John F., and Carroll C. Arnold. Public Speaking as t a Liberal Art, 2nd ed. Boston: Allyn and Bacon, *>*1968, Windes, Russs el0 R^. and Arthur'Hastings. Argumentation and Advocacy^, , , New York: Random House, 1965. ~ 0 o O' Wittgenstein, Ludwig. "Philosophical Investigations, trans. G. E. M. Anscombes^t^Oxford: Basil Blackwell, 1953. _________. The Blue and Brown Books. New York: Harper and Row, 1958. Articles , = v Mston, William P. "Language," The Encyclopedia of Philoso phy, ed. Paul Edwards, IV, 384-386. "Meaning," The Encyclopedia of Philosophy, ed. Paul Edwards, V, 233-240. Anderson, Raymond E. "Kierkegaard's Theory of Communica tion," Speech Monographs, XXX (March 1963), 1-14. Babcock, C. Merton. "The Importance of Perspective in Com munication," Central States Speech Journal, VI (Fall 1954), 3-6. Baier, Kurt. "The Ordinary Use of Words," Proceedings of the Aristotelian Society, n.s., LII (1951-1952) , 47- 70. Baird, A. Craig. "Speech and the 'New' Philosophies," Cen tral States Speech Journal, XIII (Autumn 1962), 241- 240. 3ecker, Samuel L. "Research on Emotional and Logical Proofs," Southern Speech Journal, XXVIII (Spring 1963), 198-207. Bemis, James L., and Gerald M. Phillips. "A Phenomenologi cal Approach to Communication Theory," Speech Teacher, XIII (November 1964), 262-269. 3itzer, Lloyd F. "The Rhetorical Situation," Philosophy and Rhetoric, I (Winter 19(58'), 1-14. 2B(r 3lack, Edwin Benjamin. "A Consideration of the Rhetorical Causes of Breakdown in Discussion," Speech Mono graphs , XXII (March 1955), 15-19. Blankenship, Jane. "On the Teaching of Style," Speech Teacher, XIII (March 1964), 99-102. 3ryant, Donald C. "Rhetoric: Its Function and Its Scope," Quarterly Journal of SpeejSfo, XXIX (December I&53),,,* 401-424. "Of Style," Western Speech, XX (Spring 1957), Tol-lio. Dapp, Glenn R. "General Semantics for the Debater," South ern Speech Journal, XIX (May 1954), 294-303. Davell, Stanley. "MjaSt We Mean What We Say?" Inquiry, I (Autumn 1958), 172-212. Dhisholm, Roderick M. "Philosophers and Ordinary Language," Philosophical Review, LX (July 1951), 317-328. ___________. "J.. L. Austin's Philosophical Papers," Mind, LXXIII (January 1964), 1-26. Christopherson, Myrvin F. "Speech and the 'New' Philoso phies Revisited," Central States Speech Journal, XIV (February 1963), 5-11. Coutu, Walter. "An Operational Definition of Meaning," Quarterly Journal of Speech, XLVIII (February 1962), 59-64. Dallinger, Carl A. "Psychology, Symbols, Communication— Questions or Answers?" (rev. of Psychology of Com munication and Language in Thought and Action), Speech Teacher, XIV (March 1965), 151-155. Dedman, Donald W. "Criticizing Student Speeches: Philoso phy and Principles," Central States Speech Journal, XVIII (November 1967), 276-284. Dell, George W. "Philosophic Judgments in Contemporary Rhetorical Criticism," Western Speech, XXX (Spring 1966), 81-89. Eubanks, Ralph T. "Nihilism and the Problem of a Worthy Rhetoric," Southern Speech Journal, XXXIII (Spring 1968), 187-199: 0 261 Fausti,.Remo P., and Arno H. Luker. "A Phenomenological Approach to Discussion," Speech Teacher, XIV (Janu ary 1965), 19-24. Fisher, Walter R. "The Importance of Style in Systems of Rhetoric," Southern Speech Journal, XXVII (Spring 1962), 173-182. Flew, Anthony G. N. "Philosophy and Language," Philosophi cal Quarterly, V (January 1955), 21-36. Frandsen, Kenneth. "Semantic Compatibility in an Interper sonal Communication Laboratory," Western Speech, XXI 0! A (Summei^l967), 210-223. n° „ P * > ■ CrO] (O ' Gildston, Harold,0and Phyllis Gildston.o "General Semantics and Rhetoric," Today's Speech, XIV (November0L966)Q, 18-20. Goldschmidt, Walter. "Language and Culture: A Reply," Quarterly Journal of Speech, XLI (October 1955), 279-283. Gorman, Margaret• "A Critique of General Semantics," West- ern Speech, XXXI (Winter 1967), 44-50. Gregg, Richard B. "A Phenomenologically Oriented Approach to Rhetorical Criticism," Central States Speech Journal, XVII (May 1966), 83-90. Grice, H. P. "Meaning," The Philosophical Review, LXVI (1957), 377-388. _Hampshire, Stuart, "in MemorianJ. L. Austin. 1911-1960." Proceedings of the Aristotelian Society, n.s., LX (1959-1960), i-xiv. Hare, Richard M. "Philosophical Discoveries," Mind, LXIX (April 1960), 145-162. Hartung, Charles V. "The Scope of Linguistic Study," Quar terly Journal of Speech, L (February 1964), 1-12. Hayakawa, S. I. "Crucial Questions in Speech-Communication Research," The Frontiers in Experimental Speech-Com- municafeion Research, ed. Paul E. Reid. Syracuse: University Division of the Summer Sessions, Syracuse University, 1966. Heath, P. L. "The Appeal to Ordinary Language," Philosophi cal Quarterly, II (January 1952), 1-12. 262 Hill, Archibald A. "Linguistics Since Bloomfield," Quar terly Journal of Speech, XL (October 1955), 253-260. Hillbruner, Anthony. "Plato and Korzybski: Two Views of Truth in Rhetorical Theory," Southern Speech Jour nal, XXIV (Summer 1959), 185-196. Hochmuth [Nichols] , Marie, (rev. of Meaning and Necessity, , Quarterly Journal of Speech, XLIII (February 1957), 83-84. ___________. "I. A. Richards and the 'New' Rhetoric," Quar terly Journal of Speech, XLIV (February 1958), 1-16. Howell, Wilbur Samuel. "John Locke and the New Rhetoric," Quarterly Journal of Speech, LIII (December 1967), 319-333. ^ oP qp '& . ■ <nx Johannsen, Richard L. "Richard Weaver's View of Rhetoric and Criticism," Southern Speech Journal, XXXII (Winter 1966), 133-145. Johnstone, Henry W., Jr. "The Relevance of Rhetoric to Phi losophy and of Philosophy to Rhetoric," Quarterly "Journal of Speech, LII (February 1966), 42-46. Khatchadourian, Haig. "Symbols and Metaphors," The Southern Journal of Philosophy# VI (Fall 1968), 181-190. "Vagueness/ Meaning, and Absurdity," American Philosophical Quarterly, II (April 1965), 1-11. Danger, Susanne K. "The Origins of Speech and Its Communi cative Function," Quarterly Journal of Speech, XLVI ?......"(April 1960), 121-134. Llewelyn, J. E. "Family Resemblance," Philosophical Quar terly, XVIII (October 1968), 344-346. Malcom, Norman. "Philosophy for Philosophers," Philosophi cal Review, LX (July 1951), 317-328. Mates, Benson. "On the Verification of Statements about Ordinary Language,./' Inquiry, I (Autumn 1958) , 161- 171. McKeon, Richard. "Aristotle's Conception of Language and the Arts of Language, Part I," Classical Philology, XLI (October 1946), 193-206. v&) 263 McKeon, Richard. "Aristotle's Conception of Language and the Arts of Language, Conclusion," Classical Phi lology, XLII (January 1947), 21-50. Moore, Wilbur E. "I Disagree," in "A Controversial Sympo sium: the Basic Function of a Speech Department Is Developing Speech Skills," Central States Speech Journal, X (Autumn 1958), 50-56. Murray, Elwood. "Streamlining the Speech Edifice: I," Western Speech, XXIII (Fall 1959), 197-202. ___________ . "The Laboratory for Semantic Compatibility at the University of Denver," Western Speech, XXXI (Fall 1967), 274-280. JJatanson, Maurice. "The Limits of Rhetoric," Quarterly Journal of Speech, XLI (April 1955), 127-132. ___________. "Rhetoric and Philosophical Argumentation," Quarterly Journal of Speech, XLVIII (February 1962), 24-30. Sfewman, John B. "The Area of Semantics," Quarterly Journal of Speech, XLII (April 1957), 155-164. ___________. Rev. of The Principles of Semantics, Quarterly Journal of Speech, XLIV (October 1946), 324. ___________ . "General Semantics and Academic Phagocytosis," Quarterly Journal of Speech, XLVII (April 1961), 158-163. ___________ . "The Semantic Analysis of Ordinary Language," Quarterly Journal of Speech, XLIV (December 1963), 410-416. ___________ . "Semantics and the Study of Speech," Speech Teacher, XVI (March 1967), 98-102. "Symptomatic Signals in Speaking and Writing," Western Speech, XXXIII (Winter 1969), 40-44. ___________, and Milton W. Horowitz, "Speaking and Writing," Today1s Speech, XIII (February 1965), 2-4+. Osborn, Michael M. "The Evaluation of the Metaphor in Rhet oric," Western Speech, XXXI (Spring 1967), 120-127. ___________, and Douglas Ehninger. "The Metaphor in Public Address," Speech Monographs, XXIX (August 1962), ________2 2 3 = 2 3 A .______________________________________________ --------------------------------------------------------------------- Z M Parrish, Wayland Maxwell. "Interpreting Emotion in Poetry," Southern Speech Journal, XIX (March 1954), 205-213. It t Getting the Meaning1 in Interpretation," Southern Speech Journal, XXXIII (Spring 1968), 178- 186. Perritt, H. Hardy. "Linguistics: A Lambent Glance," South ern Speech Journal, XXVI (Summer 1961), 279-284. auinton, Anthony. "Linguistic Analysis," Philosophy in the Mid-Century, ed. Raymond Klibansky*} II. Firenze: La Nuova Italia Editrice, 1961. Rogers, Carl R., and F. J. Roethlisberger, "Barriers and Gateways to Communication," Harvard Business Review, XXI (July-August 1952), 46-53. Rosenfield, Lawrence W. "Rhetorical Criticism and an Aris totelian Notion of Process," Speech Monographs, XXXIII (March 1966), 1-16. ^ "The Anatomy of Critical Discourse,1 1 Speech Monographs, XXXV (March 19-68) , 50-70 "A Game Model of. Human.Communication," What Rhetoric (Communication Theory) Is Appropriate for Contemporary Speech Communication? ed.' David H. Smith. Minneapolls: University of Minnesota, 1969, 26-41. Rosenthal, Paul J. "The Concept of Ethos and the Structure of Persuasion," Speech Monographs, XXXIII (June 1966), 114-126. Rothman, Richard M. "Language and Thought Re-visited," Today's Speech, XIV (April 1966), 11-12+. Ryle, Gilbert. "Systematically Misleading Expressions," Proceedings of the Aristotelian Society, n.s., XXXII (1931-1932), 139-170. ( Rev. of Meaning and Necessity, Philosophy, XXIV (1949), 69-76. Rynin, David R. "Semantics," Western Speech, XX (Winter 1956), 35-44. Scott, Robert L. "Some Implications of Existentialism for Rhetoric," Central States Speech Journal, XV (Novem ber 1964), 267-284. m “21)5 Searle, John. "Austin on Locutionary and Illocutionary Acts," Philosophical Review, LXXVIII (October 1968), 405-424. "Meaning and Speech Acts," Philosophical Review, LXXI (October 1962), 423-432. "Human Communication Theory and the Philosophy of Language: Some Remarks," Human Communication Theory, ed. Frank E. X. Dance. New York: Holt, Rinehart, and Winston, 1967. Sesonske, Alexander. "Saying, Being and Freedom of Speech," Philosophy and Rhetoric, I (January 1968), 25-37. Smith, Raymond G. "Development of a Semantic Differential for Use with Speech Related Concepts," Speech Mono graphs , XXVI (November 1959), 263-272. "Motivation and Communication Theory," Central States Speech Journal, XV (May 1964), 96-98. Strawson, P. F. "On Referring," Mind, LIX (July 1950), 320- 344. 0 Toulmin, Stephen E., and Kurt Baier. "On Describing," Mind, LXI (July 1952), 13-38. Wallace, Karl R. "The Substance of Rhetoric: Good Rea sons," Quarterly Journal of Speech, XLIX (October 1963), 239-249. Walter, Otis M. "On Views of Rhetoric, Whether Conservative or Progressive," Quarterly Journal of Speech, XLIX (December 1963), 367-382. Warnock, G. J. "Verification and the Use of Language," Revue Internationale de Philosophie, XV (1951). "The Philosophy of Wittgenstein,1 1 Philosophy in the Mid-Century, ed. Raymond Klibansky. Firenze: La Nuova Italia Editrice, 1961, pp. 203-207. Weaver, Carl H. "Semantic Distance Between Students and Teachers and Its Effect upon Learning," Speech Mono graphs , XXVI (November 1959), 273-281. , and Gary L. Weav%r. "Information Theory and the Measurement of Meaning," Speech Monographs, XXXII (November 1965), 435-447. 266 Weitz, Morris. "Oxford Philosophy," Philosophical Review, LXII (1953), 187-233. Wells, Roulon. "Meaning and Use," Word, X (1954), 235-249. ___________ . "Philosophy of Language," Philosophy in the Mid-Century, ed. Raymond Klibansky. Firenze: La Nuova Italia Editrice, 1961, pp. 139-145. White, Morton. "New Horizons in Philosophy," Central States Speech Journal, XII (Spring 1961), 188-196. Wieman, Henry Nelson. "The Philosophical Significance of Speech," Central States Speech Journal, XII (Spring 1961), 170-175. ___________ , and Otis M. Walter. "Toward an Analysis of Ethics for Rhetoric," Quarterly Journal of Speech, XLII (October 1957), 266-270. Wisdom, John. "Philosophical Perplexity," Proceedings of The Aristotelian Society, n.s., XXXVII (1936-1937), 71-88. Unpublished Materials Hertz, Richard. "The Existence of Linguistic Rules." Un published Manuscript. Pasadena: California Insti tute of Technology, 1969. ___________ . "Rules and Language: A Philosophical Study of Linguistic Communication." Unpublished doctoral dissertation, University of Pittsburgh, 1967. Spicer, Holt V. "Stephen Toulmin's Functional Analysis of • Logic and Ethics and Its Relation to Rhetoric." Unpublished doctoral dissertation, University of Oklahoma, 1965. Wier, John Rex, Jr. "The Relationships of Certain Selected Formulations of General Semantics to Classical Rhe torical Theories of Invention." Unpublished doctor al dissertation, University of Denver, 1964. Willard, Dallas Albert. "Meaning and Universals in Husserl's Logische Untersuchungen." Unpublished doctoral dissertation, University of Wisconsin, 1964. <§ #

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Clement Attlee'S Advocacy Of Democratic Socialism: A Reasonable Rhetoricof Revolution

PDF

Emerson'S Philosophy Of Rhetoric

PDF

Prescriptive Deontic Logic: A Study Of Inferences From Linguistic Forms Expressing Choice And Conditional Permission And Obligation

PDF

Christian Science and the rhetoric of argumentative synthesis

PDF

An Analysis Of Selected Presage Criteria Of Reciprocal Student And Teacher Ratings In Beginning College Speech Classes

PDF

The Rhetoric Of The Methodist Camp Meeting Movement: 1800-1850

PDF

A Historical And Critical Study Of The Public Address Of James Harvey 'Cyclone' Davis (1853-1940) Of Texas

PDF

An Experimental Study Of The Effect Of Listener Feedback On Speaker Attitude

PDF

An Experimental Study Of An Application Of Game Theory To The Selection Of Arguments By College Debaters

PDF

Communication Of Mexican Americans With Public School Personnel: A Study Of Channel, Code, Receiver, And Source Preferences

PDF

An Experimental Study Of The Accuracy Of Experienced And Inexperienced Speakers In Identifying Audience Behavior As Indicative Of Feelings Of Agreement, Indecision, Or Disagreement

PDF

Kennedy-Khrushchev Strategies Of Persuasion During The Cuban Missile Crisis

PDF

A Phonemic And Phonetic Description Of The Speech Of Selected Negro Informants Of South-Central Los Angeles

PDF

The rhetoric of the inward light: an examination of extant sermons delivered by early Quakers, 1671-1700

PDF

An Experimental Comparison Of Spoken Communication Developed Individually and Interindividually

PDF

An Experimental Study Of The Effect On Babbling Of Visual Interaction Between Mother And Infant

PDF

An Intensive Rhetorical Analysis Of Selected Speeches Of Robert Maynard Hutchins: 1940-1955

PDF

An Experimental Study Of The Ability Of Lay Judges To Distinguish Betweentypescripts Of Individual Idea Development And Group Idea Development

PDF

An Analytical-Historical Study Of The Factors Contributing To The Successof Mark Twain As An Oral Interpreter

PDF

A Descriptive-Analytical Study Of Ethical Standards In Contemporary American Public Address

Asset Metadata

Creator Stewart, John Robert (author)

Core Title Rhetoricians On Language And Meaning: An Ordinary Language Philosophy Critique

Degree Doctor of Philosophy

Degree Program Speech

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,Speech Communication

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Fisher, Walter R. (committee chair), McBath, James H. (committee member), Willard, Dallas (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-424864

Unique identifier UC11360924

Identifier 7023187.pdf (filename),usctheses-c18-424864 (legacy record id)

Legacy Identifier 7023187.pdf

Dmrecord 424864

Document Type Dissertation

Rights Stewart, John Robert

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA