This is an offprint from: Yishai Tobin (ed.) FROM SIGN TO TEXT A SEMIOTIC VIEW OF COMMUNICATION John Benjamins Publishing Co. ‘Amsterdam/Philadelphia 1989 (Published as Vol. 20 ofthe seri FOUNDATIONS OF SEMIOTICS) ISSN 0168-2555 / ISBN 90 272 3292 X © Copyright 1989 - John Benjamins B.V No part of this book may be reproduced in any form, by print, photoprint, microfilm or any other means, without, written permission from the publisher. The challenge of polysemy Walter Hirtle ‘thin the system anc ony within the ystm the forms have a meaning of their own, pre: destning them for whichever particular contextual sensei required in discourse (Guillaume 1984:81) In intellectual endeavours, itis always prudent to have a clear view of one’s postulates. In this colloqui wasted effort to take a few ing point, the ‘SIGN, a question which has often been examined in the past but w! requires certain clarifications as I intend to show. However, before begi ning I wish to make a change in terminology in order to avoid confusion. Since I am accustomed to using the term sign in the common-usage, pre- Saussurian sense of ‘the physical means of expression’, [ am afraid that any attempt on my part to use it as many people here do, in the Saussurian sense, would lead to the inconsistencies found in authors who try to switch usage in mid-stream, and even in the Cours de linguistique générale itself. T shall therefore make use of the term word although I am quite aware that the word is not the only type of SIGN. The point I want to make concerning the word can however be applied, mutatis mutandis, to other types like the morpheme. Let us begin by a citation which brings into focus the problem posed by the word. A grammar published a few years ago begins with the following statement: Aword on (Mitins, 1962: If we accept this as a faithful reflection of the nature of words, if indeed the meaning of a word is derived from its co organizers of the present colloquium with getting things the wrong way m is not meaningful; what it means depends on it context136 WALTER HIRTLE round since they should have given it the name: “From Text to Sign”. Quite ew involved here, and itis worth to discern them more clearly. ‘meaning from the context is not an snses in different contexts. The example given by ‘the word operation which can have senses as varied as surgical operation’ or ‘financial operation’ depending ‘And singe there is no way of telling from the or cited, oon the context and word which of the senses is intended, the approach of the al above attributes the source of the meaning to the context and , solution which seems to correspond exactly to our experience as readers or listeners when we try to understand @ text. ‘And yet this seemingly satisfactory solution leads to a contradiction with another aspect of reality. After all, a context is made up of words, each, according to this approach, arriving in the context in quest of a mean- ing. Unless Lam much mistaken, if one combines a number of meaningless words to form a context, the context itself will be meaningless. And yet we know by experience that sentences do convey meaning, that we can com- municate by means of language. This shows that words do not get their ‘meaning from the context. spasse, some scholars have appealed to the extra experience tel ations that are fami is quite unknown to the (as when one receives a telephone call or hears a cry in the night). We can only conclude that words do not get their meaning from the extemal situation It seems, then, that both the contextual approaches to meaning conflict th our experience in using language (and that the organizers do not have loquium standing on its head). This takes us back to the common sense view that a word on its own does have meaning, and that itis this ‘meaning which it contributes to the context. This approach postulates that ‘one form has one fixed meaning because only on this basis can we account for the fact that communication is possible in novel and unknown situations ‘THE CHALLENGE OF POLYSEMY BT ‘and for the fact that even when a word is out of context it conveys some meaning. This is the comfortable, traditional approach to the problem, but it too has its difficulties. Thus, the fact that a word can have different senses in different contexts would appear to contradict the one formvone meaning postulate. However, rather than admit that their position is untenabl proponents of this approach generally resort to one of two exped ther of which is very convincing. ‘One way of defending the one form/one meaning po reduce the different senses of the word to one of the obst the author of an article entitled “Every any means every” (Savin 1974), argued that in every one of its uses, any has the sense of ‘every’. the sense of any in an example like Pick any card of ‘every’ and so must be left out of account. Indeed this bly leads to a type of reductionism which leaves out part of is to maltiply the form, to declare that there is a seperate form, a hom ‘onyim, for each discernable sense of the word or morpheme. For example, it has been claimed (Ladusaw, 1980: 67ff) that in English we have not one but two any's, corresponding to two of its discernible senses, Both of these supposed words are written and pronounced in exaetly the same way. When applied to grammatical morphemes, this expedient has led to the proposal ‘that in English we have three progressive forms. Unfortunately, this way of defending the one formone meaning position runs into a very practical problem: how is the listener to know which of the homonyms is appropriate language. iply the form to equal number of contextual meanings found, nor reduce the number of con- textual meanings to one, the counterpart of the single form found, must we abandon the one form/one meaning position? Impossible, because without138 WALTER HIRTLE this one-to-one correspondence between the mental and the physical in a ‘word, communication as we know it would be impossible. This is the chal- Tenge of polysemy. ‘The fact of polysemy poses the fundamental problem for language analysis because it concerns the very nature of the basic entity of language: ‘the combination of meaning and form, the symphysy of mental and physical that constitutes every word or morpheme, Furthermore, it is a problem which is commonly met in science: how to derive a plurality of effects or consequences from a single cause or condition. One is reminded of Eins- tein's dictum: “The grand aim of all science is to cover the greatest number of empirical facts by logical deduction from the smallest number of hypoth- ‘ses of axioms.” Indeed this problem has already been confronted and eleg- tantly solved in one of the most highly developed sectors of linguistics, com- parative grammar, and it is worth recalling the main lines of how it was sol ved there because a similar type of solution can be proposed for the prob- Jem of polysemy. In comparative grammar, it will be remembered, striking phonetic cor- respondences were observed between words in different Indo-European languages. To explain these correspondences, to make it conceivable how the reality observed in texts could have come about, nineteenth century comparatists tried to imagine a single prior condition which might have tiven rise to them. Concretely, they tried to reconstitute in the proto-lan- guage, which was not observable, a phoneme which could feasibly develop in different ways to produce the result observed in texts. The essential here, then, is the reconstituting of a single prior condition to account for a plural- ity of observed consequences, the prior con conceivable only, the observed consequences Since this method is not only accepted as valid but has even been consid- tered as a model in other fields of scientific endeavour, it can serve us as, guide in our attempt to meet the challenge posed by polysemy (cf. Valin 1964 for a detailed account). In the case of polysemy, the perceivable facts, the data to be explained, are the different senses of a word or morpheme, its contextual ‘meanings. We must therefore seek something in the field of meaning which will provide a prior condition, a i which all the contextual meanings will appear as possible consequences. ‘The meaning we are secking must somehow exist before the contextual ‘meaning arises because a condition necessarily exists before its conse- } I | ‘THE CHALLENGE OF POLYSEMY 139 quence. ‘The linguistic framework within which we can conceive of this goes back ultimately to Saussure, who first proposed that language is @ dichotomy consisting of langue and parole. However, it fell to Guillaume to make explicit the temporal relationship between the two (cf. Hewson, 1976). Guillaume’s lifelong concern with exploring time in language, and language in time, ted him to realize that Ja langue, or rather tongue (to use the English term), being the prerequisite condition for discourse (a common alternative to the French parole), must exist before an act of language can produce a bit of discourse. It follows, therefore, that it is in tongue that we must seek the prior condition of all the contextual meanings of a word and not, as so many do, among the contextual meanings in discourse. Just what does this imply — to seek in tongue the prior conditions for implies that the meaning of a word t as comparative grammar pos- tulates the prior existence of a form in a proto-language, so we must postu- ts meaning in the subconscious system Prior to its use in discourse since we cannot use something t! already exist. In the second place, ‘temic-meaning-in-tongue never emerges as such into consciousness and so can never be observed directly; just like some form in a proto-language, ust be reconstructed on the basic of the evidence of attested usage since any theoretical construct must be confronted with data. In the third place, the potential for producing all the actual senses in discourse since the prior condition permits a certain range of consequences. Guillaume (1984: 81) describes the relationship from the point of view of the speaker undertaking, an act of language: {tcannot be too strongly emphasized that, in tongue, the systemic meaning of a form is totally acquired whereas the total contextual meaning is yet 0 be acquired. The role of tongue is to permit a tairly wide variety of contex- tual meanings in discourse by presenting forms i meaning makes possible a certain range of con ‘This, then, is one way to meet the challenge of polysemy - in fact, the only way, to my knowledge. No other way has been proposed permitting us to reconcile what we know of polysemy with what we know of language as abad WALTER HIRTLE a meaning-expressing activity, a means of communication, And this post- ulating of a potential meaning permitting a certain range of contextual ‘meanings has provided the key to @ number of grammatical problems such as the imperfect in French (cf. Guillaume 1971: 119-133), the article (cf. Hewson 1972), some/any and number in English (cf. 1988, 1982). Indeed, so far this conception of meaning has proved heuristically valuable wherever it has been applied. Its only disadvantage to date has been th: conflicts, not with the facts of language, but with the ideas of many ling and other students of language. The problem with this approach is that it entails a new view of lan- ‘guage, a view which begins by taking Saussure’s langue/parole distinction seriously enough to explore it and discover that it does not constitute a dichotomy, as Saussure would suggest, but a binary whole. In this view, the whole problem of accounting for usage is one of discerning the passage from one part (tongue) to the other (discourse). That is to say, 10 explain the use of a given word, we must describe how the speaker, from the word. n tongue, actualizes it es one constituent of the particular ing. By means of the act of language, the speaker calls on ite possibilities to create a momentary bit of discourse with its expressive particularities. Without the potential of tongue, without, 1¢ subconscious set of mental programmes at the permanent dis- posal of the speaker, thete would be no discourse as we know it, no texts. Does this mean that everything in a text is provided for by the system in tongue? Certainly not, The various means of expressiveness, like the rhythms of a given poem which depend on the inventiveness of the poet, ed in tongue as a permanently available |. In that case, we would all be poets. How- istic framework of any text, its syntax, can only be what the system of the word permits it to be. That is to say, a word can only be used in discourse in ways permitted by its system, and this system is certainly hed in tongue. As a consequence, although we may not all be poets, ‘we are all speakers of our mother tongue. Such notions may not appear to involve anything radically novel. Indeed, many scholars are even willing to admit this view of language as a theoretical necessity and yet surprisingly few are prepared to accept the methodological progression it imposes. Most of us, I suppose, would agree ‘that in order to analyze a-text, one must already have made some analysis of the sentences composin larly, not many would attempt to s ‘THE CHALLENGE OF POLYSEMY 141 ihout fst analyzing its component lain the make-up of a sentence wit! ly . cre utes And yet teltvely fo are prepared to take the nex e ed step, which consists of acknowledging the priority ofthe word and the nee textual analysis is, then, c sie vew of Tanguage, that is, a view which attempts to the facts. It brings us back to the theme of the present com Sign to Text, and shows that we have not got things bas mn concerning the st ‘wards, It does, however, raise a fundamental question concerning, ing oi what do we understand by the term SIGN? Do we understand it asan clement of discourse with its ever-changing, chameleon-like combina tion of form and content, oF do we understand it s 2 unit of fongue with 8 stable, one-to-one correspondence of form and meaning thanks | ea Xe answer we give the speaker can create discourse? On the answer we gi Bes will depend the very nature and value of our syntactic and textual analysis REFERENCES, structure sémiologique et structure psychi- Guillaume, G. 1971. Legos de linguisique: r nu ue jrangaive I. Québec: Presses de Unive va : on ee S Sea tonndatons fore Sconce of Language. Anstersam and Pile Hewson, J. 1972. Article and Noun in English. The Hague: Mouton. Howson, 3, 1976, “Langue and parole since Saussure” in Htstoriographia Linguistic, 3, Sis Et Hi To, Number sind nner Space: Study of Grammatical Number in Eras us Pee de Uns Lael, Qutb i vices Sone ely: merge se" La 2p ET elk, WA. BR, Pty Sey a Ihr epee tna Unverty Lng Cl ; Bn, Londo: Meta. natn HE, Graar of Moder al cs i sete Mone cera mens cn Prodienc sels e pcg "Egon CRS vat a a ntade comparave ning Rtogu eo pachonée Me tenes Que: recede nvr LaIn the FOUNDATIONS OF SEMIOTICS series the following volumes have been pub- lished and wil be published during 1989: STUDIES IN LOGIC by Members of the Johns Hopkins ntrod. by Max HE. Fisch, Amsterdam, 1983, th Devopmen of Sint ‘With an English summary. Amsterdam, 1987 HOFFBAUER, Johannes Christoph: Tentamina Semiologica, sive quaedam 6. sry: or the Secret and Swift Messenger etc. (1707), Tog. with an Essay towards a Real Character and a Philosophical Lan- semloticant. AmsterdamPhiladelphia, n.y-p. 9, PHARIES, David A.: Charles S. Peirce and the Linguistic Sign. Amsterdam, 1985, 10. DASCAL, Marcelo: Leibniz. Language, Signs and Thought. Amsterdam, 1! 11, EKSTEIN, Rudolf: The Language of Psychotherapy. Amsterdam/Philadelphi 1989. 12. BUSSE, Winfred & Jurgen TRABANT (eds): LES IDEOLOGUES. Sémiotique, opie lense Engg pendant le Réoluon rng, Antertan, 13, FONAGY, Ivan: Language within Language. An evolve approach. Amsterdam Philadelphia, n.y.p- 14, DELEDALLE, Gérard: Charles S. Peire, phénoménologue et sémioticien. Amst dam, 1987. 15, MORRIS, Charles W.: Symbolism and Realy. A Smudy in the Nature of Mi 18. DELEDALLE, Gérard (¢d.): SEMIOTICS AND PRAGMATICS, Amsterdam Philadelphia, 1989. n.y.p. 19, KOBERNICK, Mark: Semiotics of the Drama and the Style of Eugene O'Neill sy: ON THE MEDIEVAL THEORY 88. Geet aa 28, SCHMITZ, H. Water (ed) ESSAYS IN STONIFICS. Papers presentedon the sion of the 150k anniversary ofthe birh of Vitoria Lady Welby (1837-1912). Ams ‘damPhiladelphia, 1989: 1n.y.p. 24, MARTY, Robert: L’Algebre des signes. Amsterdam/Philadelphia, 1989. n.y.p.