Libl-If)" of Cona' n5 C.tlloaina in Publkation Data

Wdt lis Zeichen . Ena1i$h Clinics of Km;(I{ics.

(TopIQ in contemporlry Je1tIlotk,) Trlnslllion of: Die Welt lis Zeichcn. Indullts bibliotr.phiul rderences .nd In<Ic~ .

I. Semiotics. I. Krlm~n , Mlrt;n, dltt. 11. Title. Ill . x.in. P99.W3I1l 1917 ooLS I ISBN 0-J06..42321-9

This volume Is • Ir. nsl.tlon of Dk Wt/I ~fs ltkll,lI: K/~ssilr." dtr modtr"tll Stmlolllr.. edited by M.rtin Kr.mpn'l . KllUs Othltt. RoIInd POSM1 •• MI Thur. von UukOll. published In 1911 by Wolf Jomt Siedln Verll' GmbH. Btflln. Chlplns ,. 6. Ind 8 were orl,lnl lly pubUshtd In En,l1sh. Ind It Is the orl,lnll versions of Chapten ~ and 6 llId 1 revlstd vcnlon of Chlpter • th.t .re .~intcd herein. with ~rmi"ion of rhe copyri.ht hold" ...

Germ.n tdltlon Cl 1911 by Wolf Jobst Sledlcr 'o'nlll, OmbH, Btflin

1Cl 19l7 Plenum Pre,.. N .... York A Division of Plenum Publlshln, Corporltlon 233 Sp,ln, Streel. Ntw York , N.V. 10013

All ri,hls .nerved

No part of thil boolt mlY ~ ,I!"ptoouetd. stored In I rCl .!ev.lsyllml. or tnmsmitttd In Iny form or by any means •• I~ronlc. mechanlc.l. phouxopyln,. mk .ofilmln,. recordin,. or OIher ... llC, without written permi .. lon from the PubU1l1fl"

Prlnltd In the Uniltd Stltes of Amnk.

Contents

Foreword to the English language Edition

11wma£..t.~S~bwk

. . . . . . . . • . . • . . . . . . . . .. vi;

P~m . . .. .. ..... . ........... i. Ro/and PQstler

Klilus Orhltr

~tline of pejrce's Semi.no"li<", _ ___ ~_. _. _ . . _._._._._._._._. _. _. _ .. _.~--,1

I . Stages in the Foundation of Semiotics 2. The Foundations of Pei~'s Semiotics 3. The Imponance and Generality of Peircian Semiotics .......... . 4. Some Aspects of Pcircc's Consensus Theory ........ . .... . ... .

Notes . . . . . . . . . . . ... .. . . _ .. . .. ........ .. . . . .. .. .. .. . References ........ ... ... . . . ..................... . • .....

RQtund PQsner

2. Charles Morris and the Behavioral Foundations of Semiotics

I . Inte llectual Development and Specialization .... ... .. .. ..... . . 2. Phases of an Action ... . .. . .. ..... , ....... ... • . ' •. . • ...... 3. Types of Signs .... . .. . .. . .... . ... , ............•.. • ...... 4. Dimensions of Signification .. ... ... .. •• ...•. .•. ... . .•... .. . 5. Oimensions of Sign Use ........ .. •.. . . . .... •. ......... . .. 6. Dimensions of Value . .. .. . ...... .• .. • ...... • ...... •......

Notes . .. . .... . . ... ... . .. . .. . . . .... •. .....•............. Refe~nces ........... . ..... . •. . • . . •.. • . .. • ..• ....... ...

2 , 8

11 18 20

23

23 2. 28 33 39 42 46 53

xii!

CONTENTS

CONTENTS

Mart;" Kra~"

3. Ferdinand de Saussure and the Development of Semiology S9

I . On the Biography of de Saussure . ...... .. . . . ... .... . . . .. . .. 59 2. Semiology and Linguistics in de Saussure .... .. ..... . .... ... . 6 1 3. On the Semiological Generalization of de Saussure's Basic

Concepts .. . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . . . . . .. 66 4 . De Saussure's Semiology and Problems of Stl1.J ctura lism ., . ... , . 78 5. Perspectives of SemloJogy ... .•. . , .. ... .. ....... . .. ... .. . . . 83

Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 References .......... . ......... • . . . • . . • . ..... . ..... . .... 86

Jurg~" Traballl

4. Louis Hjelmslev; Glossematics as General Semiotics . .• • . • . • • 89

I . Lll Lll"gut as the General Principle of Semiotic Structure ....... 89 2. The Double·Art iculated or Twice·Polll'led Sign . ..... .. ... .. ... 93 3. Connotation and Metasemiology ..... ... .. .... ... ... , ..... . . 100

NO(es . .. ... . ........... . ..•........ . ...... .. ... • .. . , ... 106 References 108

Um~rto fro

5. The Influence of Roman lakobson on the Development of Semiotics ..... . ... .. ..... . .. .... . . ..... . .... . ... ... . .... . 109

I . The Hislory of an Oslracism .. .. .... . .. .•.. • . ... .. •. • . .. ... 109 2. The Quesl of Semio(ics ... ,. .... . . .. .. . ... . . .... . . . .. ... . . [11 3. The Basic Assumptions .... . . ... . .. • ...•. . . ... • .. • ....... . 113 4 . The Final DeSign .... . .. , . .. ... . .. . . . . . .. . ... . .. . . . ..... . 122

Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . • . . . . . . • . . . . . . . . . . .. 123

Thoma$ A.. St~ok

6. Kad Balller .... . .... .. ... ... .. . . .. .... .• ..• ..• . .. . .. •. .... 129

I . Background . ... .. .... . . .. . .. . ... ..... . ..... . .... .. . . . ... 129 2 . Blihler and Intimations of Semiotics . . . . .. ... ...• .. • .. .• ..... 13 1 3. Open Questions .. . ..... .. ... .. . . ... .. .... .. •.. • . . .•.. .. . 142

References .. .. .... .. ... . . . .... .. ... ... • .. . • .. • ... •. .. .. 143

CONTENTS

Thurr loon Utxkull

7, The Sign Theory of Jako b von Uexkilll 147

I. Personal History and Field of Research 147 2-----Wbatis Umwc:ltJkseJlICb? 148 3. Umwelt Research and Linguistics .. .. , ...... , . , .... , .... , " 149 4. The Problem of the Sign-Receiver, the Law of "Specific

Life-Energy," and the "Elementary Self' ,. " , " ,. " ' ,' ,. ,, . , 151 5, The "Private" Nature of Signs ", .. " .... "., .. , .... .. ... .. 153 6. The Composition of Man's Ellperienlal Universe Seen as a

Composition of Sign Processes ...... . ..................... 155 7, The Subjective Universe of the Observer as the Kc)' to the

Self·World of thc..AnimaLunder.Ob&er'talion 162 8. The Functional Circle as a Special Case of the Con~

Correlation " .". , . . . ,",., . ,',. ," " "', ." .". , . . , '" 166 9. Essentials of Jakob von UexkilU's Theory of Signs .. , ", .• ,'" 169

10 , Concluding Observations: Attempt at a CompariiOn of Uedo.lI's, Peirce's, Saussure's. and Morris' 1beories of Signs ., . , .. • ..• , 172

Notes .. . , ..... , . , ... .. , , . ... , .... , , . , .. , . , ... , , , .... , ,. 175 References 178

£ugtl1 Batr

8. Thomas A. Sebeok's Doctrine of Signs 181

I. Sebeok's Career and Theoretical Poinl of Departure .. , , , , .... " 181 2. Historical Sketch of the Semiotic Tradition ..... , ... , . , ... ,',. 134 3. Anthroposemiotics and Zoosemi()(ics .. .... .............. . , .. 187 4 . Contributions 10 the Classification of Signs and Sign Systems .... 190 5. Endosemiotics and EIlosemiotics: The Semiotic Self , . , ... , ... .. 195 6. Sebeok's ''Thomism'' .. .... ..... . .... .. .......... ......... 196 7. Forms of Life and Forms of Language ............ , .... , , .... 201 8. Totem and Taboo , .. ... .... .... , .............. , .. ... .. ... 203 9. Semiotics in Transition ........................ • . , ........ 206

Notes ...... ............... . • .. . .. , •. . • . , . , ..• , ...... , .. 207 References .. " .... . " . . , ...... , ...... , ......• ,......... 207

Glossary ............•.. , ...•. • ...•..•. , ' •. . • ' , ....• ' , .. , " 211

Author Index ..... ...... .. ... ... ..... . ' . . ...... . .... .. .. .. .

Subject Index 2.'

CHAPTER 4

Louis Hjelmslev: Glossemat·cs as General Semiotics

JORGEN TRABANT

1. LA LANGUE AS THE GENERAL PRINCIPLE Of SEMIOTIC STRUCTURE

1.1.

N t long ago. the author of a r view f the recent German edition of Hjelrn lev' major work. the Prolegom na 10 a Theory if Longua e queri d th d for the tran lati n by a kin wh th r Hj Imsle ought till to be read in lingui tics cou K t chi, 1977 . No clearer indication could be ~ und of th i reput into which glo sematic • a an important chool of European tructurali rn, h fall n. Thi general tendency howe r. t which the lingui tic of Greima land a perhap ch only exc pci n ught t be c mbated n two ground .

In the fi I place, Co riu one of th mo t minent hi IOOan of lingui tics. ha pIa ed Hj Im le ide no I r figure than Humboldt a n f the ~ und rs f g n ral Iingui tic : • For we a con inc d that glo m lic mark a d i ive

point in the hi tory of lingui tic . It cannot be pa ed over, r re ard dimply a a deviati n' from the true c u f lingui tic • since it c ntinue and devel a tradition that goe back to . . . Wilhelm v n Humboldt. . . . It th re(l m t u. that in thi r pe t Hjelm lev can be compared only ith Humboldt . . . •• for. "like Humboldt Hj Im le ee th po ibility of comprehendin all human and cultural pr blem in t nn of languag · (C riu t 1954: 17 .

I in yd Whyte .

German .

89

90 JURGEN TRABANT

Sec ndly t I would offer th foH wing argument again t ign ring Hjelm -Icv. Even if. a the review r cited above implies one ought not to read Hjelm I v .in a university linguistics course bKau eglossemati.es-regardlc s of C eriu· opinion-might ju tifiabJy be regarded a a "deviation' from the true eou of lingui lics. it seemt me nevertheles that Hjelmslev s work iindi pen ible in a univ r ity semioti s cou e. F r in my pini n. and with all .re peel f'I r Hjelmslev's importance as a linguistic scienti t. the truly original a peel of hi work i the development of a emiotic rather than a lingu; tic th ory. F r he i· nothing le thanth originat · r f that S .u . urean· de ideratum. namely a general science f ign (simiologi~) bas d on immanent a"d structural /in lIis/ie .

1.2.

A1though he described his new lingui tic science as "structural linguistics" or "immanent linguistic • tt Hjelm lev al 0 gave it the name glossemalics. t di tingui h it from traditional lin uistics. Strictly peaking glo matic i n t a lingui tic science at all: It i n t a science of the "lingui tic JanguageH (iangue lingllislique-Hjelmlev 1944:32) OT of the "natural language. · but rather a science of signs. acience f thgeneraJ emiotic structure. In Hjelm lev · s writing "language" (tongue). d fined as a glosematie obj ctis sub tantially ynonym u with si n ystem" (which Hjelmslev calls "semiotic'). Exclud d

from this usage of the term language are those specific points at. which th particularity f th lingui tic language ver the gen r I · mioti structure is detennined. I

In hi programmatic editorial in Acta Lin uislica. HjelmsJev make it absolutely clear that the object of glossematic is not. in lhefinal analy i , the • linguistic language.2 Following Sau ure' model. he in i ts that the object of lingui tics '/' objet auque/ on ,ise" - the goal) i not the parole. which is never· theles the n.ece _ ary tarting point r objet eludi~l. but the lan ue. Thi obj ct /angue is d · fin d mo exactly by Hjelm lev as follows: "If th parol . the manifestation of th langue then a langue. in turn i th manifestation of the typologicaJ clas to which il bel ng • and ultimately th manife tati n of that particular class of cia th tiTHE Ion ue. . . . the pecies (espece Ion ut is the actual and prime object of stnJcturallinguistics. A partjcular longue (/all8"e porticu!iere) i subordinate lO the type. and the type ubordinate to the pe ie . We {the editorial board of Acta Lingll; tica} hope. therefore. that research into any panicular language will be based on the structure of the type or the. peci /anglle. and will be aimed specific Ily al throwing light n (hi pedes" (Hjelms· lev. 1944: 31 f.) . . romthis ascending sequence: lanue parliclIlier -Type cia .- pecies (espe.ce·, clas of cia • it become cl ar that langue j . ome-

Ihing far more compreh nsive than the tructureof one historical language: As a pecies. /anglle embraces the linguistic universal .

GLOSS MATICS AS GENERAL S MIOTIC 91

Yet even with th specie lan ue. H jelmslev has not reached the end of his abstractions: La langue is extended further to become the structure of th cia of all sign ystems. Tb three level of ab traction of pok n languag are then only a special ca within thi general lan ue. Having establi hed the equenc of langue part;culiere ty~ and pedes. and indicated th

that which should be illuminated by . tructural lingui tic • Hjelm I c ntinue : uThi typological hierar"Chy. which a cend from the parti ular languag to t pecie Janguage theoretically end nly at the point at hich one h ani ed

at th general principle of emiological tructure. 8 th Sau ure the ry and aJ 0 the most recent researches in fonnallogic indicate that th lingui tic languag Ion ue Iin8u;st;qu represent nly ne of the po ible manife tati n f la

Ion ue, when la Ion u i und in t wide t n a em racing every ign sy t m organized as a transfonnati e structure' ibid. Figure I hows how

we could depict th hierarchy of Ion u :

1.3.

"object vi eft:

(The genera. prlne e 0

Species language

~ Type language. ••

Particular 'anguag

:~~~~~: p.rOI~' Figur 1

Linguistic language

Nonllngul.tic langu ges

Th pas age qu ed above reveal a characteri tic feature of Hj Im lev' writing: He tend to pre nt the central ideas f gl ematics and thu glo se­matics it If as th fulfillment of Sau su intenti n . Although we cannot provide a detailed account h re of Hjelm Icv exegesis of Sau u ,w can h w taking one point central t our argument, that thi interpretati n i by no

mean 0 If-evi nt Hjelm lev ould have u believe. Sau sure presented Jangu ge a one pani ular sign system among many

and linguistics as one specific area ilhin th propo d scienc of miology. But he did not claim as Hjelmslc does that th lingui tic language was only one manifestation of a very genend langue. In the passage in the Cour t which Hj ImsJcv is presumably alluding in the qu tati nave, Sau ure re~ exclu-ively to the lingu;sticlanguage. In fact he say that the pok n lan uage--by

which he means the phonic language-is one po ible material manife tation of

92 JURCEN TRABANT

language. Beyond this material manifestation "above the functioning of the various organs tt there may be, says Saussure, "a more general faculty, the faculty which governs signs and which would be the lingui tic faculty par excellence" (Saussute. 1916: 27). While this formulation of a U ign-governing faculty" appears to sugge t Hjelmslev s wide-ranging definition of langue, it actually refers quite clearly and specifically in the pas age cited only to the 'Iongu~ Iingu;sl;que ." Wheth r or not. on the b i of the Cours text, the "Iocull/linguistique' can be extended to become a "faculte sem;o!og;que" must remain an entirely open que tion. In my opinion Saus ure i till working h re on the ab er et ion plane of specie (lingui tic) Janguage and not on the high r level of semiotic struc­ture in general. 4 This also account for the other relevant passage where he in ist that "the e s nee of languag i alien to the phonic chara t r f the lin­guistic sign" (Saussure, 1916: 21 . The extension of the concept langue to mean specifically "semiotic structure' is characteristic of Hjelmslev's radical interpretation.

1.4.

In his anaJysis of glos ematics, Coseriu (1954) shows quite clearly that the focus of Hjelmslev's attempt to establish a new Jinguistic cience is not language itself, but rather sign system in general.!\ This interpretation re olves. in panicular. all those difficulties that arise when a given glossematic predicate (for example, the rh ore m of the independence of substance from fonn i applied to linguistic language or to other specific languages. In de cending to the level of concrete sign systems glossematics ceases to be glossematic , that is to say, a science of pure form. 7

In certain respects, however, Hjelmslev him If appears to be apprehen ive about the radicality and audacity of his reasoning. A symptom of this i the fact that only research into linguistic language is published in his Acta Linguistica, even though the journal was conceived as the organ of the all-embracing langue described above. He explicitly admits to this limitation 8S a concession forced by practical considerations. but this conce ion cannot be justified in term of his own theory. In other word Hjelmsl v himself has not carried his radical theory through to its logical conclusion. He appears to be implying that glos­sematics can at the same time be both a general the ry of signs and a theory of language. But this distorted self-evaluation of glo ematics prevent the truly novel aspect of the th ory-the idea of a general th ory of ign -from bearing fruit. Funhermore, it create new difficultie , such a the generalizati n of lin­guistic signs into signs per se-a question we shall go into later. Yet in pile of Hjelmslev' own evaluation of glo sematic and the problem arising from his conception of signs, the appearance of glos ematic must be seen as an imponant point in the development of modem semiotic for it marks the point at which linguistics meets the broader science of semiotics. In

Go ogle books classics 01 semiotics

~ ~ Cl Contents ~ Page93 aa

1.5.

Giossematics as semiotics represents not only a transition from one plane of abstraction to a higher one, but also the expansion on the higher plane of the data gathered at the lower level . For in glossematics. the structure of the linguistic sign is taken explicitl y as the basis of the general sign structure: That is to say. the only things I(;cepted as signs are those things that are structured in the same way as linguistic signs. In this respect glossematics is an agent of the so-called " linguistic imperialism" that has been increasingly challenged by those semi­oticians who feel that it is necessary to emancipate semiotics from its parent sciences .

Even if it is no longer possible today to grant the attribute "sign" only to those structures that. like linguistic signs. are "double articulated,"" there are still good arguments in favor of regarding the linguistic sign as the nucleus and guiding principle of semiotics.ll This does not imply . however. that features specific to the linguistic sign should be elevated to the higher plane of abstraction of signs in general. as Hjelmslev has done .

2. THE DOUBLE·ARTICULATED OR TWICE·FORMED SIGN III

2.1.

Something can only be called a sign when it is possible to distinguish in it a signifianl and a signifii. In place of the Saussurean lenns Hjelmslev introduces the terms ~xpr~ssion and cont~nr. which he calls the plan~s of the sign . In a funher interpretation of Saussure the teoos/orm and subs/anc~ are inlroduced: Language . according 10 Saussure. is a form that initiates distinctions in both the "amorphous mass" of our thoughts (content) and also in the "plastic matter" (mo'itr~ plastique. Saussure, 1916: 155) of sounds. Just how Saussure imagined the formahon process of these two formless substances is shown in Figure 2, from the Cours (Saussure 1916: 156). Saussure wanted to portray language as on~ synthetic form located between two substances . Hjelmslev. in contrast, emphasizes the differing formation of the two planes of expression and content,

Figure 2

•

Go ogle books classics 01 semiotics

~ ~ Cl Contents ~ ~

Page~ aa _Link ErFee<Jboc k IV"'VL I ~ 1rv.,",I ~ I

and defines this enlity created out of two differenl forms as the definitive model of the sign.

T11e image used by Hjclmslev to illustrate the relationship between form and substance is that of a net (form), which is thrown over the unformed maUer (called by Hjclmslev purport-also on the plane of exprcssiorl-()r matiir~). so that a form~d substanc~ is created by projecting non-substantial fann onto the formless substance. 11 is necessary. therefore, to distinguish. on both planes of the sign. between matler (formless substance), (formed) substance, and (non­substantial) form . This leads, in Hjclmslev 's model of the sign, to the followi ng six major components: expression-purpon, expression-substance (ES). expression­form (EF), content-form (CF), content-substance (CS), and content-purport . These can be represented in the following diagram (with the particular inter­depcndences indicated by arrows);

Expression-purport- ES-EF++CF-cS-Content-purport

Following on from another of Saussure' s maxims, the sole object of glassemalics is the pure, non-subslanrial form .

2.2.

2.2.1.

The relationship of fann. subSlance and maner can be illusltated using Hjclmslev' s now classic example (see Figure 3) from the area of cont~nt. con­cemi ng the naming of colors in English and Wel sh (Hjclmslev, 1943/63 : S3) . A segment of the color spectrum as linguistically unformed maller is symbolized by a rectangle (I) (although the delineation of the limits of this area of reality is actually inadmissible, as it already seems to symbolize some sort of forming process). 1M: varioos languages now throw their various "nets" onto this unfonned area of reality (11). The net thrown by the English language over this area (solid lines) is different from that thrown over the same area by the Welsh language (dolled lines):

11 III gwyrdd

f----- -- - - green- - - - - -blue glas - -f------- --grey -- ----

brown IIwyd

Purport CF CS fig ure)

• l

GlOSSEMA TICS AS GE ERAl SEMIOTIC 9S

The area is iven form in this wa a a t tality. and fonned ub lance are created a divi ible. di tingui hable part of the reviou Iy unformed u tance. In th Engli h language the c nt nt-~ rm thro n ov r the matter produces di -tingui habl colors called r en. bill. re. and bro n.

2.2.2.

It i n t alway correctly under tood. in interpretati n of glo ematics. 1

that content- ub tance i . amon oth r thing • what we tend to call 'meanin ." But c ntent-substance aI includes the so-called referent" the actual peel of reality designated by the meanin . According to Hjelmslev, the difference between meaning and reference is something that lie \ ithin the content-sub tance. and not correspond to th differe e een c ntent-form and content-ub tance. In fact Hjelmsl v him If 1954 differentiate within the u tance

betw n phy icat. a iobiological. and a c 11 tive-appreciativ level. whereby th physical level of the content- ub lance approximates to the reference and m mer two I v I to different c nc pti n of mean;n . In con t. glo malic cont nt-form i only form. It i not 'meaning • although constantly mi taken to be : It i only the framework. the net. the con tellati n of difference. Whoe er mi und tand this mi und tand th real int ntion f glos matic n mely to be a di cipline that is independent f all ub tance. c ncemed nly with pure form.

Thi primacy of pure form i based on two Sau urean t n t : firstly on hi in i tence that lan uage is fonn and not ubstance, I~ and ondly on his d finition of th formal charact r of lan ua e, which i d rived f m t vi w that in langua e there i nothing positiv only dif~ rences Sau u , 1916: 166 . Language as pure form is a radical. a priori hypothe is offered by Hjelm I v a an interpretation of Sau u . Th tru i t f glo matic. it ~ Bow • i th lan lie as pu form. Sub tance i excluded from glo matic c n ideration that i from Hn uistic seen fr m ch gl sematic viewpoint, and the tudy of u -tance is assigned to other. non-glo mati. that i t ay non-lin ui tic di­

ciplin s. Semantic a the tud of m aning-th tudy of cont nl- ub lance­i there y excluded compl tely from lingui tic -glo sematic . I

2.2.3.

The example of semantics makes it clear why glo semalics create diffi­culties by claiming to be a linguist; science rather than a general ience of s; n . If we regard glo sematics, in the way Hjelm lev ants us to, a 'lin­gui tic :' then we are confronted by the enormous problem posed by semantics. Ac ording to Hj Im lev. emantic. which with ut any doubt i an peel f lin ui tic must be regard d as a n n-lingui tic di cipline ince it studi sub-lance. But if we regard 10 sematic a general semiotics then the excJu ion

96 JORGEN TRA8ANT

of semantics from glossematics ( - semiotics) becomes entirely logical and sensible since a generaJ theory of signs is not concerned with meaning within a particular "language" or sign ystem. but rather with the formal structuraJ laws that might govern the content of the sign in general.

2.2.4.

As a cience of general sign structures. glossematics is a science of the­oretical possibilities and not of manifest reaHties. HjelmsJev's view of glo se­matic as a type of algebra should be viewed in this way. for like algebra. glossematics is a discipline of possible theoretical constructs that do not have to be made manifest in particular sub lances. Thus in the previous example. it is not the EngJi h color names that are suitable objects for glo matic ( ince they are sub tantial forms' but rather the framework used in the diagram. as a possible way of structuring a given content area. Other purely th oretical ways of structuring the same entity can be seen in Figure 4. The e theoreti ally po ible constructions mayor may not occur in reality. As Hjelmsl v him If ay with regard to his science of the general langlle: "All possibilities must h re be foreseen , including those that are virtual in the world of experienc • or remain without a 'natural' or 'actual' manifestation" (Hjelm lev, 1943/63 : 106) .

2.3.

2.3.1.

On the expression plane too, form acts as a net thrown onto the initially unformed matter. For example. if one imagine a unformed matter the potential for producing sound by means of the vocal organs. then it becomes clear that every Janguage form this matter in different way and impo e different limits. One of Hjelmslev' example concerns the continuum formed by the median profile of the roof of the mouth from the pharynx to the lips. as the area where

Figure ..

ClOSSEMATICS AS G N RA 5 MIOTIC'S

the vojc le occ1u ive con· n n are articulated. In th u uat Eu n an­guage thi matter is di tribut d over three area namely l pl., III and IkJ w ttPr .... Cl

the kimo and Latvian language di tinguish betw n two IkJ areas and many Indian languages have two III area (Hjelmslev 1943/63: SS).

Once again we get theame picture here WIth the e ntent. The unformed matter i eparated by the form into three area . becoming formed ub lance or expre i n- ub tance. Here t pre ion .. ~ on i not the phn 1 gical rh .. . netic tructure of a .Ianguag • . ut only the n n- ub tantial net thrown er th unfonned matter. Like the mantic and referenti I peet · on the c ntent p.1 ,

the phon logical and phonetic :peet 1 ng to the .b nee: lh different plan within th expre ion- ub lance, 11 and fore n a maner for glo ematic.

2.3.2.

Even n the expf i n pi n the in i t.cnoe that the n w di ip.line uld occupy it If excluively with fann i nly tenable if we locate gJmatic a ience f ign on m re g n ral I vel . F r ageneta)ci ne of ig gl emalic has nothing to do with the fonning of a u tance th ugh a par-ticular . ign y tem, and it follow from thi that ph .. Iogy and ph tic . ' the specific investigation of the ifte ign y temlanguage. would not then be glo ematic discipline. Ont other hand , if phonetic and n I BY e to be banished from lingui ti . on th ground that they were noo-Iingui tic then uch Slo ematic fonnaH· m would rightly be seen as ab urd.

As Co riu h pointed ut. ignoring tb expres ion- ub lane of natural languages is even more anti-lingui tic than ignoring the c ntent-sutance. F r the material manifestation of language as an entity in the world ' the ound; that i to ay, th phonic. ub tance i lhub tan. e of language.. With ut expres ion-sub lance the linguistic language would imply n exi t : "'The an­guage form .manife ts itself and is immanently extanl in the ex i -so tance. The cont nt-sub tance-the world in general-at exist without language" th world "neither manife t n r makes materiaJany langua form' . through this fonn the world i merely organized. In other words , the world i made percepti.bl through Janguage but also exists without the language form (Co· riu, 1954:2 0 .

2.4.

A expteSsion- ub tanceand content- ub lance are not nece arilylinked to expression-form and content-fonn re pectively (the fonn pure virtualiti that do not needsu tantialmanifestation). I the gl malic ign m el i ultimately c m posed of only two f Hjelm lev six maj r component namely expre ion-form and content-form. The ' ign function' he ch eteri tic

98 JORC£ TRABANT

dependence of igns-is the relation hip betwe n expression-form and c ntenl­form . A in Saussure's model this relationship is clo e and reciprocal. 10 The ign in i If demateriaJized into pure fonn. remain the true object of glossematics.

2,5,

The need to distinguish between the two fonns of the sign. the content­fonn and the expression-form. only occurs, of course. when the content-plane and the expression-plane are actually formed in different ways. There are struc­ture in which the expression plane and th content-plane ha e one and the ame ~ nn: Hj Imsl v call th m ., ymbol .' Th y are interpr tabl objects to which a content- ub tance can be assigned. but no content ~ nn that differs from the expre ion-form. Th interpretable ntitie serve in Hj Im lev's argument as a background again t which the specific characteristics of the sign. namely the differing forms of the two separate plane. stand out particularly well. Signicity con i ts not only in the mere distinctiveness of the two planes of expression and cont nt in symbols too, expression can be differentiated from content). but in th different fonn of the two I vel.

Thu the expre ion-plan of the verbal ign Tisch table) is con tructed in a different way from that of its content. The expression-plane is made up of the unit If} l il and I , while the content-plane i made up of the unified idea­complex Tisch. Even if on were to analyze the content of Tisch using various aspect of the content, such as "lab) ,.' "singular:' "nominative:' it would still be impossible to assign each of these three content a pects to a corresponding unit on the expression plane. for example • table" to IU. U ingular" to lil. or "nominative" to I~. The content-plane ha a different form to the expression­plane.

In contrast, there exists with symbols a one-to-one relation hip between the expression-plane and the content-plane. An example is offered by chess pieces, in which every element of the expres ion-pJane. that is every chess pi ce. corre pond to a unit on the content-plane. namely a particular function . In this case, therefore, it is unnecessary to submit the expres ion-plane and content­plane to separate analyses. as both plane present the sam form . Hj Imslev gave other examples in a lecture delivered in London in 1947: the hourly chim s of a clock. whereby the number of chim c rresponds to the number of hours. and the lights of a traffic signal, in which each color corresponds to a function "Stop!:' Go!,'" Caution!" . According to Hjelm lev, symbol are not sign,

a they do not display any difference in form between content and expression­indeed. he pecificaJly calls them "interpretable. non-semiotic entities" (Hjelm -lev, 194 163: I) 4 )-and thus are not the concern of glossematics.

Hjelmslev's definition of symbols embraces th iconicity. the similarity between si n and denotatum. which Moni sees in iconic signs (Morris, 1971 :420 ,

GlOSSEMATICS AS GE ERAL SEMIOTICS

and al 0 the 'motivatedne · of Sau urean "symbol: A i h n in lhe

example of the chess pieces which share neither "similarity · nor · natural moti­vation with their content (the che' knight no imilarity to his function), Hjelmslev s definition goe much further and i much cl arer a it redu e both iconicity and symbolicity to the level of structural feature that i 10 the ~ ture of ingle-anicu)ation in contrast to th d uble-aniculati n of ign .

2.6.

The semiotic problem posed by glo sematic when re arded as a g n raJ the ry of sign lies as we ha e n in the fact that the tructure of t lingui tic sign the existence of different fonns of expression and content the double articulati n i con id red to a fundam ntal characteri tic of ny ign. Thi partkular variant of "Iingui tic imperialism is the result both of the strang Iy hybrid asse ment which glo ematic make of itself. as described above. and aloof the glossematic interpretatj n of S u ure's po tu)ate of immanence. an interpretation that i radicaJ in two ways. Fi t. it exclude all 'external c n-id ration . particularly ychol gi al. ocial and thn I gical c n id rati n •

which in current tenns could be called the exclu ion of the pragmatic dimension. Second it exclude sub tan e (fonnali m). If th pragmatic dimen i n remain suppre sed and aspect like human conve ation intentions of the commu­nicators. and the actional charact r f sign do not come under rutiny but purely the sign alone. the sign a a thing, then the immanent tructure of thi reified entity must provide it definitional criteria. Furthermore. c n­cem for substance for the materiality of the ign-entity i excluded the way in which the sign is articulated remains for Hjelmslev the sole criterion of signicity .

That thi is a bad criterion of ignicity can be demon tTated from the br der pragmatic viewpoint u ing Hjelm lev's own xample. S en from th if actionaJ a peel, the traffic ignaJ that Hj Im lev coun as ymbol nd lh exclu from glossematics- emiotic should with ut doubt be regarded as ign. even th ugh they are single-articulated. In contrast. th action I character of telephone numbers make it difficult to agree with Hjelm lev's propo ition. mad in the 1947 London lecture, that such numbers are ign. 20

Even if contemporary emi ic re earch i not reduced to the imman nt synt elic .. semantic approach of European tructuraJism an tri t look at rh sign from a pragmatic vi wpoint, the immanentist approach has achi.ev d a deeper understanding of the inner tructure of signs and created an indi pen ible ba i for further work. Nobody before Hjelmslev had compreh nded 0 clearly the tructure of single-articulat d and doub) -aniculat d ntities. Dial tically con­

served in a pragmatically-based semiotic theory. these differences in articulation remain important feature ~ r a ign cia sification. and important crit ria of th

~ ~ Cl Contents ~ Page 100 au _ Link E}' Fee<lback

100 JURGEN TRA8ANT

efficiency of sign systems. 11 is not mere chance that a double-articulated sign system, namely "natural" language. is the richest . most important. and most universal o f ,11 sign systems .

3. CONNOTATION AND METASEMIOLOGY

3.1.

In the following section I would like to discuss two funher consequences of the highly immanent and fo""aJi stic nature of glossematic sign theory . Both are releyant to current semiotic research, and are concerned with aspects of semiotics that are explicitly excluded by glossematics.

3.1.1.

I. 1be (general structuralist) exclusion of "~xlernal." in the wider sense pragmatiC'. facts leads to the characteristic glossematic doctrine of the C'onnotaliv~ sign .

2. 1be typical glossematic exclusion of subslanu makes it necessary for the analysis of substance to be linked to (fo""al) glossematic analysis [!J

as m~lastm;o(o8Y'

In spite of Hjelmsley's exclusion of the pragmatic dimension (whicti is by no means unique to him), he is not blind to the reality of language and to differences in languages and sign systems that derive from real speakers and real listeners. Hjelmslev tries to link this awareness to the specifically structuralist immanence postulate!' by incorporating the pragmatically determined differences in semiotic material (differences between nations, classes. regions, communi­cative situations . and indiYiduals) by means of his theory of the connotative sign . according to which the excluded "external" information about the speaker and listener reappears as an immanent quality of the sign . This takes place on an additional conlml-plant , which vaults over the dtnolal;\'t sign to form the expression-plane of the C'onnolal;\'~ sign . As an expression of the connotative content can serve the substances or the fonns . or the fo""s and substances of the denotative sign.

By "connotation" Hjelmslev is not referring to the so-called t molil,t mtQn­ing. the psychological appraisal of a particular thing.11 but to something far more general, namely everything that the sign says about the speaker: "He is an Englishman , he comes from London. he is a linguist. he expresses himself carefully" and so on . Following the immanence postulate, this infonnatio n about the speaker is comprehended as something that the sign says about itself: '" am

(:0 '" Ol

~ ~ Cl Contents ~ Page 101 au _ Link E}' Fee<lback

GlOSStMATlCS AS GENERAL SEM10T1CS 101

English , from London, linguistically employed, rept"e5ent refined speech" etc . In this extended concept of "connotation" every sign participates as a matter of principle in several connotative semiotic structures .

Now if the connotative struclure is 10 be a sign , il must be built like a sign and must have lhe strata of a sign: expression-substance, expression-fonn, content-form. and content-substance. Above 811 . it must possess diffaently formed planes of expression and content. that is to say a connotative expressionlorm and a connotative contemlorm , as in Figure S:

C1n. cs

COIl" . Cf

I 'To" \ conft. ES

tl'Ml c1eno"Uv, .Igft :

den . ES_den . EF_dtn. Cf_c\tn cs

Figure 5

The assumption that the connotative expression-plane and the connotative contenI.-

plane are differently formed , in other words the assumption of a sign-like structure [!]

of connotation. is one of the most problematic concepts in glossematic ~ign theory .2.1 For it must be questioned whether Ihe connotative entities do actually manifest different ronns of expression and content, or whether they are symbol­structum;, that is to say "inteTpretatM' structures 10 which only content-substances can be assigned , standing in a ORe-to-one relationship to the interpreted entities. like the traffic regulations to the traffic signals . As a simple and illuminating eltBmple one can assign to a given word or text the connotative content-substance "Gennan." that is to say the "actual nolions of social or sacred character that common usage attaches to the concept 'Gennan language' "(Hjelmslev . 19431 63: 119). Taken this way. this relationship would be identified as a "symbol." But since it is supposed to be a "sign," a content-Jorm must be assumed.

What is the fann ofthe connotative content "Gennan"? Hjelmslev's answer would be: the mutual dependences (junctions) of connotative contents . for exam­ple the relationship between the content "Gennan" and the content "French," "Russian" and 50 on. Unfortunately. Hjelmslev does not indicate what such a fonnal. mutual demarcation of connotative contents would look like .l-O

Be that as it may, he does put content-fonn in a connotative context, and here too the real glossematic analysis of the connotative sign is an analysis of fonn . After the analysis of denotative signs (seen as connotatively homogeneous) is conclUded, the connotative contents-which had hitherto been systematically excluded--are exposed to the same formal analysis: "After the analysis or the denotative semiotic is completed. the connotative semiotic must be subjected to an analysis according to just the same procedure" (Hjelmslev, 1943/63: 119) .

•

102 ,URGfN TRA8ANT

The gJossematic interest in the connotative sign is an intere t in the forms of the connotative sign se Figure 6 :

conn. CS

{cot. CF glossematlc analysis o~ • the connotative slg" t

conn. EF

t conn. ES

figure 6

The connotative sign-form does not itself contain the (substantial) interpretations but only the rather dubious mutual d pendences of such interpretations. It i . in other words an "empty fonn. For this rea on. the excluded pragmatic infor­mation it If i not considered in r course of the glossematic in the narrow sense) treatment of connotative signs. This is only done in the analysis of con­notative content-substance.

Allhough the analysis of substance is not a g/ossemalic task, Hjelmslev doe not banish it entirely from science. but rather categorizes it as a separate. non-gl0 sematic discipline that has to be linked with the glossematic analysis of form. This holds both for the denotative and the connotative sign. Just how thi anaJy i of subslance i related to the analysis of form will be shown in section 3.2.

3.1.2.

Today the pragmatic and actional character of Janguage and of signs is considered to be of such central importance, that attempts have been made to define signs in tenns of actional trucrures. In such attempts the sign and the languag are from the outset much more intimately linked with the acting man than in Hjelm lev 's structurali tically immanent m thod. in which the pragmatic aspect of igns is admitted to exist. but regarded as an extra dimension of the ign that disturb the functional homogeneity of the object and is therefore initially

excluded from the true core of linguistics and semiotics. Only after it has per­formed it own pecified task doe glossematics turn to the connotative a pect (and even then only very formally!).

Clearly "denotation " essentially meaning "meaning." and. on the level of speech · reference" (or what BUhler has called "representation'). is the principal function of the linguistic sign. In this respect the characteristic method of truc­turalism in proceeding from denotation. and the neglect of the addilionaJ. "prag­matic" apect of communication is understandable. Yet it should al 0 be empha ized that the act of denotation the representation of reality by means of the sign . is also an actiona/ proce s. And it is an action carried out with respect

GlOSSEMA TICS AS GENERAL SEMIOTI S 10)

to the others 0 tbat the dimen ion of acting pe ons--the pragmatic dimen i n­cannot be een as something added or extra but as omething es ntia) to the sign .26

3.2.

3.2. 1.

Although he ha bani hed ub tan e from the glossematic paradi • Hjelmsl v neverthele s considers the analy is of the ub tance of si ns to be an important and socially significant scientific a tivity. He does not con ider it to be unscientific merely unglo enuJtic. In hi irnpre iv Vice Rectorial addre of 1953 on lingui tic cont nt-f, rm a ociaJ factor Hj Imsl v. 1953 h d w attention. for exam pi • to the political rei vane in our time of sci ntitic re arch into meaning content- ub tance . Hjelmslev relal b lance analy i to the glossematic analy i of th pure form of language and ign in a complicated way, namely via a third-le el sign tructure the so-called metasem;olog .

Linguistics-glossematic a i ntific metalanguage, whose object-language is a sign sy tern, is a second-level tgn structure similar to the connotative sign. (n contrast however to the connotative ign. in which th expression-pia i made up of sign (out of the sub tances or forms of the in. r both. Hn ui tic as a sign structure ha a content-plan of ign but only the form of th in, a the ubstance are by definition not a matter for lin Ut tics. Th scientific rneta-sign- truClure is called by Hj Im lev emiolo . Linguistic j a miology who e tructure refie t the pattern in Figure 7 whereby we are differentiating here in m la-sign only between the pJanes and not. in addition. belween fonn and substance):

den. ES~den. EF~den. CF~den. CS

• semiological­linguistic content-plane

t 5 mlologlcal­linguistic expression-plane

Figure 7

Now it is po ible to peak from a higher plane about semi logy it If: the scientific meta-lang\1ag semiology, lingui tic becomes, in turn, a c nt nt­th obj Cl of di cu sion . Th re ull i a third-le el ign lru tur • m tas miolo ...

104 JURGE TRABANT

In this di cussi n on semiology it finally becomes possible following Hjelm lev. to speak about those things that had to be excluded. that is, the sub lance. Since a discussion of the semiotic treatment of the linguistic form would merely repeat what has already been said, metasemiology can even be dedicated primarily to sub tance. The substance that were excluded from the gro ematic treat m nt of object language can finalJy be d ribed on a meta-plane. Figure 8 depict thi complex relationship. The disciplines that describe the den tatlv sign substances­phon tics and phonology on one id. and semantics on the other-are not integrated into linguistic-glo sematics. but Jayered as it were on top of it. The semiotic di ciplines concerned with ub tance emerge in the hierarchy abo e the layers of semiotic analy is c n emed with form. Phonetics/phonology and semantic are not the customary partial metalanguages. but partial meta­metalanguage .

But to regard phonetics/phonology and mantics not as straightforward linguistic disciplines, but rather as meta-inve tigations loo ely related to Hn­gui tics. i a total distortion of normal linguistic research. The metaglos matic rescue operation aimed at saving substance analysis, and the attempt to reunite with glossematic tho e disciplin s that in any ca are barely conceivable as non-lingui tic would all be unn cessary if HjeJm le did not persist in viewing lingui tic language and other concrete sign y terns as pure form. If he ere to retain on th plane of pure fonn only the general semiotic tructure. the langue a described above in 1.2 then the lingui lie language could ettle wher il

~ ... den. EF ~ .. den •

phonetics phonology

.. emlologtcal - Ilngulstic

content-plane

~ semiological-linguistic expression-plane

metasemiological content-plane

~ metasemiological expression-plane

Figure 8

semantics

G OSSEMA TICS AS GENERAL SEMIOTICS 105

long. n th low r plane of fonned ub tance • and phonetics/phonology and semantics would be what they actually are. namely lingui tic di iplin .

3.2.2.

A we have n the true glos ematic study of c nnot tive ign limit it elf to rhe analy is of form. The study of the excluded ub tance i related to the analysi of pure fonn (sub tance here mean. content- ubstance as the n­not tive expre ion- ub tance i anaJyz d in any ca , ince it i the d n li e ign in the me way a with d notative ign t via a meta em;olog ': HJu t as

the metasemi logy of denotative semiotic will in practice treat the objec of phonetic and semantic in a reinterpreted fonn 0 in the metasemiotic11 of connotative emiotics, the large t part of specifically sociologicallingui tic and

au urean external lingui tics will find their place in reinterpreted fi rm t

Hjelmslev t 1943/63: 125 . The metasemiology of connotative signs analyz.e the variou "geographical and hi t rical, political and social, acral and psycholog­ical c ntent-purports' ibid.), which are linked to the nation, region, tyle. personality mood, etc. Hj Imslev singles out ociology ethnology and p y­chology a th di ciplines that c ntribute m t to the analy i f conn tative content-sub tance.

Figure 9 a rep entati n f conn talive meta miology i complicated by the fact that the connotative sign i it If a second-level sign tructure, making conn tative miology a third-level sign tructure and connotative metasemlolog a fourth-level sign tructure:

cotn. CF I conn . onn. lem 0109 ca' ... -.... lologlca'

xpr IIlon-conn. EF content- plan plane

t conn. AS ...

den. ES~den. EF ....... cMn. CF ...... d n. CS

Figur 9

~'o IIngul,Uc • • thno-lIngulltlcs p'ycho-lIngulltlc.

la- ..... la-lot. 101.

cont nt- •• pr ,ao,,-plane pia

The criticism made above with re peet to phonetics/phonology and semantics can also be applied in part to the analysi of connotative sign-substance. For glos matic fonnalism d mands that substance which has been excluded from the formal core of semiology be approached via a high r mela-plane which would be entirely unnece ary but for the a priori exclusion.

106 ,ORGE TRABANT

On the other hand. it cann t be aid that the disciplines de cribed as "connotative metasemiology' are generally considered to be linguistic disciplines in the way that ph netics/phonology and mantics are. The e di cipJin • for which names like socio-linguistics. ethno-Iinguistics. psycho-Jingui tics. pragma­linguistic • and 0 on have been found. are certainly not regarded by many linguist a belonging to the inner c re of linguistics. Although no other direction in modem lingui tic has supported Hj Jmslev s extreme fonnali m it ha never­th I been characteri tic f Jarge clion of tructurali t research from Sau ure to Chomsky to exclude from true lingui tics the so-called "external factors"­the pragmatic element taken in a very wide sense. HjeJrnsle • propo al of a connotative metasemiology. in contrast. is an attempt to overc me the xclu iveJy immanent viewpoint. a fundamental attempt, as Coseriu ha pointed out. "to comprehend all human and culturaJ problems in tenn of language." Today' hyphenated di cipJine could be een as further. single step toward overcoming rh immanent viewpoint. Tb hyphen in rh current titles of these disciplines "symbolically" reflect the fact that these efforts are still regarded as som thing alien. grafled on to the true inn r core of lingui tics. Which, in the structuralist tradition. still prefers to exclude th disturbing facts concerning the speaker and the listener,

But only from a Jinguistic tandpoint that integrate these 'external" factors i it po ible to supersed the oPPO ition of "external" and "int mal.' Only when this opposition is removed wiJl the scientific (reatmenr of the "externaJ" facrors of language becom a central concern. and only then will the anaJysis of the connotative content-sub tance emerge a a true linguistic endeavor rather than som thing peripherally related to linguistics.

NOTES

I. Cf. H,;elm le . 194 16 : 109 f . 2. Hjelmslev, 1944. Sec also Chapr . 21 of the P,oltgomtna. which argu Ih t gl semalics i

c ncemed jlh "lanlEu lEe in far broader sen .. (102). namely with "an structure whose form i an I g u 10 th t o( 'natural' lingua e" (102). "any semi lie- an slruclure thal i analogou t a langulse and li fie the given definiti .. (107). Thu gl malic. "lingui tic in the bro der sense:' is .. 'semiolog . on an ;mmantnt basi .. ( 10 ), the explicit realization of the Saussurean semiology desiratum (107 r.)

3. On Hjtlm lev' c ncq>ti n of'" nguage type," see Hjelm lev, 1968: I J 2. 4. C nu's pi cement of SIU sure' "racull~ Jin,ui tique par excellence" on the same plane of

I lion H.;elm le\" lang"t (Co riu. 19S4: 201) hould be qualified. In m vie Hjtlm· lev' lan/f"t i located one pi ne r cj n higher.

S. See Coseriu 19S4: especiall 201. 218. 6. 5e"e I e 2.2.3. and 2.3. AI compare Coseriu. 1954 and Trabant. 1970. 7. Cf. Coseriu. 19S4. e pecilll 227 fr. 8. "rei 'impose cepend nt une ~rve d ordre pratique. C'e lala"I",/i" 11; l iqw ui c itue

le d maine de noue re ue. On ne il pas dans quelle me ure cene re triclion. im~e pat I. lradilion. correspond i une realit~ , .. C'e t tin i que n Ire org ne con ere ~ I1 langue

~ ~ Cl Contents ~ Page 107 D ..

linluiSliquc , ~t qu 'il n' ldme«1'I lel reo:hero:bel l ur les languc l IIOfI Iinguistiquel que clans II

mesu~ ott ~$ reo:hen:hes conlribuent direclemenl 1 J'4!rude linauiSlique proprement ditt"

(Hjelmdev. Ill«: 321-9. Ir HjellT\$lev had relll y ronceived his Ac,u U"I"iJliro tu "A~u. Sem;Olk:I ,~ then resell'l:h into

semial icl would ha~ twn I quite d;ff~nt course. hlvinl hid its own ;oumlll fOf" thiny Y«n prior to the Ippearlnce of S,miQ/lc:u (i.e . sinct: 1939).

10 Apen rrom SIII5sure ', prrpemnry comments on I !-eneral sden«; ohi!-"" , glossematks remains

the principill European contribution-which is. 11 the same time . the contributkln ortinpistic_ 10 modem gerxl'll semialics. 1M American contributions 10 semiotic. derive fmm philosoplty

Ind possibly psy~hology .

11. cr. &:0. 1972:2)1ff. 12. Throu.h the Inalysis of linguistic signs, 1 have abstractcd ccrtain ectionll criteril IS definitive:

critenl for SigM in gtrxral : see Tnlbant. \976. 13 . For eumple Gre imn, 1966: H . 106; see llso Tl'lbint , 19TI:9f. 14. Saussure 1916: 157. On the various .... 1)'1 rA inletplelinl thi s sentenee see Co5triu. 1954: 220. 2JI . 15. Hjclmslcv. 194J16J: 79: " LinJUiSlics must sec its main wk in esublishi", IIC~ ordleeJtjR5Siort

and I science or the content on an intemll aNI functional basis; it I'I11I$I esIabIiot,h I science of the CXprtS!1on wilhout ",vinl re<.'OIIne 10 phonetic or phcnomcnolotlic&1 pmnisse5. I science or the content wilhout nmulos ic&1 or phcilo",enological premisses ... Such alinguiSlKl, tu dillinJuished rrom CQII~nlionaIlinlllistit"s . would be one ..... hose science of the cx~ion is not. phonf:tics and whoot !l("ienec of the (.'(Jftt!It it. not 1 Kmlntic:s."

16. Coseriu. 1954: 216: MAs JOOn as he introduces t.famplt s he lelves this plane (of a\&dnJ, and cannot IVQid tIw: difficulties thu &rise fmm his &aempt 10 apply the Ill@eln or ~ form 10 'fonns of subswtcc' (i.e. the cxunplc5 1 .~

17 . The abstrICtkln level of the Pnrogue $ChooJ.·s phonology comn off t.i1y whh Hjelldev. who dismisses cXlct.ly th is level of runctionll language description 15 -ficlion.~ th.II. is 1$' nonscnsical uep in the formali:r.alionoftheeJjftSSion·pla:nc.as • ~ wbicto Utould , therefore. be el iminMCd. For him only the phonetic plane is vl lid IS '\I5&ge" (see Hjelm5lev I94J; ssr.).

18. Atoordin. 10 die glos.scmadc d assification of functions , the reIMionship bctwem subsanr:e and form il; onc of Mdccamination." thaI il lO say. I rel:i1lionship between one vai&bk- (subsImcc) and onc t'OI\5IInt [form) rsec Hjclms~ 194J/f!3: 41), whicto is l ymboIiud by 1II1n"O'W_.

19. A relltionship betW'Cttl tl/o'O constants , an " intt .. :lepeIlOenec" is Iymboliud by I double·he&ckd

~-. 20. For an lnempllO defirx the OOIICCpt of tIw: si!O on the basil or lCtionll theory , JO: TrabMt. 1976.

Thm arc CCfUinly sections in Hjclrmlev' s writinp in whicto sipity is nee fully idenlified 'NidI tIw: iTllcmal SlnlCture or the doublc· llticulMCd sign. but wttere rIlher the function or MJ/, nDliot! . or the "relationsh ip 10 the non-lin.uistic flt'lOn which SIIITOUnd iI~ Ippears 10 be liken IS definitive

(or signicity ;n ~ (HjelmsIc Y. 1943!6J: 471. This opposilklcl or I wuctunI ~"'" IrtmO 1IJ1u- "ldiol\ar definition of , igM iI imimaled in the fonnulMion or langu&.&e 11 a '"fipte system with siJll·purpose" (ibid.: 1021.

21. Cf. , for cxtmple. Hjelmslcv. 1943!ti3: 4ff. 22. Acrording lO Hjehnslcv. 1954; 60. lIIis is nther an clemcnc of the dentlCllive: C'OI\tIent.subscancc:. 23. I ... ..., mywlf tppIicd tho: idea of the OOIlOOlaIi..., sign 10 the Sln>CtUrC of literary IeXts; 5« Tr&bwtc ,

1970. 1 now htve doubts. on the onc hand . ..... hether litmvy ICxts should. in tIw: conlext or ac:rioftCII 11wot)'. be re,arded IS signs It &11; 5« Trabant. 1976: 95ft On the othe-r hind. tIw: Jlrwfllrul

signicity orthc connogtive siln. thI1 is 10 say Ihe lSMImp:ion or. CQIIntlCIlive contelll. rormdiffm", rrom the form of the expression on ..... hich the siplicity of the 00IUI0IIli~ ~gn depends. $ttIt'IS

equally probIetTlllicaJ . 24. Cf. Hjclmslev. I94J1tJ; \1 9. The ~Iationot,hip between conootMive conICT1I5 is probIbIy I ~I.·

tionot,hip betwccn variants n'oMlellatioln. AccordinllO the Ilos.sclTlltic inlt'Iutation. ~ reI&­tionsh ip$ arc not scientificllly dncrit.blc 11:1 !hey unoont 10 the ab$mcr; of depelldcnecs . Cf. Hjelmslcv. 19S4I63: 83. Il!d TrabIn! , 1970: 29. n . 28 .

•

~ ~ Cl Contents ~ Page 108 au _ Link E}' Fee<lback

108 rORGEN TRABANT

25. For eurnple T ....... , 1976. 26. On cM ~laIionship of ~ 10 cM cxher semiotic dimensions. and in panicul .. 011 !he Minle­

pliveM pnII!I"IMic: sundpoilll. tee Tr.t.nI , 1976: 46{.

27. Hjelmslev is nor quile precise hm: As !he Milym 0( the form of conncntlYe signs is ilself I Mmetaemiotic 01 connotMive sians," Hjelmslev should speak 01_ "mela-melasemiol ic 01 connotItive

signsH or of I "mcClsemioklgy of conllOUllive signs." 28. Hjelmslev reprds Ill)' ~levds ~ the level 01 mewemioloJy 10 be meanillJles5 (Hjelmslev,

1943163:125).

REFERENCES

Coseriu. E. (1954). HFonnI)' suStlnCil en leas sonidos dellenglllje." In: E. Coseriu, Ttor(Q dtl IttlllUJjt Y li"liif,'ira ItMral. Madrid. 1962: 115-234.

&::0. U. (1972). EI"fiihflUll ItI dit StmlOlilt. Munich. Greimas, A. I . (1966). Slmtulliqw ,",fe·llInllt . Plril . Hjelmslev, L. (1943163). Prolt~_rIQ la a TI/n)ry of LAtI~lIOlt . Madison. Wisc. 1963' (Dlnish:

I~I ~ Hjelmslev. L. (1943). " t..an,ue Cl piI"Ole." ln: Hjelmslev. 1971 : 77-39. Hjelmslcv. L. (1944) . "Lin,uistique SltVdUraIe." In: HjtJmslev, 1971 : 28-33 I - Edilori_1 of Ani

Lin8uistic_j . Hjelmslev. L. (1947). "u s~ure fondimentlle du I_nllle" (French Innsl_lion of _ London

ie<:ture of 1947). In: L. Hjelmsiev. Pro/I,ombtts d IItlt IMor/t dll latl/rQll. Parh, 1968f71 : 177- 231 .

Hjelmslev. L. (1953). "1I forme du conlenu!lu IIngage comme fXleuT 5OCial." In: Hjelmslev, 1971 : 97-104.

Hjelmslev. L. (1954). " La Itralification du lan, • . " In: Hjelmslev. 1971 : 44-76. Hjelmslev, l. (1968). Dit SprOCM. Darmstadt (Danish: 196.3 ). Hjelmsiev. L. (l971J. EssuisU"'llisfiques. Paris. Kouc hi , T. (1977). Review of L. Hjelmslev, ProltlamtrIQ tll tltltr Sprorh,htor/t (Munich. 1974)

and L . Hjelmslev. ANjI4'le :.r SprachwiJltlUC"ha/' (Sluttgart , 1974), Histr"iO/rrophio U,,· ,uis,ko. 4(1). 91- 105.

Moms. C. W. (1971). Wril;"IS Oft /ht GlMrof Tlreor)" of SI,fIS. The Hlgue. Sauuure. F. de f 1916). C_" de Iitl/ruiSfiqllt M"'rolt . Parili. 1962.' TraNnl, J. (1970). Zur Stm/oI0l/t des li/uor/KIw" KII"s"'"t!rb~fOlStf1/O,Ur.1Id Liltrrmmheorie.

Munich. TtabaM. J. f 1916). Eltmt"'t der Stmiotilt. Munich. Trabant. J. ( 1917). "Stmiolik der Poesie in Ftlnkteich." In: A. Noyer· Weidner (Ed .), Aufl4lu : .. r

UltrO,urwilStflSrha/l , I. WieJbaden: 1- 32.