Contemporary Music Review 1997, Vol. 16, Part 3, pp. 5-25 Reprints available directly from the publisher Photocopying permitted by license only

�9 1997 OPA (Overseas Publishers Association) Amsterdam B.V. Published in The Netherlands

by Harwood Academic Publishers Printed in India

The implicit musical semiotics of Marcel Proust

Eero Tarasti Univers i ty of Hels inki

Because of his awareness that the surface of every narration is a tissue of signs, Proust is able to present a semiotic analysis of musical enunciation in his description (in La prisonni~re) of a performance of the fictional Septet by Vinteuil. He shows music as a fait social total, in which the enunciate is in dialogue with the enunciation; and thus he discusses different ways of listening, different aspects of the composer as man and as musical subject, different metaphors of the music (spatial, visual or aural), different philosophical reflections, as well as the strictly musical characteristics of the enunciate, aspects of repetition and cyclic form. Musical communication is presented as a multilevelled whole in which one moves freely from one level to another, without the control of a unilinear generative principle such as that presented by Ingmar Bengtsson. A semiotic analysis is provided of the passage from Proust's novel, showing the interaction of actors, modalities and isotopies.

KEY WORDS Enunciate, enunciation, musical communication, semiosphere, modality, poiesis.

Towards new areas of sign theory

H a r d l y a n y scholar of the d o m a i n w o u l d da re to a rgue if I s a id tha t Marce l

P rous t (1871-1922) w a s the g rea tes t s emio t i c i an of al l t ime so far. F r o m a

semiot ic v i e w p o i n t he has been a l r e a d y i n v e s t i g a t e d a.o. b y Gi l les De leuze

(1970) a n d Jean-Jacques Na t t i ez (1984). If w e p a y a t t en t ion to the sur face of

the P rous t i an na r ra t ion , i ts themat ic a n d f igura t ive level, w h a t a re i n v o l v e d

are o b v i o u s l y s igns, no t on ly of a ve rba l na tu re , i.e. w h o sa id w h a t to w h o m ,

b u t a lso o the r messages ; ru les of hab i t s a n d po l i t eness , food , l andscapes , pa in t ings , ges tures - - a n d of course l ikewise mus ica l s igna l s a n d symbol s .

Never the less , a se r ious m u s i c o l o g i s t m i g h t w r i n k l e h i s / h e r fo rehead in

scornful su rp r i se if s o m e o n e sa id tha t P rous t w a s a lso one of the grea tes t

mus i c ana lys t s there has ever been . Af te r all , even in the case of the f amous

6 Eero Tarasti

"petite phrase" of Vinteuil's sonata, he makes no other musical observation than that there are five notes of which two are repeated. H o w can one even mention such a musical amateur while pondering new methods of musical research?

However, I am quite convinced that Proust realized something quite essential about the nature of musical communication itself; under his guidance we can set about elaborating entirely new starting points for our research beyond the old semiotico-generative methods. Old? Of course I do not want to deny the validity of semiotic and generative methods just when they have ripened to a level on which they really function as a useful tool for music analysis (at least I hope so; see Raymond MoneUe's Linguistics and Semiotics in Music, 1992, and my own bookA Theory of Musical Semiotics, 1994). The "old' method will maintain its place in the description of certain kinds of musical texts at the same time that it also opens a little the gate outside the s tudy of the musical text or 'enunciate' , towards a theory of enunciation, i.e. poiesis and aesthesis.

Yet what we need now is a new theory, which is semiotic in a still more general sense than earlier methods of score analysis and which would melt together poiesis and aesthesis as well as observations of the neutral level - - all levels, indeed, on which generation, enunciation and enunciate may be situated.

In such a total musical semiotics, no reading or code as such will be more justified than any other, but each will emerge as music becomes a sonorous and spiritual reality, i.e. as it starts to exist beyond the enunciate, the score. All the aspects of this process can likewise constitute an object for artistic transformation - - and this is what happens in the music of our time. The poetic function of music focuses in no case exclusively on the structure of the message itself, on its self-reflecting form, as Roman Jakobson supposed, bu t it lies precisely in the fact that all functions, levels and phases of communication can be freely varied.

The remark by Gianni Vattimo that the relevant aspect in art is no longer the work itself, but the experience, the event, merely emphasizes the other pole of the matter (see Vattimo, 1985); it is a statement typical of the cognitive era, but it does not yet represent any universal prismatic model in which the most essential features are the very shifts of the viewpoints. Here, the most important points in the entire aesthetic communication are those places where the standpoint, the level the approach, the function, the category, the modality

any stable or constant aspect - - changes. For instance, almost the only device remaining for a composer is to leave

a trace of this in the notation or to give some kind of indication when h e / she wants to shift the code or isotopy. Furthermore, the artist cannot anticipate what the other subjects of communication are doing. For if communication in the arts is prismatic, kaleidoscopic, it is naturally also polysubjective.

The implicit musical semiotics 7

Such shifts of codes are, in fact, exemplified by certain ambiguous events coded in the score, in which some passage, verse, phrase, section can be listened to in two different ways (the 'duck/rabbit" phenomenon mentioned by Wittgenstein in his psychological remarks - - see Wittgenstein, 1980, p. 16e

and by the Gestalt theorists). For example, the main theme at the beginning of Chopin's G minor Ballade canbe imagined to start, after the introduction, as late as the moment when the G minor tonic is reached - - according to a spatial way of listening-- or as early as the point where the broken dominant seventh chord begins as an anacrusis to the tonic chord of the strong beat, which would represent an actorial way of listening (see Figure 1)

t ~ actorial segmentation r o)

t

Figure 1

Sometimes there is an inbuilt surprise in a piece, when suddenly the "right' isotopy is revealed to the listener. This may be illustrated by the beginning of the Piano Quartet in A major, op. 30, by Ernest Chausson (1897; see Figure 2a).

Animl~ ~ i z z .

J 3 ,~ p i z z . c I r c o

- I ' - I , - :'

A n i m ~

~-- . ' , . = ~ : r. r r -~ ~ ,.11 1 I I I J -J

.f.

~ . . . . ~ . . ; ~ i I I I

.~ , - , - . : " - ~ J ~ i J ~ J ' l t I I , I i " . . . . . =

Figure 2a

8 EeroTaras t i

In the first 26 bars we do not know in which style, which year, which counh'y we are, until the impressionist ninth chord in bar 27 (Figure 2b) shifts us at once to the correct topos of French late r o m a n t i c i s m and b u r g e o n i n g impressionism. One lone chord can serve as a sign or index of such a code shift - - or rather its discovery.

r l - - I ~ - - n i l

e n d i - m i - n u �9 a .

p

Y

o ~ .

0 ~ _ j O

re te nu

, , , , ; F I I J i / .,.i i'~ l I [ J r "iml 1, - m ~ i I i - ~ L ~ , --I ' I ~ - - I i "

~r

e n d i - m i - n u - a n t .

F i g u r e 2b

Let us assume that we are going to s tudy the semiotics of opera. First, we could naturally prepare a musico-kinetic bass-line sketch of the opera - - let us say b y Wagner or Verdi - - chosen by us, which can be described in the Schenkerian manner as a chord progression on an U r s a t z . O n e could then trace the bass-lines or Urs/itze of other levels of the opera, narration, drama, rhythmics, t imbres etc. In a word, we could por t ray the energetic course of the piece.

However, beside such an approach we are necessarily in need of another approach, of the kind that moves smoothly f rom one level to another, from one aspect to another, since receptionis a manifold, multi-dimensional process and it constantly changes its form in the most varied ways. If we proceed in a stiffly generative way and pass systematically through all the phases, our result does not correspond to the reality in which we live, and which we introduce into our reception process of an artwork.

These two approaches can be united; on one side there is a deep narrative basic stream, which runs now faster, now more slowly, now more broadly,

The implicit musical semiotics 9

now more quietly and, when narrative lacunae are formed, even throws itself to be taken by the stream. On the other hand, we meet on the surface levels the play of all the semiotic mechanisms; we are like Walter Benjamin's "absent- minded" receivers (Benjamin, 1974, pp. 471-508), or are like Theodor Adorno's types of listener, who vary according to the occasion (Adorno, 1962). In other words: sometimes the important thing is a motif, sometimes the whole sound field, sometimes a scenic gesture, sometimes a speech, sometimes a visual effect of the scenery or lighting - - in short, any aspect whatsoever of the enunciation or virtuality of the utterance.

Such a model would therefore combine enunciation and enunciate because of their constant interaction, competing for pertinence and attention in the artistic communication.

We are thus not helped by the mere statement that signs do not signify anything except as a part of their significant environment, or field, on the semiosphere (see Uspenki and others, 1973), the perfusion of signs (C.S. Peirce, quoted in Sebeok, 1977, p. ix), or the isotopy (AJ. Greimas, 1979, pp. 197- 198) - - but we also have to admit that what is significant and becomes significant, is entirely dependent on the interaction between the enunciation and the enunciate. No part of the enunciate is pertinent as such, by any cogency of its own, unless it is allotted a position in the enunciation. Accordingly, some element, totally unimportant in terms of the structure of the enunciate, can for some reason become an essential item in the process of enunciation, since it gives an impulse, serves as an index for a completely new semiotic development. In such an analysis everything is moving, flexible, changing, like reality itself. The logic of such an analysis recalls the m~moire involontaire of Henri Bergson or Marcel Proust.

Now we may understand why the generative course is after all impossible; not because it could be shown somehow logically inconsistent, but because there is no musical or poetic object to which it could be applied. We must accept the idea that musical reality - - even its smallest element, the sign is a totality, afait total social (Marcel Mauss, 1950), and not only a social fact but also psychological, physiological, physical, mathematical, semiological, acoustic, cognitive etc.; and thus we cannot at all isolate the musical object. It is paradoxical that there is an extensive study entitled Trait~ des Objets Musicaux (Schaeffer, 1966)!

The musical object does not exist, since music is always simultaneously, concomitantly and consecutively present in all the human subjects attending to its course, as well as apparently objectively present as a notation, recording or acoustic stream of sound. Taking it into account that these various spheres of reality each follow their own laws, it is impossible to imagine that anyone could create a generative trajectory covering the whole reality including all its pertinent traits.

But what, then, is the object of musical semiotics?

10 Eero Tarasti

I believe that all the aforementioned features of musical reality are semiotic by nature, while at the same time I believe that their inner laws are most probably different. It may be that some of the concepts offered by certain theories of semiotics are valid in many segments of reality at the same time, even to the extent that it is tempting to consider them universal.

Nevertheless the universe of semiotic phenomena takes shape so variously and incalculably on each occasion and in each situation of communication, that this whole cannot any longer be dominated by explicit rules. The only directive is that these parallel semiotic realities are actualized by impulses stemming from various sources, in such a way that a definite meaning of music can never be fixed. Meaning J signification-- is merely based upon this alternation of different semiospheres and their incessant changes. Musical signification is not an object to be set in front of us whose structure and layers one can see by cutting through it with a knife. The fiction of universality of meaning occurs persistently in some sciences and limits the validity of their results. The belief that one can look inside the human brain with surgical tools and in this way spread out what "really" happens in our mind, forms the basis for a view of the human mind as a huge computational network.

However, letus return to the problem of the relationship between enunciation and enunciate, as far as it appears in the correlation of a musical performance and the music to be performed. If we listen to a piece like Brahms's Clarinet Sonata, we are operating on the level of the musical enunciate when we follow, say, the unfolding of its main theme, modulations and chord changes. But as soon as the clarinettist plays a little bit louder or the pianist makes a striking gesture during a pause in the music or someone starts to cough in the audience, our attention automatically shifts, because of this impulse - - sign - - to the enunciation itself, to those events which bring about the enunciate or utterance and upon which its existence is, in fact, all the time dependent. If the clarinettist wants by this sign to express some pathetic state or emotion, if the pianist wants to project the gestural, kinetic aspect of the music, if the coughing in the audience means a weakening of attention, if the noise of somebody entering is to focus attention on the social milieu of the audience-- then these impulses correspondingly catalyze and open various semiotic paradigms or realities, emotional- aesthetic, gestural, physiological, social. All these spheres belong to what Greimas calls the "natural world" (le monde naturel), (Greimas, 1979, p. 233) which is itself already semiotized. But is there any logical consistency in the actualiTation of these worlds? Is it in the end impossible to portray semioticaUy the alternation of enunciation and enunciate, apparently such a limited problem?

To my mind at least, Marcel Proust has succeeded in doing this. Next I shall scrutinize his model of "semiotic" music description, which I shall try to make as clear and explicit as possible.

The implicit musical semiotics 11

Music at the salon o f M a d a m e Verdurin

Music plays an important role in Proust 's long novel A La Recherche du Temps Perdu. Nevertheless, it is not an accident that music attains its most central role in relation to the other arts only at the end of the series, in the scene at the bourgeois salon of Madame Verdurin in the volume La Prisonni~re. Earlier in the series, Proust had embodied literature in the form of the author Bergotte, and painting in the output of Elstir. They are naturally both fictional figures, as is the composer Vinteuil. However, it is music which becomes the central art form as part of a "semiotics" of the passions. The scene I mean takes place in the aforementioned novel on pages 62-85 (A La Recherche du Temps Perdu, vol. VI, 14th edition, Paris: GaUimard, 1923). It consists of 148 sentences (as the criterion of a sentence in this segmentation I have chosen a full stop m which does not tell us much, since in Proust one sentence can be a single line or a whole page).

The content of a sentence can be summarized according to which level it occupies in our model of musical communication: that of poiesis, aesthesis, neutral l eve l aesthetic ideas, performance situation, notation, reception, physiological states, ethical observations, functions of communication, the levels of the piece i t se l f - - symbolic, semantic, metric, syntactic, phonetic etc.

The description is realized through the narrator "I" of Marcel Proust, who can be compared to an anthropological field worker, an explorer, who in his diary notes everything he experiences in the smallest detail, and allots each of his observations to many different semioses, not limiting himself to mere aesthetic or musicological observations, i.e. to the levels of semantics and syntax. Since Proust therefore moves from one sphere of reality to another, he is expressly a semiotician and anthropologist.

In this first quick analysis I do not adhere to any semiotic metalanguage, but I leave the theoretical interpretation in the true sense still open. I thus anticipate that one can draw from Proust 's description an entirely new, so far unknown, more exhaustive model for music, and that it must not be bound to any ready-made, pre-existent semiotic theory.

Sentences 1 4 (or Section 1)

The musicians sit on the stage. The listeners prepare themselves for the performance, or the whole situation is depicted; the expectations of the audience are portrayed:

Morel ~tait d~ja monte6 sur l'estrade, les artistes se groupaient, que l'on entendait encore des conversations, voire des fires, des qu'il parait qu'il faut ~tre initi~ pour comprendre'.

12 Eero Tarasti

Aussit6t M. de Chaflus, redressant sa taiUe en arri~re, comme entr6 dans un autre corps que celui que j'avais vu, tout A l'heure, arriver en trainaillant chez Mme Verdurin, prit une expression de proph~te et regarda l'assembl6e avec un s6rieux qui signifiait que ce n'4tait pas le moment de rite, et dont on vit rougir brusquement le visage de plus d 'une invit6e prise en faute, comme une 61~ve par son professeur en pleine classe. Pour moi l'attitude, si noble d'ailleurs, de M.de Charlus avait quelque chose de comique; car tant6t il foudroyait ses invit6s de regards enflamm4s, tant6t, afin de leur indiquer comme un vade mecum le r61igieux silence qu'il convenait d'observer, le d6tachement de toute pr4occupafion mondaine, il pr6sentait lui-m~me, 6levant vers son beau front ses mains gant4es de blanc, un mod61e (auquel on devait se conformer) de gravit6, presque d6ja d'extase, sans r6pondre aux saluts de retardataires assez ind6cents pour ne pas comprendre que l 'heure 6tait maintenant au Grand Art. Tous furent hypnoKs4s; on n'osa plus prof6rer un son, bouger une chaise; le respect pour la musique - - de par le prestige de Palam6de - - avait 6t4 subitement inculqu6

une foule aussi mal 61ev6e qu'614gante.

(p. 62)

[Morel had already mounted the platform, the musicians were assembling, and one could still hear conversations, not to say laughter, speeches such as "it appears, one has to be initiated to understand it." Immediately M. de Chaflus, drawing himself erect, as though he had entered a different body from that which I had seen, not an hour ago, crawling towards Mine Verdurin's door, assumed a prophetic expression and regarded the assembly with an earnestness which indicated that this was not the moment for laughter, whereupon one saw a rapid blush tinge the cheeks of more than one lady thus publicly rebuked, like a school gift scolded by her teacher in front of the whole class. To my mind, M. de Charlus's attitude, noble as it was, was somehow slightly comic; for at one moment he pulverised his guests with a flaming stare, at another, in order to indicate to them as with a vade mecum the religious silence that ought to be observed, the detachment from every worldly consideration, he furnished in himself, as he raised to his fine brow his white-gloved hands, a model (to which they must conform) of gravity, already almost of ecstasy, without acknowledging the greetings of late-comers so indelicate as not to understand that it was now the time for high Art. They were all hypnotised; no one dared utter a sound, move a chair; respect for music - - by virtue of Palam6de's prestige - - had been instantaneously inculcated in a crowd as ill-bred as it was exclusive.]

( t r a n s l a t e d b y C.K. Scot t Monc r i e f f )

Sen tences 5 -6 (Sec t ion 2)

The p e r f o r m a n c e is m o s t l y d e s c r i b e d t h r o u g h the c o n s c i o u s n e s s of the na r r a to r

" I ' . H i s fa l se e x p e c t a t i o n s a n d e r r o r s r e g a r d i n g t h e p i e c e a r e p r e s e n t e d ; th is

is t h e f i rs t b u t n o t t h e l as t t i m e w h e n the i d e n t i t y of t he m u s i c a l w o r k is

d e s t a b i l i z e d - - n o t e v e n t h e m a i n p r o t a g o n i s t k n o w s w h a t h e w i l l hear , a n d

n o t o n l y d o e s h e n o t k n o w b u t h e t akes t h e p i e c e to b e a n e n t i r e l y d i f f e ren t

o n e ( sec t ion 2).

The implicit musical semiotics 13

Sect ion 7 (Sect ion 3)

One returns to the por t raya l of the reception, and to the posi t ion and at t i tude of one listener, M a d a m e Verdurin; this narrat ive m o v e m e n t be tween the ' T ' and other actors, the actorial "d i sengagement" as one w o u l d say reflects the b lending of the I wi th the rest of the audience, since he belongs, in the first place, to the same musical reference group.

Mme Verdurin s'assit a part, les hemispheres de son front blanc et l~g~rement ros~, magnifiquement bomb,s, les cheveux ~cart~s, moiti~ en imitation d'un portrait du XVIIIe si~cle, moiti~ par besoin de fraicheur d'une fi~vreuse qu'une pudeur emp~che de dire son ~tat, isol~e, divinit~ qui pr~sidait aux solennit~s musicales, d~esse du wagnerisme et de la migraine, sorte de Norne presque tragique, ~voqu~e par le g~nie au milieu de ces ennuyeux, devant qui elle allait d~daigner plus encore que de coutume d'exprimer des impressions en entendant un musique qu'elle connaissait mieux qu'eux.

(p. 63).

[Mme Verdurin sat in a place apart, the twin hemispheres of her pale, slightly roseate brow magnificently curved, her hair drawn back, partly in imitation of an eighteenth century portrait, partly from the desire for coolness of a fever-stricken patient whom modesty forbids to reveal her condition, aloof, a deity presiding over musical rites, patron saint of Wagnerism and sick-headaches, a sort of almost tragic Norn, evoked by the spell of genius in the midst of all these bores, in whose presence she would more than ordinarily scorn to express her feelings upon hearing a piece which she knew better than they.]

(translated b y C.K. Scott Moncrieff)

Sentences 8-23 (Sec t ion 4)

The first l istening per iod - - ' pe r iod of at tent ion ' - - of the narra tor is mos t ly situated on the epistemic axis of knowing. He is at first as though in an u n k n o w n realm - - a spatial m e t a p h o r m in which he finds a familiar pa th th rough a theme b o r r o w e d f rom an earlier piece.

The reference to the sonata of Vinteuil means that one steps outs ide the musical object. The small phrase is used as a k ind of interpretant , by which the text opens to the listener; it forms a rupture, a cleavage, th rough which one penetrates into the object. It is crucial that a l though the listener feels himself as t h o u g h in an u n k n o w n country, this small phrase of the Sonata helps h i m to orientate himself there, i.e. to achieve spatial d isengagement . Since the small phrase is, f rom the poin t of v iew of its aesthetic informat ion value - - its "knowing" or savoir ~ o u t w o r n for the listener, it assumes, on

14 Eero Tarasti

the other hand, a new meaning in this so far unfamiliar context. The small phrase functions therefore like the leitmotifs of Richard Wagner; they adopt new meanings in various contexts, on different levels of meaning inside the work.

The unfamiliar music of the Septet is at first described with colours "red" w and light phenomena - - "dawn". Thus new interpretants are sought, taken now from a reality external to the music.

Then (sentence 20) the first syntactic observation is made on the musical text: it is called 'a song of seven notes'. Thereafter it is described as a kind of "call" or as the crow of a cock. But the motif is at the same time changing its meaning - - meanings constantly change in the Proustian text, since he describes them precisely as they take shape in time and in a certain concrete situation.

In passing, the narrator also presents a remark or hypothesis about the poiesis of the motif, since he supposes that its origin is the sound of the bell at Combray.

Thereafter the nar ra to r - - the 'T" or ego - - returns to convey his own reception of the music, with clearcut aesthetic evaluations; he uses the thymic dysphoric category when he says that the motif does not please him, and seems almost ugly. This sudden judgement of taste gives him an impulse to shift from the inner musical world, from the enunciate or utterance, to the world of the enunciation, to the performance situation of the musical piece. This is characteristic of the Proustian narrator; as soon as something reaches a certain limited meaning, it ceases to be interesting and causes a reaction of boredom, a shift to another level of reality. Everything which is fixed, determined and unambiguous is already dead for him and has lost its true purport .

Sentences 24-29 (Section 5)

His attention is caught againby Madame Verdurin's reaction and particularly the non-verbal behaviour caused by the music. This is exactly that musical behaviour described by the anthropologist Alan P. Merriam.

In fact the whole body of Madame Verdurin is "talking", "telling", and it tells even louder than words what she does not expressly say. In other words, it reveals her as a competent listener, who knows what is involved; as a contrast there is the incompetent herd of other listeners, still at this stage anonymous.

Sentences 30-32 (Section 6)

Now one moves to the level of the physical activities of the musicians and the performing act itself is depicted, i.e. the concretely physical aspect of the enunciation; it might, of course, have been investigated by exact methods

The implicit musical semiotics 15

of measurement in the manner of Paul Bouissac (1973), but Proust approaches even this area by means of metaphors and interpretants. Through the reference to a curl fallen over the forehead of the violinist Morel, one moves again to a new level, a description of the reception.

Sentences 33-34 (Section 7)

The narrator wonders what effect this curl might have on Baron de Charlus, but he is only able to see Madame Verdurin. Thus his thirst for knowledge is not satisfied.

Sentences 35-37 (Section 8)

An extensive musico-philosophical reflection follows now, in which the narrator presents his profound philosophical remarks; however, they proceed all the time in parallel with the musical piece.

First (Sentence 36), the narrator realizes that what is involved is a masterpiece, in relation to which all the previous works by Vinteuil were only drafts.

Next (Sentences 37-47), the narrator compares V'mteuil's pieces to his earlier loves, among which only the last one for Albertine is equal to the Septet.

The narrator "I" presents his "actorial theory" : of all the persons we know, we carry in our minds a kind of sign, a double. He sinks from the musical world to reflections on the 'inner actorial sign' of Albertine. Whenhe imagines how this "pseudoactor' Albertine falls asleep, a signal from the music also penetrates into his consciousness. This sign serves as a kind of bridge and shifts him again to ponder on the creation m the poiesis - - of the piece, on V'mteuil, especially with regard to his sleeping daughter.

This image is strengthened by a musico-historical association-- a reference to Robert Schumann.

Sentences 48-59 (Section 9)

The narrator expels Albertine's "double" sign or pseudoactor from his mind and sets about thinking of the composer of the piece, Vinteuil, now not as a real person n as he was above n but precisely as a subject of enunciation (composing).

Of Vinteuil's style he states that it is inimitable and that it cannot be captured within any generative rules. The narrator ego therefore does not believe, on this level, in the 'identity', the universal categories, from which an art work

16 Eero Tarasti

emanates, but instead takes as Vinteuil's distinctive feature a kind of lasting novelty (durable nouveaut~).

The narrator states how Vinteuil as a real man was completely different from what he was as a composer - - or perhaps this is Proust, stressing the difference between the real person and the artist's identity. As a real man Vinteuil was timid and sad, whereas as a composer he was joyful. The "thymic" state of this joy is transmitted from the work to the listener. Thus Proust believes that the purely artistic emotional state, that of the subject of the enunciation, can be transmitted to the listener via musical signs.

In this way Proust confronts art and life, and suggests that the concrete performance of Vinteuil's piece by means of instruments that are animated by his music seemed to prolong his own life - - even though the extension was unreal.

Sentences 60--84 (Sect ion 10)

The narrator moves now from Vinteuil's composer ego to examine the Septet itself. First, the Septet is compared to the Sonata, and the two semiotic objects are scrutinized side by side.

Although the works are apparently quite different, they have the same semantic gesture or Ursatz, a prayer of hope. The narrator admits that a strict musicological or "musicographic" description could have explained the genealogy of these themes or the way in which they had been derived from the same nuclear motif, but such reasoning, even if congenial because of its scientific flavour, is not semiotics, since it takes place only on the level of the signifiers, and does not account for the signified, the content.

Consequently, according to Proust, when great variety and diversity prevail on the manifest level of Vinteuil's work - - on the level of sound or the signifier, the level of content reflects great unity.

Vinteuil's output Signifier = variety Signified = unity

The hidden unity of Vinteuil's output is also reflected on the acoustic sound level as his typical "accent".

At the same time it is stated that this accent is a sign of the individuality of the soul and that it is characteristic not only of the composer but also of the performer - - the example is given of singers.

The accent in question, which distinguishes Vinteuil from all others but also guarantees that his identity is maintained always the same, represents an enigma which is then pondered by the narrator.

He asks how it can be learned or heard. He remarks that each artist has it in the background, like his own 'semiosphere', and he depicts it as his lost

The implicit musical semiotics 17

fatherland. In Greimasian terms: the composer n and similarly the performer - - is like a subject who has been separated from an object and who tries all the time to reach this original lost object, with varying success.

When he notices that Vinteuil, in his Septet, has come closest to this lost, separated object, the narrator refers to the enunciation itself, in which even the violinist, Morel, had got near to it. His performance is characterized by the qualities of moral strength and intellectual supremacy. These qualifies are - - as in the comparison of Vinteuil's real personality and his artistic ego - - completely opposed to his true character, which is sketched elsewhere in the novel.

Nevertheless, the narrator notices that his theory of the "lost fatherland' could not be justified by reason.

Next the lost fatherland is likened to an unknown star. Listening to various pieces by various composers is like making real journeys from one star or planet to another. Music becomes the only means or sign process - - besides painting - - through which we can look into the individuality of a human soul.

Sentences 85-100 (Section 11)

An interval between two movements follows now in the performance. The audience is clearly split into two groups, unlike at the beginning; the narrator has had a deep artistic experience, whereas the other people had not experienced anything corresponding and exchange meaningless news with each other.

The social situation appears now entirely devoid of signification on the side of the inner experience of the n a r r a t o r - - the enunciate has now become superior to the enunciation.

Sentences 101-112 (Section 12 )

The end of the Septet is now described, and once again one moves to the analysis of the musical enunciate itself. It is most closely examined in the light of various themes, phrases, the recurrence of the inner actors of music, i.e. on its actorial level.

The music is now characterized by a violet colour. Likewise the musical form is portrayed as a fight between these actors. The syntactic structure of the music is viewed as the approaching and distancing of various motifs, as movement in general - - one motif is repeated five or six times - - as the return of the mystic call of the beginning. In other words, what is involved is cyclic form, the characteristic device of C~sar Franck and his pupils.

18 Eero Tarasti

In the end, the rejoicing motif wins; it has a purple colour. Joy, the signification of the motif, makes the narrator-ego shift within his own real ego, and ask whether he can ever participate in such a supernatural joy. In other words, in the Proustian world one is constantly gliding, imperceptibly, from one level to another, from one semiosphere to another.

Sentences 112-135 (Section 13)

One moves now to the level of poiesis; the narration tells of the reconstruction work of Vinteuil's daughter and her friend, the writing and collecting of th e whole score on the basis of sketches, Accordingly, one now returns to the level of the material production of the piece, while at the same time one ponders why these authors set about doing this laborious work.

Sentences 136-141 (Section 14)

Observations on the posthumous reputation and reception of Vinteuil, compared to other masters. It is stated that without the Septet he would not have become the immortal genius that the narrator and society deem him. The force behind the cornerstone of his reputation, again, is a combination of ingenuity and vice.

Sentences 142-148 (Section 15)

The same combination also brought about the performance of the Septet; its invisible but real cause was exclusively the relationship of Baron de Charlus to Morel, the wish of the Baron to arrange for the musician an opportunity to perform for high nobility in order to be granted a particularly distinguished decoration.

Therefore, it is only at the very end that Proust describes what was first in chronological order: the reasons which led to the whole situation. By this means, Proust wants, on one hand, to show that these reasons are not immediately observable by us, but only to be inferred on the level of thirdness. Moreover, he shows that even the most profound and subtle experiences are always dependent on the material and inferior sphere of ordinary reality.

The Proustian "musical model" explicated in the Greimasian manner

Before I start to summarize Proust's description, it is proper to make some general remarks on his "musical model".

19

C~

I ~176

[ ~ ~-

t~

r~

20 Eero Tarasti

1. Musical communication for him is a multileveUed whole in which one constantly moves from one level to another.

It is impossible to say of any level that it is more "justified" or important than any other. Particular attention has to be paid, in further analysis, to how these shifts take place or what makes the narrator-ego, the most central musical listener, move his attention from the enunciate to the enunciation etc.

The Proustian musical listener is a modern, absent-minded listener, and his attention does not follow logically any determined line or semiotic process - - for example he does not discern the aesthetic associations of the piece until the end, nor perceive its syntactic course as a fight among various themes, but glides from one point of view to another. 2. Musical communication is by nature a continuous process. This becomes obvious on the level of the text itself; phrases follow each other without any divisions into paragraphs or groupings, they pass by like a continuous stream of words.

In the Proustian description all the human agents of the communication are present in the same way as all other musical ingredients - - f rom a composer 's living conditions to his professional work, the reconstruction of the score or notation, the physical conduct of the players, the portrayal of the music itself as a timbre, the elucidation of the compositional structure, its comparison to other musical pieces and composers, and finally the receivers and their verbal and non-verbal behaviour. 3. In Proust musical communication does not, however, proceed in a linear way from sender to receiver. If one thinks of the chain of musical communication such as it is presented by a . o . the Swedish musicologist Ingmar Bengtsson (1973) - - in terms of the general scheme of information theory, of which we of course have numerous variants (Figure 3) m then the difference of the Proustian view is striking. In Proust the communication goes as follows:

Performers - listeners - narrator - listener - narrator (composition) - listener - performers - listeners- narrator (composition; subsidiary character) -composer - composition - composer - performer - listener - composer - composi t ion/ authors - composer /authors - performer/l istener.

Naturally, this scheme enormously simplifies the whole Proustian description, but it reveals to us one important thing: in his model one is constantly passing crosswise among different entities of communication. The boxes in the original Bengtssonian model m displaying the direction - - prove to be completely illusory. Musical communication does not have any direction in the proper sense, since the normal course canbe interrupted at any time, from any impulse, causing either a flashback to some previous or subsequent place in the communication chain or moving us suddenly totally outside it.

The implicit musical semiotics 21

In this model the musical subjects and objects are therefore not bound by any unilinear generative course but move freely within the limits of these roles.

From this we can only conclude, as I stated at the beginning, that if we want to pursue a complete and exhaustive description of any musical reality, we have to take into account this Proustian multifariousness and mobility. 4 . There is one more crucial difference between Proustian musical communication and the 'normal' transmission of messages. In Proust's vision the maker of the model, the analyst, has been included in the model as one of its essential parts, through the narrator-ego. In the Bengtssonian model the researcher has faded into the background as its anonymous sender and creator. Every semiotic inquiry has in fact to start, to my mind, from the same situation as any narration: who is the subject of this semiotic analysis, examination, narration? If the model is missing its author, it is certainly deficient.

Making the author, the scientific narrator-ego, explicit in the research itself, is not at all a matter of unscientific subjectivity, but on the contrary it accomplishes the Hjelmslevian criterion of the explicit nature of scientific discourse better than any apparently objective, positivist description. 5. If one wants to apply the Proustian model to any new object, one should therefore take into account the entire situation of the musical enunciation, i.e. the performance, those who perform the piece, in which place, when, who the listeners are, what they experience, how the different degrees of musical competence of the listeners influence the decoding of the musical message, what the structure of the piece expresses, how the piece is related to the other works of the composer, to which musical genre it belongs, and what is the general role of its traditional genre, how it has been made, which kind of melodies it contains, which kind of dynamics, chordal processes, rhythms, what is its temporal nature in relation to the inner time conceptions o f the listeners.

But would such a description be a satisfactory semiotic analysis of the performance situation? What would make it particularly semiotic?

Such an analysis would be changed from a mere taxonomy into semiotics undoubtedly by the presence of the meaning and signification of the whole process, the facts that music as a sign exists in aninfinite number of simultaneous modes of being and that it can be compared to the "double" of entities, as described by Proust, or in fact, to the sign of the reality which we have in our minds; on the other hand, "semiotic" is an apt term because the mode of existence of these signs in our minds conveys their individual "accent", as Proust says. This means that we all conceive the communication in slightly different ways, but that in spite of these differences there is also something stable in these modes of existence. We have tried to conceptualiTe these constant,

22 Eero Tarasti

universal laws, us ing the terms of our semiotic meta language , b y wh ich we por t r ay this aspect of the reali ty which is significant to us. 6. As a last remark about Prous t ' s description, w e m u s t r emember that he after all gives us an analysis of a nonexistent compos i t ion b y a nonexistent compose r (I am grateful to m y col league Erkki Sa lmenhaara for r emind ing me of this impor tan t aspect). Therefore, in this view, music is essentially a pure p roduc t of ou r mind and no th ing else.

Finally I summarize the Proustian descr ipt ionby focusing on its mos t essential contents. Hereby I a m us ing formaliTations s t emming f rom Gre imas ' s theory ("universalities").

The number s refer to sections in general; I give first the pos i t ion in the chain of communica t ion in wh ich the descript ion takes place.

A s e m i o t i c a n a l y s i s o f M a r c e l P r o u s t ' s d e s c r i p t i o n o f a s o i r e e a t

M a d a m e Verdurin's (La Prisonni~re, pp. 62-85, Gallimard, 1923)

1. Pd = the physical activities of the performers (d = to do) m(e) = "expectation", "preparation" as modalities Lm(e) = the introduction of the listeners (L) and their expectations

2. $1 = subject (1) or the narrator re(e, not-to-know) = the modalities of this subject are "expectation" and "ignorance"

3. LS1 = the "listener subject 1" or Madame Verdurin The basic isotopy in these first three sections is '1society", since their different subjects are seen as members of this collective actor, so far without any indivfdual characteristics, as possible destinatees of music experience.

4. Slph = the musical, phenomenal experience of the subject $1 is described re(not-to-know) NOsp - - music as a spatial object whose modality is "unknown" re(know) Mal = one familiar actor appears in music ("la petite phrase") Mi(ex) = music is provided with an external interpretant ("colour") Ma(phemes) = a phemeanalysis of the musical actor or its syntactic properties is made m(dys)Na = a dysphoric evaluation of the musical actor trans(unex) = a transition which is particularly indexical; i.e. $1 gets bored which causes a "move" to the next place in the communication

5. LSld(non-verb) = the non-verbal conduct of the listener $1 (Madame Verdurin) is depicted (d = to do) m(know)LS = the modality of the listener $1 is "knowing" i.e. her musical competence

6, Pd = the physical behaviour of the performers is depicted 7. LS1, LS2, $1 = the listeners $1 and $2 (Mine Verdurin and the Baron de Charlus)

and the narrator 8. Slph = the inner world of the narrator

m(euph)M(can + know) = the music is acknowledged as a masterpiece ("+know") and as it were the end-point of various sketches ("can")

The implicit musical semiotics 23

PseudoA = the "double" of Albertine: a pseudoactor trans(index) - - PseudoA = pseudoactor is rejected or negated

9. C("real actor" v. "composer actor") = the composer is introduced inhis two opposed roles, as a real person and as a composer

10. M(O1 v o l ) = the music itself is examined - - Object 1 (the Septet) is compared to object I (the Sonata); these reflections are given without any "modalizations" of the narrator, Subject 1 dC (composer actor) (S V O; S-O) = it is said that the author inhis role as a composer actor is disjuncted from his object and that he is trying (d=clo) to reach i t dP(S O) = it is said that the performer subject (Morel) has inhis interpretation reached his object ("the lost fatherland") P(rea/actor v. performer actor) = the performer is presented inhis two roles as a real person and as a performer;just as in the preceding analysis of the author or composer, it is stated here that the performer actor is euphoric whereas the real actor is dysphoric

11. L[m(euplO $1 v. m(dyspl0] = the listeners have been divided into two groups: Subject S1 (the narrator) whose modalities are euphoric whereas others are modally (thymically) indifferent or dysphoric

12. M[A(phemes: a l v. a2)] = a phemeanalysis of the music; a fight between two actors or themes Slph[m(euph)al] = subject S1 experiences the theme or actor I as a motif of joy or something euphoric.

13. C[Onot.(A2,A3dO)] = the authors or composers, actors A2 and A3 (V'mteuil's daughter and her friend) are portrayed since they have "done" or prepared the Object, the score of the Septet

14. mC(AI) = the evaluation of the afterfame of the composer actor A1, Vinteuil m[C(real A2,A3)] = the dysphoricization of the real actors A2 and A3 P a reference to their vice

15. mP[(dysplOrealA], mL(dysph)realS2 = the dysphoricization of the real performer actor and the real listener actor $2

References

A d o m o , T h e o d o r W. (1972) Einleitung in die Musiksoziologie. F r a n k f u r t a m M a i n

Bengtsson, I n g m a r (1973) Musikevetenskap: en 6versikt. Stockholm: Scand inav ian

Un ive r s i t y Books

Ben jamin , Wa i t e r (1974) D a s K u n s t w e r k i m Z e i t a l t e r s e i n e r t e c h n i s c h e n

R e p r o d u z i e r b a r k e i t . In Gesammelte Schriften, 1/2, F r a n k f u r t a m M a i n :

S u h r k a m p , p p . 471-508

Bouissac, Pau l (1973) La mesure des gestes: P r o l e g o ~ d la sbniotique gestueUe. A p p r o a c h e s to Semiotics , 3. The H a g u e , Paris: M o u t o n

Deleuze , Gil les (1979) Proust et les signes: perspectives critiques. Paris: Presses

Univers i t a i res de France

24 Eero Tarasti

Greimas, A.J. & Court~s, Joseph (1977) S~miotique: dictionnaire raisonn~e de la tMorie du langage. Paris: Hachette

Mauss, Marcel (1950) Sociologie et anthropologie. Pr&~d~ d'une introduction l" oeuvre de Marcel Mauss par C. L~vi-Strauss. Paris: Presses Universitaires de France

MoneUe, Raymond (1992) Linguistics and semiotics in music. Contemporary Music Studies, 5. London: Harwood Academic Publishers

Nattiez, Jean-Jacques (1984) Proust musicien. Paris: Bourgois

Proust, Marcel (1923) La prisonni~re. A la recherche du temps perdu. Vol. VI. edition. Paris: Editions de la NouveUe Revue Fran~aise

Proust, Marcel (1934) Remembrance of things past. 2 vols. Translated by C.K. Scott Moncreiff. New York: Random House

Schaeffer, Pierre (1966) Trait~ des objects musicaux. Paris: Le Seuil

Sebeok, Thomas A.(ed.) (1977) A perfusion of signs. Bloomington: Indiana University Press

Tarasti, Eero (1979) Myth and music. A semiotic approach to the aesthetics of myth in music, especially that of Wagner, Sibelius and Stravinsky. Approaches to Semiotics, no. 51. New York, The Hague: Mouton

Tarasti, Eero (ed.) (1987) Semiotics of music. Special issue of Semiotica, 66(1-3)

Tarasti, Eero(ed.)(1987-1988) La musique comme langage I-II. Special issue of Degr~s, 52-53

Tarasti, Eero (1988a) On the modalities and narrativity in music. In Rantala, V., RoweU, L. & Tarasti, E., (eds.) Essays on the philosophy of music. Acta Philosophica Fennica, 43. Helsinki: Societas Philosophica Fennica, pp. 155- 168

Tarasti, Eero (1988b) Basic concepts of studies in musical signification: a report on a new international project in semiotics of music. In Sebeok, T. & Umiker- Sebeok, J. (eds.) The semiotic web. Berlin: Mouton de Gruyter, pp. 405-581

Tarasti, Eero (1989a) L' Analyse s~miotique d'un prelude de Debussy: La terrasse des audiences du clair de lune. La mise en ~vidence d'un parcours narratif. Analyse Musicale, 16, pp. 67-74

Tarasti, Eero (1991) Beethoven's Waldstein and the generative course. Indiana Theory Review (a special issue on Narrativity in Music, ed. R. Littlefield) pp. 99-140

The implicit musical semiotics 25

Tarasti, Eero (1992) A narrative grammar of Chopin's G Minor Ballade. In Minds and Machines, 2(4), CamiUeri, L. (ed.) a special issue on Music and Cognition, pp. 401426

Uspenski, B.A., Ivanov, I.I., Toporov, V.N. Piatigorsky, A.M., Lotman, J. (1973) On the semiotic study of cultures ( as applied to Slavic texts). In Jan van der Eng & Mojmir Grygar (eds.) Structures of texts and semiotics of culture. The Hague: Mouton

Vattimo, Gianni (1985) La fine della modernitY: nihilismo ed ermeneutica nella cultura post-moderna. Milan: Garzanti

Wittgenstein, Ludwig (1980) Remarks on the philosophy of psychology. Vol. 1. G.E.M. Anscombe & G.H. yon Wright (eds.) Chicago: Chicago University Press.