Marcel Henaff

The Mythologiques: Between and music^

When Claude l^vi-Strauss decided to use the metaphoric model and a certain vocabulary of music in his analysis of , he implicitly consi- dered the mythical narrative, like music, as a symbolic device. does not belong to the order of signs, which can be interpreted as having and conveying an encoded, signified meaning. Instead, it organizes the world for its practitioners; its Performance is each time an act of Organization of things and beings. Like music, myths are in a relation of transforma- tion: since myths - according to Levi-Strauss - belong to the order of symbolism, they can be 'translated' only within that order. Thus translat- ing means generating new symbolic elements, i.e. producing a transfor- mation. Myths are reinterpreted from one Variation to another, just like variations on themes in a piece of music. The task of the mythologist is thus comparable to that of a music performer, whose interpretation is itself part of the system to which the interpreted structure belongs - so mythological analysis is an extension of the network to which mythical narratives belong. The underlying and groundbreaking implication of this music-myth relation proposed by Levi-Strauss is of a semiodc nature: in that myths, as facts of language, cannot be translated but only transformed into vari- ations within the same system, the symbolic nature of language is chal- lenged and redefined as independent from and unique with respect to other symbolic systems.

CORRESPONDENCE: Marcel Henaff. University of California, San Diego, USA. EMAIL [email protected]

The greatest surpiise caused by the publication of the first volume of Levi- Strauss' Mythologiques'^ in 1964 was not first and foremost the apparently

Published in Japanese "SHINWA-RONRI: gengogaku to ongaku no aida de", transl. bv Katsunori Izumi, in LEVI-STRAUSS "SHINWA- RONRI" NO MORI [Ihrou^h the forest of />w-.S>ra«.fj'Mythologiques]; Ed. Watanabe Ko7,o & Kimura Hideo, Misuzu Shobo Publisher (2006), Tokyo; pp. 30-57. Translator's note: The series Mythologiques '•Kis translated by John and Doreen Weightman and published by Harper and Row, New York under the heading Introduclion to a Säence of Mytholo- gy. It includes 'l'he EMW and the Cooked (transl. of \je Cm et k Cuit 1969); From Hon^ to Ashes

Cognltive Semiotics, Issue 3 (Fall 2008), pp. 20-35 THEMTTHOLOGIQUES | 21 overwhelming nature of the project, which aimed at nothing less than accounting for most myths of the two Americas. Nor was it the use of the linguistic model already proposed in the foundational article of 1949 "The Structural Study of Myth".^ It was the use of the model and the vocabulary of music. This change of theorerical apparatus and even of methodology, from a scientific discipüne (as linguistics understood itself to be) to an art (as music is) has often been misunderstood, as Levi-Strauss himself noticed/ Non- specialized criticism saw in that change a commendable effort to present a difficult matter in an accessible way and to render appealing an enterprise otherwise very austere. In the discipline, the supporters of classical empiricism rather saw in it one of those new theoretic fancies, of which French structural- ism was suspected to be fond of. Fortunately, competent readings were not lacking either. For that, it would be enough to read attentively Levi-Strauss himself, who in his "Ouverture" explained himself in a precise and rigorous way about this myth-music relation. There was nothing decorative about it; on the contrary, what was at stake there concemed both the theorerical status of the mythical narrative and the method of analysis. However, in this shift something seems to have escaped Levi-Strauss himself. Something that he never formulates explicitly, nor even seems to consider, and which yet constitutes the foundation of his new approach. What then? The foUowing: the proposed idenrification between myth and music considered the mythical narrative as a symbolic device, i.e. as an operating device which aimed at producing an effect, and not as a discourse aiming at formulating an utterance. It is this implicit choice that is confirmed by Levi-Strauss' own hypothesis, according to which myths are in a reciprocal relation of transformation (in the specific sense that he attributes to this concept) and that, just like variations in music, they interpret each other — or translate each other — by producing new versions. Such a conception implies a greater consequence: it compels one to ask oneself

(transl. of Du Miel aux Cendns 1973); 'l'he Orifftt of Table Mannen (transl. of L'Origine des Manieres de Table 1978); and The Naked Man (transl. of l.Homme Nu 1981). 3 Originally published in English, "The Stmctural Study of Myth", Journal of American , vol. 68, No. 270 (1955); Reprinted in Anthropologe structurak (1958: ch. XI). 4 "The relationship between myth and music on which I insisted so much in the initial section of The VJXW and the Cooked and also in the final section of T'Homme Nu - there is not yet an English tide because it is not translated - was probably the topic which gave rise to most misunderstandings, especially in the English speaking world, though also in , because it was thought that this relationship was quite arbitrary". Levi-Strauss (1978). Mjth and Meaning, ch. 5 ("Myth and Music"), p. 44. (Translator's note: The English tianslation of L'Homme Nu was first published only in 1981, i.e. three years after Myth and Meaning, which explains Levi-Strauss' remark in this passage.j 22 M. HDIAFF

about the enunciadon position of the mythologist. Can he pretcnd to be an interpreter, able to fonnulate a signified {signifie) about the myths he studies? In other words: is it possible for him to claim a theorerical position, according to which he could explain what the myths mean? But to suppose this would be not to take seriously the fact that rr^ths do not aim at Sewing something but at producing it-, they work at setting up a world. What is, then, the task of the mythologist? This is Levi-Strauss' answer: he intervenes not as a commentator but as an performer of a piece of music. Just as the latter interprets a piece by playing it, so does the mythologist relate to the myths by narrating them or by developing them. But is this actually his only task? This shall be the ultimate issue of this discussion; because it is the actual status of critical knowledge that is at stake in this shift from a linguistic model to a musical model.

The methodological revival: the linguistic model The revolution proposed by Le\H[-Strauss in his analysis of myths from 1955 onwards, seems at first similar to the one he performed in the domain of kinship. Here also, he invokes the linguistic model, first of all. By rejecting the traditional types of Interpretation, which aimed at explaining a mythical narrative by focusing on the content alone (figures, actions, themes), Levi-Strauss performed a decisive revolution. He dismissed the p^chological explanations, according to which myths were expressions of our fundamental feelings and of our inner conflicts, as well as the purely sociological explanations, which assumed that myths only reflect the conditions and contradictions of a social group (even if this can make up one of the dimensions of myths), or even the symbolist interpretations, which claimed that myths express archetypes of human nature. He refused just as much the rationalist reductions, which turn myths into imaginal and somewhat naive transpositions of natural phenomena. Finally, he strongly questioned the Junctionalist theories, according to which myths would first and foremost translate the material needs of individuals and groups. Certain aspects of these different approaches can and even should be taken into account, but on the condidon of being considered in a completely different manner. For this, Levi-Strauss proposed to depart from this ver)' simple obvious fact: rr^ths an narratives, so they are Jacts of language. As narratives, they enunciate a succession of events. It is important first of all to understand what makes up the logic of their narration. Now, the latter seems obscure. What indeed strikes every reader of myths is the apparendy absurd or poindess THEMYTHOLOGIQUES | 23 nature of many episodes or details. The attempt to reduce this absurdity or this poindessness through interpretarions that apply to the characters or the images, in short, dedphering the contents on their own, would be to refer them to a meaning which would precede or go beyond narrative. Therefore, there would supposedly be a secret meaning only decipherable by subde Interpreters. For centuries there were enough candidates for this "hermeneutic illumination"; hence the different schools of Interpretation mentioned above. However, if instead of projecting external interpretative schemas we ask in the first place exacdy how the details and episodes of these narratives are related to the Clements of context, which are then linked to segments of other mythical narrations, then the narrative wholes tum out to be quite different: they organize the references to a given environment, but they do so within the more complex device of their links to other narratives. Sometimes they do so only on certain levels within those other narratives. These latter, in this network, may contradict or complete each other, oppose each other or repeat themselves, reply to, or invert each other. The contents are thus well summoned, but rather as semantic cores irradiating Clusters of relations in different groups of myths. Two methodological concepts dominate this approach. The first is that of a ^stem (yet it would be recommendable to add to it that of a networ/^). At this level, the discipline that serves as a model is in fact linguistics, more precisely . The second major concept is that of tranrformation-, it stems both from mathematics and from the natural sciences. But, as we shall see in short, it is from music - thus from an art, from a practice - more than from a scholarly discipline, that Levi-Strauss demands a fiinctioning model. This choice is surprising and requires further explanation. Yet let US retum to the linguistic model and point out what is essential about it. Levi-Strauss formulates the hypothesis that myths of the same cultural area form a system (as we say about language). Interpretation is to be found on this level: in the reciprocal relation of the narratives and their variations. Precisely while he reminded us of the three fundamental rules to observe in the act of analyzing myths, he defined an approach that was rigorously opposed to all ancient traditions of the gloss. These rules may be summarized as follows: 1) do not reduce a mythical narrative to just one of its levels; 2) do not isolate a myth from the group of other myths with which it is in a relation of transformation; 3) do not isolate a group of myths from the other groups of myths to which it is connected. In short, to understand a narrative within a given tradition is to 24 I M.HENAF

seize, first of all, the entire set of other narratives that are linked to it; it is to situate it in a network. Just like in the case of kinship systenis, for l^vi-Strauss what was at stake was not to ask structural linguistics for directly transcribable schemas, but rather to define an orientation of method. Such a request seemed more well- founded because there was a very precise resemblance between the difficulties of traditional linguistics in the West and those of the Interpretation of myths. Traditional linguistics (understood as that of the philosophical reflection since Antiquity) searched for a relarion between sound and sense. A research fated to failure, since the same sounds can be found in different languages, associated to very different meanings. The methodological revolution was accomplished when the problem was put otherwise, i.e. when one understood that there could be no meaning in the sounds taken in isolation, but only in the rules of their combinations. The same goes for myths, explains Levi-Strauss: the themes and figures have no meaning in themselves (therefore, it is poindess to search for an archetypal or universal signification for them). Once again one needs to search for meaning in the rules of combination or composirion of the Clements. NaturaUy, this does not mean a pure and simple transposition of the method of linguistics to mythology. It is rather a model that allows one to situate terms in a system of relarions. Just like in the domain of kinship, the point was to think of the system as such, i.e. the priority (which does not mean the precedence) of the reladons over the terms. So in the case of myths, narrative ought to be considered as formed by consotuent units, whose relevance is determined by the system. But how are these units to be defined? What is the system in question? Mythical narrative, as a fact of language, is subject as such to common linguistic analysis and on this level it reveals no more than any other linguistic utterance. However, precisely the sentence, which constitutes for linguistics the largest unit over which it can legislate, is on the contrary for discourse analysis, being the smallest unit from which it can devise combinations. The superior level of the one becomes the inferior level of the other. In short, is it not the sentence that ought to be treated by the mythologist as the by the linguist? That is to say, as an oppositional, relative and negative unit? But the sentence is a discursive unit. Thus the sentences "The bird sings" or "Socrates was a wise man" possess a meaning on their own: each of them is already a discourse. Now, by proposing to call nytheme the minimal narrative unit that defmes the level to which other levels can be reduced, Levi-Strauss intends to THEMYTNOLOGIQUES | 25 confer to it the status of a phoneme. So, the is not identical to the sentence any more than to the ward. It is not a linguistic entity, but it shares with it the logical form. It is not a signified or a theme either, but it is not separable from the figures that inhabit the narrative or from the motifs that abound in it. It is inseparable from both of these, and yet it is something eise. This is what makes the analysis difficult. The mytheme cannot be isolated as an object. It situates itself in the combinations and the transformations in which it is the recurrent element. Its existence is more logical than thematic, it consists of the Operation and the interplay of relations: "In any hypothesis, one would be wrong to put the mytheme in the same category as the word or the sentence — entities whose meaning or meanings can be defmed, albeit ideally (for even the meaning of a myth varies with the context), and arranged in a dictionary. The elementary units of mythical discourse certainly consist of words and sentences, which - in this particular use and without overextending the analogy - are however more in the order of : meaningless units that are opposed within a system, where they create meanings precisely because of this Opposition". ^ The analog) of the ^theme mth the phoneme is thus of a purely methodoh^al nature. In the analysis of myths such as he practices it, Levi-Strauss has not searched to systematically isolate the in order to then show their combinations. The result of this would have been artificial, considering the plasticity of narratives and the richness of their metamorphoses. This gap between the announced method and the practiced method seemed to be an indication of a difficulty not yet overcome. Many critics have stressed it, sometimes with severity; often also by revealing themselves incapable of taking into account the limits indicated by Levi-Strauss himself. Taking the phoneme as a model could in fact present a risk in the sense that this unit is not only oppositional or relative, but especially negative, in short, it has no content of its own. The phoneme does not correspond to any signified. It would seem absurd to say so much about the mytheme, except if we forget that here the analysis is not situated on the linguistic level, but on a completely different level: that of mythical discourse. Levi-Strauss, in fact, does not say that the mytheme is deprived of linguistic sense, he even expressly asserts the opposite. In fact, everyone understands sentences like: "his head rolled and became a comet" or even "the hero climbed onto the tree and could not get back down".

5 \je RegardEloigne 199). [English translarion: The ViewfivmAfar{\^%S: 145).] 26 I M.HQ4AFF

Levi-Strauss says only the following: a narrarive segment or a semantic dement of a myth (plant, animal, body part, mineral, tool, landscape, etc.) does not give US by itself any information outside the systeni where it occurs and outside the relations of transFormation in which it is inscribed. In this perspective, its status is analogous to that of the phoneme. This analogy allows the radicalization of the method with respect to previous approaches. It is on this level that the relarion with phonology is established and is limited. Otherwise, we see that the other resources of the linguistic disciplines are used: morphology, syntax and, of course, the lexicon.^ We could point out that by using the term "language" to name the mythical whole, Levi-Strauss was thinking about language as a system, in the sense of Saussure, and thus as a Virtual device of differences and oppositions that each discourse mobilizes and actualizes. However, it is not a connection of this idnd that is assumed between myths, but - in a much more interesting way - a relarion of transformation. The great originality of Levi-Strauss — by which he completely renewed the issue of myths and symbolism - was that he liberated the study from the task of extracting significations, at least those which ancient hermeneutic traditions of interpretadon tried to estabüsh. Or, more accurately, the question of the sense of the narrarives is no longer separable from the way in which they build a universe. What matters, thus, is their coherence, their capacity to integrate the greatest diversity of elements and experiences in the system. The rigorous task of the mythologist, according to Levi-Strauss, is to expose this device and not to comment on it. To State what this principle "signifies" - if we still allow this term - amounts to understanding how it produces an intelligible order by taking over the contingent materials presented. Indeed, Levi-Strauss sensed the difficulües related to the use of the linguistic model and of the concept of signification associated with it (since all discourse aims at expressing a message). He did not explicidy formulate himself the

6 This is what Levi-Strauss vigorously reminds of in "La structure et la forme" (i960]. In this text (reprinted in Anthropobgie Struäurale Deux 1973), Levi-Strauss sets his difference with respect to Morphohg^ of the Volktak by Vladimir Propp. He blames the latter for proposing general narrative models that can apply to any sort of tales, instead of starting from different tales and from their particular contexts to be able to draw recurrent relations, i.e. structures. Propp wanted to discover molds in which he could fit all sorts of Contents. Levi-Strauss argues that Contents - the lexicon - are always contingent, not deductible. On the other hand, the kinds of relations between these Contents show an ability of ordering, assuming logical Operations that are the same of all human minds. The structural approach is thus the most attentive to the materiality of facts and of empirical data, while being the most ambi- tious about the intelligence that is displayed even in the most ordinary experiences of life: from cookery to initiation , from marriage to the narration of myths. THEMYTHOLOGIQUES | 27 reasons for these difficulries. But he found a way out by resorting to another model: music. Everything seems to show that he has remarkably solved a problem without really having perceived the facts, nor therefore suffidendy fonnulated the Solution. Hence the obscurity of certain formulations, which are extraordinarily intuitive and yet unsatisfactory in their methodological wording. It shall be up to us to clarify the issues.

Reasons and resources of the musical model It would be inaccurate to State, as others have done, that with the Mytholo^ues Levi-Strauss goes from the linguisric model, which had prevailed untU then, to a musical model. This is only partially tme in that it is still indeed the model of language as a system, on the methodoloffcal level, that remains relevant: it gives the f^theme an analogous status to that of the phoneme, that is to say, it makes it a minimal unit, purely differential, oppositional and semantically empty. Why, then, resort to music? Levi-Strauss seems to demand from music to provide him with a model of exposition. He explains himself as to this choice in the introduction (named "Ouverture") of the Mythologques, and he retums to it in his general conclusion (named "Finale"). We should nodce that after having chosen to name the different parts or moments of his outline as musical genres ("The 'Good Manners' Sonata", "Fugue of the Five Senses", "A Short Symphony", etc.) Levi-Strauss abandons this practice in the following volumes, as he would have had to simulate excessively, in an explanatory discourse, a form of language depending on a given art. However, this form of exposition would recurrendy prove something eise that Levi-Strauss had sensed but apparendy did not clearly identify: it is also a form of analysis, or rather it is what makes inoperative, indeed iUegitimate, the traditional forms of analyzing myths. In fact, for Levi- Strauss we cannot comment on myths without pretending to know more than them, without conferring a significadon that exceeds them and that is external to them; which amounts to missing them, not understanding them. From this perspective, music provides a good example: a musician "understands" and "analyzes" a given piece not by unfolding a discourse about it (which is legitimate on another level) but by taking it up again and by tran^orming it in a new composition. By asserting the same about mythical narrative, Levi-Strauss identified it, without saying so, with a symbolicprocess. The whole question is in fact how a meta-language — a critical discourse — can intervene on or account for such a process: can meta-language dissect it 28 I M.HÖMFF

without necessarily missing it? Or even expose it without necessarily taldng part in it? Outside? Inside? We shall see how Levi-Strauss struggled with this dilemma and how he asked music to offer him the niodel of a Solution. But let US retum to the problem of exposition; we shall see that it leads to that of symboiism. As all narratives do, the myth integrates a synchro- ny/diachrony Opposition. It presents a succession of events and actions. It is narration. Such is its melodic line. But this line is not yet all there is to it. Narrative implies a multitude of levels which Levi-Strauss names codes (for example: technical code, cosmic code, marriage code, culinary code, botanical Code, zoological code, etc.). Thus as a polyphonic piece, the myth ought to be read not only linearly (enchainment of narrative sequences, analogous to a melodic suite) but also from top to bottom (the codes or semanüc strata in the myth are comparable to the harmonic levels of the musical composition). This analogy with respect to the simultaneity of the levels, already outlined in the 1955 article, is certainly the most visible one, but it is perhaps not the most interesting one. There is another more important relation, which concerns language and time. Both myth and music can be said to be languages under this precise consid- eration that they ensure communication between those who listen to them (which presumes - initial condition of all language - a stability of the signs within a given Community). The analogy stops there, since unlike articulated language music is not in position to generate a signified {signifii) that, without detaching itself from the signifier {signifiant), does not however merge with it. Language produces significations, which the subject can abstract from the act of enunciation. Music, by contrast, always captures the listener immediately within and by the sound matter; it only exists in the Performance. Now, it is through analogous ways, says Levi-Strauss, that myth, while pertaining to articulated language, produces its effect. What does this mean? This: the narrative of the myth, just like the Performance of music, only exists as it unfolds in a temporal succession. Through the processing of sounds, which are chosen and used following tonal scales, music incessandy mobilizes the physiological rhythms of the listener, creating expectations, pauses, tensions and their solutions. This direcdy sensorial action is at the same time the intelligible Operation of music. No signified (as is the case of articulated language) detaches itself from it which would enable its translation into another code. How does mythical narrative emerge for this model? It cannot be from the angle of its belonging to ordinary language (whose syntax and lexicon it does THEMYTHOLOGIQUES | 29 not disrupt). Thus this can only happen on another level, which makes language enter into another system or into another dimension. On the one hand, listening to a myth, like listening to music, unfolds in an irreversible time and yet it "transmutes the segment devoted to listening to it into a synchronic totality, enclosed within itself'.^ But this likeness of temporality (which begins in the physiological rhythms) is paired with another one, namely producing the unity of the intelligible and the 'sensible'. At the very level of figures and in a multiplicity of natural elements, the myth enunciates a logic, a cosmology, a , an ethic, an aesthetic, etc. Going through the narrative sequences, going beyond them, the listener is faced with this simultaneity of schemes as a device of thought inherent to the events and to the images evoked. It is this immediate 'meaning-experience' as sensation and as knowledge that brings myth and music doser together. Between the two forms, however, there is a fundamental difference of application: "Just as music makes the individual conscious of his physiological rootedness, mythology makes him aware of his roots in Society. The former hits us in the guts; the latter, we might say, appeals to our group instinct. And to do this, they make use of those extraordinarily subde cultural mechanisms: musical Instruments and mythic pattems".® It is thus important to understand this use of the musical model, if we want to grasp the coherence of the thought process proposed and put into practice by Levi-Strauss, who finally justifies it well when he claims that what is specific to myth, just like what is particular about music, is that it cannot be transposed or translated into another signifying system. Can music, whose "peculiar quality is to express what can be said in no other way"', be related with myth, which is enunciated in language? On what level can we reveal an isomorphism? It can only be on a meta-linguistic level, that of the narrative system. It is only on this level that the myth operates on time in an identical way of that of music. Why? Because what is peculiar about such an Operation is that it is not replaceable, it does not exist in any way other than as Performance. In other words, it exists only for its addressee and during the time of its accomplishment: "In both instances the same reversal of the relation between transmitter and receiver can be observed, since in the last resort the latter finds himself signified by the message from the former: music has its being in me, I listen to myself

7 Ije Cru et le Cuit, op.cit. p. 24. (EngUsh translation The Raw and the Cootaed, op.cit., p. 16.) 8 Cru et te Cuit, op.cit. p. 36. fEnglish translation: The Raw and the Cooked, op.cit., p. 28.] 9 l je Cru et le Cuit, op.cit, p. 40. [English translation: The Raa/ and the Cooked, op.cit., p. 31.] 30 I M.HÖJAFF

through it. Thus the myth and the musical work are like conductors of an orchestxa, whose audience becomes the silent performers". The question we could raise would thus be the following: does that not also go for the reception of every work of art, as in the perception of a painting or the reading of a novel? Where is the difference? What is specific about music among the arts and myth among discourse genres that singles them out in their domain and brings them together? It all happens as though in the case of myths, unlike the problems of kinship or those posed by "totemic" classifications, something bothered Levi-Strauss in the outline of the logical device that supports the activity. As if, in this case, unlike in the previous fields of research (such as kinship, social organizations, natural taxonomies) there was an irreparable loss connected to the analysis, as if the revelation of the internal architecture of myths and of their functioning would be done at the cost of their direct enjoyment. This is clear at the end of the Ouverture of the Mythologiques, in the surprising invitation directed to the reader to forget the book he is about to read, so that "he will find himself carried toward that music which is to be found in myth"." This would necessarily be done by winning on one level and losing on another one. What is won is "the hidden significance" of these narratives, what is lost is "the power and majesty that cause such a violent emotional response when it is experienced in its original State, hidden away in the depths of a forest of Images and signs and totally füll of bewitching enchantments, since in that form at least nohodj understands it"}'^ Is this a paradox, modesty or lucidity? This explanation advanced in the Ouverture is undoubtedly not sufficient. The ¥inale proposes another one, which without renouncing the first, modifies it or completes it in an original way (and beyond any doubt the years spent reading and writing out hundreds of myths may have in the meantime produced this change of view in the author). It is no longer just a matter of opposing analysis to Performance. It is now important to assert that myths, like music, are not translatabk to any code other than their on>n. This is undoubtedly a consequence of the importance acquired by the concept of transformation: "The myths are only translatable into each other, in the same way

10 Ije Cru et le Cuit, op.cit, p. 26. [English translation based on 'l'he Kaw and the Cooked, op.cit., p. 17.1 11 Ije Cru et le Cuit, op.cit, p. 40. [English translation: i'he Raw and the Cooked, op.cit., p. .3t-32.| 12 Iji Cru et le Cuit, op.cit, p. 40. [English translation based on: l'he Raw and the Cooked, op.cit., p. 32.1 TNEMVmOLOGIQUES | 31 as a melody is only translatable into another which retains a relarion of homology with it".'^ Now, by presenring myths (and their devices) in this way, Levi-Strauss, without admitting it himself, was doing nothing but defining them as a symbolkprocess, which is characterized essentially by its operatory power. What is generally understood by symbolism, then? Or rather, what should one understand by it? We can certainly even depart from the notion of ^mbol which is normally defined in Opposition to sign. Ducrot and Todorov present things like this: "The practical test that allows us to distinguish between a sign and a is the examination of the two related elements. [...] In the Symbol, [...] they must be homogeneous. This Opposition sheds light on the problem of the arbitrary aspect of the sign [...]. The relation between a signifier and a signified is necessarily unmotivated\ the two are different in nature and it is unthinkable that a graphic or phonic sequence should resemble a meaning. At the same time this relation is necessaiy, in that the signified cannot exist without a signifier, and vice versa. On the other hand, in the symbol, the relation between the symbolizer and the symbolized is unnecessaiy (or arbitrary), since the symbolizer and the symbolized (the signifieds flame and kve) exist independently of each other; for this reason, the relation can only be motivated. [...] These motivations are customarily classified in two major groups [...]: resemblance and contiguity".'"* Unlike the sign, the sjrmbol presupposes a fusion between the sensitive support and the thing evoked. How is a symbolic device an eßäent device, or rather in what way or why must it be efficient to be claimed symbolic? No one answered this question in a better way than Dan Sperber in IJS SymhoUsme en General.^^ Sperber's main thesis is contained in this assertion: " are not signs. They are not paired with their interpretations in a code structure. Their interpretations are not mean- ings".'^ The mistake of semiology was to handle facts of symbolism as if they were about signs and their signification. Semiologists then asked themselves what the message of such signs was. This led to deciphering symbols by annuUing the specificity of symbolism while keeping this evident question unanswered: why does symbolism exist, if it must be reduced to something

13 l.'Hamme nu, op.cit. p. 577. [English transladon: The NakedMan, op. cit., p. 646.) 14 Ducrot & Todorov (1972). Dictionnaire engcbpeSque des säences du langage, pp. 134—135. pnglish translation based on: P.n

eise? Levi-Strauss seems to have clearly sensed this risk and tried to avoid this semiological reduction in his analysis of myths. Sperber thus asks die following: if a symboüc device is not a system of signs and does not aim at formularing a signification, what is it then? He answers: "Symbolism conceived in this way is not a means of encoding information, but a means of organizing it".'^ Elements appear in systems of homologies, oppositions, inversions, symmetry and other logical relations. These Operations are not to be the object of search for signification. Thus "a symbolic Opposition must not be replaced by an Interpretation, but placed in an organi^tion of which it constitutes an dement".'® In short, symbolism is primarily an operative system, an Instrument of Organization. We do not ask of an Instrument what it signifies, but rather what it is for, what it allows to produce. Symbolism produces an Order in a set of sensorial elements, an order that is intelligible with regard to homologies, oppositions and inversions etc. that it puts into play or establishes. This is precisely what Levi-Strauss calls "untamed thinking" ("la pensee sauvage"). This is how he goes about the use of natural species for introducing a system of differentiations in human society (what was traditionally called a "totemism"). This is also the case of a , which does not aim at saying something, but at keeping or modifying the course of things. Finally it is also the case of sequences and levels of a group of mythical narratives, which do not aim at expressing the world, but at organizing it.'' In front of the facts of symboUsm, the traditional questions of semiology (like those of hermeneutics) consisted of asking oneself: what does this mean? In short: what is the message of this Code? Now, the problem is not to know what the symbolic elements signify (because this would be confusing them with signs), but to understand their function. Levi-Strauss clearly says so: "Of the sun and the moon, the same thing can be said as of the coundess natural beings that mythical thought manipulates: mythical thought does not seek to give them a meaning - it expresses itself through them",^" that is to say, it exposes itself as such; or rather, it operates thanks to them. This is why a symbolic phenomenon can onlj be affected ly another one of the same order. The relations between them are those of transformation, which means that

17 Ibid. p. 82. fEnglish translation: p. 70.] 18 Ibid. p. 82. [EngUsh translation: p. 70.] 19 This is especially true about "myths" of oral traditions. The question is complctely different with respect to the mythologies re-elaborated by written traditions. 20 Anthropologie Structurale Deux (1973: 261). [Tänglish translation bascd on Structural Anthropologe, vol. II (1976: 221).] THEMYIHOLOGIQUES | 33 they are among themselves as variations of one same scheme: this goes as much for the variations of a myth, as for those of a rite. These variations do not aim at saying something different, but at completing or expanding an Operation. Narratives that seem absurd from the point of view of signification become coherent from the point of view of their Operation. Hence the analogy myth/music. A Variation in music 'translates' a theme or a sequence into sound forms and according to particular rules. This is precisely a pure symbolic process, understood as Performance. Such is exactly the new version of symbolic efficacy, which Levi-Strauss had mentioned in texts^' from the 50s and that he finds again, without admitting it clearly, in the functioning of the mythical narrative. A myth (or rather a group of myths) organizes elements, it builds an order. It has no meaning by itself, it is at the most what makes a meaning possible by the composition it produces: "A myth proposes a grid, definable only by its rules of construction. For the participants in the culture to which the myth belongs, this grid confers a meaning, not to the myth itself but to everything eise: that is, to the Images of the world, sodety, and its history, of which the members of the group are more or less aware, as well as to the questions with which these various objects confront the participants. In general, these scattered elements fail to link up and usually coUide with one another. The matrix of intelligibility supplied by myths allows us to articulate those elements in a coherent system".^ It is the discovery of this performativity of the mythical narrative that has very evidendy led Levi-Strauss to underline the analogy between myth and music, and to suggest the fading of the mythologist before his object, since we cannot do the exegesis of a myth without dissolving it, and the only possible task is to restore its operatory device. Hence no longer resorting to a discipline like linguistics, but to an art like music.

Conclusion: the position of the interpreter Myths relate to each other like the pieces of a device, not like references from signified to signified. They do not aim at saying but at organizing and trans- forming. It is in this sense that they "think each other" and even that they

21 "L'efficacite symbolique" [1949] in Anthropolo^ structurak (1958: ch. X) [English translation: "The Effectiveness of Symbols", in Stmäural Anthnpoh^, New York: Basic Books (1963: ch. X).); See also "Le sorcier et sa magie" [1949] in ibid. ch. IX. [English translation: "The Sorcerer and His Magic", ibid. ch. IX.] 22 IJ YUgardMgne 200). [English translation: Vietvfrom Afar 145-146).] 34 I M.HENAFF

"think themselves in us" beyond our awareness. We receive them at the ver\- level where they operate. Wanting to give them a meaning instead of letting them ort would be wanring to convert the symbol into a sign, replacing a process by a signification. Exacdy as if we tried to replace a musical phrase by a discourse, although it can only be "translated" by a Variation, i.e. by a transformation regulated on the same level of forms (rhythm, tonality, melody, theme, for example): "Myths are only translatable into each other, in the same way as a melody is only translatable into another which retains a relation of homolog)' with it".23 Consequent with this analysis, Levi-Strauss concludes that the book of the mythologist enters into the play of the variations of the narratives that it reports; or rather: this book confirms "the reciprocal translatability of several myths. This is why it would not be wrong to consider this book itself as a myth: it is, as it were, the myth of mythology''.^-» A surprising Statement, which has caused numerous and very lively debates. This means that the "interpretaoon" is not a meta-discourse, but that it remains an Intervention on the level of that which it is about. Interpretation also belongs to the system of transformations of forms into other forms. The position of the Interpreter as the one who could teil the meaning of the object considered, is thus made clear. His role is only to bring narratives in contact, to enhance their interaction: in short, to favor their reciprocal "translation". He is only a go-between, a catalyst of relations. We thus better understand the interpretative restraint of Levi-Strauss; a reticence that often disappointed his readers. However, while he refuses to take the position of the one who would know the "trudi" about the myths he reports, at the same time Levi-Strauss invites his readers to share a rieh experience: to enter with him in the narrative movement and to let themselves be moved by "the music that is in the myths".

Translatedfrom the French bj Ana Matgarida Ahr an tes

References

Ducrot, O. & Todorov, T. (1972). Diäionnaire eng/clopedique des sdences du langage. Paris: Seuil. jEnglish translation by C. Porter (1979). Engäopedic Dictionary of the Sdences of Ijinguage. Baltimore and London: The John Hopkins Universit)' Press.]

23 I,'Hamme nu, op.cit. p. 577. [English translation: The NakedMan, op. cit., p. 646.] 24 / ^ Cm et U Cuit, op.cit. p. 20. [English translation l'he Raw and the Cooked, op.cit., p. 12.] THEMYTNOLOGIQUES | 35

Levi-Strauss. C. (1955). The Structural Study of Myth. joumal of American t'olkhre 68 (270). (Originally published in English.] Levi-Strauss, C. (1958). Anthropologe Structurak. Paris: Plön. [English translation by C. Jacobson (1963). StructuraJAnthropologe. New York: Basic Books.] Levi-Strauss, C. (1964-1971): Mjtholo^ues. Paris: Plön. [English translation by J. Weightman & D. Weightman. MythoU^ues. Introduction to a Science of Mythologf, New York: Harper and Row.] 1964. Le Cru et le Cuit. \The Rav and the Cooked. 1969.]; 1966. Du Mielaux Cendres. [Frvm Hon^ to Ashes. 1973.]; 1968. IJOrigne des Manieres de Table. [The Origin of Table Manners. 1978.]; 1971. L'Homme Nu. \The Naked Man. 1981.] Levi-Strauss, C. (1973). Anthropologe Structurak Deux. Paris: Plön. [English translation by M. Layton (1976). Structural Anthropologe, voL IL New York: Basic Books.] Levi-Strauss, C. (1978). Myth and Meaning. Toronto: University of Toronto Press. Levi-Strauss, C. (1983). Le Regard Eloigne. Paris: Plön. [English translation by J. Neugroschel & P. Hoss (1985). The ViewfromAfar. New York: Basic Books.] Sperber, D. (1974). Le Symholisme en General Paris: Hermann. [English translation by A. L. Morton (1975). Rethinking Symbolism. Cambridge: Cambridge University Press.]