Michael Riffaterre Transposing Presuppositions on the Semiotics of Literary Translation - Классики о переводе - Каталог статей

Michael Riffaterre Transposing Presuppositions on the Semiotics of Literary Translation Literary translation is different from translation in general for the same reasons that literature is different from nonliterary uses of language. Literature is distinguished from them, first by the semi-oticization of discursive features (for instance lexical selection is made morphophonemically as well as or more than semantically). Second, by the substitution of semiosis for mimesis (this covers the consequences of the indirection of meaning that is the pivot on which literariness turns). And third, by textuality that integrates semantic components of the verbal sequence (the ones accessible to linear decoding)—a theoretically open-ended sequence—into one closed finite semiotic system, the very existence of which is not manifest until readers become aware of the connection between the text and an intertext. It is only then, in the light of that intertext, that the discrete meanings of the words, phrases, and sentences composing the text assume new functions in its general scheme. Only then do these discrete meanings yield to an overall significance resulting from their implication of or opposition to the intertext. Literary translation must reflect or imitate these differences. First, it must semioticize forms and sounds like the original, although in a different system (which is easy: equivalent alliterations, for example, demand only a repetition of sounds, but not necessarily the same sounds as in the source text). This aspect, I will lay aside. Second, literary translation must render both meaning and significance. The literary text requires a double decoding, at the levels of both systemic structure and of its component parts. This decoding too must be translated in a way that will induce the reader of that translation to perform likewise a double decoding. The signals guiding readers in such a decoding in the original must be reproduced in the translation. Literary translation must also reproduce those features of the original text that are the traces left by its production. That is, the translation, like the original but not necessarily in the same way, must be visibly derivable from the formal or semantic given that determined that production. Other forms that we empirically sense to be signs of literariness also must be rendered, even though they do not directly or primarily carry meaning: for example, the signs indicating the genre the text belongs to, the signs making obvious that it is artifact rather than a plain representation of reality. A simple way to rephrase these problems and define their constraints is to express them in terms of presuppositions—the implicit and requisite conditions of the text. A translation presupposes a source text. Within the source, its literary features in turn presuppose, and they function because of what they presuppose. The text's artifice, its being an artifact, presupposes an author. Whether implied or represented, this author is translatable, as style if implied, or as mimesis if represented. Style and topic presuppose a genre. Indirection of meaning (the subordination of word and sentence meanings to textual significance, and that of literality to figurativeness or symbolism) presupposes the sociolect and the intertext, the first because it contains the rule and the norms indirection violated, the second because it represents the substitute authority presupposed by the means used to violate the sociolect's authority. No literary translation therefore can ever be successful unless it finds equivalencies for these literariness-inducing presuppositions. Some equivalencies however may not be found in the target language at the same level as in the source language: equivalencies to lexical features of the original may have to be found at the syntactical level in translation, and the reverse is true as well. I will concern myself with these only, and propose the view that in order to palliate the fact that words translated term for term only rarely exhibit the same presuppositions as in the original, the translator must transpose presuppositions. He will do so either by taking them up directly, or by shifting the burden of presupposing to another segment of the text. In the first instance, the presupposed word is translated, rather than the presupposing one. This, certainly, is the practice of successful translators. In Milton's translation of Horace (Odes, I, V), than which one more perfect is hard to imagine, his only two departures from the source text illustrate the point: Quis multa gracilis te puer in rosa perfusus liquidis urget odoribus grato, Pyrrha, sub antro? cui flauam religas comam, simplex munditiis? heu quotiens fidem mutatosque deos flebit et aspera nigris aequora uentis emirabitur insolens, qui nuc te fruitur credulus aurea, qui semper uacuam, semper amabilem sperat [ . . . ] What slender Youth bedew'd with liquid odours Courts tbee on Roses in some pleasant Cave, Pyrrha for whom bind'st thou In wreaths thy golden Hair, Plain in thy neatness; O how oft shall he On Faith and changed Gods complain: and Seas Rough with black winds and storms Unwonted shall admire: Who now enjoyes thee credulous, all gold. Who alwayes vacant, alwayes amiable Hopes thee [ . . . ] Milton uses "golden" both for flauam "blonde" (1.4) and for aurea (1.9), "golden" the first for the hair, the second for the girl. This one rendering of two different words, I believe, is due to the pressure of semantic overdetermination: we have one physical and one moral representation, but the significance stems from the first symbolizing the second. Milton, of course, is aware that Pyrrha means "redhead, she of the flaming hair," so that the female protagonist is less a person than an embodiment of deceptively pleasant appearances. It cannot be by chance that "credulous" and "all Gold" should be linked together—Cupid's treachery teaches us that all that glitters is not gold. The second departure is bedew'd, instead of "bathed in scents." Clearly it transposes to the lover the features of a flower that is presupposed by odours, thus avoiding in the target language the effeminate connotations that being drenched in perfume would suggest to the British. The power of implication (the latency of the repressed flower) is such that a modern translator makes a telling mistake: instead of liquid odours, he writes liquid flowers. Transposing presuppositions will mean either making the implicit explicit or a lateral displacement whereby the semiotic detour, a figurative turn of phrase for instance, will be replaced by a me-tonym of the representation that was blocking the way. I am not suggesting that we replace translation with paraphrase, for the paraphrase would still be linear. Nor am I suggesting a commentary which would mean separating the linear reading from the intertex-tual decoding of significance, keeping the first (not literary per se) in the text and relegating the latter (the essence of literariness) to a footnote. I am suggesting a limited periphrasis built around the matrix word of which the periphrasis is the transform. Whereas a normal periphrasis represses a lexeme (which remains present to the reader's mind as a referent, as the latent matrix of the ambages he is forced to follow) by transforming it into a syntagm, the translator's periphrasis will retain the lexeme's literal meaning and will develop the significance, the implications of the presupposed, in the syntagmatic derivation proper. Nowhere can we find a better example of the role played by presuppositions than in the case of specific, detailed literary descriptions. Their function, insofar as they are specific, runs the gamut from creating the illusion of reality to creating an unacceptable representation. In the former instance, such descriptions make for verisimilitude. In the latter, the subverted mimesis generates an idiolect which, besides representing (however abnormally), gives its visible, unique shape to the text's significance. In both cases, however opposed they may seem to be, accuracy in depicting and in naming is equally essential in establishing the given from which every pertinent, significant aspect of the source text is derived. Translators are here faced with the peculiar difficulty of imitating or duplicating text production in the target language: the translation of that part of the original that is derivative, which is to say, generated from the given by a series of variants (these being literally translations of synonym into synonym, albeit within one language, as I pointed out earlier), must be as derivative as it is in the original. Otherwise the unity of the text, its function as a semiotic agent of transformation, as one unit of significance, will be lost The difficulty is not just that the translator must find a generator with a productivity equal to that of the original, but that he must define precisely what is at work in the original and which semes of the matrix word are activated. My point is that it is not the visible, explicit features of the matrix that make it functional, that makes it the motor of text production, the starting point of rhetorical amplification, but its presuppositions. The pressure that these bring to bear on the text's future development results from their being repressed by the very act of naming the word that refers to them, thus placing the burden of imagining them on the reader. The derivation from this starting point will be constituted by that which the matrix's explicit features entail, but these entailments themselves are none other than the lexical and syntactical surfacing of the repressed presuppositions. Thus is generated a celebrated tableau of Latin poetry: the Ocean nymphs in the opening scene of Catullus's poem 64, the "Epithalamion of Peleus and Thetis," the parents of Achilles. The entire main stream of epic poetry flows symbolically from Thetis' bosom, leaving no doubt that the poem's thrust is to be a tale of beginnings, to be a type of splendid after-the-fact preface to Homer, the Oresteia and Virgil. Every detail thus is charged with the symbolism of the world of myths that the reader will be able retrospectively to connect with it. This river of images rises from the very first scene, in which the Argonauts sailing to the conquest of the Golden Fleece are sighted by Nereids. One of them spots Peleus, falls in love with him and thus Achilles will be born. The motif itself includes a double transgression. First, the symbolic and dramatic crossing of a frontier: a goddess marries a mortal. Then, a creature from the sea, she has to overcome yet another barrier, which separates the submarine world from the human, the fantastic from the natural. The power of this miracle (and its demonstration that Eros conquers all) can be verified in modern tales, since it still endures, from the mermaid of Hans Christian Andersen to Lois Lemaris, the siren-like temptress in Superman, to the amphibious maiden of the 1984 movie, Splash. From the depths, therefore, rise the Nereids, among them Thetis: "the ship's ram had hardly furrowed the windy plains, the waves torn open by the oars were just turning white with foam, when the Nereids emerged from the boiling abyss to marvel at this monster." And now the detail that I think begets the rest: hac, illa atqae alia uiderunt luce marinas mortales oculis nudato corpore Nymphas nutricum tenus extantis e gurgite cano (16-8) [on that day, on the next and then once more, mortal eyes were able to see the sea Nymphs, standing out from the white abyss with their bodies exposed as far as the breasts]1 No sailor could resist such an impudent come-on: "Then is Peleus said to have caught fire with love of Thetis, then did Thetis not disdain mortal espousals." This is the F. W Cornish translation (Loeb Classical Library). Its lofty tone makes up for the graphic depiction of scenes of exhibitionism, for a glimpse at the power of raw sex, from which so many woes issue: Helen willingly ravished, and the fall of Troy, Agamemnon felled by Clytemnestra's axe and the Eumenides unleashed, Dido seduced and the rise of Rome and Carthage's ruin, etc. All this derives ultimately from the one detail translators refuse to come to terms with: the nymphs are shown upthrust above the waves baring their bodies naked as far as the breasts—"nutricum tenus." Every translation I have seen is inaccurate even when it dares to be specific. Lefevere: "as far as the nipples," and on second thought, "as far as the breasts," treading lightly as it were.2 Burton averts his gaze and concentrates on the missing bra: "exposed bodies denuded of raiment Bare to the breast." Hart-Davies resorts to neutral commonplaces of ekphrasis: "fair naked forms with white breasts glistening." Sisson, on the contrary, more perceptive or cynical, sees the nymphs as women natives spotting Captain Cook's sailors, and he calls a spade a spade: "naked sea-nymphs sticking out of the water showing their tits." Despite, or rather because of the vulgarity of "sticking out" and of "tits," Sisson's female flashers do suggest accurately the frenzy of mutual hot pursuit, in which Peleus's lust matches his future bride's. But Sisson's explicitness and the others' evasions equally miss the other implications of Thetis's exhibition, of "nutricum tenus," for it is, I think, this phrase that gives the whole scene its significance. Tenus, the preposition, up to or indifferently down to, in such a context is the very mechanism of sexual implication. It does not matter whether undressing starts from the top or from below the waist, unbuttoned blouses or lifted skirts, in any erotic text the indication of a limit to the baring of flesh is a presupposition of something more to come and more to expose. This is the suspense of striptease itself, the first step of voyeurism, the anticipation of desire. Furthermore, in the semiotic system of the body, the feminine breast is not bare, it is not shining forth erogenous unless the nipples stand out. Passing this point is the difference between a mere decollete, however promising the cleavage may appear, and aggressive desire. But what is it that you see? The word used by Catullus to speak of desirable breasts, of erotic symbols, is, to say the least, peculiar. So peculiar that not a single translator heeds it. Instead of speaking of breasts, or of nipples as in an equally hot passage fifty lines below (Ariadne panting for Theseus), Catullus uses nutrices, and nutrix emphatically and precisely designates she who suckles an infant, a nurse, not a synecdoche for nurse, not a milk-filled breast, but the whole person. These mermaids are naked to their nurses inclusive. It would be hard to find a worse letdown, at least in most Western sociolects, for they tend to separate sex from motherhood, and most readers therefore keep them apart in their fantasies. Probably also for Catullus's contemporaries, since this seems to be a once-only boldness of expression, and other literary uses of the word, figurative or literal, refer to the whole nurse from head to foot, and a professional one to boot. Translators seem to assume that by identifying nutrices as a metonymy, they are free to ignore it, since a literal translation would be grotesque, and free to substitute the generic word, breast, to which the metonym refers indirectly. The trouble is, indirection is precisely what points to the presupposition. It does so the more powerfully because nutrix is so inseparable from a vast descriptive system of words all very visibly its cognates (nutricatus, nutricium, breast feeding and also the nurse's fee, nutricula, the little mother etc. down the paradigm of compounds of nutria "to feed"). So nothing can blunt the oxymoron, or get us acclimatized to the audacity of the image. The whole weight of presupposition is that this lusting sea-goddess will not just be a wife to the hero, a development that is never more than anecdotal in mythology, but a mother. She therefore will be the source for one of those genealogies of heroes that are significant in mythology. They are semiotic rather than narrative, for they correspond to the impulse in mythopoesis to multiply characters that are exemplary in some way. They become the means of revealing the continued agency of Fate. Thetis will generate a line of terrifying goddesses, demi-goddesses and illustrious women sinners whose sexuality will change the face of the world, from generation to generation, and destroy empires. It cannot be by chance that towards the end of the epithalamium, on the morning after the wedding night, the task of verifying that the bride is no longer a virgin should be entrusted to her nurse (the same word nutrix is used again, now literally), nor is it by chance that the poem, on the whole cheerful and festive, should end on a pessimistic tableau of mankind as imperial Rome has corrupted it. It cannot be by chance that at the clausula, unexpectedly and seemingly gratuitously, we should have the image of an incestuous mother "mater substernens se impia gnato," this could not be more graphic— the mother lying supine under the body of the son of her womb. There is a passage in Trollope (The Duke's Children, chap. 72) where not dissimilar anxieties are made explicit, evidently because a narrative always deploys what poetry implies, and because psychological analysis, that fiction favours, bare the character's inmost fears. This passage suggests that I am not reading too much into Catullus's two words: But this girl, this American girl, was to be the mother and grandmother of future Dukes of Omnium,—the ancestress it was to be hoped, of all future Dukes of Omnium! By what she might be, by what she might have in her of mental fibre, of high or low quality, of true or untrue womanliness, were to be fashioned those who in days to come might be amongst the strongest and most faithful bulwarks of the constitution. Presupposition thus organizes the whole poem, regulating a derivation that pervades everything and the end of which dictates the clausula and affirms textuality. It seems to me that no translation that would be true can neglect it, and that it is revealing that translators should have recoiled from such a total experience. It is precisely the locus of originality, the source of all poetic power in a text that otherwise would be a mere epyllion, a graceful exercise on a literary genre. The poem is much more than that because a presupposition has been substituted for the only one the sociolect proposed as a metonym of sexuality—the ultimate ungrammaticality triggers the semiosis. Little wonder a literal translation should here be twice unbearable, first because the image is ludicrous, and second because the only way to be blind to its ridicule is to subordinate it to an even more troubling semiosis—that sex is the hereditary form of fate. Little wonder; but it is only more imperative that such a translation should be attempted: the only way, I suppose, would be a periphrasis suggesting that these jutting breasts are also a promise or threat of a harvest of heroes. Some may still regard periphrasis as a translator's cop-out, a facile way out of a difficulty. No doubt a more elegant solution would be to maintain in the target language an implicitation matching the original's clever pointing to the presupposed. But such striving must be in vain, and indeed lead to the loss of the original's connotation, if the presupposed is intertextual. In such instances, the translator cannot hope to avoid providing an explicit equivalent, within the text, of what the original merely alluded to by its implicit reference to an intertext. In most cases it is not possible to find a comparable intertext in the target language, in which the literary canon is bound to be totally different from that in which the original text is immersed. Take the following lines from a Maurice Fombeure poem entitled "Présence des Automnes" ("Presence of the Fall Season," or better, "Presence of Autumns") in which the plural refers to a recurrence, the natural cycle that enables us in the Fall to remember other autumns as well and experience anew their melancholy: Le brouillard noie les cachédrales [...] Je songec aux brumes septembrales Dessus les vignes de chez nous. Fog is drowning the cathedrals [...] I dream of the mists of the vintage-season Upon our village's vineyards. One may wonder why I do not simply say "September mist" and why I invent a village instead of "the vineyards back home." Indeed, had I been more timid, the more obvious version I have just proposed as an alternative would still have translated accurately the surface of the text or its linearity. A literary text, however, cannot be cut off from its intertext, and a literary reading of it in an intertextual reading. An intertextual overdetermination takes charge, eliminating any chance that the adjective septembre might only refer to September, or that chez nous might only refer to home, that plain referentiality be a defensible interpretation. Intertextual overdetermination links together two representations of homely values: chez nous, "at home" or "back home" ("back home" is preferable because the first line, and the intervening lines make it clear that the speaker is in Paris and therefore harks back to the country, that is, to a nostalgic topos); chez nous is on a par with the preposition dessus, "upon". The reader expects sur. "Expects" or rather "expected" would be more accurate, for the reader discovers a frustrated expectation a posteriori, after the text has given him a turn of phrase that in retrospect he might have expressed otherwise or more naturally. More precisely still, since the reader is unlikely to stop and ponder, what occurs is that he notices the relative ungrammaticality of dessus, of an adverb used where a preposition should be, and he therefore recognises it as the equivalent of sur (sur les vignes), which is what one would naturally say. The reader intuitively performs this translation within his own language (such "translations" are our mode of perception of stylistic registers, of figurative discourse, and generally, of textual idiolects). He performs also an interpretation, because dessus used as a preposition is an archaism, and one known to him from folk songs. Dessus and chez nous connote traditional, quaint values, the values of life in the provinces, sheltered from the corrupt influence of the big city. The proper strategy for the translation is to shift the homely, comfortable connotations of the archaic preposition to "back home," and to make "back home" the village of our innocence, like Proust's Combray, the place where grandparents waited for the young ones raised in the city to come back to a bath of purity, etc., etc. I am thus transposing the presuppositions of an obsolete grammatical connective to a location symbolic of one's youth or roots. This harking back to an effective past is confirmed by the appearance in the next stanza of a shepherd wrapped in his cape—a character representing rustic simplicity, for the name of his cape or great-coat, houppelande, is another archaism and one that is used virtually only for folklore tableaux. It is perceived as designating metonymically a rustic character, whose silhouette merely translates a favourable, eulogistic idea of the past into a human being code. What about septembrale? The adjective is not just an adjective corresponding to the noun septembre. For one thing, no other month-name has ever produced an adjective in French, so that sep-tembral stands out as an exception, one that its ending underscores. No French reader is likely to pass it by without noticing it. Most will recognize it for what it is, that is, a quotation from Rabelais, who apparently coined the word. It is therefore a purely conventional sign, an index pointing to artificiality as a component of literariness, much more than just a word referring to a season of the year. Nor is this all. Rabelais uses it only once, in connection with the word purée, "a mash": purée septembrale. The two words are so inseparable that they function as a playful compound, a funny periphrasis for wine (purée then is the stuff that has been crushed and pressed in the wine press, a synecdoche for the wine that it will produce after fermenting). The periphrasis is the better known because it is a euphemism, and a playful one used by one of Rabelais' high-living monks to speak of a drunk while seeming to mince words, with a mock respect for propriety: he is drunk not because he drank too much, but because he sniffed the fragrance of a certain delicious purée. The fortune of the quotation has been enormous, since in any society overindulgence is the occasion for euphemism. To summarize, Fombeure is using the code of the wine harvest to speak of the Fall, as Keats had used the code of winnowing to describe the hair of his allegorical Autumn lifted by the breeze. In making my decision, as a translator, to spell out the wine harvest, to extract it from the adjective, I may still appear to read too much into Rabelais' coinage. Why could Fombeure not borrow a valorized and unique adjective, cleanse it from its winery connotations, and fit it to a new, serious purpose? This nonce-word would then be only a literary word, and presuppose a literary, exemplary autumn, a theme, with its attendant precise imagistic connotations, and its precise melancholy mood, rather than a season that would have to be for all men? Why can I not accept an author's choice to separate septembrale from its puree? My first answer is that this September fog spreads over vineyards. This, however, would not suffice, and the text needed stronger associations to produce its final, fully motivated form that we will not be tempted to tamper with. Overdetermination does the trick, and it is provided here by the echo from another verbal joke. The stanza evokes an autumn of fog and mist. It functions as a description first of all because it unfolds a vaporous paradigm that continues in the next stanza with le ciel fumeux, "a sky as if smoky." And it so happens that there exists a word jokingly used to designate a thick fog, a borrowing from the colloquial English pea-souper, a borrowing as intertextual and as conspicuous as the one from Rabelais. This word is purée in purée de pois, "pea soup." Septembrale as well as brouillard presuppose the same noun, the relatively rare word purée—rare, that is, in literature, unless it is used as a quotation and a humorous device. Purée, the silent, repressed link, steeps in Gallic tradition the image of a nostalgic autumn. Purée, equally pertinent to "fog," to the "wine-harvest," and to Gallic wit, represents now not just a presupposition but an interpre-tant, that links the metonyms, fog, vines, countryside, with the kernel word of the title—Autumn—an interpretant that replaces a reference from words to things with the presuppositions of an object that is itself a system of signs. Again, as with Keats, the season is represented through its atmospheric features and its seasonal business. Since all this is achieved through the twist of an intertextuality that is not available in the target language, the only solution is to make the presupposed tableau explicit and complete, which I did. Fombeure's next stanza corroborates this interpretation. It goes on representing autumn by further developing the paradigm of a metonym of it. This time, the metonym is the southward flight of migratory birds: the geese are leaving. Here again the text proposes the image indirectly, through an intertext, but one organized by virtue of a structure so generally known and whose variants are so widespread in Western culture that the translator will have no trouble finding equivalents. Dans le ciel fumeux et 1éger Crient les migrateurs isocèles. [Through a sky of light smoke the isosceles migrators go trumpeting. Or, Through the light smoky sky the call is heard of the isosceles migrators.] I am afraid that my variation will not prevent "isosceles" from sounding contrived. And yet I cannot and must not avoid it. It has the power of catachresis, turning as it does periphrastically around the familiar motif of the geese flying in triangle. It is only with the last lines that the image is retranslated into its ordinary version-repeated, but with a variation, which device elegantly serves as .1 clausula. The clausula makes it clear that the whole poem is centred on the nostalgia for an elsewhere. This is the theme for the whole-collection the title of which (À dos d'oiseau [Riding on bird-back]) is now referred to by the clausula. At this point, Fombeure's choice of a geometrical technicism, so strikingly at odds with the rustic, colloquial, homely context, appears compelling. For "isosceles" is now identified with two presuppositions so widely exemplified that they effortlessly transcend linguistic borders. What the image presupposes here is first that earthbound observers can tell migratory birds from their flying in triangle formation (a spectacle that in turn entails or completes the conventional tableau of autumn). Second, the flight formation itself presupposes an aesthetic and ethical view of nature. In this view, an ordinary, natural spectacle is perceived and celebrated as an extraordinary, natural wonder. The presupposition is that these birds have by instinct an ideal navigation system. Hence an anthropomorphic praise of their natural ingenuity: they are to be admired insofar as they observe a regularity that should be the privilege of man. These birds twice qualify as a literary motif: because they stand for nostalgia, and because they act in a human manner. This causes a structural transformation whereby the theme of animals equating man is interpreted as animals superior to man. This transformation we recognize as subsidiary to an even more generally applicable structure, the paradoxical equation natural is preternatural, which is also basic to didactic poetry. This being a fundamental structure of the imagination, translation will raise no difficulty. To be sure "isosceles migrators" may always look artificial in English, but if so it remains faithful to the artifice of the French original. In both languages, however, this artifice is balanced, corrected, and given a perverse appropriateness by the fact that a geometrical term for birds in the air happens to be also a variation on spatial geometry, géeométrie dans l'espace. Textual overdetermination enabling readers to interpret (that is, to translate into simpler or more accepted phraseology) obscure or unusual images within their own language, and to find legitimacy in bizarre metaphors, is the translator's surest guide. Difficult, almost untranslatable passages bespeak the need for a gloss. Not a gloss in the guise of a Nabokovian footnote à la Pnin, but a periphrasis. The near untranslatable simply demands that we replace the lexeme-for-lexeme substitution with a syntagm-for-lexeme substitution. Presupposition lends itself naturally to this strategy, since presupposing requires a syntagm anyway. Descriptive systems stand by ready to supply such syntagms. Already with the source language, presupposition is actualized just as naturally in the mind of the reader as it is in the text, a parallel development made possible by the common model of the descriptive system. The translator's solution therefore should be to actualize the relevant parts of a system. Perhaps the simplest way to state the difference between literary and non-literary translation is to say that the latter translates what is in the text, whereas the former must translate what the text only implies.
1 2 3 4 5 Категория: Классики о переводе \| Добавил: Voats (20.09.2011)
Просмотров: 2632 \| Рейтинг: 0.0/0