AFTER BABEL ASPECTS OF LANGUAGE AND TRANSLATION GEORGE STEINER Oxford University Press LONDON OXFORD NEW YORK OXFORD UNIVERSITY PRESS Oxford London Glasgow New York Toronto Melbourne Wellington Nairobi Dar es Salaam Cape Town Kuala Lumpur Singapore Jakarta Hong Kong Tokyo Delhi Bombay Calcutta Madras Karachi 0 9 8 7 6 Copyright O 1975 by George Steiner Library of Congress Catalogue Card Number: 74-29207 First published by Oxford University Press, New York and London, 1975 First issued as an Oxford University Press paperback, 1976 Printed in the United States of America Chapter Five THE HERMENEUTIC MOTION i The hermeneutic motion, the act of elicitation and appropriative transfer of meaning, is fourfold. There is initiative trust, an investment of belief, underwritten by previous experience but episte-mologically exposed and psychologically hazardous, in the meaning-fulness, in the 'seriousness' of the facing or, strictly speaking, adverse text. We venture a leap: we grant ab initio that there is 'something there' to be understood, that the transfer will not be void. All understanding, and the demonstrative statement of understanding which is translation, starts with an act of trust. This confiding will, ordinarily, be instantaneous and unexamined, but it has a complex base. It is an operative convention which derives from a sequence of phenomeno-logical assumptions about the coherence of the world, about the presence of meaning in very different, perhaps formally antithetical semantic systems, about the validity of analogy and parallel. The radical generosity of the translator ('I grant beforehand that there must be something there'), his trust in the 'other', as yet untried, unmapped alternity of statement, concentrates to a philosophically dramatic degree the human bias towards seeing the world as symbolic, as constituted of relations in which 'this' can stand for 'that', and must in fact be able to do so if there are to be meanings and structures. But the trust can never be final. It is betrayed, trivially, by nonsense, by the discovery that 'there is nothing there' to elicit and translate. Nonsense rhymes, poisie concrete, glossolalia are untranslatable because they are lexically non-communicative or deliberately insignificant. The commitment of trust will, however, be tested, THE HERMENEUTIC MOTION 297 more or less severely, also in the common run and process of language acquisition and translation (the two being intimately connected). 'This means nothing' asserts the exasperated child in front of his Latin reader or the beginner at Berlitz. The sensation comes very close to being tactile, as of a blank, sloping surface which gives no purchase. Social incentive, the officious evidence of precedent— 'others have managed to translate this bit before you*—keeps one at the task. But the donation of trust remains ontologically spontaneous and anticipates proof, of ten by a long, arduous gap (there are texts, says Walter Benjamin,which will be translated only 'after us'). As he sets out, the translator must gamble on the coherence, on the symbolic plenitude of the world. Concomitantly he leaves himself vulnerable, though only in extremity and at the theoretical edge, to two dialectically related, mutually determined metaphysical risks. He may find that 'anything* or 'almost anything' can mean 'everything'. This is the vertigo of self-sustaining metaphoric or analogic enchainment experienced by medieval exegetists. Or he may find that there is 'nothing there' which can be divorced from its formal autonomy, that every meaning worth expressing is monadic and will not enter into any alternative mould. There is Kabbalistic speculation, to which I will return, about a day on which words will shake off 'the burden of having to mean' and will be only themselves, blank and replete as stone. After trust comes aggression. The second move of the translator is incursive and extractive. The relevant analysis is that of Heidegger when he focuses our attention on understanding as an act, on the access, inherently appropriative and therefore violent, of Erkenntnis to Dasein. Dasein, the 'thing there', 'the thing that is because it is there', only comes into authentic being when it is comprehended, i.e. translated.1 The postulate that all cognition is aggressive, that every proposition is an inroad on the world, is, of course, Hegelian. It is Heidegger's contribution to have shown that understanding, recognition, interpretation are a compacted, unavoidable mode of attack. We can modulate Heidegger's insistence that understanding is not a matter of method but of primary being, that 'being consists in the 1 Cf. Paul Ricoeur, 'Existence et hermeneutique' in Le Conflit its interprita-tions (Paris, 1969). 298 AFTER BABEL understanding of other being* into the more naive, limited axiom that each act of comprehension must appropriate another entity (we translate into). Comprehension, as its etymology shows, 'comprehends* not only cognitively but by encirclement and ingestion. In the event of interlingual translation this manoeuvre of comprehension is explicitly invasive and exhaustive. Saint Jerome uses his famous image of meaning brought home captive by the translator. We 'break' a code: decipherment is dissective, leaving the shell smashed and the vital layers stripped. Every schoolchild, but also the eminent translator, will note the shift in substantive presence which follows on a protracted or difficult exercise in translation: the text in the other language has become almost materially thinner, the light seems to pass unhindered through its loosened fibres. For a spell the density of hostile or seductive 'otherness' is dissipated. Ortega y Gasset speaks of the sadness of the translator after failure. There is also a sadness after success, the Augustinián tristida which follows on the cognate acts of erotic and of intellectual possession. The translator invades, extracts, and brings home. The simile is that of the open-cast mine left an empty scar in the landscape. As we shall see, this despoliation is illusory or is a mark of false translation. But again, as in the case of the translator's trust, there are genuine borderline cases. Certain texts or genres have been exhausted by translation. Far more interestingly, others have been negated by transfiguration, by an act of appropriative penetration and transfer in excess of the original, more ordered, more aesthetically pleasing. There are originals we no longer turn to because the translation is of a higher magnitude (the sonnets of Louise Labé after Rilke's Umdichtung). I will come back to this paradox of betrayal by augment. The third movement is incorporative, in the strong sense of the word. The import, of meaning and of form, the embodiment, is not made in or into a vacuum. The native semantic field is already extant and crowded. There are innumerable shadings of assimilation and placement of the newly-acquired, ranging from a complete domestication, an at-homeness at the core of the kind which cultural history ascribes to, say, Luther's Bible or North's Plutarch, all the way to the permanent strangeness and marginality of an artifact such as THE HERMENEUTIC MOTION Nabokov's 'English-language' Onegin. But whatever the degree of 'naturalization', the act of importation can potentially dislocate or relocate the whole of the native structure. The Heideggerian 'we are what we understand to be' entails that our own being is modified by each occurrence of comprehensive appropriation. No language, no traditional symbolic set or cultural ensemble imports without risk of being transformed. Here two families of metaphor, probably related, offer themselves, that of sacramental intake or incarnation and that of infection. The incremental values of communion pivot on the moral, spiritual state of the recipient. Though all decipherment is aggressive and, at one level, destructive, there are differences in the motive of appropriation and in the context of 'the bringing back*. Where the native matrix is disoriented or immature, the importation will not enrich, it will not find a proper locale. It will generate not an integral response but a wash of mimicry (French neo-classicism in its north-European, German, and Russian versions). There can be contagions of facility triggered by the antique or foreign import. After a time, the native organism will react, endeavouring to neutralize or expel the foreign body. Much of European romanticism can be seen as a riposte to this sort of infection, as an attempt to put an embargo on a plethora of foreign, mainly French eighteenth-century goods. In every pidgin we see an attempt to preserve a zone of native speech and a failure of that attempt in the face of politically and economically enforced linguistic invasion. The dialectic of embodiment entails the possibility that we may be consumed. This dialectic can be seen at the level of individual sensibility. Acts of translation add to our means; we come to incarnate alternative energies and resources of feeling. But we may be mastered and made lame by what we have imported. There are translators in whom the vein of personal, original creation goes dry. MacKenna speaks of Plorinus literally submerging his own being. Writers have ceased from translation, sometimes too late, because the inhaled voice of the foreign text had come to choke their own. Societies with ancient but eroded epistemologies of ritual and symbol can be knocked off balance and made to lose belief in their own identity under the voracious impact of premature or indigestible assimilation. The 30o AFTER BABEL cargo-cults of New Guinea, in which the natives worship what airplanes bring in, provide an uncannily exact, ramified image of the risks of translation. This is only another way of saying that the hermeneutic motion is dangerously incomplete, that it is dangerous because it is incomplete, if it lacks its fourth stage, the piston-stroke, as it were, which completes the cycle. The a-prioristic movement of trust puts us off balance. We 'lean towards' the confronting text (every translator has experienced this palpable bending towards and launching at his target). We encircle and invade cognitively. We come home laden, thus again off-balance, having caused disequilibrium throughout the system by taking away from 'the other' and by adding, though possibly with ambiguous consequence, to our own. The system is now off-tilt. The hermeneutic act must compensate. If it is to be authentic, it must mediate into exchange and restored parity. The enactment of reciprocity in order to restore balance is the crux of the metier and morals of translation. But it is very difficult to put abstractly. The appropriative 'rapture' of the translator—the word has in it, of course, the root and meaning of violent transport— leaves the original with a dialectically enigmatic residue. Unquestionably there is a dimension of loss, of breakage—hence, as we have seen, the fear of translation, the taboos on revelatory export which hedge sacred texts, ritual nominations, and formulas in many cultures. But the residue is also, and decisively, positive. The work translated is enhanced. This is so at a number of fairly obvious levels. Being methodical, penetrative, analytic, enumerative, the process of translation, like all modes of focused understanding, will detail, illumine, and generally body forth its object. The over-determination of the interpretative act is inherently inflationary: it proclaims that 'there is more here than meets the eye', that 'the accord between content and executive form is closer, more delicate than had been observed hitherto'. To class a source-text as worth translating is to dignify it immediately and to involve it in a dynamic of magnification (subject, naturally, to later review and even, perhaps, dismissal). The motion of transfer and paraphrase enlarges the stature of the original. Historically, in terms of cultural context, of the public it can reach, THE HERMENEUTIC MOTION 30I the latter is left more prestigious. But this increase has a more important, existential perspective. The relations of a text to its translations, imitations, thematic variants, even parodies, are too diverse to allow of any single theoretic, definitional scheme. They categorize the entire question of the meaning of meaning in time, of the existence and effects of the linguistic fact outside its specific, initial form. But there can be no doubt that echo enriches, that it is more than shadow and inert simulacrum. We are back at the problem of the mirror which not only reflects but also generates light. The original text gains from the orders of diverse relationship and distance established between itself and the translations. The reciprocity is dialectic: new 'formats' of significance are initiated by distance and by contiguity. Some translations edge us away from the canvas, others bring us up close. This is so even where, perhaps especially where, the translation is only partly adequate. The failings of the translator (I will give common examples) localize, they project as on to a screen, the resistant vitalities, the opaque centres of specific genius in the original. Hegel and Heidegger posit that being must engage other being in order to achieve self-definition. This is true only in part of language which, at the phonetic and grammatical levels, can function inside its own limits of diacritical differentiation. But it is pragmatically true of all but the most rudimentary acts of form and expression. Existence in history, the claim to recognizable identity (style), are based on relations to other articulate constructs. Of such relations, translation is the most graphic. Nevertheless, there is unbalance. The translator has taken too much—he has padded, embroidered, 'read into'—or too little—he has skimped, elided, cut out awkward corners. There has been an outflow of energy from the source and an inflow into the receptor altering both and altering the harmonics of the whole system. Peguy puts the matter of inevitable damage definitively in his critique of Leconte de Lisle's translations of Sophocles: 'ce que la realite nous enseigne impitoyablement et sans aucune exception, c'est que toute operation de cet ordre, toute operation de deplacement, sans aucune exception, entraine impitoyablement et irrevocablement une deperdi-tion, une alteration, et que cette deperdition, cette alteration est 302 AFTER BABEL toujours considerable.'1 Genuine translation will, therefore, seek to equalize, though the mediating steps may be lengthy and oblique. Where it falls short of the original, the authentic translation makes the autonomous virtues of the original more precisely visible (Voss is weak at characteristic focal points in his Homer, but the lucid honesty of his momentary lack brings out the appropriate strengths of the Greek). Where it surpasses the original, the real translation infers that the source-text possesses potentialities, elemental reserves as yet unrealized by itself. This is Schleiermacher's notion of a hermeneutic which 'knows better than the author did' (Paul Celan translating Apollinaire's Salome*). The ideal, never accomplished, is one of total counterpart or re-petition—an asking again—which is not, however, a tautology. No such perfect 'double' exists. But the ideal makes explicit the demand for equity in the hermeneutic process. Only in this way, I think, can we assign substantive meaning to the key notion of 'fidelity'. Fidelity is not literalism or any technical device for rendering 'spirit'. The whole formulation, as we have found it over and over again in discussions of translation, is hopelessly vague. The translator, the exegetist, the reader is faithful to his text, makes his response responsible, only when he endeavours to restore the balance of forces, of integral presence, which his appro-priative comprehension has disrupted. Fidelity is ethical, but also, in the full sense, economic. By virtue of tact, and tact intensified is moral vision, the translator-interpreter creates a condition of significant exchange. The arrows of meaning, of cultural, psychological benefaction, move both ways. There is, ideally, exchange without loss. In this respect, translation can be pictured as a negation of entropy; order is preserved at both ends of the cycle, source and receptor. The general model here is that of Levi-Strauss's Anthropologic structural which regards social structures as attempts at dynamic equilibrium achieved through an exchange of words, women, and material goods. All capture calls for subsequent compensation; utterance solicits response, exogamy and endogamy are 1 Charles Peguy, 'Les Suppliants paralleles' in Oeuvres en prose 1898-190S (Paris, 19.59), 'i P- 890. This analysis of the art of poetic translation first appeared in December 1905. Cf. Simone Fraisse, Piguy et It monde antique (Paris, 1973), pp. M6-59- THE HERMENEUTIC MOTION 303 mechanisms of equalizing transfer. Within the class of semantic exchanges, translation is again the most graphic, the most radically equitable. A translator is accountable to the diachronic and synchronic mobility and conservation of the energies of meaning. A translation is, more than figuratively, an act of double-entry; both formally and morally the books must balance. This view of translation as a hermeneutic of trust (ilancement), of penetration, of embodiment, and of restitution, will allow us to overcome the sterile triadic model which has dominated the history and theory of the subject. The perennial distinction between literalism, paraphrase and free imitation, turns out to be wholly contingent. It has no precision or philosophic basis. It overlooks the key fact that a fourfold hermeneia, Aristotle's term for discourse which signifies because it interprets, is conceptually and practically inherent in even the rudiments of translation.