Translation : between what can be translated and what must be translated

Starting from a disconcerting interpretation of Jacques Derrida, our analysis aims at investigating and also tries to explain the blockage which appears in the English, French and Romanian translations (signed by Maurice de Gandillac, Antoine Berman, Laurent Lamy, Alexis Nouss, Harry Zohn, Steven Rendall, Martine Broda, Catrinel Pleșu etc.) of a well-known text of Walter Benjamin, Die Aufgabe des Ubersetzers, when translators transpose in their target languages the two quotations given by Benjamin: one of Mallarme, left untranslated in the source text, and another, signed by Pannwitz. The fact is that both quotations have something in common: a discoursive form which results from an unusual syntax, as if they were already, in a certain sense, „translations”. As if the translators feared—a feature of the translator’s psychology?—not to render their text sufficiently accessible, even when the source text is not intended to be accessible. Hence the painful dilemma of the intentional fallacy (not only of the text to be translated).


Text intentionality: an equally thorny issue…
As it was designed starting with the 70s, translation studies and translation practice itself had a lot to gain in terms of managing the translatable -untranslatable binomial (which comparatism or linguisticsinspired models failed to manage), mostly by making the modalities of perceiving connections between the target-and the source-text more flexible.In other words, by making the representations of the fidelity/infidelity relation more flexible.
A first step was made in the 70s, when Roland Barthes proclaimed, on behalf of structuralist literary theory, the "death of the author": thus, the very pertinence of the older notion of "intentionality" was questioned.In other words, he questioned the operational character of the text/author relation from a theoretical perspective, but also the author's "responsibility" for the sense and signification of the text 1 .What the American New Criticism coined as intentional fallacy-intentional utopia-was bound to prejudice literary theory, because the latter essentially reflected the conflict between literary explanation (analysis of author's intentions, of what he meant to say in his own text) and literary interpretation (description of text meanings regardless of author's intentions).Therefore, the exclusion of the author should have contributed to the autonomy of literary research in relation to history and psychology.Furthermore, due to the postulates of literarity and autotelism inspired by Russian formalism and by Roman Jakobson, it should have had "internal", "immanent" bases, while literarity should have resulted as an autonomous element, from the very particularities of organizing the material comprising the work.Deprived of "origin", the text exploded under the pressure of a polysemy viewed as constitutive principle, thus eventually revealing its own intertextual nature.Only a new authority-about to be designed as a theoretical concept-would have enabled it to gain coherence, thus reconstituting its unity scattered in a "mosaic of quotations": the reader.
The hermeneutics and aesthetics of reception would brilliantly confirm that text signification prevails over the author's intentionality because it accumulates-through its diachronic existence-new layers of meaning impossible to anticipate by the author and even less by its first readers.Hence, the sense of a text coagulates by the way it is "interrogated" by a historically conditioned subject, he himself also concerned (in his own context) with deciphering the "question" to which the text aimed to answer.The text no longer comprises only a sense-unchanging and interpretable as such from one reception to another-, but also a signification, due to which it articulates around a situation, thus being contextualized by the coordinates of each new reception.Whereas sense facilitates reception stability, signification explains the variations of text reception: sense is singular, while signification is open, plural, inexhaustible, and ultimately infinite.However, whereas the adequacy and depth of an interpretation depended on its capacity of revealing the coherence and complexity of a text, the coherence criterion itself-that had seemed able to eliminate the author's authority-could not be constructed without using what Antoine Compagnon had named the "presumption of intentionality" (Compagnon, 1998, p. 13-101), or at least the probability of an intention.Without such a premise, the coherence criterion was deprived of any conceptual pertinence.
It is no coincidence that translation practice also acquired an increasingly intense theoretical reflection in a period that had seemed ready to embrace the hegemonic and scientific claims of linguistics, by ascribing them the powers of an exhaustive knowledge method, recently dethroned by those who drew attention on the dangers of the linguistic "spell" (Pavel, 1988).After a period when the expectations of linguistics-looking for an experimental sample meant to illustrate its postulates-seemed to come true, the limits of an exclusively linguistic translation model became apparent.The Saussurean perspective itself-that viewed languages as closed systems, within which the sign acquired a signification (value) that coagulated exclusively due to its relations with the other language signs-theoretically confirmed the aporia of untranslatability.At the same time, however, it deepened the gap between translation practice and its theoretical impossibility: linguistic differences between languages could only reinforce the hypothesis of untranslatability, legitimated by Roman Jakobson by launching the formula according to which languages differ not in what they may convey, but in what they must convey2 .
Looking beyond the strictly linguistic perspective, (which would have been cautious in terms of accepting the possibility of translating the same term using a series of equivalences whose selection should depend on conformity to its context) reflection on translation shifted more and more toward a "communicational" perspective, which assimilated verbal significations to variables subjected to the influence of external factors.Taking into account that a text is not constructed only based on the action of linguistic mechanisms, it also includes an entire palette of extralinguistic elements, due to which linguistic competence is associated with textual competence.Thus, the notion of meaning hypothesis underscores the idea that utterance comprehension comes not only from the knowledge of a language, but also from the knowledge of a world.
The criterion of coherence and complexity cannot be justified in the absence of intentionality, for it would be a consequence of chance (Compagnon, 1998, p. 97): however, precisely this capacity of intentionality-in-coherence seems to raise a problem in translation and it often adds-on the reboundmore fuel to the untranslatable phantasm 3 .It is a fact that over time, during all eras, untranslatability reverberated either on the character (form) or on the meaning (content), and in the worst case on both.Ultimately, the temerarious Vocabulaire européen des philosophies (Cassin, 2004) elaborated by a team of researchers coordinated by Barbara Cassin bears the mark of this ambiguity, while it announces from the subtitle that it is also a dictionary of untranslatable terms...

…as the euphoria of unlimited translatability
Nonetheless, it is equally wrong to be misled in the euphoria of unlimited translatability, which seems equally risky as proclaiming untranslatability to be the inescapable end of translation.Whereas it is true that not everything can be translated, one should also admit untranslatability when it is entailed by the very intentionality of the text to translate and one should wonder, by reversing a splendid observation made by Jacques Derrida4 , if the text "wishes" to be translated as translatable.
Hence, I would like to get a better insight into the well-known essay of Walter Benjamin called Die Aufgabe des Übersetzers (Benjamin, 1972), and especially into the way in which the French, English, and Romanian translators proceeded for two quotations within this text.The essay was translated into Romanian; there are actually two versions signed by the same female translator, Catrinel Pleșu, but published by two different publishing houses, both under the tile Sarcina traducătorului.I will also analyse the French version called La tâche du traducteur (1971), signed by Maurice de Gandillac (Benjamin, 1971(Benjamin, , 2000a)), and the most recent version translated by Martine Broda (Benjamin, 1991).I will also refer to two other French versions, one of which is a partial translation by Antoine Berman done during a seminar of translation studies concerning the interpretation and commentary of Benjamin's text, held in the winter of 1984-1985 5 .Finally, I will also evoke two English versions entitled The Task of the Translator (Benjamin, 1968), signed by Harry Zohn, and The Translator's Task (Benjamin, 1997), signed by Steven Rendall.
Though not listed in the dictionary of untranslatable terms edited by Barbara Cassin, the German word Aufgabe seems to benefit from a broader semantic sphere than its Romanian, French, and English equivalents, considering that Jacques Derrida associates its meaning not only to a task, but also to a mission, a duty (in the two meanings of the Romanian words sarcină, datorie, but also misiune).This word also denotes a commitment by virtue of which the translator contracts a duty: his task is to liberate himself from it through a donation (Derrida, 1998, p. 211).Inspired by Derrida's comment (who also noted that the German verb aufgeben refers not only to a donation, but also to abandonment (Derrida, 1998, p. 212), and taking advantage from the resources of the word family of don, a more recent retranslation of Benjamin's text, made by Alexis Nouss and Laurent Lamy, dares to propose the title L'abandon du traducteur (Lamy & Nouss, 1997).This translation also actualizes throughout the entire text the polysemy of Aufgabe, by putting the stake on the tâche -abandon (task -abandonment, renouncement) couple and by lexicalizing the to-and-fro between translatable and untranslatable, between task and its failure, disguised in the semantic core of Aufgabe, but possible to decipher in the very intentionality of the source-text, in its coherence, which functions on both semantic trajectories.
My analysis will focus on two quotations present in Benjamin's text, namely on their transposition into Romanian, English, and French (for the second quotation).Toward the end of his essay, Walter Benjamin cites a sentence from Mallarmé in French without translating it, after invoking at the beginning of the paragraph, in Latin, ingenium (see Appendix A).The Latin term is translated by Harry Zohn into English (philosophical genius) and by Gandillac and Martine Broda by génie: this reluctance of preserving the Latin term illustrates the distance separating us from the times when Latin was the key language of science and philosophy.On the other hand, the option of the two French translators opens wide the gates of what is called ethnocentric, naturalizing translation, because it pushes the reader toward a Franco-French concept, that "ideological arrangement" 6 still deeply rooted in French mentalities, thus subtly getting the reader further away from the meaning to remember: the one of an innate feature of philosophical spirit.
In exchange, the Romanian version suggests-by reversing the possession relations-that ingenium belongs to philosophy ("ingenium al filosofiei"): it is one thing to say that philosophisches Ingenium, that a part of our spirit is philosophical, and a very different thing to state that philosophy possesses one...
Walter Benjamin fails to indicate the exact source of the quotation.Alexis Nouss provides it in the notes accompanying his translation: the excerpt is taken from "Crise de vers", part of the volume Igitur.Divagations.Un coup de dés (Lamy & Nouss, 1997, p. 23; the quotation is taken from Mallarmé, 1976, p. 244).Antoine Berman, who had found it before Nouss, declared during the seminar that he was astonished of the "altered, censored" (Berman, 2008, p. 157, and on the next page he reiterates the accusation: "Benjamin censored it") character of the quotation!The accusation was definitely excessive, but the fact remains that Benjamin "cuts" the quotation where Mallarmé, the poet, associates-not surprisingly-"pure language" with poetical language, with verse 7 , the only one endowed with compensating virtues, which "makes up for what languages lack" ("rémunère le défaut des langues", Berman, 2008, p. 158): but that was not Benjamin's pure language.
Benjamin's decision not to translate the sentence did not go unnoticed: it stirred wonder, curiosity, various interpretations, and a fiery desire to understand expressed by all those who studied Benjamin's text.Jacques Derrida, Antoine Berman, Alexis Nouss-they all deciphered it in a way or another as a brilliant illustration of the untranslatable, which they each explained from their own perspective...
In principle, the decision not to translate the French quotation loses its entire symbolic value in one language: French.Nonetheless, the Romanian version also makes Benjamin's quotation obsolete, by translating Mallarmé's quotation in the first edition, published in 2000.In the second version published after two years-which did not actually revise anything-the original quotation was provided, but with a Romanian translation in the only footnote (p.45) of this version.Furthermore, because it does not mention whether the note belongs to the translator (editor?), it suggests implicitly that it reflects a similar gesture to that of the original author!Regardless, through this decision, the translator alters the deepest layers of meaning in Benjamin's text; moreover, she manages to make all paratexts attempting to decipher its symbolism-from Jacques Derrida to Antoine Berman, including Alexis Nouss and Laurent Lamyutterly incomprehensible!All the more as Benjamin's text is a preface, actually the preface to his own translation of Tableaux parisiens by Baudelaire.By choosing not to translate Mallarmé's fragment, thus proclaiming its untranslatability (which, in Derrida's words, "he has left shining in his text like the medallion of a proper name" 8 -the proper name is untranslatable!),Benjamin implicitly posits that its meaning cannot be transposed into another tongue without "damage" (Derrida, 1998, p. 213).He goes on by saying that it is impossible to translate it and that "in Mallarmé's text, the effect of being proper and thus 6 Starting from the 16 th century and culminating with the discourse on the universality of Rivarol's French language (1782), the thesis of French language genius consolidated over time and it turned into a persistent cliché, based on two support points: clarity and the principle of "natural" order, leading toward the postulate of its universality.Cf.Meschonnic (1997, p. 227-240).
untranslatable is tied less to any name or to any truth of adequation than to the unique occurrence of a performative force" 9 .Through this gesture, Benjamin destabilizes the concept of translation itself and he confirms to some extent the interpretation provided by the translation of Nouss/Lamy, who decoded Aufgabe more like an abandonment than a task (or maybe even both).
A question remains: even Antoine Berman wondered why Benjamin did not translate Mallarmé's fragment, all the more as-he said-it is not linguistically or stylistically untranslatable, though it develops a syntactical structure that is "if not absent, at least strangely rarefied" 10 .According to him, the real reason for not translating it must be sought elsewhere, in its very sense, which any translation or any decision regarding translation would have contracted it "in an ironic register", because the sentence "concerns the imperfection of tongues, that is their multiplicity, thus what justifies both the necessity and the impossibility of translation" 11 .The Romanian translator's decision destroys the very coherence of Benjamin's essay, because translating this sentence (designed as a deviation from the norm) is the same as translating something that has already been translated to a certain extent.Actually, Derrida among others understood it very well: one does not translate a translation and anyway not a text that posits the impossibility of translation.Maybe Benjamin chose not to translate it also because he already perceived it as a translation (Derrida, 1998, p. 218)...
As the other translators, Steven Rendall wonders why Benjamin provided a quotation from a great French poet without translating it and by inserting it within a preface preceding his own translation of another great French poet.Among other possible explanations, Rendall suggests that there was no reason for translating it, because it already involved a decontextualisation similar to the translation process (Rendall, 1997, p. 179).Hence, translating the quotation would have contradicted Benjamin's intention and project, and no translator of the English versions I consulted surrounded to this "urge".I did happen to find, though, a word-for-word English translation of the quotation, which transposes perfectly the altered syntax, but it is part of a philosophical work by Samuel Weber 12 called Benjamin's -abilities: Languages, imperfect insofar as many, lacking the highest: thinking being writing without accessories, neither whispering but silent still the immortal word, the diversity, on earth, of idioms prevents no one from offering the words which, if not, would find themselves, in a single stroke, itself materially the truth (Weber, 2008, p. 75-76).
I could have at least ascribed a good intention to the Romanian version, like the "pedagogical" decision of helping the Romanian receptors, had the translator opted for a literal transposition of Mallarmé's fragment 13 .However, far from literal, it is full of intentionality that contradicts blatantly not only what the quotation wants to say, but also the intention of the source-text, which concerns the translator's task.What it does not contradict it trivializes; it cancels the laconism of the original text, it rationalizes syntax by introducing verbs where the source-text eludes them, and it practices coordination: "imperfecțiunea limbilor constă în pluralitatea lor și în lipsa celei supreme", which is redundant.Finally, it diminishes considerably its negative register: manque (turned into the noun lipsa); sans accessoires, ni chuchotement (fără accesorii și fără șoapte-a coordination that makes the rhythm monotonous, thus eliminating the dramatic note); empêche personne (împiedică pe toți); sinon (altfel).Therefore, the text is entirely reorganized following a linear, classic view of syntax and word sequence.Furthermore, the phrase tacite encore l'immortelle parole is mistranslated by cuvîntul nemuritor ramîne încă subînțeles, thus intercalating a verb again.As for the phrase diversité, sur terre, des idiomes qui empêche personne de proférer les mots qui, sinon se trouveraient, par une frappe unique, elle-même matériellement la vérité, it is transposed into Romanian by diversitatea idiomurilor pe pămînt împiedică pe toți să profereze cuvinte care, altfel, la o atingere unică, s-ar materializa ca adevăr, which is meaningless.Actually, the French text states that the diversity of idioms does not prevent anybody from offering the words which, such diversity should not exist and only a supreme tongue should exist, would materially be Truth itself.Whereas Mallarmé's text is untranslatable, says Samuel Weber, it does not mean it cannot be transcoded, but by seeking to render its meaning faithfully one misses the essential: namely the way in which the meaning of the text is meant.It comprises a threefold deviation from grammar rules: the sequence of words is reversed, so is the syntax (seemingly, the subject does not accord with the form of the verb), la suprême is the real subject, les langues imparfaites, the object...The Romanian version fails to highlight any of the aforementioned aspects and it seems to have been caused by some kind of "recklessness".Its only merit is that it demands a new version, which will be able to pinpoint the untranslatable where it is-so to speak-"premeditated", within the project, not within the object...

Re-cognizing the poetics of the text to translate
Within the so-called functionalist theories of translation and within the debate on translation critique and evaluation, there has been increased focus on the type of text to translate, on the intentionality it expresses.Intentionality has been considered the premise of translation techniques adequacy, within an equation that completely lacks the threat of untranslatable, managed as a variable that can be isolated and eliminated in the translation process.These typologies focusing on the criterion of text function are implicitly dominated by the belief that all translational processes involve inevitable losses: the underlying ideology of translation techniques seems to be obsessed with "managing" the (theoretical) impossibility of translation.Thus, it determines the elaboration of strategies aiming to limit losses, considered secondary (hence, acceptable), provided that the translation act focuses on producing an equivalent target-text, able to recompose the predominant function of the source-text.Hence, the shifting of translation toward the dominant function of the text to translate becomes the basic translational principle: translation adequacy is no longer justified in relation to the source-text, but with its skopos, which dictates to a translator the strategies to follow14 .The faithfulness-translation binomial, just like the question of untranslatable, are thus de-dramatized through an operation of shifting: faithfulness no longer associates directly (through an inevitable relationship of filiation) the target-text to the source-text, while "treason" is justified by translation techniques that no longer require direct conformity.In exchange, the receptor's position is consolidated insofar as the function of a text is also determined considering its effect on it 15 .
The act of translating a philosophical text seems to want to mask a contradiction that conceals some kind of "desecration" of the philosophical intentionality of the text per se, namely to convey universal contents.He himself a translator of Jürgen Habermas and Theodor W. Adorno, Jean-René Ladmiral considers the translation of philosophy a "scandal", an almost indecent gesture which alters Reason, considering that the very possibility of translation involves operations that dissociate "the conceptual signifiés of philosophy (of a philosophy?) from the signifiants of the language of departure or source language […] to facilitate their subsequent 'reincarnation' into other foreign signifies of the language of arrival or target language" (Ladmiral, 1998, p. 990;cf. also Ladmiral, 1989, p. 6).Such scandal would even be twofold: on one hand, the discourse that aims to explain a universal Reason cannot avoid the "historical and cultural particularity of national traditions"; on the other, it is forced to be embodied in "linguistic accidentsjustly called idiomatic-of natural languages" (Ladmiral, 1998, p. 990).Hence, the translation of this type of text highlights -on the level of scandal and derision -the inner tension of the universalizing project of philosophical rationality, forced to accept being transposed into the signifiers of a particular language.The special profile of this "scandal" delimits the specificity of philosophical texts and confirms the tension caused by uncertainties surrounding the very possibility of translating them.Such tension is all the more significant as, in this type of text, "the rational transparency of conceptual signifiés as purpose of the philosophical discourse […] will be shadowed by the contingent 'impurities' of the linguistic signifiant" (Ladmiral, 1998, p. 983).Therefore, philosophical text can be defined as a signified-oriented text, which is underscored by the valorisation of the language's capacity of becoming a meta-language.In other words, the philosophical discourse referent is its own signified -hence the "scandal" of translating this category of texts (Ladmiral, 1998, p. 988).Philosophical texts can be included in the category of literary texts in the broad sense; however, through the "technicity of the philosophical 'jargon'" (Ladmiral, 1998, p. 988) they use, they are also included in the category of technical texts, but from which they are distinguished because the subject speaking is omnipresent in the text.
Aiming to short-circuit the untranslatable, many consider philosophical texts a stratified structure, which would involve different translation techniques: the discourse of such a text appears, on one hand, as an architectonics made of a level comprising specialized language and, on the other, as a narrative structure, as a story related to the literary type of text.Hence, the translation of philosophical texts would represent a very special case of combination between literal translation techniques-for transposing terminologies-and so-called "free", "idiomatic" translations, for the translation of the textual device.Hence, the translator would have the task of dividing the language used in the text, namely the aspects related to the common lexicon of the source-language to the author's word (that system of indicators making up the mark of a particular subjectivity) (Ladmiral, 1994, p. 223).However, such a view of philosophical text translation entails a set of risks, which are far from insignificant: its underlying ideology privileges meaning to the detriment of text poetics, of the modality of its constructs, thus generating a series of target-text distortions, of explanatory deformations, annotations, peri-and paraphrases, which may destroy the original configuration.
Hence the imperative of translating not only text rhetoric, but also, in the words of Henri Meschonnic, text poetics (he says translators are prone to taking one for the other): "The task of Humboldt's translator is to recognize this poetics.To recognize it like poetics.Not like rhetoric.Thought labour gives birth to a poetics if it transforms language values into discourse values, specific only to its discourse.And if language categories remain language categories, rhetoric games emerge.This banality -that one cannot separate a thought from its writing.The translator has the task of not taking poetics for rhetoric, at all levels distinguished by traditional linguistics" (cf.Meschonnic, 1999, p. 350).Upon examining the French translations of the texts written by Wilhelm von Humboldt, Meschonnic concludes that the differential translation of philosophical text seen as the sealed superimposition of two layers (on one hand terminology, on the other its formal textualization) entails serious dangers.Whereas terminology was translated by observing correspondences, "operators, logical rhythm [...] are treated like an element where variation is not important.The text is respected in its rigour, but such rigour has only a substantial, conceptual nature.Hence, a certain idea on philosophical text emerges: besides technicity-that involves the confusion between concept and word-all the rest is literature.Meaning rhetoric.Such a view of language is related only to the sign.To the primacy of the signified identified with the sign" (Meschonnic, 1999, p. 382).
This method conceals a double (two-faced?) ideology on the act of translation.Henri Meschonnic seems to suggest it by positing that, in the end, the translation of specialized lexicon through correspondences expresses the confusion between concept and word, insofar as the guarantee of translation success is anchored in the belief that rendering the sense depends on the possibility of maintaining them in all occurrences of the source-text.Jean-René Ladmiral-whose well-known translational "theorems" (Ladmiral, 1994) militate for the right of practicing "annexionism" in translation-sees this procedure as the extension of "substantialist metaphysics" 16 on language, which tends to sacralize the source-language, to over-evaluate its expressive valences, which increases the danger of falling into the trap of translatableuntranslatable aporia.Hence, dramatizing the gap between what is and what is not translatable will undermine to a greater extent the endeavour of philosophical text translation that, through its very nature, expresses within a tensed register "the coincidence between individual singularity and universality" 17 .

How does one translate a "translation"?
The second quotation provided by Walter Benjamin is a fragment from Die Krisis der europäischen Kultur by Rudolf Pannwitz (Pannwitz, 1917).Again, the syntax is altered; furthermore, punctuation marks are absent and there are no capital letters for common nouns or first words of sentences.This was the style adopted by Jakob Grimm or Stefan George (Berman, 2008, p. 159).The quotation comprises two sentences that end in a full stop, but none of the sentences begins with a capital letter.The two quotations, which share the same type of formal structure, have quite many points in common: by ignoring syntax, both of them rely on words, which are ripped out of the web of syntactical structure.They are displayed as names whose primary relation is not to other words in the language but to things 18 .However, for Benjamin, too, translation is a word-related operation.One must look for the reason why the author chose these two quotations: "[...] Pannwitz's German, like Mallarmé's French, subverts the linguistic structure of the language, and particularly its syntax-and might thus be seen as having already achieved the kind of liberation of the word that translation is supposed to provide" (Rendall, 1997, p. 180).
Restoring-through translation-the syntactical order of the two quotations (one if which is not to be translated) destroys not only their objective, but also their raison d'être in Benjamin's text.As I have mentioned before, in a certain way, the two quotations are already (intralinguistic) translations, because they disrupt syntax and liberate the word by translational principles similar to those predicted by Benjamin.That is why they are untranslatable, and Benjamin states it clearly: "Übersetzungen dagegen erweisen sich unübersetzbar nicht wegen der Schwere, sondern wegen der allzu großen Flüchtigkeit, mit welcher der Sinn an ihnen haftet" 19 or "Übersetzung ist eine Form.Sie als solche zu erfassen, gilt es zurückzugehen auf das Original.Denn in ihm liegt deren Gesetz als in dessen Übersetzbarkeit beschlossen" 20 . 16Jean-René Ladmiral, Traduire: théorèmes pour la traduction, Éditions Payot, Paris, 1979, apud Brownlie (2002, p. 306). 17Ibid.Cf.Jean-René Ladmiral, La traduction philosophique, in "Revue de Phonétique Appliquée", apud Brownlie (2002).  1Cf.Rendall (1997, p. 179).Rendall wonders why Benjamin chose this quotation from Pannwitz, though it would have made more sense to invoke Schleiermacher or Humboldt.
Zohn's version: "Translations, in contrast, prove to be untranslatable not because of any inherent difficulty but because of the looseness with which meaning attaches to them" (Benjamin, 1968, p. 80).Rendall's version: "Translations, on the contrary, prove to be untranslatable not because meaning weighs on them heavily, but rather because it attaches to them all too fleetingly" (Benjamin, 1997, p. 164). 20Benjamin (1972, p. 9).Nouss/Lamy version: "La traduction est une forme.Pour la saisir comme telle il faut revenir à l'original.En effet, c'est en lui que repose sa loi, telle qu'elle est contenue dans sa traductibilité" (Lamy & Nouss, 1997, p. 14).The version of Catrinel Pleșu: "Traducerea este o formă.Pentru a o înțelege ca formă, trebuie să ne întoarcem la original, căci în el este cuprinsă legea care guvernează traducerea, și anume traductibilitatea sa" (Benjamin, 2002, p. 39).Unlike Nouss/Lamy, who maintain strictly the jerky structure of this three-sentence concatenation with an argumentative crescendo, the Romanian version chooses to connect the last two sentences of the source-text, thus reducing their impact through excessive explicitation and missing their very rhythm and poetics.Rendall's version: "Translation is a mode.In order to grasp it as such, we have to go back to the original.For in it lies translation's law, decreed as the original's translatability" (Benjamin, 1997, p. 152).Zohn's version: "Translation is a mode.To comprehend it as mode one must go back to the original, for that contains the law governing the translation: its translatability" (Benjamin, 1968, p. 69).
Of the six translations consulted, only one dares to keep the form of Pannwitz's text (see Appendix B).The others merely try to suggest its formal particularities to a lesser or greater extent, but they do end up introducing commas and full stops here and there.Antoine Berman rephrases the sentences, which now begin with capitals, but he does not introduce punctuation marks within them; in exchange, he translates, according to the norm, wort bild ton by mot, image et son (an et precedes the last term of the enumeration).Nouss and Lamy massively alter the sentence structure, by multiplying full stops and by introducing two semicolons; in exchange, they eliminate the ellipsis of the source-text.Finally, Gandillac's translation preserves both the punctuation and the sentence structure of the source-text.The versions of Zohn and Catrinel Pleșu follow the norms to the letter and they alter all deviations of the source-text (orthographic, syntactic, sentence structure, capitals at the beginning of sentences).Only Rendall dares to transpose integrally Pannwitz's text, namely its form, but he chooses-rather surprisingly-to suppress to first term of the wort bild ton sequence.However, Rendall too fails to find a solution of "transposing" the lower case letters for common nouns used in the source-text.Nobody seems to have thought to the (plausible, though) possibility of keeping the bizarre character of the German nouns written with lower case letters by translating into English, French, Romanian, those nouns using capital letters.Naturally, one may object that making such a decision for languages where common nouns begin with small case letters would ascribe them a symbolic or allegoric value, by personifying them.However, should such a technique be applied to all nouns and to the beginning of sentences, the reader may realize that the purpose is not to provide a symbolical connotation to nouns, but that something else is at stake.Of course, writing common nouns in German using small case letters is a decision immediately perceived as disrupting orthography, while writing them with capital letters in French, English, or Romanian could never be perceived as a deviation from the norm.
Some translators feel the need to justify themselves, which means that the decision of normalizing translation makes them uncomfortable: Berman (2008, p. 179) partially retranslates Gandillac's version, by trying to keep the oral character of the source-text.He humbly states that it should not be "combed", "but left partially dishevelled" (p.178).Nouss and Lamy, who dared to translate the title by "L'abandon du traducteur", seem timorous all of a sudden and they admit in a footnote (without fully justifying this decision) that they were not "fully" faithful to the source-text, "since we restore a punctuation whose omission would lead, in our opinion, to confusion".
However, the form-reproducing the oral register-chosen by Pannwitz is related to what he is trying to convey, and Zohn or Catrinel Pleșu failed to pinpoint and thus to translate it: by both form and content, Pannwitz's text refers to the oral essence of language, where "word, image, and sound meet" (Berman, 2008, p. 179).The mention of the word "dialect" should have definitely drawn their attention.According to Antoine Berman, the quotation is an authentic "historical story" (p.178) on translation, with a twofold perspective: on one hand, through the "collision", the commotion of languages for which it testifieswithout trying to institute it; on the other, through the "movement toward the 'ultimate elements' of language itself, where word, image, and sound meet -as from dialect to dialect" (Berman, 2008, p. 179)  21 . 21Berman does not miss the chance of mocking Meschonnic, saying that he is surprised the latter quoted the text "by ascribing it to Benjamin" and he condescendingly asserts "of course, whoever reads Pannwitz?".To twist the knife even deeper, he adds: "Here are some more historical statements on translation".However, Berman fails to pinpoint where exactly Meschonnic made such confusion.Upon reading the mean observation, I first thought it was a "revealing lapsus" and that maybe Meschonnic's mistake proved, on the other hand, to what degree Benjamin's quotation and text merged, to what extent one supported the other, and that all translators had the duty of preserving this fusion...I still had a little doubt, though, and I studied Meschonnic's works until I finally found what I was looking for.In Pour la poétique, Meschonnic says: "To paraphrase a fragment quoted by Benjamin, I would say that, instead of frenchizing Sanskrit, Greek, English, we should sanskritize, hellenize, and anglicize French".Seven lines below, talking about the "dialectics of translation process contradictions", he posits that it "leads to such formulation by Benjamin" and he quotes the second part of Pannwitz's quotation: as in the case of the quote from Mallarmé, Berman was only half right and maybe it would have been more important for him to notice that Meschonnic did not ascribe to Benjamin a segment of Pannwitz's text, but that he omitted to specify that he was actually quoting what Gandillac-the translator-said that Benjamin said... Cf.Meschonnic (1973, p. 143).
Pannwitz's text refers precisely to the "oral" essence of language, and dialect is not invoked incidentally: "For all tongues are the Tongue insofar as they are dialects.There is no language-tongues (from gender to species), but tongue-dialects.The space where translation unfolds as an intention to be a letter of the tongue is the dialect.French and Chinese are dialects of the same tongue" (Berman, 2008, p. 179).
According to this reading, work-which replaces word in the English and the Romanian versionleads to a terrible mistranslation: assimilating the word to written word when the issue concerns dialects, transposing it by metonymy as work abolishes and completely annuls the value and sense of the wordimage-sound triad.The Romanian version and the English one (not to mention the one that simply eliminates wort) not only make both the quotation and the text by Benjamin unintelligible, but they also completely alter the meaning.The sequence work, image, and tone // opera, imaginea și tonul22 deviate the reading by making it scriptural, writing-oriented, thus alienating it from the dialect essence itself, namely from the orality pointed out by Pannwitz.Finally, word, image, tone are not convergent, like in the versions of Pleșu and Zohn: their hypothesis of meaning deviates massively because of the metonymical and inevitably rhetorizing assimilation of word by work, and it institutes another coherence23 , which makes dialectal orality obsolete.A translator should wish to get to their confluence, where word, image, tonality meet, not where they converge and all the less where they are convergent.This is the only way to penetrate the depths of one's own language, to unfold it, not by means of, but due to the foreign language.
Hence, the connections established between Pannwitz's quotation and Benjamin's essay are to remain forever invisible for the readers of the Zohn, Rendall, and Pleșu versions.By highlighting the dialectal, the quotation open the way toward the orality of that Reine Sprache [pure language], identified with the dialect or, more precisely, with the "dialectal essence of language" (Berman, 2008, p. 181): "From poetry to theatre, children books, novels, psychoanalysis or religious, juridical texts... the translation of works is related to orality", Berman warns (p.180).Furthermore, he insists that writing includes orality: "Writing is best listened to when orality is embedded in it.Language itself is oral language".Translation is the (only) act able to "uncover the orality of the written original" (p.180).
At the end of this endeavour, a question remains to which I have yet to find an answer: why did translators-who understood the stake of Pannwitz's quotation, who interpreted and commented it in such a brilliant manner (Berman, Nouss, Lamy and Rendall)-not dare to translate this quotation in both its letter and spirit?By translating this quotation as they did, they reached the highest point of treason, because they betrayed both the letter and the meaning.By doing so, they betrayed the very coherence of the text where the quotation is inserted, which refers to the translator's task24 .

Benjamin (1968)
For there is a philosophical genius that is characterized by a yearning for that language which manifests itself in translations."Les langues imparfaites en cela que plusieurs, manque la suprême: penser étant écrire sans accessoires, ni chuchotement mais tacite encore l'immortelle parole, la diversité, la terre, des idiomes empêche personne de proférer les mots qui, sinon se trouveraient, par une frappe unique, elle-même matériellement la vérité".If what Mallarmé evokes here is fully fathomable to a philosopher, translation, with its rudiments of such a lunaguage, is midway between poetry and doctrine.Its products are less sharply defined, but it leaves no less of a mark on history.

Benjamin (1968)
Pannwitz writes: "Our translations, even the best ones, proceed from a wrong premise.They want to turn Hindi, Greek, English into German instead of turning German into Hindi, Greek, English.Our translaters have a far greater reverence for the usage of their own language than for the spirit of the foreign works… The basic error of the translator is that he preserves the state in which his own language happens to be instead of allowing his language to be powerfully affected by the foreign tongue.Particularly when translating from a language very remote from his own he must go back to the primal elements of language itself and penetrate to the point where work, image, and tone converge.He must expand and deepen his language by means of the foreign language.It is not generally realized to what extend this is possible to what extend any language can be transformed, how language differs from dialect; however, this last is true only if one takes language seriously enough, not if one takes it lightly".

Rendall (1997)
He writes: "our translations even the best start out from a false principle they want to germanize Indic Greek English instead of indicizing, graecizing, anglicizing German.they are far more awed by their own linguistic habits than by the spirit of the foreign work [...] the fundamental error of the translator is that he holds fast to the state in which his own language happens to be rather than allowing it to be put powerfully in movement by the foreign language.he must in particular when he is translating out of a language very distant from his own penetrate back to the ultimate elements of the language at that very point where image tone meld into one he must broaden and deepen his own language through the foreign one we have no notion how far this is possible to what degree each language can transform itself one language differentiates itself from another almost as one dialect from another but this happens not when they are considered all too lightly but only when they are considered with sufficient gravity".