Repository path: workshop/translations/berman-tendances/R04-v1/translation.md · rendered 2026-09-09

Antoine Berman, the twelve tendances déformantes (1985) — the defining sentences

Contamination: high. This is the most-cited negative typology in translation studies and has a standard published English rendering (S. Heyvaert's "Translation and the Trials of the Foreign", reprinted in Venuti's Translation Studies Reader). The lead has certainly met the received English names — "rationalization", "clarification", "expansion", "ennoblement", "qualitative and quantitative impoverishment" — many times over, and those names shaped what came to mind here. Heyvaert's version was not consulted, before or during (R04 §1), but the honest declaration is high, not suspected: for several of the twelve the received English arrived before the French did, and the log below is in part a record of prising them apart.

The twelve

Berman's own French in the left column, brief and attributed; the lead's English in the right. Ellipses mark omitted material within a quotation; nothing is silently joined.

The frame

« Cette analytique part du repérage d'un certain nombre de tendances déformantes. Celles-ci forment un tout systématique. J'en évoquerai ici douze. »

This analytic begins from the identification of a certain number of deforming tendencies. They form a systematic whole. I shall raise twelve of them here.

« Ce que l'analytique a en vue, par contre, ce sont des universaux de la déformation inhérents au traduire comme tel. »

What the analytic has in view, by contrast, are universals of deformation inherent in translating as such.

« Mais cette coïncidence est momentanée: au XXe siècle, nous ne sommes plus soumis aux normes classiques, et néanmoins les universaux de la déformation subsistent »

But this coincidence is momentary: in the twentieth century we are no longer subject to the classical norms, and the universals of deformation persist all the same.

1. La rationalisation — rationalization

« Elle porte au premier chef sur les structures syntaxiques de l'original, par exemple sur cet élément sensible et modifiable du texte en prose qu'est la ponctuation. La rationalisation re-compose les phrases et séquences de phrases de manière à les arranger selon une certaine idée de l'ordre d'un discours. »

It bears first and foremost on the syntactic structures of the original — for example on that delicate and alterable element of a prose text, its punctuation. Rationalization re-composes the sentences and the sequences of sentences so as to arrange them according to a certain idea of the order of a discourse.

« Résumons : la rationalisation déforme l'original en inversant sa tendance de base. »

To sum up: rationalization deforms the original by reversing its basic tendency.

2. La clarification — clarification

« C'est un corollaire de la rationalisation, concernant en particulier le niveau de « clarté » sensible des mots ou leur sens. Là où l'original se meut sans problème dans l'indéfini, notre langue littéraire, par exemple, tend à imposer du défini. »

It is a corollary of rationalization, bearing in particular on the level of palpable "clarity" of words, or on their meaning. Where the original moves in the indefinite without any trouble, our literary language, for instance, tends to impose definiteness.

« Le passage de la polysémie à la monosémie est un mode de clarification. »

The passage from polysemy to monosemy is one mode of clarification.

3. L'allongement — lengthening

« Tendanciellement, toute traduction est plus longue que l'originale. »

As a tendency, every translation is longer than the original.

« Je veux dire par là que l'ajout n'ajoute rien, qu'il ne fait qu'accroître la masse brute du texte, sans du tout augmenter sa parlance ou signifiance. »

By which I mean that the addition adds nothing — that it does no more than swell the raw mass of the text, without in the least increasing its parlance or its signifiance.

4. L'ennoblissement et la vulgarisation — ennoblement and vulgarization

« La rhétorisation consiste à produire des phrases « élégantes », en utilisant pour ainsi dire le texte de départ comme matière première. L'ennoblissement n'est donc qu'une ré-écriture, un « exercice de style » à partir de et aux dépens de l'original. »

Rhetoricization consists in producing "elegant" sentences, using the source text, so to speak, as raw material. Ennoblement is therefore nothing but a re-writing, an "exercise in style" made out of the original and at the original's expense.

« L'envers logique de l'ennoblissement – ou complément –, c'est, pour les passages jugés trop « populaires », le recours aveugle à un pseudo-argot qui vulgarise l'original »

The logical obverse of ennoblement — or its complement — is, for passages judged too "popular", the blind resort to a pseudo-slang that vulgarizes the original.

5. L'appauvrissement qualitatif — qualitative impoverishment

« Il renvoie au remplacement de termes, expressions, tournures, etc. de l'original par des termes, expressions, tournures n'ayant ni leur richesse sonore, ni, corrélativement, leur richesse signifiante ou « iconique ». »

It refers to the replacement of the original's terms, expressions and turns of phrase by terms, expressions and turns of phrase possessing neither their sonorous richness nor — correspondingly — their signifying or "iconic" richness.

« Quand on traduit le péruvien chuchumeca par « pute », on a certes rendu le sens, mais nullement la vérité phonético-signifiante de ce mot. »

When one translates the Peruvian chuchumeca as "whore", one has certainly rendered the meaning, but in no way the phonetic-signifying truth of the word.

6. L'appauvrissement quantitatif — quantitative impoverishment

« Il renvoie à une déperdition lexicale. »

It refers to a lexical bleeding-away.

« La traduction qui ne respecte pas cette multiplicité rend le « visage » de l'œuvre méconnaissable. »

A translation that does not respect this multiplicity makes the "face" of the work unrecognizable.

7. La destruction des rythmes — destruction of rhythms

« Le roman n'est pas moins rythme que la poésie. Il est même multiplicité de rythmes. »

The novel is no less rhythm than poetry is. It is even a multiplicity of rhythms.

« la traduction déformante peut – par exemple en remaniant arbitrairement la ponctuation – affecter considérablement la rythmique. »

Deforming translation can — by arbitrarily reworking the punctuation, for example — considerably affect the rhythmics.

8. La destruction des réseaux signifiants sous-jacents — destruction of underlying networks of signifiers

« L'œuvre littéraire comporte une dimension cachée, un texte « sous-jacent », où certains signifiants se répondent et s'enchaînent, formant toutes sortes de réseaux sous la « surface » du texte même – du texte manifeste, donné à lire. C'est donc le sous-texte, qui porte le réseau des mots-obsessions de l'œuvre. »

The literary work carries a hidden dimension, an "underlying" text, in which certain signifiers answer one another and link up, forming all sorts of networks beneath the "surface" of the text itself — of the manifest text, the one given to be read. It is the sub-text, then, that carries the network of the work's obsession-words.

« Si l'on ne transmet pas de tels réseaux, on détruit l'un des tissus signifiants du texte. »

If one does not carry such networks across, one destroys one of the signifying tissues of the text.

9. La destruction des systématismes — destruction of systematisms

« Le systématisme du texte dépasse le niveau des signifiants, des métaphores, etc. : il s'étend au type de phrases, de constructions de phrase employées. L'emploi des temps peut être un tel systématisme »

The systematism of a text goes beyond the level of signifiers, metaphors and so on: it extends to the kind of sentences, of sentence-constructions, that are used. The handling of tenses can be such a systematism.

« alors que le texte de la traduction est plus « homogène » que l'original (plus « stylé » au sens banal), il est également plus incohérent et, d'une certaine façon, plus hétérogène, plus inconsistant. »

Whereas the text of the translation is more "homogeneous" than the original (more "styled", in the trivial sense), it is at the same time more incoherent and, in a certain way, more heterogeneous, more inconsistent.

10. La destruction des réseaux vernaculaires ou leur exotisation — destruction of vernacular networks, or their exoticization

« Le domaine est essentiel, parce que toute grande prose est enracinée dans le langage vernaculaire. »

The domain is essential, because all great prose is rooted in vernacular language.

« L'effacement des vernaculaires est donc une atteinte très grave à la textualité des œuvres en prose. »

The effacement of the vernaculars is therefore a very serious injury to the textuality of works in prose.

« Une telle exotisation, qui rend l'étranger du dehors par celui du dedans, n'aboutit qu'à ridiculiser l'original. »

Such exoticization, which renders the foreign from without by the foreign from within, succeeds only in making the original ridiculous.

11. La destruction des locutions et idiotismes — destruction of set phrases and idioms

« il est évident que, même si le sens est identique, remplacer un idiotisme par son « équivalent » est un ethno-centrisme »

it is evident that, even where the meaning is identical, to replace an idiom by its "equivalent" is an ethnocentrism

« Traduire n'est pas chercher des équivalences. »

To translate is not to hunt for equivalences.

12. L'effacement des superpositions des langues — effacement of the superimposition of languages

« Dans une œuvre romanesque, les superpositions de langues ont trait au rapport des dialectes à une langue commune, une koïné, ou à la coexistence, au sein d'un texte, de deux ou plusieurs koïné. »

In a work of fiction, the superimpositions of languages have to do with the relation of dialects to a common language, a koine, or with the coexistence, within one text, of two or more koines.

« On a là la quintessence des problèmes que pose la traduction romanesque – le problème exigeant une réflexion maximale du traducteur. »

Here we have the quintessence of the problems posed by the translation of fiction — the problem that demands the maximum reflection from the translator.

What each tendency damages, in this project's vocabulary

Written at translation time, before the Tier D design existed, and frozen with the log. Sense ids from wiki/goodness-senses.md. This mapping is the lead's, not Berman's — he has no such list — and is internal-judgment-only.

None of the twelve targets accuracy. Berman's quarrel is not with the translator who gets it wrong; it is with the translator who writes well. Nine of the twelve are things a competent, careful, fluent translator does on purpose. That is a property of the catalogue, not an omission from it — and it means a Berman-only operator set cannot test the one sense a jury is most likely to detect.

Translator's log

Frozen 2026-07-25, before the Tier D experiment design was written. Decisions and alternatives, not quality claims.

The two coinages that have no English: parlance and signifiance. Berman uses them as a pair, and English resists both. Parlance is a false friend of the worst kind: the English word "parlance" means idiom, manner of speaking ("in legal parlance") — nearly the opposite of what Berman means, which he defines himself elsewhere in the essay as « ce qui fait qu'une œuvre nous parle », what makes a work speak to us. The live options were (a) keep parlance italicized as a loan, (b) "speaking-power", (c) "address", (d) "the work's capacity to speak". I kept the French italicized. Choosing (b) would have made the sentence read smoothly, and smoothness is precisely the ennoblement Berman is naming; a reader who trips on the italic is in the right relation to it. Signifiance has an English cognate ("significance") that means importance — again a false friend — so it stays French too. Two of the essay's four load-bearing terms cannot be Englished without deciding to lose them. This is internal-judgment-only.

A third false friend, and this one is a trap the English literature actually fell into. In « le niveau de "clarté" sensible des mots », sensible is perceptible, palpable — not "sensible/reasonable" and not "sensitive". Likewise in tendency 1, « cet élément sensible et modifiable du texte en prose qu'est la ponctuation »: punctuation is not "sensitive" in the emotional sense, it is the part you can feel and can alter. I rendered the first "palpable" and the second "delicate and alterable", splitting one French word two ways because English has no term that is both perceptible-to-the-senses and easily-disturbed. Recording the split because the alternative — using one English word for both — would have made the essay look more terminologically fixed than it is.

Déperdition is not "loss". French uses it of heat escaping a house — a continuous leaking-away of something that had been contained, with no agent. "Loss" is neutral and countable; "depletion" is right but clinical; "attrition" imports wearing-down by friction, which is not the image. I chose "bleeding-away", which keeps the leak and the sense that the work is losing something it was holding. It is more figurative than Berman's word and I record that as a debt, not a gain.

Systématisme has no English noun at all. "System" is wrong (Berman contrasts le systématisme with le style and with le système), "systematicity" is available but abstract in the wrong direction, "pattern" throws away the claim that these are systematic rather than merely recurrent. I transliterated to "systematism", which is not English, and left it. The received English rendering ("systematic nature", by report) makes the term into a property; Berman uses it as a countable thing — « un tel systématisme », such a systematism — and the countability is the point, because a text has several.

« Le roman n'est pas moins rythme que la poésie. » French drops the article in predicate position, so the novel is rhythm rather than has rhythm. English cannot drop it in the same place without sounding broken, but "is no less rhythm than poetry is" survives, just, on the strength of the comparative. The alternatives — "is no less rhythmic", "has no less rhythm" — are both smoother and both convert an identity claim into an attribution. Kept the rough version.

« l'étranger du dehors … celui du dedans ». Étranger is simultaneously foreign, foreigner, stranger, and — as Berman's title uses it — the Foreign as such. The sentence says that exoticization renders one kind of étranger by another kind: the foreignness that comes from outside your language, by the foreignness that already sits inside it (Parisian argot standing in for lunfardo). English "foreign from without / foreign from within" keeps the chiasmus and loses that both are the same noun as the essay's title. Considered "the alien" for one limb to keep a noun; rejected because it splits the term the whole essay turns on. The project has this exact problem already and did not notice it was the same problem: S-venuti-invisibility uses "foreignization", which is Venuti's English for the same French/German lineage, and English forces "foreign-ization" (a process done by a translator) where étranger is a condition of the text.

Hyphens I kept. Berman writes re-composer, ré-écriture, ethno-centrisme, mots-obsessions, phonético-signifiante. Each hyphen is doing work — re-compose keeps "compose again" audible under "reorganize"; ré-écriture keeps writing inside rewriting. English tolerates all five, so they cost nothing, and dropping them would have been a small silent rationalization of the very essay that names rationalization.

Where I had to choose between Berman's syntax and English order. Tendency 5's defining sentence is a single 40-word noun-phrase chain with a « ni … ni … » correlative suspended across a parenthetical adverb (« ni, corrélativement, leur richesse signifiante »). English can hold "neither … nor" across a parenthetical, but only with dashes, which Berman does not use here. I used dashes. That is exactly tendency 1 applied to Berman — I repunctuated to a certain idea of the order of a discourse — and I could not see how to avoid it. Recording it because the experiment this translation feeds is about detecting that operation.

What I did not translate and why. Berman's examples (Arlt, Conrad/Gide, Roa Bastos, Thomas Mann, the Sept Fous augmentative network) are the essay's evidence and are what makes it persuasive; they are omitted here because the copyright constraint is real and the twelve definitions are what the study limb needs. A reader of this page is getting Berman's taxonomy, not Berman's argument. The argument's load-bearing move — that these are universaux rather than period norms — is quoted, because the study limb depends on it: it is the licence for applying a French catalogue built on Spanish, German and English novels to a Japanese and an Old English passage.

One thing the translation forced that I did not expect. Rendering the twelve one after another made it obvious that they are not twelve. Numbers 1, 2, 3 and 7 are one operation seen from four angles (Berman says so of 2 and 3 himself: « C'est un corollaire », « C'est là une conséquence »); 10, 11 and 12 are one domain at three grain sizes; 5 and 6 are a genuine pair. What is actually being catalogued is closer to five independent operations, and the pressure toward twelve looks like the pressure of a list. That matters for the study limb: an operator set drawn from Berman is narrower than "twelve", and treating each number as an independent test would overstate the coverage. internal-judgment-only.