Is That a Fish in Your Ear?

by David Bellos

  The degree to which either of these ideas of translation can affect the receiving language and culture doesn’t really depend on their intrinsic merits as translation methodologies or on the brilliance of their users. It depends on volume. Leaving the special features of Bible translation to the side, we can say that the reciprocal flow of translations between any two languages is never equal and in most cases utterly unbalanced. The direction of flow is the key to understanding which way is UP, and what happens down below.

  SIXTEEN

  Translation Impacts

  Some Bible translations have had profound and lasting effects on the receiving language. Luther’s Bible is considered the first monument of modern German, and the King James Bible remains an inescapable reference point in the history of English. However, such impacts are not typical of translation in general. Individual translators do not often produce the smallest ripple in the target culture. However, continuing waves of translated works in particular fields always leave the receiving language in a significantly different shape.

  That was clear to Friedrich Schleiermacher when he set out to explain how the Greek classics should best be translated into German. It wasn’t any one book that would make the difference, he insisted, but only large-scale translation of Greek philosophy and drama that could help the German language “to flourish and develop its own perfect power through the most varied contacts with what is foreign.”1 But depending on the relation between the original and the receiving society, the target may get hit in radically different ways.

  English-language translations of French critical theory from the 1960s to the 1990s, for example, have made abstract discourse about literature in English sound much more like French than it ever did before. In the reverse direction, the language of celebrity journalism in French has been quite transformed by the mass import of English-language styles: la presse people (pronounced pi-pol) exhibits unmistakable signs of what is now denounced as the homogenization of tongues.

  It’s not just a matter of vocabulary. A small but quite profound change in the way dialogue is introduced in Swedish narrative can be traced back to its source in translations of English-language novels.2 Constructions of the following type are ten a penny in modern English fiction of all kinds:

  1. “Don’t try,” she said with disdain.

  2. “It doesn’t matter,” he said calmly.

  3. “And now you must go to sleep,” he said in a tone that was friendly but authoritative.

  4. “Get out,” said Frank abruptly.

  The grammar of Swedish does not make this kind of construction impossible. However, placing a verb of saying together with a modifier (“with disdain,” “calmly,” “abruptly”) after direct speech is fairly unusual in Swedish novel-writing style. In a representative corpus of thirty novels written originally in Swedish, the construction occurs 64 times, but in a parallel corpus of about the same size consisting of novels translated into Swedish from English, it occurs 484 times. This “fingerprint” of English—of one of the English novel’s habitual “dialogue props”—has now been integrated into original writing in Swedish, where it is characteristic not of literary fiction in general but of detective fiction in particular. It’s one of those small yet significant merg-ings of language and style that are often attacked as unintended results of globalization. But Swedish detective fiction has had sweet revenge. Its language may have been infiltrated by an English-language device for the presentation of dialogue, but hard-boiled Swedish crime fiction by Henning Mankell and Stieg Larsson has now conquered the world’s bestseller lists.

  Merging of another kind has been vigorously proposed by the American lawyer Preston Torbert. In his work for U.S. companies doing business in China, he has had to deal with hundreds of contracts that had to be written in two languages—English and Chinese—and have equal validity and force in two jurisdictions. It’s a tall order because their legal traditions have grown up in isolation over many centuries and don’t have many matching terms.3

  One difficulty arises from what is called the “class presumption” in American law. If a contract says that one of its clauses applies to “any house, apartment, cottage, or other building” on some piece of land, for example, that “other building” means, by the force of the class presumption, only another building of the class constituted by “house, apartment, cottage”—that is to say, a residential building. This construction of the sentence is contrary to English usage in a nonlegal context, where “other building” may plausibly refer to a factory, a space station, or a folly.

  Chinese does not have a term for “class presumption,” and its legal culture does not allow for it, either. If the restriction expressed in English is translated without additional modification, the Chinese characters for other building refer equally plausibly to a factory or a workshop as to a residential building, a meaning that the “class presumption” of American legal English specifically excludes. You could, of course, insert additional Chinese characters to say “or any other similar building,” “or any other building of the same class,” “or any other residential construction.” But if it came to a dispute in court, a smart lawyer might be able to claim that the two versions of the contract were not exactly equivalent, since the English contains no words that correspond to the added characters.

  The solution proposed by Torbert is to draft the English in such a way that its Chinese translation is not a problem—that is to say, to modify the source-language text to make it better suited to translation into the target language. Moreover, such a change would make American legalese less arcane, which is of benefit to everybody. The solution is so simple that it makes you wonder why American contracts have not always said “house, apartment, cottage, or other similar building.” Torbert’s answer is that it is because legal drafters have not had Chinese to help them until now. Chinese can teach English-language lawyers how to say what they mean.

  Translation impacts such as these are obviously tiny. French, English, Swedish, and Chinese have not been altered by them, just lightly massaged at the edges—at least, so far. But the translation of the Gospels into Bosavi, a language spoken by small communities of rain-forest dwellers on the Great Papuan Plateau, has had much more far-reaching effects.4

  Before the Bosavi were converted to Christianity in the 1970s, their culture (somewhat like that of ancient Rome) did not recognize sincerity as a concept. It was what people said in public that was taken seriously; private thoughts and the conformity of outward behavior with inner states was not a concern. But sincerity—the correspondence between saying something and meaning it—is integral to the message that Christian missionaries brought. The Asia Pacific Christian Mission regarded vernacular languages as “the shrine of a people’s soul” and was therefore committed to teaching the gospels in Bosavi. However, none of the missionaries was a field linguist, and none became fluent in the language. In addition, Bosavi people in general spoke no other tongue: for trade contacts, they had always relied on speakers in bordering villages who could translate through a neighboring language and, in more recent times, on the regional contact language, Tok Pisin.

  The missionaries used the Nupela Testamen, the New Testament translated into Tok Pisin—not from Latin or Greek but from the simplified English text called the American Good News Bible, first published in 1966, aimed at children and uneducated adults. Use of the intermediary language limited the mission’s initial contact to a small group of young Bosavi men who had worked outside the area and acquired some Tok Pisin. The missionaries taught them basic literacy and then set them on the road as missionizers themselves. At the rudimentary services these new converts organized in the small villages, they read aloud from the Nupela Testamen, then improvised an oral translation into Bosavi, either in small sections or after a whole passage. Given the translation method, it’s not surprising that they introduced many Tok Pisin words and ways of saying into Bosavi as they went along. But the true impact of these “language turners” li
es at a deeper level than that.

  Bosavi is one of the many languages that possess evidentials, grammatical forms that indicate how something is known—by sight, by hearsay, or by deduction (see here). Tok Pisin, by contrast, does not. So when it came to improvising a Bosavi version of the Tok Pisin version of the Good News version of a Bible story focused on the difference between what people thought and what they said, the newly minted Papuan missionaries had a huge problem, exemplified by the phrases in italics below:

  Jesus said to the paralyzed man, “My son, your sins are forgiven.” Some teachers of the law who were sitting there thought to themselves, “How does he dare talk like this? This is blasphemy!” God is the only one who can forgive sins. At once Jesus knew what they were thinking, so he said to them, “Why do you think such things? Is it easier to say this to a paralyzed man, ‘Your sins are forgiven,’ or to say ‘Get up, pick up your mat, and walk’? I will prove to you then, that the Son of Man has authority on earth to forgive sins.” So he said to the paralyzed man, “I tell you, get up, pick up your mat, and go home!”

  Tok Pisin uses na long bel belong, literally “in belly of them,” to express “in their hearts” or “in their minds,” and in the Nupela Testamen this phrase stands together with tingting, “think,” to express the fact that the teachers of the law “thought” something without saying so. The Bosavi oral translators couldn’t say anything quite so ungrounded in evidence. One recorded version has Jesus knowing by direct visual evidence what the men of law were thinking—the evidential suffix -lo: b is added to the verb for “think.” Also, it adds a tag, a: la: sa: lab, to the whole line, meaning something like “it says,” or at any rate grounding the source of the knowledge not in the actual speaker but in some external authority. But versions varied among different preachers and occasions quite considerably, until a formal borrowing (a syntactic calque) from Tok Pisin became accepted as a new way of referring to “inner thought,” thought not evidenced by words spoken aloud: kufa, literally, “of belly,” prefixed to the verb for “think.” The effort of translation has altered the language of Bosavi, and with it, a whole mental world. “Private thoughts” are now “belly-think” in Bosavi, or, to put it the other way around, thanks to frontline and improvised language mediation, what a speaker of Bosavi can now do with his belly has undergone a huge change.

  Changes brought about in the life of the Bosavi by the missionary effort obviously go far beyond the grammar and vocabulary of their language. However, the change in the way speakers of Bosavi can now conceptualize and refer to “inner life” is not only an effect of conversion to Christianity but also a direct impact of translation—the translation of the gospels from Tok Pisin into the Bosavi tongue.

  Most commentaries on the effects of translation on receiving cultures of the remote or recent past use words such as enrich, extend, and improve to describe how the target was hit. But when we can see and hear it happening in our own present time, quite other metaphors crop up: distort, mangle, and homogenize come to mind. The role of evidentials in Bosavi grammar has been irreparably diminished by the improvised calques from Tok Pisin that provide a way of talking about things that have no evidential status at all. From some points of view that has mangled a unique and irreplaceable mental world. Similarly, we could say that the mass import of English-style celebrity gossip into French media has produced a stylistic monstrosity that cheapens the language itself. However, in other times and places, much greater lexical and stylistic changes of the same nature have given rise not to lamentation but to feelings of the opposite kind. For example, Japanese translators imported many scientific terms from European languages in the late nineteenth century, and most users of those new terms considered their language had been enriched by them. Similarly, in the fourth to eighth centuries C.E., Syriac (a Semitic language closely related to Aramaic) is said to have flowered in the hands of Severus Sebokht, a bishop, scholar, and translator who imported quantities of Greek words and expressions together with the mathematical, medical, and astronomical knowledge of the ancient Greeks that the Latin West had ignored (and would not rediscover for centuries, until Arabic translations of those Syriac translations of Greek science were translated once again in the middle of the twelfth century C.E., in Toledo [Spain], by Gerard of Cremona, into Latin, for wider distribution throughout Europe).5

  The Christian fundamentalists who converted the Bosavi people may indeed believe they enriched the language of the souls they have saved; and I suppose there may have been Syrian naysayers all those years ago who thought the mass import of Greek terms had wrecked their own ancient tongue. But the fact is that attitudes toward language change induced or accelerated by translation are not motivated exclusively by feelings about language or about translation. They arise from deeply seated and far less tractable ideas.

  The first of these is the place you think your language ought to occupy in the hierarchy of translation tongues. For many people, especially those caught in the mind-set of a monolingual European nation-state, this is a sensitive topic; because the imagined rank of a language often conflicts with reality, this can give rise to collective hypocrisy and spite. French people who look down on the use of English words that they nonetheless import by the bucketload are in this kind of plight. They are not alone.

  The second major constituent of attitudes toward language change propelled by translations is the value you place on what it is that the new vocabulary brings. Translation impacts on a receiving language can’t really be separated from the impact that the translated material has. At different times, translations may flood receiving cultures with Hollywood glitz, shipbuilding techniques, religious salvation, saucy stories about Marie Antoinette—just about anything that’s ever been thought worth writing down. The value you attach to the linguistic traces of such flows is subordinate to your need or desire for the material that the translations in question make available for the first time.

  The damage done to other cultures by lopsided translation flows is no different from the benefits brought to receiving languages by lopsided translation flows. The real damage and the real benefits lie not in translation as such, or in its impacts on receiving languages, but in the nature of the works that translation spreads.

  SEVENTEEN

  The Third Code: Translation as a Dialect

  What language do you speak? That sounds like a merely factual inquiry with an uncomplicated answer, whatever it is. But as I was reading an American newspaper during the financial crisis of 2008, I learned that the U.S. treasury secretary was about to unveil the big megillah to put an end to the tsunami that was rocking Wall Street at the time. What language was that? Well, English—but only sort of. It was also, marginally, in Hebrew (mediated by Yiddish) and in Japanese, too. I can translate it into French—M. Paulson s’apprête à dévoiler la bonne méthode pour calmer la tourmente des marchés—but that doesn’t prove the sentence was in English, only that I understood it. I can back-translate the French sentence in any number of ways—but that would only show that “English” is a far from determinate thing.

  Translators working into English are confronted on every page with decisions about the nature, scope, identity, and audience of the language they are writing. I write in a personal idiom that bears traces of my upbringing in England, my long stay in Scotland, and my present life on the East Coast of the United States. When I write a translation, however, I have to make choices in every paragraph about what variety of written English to use. As is well known, spellings, numbering systems, greetings, and curses, as well as several hundred common vocabulary items, have different forms in different parts of the English-speaking world. It drives me mad. How do I know what is “English” and what is something else?

  The practical solution is this: I write the way I like, and then a skillful copy editor amends my prose to make it conform to the style appropriate to the output and the target audience of a particular publishing house. But that is only the outward form of the sol
ution. The target audience of most English-language publishing houses, for most of the books they put out, is indeterminately large, and includes American, Australian, Indian, Canadian, and South African readers—each large grouping feeling most at home in significantly different varieties of the spoken and written tongue. So what gets edited out in any of my translations—and in any translated literary or nonfiction work of more than local interest—are those quirks of language that mark it as belonging to any geographical variety of English. In other words, I get de-Britted if I am being edited for U.S. publication and de-Yanked (a less difficult job, since my Americanisms are few and far between) when a London publisher takes the lead. What you get at the end of the process is “English-minus”—ideally, a common center ground of the English language, stripped of vocabulary and turns of phrase that are not understood or understood in different ways in any part of the messy spread of what is still called, for want of a better word, the English tongue.

  The language of translations in English is therefore not a representation of a language spoken or written anywhere at all. Because its principal feature is to be without regional features, it’s hard to see from outside—and that’s precisely the point of this sophisticated stylistic trick. “Tranglish” is quite different in nature from the clumsy International English of social science and global journalism. It’s smooth and invisible, and it has some important advantages. Detached with skill and craft by professional language doctors from any regional variety of the tongue, it is much easier to translate than anything actually written in “English” by a novelist from, say, Queensland, Ireland, Wessex, or Wales. But as it is already translated (from French, in my case, but this would be just as true if I were working from Russian or Hindi), any remaining strangeness in the prose, in the ears of a speaker of any of the myriad varieties of English the world over, is automatically construed as a trace of the foreign tongue, not of the translator’s identity. The “translator’s invisibility,” eloquently denounced by Lawrence Venuti as a symptom of the anti-intellectual, antiforeign bias of Britain and America,1 is also the unintended result of the unbounded nature of the English language itself.