by David Bellos
FOUR
Things People Say About Translation
It’s a well-known fact that a translation is no substitute for the original.
It’s also perfectly obvious that this is wrong. Translations are substitutes for original texts. You use them in the place of a work written in a language you cannot read with ease.
The claim that a translation is no substitute for an original is not the only piece of folk wisdom that isn’t true. We happily utter sayings such as “crime doesn’t pay” or “it never rains but it pours” or “the truth will out” that fly in the face of the evidence—Russian mafiosi basking on the French Riviera, British drizzle, and family secrets that never get out. Adages of this sort don’t have to be true to be useful. Typically, they serve to warn, console, or encourage other people in particular circumstances, not to establish a theory of justice, a weather forecasting system, or forensic science. That’s why saying a translation is no substitute for the original misleads only those who take it to be a well-known fact. It’s truly astounding how many people fall into the trap.
When you say “crime doesn’t pay” to a teenager caught filching a DVD from a market stall, it does not matter whether you believe this to be true or not. You are trying to steer the young person toward acceptance of the eighth commandment and using a conventional phrase in the service of that moral aim.
Similarly, a schoolteacher who has just caught his students reading The Outsider in English when they were supposed to be preparing their lessons by reading Camus’s novel in French may well admonish them by saying in an authoritative tone of voice, “A translation is no substitute for the original!” The students know it’s not true because they have just been caught using the translation as a substitute for the original. But they also understand that the teacher used a piece of folk wisdom to say something else that really is true—that only by reading more French will they improve their language skills. The teacher means to spur them into greater assiduity, not to speak the truth about translation.
Students eventually graduate and get jobs, and soon enough some of them start writing book reviews. In those circumstances, when they have to write about a work of foreign literature translated into English and are lost for a phrase to use, they may parrot the warning they first heard at school. In common with all things people say and write, however, the force of the saying that “a translation is no substitute for an original” is completely altered when the context of utterance is changed.
In its new context, it means that the writer of the book review possesses sufficient knowledge of some original to be able to make a judgment that its translation is not a substitute for it. Whether or not the reviewer really has read the original work, the assertion that the translation does not constitute a substitute for it puts the reviewer in charge.
Using the adage in this way obviously affects the meaning of the word substitute. If, for example, I said, “Instant coffee is no substitute for espresso made from freshly ground beans,” I would be wrong, in the sense that the purpose of instant coffee is to serve as a substitute for more laborious ways of making the drink; but also right, as long as the word substitute is understood to mean “the same as,” “as good as,” or “equivalent to.” Instant coffee is clearly not the same as espresso; many people regard it as not as good as espresso; and because preferences in the field of coffee are matters of individual taste, it is not unreasonable to treat powdered coffee as not equivalent to espresso. We do often say all these more explicit things about coffee. But it is not so straightforward when it comes to translation.
People who declare translations to be no substitute for the original imply that they possess the means to recognize and appreciate the real thing, that is to say, original composition as opposed to a translation. Without this ability they could not possibly make the claim that they do. Just as an inability to distinguish two types of coffee would deprive you of any possibility of comparing them, so the ability to discriminate between “a translation” and “an original” is a basic requirement for anyone who wants to claim that one of them is not the same as, equivalent to, or as good as the other.
In practice, we look at the title page, jacket copy, or copyright page of a book or the byline at the bottom of an article to find out whether or not we are reading a translation. But in the absence of such giveaways, are readers in fact able to distinguish, by the taste on their linguistic and literary tongues, whether a text is “original” or “translated”? Absolutely not. Countless writers have packaged originals as translations and translations as originals and gotten away with it for weeks, months, years, even centuries.
Fingal, an Ancient Epic Poem in Six Books appeared to great acclaim in 1762. For many decades, it was held to give precious insight into the ancient culture of the original inhabitants of Europe’s northwestern fringe. Figures as eminent as Napoleon and as learned as the German philosopher Johann Gottfried von Herder were entranced by the authentic folk poetry of the “Gaelic Bard.” But they were wrong. The story of Ossian hadn’t been invented by Celtic poets at all. It was written in English by a minor poet called James Macpherson.
Horace Walpole had a shorter run. In the introduction to the first edition of The Castle of Otranto (1764) he claimed his novel was but a translation of an Italian work first published in 1529, and he promised to make it available if his work met with any success. It did—in fact, it was a bestseller and spawned a whole genre of writing called “Gothic horror.” A second edition was needed, and so the author had to eat humble pie. He could not produce the Italian original, for there was none. He, too, had written his “translation” in English.
Even grander deceptions speckle the history of many literatures. The Letters of a Portuguese Nun, first published in French in 1669, purports to be a translation, even though the original was never produced. This exquisite spiritual text fascinated readers for three centuries and was translated from French into many other languages—one version was done into German by the poet Rainer Maria Rilke, who never even suspected that he had been taken for a ride. The letters had in fact been written in French by Guilleragues, a friend of Jean Racine. The hoax was not unraveled until 1954.1
A more recent example of pseudo-translation in French is provided by Andreï Makine, whose first three novels, published between 1990 and 1995, were presented as works translated from Russian by the fictional Françoise Bour. In 1995 Le Monde revealed that they were French originals and thus cleared the way for Makine’s fourth novel, Le Testament français, to win the Prix Goncourt, which is awarded only to writers of French.
Pseudo-translations can be hard to kill off once they have come to life. In Soviet Russia, the poet Emmanuel Lifshitz felt he could express himself more fully by writing as if he were someone else—as James Clifford, an Englishman who did not exist. Originally printed in The Batum Worker, the twenty-three poems purportedly translated from English were reprinted in Moscow with a short biography of the poet, which tried to give the game away in its closing sentence: “Such could have been the biography of this English poet, who grew up in my imagination and who has materialized in the poems whose translation I ask you to consider.”2 But even clues as big as that can be missed by readers who really want to believe they can tell the difference between originals and translations. Lifshitz did not include the Clifford poems in collections of his own verse, and that is perhaps why James Clifford lived on in literary circles as a well-known English poet for many years. In conversation with Lifshitz, Yevgeny Yevtushenko mentioned how well he remembered the melancholic Englishman—a true eccentric.3
Examples of the reverse process, passing off translations as original works, are probably just as numerous. Three novels by the multilingual writer-diplomat Romain Gary that were purportedly composed in French (Lady L., 1963; Les Mangeurs d’étoiles, 1966; and Adieu Gary Cooper, 1969) had actually been written and published in English (as Lady L, 1958; The Talent Scout, 1961; and The Ski Bum, 1965, respectively) then secretly
translated by a senior editor at Gary’s French publishing house. How many translations have been misrepresented as originals and never rumbled? It can’t be the case that every deception of the kind has already been unmasked.
Authors have many reasons for wanting to pass off original work as a translation and a translation as an original. Sometimes it helps to get through censorship, sometimes it is to try out a new identity. It can serve individual or collective fantasies about national or linguistic authenticity, and it can be done just to pander to a public taste for the exotic. What all such deceptions underscore is that reading alone simply does not tell you whether a work was originally written in the language you are reading it in. The difference between a translation and an original is not of the same order as the difference between powdered and steamed coffee. It’s more than just an idea. But it is not at all easy to demonstrate.
The idea that a translation is not a substitute for an original work must also be subjected to another critique. If the adage were true, then what would users of a translation get from reading a translation? Not the real thing, obviously. But they would not even get a substitute for it—not even the literary equivalent of powdered coffee. Asserting the irreplaceable nature of a literary original condemns those who cannot read the language in question to the consumption not of Nescafé but of dishwater. No opinions would be worth holding except by those who read works in the original.
Yet the examples of Cervantes (Don Quixote claims to be translated from the Arabic), Walpole, Macpherson, Gary, Guilleragues, Makine, Clifford, and countless others demonstrate that nobody can be certain that what he has read is an original.
Ismail Kadare tells another story about the indistinction of original and translated texts in his memoir-novel, Chronicle in Stone. As a ten-year-old, he was entranced by a book he’d been given by an uncle. With its story of ghosts, castles, murder, and betrayal, it appealed to him immensely, especially as it seemed to explain some of what had been going on around him in the fortress city of Gjirokastër over the preceding years of war and civil strife. The book’s title? Macbeth, by William Shakespeare. Young Ismail could see Lady Macbeth down the street, wringing her hands on the balcony, washing away the terrible things that had happened in her home. He had no idea that the play had originally been written in English. In childish fascination with a text he reread many times, Kadare copied out the unsuspected translation by hand, and nowadays, when asked by interviewers which was the first book he ever wrote, he always answers, with only half a smile, Macbeth. To this day, Kadare has not learned to speak English, but he counts Macbeth as the founding experience of his own life in literature. Whatever the quality of the translation that so inspired him, it clearly did not have the effect of dishwater. It was more like an elixir.
Why then do people still say that a translation is no substitute for an original? The adage might conceivably be of use to people who consciously avoid reading anything in translation, as it would justify and explain their practice. But since there is no reliable way of distinguishing a translation from an original by internal criteria alone, such purists could never be sure they were sticking to their guns. And even if by some stroke of luck they did manage to keep clear of all but original work in their reading, they would end up with a decidedly peculiar view of the world—if they were English readers, they would have no knowledge of the Bible, Tolstoy, or Planet of the Apes. All the adage really does is provide spurious cover for the view that translation is a second-rate kind of thing. That’s what people really mean to say when they assert that a translation is no substitute for original work.
FIVE
Fictions of the Foreign: The Paradox of “Foreign-Soundingness”
For most of the last century, reviewers and laymen have customarily declared in order to praise a translation to the skies that it sounds as if it had been written in English. This is hollow praise, since the selfsame community of reviewers and laymen has often shown itself unable to tell when an alleged translation was written in English. All the same, the high value placed on naturalness and fluency in the “target,” or “receiving,” language is a strong feature of the culture of translation in the English-speaking world today. But there are contrarian voices. If a detective novel set in Paris makes its characters speak and think in entirely fluent English, even while they plod along the boulevard Saint-Germain, drink Pernod, and scoff a jarret de porc aux len-tille s—then something must be wrong. Where’s the bonus in having a French detective novel for bedtime reading unless there is something French about it? Don’t we want our French detectives to sound French? Domesticating translation styles that eradicate the Frenchness of Gallic thugs have been attacked by some critics for committing “ethnocentric violence.”1 An ethics of translation, such critics say, should restrain translators from erasing all that is foreign about works translated from a foreign tongue.
How then should the foreignness of the foreign best be represented in the receiving language? Jean d’Alembert, a mathematician and philosopher who was also co-editor of Diderot’s Encyclopédie, came up with an ingenious answer in 1763:
The way foreigners speak [French] is the model for a good translation. The original should speak our language not with the superstitious caution we have for our native tongue, but with a noble freedom that allows features of one language to be borrowed in order to embellish another. Done in this way, a translation may possess all the qualities that make it commendable—a natural and easy manner, marked by the genius of the original and alongside that the added flavor of a homeland created by its foreign coloring.2
The risk of this approach is that in many social and historical circumstances the foreign-soundingness of a translation—just like the slightly unnatural diction of a real foreigner speaking French (or English, or German … )—may be rejected as clumsy, false, or even worse.
In fact, the most obvious way to make a text sound foreign is to leave parts of it in the original. Such was the convention in Britain in the Romantic era. In the earliest translation of the novel now known in English as Dangerous Liaisons, for instance, characters refer to and address one another by their full titles in French (monsieur le vicomte, madame la présidente) and use everyday expressions such as Allez!, parbleu!, and ma foi! within sentences that are in other respects entirely in English. 3 Similarly, in recent translations of the novels of Fred Vargas, the lead character, Jean-Baptiste Adamsberg, retains his French rank of commissaire in charge of a clutch of brigadiers, but he talks to them in English.4 Following the same logic of selective foreignism, German officers in most Second World War movies made in Hollywood speak natural English interrupted at regular intervals by Jawohl, Gott im Himmel, and Heil Hitler.
The device may be taken much further, in popular as well as classical works. The dubbed Italian version of Singin’ in the Rain, though it performs miracles of lip-synch in the translation of witty patter, leaves the sound track of the title song in the original English. A famous modern production of King Lear in Chinese has Cordelia speaking Shakespeare’s lines—she speaks the truth to her father in the true language of her speech.5
In general, however, translations only simulate the foreign-soundingness of foreign works. In fact, the challenge of writing something that sounds like English to speakers of other languages can even be met by not writing English at all.
English is heard around the world in pop songs, TV broadcasts, and so on by millions of people who do not understand the words of the lyrics, jingles, and reports. As a result there are large numbers of people who recognize the phonology of English—the kinds of sounds English makes—without knowing any English vocabulary or grammar. Some forty years ago, an Italian rock star performed a musical routine in which he pretended to be a teacher of English showing his class that you do not need to understand a single word in order to know what English sounds like. Sung to a catchy tune, Adriano Celentano’s “Prisencolinensinainciusol ol rait” is a witty and surprising simulation of what English sounds like
—without being in English at all. However, the transcription of “anglogibberish” in textual form represents English-soundingness only when it is vocalized (aloud, or in your head) according to the standard rules for vocalizing Italian script. “Prisencolinensinainciusol ol rait,” which can be found on many currently available websites and in some cases with one of its possible transcriptions, is a specifically Italian fiction of the foreign.
It is equally possible to produce gibberish that sounds foreign to English ears. A famous example is the song sung by Charlie Chaplin in Modern Times (1936). Having got a job as a singing waiter, the hapless fellow finds himself on the restaurant dance floor with the band thumping out a French music-hall tune, “Je cherche après Titine”—but he does not know the words. Chaplin dances, mimes, looks perplexed. Paulette Goddard, in the wings, mouths, “Sing!” Our lip-reading is confirmed by the intertitle: “Sing! Never Mind the Words!”
Chaplin then launches into a ditty in Generic Immigrant Romance, which for English speakers only can be represented thus:
Se bella giu satore
Je notre so cafore
Je notre si cavore