How much really gets lost in translation?
The New Yorker August 4, 2014
Once, in a restaurant in Italy with my family, I occasioned enormous merriment, as a nineteenth-century humorist would have put it, by confusing two Italian words. I thought I had, very suavely, ordered for dessert fragoline—those lovely little wild strawberries. Instead, I seem to have asked for fagiolini—green beans. The waiter ceremoniously brought me a plate of green beans with my coffee, along with the flan and the gelato for the kids. The significant insight the mistake provided—arriving mere microseconds after the laughter of those kids, who for some reason still bring up the occasion, often—was about the arbitrary nature of language: the single “r” rolled right makes one a master of the trattoria, an “r” unrolled the family fool. Although speaking feels as natural as breathing, the truth is that the words we use are strange, abstract symbols, at least as remote from their objects as Egyptian hieroglyphs are from theirs, and as quietly treacherous as Egyptian tombs.
Although berries and beans may be separated by a subtle sound within a language, the larger space between like words in different languages is just as hazardous. Two words that seem to indicate the same state may mean the opposite. In English, the spiritual guy is pious, while the one called spirituel in French is witty; a liberal in France is on the right, in America to the left. And what of cultural inflections that seem to separate meanings otherwise identical? When we have savoir-faire in French, don’t we actually have something different from “know-how” in English, even though the two compounds combine pretty much the same elements?
These questions, about the hidden traps of words and phrases, are the subject of what may be the weirdest book the twenty-first century has so far produced: “Dictionary of Untranslatables: A Philosophical Lexicon,” a thirteen-hundred-page volume, originally edited in French by the French philologist Barbara Cassin but now published, by Princeton University Press, in a much altered English edition, overseen by the comp-lit luminaries Emily Apter, Jacques Lezra, and Michael Wood. How weird is it? Let us count the ways. It is in part an anti-English protest, taking arms against the imperializing spread of our era’s, well, lingua franca—which has now been offered in English, so that everyone can understand it. The book’s presupposition is that there are significant, namable, untranslatable differences between tongues, so that, say, “history” in English, histoire in French, and Geschichte in German have very different boundaries that we need to grasp if we are to understand the texts in which the words occur. The editors, propelled by this belief, also believe it to be wrong. In each entry of the Dictionary, the differences are tracked, explained, and made perfectly clear in English, which rather undermines the premise that these terms are untranslatable, except in the dim sense that it sometimes takes a few words in one language to indicate a concept that is more succinctly embodied in one word in another.
Histoire in French means both “history” and “story,” in a way that “history” in English doesn’t quite, so that the relation between history and story may be more elegantly available in French. But no one has trouble in English with the notion that histories are narratives we make up as much as chronicles we discern. Indeed, in the preface, the editors cheerfully announce that any strong form of the belief to which their book may seem to be a monument is certainly false: “Some pretty good equivalencies are always available. . . . If there were a perfect equivalence from language to language, the result would not be translation; it would be a replica. . . . The constant recourse to the metaphor of loss in translation is finally too easy.” So their Dictionary is a self-exploding book, like one of those kinetic works of art that Jean Tinguely used to make, where the point of the work is to watch it self-destruct in the museum garden.
Yet Tinguely was a considerable and entertaining artist, and this is a considerable and entertaining book, full of odd words beautifully, at times owlishly, annotated. Though most of its instances are in English, French, German, and Italian, the editors still have time for a discourse on the nuances of the Romanian dor, a word whose approximate meaning is “longing” but that is “a lyrical expression of the feeling of finitude, between folk metaphysics and philosophical reflection, and self-consciously Romanian.” Derived from dolus, a vernacular-Latin noun meaning “suffering,” dor indicates, according to the classicist Anca Vasiliu, something more particular and estranged than melancholy, less sentimental and self-indulgent than Weltschmerz—and, it occurs to a reader, strikes the central emotional tone of both the French-Romanian aphorist Emil Cioran and the Romanian-American artist Saul Steinberg. They are masters of dor.
Not far away, there is a fascinating entry on the relationship between the Freudian concept of Trieb and the English “drive,” the word often used to translate it. The thought is that “instinct” might much more closely match Freud’s original meaning, and that using “drive” so relentlessly has made American Freudianism a far more driven practice than the original. Rather nicely, in a book devoted to words, logos, translated in the King James Version of the Gospel of John simply as “the word,” turns out to be the most all-purpose of items. Twenty-three alternate meanings for it are listed in English alone—it is, the editors say, a model of “polysemy,” packing multiple meanings into a single sign, and managing to suggest both words themselves and the wider shores of wisdom that words articulate. Meanwhile, many fine pages are devoted to the fine shades separating “piety” from “pity,” provoking in a reader the thought that where the Christian desire was to discriminate one from the other—moral duty from moral impulse—modern humanism is essentially an attempt to return the words to a single meaning.
Some words have surprisingly specific authors. “Spleen” extends, like a suspension bridge, between the twin piers of Shakespeare and Baudelaire: Shakespeare took it out of ancient medicine and gave it life as an expression meaning masculine overcharge (“Quicken’d with youthful spleen and warlike rage”); three centuries later, Baudelaire gave the word a second life, adding to the original meaning an overlay of beetle-browed irritation. Sprezzatura is “a word that is untranslatable par excellence,” with not even a close translation outside Italian, but then it is not exactly in Italian. The term is, the editors explain, a self-conscious invention by the sixteenth-century writer Baldassare Castiglione, to label his own idiosyncratic concept of a gentleman’s seeming indifference to polish; he made it up as much as Lewis Carroll made up mome raths.
Significant patterns of meaning do emerge from the mists. Some words alter because of the stray contingencies of time, others because of the specificities of political history. To take those two introductory instances: although one could try to draw significance from the fact that spirituel means not “spiritual” but “witty,” the truth is that it’s just an etymological accident. The same meaning lingers in our word “spirited.” (When Shakespeare has Brutus say that Antony has a “quick spirit,” he means that he’s very bright, not very good.) The French are as unlikely as we are to think that a witty man is a spiritual one, and if there’s a moment’s confusion it gets clarified by tone and touch.
On the other hand, though the word “liberal” in French does not mean what “liberal” means in English—a French liberal admires Margaret Thatcher, is a critic of the welfare state, and supports the free market whenever possible—the difference is historical in another sense. France took a statist turn very early, so speaking up for liberty meant speaking against the state. In America, given the centrality of the struggle for civil rights, speaking up for liberty has often meant speaking up for the state. A liberal in France does stand for most of the same things as a liberal in America or in England; he just seems farther to the right, because the situation in France is, for historical reasons, skewed so much farther left.
Of all these words, one and one alone seems truly to stump the editors, and that is the German word Dasein, made famous in Heidegger’s “Being and Time,” and usually left by bewildered translators in the original German, and meaning—well, what it means bewilders even the Continental editors. Calling it a “paradigm of the untranslatable,” they struggle over six double-columned pages to explain, anxiously, what it might mean: roughly, “life,” but with an overtone of anxiousness (“the very being of the being that we are, essentially or inessentially, not in the sense of an identity, but in proportion to a being that we have ‘to be’ ”), the whole leading to the sneaking suspicion that Dasein is untranslatable because it has no particular meaning in the original language, either. A truly untranslatable word, it seems, may be the sign of an unsustainable concept.
A spectre haunts this book, however. It is the spectre of Benjamin Lee Whorf and the theory of linguistic relativism to which he gave his name. Whorf was an amateur American linguist in the first half of the twentieth century who became obsessed with the idea that the system of tenses in the Hopi language gave the Hopi a different view of present, past, and future. (His understanding of Hopi grammar turns out to have been rudimentary.) Whorfianism came to refer to a larger idea derived from this notion—the idea that our language forces us to see the world a certain way, and that different languages impose different world views on their speakers. It’s a powerful idea in the pop imagination. It sounds right when you say it.
Yet “Whorfian” relativism, at least in its strong forms, is one of those ideas that disappear under any kind of scrutiny. After all, if we were truly prisoners of our language, we shouldn’t be able to use it to see its limits clearly, or to enumerate the concepts that it can’t conceive. The ghost of Whorf haunts every page of the “Dictionary of Untranslatables”—only to be exorcised by the authors, only to return. For instance, in a long section on the use of the English suffix “-ing,” we are told that our possession of such a form means that we have instant access to an idea of change-in-motion. The title of E. P. Thompson’s classic “The Making of the English Working Class” is cited as an example of how English lends itself to stories of historical change as action. In French, Thompson’s book would be called “The Formation of the English Working Class,” and if this is a big difference it’s hard to hear.
In recent years, there’s been some empirical support for mild versions of the Whorfian idea. Given more color names—“aqua,” “teal,” and “periwinkle,” in addition to “blue”—we do seem to respond to more colors, or at least to group the colors we’re shown more finely. In other words, having many words for shades of blue helps you tag the memory more easily and retrieve it faster, though it doesn’t mean that you really see more shades than the next guy. (Common sense tells us this already about, say, wine tasting: when we’re given new terms—there’s tar, tobacco, and rosewater here—we’re more likely to say, “Oh, yeah, I smelled that!” than “Oh, now I smell something new.”) The names help us sort the steady perception into manageable bits. Similar studies have helped rehabilitate Whorf, at least a little.
In response to this revival of pop Whorfianism, the linguist John H. McWhorter has written a short, sour, brilliant little book, “The Language Hoax” (Oxford), which attempts once again to explain to a general audience why the idea of linguistic relativism is empty. Patiently dissecting a bad idea, McWhorter explains that even the most robust Whorfian effects, like those in color sorting, involve, at most, microsecond differences in doing tasks—less time than it takes for kids to start laughing at the wrong word in a trattoria. It is no more surprising that a tribe whose language doesn’t have numbers can’t do math, he says, than that a “tribe without cars doesn’t drive.” McWhorter also puts his finger on a core problem: “A difference in thought must be of a certain magnitude before it qualifies realistically as a distinct ‘worldview.’ ” Even if it were true that the participle in English made us a mite more likely to think “actively,” differences in ideology and belief overwhelm and obliterate those lexical tints. Buddhists and Stoics have no trouble at all making their fatalism felt in English.
McWhorter makes all the right arguments, and he makes them clearly. And yet some version of the case persists, as imperishable as a zombie. What’s the allure of linguistic relativism? There may be solace in imagining ourselves prisoners of circumstances beyond our control—of language or horoscopes, of God or Capital—and so relieved of responsibility for what we do next. It may also be that linguistic relativism gives a kind of cheap knowingness that we all enjoy: you’re a prisoner of your tongue, but I’m the one who can show that you’re imprisoned. In truth, language seems less like a series of cells in which we are imprisoned than like a set of tools that help us escape: some of the files are rusty; some will open any door; and most you have to jiggle around in the lock. But, sooner or later, most words work.
Curiously, McWhorter only briefly dismisses the author who continues to give linguistic relativism its greatest cachet among literate people: George Orwell, whose essay “Politics and the English Language” (1946) made the claim that the debasement of thought cannot be separated from the debasement of language. Criticizing Orwell is as offensive to most humanists as criticizing Aquinas is to Catholics, but the essay gives mere obfuscation a cognitive power it never had. Orwell rightly detested double-talk, cheap euphemism, and deliberate obscurity—the language of “strategic hamlets” and “enhanced interrogation,” and all the other phrases that are used to muddy up meaning. But euphemism is a moral problem, not a cognitive one. When Dick Cheney calls torture “enhanced interrogation,” it doesn’t make us understand torture in a different way; it’s just a means for those who know they’re doing something wrong to find a phrase that doesn’t immediately acknowledge the wrongdoing. If the strong form of linguistic relativism were true, then not having the correct phrase or being forced to use a weird one would change our perception of what’s taking place. There’s no evidence that this happens. Whatever name Cheney’s men gave torture, they knew what it was. A grotesque euphemism is offensive exactly because we recognize perfectly well the mismatch between the word and its referent. It’s an instrument of evasion, like a speeding getaway car, not an instrument of unconsciousness, like a blackjack.
If lucid writing is the sign of a moral state, it’s the moral state of hard work, keener effort, acquired craft—a desire to communicate rather than intimidate, to have fun with a fellow-mind rather than bully a disciple. Sane and shapely sentences are good because they’re sane and shapely. There’s no guarantee that they’ll contain the truth: lots of sane and shapely sentence makers have had silly ideas. But, like sane and shapely people and homes, they are nice to have around to look at.
Although the differences between languages are minute, and can’t be elevated to either a prison house or a world view, it is in language’s minutiae that the small gestures of art live. In the 1939 French film noir “Le Jour Se Lève,” for instance, there’s a moment when the concierge, pressed by the police about a piece of furniture in the killer’s apartment, tells them that it’s “une armoire comme toutes les armoires”—“An armoire like all armoires.” The English subtitle has him saying, “It’s just a regular armoire.” This is the sense of it, but there is something universalized and pseudo-systematic in the French form that is part of the flavor of French life. The concierge is making a categorical point about furniture.
Back in the social sciences, there are studies to support our sense of such differences—not in cognitive view but in cultural flavor. Bilingual people, for instance, seem to narrate stories very differently in their two languages. Russian émigrés to America seem to use more collectivist nouns when they’re speaking Russian, more individualistic ones in English; bilingual French-English speakers tend to tell the same stories with an emphasis on “achievement” in English, and on “aggression toward peers” in French. (The English story is “I done it!”; the French version is “And the bastards tried to stop me.”) This seems little different from the truth that the rituals and habits of playing hockey (cursing, stoical indifference, then handshaking) are different from those of playing soccer (whining, faking injuries, etc.). Obviously, immigrants have sociolinguistic habits among themselves that are different from the social habits acquired by speaking their new language. We don’t speak French or Italian if we don’t know the way to speak French and Italian.
We are not captives of our tongues, but we are citizens of our languages. And citizenship is a broad concept that includes behavior and rituals. We approach the secret life of another language more intimately on first approach than after we have married into it. Learning a new language is like learning a new city: you see things you’ll never notice, or need, once you go to live there and are habituated by routine. There’s even a very real sense in which it is easier to “think” in a foreign language if you don’t quite speak it than if you do. The gender properties of French, for instance—tables are feminine, candlesticks masculine—leap out as you learn the language. They can seem to eroticize the world. But they quickly recede, and become neutralized as standard features. (Just as English speakers don’t really notice the encoded oddities of English; for instance, that we don’t have a future tense and plans have to be conjugated as acts of will—“I’ll be going later.”) Anyone whose grandparents reverted to Yiddish in moments of exasperation will recall how this dual linguistic citizenship works; your grandfather did not say “What a gonif!” because there were no curse words in English.
The rubber meets the road—or la gomma tocca la strada, as we Italian speakers say—when it comes to translations of important poetry and literature generally. No matter how well we have immunized ourselves against empty Whorfianism, when the German sighs and tells us that the Rilke we know in English is nothing like the real Rilke, we tend to credit it, if only because we know that the Keats the German speaker knows can really be nothing like Keats. (“Schönheit ist Wahrheit, Wahrheit Schönheit” is the German for “Beauty is truth, truth beauty,” but there is nothing beautiful about it: just a look tells us that the balance of the monosyllable against the disyllable, what we might call the echt sprezzatura of the original, isn’t there.)
Near-contemporary translations of great texts often seem to capture them best, whatever their small errancies. New York Review Books has just put out a new edition of John Florio’s translation of essays by Montaigne. Florio, a friend of Shakespeare’s, can sometimes be compacted and muddled—there have been many more “faithful” translations since—but the rendering rings right. The background assumptions are so confidently registered that the foreground turns can be pictured in the proper light, too. The rhythm in Florio’s Montaigne is pleasingly Shakespearean in its play between the forthright and the fancy, a play easily lost by mere homogenized lucidity: “I cannot settle my object. It goeth so unquietly and staggering, with a natural drunkenness. I take it in this plight, as it is at the instant I amuse myself about it. I describe not the essence but the passage.”
Similarly, though C. K. Scott Moncrieff, Proust’s first translator into English, apologized to Proust for his “imperfect” French, and though it’s often said that Scott Moncrieff is self-consciously poetic where Proust is not, his contemporary translation of Proust remains much the best ever achieved, exactly because the aesthete’s point of view was so deeply in his blood and bones. He is self-consciously poetic in part because English aestheticism, made under the moon of Wilde, was always damper, more elaborately, self-consciously poetic than the drier French equivalent. He missed the idiomatic equivalences sometimes. (When Swann, in the last sentence of “Swann in Love,” confesses, with shattering banality, that Odette, the woman he has sacrificed so much for, had never really been his type, Scott Moncrieff has the awkward literalism “not my style.”) Yet though he gets small things wrong, the murmuring rhythm and violet color of the book have never quite been matched by the more accurate later renditions. He finds the right English sound to match the French one.
In a fine 2011 study of translation, “Is That a Fish in Your Ear?,” David Bellos points out that, despite the endless insistence that the real thing is always lost in translation, we readily translate everything, and all the time. “Think of a great poet, and you’ve almost certainly thought of a translator, too,” he writes. For all the supposed incommensurability of languages, we guide poems from one to another every day. Even if one accepts that these are only partial victories, is there another kind? Perhaps the truth is that poetry isn’t as exclusively “poetic” as we often like to pretend, just as the “poetic” part of philosophy is bigger than philosophers sometimes want to think. (When we read Hume, the patient humor is inseparable from the moral point, that skepticism has no need to be hysterical.) Poetry contains as much wisdom as it does word magic: Szymborska in English may be nothing like Szymborska in Polish, but we read her for the good counsel as much as for the choices among words.
Citizens of our languages, we act as citizens do, participating, reforming, accepting the rituals and celebrating their alteration, occasionally even voting for new rules and rulers. No words are entirely untranslatable; none are entirely transparent. A pragmatic view of how words work is the only view of them that accounts for our persistent tiny triumphs and sudden comic errors. Sometimes they obscure; sometimes they’re plain; often they fail us. When the wild strawberries are in season, you hope you find the words to get them on your plate. When the beans come, you eat the beans.