I spend most of my working time translating literature from French: fiction, poetry, or nonfiction. We literary translators suffer many indignities. Reviewers often ignore the translation altogether, as if they were reading a novel written in English. They may praise “the style” of a novel written in, say, Hungarian, whereas they’ve read the English version and what they’re praising is what the translator wrote to put it into English. Or they may praise a translation for its “smoothness.” But what if the original was not smooth at all? When translations are publicly evaluated, it is usually by people who don’t know the original language, or not very well, since those who know it well read the text in the original. Translation prizes are often given without considering the relationship between the original language and the English text. A novel in lyrical, rhythmic English may be given a prize by people who have no idea that the original was written in intentionally prosaic, gritty, non-lyrical Estonian. This is, of course, a worst-case scenario, but it is perfectly possible.

The year I was one of the judges for PEN’s translation prize, we gave it to Philip Gabriel’s translation of Murakami’s Kafka on the Shore. I found the novel in English magnificent, but I wanted to know if it really rendered the Japanese, a language I do not read. I was told in no uncertain terms that this was not ALTA, where translations are first evaluated by people who know the source language, but PEN. Now, PEN is a great, worthy organization and I support it with all my heart and wallet. But the misunderstanding of what’s involved in translation, or rather, the ethics of translation—and by accomplished literary translators, no less! stuck in my craw. I went behind their backs and asked an acquaintance who taught Japanese at Berkeley to take a look at the original. He did, and reassured me. I voted for it and everyone else did, too. So PEN got it right—at least that time.

Literary translators need to render the connotations, the tone, the rhythms of the writer while sticking as close as possible to the denotative meaning. Getting that meaning is, as I implied above, part of the ethics of translation. It is part of the unwritten pact between translator and reader. But a literary text is, of course, far more than the simple meaning of its words: it produces a complex esthetic, emotional and intellectual effect on the reader. My goal is to produce a similar effect on readers of English to that which the original French text had on readers of French. I realize this statement is not without problems (which readers, for example?) but I think it’s what most working translators really try to do as they work, as distinguished from what they may say about translation after their work—especially if they’re academics.

Thus, when I translated Jean Guéhenno’s Diary of the Dark Years, the journal a prominent left-wing intellectual and teacher of French classics wrote under German Occupation, I tried to get the English equivalent of his literary, tightly controlled, often eloquent French, so unlike the telegraphic, casual style of most diaries. It is no accident that it was also unlike the vulgarity of the Vichy propaganda in the newspapers and magazines Guéhenno read every day. Deposition, the Occupation diary by the anarchistic novelist Léon Werth, is entirely different, though it has the same contempt for Vichy and hatred of the Nazis. Here I tried to render Werth’s varied, lively style and his terse, ironic, quips in equivalent English. Now I’m working on a collection by a surrealist poet who was officially “excommunicated” from Surrealism at the age of 22 and died from an infected needle at 36.

Producing something close to the effect of French poetry in English obliges the translator to negotiate between the denotative meaning and sound patterns, whether regular and rhymed or irregular and “free.” No sound pattern, no poetry. That’s the true part of Robert Frost’s over-quoted (and misquoted) dictum that poetry is what gets lost in translation: if the sound doesn’t work in English, the poetry disappears. But anyone who has read Ron Padgett’s recent translations of the French Modernist Guillaume Apollinaire, for example, knows that sometimes poetry does not get lost in translation. These English poems let you understand why the French rank Apollinaire with the great poets of their literature.

Rendering Roger Gilbert-Lecomte’s strong, sometimes wild voice in English has its own special problems, but I doubt if they would interest anyone but a fellow translator. That’s why going to ALTA, the conference of the American Literary Translators Association, has been such a pleasure for many years. At last, people who understand what we do! Women and men who are concerned about literary translation, who love it and talk about their practice, the problems involved and suggestions for the best way of handling them, practical advice about publishing, and so on. An added bonus is the non-stop bilingual translation sessions, where you can pop in and hear Chinese or Romanian poetry read in the original and in English any time you like while panels are going on elsewhere, like a three-ring circus. Since literary translation pays next to nothing except for the few folks who’ve managed to get a job translating bestsellers for big publishing houses, most of our members are academics whose day job is teaching at a college or university. MLA would be our natural conference. But I found early on that when you go to MLA, people talk about their latest book or article. At ALTA, you hear enthusiastic recommendations of new translations from literature all over the world, not necessarily their own translations. In MLA panels, they read lectures that one could more easily follow in print. Reading papers has always been discouraged at ALTA: we talk to our colleagues, not at them. Translators are trained listeners and readers, not lecturers; we listen to what the author says. No one reads a piece of writing more attentively than the translator, who weighs every word, every expression, every paragraph. We have to.