Recently, translating a pay piece I didn’t pick, I had what seemed like some new thoughts on an old translator’s conundrum: to alter or not to alter. I tend to take fewer liberties with pieces I don’t pick, because the voice may not speak to me; it’s a glove I don more gingerly. Strange, that interpret should have two meanings and translate only one, and that one imply a one-to-one mapping, as in, “That doesn’t translate.” The professional sense of the terms are still widely confused, and most people (who, on learning what I do for a living, ask “Oh! Have you tried working for the UN?”) still use them interchangeably, but the distinction seems obvious once drawn—translators write (ghostwrite) and interpreters speak (are ventriloquized); both channel, paraphrase, yet I found myself tempted to tell an editor who asked for sentences somewhat freighted and opaque in an original to be further elucidated, “My job is to translate, not interpret the text.” That is, although the act of translation is one of comprehension, of explaining a text to oneself (indeed, as all reading is the rescuing, mining, or recovery of meaning from a text into one’s own private unarticulated language; to that translation adds periphrasis in a communal tongue) it seems the final translation should not go further in the direction of explication than allowed for in the original.
A literal translation, however awkward, hews closer to the letter (and arguably strays farther from the spirit) than a literal interpretation, which implies not only a conscious reading (an opinion, a decision, a cognition) and a processing, but by implication leeway, the possibility of other interpretations. An interpretation is relative; a translation is (relatively, semantically speaking) absolute. Hence perhaps the adage that if a book is bad it must be the translator’s (not the author’s) fault: a fault paradoxically of insufficient interpretation.