As far as endeavors go, translation has all of the perils one might fear for and virtually none of the guardrails one might hope for. Reasonable activities—baking a blueberry pie, say—have achievable success criteria and known dangers. You know a good final product when you see it: a golden crust; a tangy, fruity filling; a crowd of dinner guests asking for seconds. You also know when you’ve messed up: the crust is burnt or undercooked and the filling is a gummy layer of mush.
Translation has none of these helpful signals. In fact, in translation all of the dangers are hidden and all of the success criteria are elusive at best. Bad translations are easily produced and just as easily condemned. Good translations, often invisible. The experts, sadly, disagree. When a translation isn’t going right, it can be vexing to try to figure out why: there is no algorithm to debug, no recipe to follow, no baking time to tamper with.
I worked for years in commercial translation in tech—I led localization teams, groups of translators across many languages, who translated software applications and their associated content. It was challenging work, especially since translators were often working without the benefit of context. The word “report,” for example, might have myriad meanings (as a verb, to report a newspaper story or report on a team’s performance; as a noun, an informational report, or a direct report working under a manager). These ambiguities intersected in unexpected ways and compounded each other. As a translator once pointed out to me, the phrase “raise an issue” could mean two opposite things: cause a problem or flag a problem that needs to be solved.
The silver lining of this particular sort of ambiguity was that we were all collaborating against it. The copywriters sought to write unambiguous copy. If phrasing was particularly confusing, we could edit the source text to be more straightforward. And as a general rule, if something was ambiguous in English, we would make quite sure to iron out that ambiguity in the target language.
In literary translation, on the other hand, the source text is immutable and often deliberately ambiguous. Rather than flattening the text, the translator’s job is to capture its multiple aspects, however impossible that might be given the constraints of the target language.
I have recently started an MFA program in poetry, and have been torturing myself with some poetry translation on the side. Translation is like High Intensity Training for a poet: not only do you have to understand the nuances of two languages, but you also have to, by some kind of alchemy, negotiate the safe passage from one to the other. Every final product is a failure in some sense, but having made your trip to the underworld—and having returned more or less unscathed—you can see the structure of the cosmos slightly more clearly. As David Bellos wrote in his book Is that a fish in your ear?, “Think of a great poet, and you’ve almost certainly thought of a translator, too.”
Poets thrive in the world of semantic ambiguity, a thing to be tamed, trained, or charmed, depending on the situation. In many cases, overworking the target language to capture all of the subtlety of the original is a fool’s errand. Like the stewards of an old-growth forest, all a translator can do is work with the perversity of natural law to cultivate whatever life-forms the circumstances demand. With enough respect, the ambiguity of the target language achieves for the reader something like the ambiguity of the source language. The rainforest will never be the taiga, but a similar story rustles in both canopies.