“The Babel fish is small, yellow and leech-like, and probably the oddest thing in the Universe…if you stick a Babel fish in your ear you can instantly understand anything said to you in any form of language.” – Douglas Adams
“How many years since I heard this ancient tongue/A loving murmur, understood by none.” – Nautilius Pompilius, “Atlantis”
Who among us has not wished for a magical earworm with which to understand the friends, family and coworkers who (allegedly) share our own language, much less those who speak a different one? The funny thing about the Babel fish is that it isn’t really a magical animal—it’s a perfectly evolved, algorithm-like creature that converts sound waves into electricity, and then back into intelligible speech. It even logically disproves God’s existence! Although written B.I. (Before Internet), the purely material nature of Adams’ Babel fish still presents an interesting question for those of us who translate: Is the goal of translation a perfect transposition of words from one context to another, or does the allusion to Babel suggest a more fundamental desire for a restoration of human communion, lost or squandered across time and space? This may seem like a distinction without a difference (after all, why can’t it be both?), but maybe the better question is, is a restoration of common understanding possible through purely material, mechanical means—which seems to be the promise of algorithms like Google Translate?
While the closest thing we currently have to the Babel fish resides in our pockets rather than in our ears, it is true that, for the most part, Google Translate works pretty well when you find yourself in a new country and you need to read a menu. I still reach for my hardback dictionary, though, when sitting down to translate something. Why do I still reach for my Oxford doorstopper rather than for the (usually) highly accurate internet options available? The problem, at least for me, continues to be that no word resides in a grammatically pristine isolation chamber. Words may be legible to an algorithm either individually or in small strings, but language is inhabited and haunted by history, religion, and even geography. At another level, does the desire for instant translation itself demand a flattening of real differences between humans: Should a Japanese man who has lived in Tokyo all his life be able to immediately understand a Swahili-speaking woman from Mozambique? Or, would this actually be a false kind of understanding—a pure linguistic exercise completely abstracted from time, place, and experience, and thus devoid of the meanings that lie behind the bare word?
Perhaps this is just pure Luddite thinking on my part (this was never actually intended to be a screed against machine learning), and maybe the machines will eventually replace human translators. If we are stuck, at least for now, with the often wobbly efforts of humans, what is to be done? If translation is a process of understanding, or even unveiling, then human involvement in that process makes it by definition unreliable and prone to manipulation. You have to read a text through someone else’s eyes, which requires a good deal of trust: How do you know that my understanding is correct? Would you trust me more or less if I told you that the song lyric translated at the top of this post is not a perfect one-to-one reproduction, and that I fooled around with the syntax to try to preserve some of the original rhyme and meter? Although I have very little practical advice, the thoughts below are some preliminary attempts at balancing veracity and meaning in translation.
How visible is the translator? You may be relying on the translator’s guidance, but you shouldn’t have to think about him very often. Clunky or anachronistic vocabulary (both of the too archaic and too trendy varieties) is often a tell. Nineteenth century translations of Latin and Greek works are infamous in part because of their overreliance on dictionaries that provided highly stylized recommendations for entire chunks of text. At the same time, if your translation of Cicero reads like it came off of Twitter…yikes, chief!
Is the translation trying to be completely exact? One translator of classical poetry identifies a “spirit behind the word and dominating the word which eludes the ‘faithful’ translator….” Now, the only thing worse than reading a translation is reading a review of a translation, BUT if a particular translator is promising exact faithfulness, it’s a decent bet that they miss the “dominating spirit.” This, unfortunately, is something of a trend at the moment, so seeking out older translations published between the 1920s and the 1970s can help to correct for this sometimes mechanistic approach.
While I don’t really recommend reading reviews of translations (at that point, you might as well go ahead and just learn the language—it’s more rewarding and less painful), it helps to know something of the author. Dostoevsky, for instance, is famous for a hectic, feverish writing style, as if he couldn’t get what was in his head down onto paper fast enough. If the translation of Crime and Punishment in your hands doesn’t make you slightly nervous at the breakneck pace, put it down.
Finally, to return to the beginning of this series of ramblings, do you understand? Not: Can you read the words on the page? Rather, have you entered into someone else’s world? Have they been unveiled to you in some way? If not, then it really is just babbling.
Courtney Ring is a recently-turned freelance editor and translator, specializing in academic publishing and the Russian language. More of her writing and translation can be found in the near future at clringediting.com. In her spare time, she enjoys embroidery, listening to music, and long walks in the woods.