What makes a good translation?
What makes a good (or even great) literary translation? At first glance it seems an easy question. Of course, a good translation is one that accurately renders into another language the original work.
Or is it?
Firstly, accuracy is something of a myth. It assumes that each individual word, with all the freight it carries (associations and implications, echoes of Latin or Greek, its appearance in famous phrases or proverbs) has an exact equivalent in another language. In fact, words in translation are rarely the same down to the molecular detail. Most fall somewhere on a scale from ‘roughly equivalent’ to ‘not really equivalent at all, but the closest you’ll get’. To take a random example, ‘sausage’ in English can cover a range of Polish products which range from an English-style uncooked pork sausage to a smoked salami-style. Not to mention the difficulty of translating expressions like ‘silly sausage’ or ‘selling the sizzle not the sausage’ (or less polite phrases!). In those cases a translator would need to look for a local equivalent of the phrase which most likely would have no relationship to sausages at all. Sitting down with the dictionary and translating word for word will get you an accurate translation, but not a successful one. Sometimes you need to take a different road to arrive in the same place.
So should a good translation be judged on whether it does end up in that ‘same place’ – whether the translator has successfully adapted the spirit and meaning of the text, rather than the author's every word? It seems more plausible, but once again it slips away once you dig into the question of how to assess such a thing. I can imagine many different opinions on what exactly is the spirit and meaning of a text, and the more complex it is, the more possible interpretations. It’s certainly a higher-risk approach, and it can fail spectacularly if the translator's view of the text doesn't line up with that of the reader.
What about judging a translation as an independent work, forgetting all questions of accuracy and forgetting that it’s a translation at all? It's not necessarily a bad idea. Awkward phrasing and sloppy wording should not be excused just because a book is translated from another language. So it is important that a translation reads smoothly and well, but this approach can also be taken to extremes. I read recently about a French edition of John Dickson Carr’s The Emperor’s Snuff Box in which the translator simply made up a new first paragraph, which hardly seems fair to Carr. This kind of thing was more common in the past, but most people now would agree that there is a limit to the creativity that should be applied by translators, even if they believe they are respecting the spirit of the original text or providing helpful context to readers.
Like most things, if there is an answer it is likely to fall somewhere in the shades of grey. There is definitely a place for subjectivity and even creativity in translation (relax, translators – Google isn’t coming for your jobs just yet). Accuracy also plays an important role, albeit in reality one that is more subjective than it initially appears. And a translation should also stand on its own two feet and be able to be considered as an independent work. In some ways, assessing if something is a ‘good translation‘ is not so different to evaluating whether you are watching a good or merely average violin solo. You apply a set of recognised criteria, give due respect to the role of experts, consult with your ‘gut feeling’ and….you decide.