I love translation (I mean, obviously, considering my career goals). I don’t want to admit it has any problems; there are too many common oppositions to it already out there. I always want to defend it with all my heart against those who say translations are inherently flawed. After all, you can’t have a translation that’s both beautiful and faithful, can you? It’s gotta be one or the other. Actually, I don’t believe that at all. I believe you can have both, and that’s the philosophy that guides me as I translate. Some people dismiss translations as ultimately imperfect no matter what you do–and maybe use that as license to get lazy–but I believe a translation can surpass that, and that you can create a definitive, Platonic, close-to-perfect translation. (Some would say this is naiveté. I will still strive for it.) To that end I try very hard to achieve a balance between beauty and accuracy, in search of the close-to-perfect translation, and if you must lean more to one side than the other, I think that it’s better to have a slightly less beautiful, more accurate translation than one that’s slightly less accurate, more beautiful. There are those who would disagree with me on that point. But while I prefer both, I would choose accuracy as most important any day.
But there is a big issue with how translations are published. Accordingly with my beliefs outlined above, most of the time I disagree with translations made by other people. If I’m just reading it, I might think, “Oh, what good writing, what good English–this flows so well. This person did a great job!” and judge it solely on those merits. And I wouldn’t be alone in that thought. But then… I’ll compare against the original. And just about inevitably I’ll discover all kinds of things left out, embellished needlessly, mistranslated, and so on. Sometimes I can even tell when those things exist just from reading the English. I’m even noticing it in 1Q84, a major work translated 2/3 by a Harvard professor of Japanese! (One example, a character says something like “I felt like children in a Dickens novel, abandoned” and I believe it should have been a singular child–Dickens only wrote about one abandoned child per novel pretty much. In Japanese you often have to figure out based on context if a word is singular or plural).
But here’s the thing… unless you know the source language, you’re never going to discover that bad translation. Knowing only English, you will judge the English, as it’s all you can judge. And if the English sounds good, you will assume–and have no choice but to assume–that the translation must be solid as well. But that doesn’t mean it is! Often it’s not–at all! But well written English is covering that up.
Likewise, a translation could be extremely accurate and faithful, but if it’s not written well, people will judge it a bad translation, and call it clunky and so on. (I’m guilty of this too–I dislike stilted, dry academic translations; the ones of Kokoro and Snow Country stand out to me, though I believe Snow Country contains mistranslations as well. I haven’t re-read Kokoro since learning Japanese so maybe it does too.) That’s absolutely a problem as well; beauty is still important in translations. You simply have to try very hard to get both.
But the real problem is that the publishing house’s editors, same as the readers, are incapable of judging a translation’s accuracy–only its beauty. I saw this all the time at TOKYOPOP. There, freelance translators were contracted to complete a translation of a volume of manga and create a script. Then, a freelance rewriter came along to polish up that English from the translation and make it sound natural. So step 1 was translation (accuracy), step 2 was localization (beauty). (However, this second step wasn’t spelled out to the translators as clearly as it should have been, as many of them–at least the ones I worked with–took it upon themselves to localize as well, when they should have provided nothing more or less than a neat, non-embellished, accurate and faithful translation.)
And that’s all fine if the translator can be relied upon to produce a good–beautiful and accurate–translation. But that’s not always the case. Even professional, experienced translators can screw up–but if there’s no one to catch those mistakes, that’s not good. As the rewriter looking over that manga translation, I’d routinely uncover tons of mistranslations. Tons and tons. But as I was often the only other one in this entire process who knew Japanese, no one would have pointed them out if not for me! Of course, a few TOKYOPOP editors were more or less fluent in Japanese. But it wasn’t a job requirement; it wasn’t a necessity, so many weren’t and things just went over their heads. I firmly believe that if your regular job involves working with translations of [x] language, you must be fluent in [x] language in order to do that job. To this day I don’t understand why this isn’t valued more; why people like me, who can do a QC on these translations, are not in higher demand. I believe anyone who wants to be or is already a manga editor should be fluent or highly advanced in Japanese. It’s just an absolute necessity in my mind.
I imagine it’s the same with publishers of translations of Japanese-language novels like 1Q84, or really any publisher of translations in any format (but it’s probably worse if the publisher puts out translations coming from multiple source languages; the editors can’t be expected to know all those). The only one who’s in charge of the actual translating is the translator, and if you don’t know the language (and most editors don’t), you have NO WAY of knowing if their translation is actually good; if their language skills are actually that good. No way! No way of knowing. That is frightening. All editors can do is judge the quality of the English, which doesn’t equate to the quality of the actual translation. Even if the translators initially pass a translating test–judged by a speaker of that language–before coming on board, they can still make mistakes in the actual job, and no one will be there to catch them. Certainly not in the case of a 900-page behemoth like 1Q84.
So I think one of the worst aspects of the published translation industry is this complete lack of quality control. There is no one to check over a translation once it’s made, and to me that just seems totally reckless and dangerous.
The only solution I can think of to this problem, even though it’s a small one, is to offer myself as a recourse: as someone who will translate with an eye to both beauty and accuracy. At my current job I regularly call owners and founders of various types of businesses and ask them questions so I can write about them; I usually ask about how they decided to get into their industry and start their business. One frequent answer: “I wanted to provide a level of service and quality that I wasn’t seeing in my industry at the time. The lack of it was frustrating to me as a regular customer in search of those services, so I decided to step up and offer it myself.” In other words: if you want something done right, you have to do it yourself.
I can’t transform the industry all on my own but I hope to at least make some small contribution to the world of beautiful, accurate translations. Unfortunately, translating is also very subjective; what seems both beautiful and accurate to me may not seem so to someone else. That’s probably the reason why I’m not able to ever think someone else’s translation–even one made by pros and experts–is as good as one I could make myself. Which is a problem–I’d love to have a translator I wholeheartedly admire, but I currently don’t. I’m just too picky. But hopefully that’s a quality that will make me a good pro translator myself.
In any case, if you don’t know the source language you’re not capable of judging a translation as “good” just based on how well its English flows.