Having been in the online translation scene for the past 5 years, I've seen many, many translations, and since the start of MangaHelpers, I've seen the incredibly rapid birth and growth of speed translations and speed scanlations.
As a result of manga becoming easily obtainable, and more schools taking up Japanese in their curriculum, there have been quite a few translators out there with a year of Japanese under their belt who believed they have sufficient skills to, well, translate.
Aside from not having enough experience with the language to translate it properly, a lot of translators (myself included) tend to have problems with English. Grammar being wrong, punctuation horrible, and sentences not sounding fluid at all.
From what was sparked in a channel called #Translators in irchighway by a member stating how he hates MH, to me contacting a friend at Nihongoresources
(who I've talked to, on and off, about the type of translations that come out of MangaHelpers), I've decided MH needs to play a more active role in the process of scanlations from translations to the finished product: the scans. And considering the size and the amount of translations that are posted here daily, it can only be done with your help!
So I'd like to start this off with an article (beware, it's quite a bit long. But I really want to stress that ALL Translators and Scanlators read it
. Members and Leechers should read it, too, if you want to help improve the quality of what you're reading now~). For those who don't have time to read it, we'll be going over several
parts of this article in the next few weeks, so stay tuned :).
With that said, here's Pomax's
article on "The Art of Translation" originally posted at http://pomax.livejournal.com
THE ART OF TRANSLATION.
Much as anime fansubbing underwent a shift from fans working for fans (ye olde VHS days), to fans working for fun (the rm, asf and vivo days), to fans resisting the commercialisation (the speedsubbing and DVD ripping days), to fans not getting in the way of commercial release (the present), the manga translation scene seems to be going through the same progression. Of course, the phases overlap a little, so it's hard to say exactly where we are, but the prolific releasing and the associated quality seems to suggest we're in an interesting combination of the last three phases.
With scanlation having moved away from tankoubon ('anthologies' or 'collections') to individual chapters from the syndicating monthlies or biweeklies, the speed scanlation groups have started moving in, sacrificing quality for quantity, and doing something probably no one had realised was possible: shaped the expectations of the audience, not just for scanlations, but for official licensed products too.
Hopefully as someone with an understanding of Japanese, as well as language in general, my arguments will carry a bit more weight than a random user complaining on a manga forum, but it's time an uncomfortable truth was discussed: the kind of English used in modern scanlations is, on average, not English in the slightest. With the emergence of speed scanners the quality of translation has gone down so much that we're seeing this nonsense English back in actual official, licensed, English speaking country distributed products that we are charged money for to own - made possible because we're collectively responsible for indoctrinating the scanlation audience with nonsense English, breeding an expectation that is preventing the medium from being taken serious. After all, if you can't bother to translate something to real English, clearly it's not worth paying attention to.
SO WHERE DOES IT GO WRONG?
Essentially, it's going wrong with enthusiast manga translators not translating from Japanese to English. Allow me to explain.
It is relatively easy to translate words. If we don't know how to translate certain words, we rely on dictionaries or native speakers to help us find alternative meanings that we might use in order to translate them within the setting of a sentence, and we'll typically end up with a coherent sentence.
But the story in a manga isn't composed of individual sentences. No running text is. Instead, it consists of sentences linked by what is known as discourse, and translating is about preserving discourse. You cannot translate any other way without doing a bad job at translating.
"Discourse" is the collection of underlying thoughts and intentions that motivated someone to write a text in the first place. Discourse is what makes a collection of sentences feel like "a text", instead of loose sentences. Put concisely, discourse is what an author "wanted to say", rather than what the words that he or she used to say it. Crucially, it is the presence or absence of identifiable discourse that determines whether or not a translator did a good job, or simply isn't a translator.
And here we hit a snag.
THE PROBLEM OF BREEDING EXPECTATION
Continuous use of certain practices sets up a 'common ground' for contributors and consumers in a particular setting. That sounds complicated, but what it really means is that if for instance a group of people says "desu" at the end of every sentence, after a while people stop noticing how odd it is, people in the group stop questioning it, and when people outside the group remark on it, it's defended as a group-defining aspect. Americans have this with gun ownership, Canadians have it with using "eh", and the scanlation scene has it with completely nonsensical English...
If you consider the English used in scanlation to be just fine and you're wondering at this point what the actual problem is, allow me to go into more detail. In the scanlation world, translation started out as a hackjob: people wanted to be able to read manga, so suboptimal translations were just fine, people just wanted to be able to enjoy manga a little, and poor English was accepted as long as the reader could roughly understand what was going on.
And that was fine.
However, these days manga translation is big business. There are plenty of scanlation groups that will point out they are 'better' than other groups, and that they take pride in the quality of their work, and it is at this point that things become a problem: if you start to take pride in your work enough to assert that you are better than others, perhaps it is time to let go of the argument that "it's good enough for the audience" and actually start doing what you're pretending you're doing, and start translating Japanese to English.
Examples of nonsensical English are legion. Read any chapter of translated manga and you're likely to run into anything ranging from minor harassments of the English language to full blown rape homicide, leaving behind a trail of bleeding interpunction and a chalk outline of what was once language.
SO WHERE DO THE TRANSLATIONS GO WRONG?
Actually, in many places. There are a number of things that a translator has to keep in mind, and try to avoid at all cost:
- Idiomatic constructions should not be translated literally
- Sentences should not be translated "per bubble", since the bubbles work for how Japanese sentences are chopped up, not how English sentences are chopped up.
- Words that are already entailed by translated words in a sentence should be left out.
- Verb mood, tense and even polarity may need to be changed to express the same grammatical construction.
Japanese and English are two different languages. This sounds elementary, but all too often people do not seem to realise this when they translate every word in a Japanese sentence and consider the translation done, even though this leaves the English translation with too many words; a sentence such as "even if by doing so we succeed, it will become difficult" is not an unusual sentence in a translated manga, even if it's not an English sentence at all.
This might sound weird to some people. Even if you don't read scanlations at all, you might be wondering how that statement could be true - after all, it uses English words, and you can figure out what it means.
However -- and this is a big however -- proper English does not need "figuring out". Real, natural English sentences are understood exactly because they are natural language. They don't have to be figured out before you know what they mean, because the way the words are arranged immediately reveals how you should interpret them. Arranging them differently, because the original phrase was not in English, just makes for unnatural and potentially highly convoluted sentences; the entire book "Everything is Illuminated" by Jonathan Safran Foer (2002) used the idea of writing out a non-English speaker's thoughts and sayings using English words and it was a mind-splitting pain to read.
Now, in the world of novels, this work could succeed because editors demand high quality English. You're not going to get a lemon fanfic or blog post published purely on "what the author has to say", he or she better damn well have some writing skills, being able to instill a sense of discourse in the reader. This also means that when a novel is intentionally written in broken English, this has to be done for a good reason, and Jonathan Safran Foer found one such reason.
In scanlation, it seems things are exactly the opposite; the quality demand by group leaders for their translators seems to be 'as low as still allows figuring out by the reader', and reasons to actually produce good translations are few and far between.
Let us look at that earlier sentence, "even if by doing so we succeed, it will become difficult", in more detail. We can pretend it is a translation for the Japanese phrase しても難しくなるぜ. The English sentence would be a near literal translation of this, but a good translator doesn't actually translate the sentence, he or she translates what the sentence means, and he or she translates them by using words and constructions in the target language that make sense in that language, not ones that map closest to the source language.
In general, an English sentence isn't English because it uses English words, it is an English sentence because it uses the right words and the right grammatical constructions, where "right" is a subjective term, but comes close to "what a professional editor would not consider wrong". So in this particular case, we can identify at least four problems with the aforementioned sentence "even if by doing so we succeed, it will become difficult":
1) it has too many words,
2) the words are in the wrong order,
3) a pronoun is used where it shouldn't be, and
4) the wrong verb tense is used.
In proper English, the sentence should have actually been: "Even if we succeed, things will be difficult".
Why? Because the act of 'succeeding' entails performing an action in English. We don't need to, and in fact shouldn't, add "by doing so" to the sentence. Second, "even if by doing so, we succeed" is simply the wrong word order. In Japanese the operative information comes later in the sentence, in English it comes earlier in the sentence: "we succeed by doing so" is natural, "by doing so, we succeed" is far less so. Now, mind, that is not to say you cannot use "by doing so, we succeed", but placing operative information later in the sentence is formal English. Unless the material you're translating is formal Japanese, carrying the ordering over to the translation is being a bad translator. That could simply be because no one ever told you, but that doesn't excuse the end result. It just means you can learn and improve.
Third, the pronoun 'it'... the dreaded pronoun 'it'. There's of course nothing wrong with this pronoun itself, but translators abuse the hell out of it. In English, we use 'it' for contextual back reference, and then only if we do not have an idiomatic expression to fall back on - in this case, we use "things" instead of "it", because that's what we use. Context in Japanese doesn't warrant using 'it' in English at every turn. Think before you commit to a translation.
The same goes for the last mistake. While the Japanese sentence used a verb form that translates to "will become", the English "will be" entails becoming, so using "things will be difficult" is perfect; "things will become difficult" on the other hand uses a double future tense.
WHAT'S IN IT FOR ME?
I can understand you may read this and wondered why it would be worth to analyse every sentence in this way, when the meaning is sort of there in the original sentence, but the point is not that you - as a reader - can figure it out, the point concerns translators - my problem is with translators who don't bother to learn how to translate. A proper translator learns to think about translations by analysing what he or she produces:
Is what I translated English, if I ignore what the Japanese line reads?
- Did I put in words that are already entailed by others, if I ignore what the Japanese line reads?
- Did I leave off words that are required by idiomatic expressions, if I ignore what the Japanese line reads?
- Are my words in the right order, if I ignore what the Japanese line reads?
- Should I use a pronoun, or should I use actual nouns for contextual omissions, if I ignore what the Japanese line reads?
- Am I using verb mood, tense and polarity correctly in English, if I ignore what the Japanese line reads?
- Does this sentence link up to the previous sentence, or are they unconnected, if I ignore what the Japanese line reads?
The quintessential requirement here is "if I ignore what the Japanese line read". A good translator will produce a text that stands on its own, without needing to be justified with "I know it sounds quirky, but otherwise it wouldn't say what the Japanese line said". A translation takes an idea from one language, and expresses it in another language. If the idea cannot be expressed in the target language, and this happens, then deal with it by coming up with an alternative text, expressing the same idea, rather than using unnatural constructions just to mimic a language that the audience doesn't understand.
Of course, you can rely on an editor or a proofer (technically these are different things, an editor checking whether discourse isn't broken, and a proofer making sure there are no typographical mistakes), but the problem starts at the translator. A good editor can turn a bad translation into a good English story, but it won't be a translation anymore. The best editor in the world cannot turn a bad translation into a good translation. They can only turn bad English into good English, so as a translator it is your responsibility to make sure that your translation is as good an English text as you can get it in the first place. Editors aren't there to catch your mistakes, that's what other translators are for.
WHY SHOULD I CARE?
I know many people will go "we're only doing this for fun, why are you harping on doing it right", to which I have only one reply really... just because you're doing it for fun, doesn't mean you're not f*cking it up for everyone else.
That's right, I said it.
Don't get me wrong, doing something "just for fun" is fine, as long as you don't pretend that once you get a real translation job, you'll change your method of work - because you won't.
There are, certainly, good translators out there, and there have been good scanlation translators who've gone on to get a job in the translation field, and deliver good work there, too. I salute these people, they do their profession justice. But, and this is a universal truth not just for translators, but any artist: they didn't become good translators after getting a real job, they were good translators to begin with.
Just like musicians become good at their music before the industry lets them on board, and visual artists must become good at their art before being able to sell their art, if you're a bad artist -- regardless the subject -- then getting paid for your art doesn't magically make you put all the things you know good artists do into practice. You'll still be a bad artist. Only now you're getting paid for it, even though you don't deserve it, and that means that suddenly you won't listen to critique anymore because the paycheck validates the quality of your work.
It does not.
In fact, and this is the really sad part, people tend to use their paycheck as a justification for delivering poor quality. Rather than working towards getting as good as possible at their art, they now only work towards keeping that paycheck, no longer being interested in improving or not, as long as they get paid.
And this is the reality of the situation: bad scanlation translators who find their way to a real manga translation company such as Viz or Dark Horse will end up creating just as bad translations as they did before, and guess what? Those translations make it to print. Yes, even commercial manga translators can produce nonsense translations, because they were bad translators before they got hired, and they didn't improve after they got hired because no one tells them to.
The convoluted part of this writing is that these people can be hired because the standard of the audience has been primed to be low. Scanlation has made the readers accept that nonsense English is not just acceptable, but is the golden standard, and licensed manga translators are hired at this competency level - because scanlations primed the audience to expect low quality, companies can get away with hiring people on the basis that "the audience won't mind, they're used to this anyway".
So the bottom line is that because we're translating "just for fun", we've shaped an industry where mediocrity is rewarded by us, with our own money.
Translation is an art. You can do it just for fun, or you can do it to make a living, but either do it right or don't do it for an audience. The internet certainly allows you to publish anything you think others might like, but while art critiquing in general on the internet has become mostly a standard (image boards are prolific, youtube is filled with performers learning from their audience's comments), for translation this seems to not have taken off (hopefully, yet).
It's easy to spot a hand that looks unnatural in a drawing, or spot a series of chords that "sound wrong" in a song, but reading a translation, comparing it to the original, is a lot of work - the unfortunate reality of the matter is that just because no one comments on how bad your translation is, that doesn't mean they didn't notice. It just means you, and everyone before you, bred an audience that just doesn't care anymore. You are, essentially, not taken serious, because "it doesn't matter".
So please, be self-critical. Listen to anyone who points out mistakes in your translations, because that is the only way you're going to learn from your mistakes. Unlike physical activities, you don't get better at translating by just doing it a lot, you need that feedback to know what you're doing wrong, or you'll just keep making the same mistakes over and over again. Without correction, every repeated bad translation will make you more likely to use that same translation a next time, entrenching the mistakes. Seek out corrections, because waiting for people to comment on them isn't going to help you improve.
Read. Read English and Japanese books, know what sounds right in English, and what sounds right in Japanese, and most crucially, realise where the differences are. Both are just languages to express similar ideas, so how can you best express these ideas? Just because one language uses an "if" construction doesn't mean you need to find an "if" construction in the other language when the thought behind the expression isn't actually a true conditional - translation is an art for a reason: it's hard.
As parting words I suppose there's not much left to say... Mangahelpers.com is a great start, offering a place for translators to get feedback on their translation, and working with publishers towards a situation where scanlation and licensed publishing can coexist, but you need to put in the effort yourself.
Do we really consider manga to be so uninteresting a product that we simply don't care about the quality we, and by extension the industry, delivers? Keep expectations low, and anything mediocre is considered good?
Thank you for reading.
- Mike "Pomax" Kamermans, of nihongoresources.com
And again, a thanks for reading all of this from me too! Quite long, but very important if you ask me! And thanks Pomax for taking the time to write this ;).
Just in case you're wondering who Pomax is, well you can read about his achievements here
. Or just know from me, that he's a guy that's constantly helping people understand Japanese with his site not to mention writing a 285 page book on Japanese language, grammar and syntax
... for fun 0.o.