The new forums will be named Coin Return (based on the most recent vote)! You can check on the status and timeline of the transition to the new forums here.
The Guiding Principles and New Rules document is now in effect.
Translation theory and ethics
Nostregar
Registered User regular
Registered User regular
Something not many people think about on a daily basis, but which happens to be my area of specialty and study in my East Asian Studies degree. I'm going to be going on to get a PHD in EAS with this as my main focus, so I thought it would be fun to come on here and talk to you all about it since you're generally intelligent people. I don't get to talk to many people about this subject, so hopefully this thread will actually help me revise my views in a way I haven't been able to so far.
For the purposes of this thread, I'm going to be focusing on East Asian languages, my experience being with Japanese. I only say this because the issues that arise when translating those languages are very different from those encountered when working with European languages.
Most people assume that translation is pretty normal and don't think about it. You see a book on a shelf that says it was translated from another language and the natural assumption is that it's pretty much the exact same as the original, just in your language.
I contend that translation, as a process, has some inherent issues that make it a generally flawed thing and which needs to be seriously reconsidered.
My first problem is that translated literature, film, and whatever else give us a window into another culture (or so we think) and so any issue in the translation will necessarily skew our view of that culture. This is why the other points are important - they are minor taken by themselves, but as a whole skew our picture of other cultures. If translated materials were not so important to how we perceive the rest of the world, it wouldn't be as big a problem.
Problem two: that information is necessarily lost in the translation process. Words have certain connotations and cultural meanings associated with them which are necessarily lost when translated because, well, it's not the same word or culture any more. The example that comes to mind here is the English translation of Murakami's "The Wind-Up Bird Chronicles" (originally Japanese, obviously). The book is written in first-person narrative style. In the Japanese version, the main character refers to himself as "boku" which is translated as "I" in English. This loses a whole lot of the meaning of that word though. Japanese has three words that males can use for "I" (and the narrator is a male). Watashi, the standard one, watakushi, a more formal version, and boku, an informal and somewhat childish one. The narrator using the word "boku" suggests to the reader a sense of childishness which really impacts the story very strongly (as anyone who has read the book can probably attest to). "I" does not carry any of that. This is an example of "lost information".
Problem three: Similar to two, but slightly different. Words do not "mean" other words. "Boku" does not mean "I". They are equivalent in many ways, but "mean" implies that they are somehow intrinsically connected, which is not the case. A good example here is the word "aoi" in Japanese. The example I use comes from Ezra Pound's translation of the poem "The River Merchant's Wife" by the Chinese poet Li Bai. It was first translated from Chinese to Japanese, and then Pound translated Japanese to English. I'm just focusing on the second translation.
A line in Pound's translation reads: "You walked about my seat, playing with blue plums." The word translated as "blue" here was "aoi" in Japanese. Aoi can be translated as either blue OR green, but like the word "green" in English can be taken to connote unripeness (as would be appropriate in this case, given the rest of the poem). Pound took the word "aoi" to "mean" the color, not considering the connotations. Similar loss occurs when, in English, the politeness levels and status markers present in nearly every word of Japanese are lost, losing a major piece of information from the original. Even if both words correspond to the same action (run, drink, etc) they do not MEAN the same thing because of the politeness markers.
Problem the fourth: Orientalism. This is a concept that most people are unfamiliar with, and rightfully so since it isn't really taught except in the EAS programs. It is essentially the idea that the people of the Occident (Western cultures such as US, England, Spain, France, so on) do not actually recognize the Orient (Japan, China, etc) but rather their idea of the Orient. It's very difficult to paraphrase, but it kind of says that we have taken their culture and possessed it and altered it to make it our own construct with no regard for the original. How we think of Japan is not how Japan actually is, but we think we know how they should be better than they do. Clearly this is more true of some people than others. If you're interested in knowing more about it beyond my shitty brief, read Orientalism by Edward Said.
In translation, I take Orientalism to be in the process of translation itself. Because information (as previously shown) is discarded, somebody must be doing the discarding. In essence, the person is deciding what is important or not important - how to translate "aoi", for example, means excluding some of the meaning. The translator decides what portions of the words and text need to be presented in the target language (for me, English). This would not be a problem if it wasn't being presented as what the original said. It isn't. It's an interpretation. It is the person filtering the work through the lens of the Western mind. It is, in a very real way, taking the original text and making it ours (culturally) without regard for the differences. That is basically the definition of Orientalism.
I know that is my most contentious and also most poorly worded claim, so feel free to ask for clarification if I did a shitty job describing it.
Problem five (and final) : ethical questions. How do you translate a direct quotation and continue to call it a direct quotation? It isn't what was said because it has been changed. How do you call a work a translation if it has been altered to make sense to the new audience? Do we need to assume alteration during translation? Most people I know don't assume that, but I and some others do. Which is the more common view?
I have a few ideas about solutions, but I want to see what people say about all of this first. I'm NOT saying translation should not occur, just that it needs to be done differently.
Anyway, what do you all think?
For the purposes of this thread, I'm going to be focusing on East Asian languages, my experience being with Japanese. I only say this because the issues that arise when translating those languages are very different from those encountered when working with European languages.
Most people assume that translation is pretty normal and don't think about it. You see a book on a shelf that says it was translated from another language and the natural assumption is that it's pretty much the exact same as the original, just in your language.
I contend that translation, as a process, has some inherent issues that make it a generally flawed thing and which needs to be seriously reconsidered.
My first problem is that translated literature, film, and whatever else give us a window into another culture (or so we think) and so any issue in the translation will necessarily skew our view of that culture. This is why the other points are important - they are minor taken by themselves, but as a whole skew our picture of other cultures. If translated materials were not so important to how we perceive the rest of the world, it wouldn't be as big a problem.
Problem two: that information is necessarily lost in the translation process. Words have certain connotations and cultural meanings associated with them which are necessarily lost when translated because, well, it's not the same word or culture any more. The example that comes to mind here is the English translation of Murakami's "The Wind-Up Bird Chronicles" (originally Japanese, obviously). The book is written in first-person narrative style. In the Japanese version, the main character refers to himself as "boku" which is translated as "I" in English. This loses a whole lot of the meaning of that word though. Japanese has three words that males can use for "I" (and the narrator is a male). Watashi, the standard one, watakushi, a more formal version, and boku, an informal and somewhat childish one. The narrator using the word "boku" suggests to the reader a sense of childishness which really impacts the story very strongly (as anyone who has read the book can probably attest to). "I" does not carry any of that. This is an example of "lost information".
Problem three: Similar to two, but slightly different. Words do not "mean" other words. "Boku" does not mean "I". They are equivalent in many ways, but "mean" implies that they are somehow intrinsically connected, which is not the case. A good example here is the word "aoi" in Japanese. The example I use comes from Ezra Pound's translation of the poem "The River Merchant's Wife" by the Chinese poet Li Bai. It was first translated from Chinese to Japanese, and then Pound translated Japanese to English. I'm just focusing on the second translation.
A line in Pound's translation reads: "You walked about my seat, playing with blue plums." The word translated as "blue" here was "aoi" in Japanese. Aoi can be translated as either blue OR green, but like the word "green" in English can be taken to connote unripeness (as would be appropriate in this case, given the rest of the poem). Pound took the word "aoi" to "mean" the color, not considering the connotations. Similar loss occurs when, in English, the politeness levels and status markers present in nearly every word of Japanese are lost, losing a major piece of information from the original. Even if both words correspond to the same action (run, drink, etc) they do not MEAN the same thing because of the politeness markers.
Problem the fourth: Orientalism. This is a concept that most people are unfamiliar with, and rightfully so since it isn't really taught except in the EAS programs. It is essentially the idea that the people of the Occident (Western cultures such as US, England, Spain, France, so on) do not actually recognize the Orient (Japan, China, etc) but rather their idea of the Orient. It's very difficult to paraphrase, but it kind of says that we have taken their culture and possessed it and altered it to make it our own construct with no regard for the original. How we think of Japan is not how Japan actually is, but we think we know how they should be better than they do. Clearly this is more true of some people than others. If you're interested in knowing more about it beyond my shitty brief, read Orientalism by Edward Said.
In translation, I take Orientalism to be in the process of translation itself. Because information (as previously shown) is discarded, somebody must be doing the discarding. In essence, the person is deciding what is important or not important - how to translate "aoi", for example, means excluding some of the meaning. The translator decides what portions of the words and text need to be presented in the target language (for me, English). This would not be a problem if it wasn't being presented as what the original said. It isn't. It's an interpretation. It is the person filtering the work through the lens of the Western mind. It is, in a very real way, taking the original text and making it ours (culturally) without regard for the differences. That is basically the definition of Orientalism.
I know that is my most contentious and also most poorly worded claim, so feel free to ask for clarification if I did a shitty job describing it.
Problem five (and final) : ethical questions. How do you translate a direct quotation and continue to call it a direct quotation? It isn't what was said because it has been changed. How do you call a work a translation if it has been altered to make sense to the new audience? Do we need to assume alteration during translation? Most people I know don't assume that, but I and some others do. Which is the more common view?
I have a few ideas about solutions, but I want to see what people say about all of this first. I'm NOT saying translation should not occur, just that it needs to be done differently.
Anyway, what do you all think?
Nostregar on
0
Posts
Some subtlety will always be lost.
Whatever. There's so much English stuff that you're not missing much.
This is a big thing in my field (Classics), especially given that most students in the UK actually do whole Classics degrees now where all the texts are in translation. I looked at aspects of this for my undergrad thesis and as part of my Masters.
I'll go through the questions properly later, probably after dinner (me hungy).
One thing that pisses me off: transliteration from non-Romanic languages with non-alphabetic symbols. You can't write Koran, you can't even write Quran. You have to write Qur'ān. It's not Shiva, or Siva, it's Siva with a little carrot over the S. Fuck that. Unless you are already a linguistic expert and know how those symbols are pronounced, they do nothing except confuse you and clutter up the page. I accept that the voice inside my head when I read the translated words is not going to match up to the voice inside an Arabic- or Sanskrit-speaker's head reading the original text. It's not worth ruining my font.
It depends on what you mean by "really really really understand". If you mean "laugh at language/culture specific jokes", sure... but that's more an involuntary response. I wouldn't call that "understanding". I don't think any meaning is lost with a sufficient translation. Perhaps your meaning is that someone can't appreciate the humor and cultural background of a work at an unconscious level...?
Same thing happens in French. I know Québec-French movies have to be subtitled for France-French audiences. On the flip side, a lot of France expressions make no sense in Québec. My friends and I once got a France version of a popular Québec humourist's album, and we actually needed to get a French girl to sit down and explain what the fuck the jokes were supposed to mean.
So I do wonder to what extent translation actually obscures the broad, important meanings of text. Little things, sure. And sometimes those little things end up important (a counterpoint to what I said from the Bible: the very beginning is usually translated "In the beginning, God created the heavens and the earth," but a lot of scholars nowadays say it should be translated "In the beginning, when God created the heavens and the earth." The first implies creation ex-nihilo (God bringing all of creation into existence from nothing), but the second implies that stuff already existed when God began "creating." The second is also consistent with cultural norms at the time, which understood divine creation as an act similar to sculpting—you organize shapeless matter into a functional form, but you don't actually create the matter itself.)
(But see, this is all stuff you can learn from learning about ancient Mesopotamian and reading footnotes! You don't actually need to learn ancient Hebrew.)
To make a comparison, some fan-created subtitles for anime shows will include notes on their choice of translation for words or concepts that carry connotations that might be lost in an english translation. I'd like to see something like that for books as well. My edition of the Wind-Up Bird Chronicle didn't have anything like that, just the translated text.
Q. Why are penguins' shops so busy ?
A. Because the fish fillet.
...Because the fish fill-ay?
I do think translation, in especially complex prose, definitely causes a loss of nuance. It's kind of like going to the Opera and reading a playbill with an English translation while the Italian next to you is weeping.
So... *shrug*
Also, if you think Japanese only has three words for "I", then you're about... I don't know... ten or so short.
True to an extent, but there are ways to mitigate it.
Re: BobCesca - Yeah, I've read a lot of the stuff. I had to do a presentation on this topic at an EAS conference this past spring, so I read up on my shit
But I think that this mentality is a problem. Why accept crappy translations? I think there are better ways to do it, or even if there aren't we should at least be trying to improve. Accepting it and saying "eh, whatever" is not a great approach to the problem. We need to worry about this precisely because translation is so important.
Re: CheerfulBear - Sure have. Actually quoted that essay a couple times in the aforementioned presentation. Really great essay.
Absolutely. Like I said, I was just using the Japanese language issues as an example. You're totally correct, language issues exist between any two cultures.
That's my meaning. A good translation is just gonna have a fuckton of footnotes. If the footnotes are good, the meaning is all there. It's just presented in a dry way that isn't necessarily going to provoke the same reaction that an intimate understanding of the language would bring about.
There was a few paragraphs explaining how they'd gone about translating the idea of the sort of woman who drives a Morris Minor into Dutch, where the cars had never been imported and ended up changing it into something like the sort of woman who drives an old bike (and in turn creating a new pun the author had never thought of based around the word for 'Granny' and 'Bike'). Seems a fascinating subject even before you start adding such wildly seperate cultures as european and far east asia.
Names are also something that seem quite interesting as well, given how in the UK more often that not, your name is pretty much chosen from a list of suitable sounds that are recognised as names despite the fact that they are just (bastardised versions of) words in another language. Yet I've some friends who's parents don't speak english as a first language and are named using words from that same language (Gujarti I think). Seems odd to think of someone just called something rather than a Name, and wierder still that we never translate people's names (though I can understand why, given how I'm primarily thinking of the Bible and most european Names are versions of the ones in there).
there will be much posting, I suspect, but probably over the weekend.
I also don't really know if there is an objective standard for translating stuff. Take Plato. I understand his Greek was quite complex. Moreover, the original had no paragraphs or punctuation. Which is a better translation—one that is absolutely faithful to the original (requiring reams of footnotes to explain the context of "arete" and shit), or one that gets the basic messages across in a user-friendly, easy-to-read way? For my purposes, as a reader casually interested in philosophy, I think the latter. But for someone deeply interested in Plato and the intricacies of Greek philosophical thought, the former is probably better. (Another example would be ancient mythological texts like the Bible or the Epic of Gilgamesh—which way is better? They both have their purposes.)
Sometimes you read foreign texts because you want to experience the depths of their beauty and complexity. Other times you read them just because you want to get the gist of what they're saying. I think translators (and publishers) should be forthright about which "style" of translation they employed, and write about their method—but I'm not sure you can objectively say one is better than the other.
Menus at vietnamese restaurants scare me. It's like someone drank a fifth of whiskey and opened up charmap.exe.
Some, certainly, and your example makes a good case for it. However, things like my "boku" example do require some knowledge of the language.
Footnotes are really kind of a translator's shame. I know many people would disagree with me on this, but most of the stuff I've read tends to agree. At the point where you're using footnotes you've removed the reader from the actual text and are now just providing factual information. It's better than nothing in some cases, but they're really not a great solution.
I'm well aware. I was trying to keep most of that brief since I know the majority of people do not care about the precise number of words for given things in a language they neither know nor care about. Those three were the major ones that made sense for the example.
I think what you say at the end there is more important than you assume. Provoking the same reaction is one of the main goals of a translation. I know some people like footnotes, I just happen to dislike them for the reasons I said up a bit.
Re: BobCesca - Right on.
To add some more stuff to the discussion, one of the other major issue translating Japanese to English is how much information is carried in verbs in Japanese as compared to the noun-centricity of English. There's absolutely nothing you can do other than translate as best you can, but you definitely lose...something...during that shift.
I've come up with a couple solutions to these translation issues, none of which are perfect but which I think help.
1.) Presenting two translations at a time. Have a native English speaker translate the Japanese to English and then a native Japanese speaker do the same work. The two will almost certainly pick different words because of different understandings of the text, and the reader who reads both side by side will be able to use those as a kind of bracket and fill in the missing meaning in the middle. It's a hassle to read, but it would definitely be one solution.
2.) Present the English translation alongside the original in Japanese. As in, left page is in Japanese, right page in English. The person will almost always not be able to read the left page but it serves as a constant reminder that you're missing something because you look at the Japanese and have no idea what it says. I think the psychological impact of this is huge.
Paul, the author of Galatians, is writing to his cult because another group of Christians are trying to convince them that they need to circumsize themselves to be saved. Paul is of the opinion, however, that circumcision does not matter at all in salvation.
That last line? That's a dick joke. Paul is basically saying, "I wish those who unsettle you (i.e. the pro-circumcision Christians) would go all the way with their circumsizing and cut their whole penis off!"
Other translations:
"As for those agitators, I wish they would go the whole way and emasculate themselves!" (NIV)
"I would they were even cut off which trouble you." (KJV)
"I could wish that those who trouble you would even cut themselves off!" (NKJV)
"I would that they that unsettle you would even go beyond circumcision." (NAV)
None of them really effectively get across the point that Paul isn't being randomly vindictive, he's being snarky.
But to what extent does this really matter? If you're just reading Galatians to get a general sense of Paul's theology, is the joke really that important? I don't think so. That said, I'd be in favor of a longer line here that actually explained the joke—accurate transliteration be damned—just so the line doesn't come across as so random and hard to understand.
I always write it as Koran for that reason-- even though I know it's not pronounced anything like that, it just looks ridiculous to stick a bunch of Qs and apostrophes in there that no one is going to know how to say anyway.
Which doesn't mean I don't get a little twitch when I hear somebody say "the core-Anne".
I was young at the time, but this had zero impact me when I read such a Bible at synagogue. I actually remember the opposite happening: I glazed over the English text and failed to internalize the meaning of the words I could actually read because I just assumed "oh it's some weird ancient nonsense language that probably doesn't make any sense."
I basically agree with this. Different types of translation serve different purposes and what's important is finding the one that fits your needs.
As someone with a very basic knowledge of Japanese language and culture, I find stuff which keeps honorifics and provides footnotes explaining stuff like what form of I is being used more interesting.
But not always. And if I'm gonna read some novel like Cantide (that was originally in French, wasn't it?) I would just want straight english because I don't know shit about French.
The problem I always have is that I'm a literary-person, so if I was reading this letter, the joke would be very important. That's the thing, whether one should just be reading something for the "general sense" of the history, philosophy, whatever, or should also be able to appreciate the literary qualities of the text, such as, for example this joke. That's one the big things which translators have to consider, and it's difficult (for me) to make a definitive decision either way.
It's just really difficult sometimes to get everything from one language into another, and sacrifices have to be made. I just don't have to like it :P
This is actually true of the Koran-- the text itself is considered to be the true unaltered word of God, but only in Arabic. I mean, duh, right? A translation isn't scripture. It's not even a translation, it's an interpretation. If your English-language Koran doesn't have Arabic on the facing page, it's no better than a picture-book bible.
This is also why when someone quotes the Koran to you about such-and-such, they are lying.
Shaka, when the walls fell.
According to Islam, the Quran, in Arabic, is the unadultered word of God.
According to Islam, Muhammad rode up into the sky on a magical flying donkey, and Alexander the Great was a Muslim.
Which is to say, why should any rational person give a shit what Islam says about translation. The same exact issues apply to translating the Quran as apply to any ancient poetic text. Yes, you're not going to preserve the rhythm of the original Quran—you're also not going to preserve the rhythm of the Sanskrit in the Mahabharata or the Ramayana (texts that also internally claim to have been transcribed directly from the mouth of deities).
But you can still quote from them. The Quran really does say women are worth half as much as witnesses compared to men (2:282). It doesn't say it with that exact rhythm or whatever, but the relevant information is the same.
Edit: just for the record, the worst Quran translation offense is when you translate it to sound like the King James Bible. I mean what the fuck.
Oh god. Don't even go there.
Guess it depends on the person. If I had them presented side by side (left page, right page) I wouldn't have much trouble following it.
Again, guess it depends on the person. It doesn't have that effect on me. Another question to ask though is would it have been different if you were reading the text by choice, not because you had to? Also, "ancient foreign languages don't make sense" is a mentality that would ruin really any attempt at translation that includes reference to the original language, even if it's in footnotes. Right? I know this is how many younger people think (hopefully you've revised that opinion since), but that doesn't make it good.
I think this problem could be very easily solved by just changing the label. If it's not going to make the efforts to preserve the original and opts instead to simply provide an English version, call it "[Translator's name]'s interpretation of [Book title] by [author]". Longer and kind of annoying? Sure. But it makes it clear that it's not a true translation, so all is well.
It would be very important to me as well. I'm honestly kind of surprised that some people just don't care and find it unimportant - if you're knowingly removing pieces of the text, however minor, you can't really call it a translation.
I'm a translator at the moment and it's a really big issue for me.
Yeah, that's basically the same issue I had with translating direct quotations from what someone said. It is no longer what they said.
Well, I'm assuming no one here considers the Koran to be factual (no offense intended, obviously). What's relevant is the beliefs of a Muslim-- and in strict terms, a Muslim doesn't agree that the English represents the word of God. That's what I'm getting at.
Also, I would also prefer references to the original language that are in English letters and are not transliterated with funky symbols. For example, Genesis 1, God creates a "dome" or a "firmament" that he calls Sky. That should have a footnote explaining that the original Hebrew word is raqiya and the root of the word is related to hammering out metal. It should not spell raqiya like râq'iy`àh. Nor should it spell raqiya like רקיע because fuck that, how do you even pronounce that?
You make a pretty fair point about the Hebrew, and I'm willing to concede that that might be a good solution in that case. Japanese actually has a distinct problem in that regard, though, since a portion of the meaning is actually in how the Japanese is written.
In brief, Japanese has three "scripts". Kanji, what most people think of as "Chinese characters" generally represent the meaning-bearing portion of a word. Hiragana, a phonetic script, is used to write conjugated endings onto the kanji, and sometimes to write entire words (also grammatical particles, etc). The third script is Katakana, which is an interesting one. It is a phonetic script like Hiragana but it is used exclusively to write words of foreign origin. It is never used to write native Japanese words, particles, or endings onto words. Seeing a word written in Katakana has a different psychological impact that a word written in hiragana or a combination of kanji/hiragana. So in that sense, the written form of the Japanese has information that a simple transliteration does not.
Kind of an interesting issue, actually, and one of the difficulties of translating Japanese.
Also, I agree, presenting something like that as an interpretation is awesome.
Edit: So that first portion was saying that even just transliterating the Japanese loses something.
Yeah, I suppose I misstated that. You can obviously talk about the content as well as any other translated ancient text. What I meant to say is more that describing what Islam says, or what a Muslim believes, is erroneous, since they explicitly do not believe that the English translation has scriptural validity.
And of course if nothing else, it's a good rule of thumb for the general rules that apply to any ancient translation, that is that they are horrible and should not be trusted.
Huge can of worms.
You want a note on the beliefs of a modern wahabist Muslim, a modern Alevi Muslim, a Muslim contemporary to Muhammed, all three? The interpretations would vary greatly when referring to a passage even in the original Arabic.
Perhaps a better solution, if you want to preserve the levels of formality, would be to use different fonts? Or a different layout on the page? (Of course, this is not to be done lightly; I'd rather read an unnuanced translation than parse a shitty page layout.) Online translations could also do a lot of shit with the Internet.