Localisation and Translation

[anthony]

There's a lot of controversy around this subject and nothing very intelligent being said.

The recent issue in short. Cabals of transsexuals and other such debris of the western world are deliberately invading the connecting joints between Japanese pop culture and the rest of the world to leave their own distinct marks upon these highly desired and prestigious works. When called out or criticised they alternate between saying that they are perfectly faithful, and saying that being perfectly faithful is actually impossible so changes do not matter.

It's Knights v Trannies, round 500.

As is always the case, the knight's angles of approach are wrong. He doesn't really get art or nice things, but just knows there's a certain class of person he should be angry at. He's right about that much, but wrong about everything else.



Certain egregious cases like this, painting weird western fringe identities on top fictional characters, creates a false impression of the problem. I don't read Japanese, but I'm taking the translation on the right as an attempt at being barebones literal. If anybody wants to correct please do. But point is that I believe that is the intention of this comparison image (made for non japanese readers). When some outright freak (despite seeing the issue of localisation as ambiguous I do not sympathise with these people at all) does something as obviously extreme and political in intentions to a work it makes one want to take up the reactive contrary position. That people should keep themselves out of localisation. "Just give me it in English".

Attempts at robotic translation would actually be fine I believe, better than the above (certain people even use GPT to translate now). But it would probably be awkward due to a machine's lack of nuance and humanity. Which is to say, things tend to read better when a straight interpretation is filtered through somebody and contains traces of their character. That is the first bare point that I'll make. But not the point.

As a general rule, we're interested in these particular foreign works to experience the work of the original creators, not this new person. It might sound impossible that things could possibly be otherwise, but this has happened. On the extreme end we have something like Ezra Pound's translations of Chinese works. People do not read Chinese translated by Ezra Pound to appreciate the original Chinese works. They read them to appreciate Pound. And the original work as refracted through Pound. There is obviously a very distinct and idiosyncratic element of interpretation in his "translation" work. He is giving you what is unmistakably a new version of something that already existed... And nobody minds. For two reasons of course. One being, that Ezra Pound is very interesting and many people like him, and so will also like impressions of his character. And the other being, that attempts at more or less straight translation exist and are even more accessible than Pound's work.

The current problems in localisation are who is doing the translation, a lack of clarity or consistency with regard to what they believe their job is, and the centralised one-off nature of how these works are distributed, and generally their size, tending to mean that only one localisation will ever be made for any given work.

From this understanding of the issue I think that answers are simple too.

First being more credit for localisers. This will both make their influence understood more clearly, and make them more accountable. And of course when they do a good job, will get good work more credit. I think they should be an upfront credit. Like on the title screen. Capcom's Resident Evil, Presented by [whoever]. Something like that. Japanese Man's Game, as interpreted by [some nerd who is hopefully not trans and insane]. If it's good. You can know why. If it's bad, you know why. Make it known who does these. And make it known that there is always an intermediary influence.

Secondly, clarity on what the job is. I do not mean we pin down one definition. I just think that with a bit of conceptual awareness here we can prevent people from dodging the matter. There is not a right way to do this. One can be a naive orientalist almost making their own work out of a thing like Pound, or one can do a chatGPT impersonation and try to do things direct. Another benefit of pinning names to the front of localised works would be that people could recognise intentions through the people doing the work. If your japanese game mentions a tranny, you can look at the front of the box and, if you know your localisers, make an informed guess as to whether you're reading the japanese man's vision, or some psycho neo-missionary trying to undermine them on purpose.

What I would do to fix the purpose issue, would be to try to get a mission statement out of each localiser, which we can maybe burn down to a word for the stupids. "Personal" localisation, when the translator is deliberately making their own personal impression upon the thing, "straight" for the people who want to be chatGPT and try to leave as little of an intermediary mark as possible. From those two we'd be pretty much set.

And as for the third, this is the most difficult one. Re-localising a big game would be a lot of work. Redubbing an anime. Translating a manga. Or a book. Lots of work to do. Depending on the medium maybe a lot more production in addition to getting the new words down. I believe that smaller and simpler works in this direction being pulled off by fans would be a good start just to send the message that options can be valuable and appreciated, and that there are people who might like something one way that they wouldn't like another, so choices matter for the big time, and maybe multiples could be worth it in some cases. For an example of fan work, certain old video games have "undubs" that you might see around if you're into emulation. People combining Japanese voice acting with english text to create new versions of old games that never existed. I played Evergrace this way.

And that leads into the maybe obvious example of audience choice in localisation. The eternal one. Subs v dubs. Subtitles of course being their own localisation. This argument tends to take on its own knightly character, the focus is on perceived purity when either way you're getting the thing as interpreted by an intermediary. But of course through new performances (and perhaps a more altered script, common argument) you are getting more of the intermediary. So yes, a less pure work. But that doesn't mean worse.



There are some dubs I greatly enjoy. I enjoy them because they are so heavily localised that I can feel a new creative presence asserting itself within the work. A point that never seems to get raised, if that new voice is really good I don't mind. And because this is an anime people only watch over the internet now the original dub and subtitles of the original script are also readily available. I can watch Cyber City Oedo according to its original creators online. Or I can watch Cyber City Oedo as presented by Manga Entertainment. And I prefer the latter. Because Manga Entertainment are cool.

What it all comes back to is associating the localisation process with the individuals behind it, and then judging them individually as good or bad influences. The knight mistake is to act like personal influence is the problem.

Last thought I'll get out for some perspective. I accidentally created a localisation conflict recently and like hearing what people think of it.



Top image is an original, high quality, faithfully scanned panel from the manga Gunslinger Girl. Below it, is the same panel from the first scanlation that I read. The text is slightly different, but far more significantly, it looks strikingly different. The scanning process has crushed the darker sheeds into deep blacks, the fine details are rendered harsher. The lines look thicker. The clean, digital quality is lost to an ink-like blurring. Guess which one I'd rather read?

A Japanese friend who recommended the manga to me was not pleased with my choice, but my answer to him is my answer on all questions of localisation. I am aware of it. I am aware that I am not reading the pure Gunslinger Girl. I am reading a Gunslinger Girl which has been manipulated and distorted by the different hands it has passed through. I like the distortions as distortions. The creative intentions of the artist are something I appreciate. But I am also fascinated by the manipulation of those same intentions. There is more to appreciate now. And if I feel like at any point there is so much that I'm losing some vital element of the original, I can always go back.

Does this illuminate the point at all? I like sharing the comparison images but hope it was somewhat useful.

To summarise.



The problem with left is not that the localisation is unfaithful. It's that it is idiotic.

[Guest]

Yes some localizations are poorly done. For JRPGs it is not such a big deal because the writing is not very important (you can understand any of the good ones without a translation, even going back to PSX days.) I don't especially care if localizers put in memes, and one of the better current JRPGs has memes inserted. The Japanese do the same in their own works, the memes are (obviously) just different.
Adding pointless moralization is more annoying, but I haven't seen this much. However, I pre-emptively skip many things so it is possible I just have not seen them.
A funny aside: One group who is championing AI translations and calling localizations pozzed has actually put forward a pozzed localization of their own. The most amusing part of this is that the series itself will make their "translation" even more ridiculous once the next segment is released in a form they can digest. One should remember that these groups are full of...very low-level Japanese readers.

[Mladorossi88]

There are very few instances I can think of in which the localization has actually been good. One of these was translating "baba" as "hag" rather than "old lady" (as some other translations did.)

[JohnTrent]

As a general rule, we're interested in these particular foreign works to experience the work of the original creators, not this new person. It might sound impossible that things could possibly be otherwise, but this has happened. On the extreme end we have something like Ezra Pound's translations of Chinese works. People do not read Chinese translated by Ezra Pound to appreciate the original Chinese works. They read them to appreciate Pound. And the original work as refracted through Pound. There is obviously a very distinct and idiosyncratic element of interpretation in his "translation" work. He is giving you what is unmistakably a new version of something that already existed... And nobody minds. For two reasons of course. One being, that Ezra Pound is very interesting and many people like him, and so will also like impressions of his character. And the other being, that attempts at more or less straight translation exist and are even more accessible than Pound's work.

The current problems in localisation are who is doing the translation, a lack of clarity or consistency with regard to what they believe their job is, and the centralised one-off nature of how these works are distributed, and generally their size, tending to mean that only one localisation will ever be made for any given work.

From this understanding of the issue I think that answers are simple too.

First being more credit for localisers. This will both make their influence understood more clearly, and make them more accountable. And of course when they do a good job, will get good work more credit. I think they should be an upfront credit. Like on the title screen. Capcom's Resident Evil, Presented by [whoever]. Something like that. Japanese Man's Game, as interpreted by [some nerd who is hopefully not trans and insane]. If it's good. You can know why. If it's bad, you know why. Make it known who does these. And make it known that there is always an intermediary influence.

Enjoying this discussion, and though I do not have much to add, I will limit my comments to this section of the post. Vincent Gallo once remarked that the use of film credits today is utterly pointless and bloated: he mentioned that the sole purpose of film credits is a mnemonic one, where it presents the names of those who played an important role in the film's creation. The film aficionado who is viewing these credits is able to gather the details of the most essential characters who sustained the film's vision. With the film credits seen today, this factor is ignored, becoming instead a list of anyone who happened to be employed during the production of the film. It now rewards the inessential. I would say this has become a major Western tendency in relation to art, which is why it took roughly until the '60s before auteur theory became a standard in American film criticism. In the case of what Gallo would view as a traditional film credit, there is still a specifically American difficulty in interpreting the names: who is the one that really spearheads the project and brings it to fruition, and who had contributed in a secondary fashion to this guiding figure? To someone who regards films as a technical and collaborative feat only, there is not much rationale in place to justify who exists as the most influential part. This may explain why American filmgoers do not really consider the names — the fixed idea of movie franchises and IPs imprinting on American minds more than the men who helped to shape the art. For instance, a person who goes to see the Sequel Trilogy of Star Wars that does not consciously see the difference between Disney Star Wars and the films created by George Lucas. To them, it's a seamless whole.

[Yumetaro]

Discotek Media is a "fan-driven" company that licenses and sells anime, including an English dub of popular 2007 shoujo romantic comedy Lovely Complex. This dub has become quite controversial on X after Brendan Blaber, aka JelloApocalypse, wrote a public Patreon post talking about his experience co-writing the script for the dub as well as directing lead actors for a third of the season. Let's see what he thinks about what he made.

"Frankly, having the dub come out this good in the end is a little frustrating. Anyone who watched this show as a kid and has fond memories of it and revisits the dub will think "Wow! It's just as good as I remembered!"

No it isn't. We made it good. This show SUCKED!!!"

If you asked a chud to describe an English localizer, his description would look a lot like Brendan. Libtarded, smug, and contemptuous of the material that he's working with as well as the person who made it. His explanation of his work and ability to fit neatly into stereotypes presents a unique opportunity to better understand how English localizers are approaching their work, how they understand art, and what kind of people they are generally. Translation and interpretation leaves traces of their character that we can people watch through, and I'd be interested to see what the Amarna Forum thinks of this. Here's some things I focused on.

The most important thing here is that Brendan doesn't understand Lovely Complex at all. It's not that I'm really familiar with this IP or anyone who worked on the anime and can tell you everything he doesn't understand (I read one chapter of the manga and decided it wasn't for me), but he keeps saying over and over again that he doesn't understand why anything in Lovely Complex is the way it is.

"I did my absolute best to make sure Risa and Otani had enough positive characteristics that them falling for each other made some kind of sense."

"Fun fact! We canonically wrote and directed Risa as a psychopath who doesn't understand empathy and Otani being in a cycle of abuse without realizing it. Because that's the only way their highly inconsistent actions make any kind of sense!"

His failure to comprehend why people like this pretty popular anime is never a chance to learn or reflect. This dub only exists out of the good will of some CEO towards the original sub, and he can't even give a token explanation as to why that good will even exists. I would say that this is a reflection of his lack of media literacy, or even empathy. But Brendan isn't going to learn from that, instead it's the mangaka Aya Nakahara who lacks empathy, a person that Brendan never even refers to by name.

"This is one of those series that was written by a very strange person, kinda like 50 Shades of Grey, and it just gets stranger and stranger the longer you look into it. The way all the characters talk and interact is wrong. The way Risa is painted as the good guy despite makes exclusively bad decisions makes everything feel like it's written by a woman with "I HATE drama!!!" in her Twitter bio and then five separate callout tweets right below it. You get the sense the author sees the world the same way Risa Koizumi does, and Risa Koizumi as a character only makes sense if she is a psychopath who does not understand human empathy."

Riveting.

I don't think that the fact that Aya Nakahara is never mentioned by name is a minor detail. I think Brendan might see Lovely Complex as an amalgamation of "intents" that are divorced from the person who made it. As an example, a good amount of this Patreon post is handwringing about transphobia, but in a comment below he declares "See the thing is we didn't change the intent of any of the scenes. Even the original Seiko stuff was, in the end, supposed to be *pro-trans*. It was just terribly done 2007 pro-trans writing." We can ignore the fact even if one of the characters could recite the YWNBAW speech while staring at the camera they still would've papered over it, because though Brendan's argument for his changes is post hoc it still says something about him. Lovely Complex isn't a specific person's understanding of transgenderism in a specific time or place. Once the character Seiko was introduced, in Brendan's world it either has an intent of "pro-trans" or "anti-trans" and you can correct it to more accurately reflect that intent. Presumably Lovely Complex's major failing is that it fails to fit into Brendan's understanding of how to reflect a depersonalized intent of "romantic comedy."

It's popular to observe that the people involved in these localizations are acting politically, but it's worth pointing out that almost none of the people involved in this project stood to gain just about anything from the many hours they put into it.

"Discotek did not want to order a dub because dubs are typically very expensive, so Sound Cadence endeavored to make the dub while spending as little money as possible. This production was almost entirely asking people to cash in favors. The leads both played their parts for free, and Marissa was set to write and direct the adaptation, also for free"

"I thought this would be a good opportunity to get some experience writing localization scripts in case I ever wanted to write a dub (maybe Bomberman Jetters or something, I dunno), so I volunteered to help out."

If you don't understand Japanese the people standing between you and the nice Japanese thing you want to see are receiving marginally more money than a janny. Many of the people who are aware of their existence wish they would go away and most couldn't care less. I think this is why AI is really dangerous in their line of work. Their employment is already reliant on being the best optimization of time and money over quality. If an LLM can get close enough no one will care when they're gone.

Perhaps some of you might think it's unfair of me to use this idiot who will never work again as a skeleton key. I'd like to point out that many people who are sympathetic to libtard localization or are employed in that area have commented on JelloApocalypse's post. Though every response is negative, I'm not aware of a single post from these types that explains an objection to what he did or how he felt about it. I have seen many that are upset that he gave the chuds a victory by writing all of this where the chuds can see it, one even going as far as to explicitly state that you should keep your thoughts where ever they won't be able to look at it.





I'd like to point out one last thing. Though I have seen many posts from libtards critical of Brendan for giving knights license to continue complaining online by acting exactly like them but in public, I haven't seen any of them notice that he said the worst thing he could possibly say. There are only a few knights who have noticed this. This is from his comments.

"Of note, most anime distributors *do* tend to ask that you don't *advertise* changes like making things less transphobic. This is because large swaths of the anime community are racist/sexist/trasphobic/etc and will descend upon a company's social media page for weeks for the strangest things. Discotek added a content warning on one of their discs and got flak for that, so they're understandably a little skittish now. Funi/Crunchy works like this too.

tl;dr, most anime distributors don't care if you make things nicer or more gay or add more representation. They just ask that you don't flaunt the changes or use buzzwords."

In other words, despite having tons of freedom with what they're doing, translators and localizers are being told to avoid blatantly walking into areas that will get the knights to notice and write mean things about the company on social media. It turns out that the knights are getting concessions and all they have to do to win more is to get more of them to complain louder and more frequently. Thanks for sharing Brendan.

[anthony]

Interesting. This is basically what I mean when I say localisation should be public facing and prominent. Every dub, hell maybe even sub, should come with a kind of mission statement or presentation or whatever. You know how so many books have intros or forewords by the author or someone presenting a new edition, I think works that have been localised should have that as a matter of course. Not invasively, we don't need a star wars style text scroll before the anime starts. But it should be somewhere accessible so we can look. It's nice for the people who do care, and for those who might not care, they know what's for them. Anybody working with good intentions should be for this. The only reason to oppose the idea is if you want to subvert what other people want.

These people don't want to be known as Libtard-Occupied-Localisations. They want to take things over quietly and present as JUST HOW THINGS ARE.

Our man Brendan clearly had ideas driving his work here. But to his credit it sounds like he was trying to make the thing better. Sure, his taste and sensibilities might be gay and raped, but the fact he wasn't lying makes it okay to me. He saw his job as delivering a desirable product to people, and he thought he could make it more appealing, sensible, enjoyable. That's fine. He wasn't trying to mess with your good times. He was trying to give you better times. And he admits that he doesn't understand the original appeal and that it is something different to what he gave you. This is Lovely Complex according to Brendan.

Arguably some of what he did came out as moral or spiritual censorship. The things he sought to fix were arguably just honestly presented human nature which frightened him, but he thought it would frighten you too. He was trying to help. From the sounds of it he would not freak out and call the police if you told him you liked the weird crazy neurotic bitch characterisations of the original, he would just consider your taste bizarre. Maybe I'm wrong, I haven't looked at this guy beyond what is presented in this thread, but it sounds like he doesn't believe his job is to morally police humanity. Just his taste happens to take the shape of moral policing in this case.

This era of adjusting for taste seems to have been mostly harmless. Again, I like the Cyber City Oedo dub. Changing a "what do you know?" to "Go fuck yourself" is really funny, and I can tell it's probably the localisers having fun. I'm having fun too, so good job. Oh and memorable example someone told me about recently. Apparently in Ghibli's 'The Cat Returns' the original script has the protagonist sternly admonished for laziness. In the dub her lesson is being told to be herself.

Again, this is arguably a kind of moral censorship, Japanese sensibilities being declared UnAmerican or whatever, but not intentionally. They just didn't want a stern note in what is meant to be an enjoyable experience about walking talking magic cats. The fact it's basically a fairly tale I think makes this difference easier to frame and accept. Like historic fairy tales there's drift as different cultures refract the parts according to taste and sensibilities. If anything cultural spinning is still weaker than it's ever been since communication technology is so sharp now. Tranny localisation commissars are a last gasp of resistance against this. Pure unfiltered access to another culture with alien values and sensibilities to their own is breaking through and they can't stop it. They're so shrill and extreme because the stakes are high, the forces in play inevitable, and they have everything to lose. They are the dad from footloose.

[calico]

End of Evangelion is clearly superior in sub.

[anthony]

End of Evangelion is clearly superior in sub.

I don't speak Japanese, but certain Japanese dubs feel more appropriate than their English alternatives in a way that strikes me as obvious.

Akira is one of my favourite movies. I first watched the newer English dub, which has very "voice actor" dubbing. Then I watched the older dub, which stars Cam Clarke, a voice actor, but a guy who talks like a human in all his roles, and lastly I watched the Japanese one. I would rank them in reverse order. The Japanese one is the best because the characters who are supposed to be Japanese street punks actually sound like Japanese street punks. Again, I don't speak Japanese, but I've seen a lot of Japanese movies. In the original Japanese the characters in Akira sound like the characters in Blue Spring. They speak like Japanese teenagers.

The old English dub understood this and tried to convey it in a localised way. Cam Clarke plays Kaneda as a kind of edgy surfer dude. There's a lot of goofy and awkward stuff in this dub but there's a rawness to Clarke's delivery that feels equivalent to the Japanese vision. Johnny Yong Bosch plays Kaneda in the newer dub. He sounds far more professional, but the cleaner sound and more performative approach really undermine the presence of the character and the tone of the film in my opinion. I do think on some level the newer sound recording tools feel mismatched with the standard of animation and when the film was made. Akira could be called something like the most advanced analog animation of all time. Pairing it with this really clean digital audio performed by a guy who sounds like he learned to talk by watching newgrounds flash animations feels really wrong.

As for Evangelion, since the main cast are so iconic and distinct, one could say that their voices are a distinct part of that. Japanese Shinji, Asuka, and Rei are wholes. How they sound is a big part of the impression they make on us, even if we don't understand the words. And I do think I like those Japanese voices more in that case too.

[Promise-Ring]

I've been playing Higurashi recently and have found many of the differences between the original Japanese and the English translation very interesting. In particular just a single short phrase that seems to be regarded as the most "iconic" from the game.



The part this release of the game thinks of as "iconic" actually happens right after this screenshot, with a big "LIES!" flashing in and out of the background ("嘘だッ" in the Japanese). The point is "嘘だッ", "噓だよ”, and "嘘だ" would all be machine translated here as the finer sounding "That's a lie", while the official translation opts for "Liar" and "LIES" at more extreme moments.

The subtitles for the anime appear to use all 3.



My first impression of "LIES" in the VN is that it's what a western cartoon with a "dark edge, dulled by its quirky sense of humor" would do. Which turned out to accurate.



In the mind of a retarded American there's something crude and tasteless you can read in their reading of the word "LIES". It's hard to put into words so just look at the video above, that's what they see when a Higurashi character says "LIES" and it's not so different from what a current day localization-mafia member might see when looking at just about any Japanese work.



Anything Anime is a great example of how awful localizer-vision is (just listen to any Anime dub that isn't ancient) but Higurashi is the best example I can think of considering we have an official adaptation of how retards see it (Doki Doki Literature club), but now isn't the time to get into that.

Obviously the localization in the above "Unicorn Overlord" example is terrible, but "LIES" isn't necessarily worse even. If we aren't trying to declare one universally better than the other, then we can even interpret that "LIES" as being a better the earlier the point in time the translated occurred is. 



I can imagine the kind of quaint old internet person that made these banners translating 噓だ as "LIES" and having it be more "accurate" than any other translation, that heart and Ryukishi07's are close to the same point. But that was then and we are now, the average white person has gotten less refined each year since Ryukishi07 typed 嘘だ in 2002, and the forever unchanging machine translated "That's a lie" remains there with him while LIES has carried the baggage of a decayed language and culture to the present day. I'm not going to say "I'm glad it's LIES/That's a lie" because this isn't a fucking Tier list, but LIES is definitively at this point in time more interesting.

[anthony]

I'm disgusted by the Unicorn Overlord localisation because it's disgusting, not because of the attitude towards the original. If Orson Welles found the original lacking and imposed his own spin upon the whole thing that he considered superior that would probably be really, really cool. But Orson Welles is not in charge of localising this kind of thing. Instead we get a bunch of gross, fat, D&D Americans.

When I read or hear any part of the Unicorn Overlord localisation I imagine the face of Patrick Rothfuss slowly being imposed over the image first as a kind of mirage, before solidifying and eventually taking up the entire screen.

[anthony]

I'd personally be opposed to Orson Welles translating Unicorn Overlord by inserting his own dialogue wherever he pleased. Maybe if he adapted it and choose to do things his own way, that would be fine. This is essentially what Ridley Scott's done for a lot of his work, and it's all brilliant. I think that a translation specifically should try to be faithful (not necessarily accurate) to the original.

Translation is adaptation. If there's one point I'm trying to make in this thread it's that there's no pure ideal, and that purity shouldn't be an ideal.