So here's a really weird question that involves an artificial conception of the Japanese language in a Spanish speaker's dream in my webcomic's universe. I'm not sure it belongs in this forum section, but here we go:
Here's the challenge I'm facing:
- My comic takes place in a future where Japan no longer exists and the language itself is functionally dead. Additionally, information about the language was intentionally scrubbed from the historical record -- think of a digital version of the destruction of the Library of Alexandria.
- Japanese cultural influences are still around (old videos of Manga, Anime, etc.,) and linguists study the language, but no native speakers.
- The image below is part of an artificial dream that was created by a corporation for a Latin American audience, and presumably not an audience that generally understood Japanese. When I made this page 10 years ago, I basically wrote variations of "this is a placeholder," "this text is ornamental," something to that effect and ran it through Google translate to Japanese. I didn't want to use the typical fake-Japanese gibberish that I used to hear in Latin American TV growing up, because that seemed awful and racist, but I don't feel like the solution I used optimal either.
- Technology has grown a lot from a decade ago, and with artificial neural networks, I am wondering whether it would make more sense for an artificial dream made by a corporation to have an awkwardly reconstructed version of Japanese. Or, would that be too much expense and bother for a corporation to do something like that?
- If I do decide to go with the gibberish (cringe,) should I approach it similarly to Simglish and go full total nonsense, or should I just fill in random words.
This is a complicated problem that I need a solution for because I have several pages of this, and I am not happy with them all. I am willing to re-letter it and improve it. Any and all advice or suggestions are desperately needed.
If they have recorded native speakers from games and cartoons then the language is going to be reconstructed very accurately. People managed functional reconstructions based on clay tablets and some of those languages don't even have surviving close relatives.
Whether the corporate would care about reconstructing it accurately for a commercial audience or not is a different question.
PixelSploiting:Whether the corporate would care about reconstructing it accurately for a commercial audience or not is a different question.
That is precisely what's driving me nuts!
As a linguist, I know that the language can be reconstructed, even if it is done so imperfectly. Using an example that's very pertinent to my own experience, the Taino language, was considered functionaly extinct for a long time (outside of small pockets within the Dominican Republic.) However, enough people spoke pidgins and dialects of Taino, and it is similar enough to Arawak, that it has been reconstructed to the point where it is much closer to the original language. However, with Spanish grammar sneaking and words from African languages influencing the language, there needed to be some clean-up before being taught in schools in a somewhat standardized manner, which still acknowledged the fact that living languages evolve.
I was also thinking that a reconstructed Japanese language extracted from fragmentary media could probably be reconstructed very accurately, even if it has some English language influence (specifically anime made for US audiences.) It would probably not be 100% normal everyday speech, as speech in scripts for film, video games and animation, don't always sound like natural language. But, unless a corporation has some cheap way for artificial neural networks to spit out accurate Japanese reconstructions for artificial dreams, I'm not entirely sure that they would sink the money into careful linguistic reconstruction.
Location:Sol 3, Milky Way Galaxy, Laniakea Supercluster
The risk with writing text in a language you don't know is always that you can't really tell how good it is. Deliberately doing it badly also comes with a risk - your audience can't easily distinguish "this translation is garbage because in-story it's supposed to be garbage in-universe" from "this translation is garbage because the author can't tell and doesn't care", which you really want to avoid. Sometimes you can clarify, either by having a character comment on it, or by having a character say something perfectly in the language, but I don't think either of those work in your case.
Since this is a visual medium, you could do something that couldn't work in reality. I think I would write the Spanish dialogue using a Japanese writing system (in addition to the subtitles in Latin letters). So "Hola, soy Aina" would be written 「ほら、そゆ・あいな」, pronounced "hora, soyu aina". This would let most readers understand that the text is being "spoken" in Japanese, but actual Japanese speakers get a fun easter egg. It's very clearly an artistic choice (since the Spanish text is right there, and presumably anyone reading your comic can at least read Latin script), so it would avoid the issue of "is this deliberately bad?" in another way. Plus, I'm kind of a sucker for writing languages in the wrong writing system.
Transcription into kana is fairly simple - the only part computers struggle with is figuring out how the English word is supposed to be pronounced, because English orthography is a dumpster fire. If it's not too much text, and if you know some basic phonology, you can easily do it yourself, and not worry about computers messing it up.
GMan003:I think I would write the Spanish dialogue using a Japanese writing system (in addition to the subtitles in Latin letters). So "Hola, soy Aina" would be written 「ほら、そゆ・あいな」, pronounced "hora, soyu aina". This would let most readers understand that the text is being "spoken" in Japanese, but actual Japanese speakers get a fun easter egg. It's very clearly an artistic choice (since the Spanish text is right there, and presumably anyone reading your comic can at least read Latin script), so it would avoid the issue of "is this deliberately bad?" in another way. Plus, I'm kind of a sucker for writing languages in the wrong writing system.
That is a freaking BRILLIANT, yet simple solution that I can handle that myself. This seems totally within the realm of what a shitty corporation would do, and it's culturally consistent with the dumpster fire that is the Latin American media handling of non-Spanish speaking characters. Thank you so much! It's going to seem a bit weird for the English version of my comic, but since it's established very early on that the story takes place in Mexico, and the previous page is always in Spanish, I can just leave the Kana translation as is.
I apologize for waking this thread, as I am well aware that Zombie threads are a no-no, but I wanted to thank everyone who helped me with the Japanese language question. I decided to go with Katakana and a name converter and I came up with some changes for the the pages starting here. Again thanks for the help!
I will be closing this thread shortly, so it doesn't come back to life. Cheers!