Quote:
Originally Posted by kyuven
*sigh* this discussion is making me want to fix the weaknesses in my translation skills.
Namely, my lack of...eh, i've said what I lack enough by this point...
Curse you American community college and high school Japanese system that doesn't teach students how to look up radicals ;_;
|
I can help there.
See, every time I translate something, I make it clear that I don't speak Japanese. I don't. Honestly. I taught myself how to read Hiragana and Katakana and I've memorized a few Kanji that are commonly repeated, then I learned participles and conjugations and off I went translating.
You may be asking yourself how someone who has only the most basic grasp of an Asian language can translate things like this. The answer is technology.
What I do when I translate is very tedious, but it gets the job done. First task is to get the words into text. Easy if it's a story from a website. Much more complicated if it's an image file. If it's an image file, I use an IME pad and draw all of the characters. It takes forever, but I haven't found a better way yet. To use an IME pad, all you need is Windows and to set up Asian languages. Vista and 7 have it automatically and it just needs to be enabled. XP is more complicated. Google answers how to do all that.
Drawing them was much more complicated for me at first, because of stroke order. Since I taught myself, I was just drawing them the way any old American would. Turns out there's a logical path in which lines go when you draw Kanji, and if you don't do it this way, the IME pad will not pick up on the character. If you do it right, it makes guesses as to what you're drawing and provides options to choose from. Clicking one adds it to your document. Repeat until you're done!
Once I have the whole document, I run it through a quick translator like QuickATLAS. It's dirty and wrong fairly often, but I can tell when it's wrong most of the time. That's when I switch over to an online dictionary like WWWJDIC and individually look up each Kanji pair to see what's confusing the quick translation. From there, it's just a little creativity to form a sentence based on what you think the sentence means. I usually translate sentence by sentence which makes the story sound disjointed at times, but my goal is to make it sound as close to the original as I can rather than make a story out of it.
Sometimes, you'll hit snags. If the resolution of the original image is too small, lines blend together on the really complex 20-stroke characters and you have no chance of working out what it is. That's when I try to find radicals I recognize and punch them into the multi-radical kanji lookup tool on WWWJDIC. I also run into problems on the complicated Kanji because I mess up the stroke order. That's when I just look up the full list of characters by stroke count in the IME pad. It also has a radical lookup function, which I use pretty frequently. And of course, one of the hardest things to do is to translate things that don't make sense outside of a Japanese context, like pronoun changes in TG manga or catchphrases / sayings they have. It takes some creativity at times, and the last option is always to just ignore it. I remember one story I translated (don't think I ever posted it, actually) where I just left off an entire page of story because I couldn't figure out the writer's intention and where it fit in to the rest of the story. I definitely skipped a paragraph or two in the last Dark Planet story I posted for the same reason. Sorry~ You get what you pay for~
Truth be told, I actually started writing a "how to translate japanese without knowing japanese" guide like two years ago and never got around to finishing it. This is a very, very summarized version of it because you've said you know at least the basics, so this should do for now.