So I had some time on my hands and I was playing around with Google Translate, which I think shows a lot of promise, but still has some weaknesses as well. The main weakness is that it’s well a machine.
I don’t know what possessed me to do this but in the Japanese to English translation interface I plugged in the katakana for Justin Bieber which looks like this:
ジャスティン・ビーバー
Below is a screen shot of Google’s epic fail translation.
Haha, that’s so weird. I tried with other names and they all seem to come out fine.
Another weird one though:
キャメロン = Cameron
きゃめろん = Melon bloody murder (wtf?!?!)
I guess kya is the screaming sound when someone’s surprised, right. So a melon is scaring someone?
I’m pretty sure that the reason for the Justin Bieber one is that because Ludacris showed up on a version of one of his songs the katakana for justin bieber and english for ludacris became associated.
Seriously speaking, I think one of the biggest weaknesses of Google translate is that it only gives you one option. There are a lot of homophones in Japan so without kanji or context it’s not easy to give a one shot answer.
Just to check I put かえる into Google translate and got “hatch”. I think the “hatch” meaning is pretty far down the list of possible meanings. The kanji is a real killer too (but not too hard when you look at the parts that make the whole) – 孵る
I guessed that might be it. Like “OMG Melon!”.
Yeah, they’ve got their work cut out for them with the homophones. That’s why people who claim that Kanji should be scrapped don’t have a clue. The Japanese language doesn’t work without it.
家に帰る = Go home
家にかえる = Frog house?
Something fun to teach, if you happen to be a teacher, is guessing the meanings of words by context. It’s not a skill they really need to use when reading.
Tim, I have the same love for kanji. Without kanji, I wouldn’t know if there was a frog in the story, something hatching, someone changing or going home or what?
Come to think of it, one of the Japanese Language Proficiency Test Level 1 kanji question types are homophones, and different kanji with the same reading. Don’t even get me started on the kanji that have the same root and same pronunciation for which the only difference is the radical, and even then the radical can be similar like tsuchihen, tehen, and gyoninben or takekanmuri and kusakanmuri.
Still lots of fun though.