On Sunday, November 4, 2012 11:01:06 PM UTC-5,
you...@cds.ne.jp wrote:
> As a native Japanese speaker, I feel it is most convincing that
> DevTeam had referred from 'gan-yaku 丸薬',
Problem is, the Dev Team were, rather obviously, not native
Japanese speakers. They were clearly much more fluent in
English than in Japanese.
Nobody fluent in English (or any other major European
language for that matter) would ever transliterate 丸薬
as "gunyoki". All three vowels are very wrong -- not
just different, but from opposite sides of the vowel
continuum (a/e/i versus o/u) in every single case.
Getting A and U mixed up, for example, is a distinctively
east-Asian trait, almost as much as mixing up R with L.
(An Engish speaker is far more likely to confuse な with
が or つ with す or the entire ら row with the だ row or
even お with う. If working with strictly written
material, an English speaker might also confuse あ with
え, because those sounds are written with the same letter
in English, though we would never confuse those two
sounds if speaking aloud or listening to spoken words.)
No, I don't buy that the original DevTeam, working in
English primarily because that was their native language,
changed お to u, a second お to o separated from the next
vowel by just one consonant, and う to i, all in the
same world. No way.
If a native English speaker made *one* such mistake, we
could speculate that maybe it was a finger-on-the-wrong-key
typo that didn't get noticed, and by the time they looked
at their transliteration again they'd forgotten the Japanese
word they looked up. That's a reach, but with one such
change it might be possible. Three such mistakes in the
same word, however, is completely implausible.
Sloppy transliteration of ganyaku could easily lead to
ganyakkoo, gonnyaku, gonnyokku, ganyakkew, or maybe even
gone-yokk-oo. But it could not lead to gunyoki, not if
it was done by a Westerner -- which NetHack was.
My best guess at this point is, they were using a paper
dictionary, and they tried to look up the kanji one at a
time, and they either got a similar-looking kanji with a
subtly different radical (Japanese has several different
pairs of radicals that look basically the same to the
Western eye but mean something completely different; we
don't have this problem with modern Internet dictionaries,
because we copy and paste; but with a paper dictionary
it would be an issue) or else they just picked the wrong
reading, which would also be easy to do.
It's also vaguely possible that they couldn't find what
they wanted in a dictionary and just made up fake words
that sounded Japanese enough to be convincing for
Westerners. It wouldn't be the first time.