From: Kenneth Whistler (kenw@sybase.com)
Date: Tue Sep 07 2004 - 12:41:42 CDT
> On 05/09/2004 18:27, John Cowan wrote:
>
> >The following links show L-shaped marks, apparently combining
> >characters, that indicate the change-of-pitch position in Japanese
> >words written in romaji. Are these novel characters, or can they
> >be identified with existing Unicode characters? Are they really
> >combining?
> >
> >http://member.newsguy.com/~sakusha/dict/martin-je.html
> >
> >http://member.newsguy.com/~sakusha/dict/kenkyusha-je.html
> >
> >
> >
> These could be 231C and 231D,
No.
> or 02F9 and 02FA (especially if they actually do indicate tone),
Yes. They indicate pitch accents, with distinctive rise or lowering
of pitch at the points indicated in the dictionaries.
> or possibly 2308 and 2309.
No.
> I don't suppose 20E7 is suitable.
Correct. It isn't suitable.
> One of them looks a bit like one of the proposed New
> Testament punctuation characters, pipelined for 2E00..2E0C, which is
> perhaps appropriate for a book "with examples like "a student of
> divinity at Oxford University.""
No. Those are brackets, not tonal modifier letters.
U+02F9..U+02FC represent precisely this kind of usage for pitch, which is
non-IPA, but which has scattered but not uncommon use.
I don't have a copy of Kenkyusha sitting on my shelf, but I believe
it also makes use of the falling pitch accent (U+02FA) shown in
the Martin's dictionary. Martin's dictionary kerns that over
the preceding vowel, but I don't believe it is best to represent
that data using an unrelated combining mark for it, simply
to attempt to get the pitch accent jammed up over the edge
of the vowel.
--Ken
This archive was generated by hypermail 2.1.5 : Tue Sep 07 2004 - 12:43:26 CDT