[Jon Babcock (Re: Correction (Was: CJK combining components (was RE: "Giga ...))) writes:]

>> >>>>> Jim Breen <jwb@...> writes:
>> >>> don't pretend to know the readings of [all] the JIS 208 kanji
>> >>> (especially the ones that don't really have any), and who find
>>
>> But aside from these, how in the world can a kanji not have a reading? I'm
>> really curious. Someone on the Unicode list had made this same, to me
>> amazing, statement, and he didn't seem to have the kanji Thomas Chan referred
>> to in mind. What gives? Have the readings just been forgotten? Examples?

The ones I had in mind were the bogus ones that cropped up in JIS X0208.
Nearly 30 years ago when the predecessor of JIS X 0208 was being
created, a decision was made to include all the kanji that appeared in
the official town/district names. The paper records of these were duly
copied and assembled and transcribed. The trouble is that a handful were
misread. In at least one case a shadow from two pieces of paper glued
together was taken as a stroke.

These "kanji" of course are now immortalized in Unicode & ISO-10646
under the source rule. How to handle their readings, etc.? One school of
thought is to use the reading of the kanji from which they have mutated.
Another is to treat them as readingless and meaningless artifacts.

An example is U+56ce, which is part of a town name in Kagoshima-ken.
In that context it's read "so". AFAIK that kanji has no Chinese source.
It's not in Morohashi. Shibano's "JIS Kanji Jiten" goes into
great detail on all the oddities. My copy is at home so I can't cross-check
to what it should be.

Jim

PS: In the first edition of Spahn & Hadamitsky they quirkily gave these
kanji "meanings" like "slithy" and "tove", to the confusion of some
readers, and the mirth of others.

--
Jim Breen [j.breen@... http://www.csse.monash.edu.au/~jwb/%5d
Computer Science & Software Engineering, Tel: +61 3 9905 3298
Monash University, Fax: +61 3 9905 3574
Clayton VIC 3168, Australia $B%8%`!&%V%j!<%s(J@$B%b%J%7%eBg3X(J