> Sun, 1 Nov 2009 21:19:44 -0800 (PST): "Peter T. Daniels"
> <gram...@verizon.net>: in sci.lang:
>
>>On Nov 1, 9:20 pm, LEE Sau Dan <dan...@informatik.uni-freiburg.de>
>>wrote:
>>> >>>>> "Bart" == Bart Mathias <math...@hawaii.edu> writes:
>>>
>>> Bart> I don't know what Peter based his comment on, but for sure it
>>> Bart> is rare to use ? alone for "sun" instead of "day" unless one
>>> Bart> is writing about sunrise or sunset.
>>>
>>> Is that relevant? We were talking about *morphemes*, not just free
>>> morphemes. The same character "?" is used to write TWO different
>>> morphemes, namely "nichi" and "hi", both meaning "sun", isn't it? Is
>>> that ideographic?
>>
>>The same shape "bow" is used in English for at least two different
>>morphemes. Does that mean "bow" is an ideogram?
>
> If the two morphemes have related meanings, probably yes.
That's a strange interpretation of "ideogram". Especially since, if the
two words have similar pronunciation and similar meaning, they are
probably etymologically the same word.
My instinctive reaction also was that 日 is not an ideograph, because it
can't be read "taiyô", the most common name for the idea supposedly
depicted.
I suppose some of these morpheme-character associations (for native
Japanese words) were coined by interpreting Chinese characters in a
limited ideographic fashion, but now that the associations for most
characters are conventionalized, this is of little relevance in the
process of everyday Japanese reading/writing. They are about as
ideographic as English "cuckoo" is onomatopoetic.
However, Japanese authors sometimes take similar freedoms today, by
writing an intended, unusual reading next to the Kanji, or when creating
names for their children (an examples I remember: write "gogatsu" (May),
read "mei").
--
A computer will do what you tell it to do, but that may be much
different from what you had in mind. - Joseph Weizenbaum