>>>>>>>>>> Thomas Chan <tc31@...> writes:

> Another place for
> similar-looking characters is the _Tong Shu_ (U+901A U+66F8) Chinese
> almanac (also known under a few variant names), which one should be able
> to get [a 2001 edition] at a Chinese bookstore or grocery in a month or
> two, which has some Taoist charms with similar outlandish characters; a
> few are excerpted in Martin Palmer's _T'ung Shu: the Ancient Chinese
> Almanac_ (Boston: Shambhala, 1986), which contains an explanation and
> partial translation of the invariant parts of the almanac.

Also, some of these Taoist 'tallies', fu2, U+7B26, can be found in the
Tao-tsang. I have only an extremely shortened version of this, the Sung, 120
chuan, Yun-chi-ch'i-ch'ien, where fu2 are found in chuan 80, among other
places. Professor Boodberg was mumbling something about these Taoist
'characters' on several occasions, but all I can recall is something along
the lines that they should be looked into if one were to study the Chinese
script. I recall no details and so far have found nothing in my notes.

>> A more serious challenge ... but I'm sure there will be more ... to
>> the idea of a hemigramic notation for *any* Chinese characters may be
>> trying to deal with characters such as the one Thomas Chan mentioned at
>> URL, http://member.nifty.ne.jp/Gat_Tin/kanji/sinji.htm 5th down from
>> the top. I'm still trying to get my mind around this one.

I should have said, 4th one down from the top. But the 5th one works, too.

> I would question whether that character is even part of the same system
> as all of the rest that we are dealing with; while its components are
> the same, the way it's assembled is not.

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

A key point. If an attempt to represent each Chinese character in terms of its
hemigrams is anything, it is an investigation into the way each character is
assembled, rather than merely another long catalog of the results.

> Vietnamese demotic characters, chu+~ no^m, are of the same system (with
> a few new components), as are "national characters" of Japan and Korea
> (although some of the latter's as used in Kugyol and other early writing
> systems may require new components as well).

Would these 'new components' all be amicable to the term 'hemigram'? I know
nothing yet about chu+~ no^m characters, but the 100 or 200 Japanese kokuji
would seem to require but few new components.

> Zhuang characters (how extensive are they?) also
> seem to follow the same rules. On the other hand, the Khitan, Jurchen,
> and Tangut characters seem to require much more additional components,
> and different manners of construction; one might also say the same of
> some of the newly created characters used in post-1950's China.

Could you say something about what the new methods of assembly might be. No
matter what definition is used, the newly created post-1950's characters must
be considered 'Chinese' and a system that could not deal with them would not
be complete.

Also, in the newly created 'vulgar characters' of the past 100 years of so,
were new methods of assembly applied?

> Another character we might want to consider is U+56CD. It's reading,
> xi3, and meaning, 'double happiness' (referring to marriage), are
> well-known; yet, none of the three juggernaut dictionaries above contain
> an entry for it! About the only place it really occurs is for
> decorative purposes like a wall sign (although I am aware that the title
> of two movies, from Hong Kong, contain this character). Is it a
> character, or a not?

Well, in any event, it poses no special problems for a system of hemigram
notation, if hsi3 U+559C is already considered to be on the list of
hemigrams. Using Boodberg's annotation for reduplicated graphemes, it might be
something like, xi2 (the '2' meaning 'times 2', and not as an indication of
tone), in the same way that, for example, ts'ung2 U+4EC2, would be h2 (h =
human, Kangxi #9), and lin2 U+F9F4 would be d2 (d = dendrological, Kangxi #75),
and p'in3 U+54C1 would be o3 (o = oral-ostial, Kangxi #30), etc.

I see I have reverted to my old habit of using Wade-Giles romanization in this
post.

Jon

--
Jon Babcock <jon@...>