for chinese, korean and japanese, there are no such rules :-(
How does this work in ideogrammic languages? If a reader comes across
an unfamiliar character, are they completely stumped as to what it
might mean or how it is pronounced, or is there some system that
would allow a good guess?
yes, completely stumped :-) ... from the context, we can guess what the
the word means, but we cannot know how to pronounce the words ...
in Japan, when kanji (chinese ideograms they inducted from china) is used
in the newspapers, there will be hiragana next to the kanji, to teach the
pronunciations ...
oftentimes, the japanese pronunciations of the same chinese word,
is different.
Korea, is different, again ;-)
Korea invented a new writing system; in modern history,
the only culture that has done so (i think).
they still use some chinese words ... sometimes on
street signs, menus, mostly in period dramas ;-)
again, they pronounce most words differently as well ...
sometimes though, the japanese and korean pronunciations,
sounds similar to chinese dialects ...
for chinese words, sometimes we guess .. a little more on this later ...
can only provide anecdotal explanations, not academic or scholarly :-P
I was thinking along the lines of complex compound ideograms (or combinations) being based on simpler ones, that everyone knows, but that does extend to pronunciation?
this is the guessing part mentioned above :-)
and you're right, we can infer to some extends but never sure.
or do you have to look it up in a dictionary?
we usually/always do ;-)
i have a dictionary on my bed ;-)
sometimes to put me to sleep,
other times, to look up words ...
some very basic words are formed from symbols, like water (水), fire (火) or wood (木),
(please google how the symbols became words/radicals ... i cannot type it out ;-P)
which would then form radicals to compose words, eg:
木 wood
森 forest
树 tree
柏 cypress
板 plank
from the radicals, we can guess that the words are related to something woody,
but we cannot guess the sounds.
but among chinese, we have a saying that -- when we don't know the pronunciation,
if there is a left or right radical, say it, else say the component in the middle ;-)
most of the times, we'd be wrong, but (only) sometimes right ;-)
dictionaries are the only sure way to go, because there are some
words that are identical, but have different pronunciations and meanings:
角 (jiao) meaning: 1. corner, or 2. contest [of strength]; tussle,
角 (jue) meaning: role.
sometimes, different words can have the the same pronunciations:
(jiao, at the third tone):
角: as above
脚: leg
饺: dumpling
sometimes, the same word have "similar" meanings, but used in different contexts:
降 (jiang) : reduce, lower;
降 (xiang): cut-down, as in one is nemesis of another ...
削 (xiao): reduce, eliminate
削 (xue): peel, pare down ...
this must sound really confusing; but most of the times, we infer
from the contexts, what the words means or how they should
be pronounced (from prior knowledge)...
and related question: are Chinese dictionaries ordered by radical, and those by stroke count, as in Japanese dicts?
yes, like the japanese dictionaries, although japanese dictionaries most
likely adopted the methods from chinese dictionaries, because, china
simplified chinese characters/words, and the dictionary lookup
methods were most likely formalised during that time ...
<aside>
the chinese were very vigilant in documentations throughout history ...
during ming dynasty, emperor yong-le (永乐) commissioned the compilations of
all known knowledge at the time, into a huge encyclopedia: 永乐大典...
medicine, astronomy, philosophy, farming, construction methods, mathematics etc ...
and more "recently", in qing dynasty, emperor kang-xi (康熙) commissioned the
compilations of chinese words and verses: 康熙字典, 康熙辞海;
later his grand-son qian-long (乾隆), commissioned the compilations of
all known [major] works of literature.
point is they have been doing these for a long time :-)
</aside>
the lookup methods follow the following steps:
1. identify a radical, sometimes there are more than one (all can be used to locate a word),
2. lookup radical in radical section, which are grouped by stroke counts,
3. from there, follow link to section where all words with the same radical are listed,
4. these words will be grouped by stroke counts.
5. from there, find the the page number where the word is described;
6. for very complex words where radical is not obvious or present,
there is a section where these words are organised by stroke counts.
in simplied chinese dictionaries, the pronunciation of the words are described using
a romanised system; eg: for sounds like:
b : sounds like the bo, in bo/ring
p: sounds like the po sound in po/or
m: sounds like the mo sound in mo/re
f: sounds like the fo sound in fo/r
...
...
the tones for the pronunciations are also listed above the "dominant vowel",
like diacritics ... there are 4 tones in mandarin (cantonese has more tones).
dictionaries for traditional chinese, the words are listed purely by stroke counts
-- at least the hong kong dictionary i saw -- simple, but brute-forced, not very efficient.
the pronunciations are composed of cryptic squiggles (most likely from ancient
times ;-)
Hong Kong and Taiwan use the traditional chinese writing system,
although it is likely that Hong Kong may be converting to simplified chinese,
now that she is "re-united" with china.
the chinese characters/words used in Japan and Korea are also
traditional chinese.
China and Singapore use simplified chinese.