Chinese dictionary

[From Bill Powers (2005.08.23.2036 MDT)]

I have found a marvelous resource: a Chinese-English dictionary that
can be accessed in Pinyin as well as English.

http://www.mandarintools.com/worddict.html

The root of this URL leads to many useful aids.

Of course I tried immediately to decipher the characters in the title
of my book in Chinese. Two early discoveries: the word for control is
zhi4, and for perception is zhi1 jue2 (the numbers indicate the tone:
1-steady, 2-rising, 3 falling from low then rising, or just falling
from low, and 4-falling from high. The characters check out with two
of the characters in the title, though I'm obviously missing some
details. It's not as interesting as you may think that "control" and
"perception" both contain the syllable "zhi" The tones are different,
and the characters that go with zhi are different in the two words.

Another discovery. As I said, the word for perception is zhi1 jue2
(that is, zhijue; it has two syllables). If you look up the
individual Pinyin syllables, the meanings are all listed with all the
tones. Zhi, for example has 105 meanings. Jue, the second syllable,
has 42 meanings. This would imply that these two syllables can be
combined to give 4410 meanings. But that is not the case.

Each meaning for an individual line in the dictionarysyllable is a
different character (I assume on the basis of an incomplete sample).
So what we have here are words being made up of pairs of syllables,
as well as each syllable having multiple meanings. However, an
interesting regularity shows up. Among the 18 meanings of zhi1, the
first syllable of "perception", only one of them pertains to anything
like perception: it is given as

zhi1: /to know/to be aware

The second syllable with the rising tone has 39 meanings, among which
only one pertains to perception:

jue2: /feel/find that/thinking/awake/aware

The only common term is "aware". So to perceive is to be aware, or to
be aware of feel?. I can see that I will have some explaining to do
about "perceptual" signals that do not reach awareness.

I'm beginning to see a structure in this Pinyin language. Each
syllable has multiple possible meanings (i.e., multiple possible
characters). Attached is the table generated by looking up jue. As it
happens almost all the meanings of jue go with the second tone; there
are only three with the first tone and none with the others. But the
meanings vary all over the place, with only a few near duplications (
like /dig/ for one and /dig/pick/ for the next one).

Obviously, for most words more than one Pinyin syllable is required.
The process is very much like intersecting categories. The first
syllable of zhi1 jue2 designates not a single meaning but an
arbitrary category many up of many different meanings. The same for
the second syllable, but the set of meanings is completely different.
So intersecting the two categories, as in set theory, results in only
a single meaning for the combination. Clearly, converting syllables
into meanings involves intersecting categories, with a limited number
of meanings going with single categories (when no valid second
syllable follows?).

Well, isn't this the case for English? We have fireplace and
firetrap. You could look up all the words that have "fire" in them
and list the meanings, which of course would be all over the map. And
you could look up all the words with "place" in them, and get another
long list, probably just as long as the attached list or longer. Each
syllable may have a meaning by itself, and each different combination
has different meanings. Many combinations of syllables in English
are, given their individual meanings, nonsensical unless you happen
to know the etymology.

What we lack in English is a unique character for each meaning. But
that may be true in Chinese, too. The two characters that make up
zhi1 jue2 are different from each other, and each one can be
associated with multiple meanings of the Pinyin syllable. However,
the key is in the number of meanings that go with each character.
Often it is just one, as in jue2: to chew, or in jue2: torch
(different character). Though the same syllable is spoken or written
in pinyin in each case, the characters are different. Obviously there
is enormous potential for punning, as well as for misunderstanding,
in spoken Chinese. I think I will chew my house for the insurance. I
think Martin Taylor or his wife commented that when conversing,
Chinese will often sketch characters in the air to clarify what
they're saying. Listening to the radio must be an adventure in wrong
interpretations. I should think that the danger of misunderstanding
must lead to a lot of stereotyped phrases that are long enough and
familiar enough to eliminate ambiguities.

I think the dictionary is misleading, after all this. It's not that
jue2 has a lot of meanings -- its just that a lot of words contain
this syllable, so the syllable's meanings given in the dictionary are
just lists of all the meanings of the words that contain this
syllable. A funny way to organize a dictionary. I suppose that if a
character has a single meaning listed opposite to it, that is a
unique character, but characters with multiple meanings given do not
have multiple meanings; it simply takes more than one character to
select which of the multiple meanings is wanted. So just as in
English, a word can consist of multiple syllables. I'm sure that has
been obvious to Martin and Bruce N. and others with some expertise in
linguistics, but it wasn't to me.

The language is looking somewhat less complex right now. I don't have
much progress to report, but Xulai (Francesca, Autumn Winter's wife)
says I am doing quite well at speaking correct Chinese (as good as
her parents, she says). So I have the talents of an excellent parrot
with a poor short-term memory, and right now it feels like a brain of
parrot size, too. We are using Gary Cziko's language facility at the
U of Illinois, which allows us to see and hear each other, and type
pinyin when I can't hear what Francesca is saying (being a bit deaf
doesn't help, but I can turn up the audio). I am very slow and
Francesca is very patient, so it may yet work out.

Best,

Bill P.

jue2.html (9.98 KB)

[From Bruce Nevin (2005.08.24 15:00 EDT)]

An important difference from English is that Chinese is what is called
an "isolating" or analytic type of language, in which words comprise
just one morpheme each, as opposed to the synthetic types of languages
("agglutinative" like Turkish, "inflecting" like English and German, or
"polysynthetic" like Navajo) where words include more than one morpheme
-- a word is a stem or root that can have other morphemes affixed. Many
of the morphemes/words of Chinese are single syllables, so much of the
vocabulary is in fact monosyllabic.

Peng2 you3 "friend"
peng2 "friend"
you3 "friend"
bing4 you3 "wardmate" or "friend made in a hospital"

And so forth. (These and more seen by looking up "friend" on that
wonderful website you cited.) From this we can see that both peng2 and
you3 mean "friend" independently of the compound word peng2 you3
"friend". Looking up bing4, I see six meanings, the last of which is
"ailment/sickness/illness/disease/fall ill/sick/defect/". Not
"hospital". "Sickness-friend".

The other five meanings/characters are

- /and/furthermore/(not) at all/simultaneously/also/together with/to
combine/to join/to merge/
- /amalgamate/combine/
- /and/also/together with/
- /nightmare/start in sleep/
- /arrange/drive off/expel/

All 6 are bing4 (you can tell this by the fact that all link to a sound
file of that name), but each has a distinct character. A listener would
have to tell contextually, or by the speaker signing) which was
intended. The first three have obvious similarities in meaning. Did
different characters historically get associated with one and the same
word? As to the last two one can imagine driving off the stuff of
nightmares (of such imaginings are folk etymologies made). There is a
more tenuous association of these with the "sickness" meaning above.

These are not just syllables, but also words, and also morphemes in
compound words like

nu peng2 you3 "girlfriend, female friend"
nan2 peng2 you3 "male friend"

(I couldn't find a suitable meaning for nu, unless it's either "slave,"
"child," or "indignant," -- and with those to choose from I would be --
so I don't know its pitch.)

If there's a distinct meaning to each character, and each character
corresponds to just one syllable, then it appears likely that a syllable
tends to be perceived as having the meaning of that character, even if
it is only part of a compound word. Is that what you're finding to be
the case?

The parts of a character also are significant, but as I found by looking
at Wieger's book those meanings are more etymological in nature. The
fascinating thing is that the etymology of characters is independent of
the etymology of syllables and words.

  /B

[From Bill Powers (2005.08.24.1926 MDT)]

Bruce Nevin (2005.08.24 15:00 EDT)]

An important difference from English is that Chinese is what is called
an "isolating" or analytic type of language, in which words comprise
just one morpheme each, as opposed to the synthetic types of languages
("agglutinative" like Turkish, "inflecting" like English and German, or
"polysynthetic" like Navajo) where words include more than one morpheme
-- a word is a stem or root that can have other morphemes affixed. Many
of the morphemes/words of Chinese are single syllables, so much of the
vocabulary is in fact monosyllabic.

Peng2 you3 "friend"
peng2 "friend"
you3 "friend"
bing4 you3 "wardmate" or "friend made in a hospital"

And so forth. (These and more seen by looking up "friend" on that
wonderful website you cited.) From this we can see that both peng2 and
you3 mean "friend" independently of the compound word peng2 you3
"friend". Looking up bing4, I see six meanings, the last of which is
"ailment/sickness/illness/disease/fall ill/sick/defect/". Not
"hospital". "Sickness-friend".

This is what I was wondering about, and still wonder. I don't mean to be the wise guy who leaps from ignorance to expertise in a single bound, but isn't it possible that there's an alternate interpretation here?

Suppose we say that in this dictionary, the alternate meanings of the morphemes (?) in each line simply indicate the meanings of words in which the morpheme -- I'll just say syllable -- occurs. The syllable doesn't "mean" those meanings all by itself, but when combined with the right other syllables the combination means one of them and not the others.

My interpretation is borne out by looking up peng2. Sure enough, there are 13 "meanings" listed, only one of which is "friend." Others are "boron", "swollen", and "shed". The other word that "means" friend is you3, which has 11 entries, one of which is "friend", but others of which are "wine container", "ritual bonfire" and "to lead." Only the combination peng2 you3 unambiguously indicates "friend", because that is the only "meaning" they have in common. that is, the two-syllable word for friend is made of the syllables peng2 and you3, which is why a word for friend is found next to each of those syllables in the dictionary.

Clearly, those lists of meanings in each entry are the meanings of words obtained by combining the syllable in that line with various other syllables, so the combination of syllables narrows the possibilities down to one (or maybe just a few, with context handling the rest).

If we organized an English dictionary the same way, we would have things like this:

car /container/transport/neglectful/vegetable

rot /decay/edible root/dog breed

The meanings that are closest to being in common are vegetable and edible root, and when you combine the syllables you get carrot, which means vegetable and edible root.. Of course in combination with other syllables, car can "have" other meanings, perhaps in combination with "ton" which has its own list of meanings. One of the words listed with ton might be box, which is a kind of container. So we have the word carton.

What I'm suggesting is that there has been a misunderstanding here, perhaps based on the idea that syllables "contribute to" the meaning that a word "has". But if we simply recognize that peng2 you3 is a word made of two syllables, we can see that it is the word, not the syllables, that means "friend" just as it is the word carton, and not either car or ton, that means a container with flat sides. [Trying it, I found that only peng you, not pengyou, produces a meaning, and there is only one meaning: friend]. For onlookers, peng rhymes with dung, and you rhymes with toe.

Of course both car and ton by themselves do have meanings, which have to be obtained by determining that no other syllable is linked to them, and then by interpreting context (doing a ton on the M1 has nothing to do with weight).

My conclusion is that the situation with Chinese is nowhere near as bleak as those lists of meanings would suggest, with 105 meanings for zhi and so on. The syllable zhi does not have 105 meanings; it simply appears along with other syllables (and perhaps a few times alone) in 105 words with different meanings.

Your "bing4 you3" example suggests that you3 itself can mean friend, but if you look at the lists of meanings, it's hard to find combinations of meanings associated with bing4 and you3 other than "sickness friend" that make any sense. But it's possible that "you3," used primarily as part of a word for friend, can be used as an ending, as in "skate-o-rama". And unless context suggests a hospital, Bing4 you3 could mean any companion in illness anywhere. If you reverse the order of syllables, I'll bet you get a different meanings: you3 bing4 = grievous illness would be a guess. I just tried that word with the dictionary, with and without numbers, as two words and as one word, and in every case the dictionary had no such entry. I lost my bet. However, entering bing you as separated syllables, but not when joined, led to just one entry, the one you cited.

Actually, relaxing a constraint on the search came up with xue4 you3 bing4, haemophilia, where the ONLY listing for xue4 is "blood".Blood grievous illness = haemophilia. Here you3 in combination with bing4 obviously does not mean anything like "friend."

I'm not ready to give up my hypothesis yet, even if it doesn't look as pure as it did before. Nature is messy.

The other five meanings/characters are

- /and/furthermore/(not) at all/simultaneously/also/together with/to
combine/to join/to merge/
- /amalgamate/combine/
- /and/also/together with/
- /nightmare/start in sleep/
- /arrange/drive off/expel/

All 6 are bing4 (you can tell this by the fact that all link to a sound
file of that name), but each has a distinct character. A listener would
have to tell contextually, or by the speaker signing) which was
intended.

My proposal (that we are mostly looking at syllables and not words) greatly reduces ambiguities, because many meanings are given by pairs (or larger combinations) of syllables -- and characters. I don't think you could guess the meaning of peng2 you3 just from hearing either pung2 or you3, but the word pung3 you4 is unambiguous. You couldn;t tell which meaning of bing4 is intended from bing4 alone, because bing4 is not a word; it's a syllable which can be part of words meaning all those 6 different sets of things. In combination with other syllables the meaning becomes either less ambiguous, or completely so.

The first three have obvious similarities in meaning. Did
different characters historically get associated with one and the same
word? As to the last two one can imagine driving off the stuff of
nightmares (of such imaginings are folk etymologies made). There is a
more tenuous association of these with the "sickness" meaning above.

As long as you assume that each syllable "has meaning," you will interpret those lists as alternate meanings. But if you don't asume that, and assume instead that these are just parts of words and that meanings can't be established until you have the whole word, then it becomes pointless to look for "similarities of meaning" among syllables. The different characters show that the meanings are different enough that the words are different words.

These are not just syllables, but also words, and also morphemes in
compound words like

nu peng2 you3 "girlfriend, female friend"
nan2 peng2 you3 "male friend"

(I couldn't find a suitable meaning for nu, unless it's either "slave,"
"child," or "indignant," -- and with those to choose from I would be --
so I don't know its pitch.)

There's a nu with an umlaut that comes up in relation to women, so maybe that's a little different. But a similarity of words for woman and slave would not be surprising, given Chinese history.

If there's a distinct meaning to each character, and each character
corresponds to just one syllable, then it appears likely that a syllable
tends to be perceived as having the meaning of that character, even if
it is only part of a compound word. Is that what you're finding to be
the case?

I don't know yet -- I barely know the characters for Man and Woman, kindly shown to me by Francesca so I could identify the correct restroom. But it's clear that two-syllable pinyin words go with two-character words, and I have read elsewhere that the characters are simply syllables, only some of which are words by themselves.

The parts of a character also are significant, but as I found by looking
at Wieger's book those meanings are more etymological in nature. The
fascinating thing is that the etymology of characters is independent of
the etymology of syllables and words.

I think Insup and Martin Taylor said that the character etymology probably is simply a set of stories told to help memorization, and have no historical basis -- that is, the characters were not developed as the stories say. They may have evolved by combining simpler parts, but one character is one syllable, however it came into being.

Early visitors to China brought back wondrous stories about the Chinese characters; they believed that the characters somehow had the power to implant meaning directly into the listener's mind, without the intermediary of words. The piece in which I read that also concluded that the characters are simply syllables with no unusual properties.

Any time that a syllable is listed with multiple meanings in the dictionary, I think we can take it for granted that to pick out one of the meanings, that syllable must be combined with at least one other. We can leave the possibility open that there are genuine ambiguities that must be resolved by context, but they are not nearly as great as the arrangement of this dictionary would suggest.

Best,

Bill P.