[From Bill Powers (2005.07.29.0235 MDT)]
[Please note the hour. This is a ramble which may not hang together very
well, as I wait to get sleepy again]
Martin Taylor 2005.07.28.10.36 –
I guess it really depends on
what you mean by “meaning”. I think that we are coming to the
core of your difficulty with this question.
By “meaning” I mean the (usually nonverbal) perceptions that
are aroused in my mind at different levels as I read words. Words are
perceptions that by convention and through private experience evoke other
perceptions from memory.
Anyway, it’s inconceivable
that any language could consist only of combinations of a small set of
basic meanings.
“Combinations” suggests a building constructed with defined
blocks that are fit together to make a shape. I tend to think of
“meaning” as being the overlap of fuzzy clouds of
connotation.
You may be right about the fuzzy clouds, but I’ve always enjoyed, since
fourth grade, what I was taught were the basic rules of English, through
which it became possible to figure out what even complex sentences
denote, as well as to construct understandable sentences. The puzzlers
are relatively rare, even in English. The rules I learned can still be
retrieved if I need to (I think I can still parse a sentence in the
old-fashioned way I learned), but they are mostly habitual now. Anyway, I
think that a lot of what I do with language doesn’t really involve rules;
this could be one of those cases where one can figure out rules for a
process that isn’t actually done by following rules. A lot of it consists
of moving words around in a sentence so they are closer to the other
words dealing with the same subject and to break up combinations that
have accidental meanings for the speaker and presumably the listener, as
in “I threw my horse over the fence some hay.” The explanation
for the preferred word order in that sentence might be worked out as a
rule, but the real explanation is in the images evoked by saying it that
way. It’s not “fuzzy clouds of connotation” that make that
sentence wrong, but the images that come to mind as a result of the
sentence as it unfolds. Maybe that’s all that “connotation”
means. The lack of endings in English means that there’s no signal to say
“this is not the object of the verb.” The meaning of
“read” in “I read the book” changes depending on the
next words: “yesterday” or “every day.” So if you’re
sensitive to unwanted meanings, you recast the sentence to identify the
time-frame first: “Yesterday, I read the book”. Or you place
the object closer to the verb and make the words track the action:
“I threw some hay over the fence to the horse.”
Well, all that’s about English, and from what I’ve seen so far it would
apply equally well to Pinyin, though in a language without tenses and
plurals, it seems to me that one would have to be even more sensitive to
word selections and orders that evoke unintended perceptions. How this
applies to characters, in which the elements apparently occur
simultaneously and not in any one order, I can’t say. If you can’t change
the order of the elements, how are unintended meanings dealt with? And
how is the correct order indicated except through context? Maybe
this has something to do with the lack of tenses in Chinese. You would
have to deal with sequence in a different way. Bruce Nevin, you always
have interesting things to say about things like this. Enlighten us, or
at least me.
Where many clouds overlap, the
meaning is densest and most precise. That’s what happens when you use
context.
I don’t think it’s quite that fuzzy. Many words simply denote
perceptions, and any connotations come after the basic meaning is
experienced. That’s what I mean by saying connotations can just be
private associations between perceptions, dependent on experience but not
on language. Some are, of course, dependent on language.
Context is indeed important in guessing at the meanings of words, but it
doesn’t help much with Chinese idioms (reported by Ina) like
“large-small” meaning size. Interesting that the perceptual
variable, for which there is apparently no word, is identified by
combining words for two values it can have. A girl once asked me,
interestingly, “Are you talking about that man-woman thing?”
And anyway, once the context, or for that matter anything, has served to
identify the meaning of the spoken word, we move from connotation to
denotation. In the sentence “I read the book,” we may initially
be uncertain whether the word “read” is pronounced
“reed” or “red”, but once we settle the time period
(present or past) and realize that the word means an action that we
experience ourselves doing with books, we no longer even think of the
color perception.
So do they represent
phonemes or syllables? Do their arrangements within a character code for
sounds? What is the nature of the chunks?
Characters represent individual syllables. That’s precise. A word may be
composed of one or several syllables, so the characters themselves don’t
convey the whole meaning of a word, any more than do the syllables of
English.
All right, then, that jibes with what I have learned. I came across a
paper on the Web exploring the myth of Chinese characters, which were
once thought to have magical properties enabling them to convey whole
meanings directly into the brain. The author said the same as you: the
characters indicate syllables. Some words consist of single syllables,
but most require multiple syllables, just as in English. To me, a
syllable is an auditory phenomenon, not a written one. “Phlegm”
is a complex visual object, but a unitary auditory event, a single
syllable. There’s no indication in the written word that it’s a single
unit instead of more than one. Why isn’t it two syllables,
“fle” and “gum”? Because that’s not how it’s
pronounced.
The radicals within the
character often represent connotative fields of meaning,
I wonder if those “connotative fields of meaning” are real. Is
this how people actually see them, or is it a result of some academic who
was searching for meanings in characters and came up with an objective
analysis? I’m thinking of what Ina said about the stories behind the
characters. Those stories may be true, or they may be, as Ina said,
simply mnemonic devices to help remember the details of the character
when one wants to do that. The characters may simply be characters just
as letters of the alphabet are, though more complicated, while the
analysis of their details may have nothing to do with their subjective
meanings beyond the function of jogging memory. After all, we can make up
stories about letter groupings, too, but they have nothing to do with how
we use them. Maybe the way people use these details to remind them of
meanings fits my “General Law of Human Behavior” which goes,
“Some do, and some don’t”.
Or maybe I’m just being unreasonably skeptical.
but the reader could be misled,
(as one can in reading English orthography for sound – it was many years
before I discovered that m i s l e d was not pronounced
“mizzled”.) Same for the phonetics within a character, if they
exist.
My father had the same discovery when he was in his 60s. His
pronunciation was “missile -d” which he explained to himself as
“being sort of shot off in the wrong direction like a missile”.
A different sort of example is “honing in” on something, a
mis-hearing which people explain by thinking of an image of something
that is brought to a sharp point by honing it. I think this involves
mostly people too young to remember WW2 and the homing beacons that
brought bombers back through clouds and fog to England, “Homing in
on a wing and a prayer,” the song said, after a famous quote by
someone now forgotten. Inevitably, someone heard “homing” as
“honing”, and rationalized it so it made sense.
Meanings are perceptions evoked from memory by other
perceptions.
Letters in an alphabet – or
perhaps the commonly associated sounds – also carry connotative meaning.
Ina’s thesis was on that topic, called “phonetic symbolism”.
She tested it in four languages from different language families
(English, Korean, Japanese, and Tamil), and four dimensions of meaning (I
seem to remember they were size, temperature, pleasantness, and activity,
but I could be wrong about one or two of those). She created all the CVC
syllables that could be made from the phonemes that were more or less the
same across all four languages, and asked monolingual speakers in their
own countries to judge all those syllables on all those dimensions. She
found almost universal commonality across speakers of any language, but
no correlation across speakers of different languages (except that they
all showed a low but significant correlation with English – around
0.2).
When you ask people to look for connotations they will provide them. If I
ask you what color verious letters are, you will probably come up with
some. But this doesn’t mean those particular connotations are always
there, or serve any useful purpose – especially in the presence of far
larger effects of other kinds. More curmudgeonly skepticism?
Now think of this in connection
with the radicals, but since there are more radicals, the spread of
connotation could be tighter than with the 26 letters. All the same, it’s
not going to be precise. In English, the meaning doesn’t inhere in a
letter, a word, a phrase, or a disquisition. It is in all of them,
becoming more precise the larger the included context, as the
“clouds” overlap in different directions.
My approach is to look for the largest regularities first, then fill in
the smaller and fuzzier ones. As to “the meaning,” I don’t
think there is any such thing – it’s in the individual’s experience that
we find meanings, not in words. We work very hard to identify the exact
experience that others seem to mean by a word, but we can never be sure
that their experience is the same as ours. If meanings inhered in
communications, we wouldn’t have to work so hard to identify them. The
idea that words “carry” meaning is, I think, theoretically
indefensible. They are just perceptions.
With that, I think I can get back to sleep, if anyone else is still awake
and reading this.
Best,
Bill P.