Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

nearly every Chinese word is two morphemes = two syllables = two characters.

284 views
Skip to first unread message

Hen Hanna

unread,
Jul 8, 2016, 3:21:44 PM7/8/16
to

On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:
> .....
> Characters DO NOT represent words, they represent morphemes, and nearly every
> Chinese word is two morphemes = two syllables = two characters. There are a few
> words that are disyllabic but a single morpheme.



> Characters DO NOT represent words

e.g. 中华人民共和国 <-- I think every
single character here represents a word.

The probability that a Chinese character represents
a word is like 99.9% ?


__________________

>>> Nearly every Chinese word is two characters.
>>> Most Chinese words are (written with) two characters.


I've heard this before, and I think ...
Nearly every person
who believes this doesn't know any Chinese. ... & knows
less than 100 Chinese characters.
most definitely less than 500 Chinese characters.


I 'd like a confirmation that Mr. Daniels knows
less than 500 Chinese characters.


>>> Most Chinese words are (written with) two characters.

Does anyone have a number for this?
--- my guess is that a bit over half (50-70%) of
Chinese words are (written with) two characters.



中华人民共和国
中華 人民 共和国 2 + 2 + 3 (2 + 2 + 2+1) characters.

HH

Peter T. Daniels

unread,
Jul 8, 2016, 3:46:01 PM7/8/16
to
On Friday, July 8, 2016 at 3:21:44 PM UTC-4, Hen Hanna wrote:
> On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:
> > .....
> > Characters DO NOT represent words, they represent morphemes, and nearly every
> > Chinese word is two morphemes = two syllables = two characters. There are a few
> > words that are disyllabic but a single morpheme.
>
>
>
> > Characters DO NOT represent words
>
> e.g. 中华人民共和国 <-- I think every
> single character here represents a word.
>
> The probability that a Chinese character represents
> a word is like 99.9% ?
>
>
> __________________
>
> >>> Nearly every Chinese word is two characters.
> >>> Most Chinese words are (written with) two characters.
>
>
> I've heard this before, and I think ...
> Nearly every person
> who believes this doesn't know any Chinese. ... & knows
> less than 100 Chinese characters.
> most definitely less than 500 Chinese characters.
>
> I 'd like a confirmation that Mr. Daniels knows
> less than 500 Chinese characters.

No, I know fewer than 500 characters.

> >>> Most Chinese words are (written with) two characters.
>
> Does anyone have a number for this?
> --- my guess is that a bit over half (50-70%) of
> Chinese words are (written with) two characters.

What is your definition of "word"?

> 中华人民共和国
> 中華 人民 共和国 2 + 2 + 3 (2 + 2 + 2+1) characters.

Have you ever opened a Chinese-English, English-Chinese dictionary?

Hen Hanna

unread,
Jul 8, 2016, 3:56:49 PM7/8/16
to

On Friday, July 8, 2016 at 12:46:01 PM UTC-7, Peter T. Daniels wrote:
> On Friday, July 8, 2016 at 3:21:44 PM UTC-4, Hen Hanna wrote:
> > On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:
> > > .....
> > > Characters DO NOT represent words, they represent morphemes, and nearly every
> > > Chinese word is two morphemes = two syllables = two characters. There are a few
> > > words that are disyllabic but a single morpheme.
> >
> >
> >
> > > Characters DO NOT represent words
> >
> > e.g. 中华人民共和国 <-- I think every
> > single character here represents a word.
> >
> > The probability that a Chinese character represents
> > a word is like 99.9% ?
> >
> >
> > __________________
> >
> > >>> Nearly every Chinese word is two characters.
> > >>> Most Chinese words are (written with) two characters.
> >
> >
> > I've heard this before, and I think ...
> > Nearly every person
> > who believes this doesn't know any Chinese. ... & knows
> > less than 100 Chinese characters.
> > most definitely less than 500 Chinese characters.
> >
> > I 'd like a confirmation that Mr. Daniels knows
> > less than 500 Chinese characters.
>
> No, I know fewer than 500 characters.
>

>
> What is your definition of "word"?
>
> Have you ever opened a Chinese-English, English-Chinese dictionary?


Yes, I was getting to that.

When you look in a Chinese dictionary, every character in
中华人民共和国
is listed (as a headword).


中 --- [inside], which is a word (within / among / in / middle / center / while (doing sth) / during / (dialect) OK / all right )

华 --- [flower], which is a word
https://www.mdbg.net/chindict/chindict.php?wdqchs=%E5%8D%8E
...
and so on.


So it seems that
中华人民共和国 <-- I think every
single character here represents a word.

HH

Hen Hanna

unread,
Jul 8, 2016, 4:02:32 PM7/8/16
to

On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:
> ...
> Characters DO NOT represent words, they represent morphemes, and nearly every
> Chinese word is two morphemes = two syllables = two characters.
> There are a few words that are disyllabic but a single morpheme.


Where did you _get_ that crap? (real question)


The last sentence is an interesting puzzle.

-- A Chinese word that is disyllabic but a single morpheme.

Could you give an example? HH

Peter T. Daniels

unread,
Jul 8, 2016, 4:32:02 PM7/8/16
to
On Friday, July 8, 2016 at 4:02:32 PM UTC-4, Hen Hanna wrote:
> On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:

> > Characters DO NOT represent words, they represent morphemes, and nearly every
> > Chinese word is two morphemes = two syllables = two characters.
> > There are a few words that are disyllabic but a single morpheme.
>
> Where did you _get_ that crap? (real question)

Virtually every book published about Chinese writing (by a scholar, not a
journalist) since 1838. You might start with John DeFrancis's *Visible Speech*
(1989).

> The last sentence is an interesting puzzle.
>
> -- A Chinese word that is disyllabic but a single morpheme.
>
> Could you give an example? HH

The most famous example is the word that means 'butterfly' -- I think it's _hudie_. The Yale Sinologist George Kennedy published a beautiful essay about
it (and its type) in, IIRC, 1964/

Peter T. Daniels

unread,
Jul 8, 2016, 4:43:32 PM7/8/16
to
On Friday, July 8, 2016 at 3:56:49 PM UTC-4, Hen Hanna wrote:
> On Friday, July 8, 2016 at 12:46:01 PM UTC-7, Peter T. Daniels wrote:

> > What is your definition of "word"?
> > Have you ever opened a Chinese-English, English-Chinese dictionary?
>
> Yes, I was getting to that.
>
> When you look in a Chinese dictionary, every character in
> 中华人民共和国
> is listed (as a headword).

And some of them aren't even glossed because there's no English equivalent
vague enough to equate to the lexical space it occupies.

> 中 --- [inside], which is a word (within / among / in / middle / center / while (doing sth) / during / (dialect) OK / all right )
>
> 华 --- [flower], which is a word
> https://www.mdbg.net/chindict/chindict.php?wdqchs=%E5%8D%8E
> ...
> and so on.

And the words beginning with that character are listed, each with the usual
sort of array of equivalents found in a bilingual dictionary.

Next to the computer I keep my smallest Chinese-English dictionary -- Langenscheidt
Pocket, with the yellow plastic cover. Unfortunately it's in pinyin order
because (like all the others currently available in US bookstores) it uses
Simplified characters. Opening it at random, I find an entry for _guang_,
glossed as 'light, brightness, glory, finished, gone, bare, naked, only',
and then more than a column of words beginning with that character (intermixed
with a few words beginning with two other characters with the same pronunciation).

> So it seems that
> 中华人民共和国 <-- I think every
> single character here represents a word.

I asked you to define "word."

Hen Hanna

unread,
Jul 8, 2016, 4:44:06 PM7/8/16
to
Thank you... I see it.

>>> some disyllabic morphemes, such as 葡萄 pútáo 'grape',蝴蝶 húdié 'butterfly', and 玻璃 bōli 'glass' had been introduced in Chinese as loanwords much earlier. Loanwords are formed by dividing foreign words into syllables and ... <<<


I'm going to get a Trade Mark for
-- The great [two-character Chinese word] hoax.

HH

Arnaud Fournet

unread,
Jul 9, 2016, 3:20:57 AM7/9/16
to
of course, most frequent words have only one syllable

wo3 = me, I
ni3 = you
ta1 = he she it
zhe4 = this
na4 = that
shi4 = to be
zai4 = be here, be in (some place)
etc.

Dingbat

unread,
Jul 9, 2016, 7:36:57 AM7/9/16
to
On Saturday, July 9, 2016 at 12:51:44 AM UTC+5:30, Hen Hanna wrote:
> On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:
> > .....
> > Characters DO NOT represent words, they represent morphemes, and nearly every
> > Chinese word is two morphemes = two syllables = two characters. There are a few
> > words that are disyllabic but a single morpheme.
>
>
>
> > Characters DO NOT represent words
>
> e.g. 中华人民共和国 <-- I think every
> single character here represents a word.
>
> The probability that a Chinese character represents
> a word is like 99.9% ?
>
How many words are there in 1) Beantown, 2) Bay City and 3) Chengshi?

Peter T. Daniels

unread,
Jul 9, 2016, 8:15:16 AM7/9/16
to
On Saturday, July 9, 2016 at 3:20:57 AM UTC-4, Arnaud Fournet wrote:

> of course, most frequent words have only one syllable
>
> wo3 = me, I
> ni3 = you
> ta1 = he she it
> zhe4 = this
> na4 = that
> shi4 = to be
> zai4 = be here, be in (some place)
> etc.

I don't see _you_ offering a definition of "word," either.

Helmut Richter

unread,
Jul 9, 2016, 11:07:01 AM7/9/16
to
Am 09.07.2016 um 14:15 schrieb Peter T. Daniels:

> I don't see _you_ offering a definition of "word," either.

In *this* context: if a lexical item (such as an entry in a dictionary)
is written with two characters, both of which are also lexical items,
then this is what is meant with "consists of two words".

Chinese example: a squirrel is 松鼠 composed of 松 (pine) and 鼠 (rat). So
this is composed of two words irrespective of whether "pine rat" is a
reasonable name of a squirrel -- I like it.

German example: a pencil is "Bleistift" composed of "Blei" (lead) and
"Stift" (pen). So this is composed of two words irrespective of whether
"lead pen" is a reasonable name of a pensil.

I have no idea how frequently two-syllable Chinese words can be analysed
as two words in this sense but the question is not meaningless.

Examples like these make me think that for the usage of the word
"ideograph" it is absolutely irrelevant whether it is a reasonable name
of characters of that kind. Words mean what their users choose them to mean.

And no, I see nothing racist in the notion. If you will, everything is
racist, for instance that the Westerners device letter-based writing
systems so primitive that a given text has a meaning in only one language.

--
Helmut Richter

Peter T. Daniels

unread,
Jul 9, 2016, 11:33:07 AM7/9/16
to
On Saturday, July 9, 2016 at 11:07:01 AM UTC-4, Helmut Richter wrote:
> Am 09.07.2016 um 14:15 schrieb Peter T. Daniels:
>
> > I don't see _you_ offering a definition of "word," either.
>
> In *this* context: if a lexical item (such as an entry in a dictionary)
> is written with two characters, both of which are also lexical items,
> then this is what is meant with "consists of two words".

It's the perennial confusion caused by Chinese terminology. Here's my paragraph on the matter from my impending book:

"The word “word” is tricky when dealing with Chinese. 字 zì signifies
both ‘character’ and the thing represented by a character, which is almost
always a single syllable, one morpheme. It’s the ordinary Chinese word for
talking about units of language bigger than a sound and smaller than a
sentence; as such it has been called the “sociological word.” 詞 cí is a
technical grammatical term that has been interpreted as “syntactic word”; it
corresponds to a considerable extent to ‘dictionary entry’, so is closest to
what we mean when we say an English word is ‘a sequence of letters between
spaces’—but there are no spaces between words in Chinese texts."

> Chinese example: a squirrel is 松鼠 composed of 松 (pine) and 鼠 (rat). So
> this is composed of two words irrespective of whether "pine rat" is a
> reasonable name of a squirrel -- I like it.
>
> German example: a pencil is "Bleistift" composed of "Blei" (lead) and
> "Stift" (pen). So this is composed of two words irrespective of whether
> "lead pen" is a reasonable name of a pensil.

You're not denying that "Bleistift" is a word, are you?

> I have no idea how frequently two-syllable Chinese words can be analysed
> as two words in this sense but the question is not meaningless.

Look at a dictionary. To take some examples from the page described earlier, do
all these words -- all the two character ones beginning with the same character
-- have something in common?

light pen, cursor, brilliance, blip, disc, patronize, smooth, halo, glow, be
barefoot, clear, glossy, bright, light year, CD-ROM, CD-ROM drive, aperture,
honor, look around, fiber optics, light ray, optics, shine

Many do. But some don't.

> Examples like these make me think that for the usage of the word
> "ideograph" it is absolutely irrelevant whether it is a reasonable name
> of characters of that kind. Words mean what their users choose them to mean.

We're not talking about words. Some words do represent ideas. We're talking
about glyphs. The elements of a writing system do not represent ideas.

> And no, I see nothing racist in the notion. If you will, everything is
> racist, for instance that the Westerners device letter-based writing
> systems so primitive that a given text has a meaning in only one language.

IF Chinese writing were ideographic, THEN it would have a meaning in more than
one language. But it doesn't, so it isn't. The former contributor here Lee Sau
Dan said he could get an idea of what a Japanese text was about, from the kanji.
He could not read it as Chinese.

Helmut Richter

unread,
Jul 9, 2016, 1:26:27 PM7/9/16
to
It is a compound word, that is, a word composed of two words. Some
compounds still reflect their original meaning (such as a "Suppenteller"
is a Teller to eat Suppe from), some do not (such as "Bleistift" whose
material contains no Blei).

Excatly in the same sense, 松鼠 is a compound word composed of two words,
and the meaning "a rodent living on trees" is not obvious to mean a
squirrel but still easy to understand. A less obvious example: a 先生
(first born = teacher) will probably not be understood if only because 生
has so many meanings. But a short explanation lets one see that it is
indeed a compound word and not just a two-syllable lexeme where the
syllables happen to be words whose meanings have nothing to do with the
meaning of the compound.

>> Examples like these make me think that for the usage of the word
>> "ideograph" it is absolutely irrelevant whether it is a reasonable
>> name of characters of that kind. Words mean what their users
>> choose them to mean.
>
> We're not talking about words. Some words do represent ideas. We're
> talking about glyphs. The elements of a writing system do not
> represent ideas.

For a "Bleistift" it is irrelevant whether it contains Blei, for an
"ideograph" it is irrelevant whether it represents ideas, as long as
everybody understands what is meant. Of course, it is nicer to have
terms whose etymological background matches their meaning.

>> And no, I see nothing racist in the notion. If you will, everything
>> is racist, for instance that the Westerners device letter-based
>> writing systems so primitive that a given text has a meaning in
>> only one language.
>
> IF Chinese writing were ideographic, THEN it would have a meaning in
> more than one language. But it doesn't, so it isn't. The former
> contributor here Lee Sau Dan said he could get an idea of what a
> Japanese text was about, from the kanji. He could not read it as
> Chinese.

But a Cantonese can read it even though the writer wrote Mandarin, can't
he? But he would certainly not understand spoken Mandarin unless he has
learnt it before as a foreign language.

My knowledge of both Chinese and Japanese is very close to zero.

I remember a package of Japanese tea with pictures how to prepare it. It
contained a time of 30 秒 to let the tea steep. Now, 30 sec is too short
by the same factor that 30 min is too long. So I looked up the glyph in
an online Chinese dictionary to find the meaning of the Japanese word.
It worked -- in fact confirmed my guess.

--
Helmut Richter

Ruud Harmsen

unread,
Jul 9, 2016, 1:30:48 PM7/9/16
to
Sat, 9 Jul 2016 08:33:03 -0700 (PDT): "Peter T. Daniels"
<gram...@verizon.net> scribeva:
>IF Chinese writing were ideographic, THEN it would have a meaning in more than
>one language. But it doesn't, so it isn't. The former contributor here Lee Sau
>Dan said he could get an idea of what a Japanese text was about, from the kanji.
>He could not read it as Chinese.

I can often get an idea what an Italian text is about, from the
similarity with Portuguese, Spanish, Interlingua. But I cannot read it
as Portuguese.

Peter T. Daniels

unread,
Jul 9, 2016, 1:48:40 PM7/9/16
to
So you're appealing to etymology. If someone says any of those compound words
to you, in either language -- probably even Suppenteller -- you do not mentally
decompose it into its elements and then recognize its meaning.

An English "soup plate," BTW, is not a plate but a bowl.

> >> Examples like these make me think that for the usage of the word
> >> "ideograph" it is absolutely irrelevant whether it is a reasonable
> >> name of characters of that kind. Words mean what their users
> >> choose them to mean.
> > We're not talking about words. Some words do represent ideas. We're
> > talking about glyphs. The elements of a writing system do not
> > represent ideas.
>
> For a "Bleistift" it is irrelevant whether it contains Blei, for an
> "ideograph" it is irrelevant whether it represents ideas, as long as
> everybody understands what is meant. Of course, it is nicer to have
> terms whose etymological background matches their meaning.

German words don't come into it at all: even Fournet _probably_ won't assert
that <Bleistift> is an ideogram. How is it any different from qianbi, which
comprises 'lead' and 'writing instrument'? (Which actually makes it look like
a calque from German. Where did the Chinese first get pencils? Were they still
actually made from lead when they did?)

> >> And no, I see nothing racist in the notion. If you will, everything
> >> is racist, for instance that the Westerners device letter-based
> >> writing systems so primitive that a given text has a meaning in
> >> only one language.
> > IF Chinese writing were ideographic, THEN it would have a meaning in
> > more than one language. But it doesn't, so it isn't. The former
> > contributor here Lee Sau Dan said he could get an idea of what a
> > Japanese text was about, from the kanji. He could not read it as
> > Chinese.
>
> But a Cantonese can read it even though the writer wrote Mandarin, can't
> he? But he would certainly not understand spoken Mandarin unless he has
> learnt it before as a foreign language.

It's what he was taught in school, as a "foreign" language, just like Arab
children everywhere don't learn to write their own language. (There's a lot
of research on that very question in Israel right now.)

> My knowledge of both Chinese and Japanese is very close to zero.
>
> I remember a package of Japanese tea with pictures how to prepare it. It
> contained a time of 30 秒 to let the tea steep. Now, 30 sec is too short
> by the same factor that 30 min is too long. So I looked up the glyph in
> an online Chinese dictionary to find the meaning of the Japanese word.
> It worked -- in fact confirmed my guess.

And Sau Dan would also have been able to interpret that instruction. That
doesn't mean either that he could read Japanese (he couldn't) or that the
instruction-writer could read Chinese.

Peter T. Daniels

unread,
Jul 9, 2016, 1:49:37 PM7/9/16
to
Exactly.

Ruud Harmsen

unread,
Jul 9, 2016, 2:35:48 PM7/9/16
to
Sat, 9 Jul 2016 10:48:38 -0700 (PDT): "Peter T. Daniels"
>German words don't come into it at all: even Fournet _probably_ won't assert
>that <Bleistift> is an ideogram. How is it any different from qianbi, which
>comprises 'lead' and 'writing instrument'? (Which actually makes it look like
>a calque from German. Where did the Chinese first get pencils? Were they still
>actually made from lead when they did?)

Don't know. Having Google Translate the Chinese Wikipedia article
sheds no light. It does show something I did not know, though: GT can
transform Chinese characters into PinYin -- with tone marks!

And there is a read aloud function too!

--
Ruud Harmsen, http://rudhar.com

Arnaud Fournet

unread,
Jul 9, 2016, 6:05:41 PM7/9/16
to
Thanks to ideograms, I can understand what a Japanese document deals with, even though I would never be able to do so, if I only had the spoken version.
That's what the powerful consequence of a non-phonetic cross-language system of writing.
But the idiot-in-town PTD does not have any experience of that situation.
A.

Peter T. Daniels

unread,
Jul 9, 2016, 10:59:08 PM7/9/16
to
The idiot Fournet does not even have the vocabulary for discussing writing
systems.

Is it "thanks to ideograms" that a monolingual English-reader can identify a
large number of words in a French or Latin text?

Arnaud Fournet

unread,
Jul 10, 2016, 4:57:04 PM7/10/16
to
this is irrelevant, idiot.
Besides, the "international" words in alphabetic languages indeed work as ideograms. About everybody understands what Hotel means.
A.

Ruud Harmsen

unread,
Jul 11, 2016, 7:00:38 AM7/11/16
to
Sun, 10 Jul 2016 13:56:59 -0700 (PDT): Arnaud Fournet
<fournet...@wanadoo.fr> scribeva:
>Besides, the "international" words in alphabetic languages indeed work as ideograms. About everybody understands what Hotel means.

Interlingua, le lingua vermente idiogrammatic!

Hen Hanna

unread,
Jul 11, 2016, 2:39:32 PM7/11/16
to
>>>> What is your definition of "word"?


Did anyone actually respond to Bill Clinton's comment
-- by giving him a definition of "is" ?

( Clinton said, "It depends upon what the meaning of the word 'is' is. If the—if he—if 'is' means is and never has been, that is not—that is one thing. If it means there is none, that was a completely true statement".[4] )

HH

Hen Hanna

unread,
Jul 11, 2016, 3:19:05 PM7/11/16
to
I (too) am going to try just one more time.


Often one of the first Chinese expressions
that Westerners learn is 我爱你 (wo ai ni)

How is each of the 3 characters (我爱你 wo ai ni) not a word ?

-- wo 我
-- ai 爱
-- ni 你


http://chineseonthego.com/t_I_love_you_chinese.html
.....

woaini - I Love You

The phrase 'I love you' in Chinese consists of three characters: 我爱你 (wo ai ni).
The first is the Chinese word for I -- wo 我.
The second is the key word ‘love’ -- ai 爱.
The third is ‘you’ in Chinese -- ni 你.

To say ‘I love you’ Mandarin way, put the three characters together: wo ai ni 我爱你.

Peter T. Daniels

unread,
Jul 11, 2016, 5:54:40 PM7/11/16
to
On Monday, July 11, 2016 at 3:19:05 PM UTC-4, Hen Hanna wrote:
> > > On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:

> > > > Characters DO NOT represent words,

You deleted "They represent morphemes."

> I (too) am going to try just one more time.
>
> Often one of the first Chinese expressions
> that Westerners learn is 我爱你 (wo ai ni)
>
> How is each of the 3 characters (我爱你 wo ai ni) not a word ?

Who says they're not?

> -- wo 我
> -- ai 爱
> -- ni 你

Unless you choose to analyze the two "pronouns" as clitics or inflections, of course.

> http://chineseonthego.com/t_I_love_you_chinese.html
> .....
>
> woaini - I Love You
>
> The phrase 'I love you' in Chinese consists of three characters: 我爱你 (wo ai ni).
> The first is the Chinese word for I -- wo 我.
> The second is the key word ‘love’ -- ai 爱.
> The third is ‘you’ in Chinese -- ni 你.
>
> To say ‘I love you’ Mandarin way, put the three characters together: wo ai ni 我爱你.

That must be a terribly scientific web site.

Hen Hanna

unread,
Jul 11, 2016, 6:57:52 PM7/11/16
to

On Monday, July 11, 2016 at 2:54:40 PM UTC-7, Peter T. Daniels wrote:
> On Monday, July 11, 2016 at 3:19:05 PM UTC-4, Hen Hanna wrote:
> > > > On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:
>
> > > > > Characters DO NOT represent words,
>
> You deleted "They represent morphemes."
>
> > I (too) am going to try just one more time.
> >
> > Often one of the first Chinese expressions
> > that Westerners learn is 我爱你 (wo ai ni)
> >
> > How is each of the 3 characters (我爱你 wo ai ni) not a word ?
>
> Who says they're not?
>
> > -- wo 我
> > -- ai 爱
> > -- ni 你
>
> Unless you choose to analyze the two "pronouns" as clitics or inflections, of course.
>


I think I had a glimpse into what's making this strange notion possible.
"nearly every Chinese word is two morphemes = two syllables = two characters."



http://mylanguages.org/chinese_nouns.php

> my car (wǒ de chē) - 我的车

车 car is 1 character.
... and many nouns for body parts
... and many nouns for everything else


elbow zhǒu - 肘
eye yǎn - 眼
face liǎn - 脸


leg tuǐ - 腿
lip chún - 唇
mouth kǒu - 口


shoulder jiān - 肩
stomach wèi - 胃

Peter T. Daniels

unread,
Jul 11, 2016, 11:43:24 PM7/11/16
to
On Monday, July 11, 2016 at 6:57:52 PM UTC-4, Hen Hanna wrote:
> On Monday, July 11, 2016 at 2:54:40 PM UTC-7, Peter T. Daniels wrote:
> > On Monday, July 11, 2016 at 3:19:05 PM UTC-4, Hen Hanna wrote:
> > > > > On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:

> > > > > > Characters DO NOT represent words,
> > You deleted "They represent morphemes."
> > > I (too) am going to try just one more time.
> > > Often one of the first Chinese expressions
> > > that Westerners learn is 我爱你 (wo ai ni)
> > > How is each of the 3 characters (我爱你 wo ai ni) not a word ?
> > Who says they're not?
> > > -- wo 我
> > > -- ai 爱
> > > -- ni 你
> > Unless you choose to analyze the two "pronouns" as clitics or inflections, of course.
>
> I think I had a glimpse into what's making this strange notion possible.
> "nearly every Chinese word is two morphemes = two syllables = two characters."
>
> http://mylanguages.org/chinese_nouns.php
>
> > my car (wǒ de chē) - 我的车
>
> 车 car is 1 character.

That character obviously did not mean 'automobile' when it was devised.

> ... and many nouns for body parts
> ... and many nouns for everything else
> elbow zhǒu - 肘
> eye yǎn - 眼
> face liǎn - 脸
> leg tuǐ - 腿
> lip chún - 唇
> mouth kǒu - 口
> shoulder jiān - 肩
> stomach wèi - 胃

How many words (詞 cí, not 字 zì) are there in Chinese? How many of them are a
single morpheme? Doesn't subtracting the second figure from the first leave
"nearly every"?

Arnaud Fournet

unread,
Jul 12, 2016, 4:05:23 AM7/12/16
to
LOL

A.

Arnaud Fournet

unread,
Jul 12, 2016, 4:06:26 AM7/12/16
to
Le lundi 11 juillet 2016 23:54:40 UTC+2, Peter T. Daniels a écrit :
> On Monday, July 11, 2016 at 3:19:05 PM UTC-4, Hen Hanna wrote:
> > > > On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:
>
> > > > > Characters DO NOT represent words,
>
> You deleted "They represent morphemes."
>
> > I (too) am going to try just one more time.
> >
> > Often one of the first Chinese expressions
> > that Westerners learn is 我爱你 (wo ai ni)
> >
> > How is each of the 3 characters (我爱你 wo ai ni) not a word ?
>
> Who says they're not?

your level of BS-monging and bad faith is awesome.
A.

Hen Hanna

unread,
Jul 12, 2016, 3:54:32 PM7/12/16
to

On Friday, July 8, 2016 at 1:32:02 PM UTC-7, Peter T. Daniels wrote:
> On Friday, July 8, 2016 at 4:02:32 PM UTC-4, Hen Hanna wrote:
> > On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:
>
> > > Characters DO NOT represent words, they represent morphemes, and nearly every
> > > Chinese word is two morphemes = two syllables = two characters.
> > > There are a few words that are disyllabic but a single morpheme.
> >
> > Where did you _get_ that crap? (real question)
>

> Virtually every book published about Chinese writing (by a scholar, not a
> journalist) since 1838. You might start with John DeFrancis's *Visible Speech* (1989).


Could you give me a short text fragment by which
I can search in this file

https://books.google.com/books?id=hypplIDMd0IC

and get to a description / notion of
[ nearly every Chinese word is two morphemes
= two syllables = two characters. ]

(in the book)? Thank you. HH

Peter T. Daniels

unread,
Jul 12, 2016, 4:31:42 PM7/12/16
to
I suppose it's a tribute to the book that there aren't used copies going for
"$3.45 and free shipping" at AbeBooks of either that one or his 1984 *The
Chinese Language: Fact and Fantasy*, though each is a bit over $10, but both
of them are worth reading in full -- are you near a library?

*Vis.Sp.* talks about characters rather than about their use. My *CLFaF* isn't
beside it on the shelf but that's presumably where you should look. DeFrancis
is also responsible for the ABC Dictionary, the best-regarded Chinese-English
dictionary.

Jerome Lee Packard, *The Morphology of Chinese* (2000), provides an introduction
to the nature of “words” in the modern language.

Hen Hanna

unread,
Jul 12, 2016, 5:15:27 PM7/12/16
to

On Tuesday, July 12, 2016 at 1:31:42 PM UTC-7, Peter T. Daniels wrote:
> On Tuesday, July 12, 2016 at 3:54:32 PM UTC-4, Hen Hanna wrote:
> > On Friday, July 8, 2016 at 1:32:02 PM UTC-7, Peter T. Daniels wrote:
> > > On Friday, July 8, 2016 at 4:02:32 PM UTC-4, Hen Hanna wrote:
> > > > On Thursday, July 7, 2016 at 8:00:43 PM UTC-7, Peter T. Daniels wrote:
>
> > > > > Characters DO NOT represent words, they represent morphemes, and nearly every
> > > > > Chinese word is two morphemes = two syllables = two characters.
> > > > > There are a few words that are disyllabic but a single morpheme.
> > > > Where did you _get_ that crap? (real question)
> > > Virtually every book published about Chinese writing (by a scholar, not a
> > > journalist) since 1838. You might start with John DeFrancis's *Visible Speech* (1989).
> >
> > Could you give me a short text fragment by which
> > I can search in this file
> >
> > https://books.google.com/books?id=hypplIDMd0IC
> >
> > and get to a description / notion of
> > [ nearly every Chinese word is two morphemes
> > = two syllables = two characters. ]
> >
> > (in the book)? Thank you. HH
>
> I suppose it's a tribute to the book that there aren't used copies going for
> "$3.45 and free shipping" at AbeBooks of either that one or his 1984 *The
> Chinese Language: Fact and Fantasy*, though each is a bit over $10, but both
> of them are worth reading in full


I wonder ...
how much Chinese could John DeFrancis actually speak ?


Apparently enough to know that
Chinese has many words of one syllable,
such as wo ("I"), hao ("good"), lai ("come").


Probably enough to never state anything resembling
[ nearly every Chinese word is two morphemes
= two syllables = two characters. ]



The Chinese Language: Fact and Fantasy
https://books.google.com/books?isbn=0824810686
John DeFrancis - 1984 - ‎Foreign Language Study

(Page 46) ........ To be sure, Chinese does have many words of one syllable, such as wo ("I"), hao ("good"), lai ("come"). It also has many expressions, whatever they might be ...

Peter T. Daniels

unread,
Jul 12, 2016, 11:18:17 PM7/12/16
to
LOOK IN THE DICTIONARY HE COMPILED.

What is with you people denying that Sinologists can speak Chinese?

Ruud Harmsen

unread,
Jul 13, 2016, 2:25:06 AM7/13/16
to
Tue, 12 Jul 2016 20:18:14 -0700 (PDT): "Peter T. Daniels"
>> Probably enough to never state anything resembling
>> [ nearly every Chinese word is two morphemes
>> = two syllables = two characters. ]
>
>LOOK IN THE DICTIONARY HE COMPILED.
>
>What is with you people denying that Sinologists can speak Chinese?

It helps if they can, but it isn't required, I would venture to say.

Peter T. Daniels

unread,
Jul 13, 2016, 7:57:27 AM7/13/16
to
Two specific individuals are in question here, one of whom I have observed in
fluent, deep conversation in Chinese, the other of whom was a US government
employee in China studying the writing-system situation before 1950 who
published the standard bilingual dictionary of the language.

Hen Hanna

unread,
Jul 14, 2016, 3:32:17 PM7/14/16
to
and these were the 1st things we learned in Chinese.



Are there just 2 ppl here who know ANY bit of Chinese?

If you're the 3rd such person, pls comment on the following.


The probability that a Chinese character represents
a word is like 99.9% ?

I can only think of a few exceptions. like 的 and 吗


> my car (wǒ de chē) 我的车

的 may not be a word.

吗 in 你好吗? "Ni Hai Hao Ma?" (你还好吗?)

吗 may not be a word.

HH

Peter T. Daniels

unread,
Jul 14, 2016, 10:24:22 PM7/14/16
to
What do you mean by "word"?

garyz...@gmail.com

unread,
Jul 19, 2016, 4:20:36 PM7/19/16
to
On Thursday, July 14, 2016 at 3:32:17 PM UTC-4, Hen Hanna wrote:
Hello, I know Chinese.

No, every Chinese character does not necessarily represent a word. Here, I will use your argument.

>When you look in a Chinese dictionary, every character in
> 中华人民共和国
>is listed (as a headword).
>
>
>中 --- [inside], which is a word (within / among / in / middle / center / while (doing sth) / during / (dialect) OK / all right )
>
>华 --- [flower], which is a word
>https://www.mdbg.net/chindict/chindict.php?wdqchs=%E5%8D%8E
>...
>and so on.

"Breakthrough" is an English word. I believe that it is really two words (trust me, I totally really genuinely do, without sarcasm). So when somebody tells me, "I made a breakthrough!", I ask him or her, "what was broken?", because "break" (noun) and "through" (preposition) are two different words.

If I want to be more ridiculous, then watch this. "cow" is an English word, would you agree? When I put all the characters in "cow" into an English dictionary, then I find that each character, in fact, is in the dictionary! Wow! Here are my findings:

"c"
Definition 1 (letter): the third letter in the English alphabet
Definition 2 (abbreviation): short for "circa"

"o"
Definition 1 (letter): the fifteenth letter in the English alphabet
Definition 2 (abbreviation): alternative of saying "zero"
Definition 3 (interjection): alternative form of "oh", now uncommon

"w"
Definition (letter): the twenty-third letter in the English alphabet

Therefore, "cow" is three words!

Clearly, this is rubbish nonsense. Like what Helmut Richter wrote about German words such as „Bleistift‟, words which are on their own are not the same when compounded. (Only in Classical Chinese, an entirely different language from Mandarin, do you have it that 99.9% of words are also one-character.) So when you ask "just 2 ppl here who know ANY bit of Chinese?", it seems that Peter T. Daniels knows a great deal more than you do. In English, if I said, "I see a schoolhouse", other English speakers would not say "You see a school housing what?".

Interestingly enough, your first example uses "中华人民共和国". If you searched up this phrase on the Internet, perhaps you would find out what it really is. Using your approach:

中 Definition: middle
华 Definition: Chinese
人 Definition: person, human
民 Definition: ethnicity, people
共 Definition: together
和 Definition: together, and
国 Definition: country

Therefore, your methodology translates 中华人民共和国 into "Middle Chinese people ethnicity togetherness country".

The actual translation?

中华 Definition: Chinese
人民 Definition: people, ethnicity
共和国 Definition: republic

中华人民共和国 is "The People's Republic of China".

To answer your question about are "的" and "吗" words (by the linguistic definition), 吗 is a word, more specifically, a particle, that marks a question. 的 is considered an individual word by Wiktionary, although I could argue that it is really a suffix added to a noun to form a possessive (genitive) case. It also has many other definitions aside from a possessive marker, but in this instance, I would say it could go either way.

semir...@my-deja.com

unread,
Jul 20, 2016, 5:19:54 AM7/20/16
to
> 中华人民共和国
> 中華 人民 共和国 2 + 2 + 3 (2 + 2 + 2+1) characters.
>
> HH

This is taking a concept from one area and trying to apply it
to another.

The fact is that language is both written and spoken, and one
affects the other to some degree.

As Modern Standard Chinese has only about 420 different sounds,
the one character word is fairly pointless in much of speech, as
it is difficult to clarify which one is being used.

Opening my dictionary at a random page there are more than
a dozen different characters for shan pronounced with the
fourth tone.

So, in my opinion trying to define rigidly the concept
of "word" in Chinese leads only to argument. Better to
concentrate on the characters and use concepts based
on them

Athel Cornish-Bowden

unread,
Jul 21, 2016, 4:34:59 AM7/21/16
to
Or taking a concept from one language and trying to apply it to another.

Even in European languages the concept of a word is not clearcut.
Hungarian treats "I love you" as one word, "szeretlek". Welsh treats it
as about six: "Rwyf i'n dy garu di". Spanish treats it as two: "te
quiero" (but you can also say it in three: "yo te quiero". French
writes it as three, "je t'aime", but if the writing had developed
differently it could easily be considered one, as that's how it's
pronounced. If we pay too much attention to spacing we could regard
中华人民共和国 as a word, but I doubt whether anyone would.
>
> The fact is that language is both written and spoken, and one
> affects the other to some degree.
>
> As Modern Standard Chinese has only about 420 different sounds,
> the one character word is fairly pointless in much of speech, as
> it is difficult to clarify which one is being used.
> Opening my dictionary at a random page there are more than
> a dozen different characters for shan pronounced with the
> fourth tone.
>
> So, in my opinion trying to define rigidly the concept
> of "word" in Chinese leads only to argument. Better to
> concentrate on the characters and use concepts based
> on them


--
athel

Hen Hanna

unread,
Jul 26, 2016, 3:59:30 PM7/26/16
to
> Within an average Chinese person's vocabulary
> (which may be 30,000 or so words,
> but this number is not important)
> could you guess
> what percentage of the words are 1-character, and
> what percentage of the words are 2-characters?
>
>
> My revised guess would be:
> 1-character words ... maybe 45% (水 石 木 好 我 爱 你 ...)
> 2-character words ... maybe 20% (人民, 君子, 小人, 水滴 ...)
>
> ( both numbers may be a bit higher, like 50% and 25% )
>

_____________________


> I (too) am going to try just one more time.

M. Arnaud Fournet -- You've made a good effort. but
I think .. most of our efforts were in vain.
mostly because only 2 ppl (you and I)
know Chinese characters
or Chinese, here at Sci.Lang.

( when an ignorant, rude (and/or senile?)
person calls you names, pls refrain from
responding to him in a similar manner. )



[The Oxford Handbook of Chinese Psychology] (book) >>> Compared to nouns, most verbs in Chinese are shorter, and more than 40 per cent are monosyllabic. In English, nouns and verbs overlap more closely in word length. Thus, the monosyllabic nature of Chinese is more clearly reflected in verbs ...


> 40+ % are monosyllabic. <-------- much lower than I thought.



[PDF]Vocabulary Development in English and Chinese: A ... - CiteSeerX
citeseerx.ist.psu.edu /viewdoc/download?doi=10.1.1.470.8174&rep=rep1...
by X Zhao - ‎Cited by 10 - ‎Related articles
suggested that Chinese adults use more verbs than nouns when they speak .... increase in the size of acquired vocabulary within a short .... phonemes, and more than 40 percent of all the verbs have such a ... So the proportion of monosyllabic.



[Footnote 4]

According to Wang (1994), 44 percent of 3000 most frequently
speaking Chinese words are monosyllabic;

and in the 8822 words used by Chinese government to evaluate student’s vocabulary level, 22 percent of them are monosyllabic words (Xing, 2006).


Peter T. Daniels

unread,
Jul 26, 2016, 5:07:37 PM7/26/16
to
On Tuesday, July 26, 2016 at 3:59:30 PM UTC-4, Hen Hanna wrote:

> [The Oxford Handbook of Chinese Psychology] (book) >>> Compared to nouns, most verbs in Chinese are shorter, and more than 40 per cent are monosyllabic. In English, nouns and verbs overlap more closely in word length. Thus, the monosyllabic nature of Chinese is more clearly reflected in verbs ...
>
> > 40+ % are monosyllabic. <-------- much lower than I thought.

You're making progress in understanding the facts.

Next step: What percentage of a language's vocabulary is verbs, and what
percentage is nouns?

> [PDF]Vocabulary Development in English and Chinese: A ... - CiteSeerX
> citeseerx.ist.psu.edu /viewdoc/download?doi=10.1.1.470.8174&rep=rep1...
> by X Zhao - ‎Cited by 10 - ‎Related articles

> suggested that Chinese adults use more verbs than nouns when they speak .... increase in the size of acquired vocabulary within a short .... phonemes, and more than 40 percent of all the verbs have such a ... So the proportion of monosyllabic.

Sorry, but the Google Scholar ellipses (caused by your choice of search terms
and subsequent failure to consult the actual text, especially given that you
could download the complete study) makes that passage unintelligible.

> [Footnote 4]
>
> According to Wang (1994), 44 percent of 3000 most frequently
> speaking Chinese words are monosyllabic;
>
> and in the 8822 words used by Chinese government to evaluate student’s vocabulary level, 22 percent of them are monosyllabic words (Xing, 2006).

That is, a bit fewer than 2000 words (1,941 to be exact) are monosyllabic. How
many monosyllabic words are there likely to be outside the 8822 most common?

Do you know how many words are in a typical adult's vocabulary (in any language)?
Around 30,000.

You couldn't hold much of a conversation if you used only the 2000 most common
English words.

Franz Gnaedinger

unread,
Jul 27, 2016, 2:45:02 AM7/27/16
to
On Friday, July 8, 2016 at 10:43:32 PM UTC+2, Peter T. Daniels wrote:
>
> I asked you to define "word."

You asked Hen Hanna, but let me say a word on the etymology of English word.
When I say 'let me say a word' I mean not a single word but a series of words
and even sentences. 'Word' covers anything from one single word to a lot of
words, and the Magdalenian etymology is BIR DAP, fur BIR activity of hand DAP,
indicating a shaman or shamaness making an important announcment or having
a word with deities or ghosts or ancestors while tambourines were beaten,
tambourine from inverse DAB BIR ... This my pleading for not to take terms
too narrowly. Keep an open mind, important is that we understand each other.
Word can be a single word or a compound of two or more words or even a speech
on a single topic, a bundle of coherent words, coherent being the key word.

Peter T. Daniels

unread,
Jul 27, 2016, 7:47:58 AM7/27/16
to
On Wednesday, July 27, 2016 at 2:45:02 AM UTC-4, Franz Gnaedinger wrote:
> On Friday, July 8, 2016 at 10:43:32 PM UTC+2, Peter T. Daniels wrote:
> >
> > I asked you to define "word."
>
> You asked Hen Hanna, but let me say a word on the etymology of English word.

Of course you can say all you want about your etymological fantasies, but they
(even if they were completely accurate) have nothing to do with the meaning.

> When I say 'let me say a word' I mean not a single word but a series of words
> and even sentences. 'Word' covers anything from one single word to a lot of
> words, and the Magdalenian etymology is BIR DAP, fur BIR activity of hand DAP,
> indicating a shaman or shamaness making an important announcment or having
> a word with deities or ghosts or ancestors while tambourines were beaten,
> tambourine from inverse DAB BIR ... This my pleading for not to take terms
> too narrowly. Keep an open mind, important is that we understand each other.
> Word can be a single word or a compound of two or more words or even a speech
> on a single topic, a bundle of coherent words, coherent being the key word.

Nor in English, it can't, and I suspect not in (Swiss) German, either.

Franz Gnaedinger

unread,
Jul 28, 2016, 3:30:31 AM7/28/16
to
On Wednesday, July 27, 2016 at 1:47:58 PM UTC+2, Peter T. Daniels wrote:
>
> Of course you can say all you want about your etymological fantasies, but they
> (even if they were completely accurate) have nothing to do with the meaning.
>

The first DAP BIR tambour was the gorilla silver back tapping DAP on the fur
BIR of his chest. Beating a tambourine or a drum can convey a signal or even
proper language. I remember having read of an African tribe communicating
with neighboring tribes by beating a wooden drum in the speaking rhythm
of actual sentences, and the message is understood verbatim, word for word,
by the receiver! This amazing exchange of language may be due to what I call
micro-timing, in combination with a relatively small vocabulary, a standard
repertoir of expressions and phrases, and a lot of practice from childhood
onward. An Indian tabla played by a master evokes talking, the name of the
tabla perhaps deriving from DAP )OG or DAP LOG, activity of hand DAP having
the say )OG or LOG, together something like tapping language, conveying
a mood, but perhaps, for Indian ears, also semantic meaning. Latin verbum
'word' may combine the fur BIR with onomatopoeic bum German Bumm English boom,
a parallel being bong in the name of the bonga drum. BIR )OG or BIR LOG may
account for Italian parola 'word' French parler 'speak', the beaten fur of
a tambourine accompanying the speech of originally a shaman or shamaness.
French mot 'word' might be akin to motoric, from Magdalenian MOT for a
repetitive forward and backward movement (like the beating of a tambourine).
'Tapping language' might be the origin of meters like the iamb and hexameter.
From my perspective I see a clear connection between tambourine and word.

Arnaud Fournet

unread,
Jul 28, 2016, 10:28:51 AM7/28/16
to
Le mardi 26 juillet 2016 21:59:30 UTC+2, Hen Hanna a écrit :
> On Monday, July 11, 2016 at 12:19:05 PM UTC-7, Hen Hanna wrote:
> > On Saturday, July 9, 2016 at 12:20:57 AM UTC-7, Arnaud Fournet wrote:
> > > Le vendredi 8 juillet 2016 21:21:44 UTC+2, Hen Hanna a écrit :
> >

>
> > I (too) am going to try just one more time.
>
> M. Arnaud Fournet -- You've made a good effort. but
> I think .. most of our efforts were in vain.

No, it was in vain,
because we made it clear that PTD is incompetently pontificating on issues he knows nothing about.
And I'll add: once again.
A.


> mostly because only 2 ppl (you and I)
> know Chinese characters
> or Chinese, here at Sci.Lang.

maybe not,
after all, people at large possibly learnt something in the process.
A.

>
> ( when an ignorant, rude (and/or senile?)
> person calls you names, pls refrain from
> responding to him in a similar manner. )

At this point, PTD has opened the way for a new kind of trolls: the pontificating trollish dumb.
P for Pontificating
T for Trollish
D for Dumb

Ruud Harmsen

unread,
Jul 29, 2016, 1:30:03 AM7/29/16
to
Tue, 26 Jul 2016 14:07:36 -0700 (PDT): "Peter T. Daniels"
<gram...@verizon.net> scribeva:
>You couldn't hold much of a conversation if you used only the 2000 most common
>English words.

It is claimed that a series of Interlingua books uses only a basic set
of 2500 words (that is, any words needed but outside that set, are
explained in words from the set).

http://www.interlingua.com/facilealeger

Peter T. Daniels

unread,
Jul 29, 2016, 9:03:05 AM7/29/16
to
On Friday, July 29, 2016 at 1:30:03 AM UTC-4, Ruud Harmsen wrote:
> Tue, 26 Jul 2016 14:07:36 -0700 (PDT): "Peter T. Daniels"
> <gram...@verizon.net> scribeva:

> >You couldn't hold much of a conversation if you used only the 2000 most common
> >English words.
>
> It is claimed that a series of Interlingua books uses only a basic set
> of 2500 words (that is, any words needed but outside that set, are
> explained in words from the set).

How can that be claimed to be "using only 2500 words"?

> http://www.interlingua.com/facilealeger

Any school dictionary is going to use a limited vocabulary in its definitions,
but still it's going to define 20,000 or so words that a youngster will encounter.

Ruud Harmsen

unread,
Jul 29, 2016, 12:05:26 PM7/29/16
to
Fri, 29 Jul 2016 06:03:02 -0700 (PDT): "Peter T. Daniels"
<gram...@verizon.net> scribeva:

>On Friday, July 29, 2016 at 1:30:03 AM UTC-4, Ruud Harmsen wrote:
>> Tue, 26 Jul 2016 14:07:36 -0700 (PDT): "Peter T. Daniels"
>> <gram...@verizon.net> scribeva:
>
>> >You couldn't hold much of a conversation if you used only the 2000 most common
>> >English words.
>>
>> It is claimed that a series of Interlingua books uses only a basic set
>> of 2500 words (that is, any words needed but outside that set, are
>> explained in words from the set).
>
>How can that be claimed to be "using only 2500 words"?

2500 words and a small, clearly defined, number of others.

Il existe un dictionario que explique tote iste 2500 parolas in iste
mesme 2500 parolas.
http://www.interlingua.com/novas/2011-12-10-dictionario

>> http://www.interlingua.com/facilealeger
>
>Any school dictionary is going to use a limited vocabulary in its definitions,
>but still it's going to define 20,000 or so words that a youngster will encounter.

Antonio Marques

unread,
Aug 5, 2016, 4:27:28 PM8/5/16
to
'All the words that are not in the [2500 word] dictionary are explained
at the foot of each page'. That's not necessarily remarkable, but it
doesn't get us very far, because the explanation may be just enough to
make some sense of the text rather than knowing what the word means. E.g.
- fork: thing used to eat
- knife: thing used to eat
- spoon: thing used to eat

Daud Deden

unread,
Sep 11, 2019, 10:25:47 AM9/11/19
to
https://www.youtube.com/watch?v=_OO2PuGz-H8
Peter Gabriel's Solisbury Hill
At about 1:20 the sound of the talking drum producing a deep sound like waterdrops, due to the tension of chords being tightening and loosening.

benl...@ihug.co.nz

unread,
Sep 11, 2019, 7:26:26 PM9/11/19
to
On Thursday, July 28, 2016 at 7:30:31 PM UTC+12, Franz Gnaedinger wrote:
> On Wednesday, July 27, 2016 at 1:47:58 PM UTC+2, Peter T. Daniels wrote:
> >
> > Of course you can say all you want about your etymological fantasies, but they
> > (even if they were completely accurate) have nothing to do with the meaning.
> >
>
> The first DAP BIR tambour was the gorilla silver back tapping DAP on the fur
> BIR of his chest.

...probably the least furry part of his body

https://gorillafund.org/when-gorillas-groups-interact-anything-can-happen/

benl...@ihug.co.nz

unread,
Sep 11, 2019, 7:34:28 PM9/11/19
to
Right about talking drums in general.
However:

"Rather than employing a full drum kit, Allan Schwartzberg made do with
a shaker in one hand and a drum stick in another, which he used to strike
a telephone book. For additional rhythmic textures, Larry Fast constructed
a fake drum kit on his keyboard, which he dubbed the "synthibam"...After
all of the session musicians departed, Fast also overdubbed some additional electronics, including the synth horn orchestration."

https://en.wikipedia.org/wiki/Solsbury_Hill_(song)

Don't know whether the talking drum effect was done with the "synthibam"
or as "additional electronics".

Here's a clearer example:

https://www.youtube.com/watch?v=B4oQJZ2TEVI

Daud Deden

unread,
Sep 11, 2019, 9:30:53 PM9/11/19
to
On Wednesday, September 11, 2019 at 7:26:26 PM UTC-4, benl...@ihug.co.nz wrote:
> On Thursday, July 28, 2016 at 7:30:31 PM UTC+12, Franz Gnaedinger wrote:
> > On Wednesday, July 27, 2016 at 1:47:58 PM UTC+2, Peter T. Daniels wrote:
> > >
> > > Of course you can say all you want about your etymological fantasies, but they
> > > (even if they were completely accurate) have nothing to do with the meaning.
> > >
> >
> > The first DAP BIR tambour was the gorilla silver back tapping DAP on the fur
> > BIR of his chest.
>
> ...probably the least furry part of his body

bare pelage


> https://gorillafund.org/when-gorillas-groups-interact-anything-can-happen/

Daud Deden

unread,
Sep 11, 2019, 9:40:38 PM9/11/19
to
Thanks Ross, I didn't go deeper into the studio sounds. Back at UW, an American friend-classmate that learned Chinese and married a Taiwanese woman was a band's drummer, he recorded water dripping with many synthesized effects, amazing 'music'.
Reminds me too of the pygmy earth drum, a pit covered by a tensed skin.

Daud Deden

unread,
Sep 11, 2019, 10:01:20 PM9/11/19
to
My interest in Peter Gabriel's songs Solisbury Hill and In your eyes were inspired by his Secret World concerts on youtube videos, though I must have heard earlier versions on radio.
0 new messages